From 16ebdea5c62c24355fa6d18205a2edf098613d07 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 11 Feb 2026 13:17:47 -0800 Subject: [PATCH] docs: Reorganize ValidMind Library "portal" around sorted Jupyter Notebooks (#1168) * Pulling in restructured notebooks * Updating script for auto-populating use-cases in sidebar * 1st round of broken links * Pulling in updated restructured notebooks * Updating developer/validmind-library.qmd landing page * Massaging sidebar * More cleanup developer/validmind-library.qmd * Modifying use case script to auto-populate use-cases in panel tabs * Adjust * Setup for how-to run tests * Save point * Editing model-testing/testing-overview.qmd * Save point * Save point * More cleanup: model-testing/testing-overview.qmd * Shuffling test desc page around * Save point * Save point * developer/model-testing > developer/how-to * Populate how-to notebooks -testing script * Edit * Updated notebooks again & how-to script edit * Fixing missed broken links * Edit * Fixing broken img in training * Notebook update & adjusting developer/supported-models.qmd * Edit * Tweak * Tweakkk * Future-proofing * Last edit? * I lied * One mooooore * dsfghjks * Update site/developer/how-to/testing-overview.qmd Co-authored-by: Nik Richers * Update site/developer/validmind-library.qmd Co-authored-by: Nik Richers * Typo * One more typo --------- Co-authored-by: Nik Richers --- site/Makefile | 6 +- site/about/overview-model-documentation.qmd | 4 +- site/about/use-cases/e-23.qmd | 4 +- site/about/use-cases/eu-ai-act.qmd | 4 +- site/about/use-cases/sr-11-7.qmd | 2 +- site/developer/_sidebar.yaml | 109 +- site/developer/how-to/feature-overview.qmd | 69 + .../test-sandbox.qmd | 1 + site/developer/how-to/testing-overview.qmd | 169 ++ .../model-testing/testing-overview.qmd | 158 -- site/developer/samples-jupyter-notebooks.qmd | 126 +- site/developer/supported-models.qmd | 12 +- .../{model-testing => }/test-descriptions.qmd | 13 +- site/developer/validmind-library.qmd | 76 +- site/faq/_faq-images.qmd | 2 +- site/faq/_faq-synthetic-datasets.qmd | 2 +- site/faq/faq-integrations.qmd | 2 +- site/faq/faq-testing.qmd | 16 +- site/guide/access/accessing-validmind.qmd | 2 +- .../work-with-content-blocks.qmd | 2 +- site/guide/monitoring/enable-monitoring.qmd | 2 +- site/guide/monitoring/ongoing-monitoring.qmd | 8 +- .../monitoring/review-monitoring-results.qmd | 2 +- .../work-with-metrics-over-time.qmd | 4 +- site/notebooks.zip | Bin 64030384 -> 62713235 bytes .../1-set_up_validmind.ipynb | 4 +- .../2-start_development_process.ipynb | 2 +- .../3-integrate_custom_tests.ipynb | 4 +- .../4-finalize_testing_documentation.ipynb | 20 +- .../1-set_up_validmind_for_validation.ipynb | 4 +- .../2-start_validation_process.ipynb | 2 +- .../4-finalize_validation_reporting.ipynb | 16 +- site/notebooks/README.md | 6 +- .../langgraph_agent_simple_banking_demo.ipynb | 1501 ----------- ...start_option_pricing_models_quantlib.ipynb | 1330 ---------- .../quickstart_code_explainer_demo.ipynb | 862 ------- .../custom_tests/implement_custom_tests.ipynb | 1095 -------- .../nlp_and_llm/rag_benchmark_demo.ipynb | 1857 -------------- .../nlp_and_llm/rag_documentation_demo.ipynb | 1681 ------------- .../{ => data_and_datasets}/dataset_image.png | Bin .../configure_dataset_features.ipynb | 4 +- .../load_datasets_predictions.ipynb | 4 +- .../{ => data_and_datasets}/model_image.png | Bin .../use_dataset_model_objects.ipynb | 1003 ++++++++ .../metrics}/add_metric_over_time_block.png | Bin .../metrics}/log_metric_accuracy.png | Bin .../metrics}/log_metric_attention.png | Bin .../metrics}/log_metric_auc_1.png | Bin .../metrics}/log_metric_auc_2.png | Bin .../metrics}/log_metric_auc_3.png | Bin .../metrics}/log_metric_auc_4.png | Bin .../metrics}/log_metric_f1.png | Bin .../metrics}/log_metric_precision.png | Bin .../metrics}/log_metric_recall.png | Bin .../metrics}/log_metric_satisfactory.png | Bin .../metrics}/log_metric_satisfactory_2.png | Bin .../{ => metrics}/log_metrics_over_time.ipynb | 28 +- .../{ => metrics}/run_unit_metrics.ipynb | 2 - .../assign_scores_complete_tutorial.ipynb | 2 - .../external-data-custom-test.png | Bin .../hyperparameters-custom-metric.png | Bin .../custom_tests}/image-in-custom-metric.png | Bin .../custom_tests/implement_custom_tests.ipynb | 1093 ++++++++ .../insert-test-driven-block-custom.png | Bin ...insert-test-driven-block-test-provider.png | Bin .../insert-test-driven-block.png | Bin .../integrate_external_test_providers.ipynb | 6 +- .../multiple-tables-plots-custom-metric.png | Bin .../parameterized-custom-metric.png | Bin ...pearson-correlation-matrix-test-output.png | Bin .../pearson-correlation-matrix.png | Bin .../explore_tests}/explore_test_suites.ipynb | 4 +- .../explore_test_suites_output.png | Bin .../explore_tests}/explore_tests.ipynb | 4 +- .../explore_tests}/test_suite_output1.png | Bin .../explore_tests}/test_suite_output2.png | Bin .../run_tests/1_run_dataset_based_tests.ipynb | 10 +- .../run_tests/2_run_comparison_tests.ipynb | 8 +- .../customize_test_result_descriptions.ipynb | 2 - .../enable_pii_detection.ipynb | 4 +- .../filter_input_columns.ipynb | 0 ...tests_that_require_multiple_datasets.ipynb | 4 +- .../understand_utilize_rawdata.ipynb | 0 ...t_multiple_results_for_the_same_test.ipynb | 4 +- .../run_documentation_sections.ipynb | 4 +- .../run_documentation_tests_with_config.ipynb | 4 +- .../how_to/use_dataset_model_objects.ipynb | 1005 -------- .../composite-metric-in-template-preview.png | Bin 152975 -> 0 bytes .../images/high-pearson-correlation-block.png | Bin 280332 -> 0 bytes .../insert-test-driven-block-correlations.png | Bin 84347 -> 0 bytes ...st-driven-block-custom-class-imbalance.jpg | Bin 188649 -> 0 bytes ...t-driven-block-custom-confusion-matrix.png | Bin 248533 -> 0 bytes site/notebooks/images/my_tests_directory.png | Bin 44368 -> 0 bytes .../images/selecting-composite-metric.png | Bin 232261 -> 0 bytes ...electing-high-pearson-correlation-test.png | Bin 484028 -> 0 bytes .../quickstart_model_documentation.ipynb | 4 +- .../quickstart_model_validation.ipynb | 4 +- .../1-set_up_validmind.ipynb | 4 +- .../2-start_development_process.ipynb | 2 +- .../3-integrate_custom_tests.ipynb | 4 +- .../4-finalize_testing_documentation.ipynb | 20 +- .../1-set_up_validmind_for_validation.ipynb | 4 +- .../2-start_validation_process.ipynb | 2 +- .../4-finalize_validation_reporting.ipynb | 16 +- .../use_cases/agents/agentic_ai_template.yaml | 343 +++ .../agents/banking_test_dataset.py | 32 - .../agents/banking_tools.py | 4 +- .../agents/document_agentic_ai.ipynb | 2197 +++++++++++++++++ .../agents/utils.py | 0 .../capital_markets_template.yaml | 227 ++ .../quickstart_option_pricing_models.ipynb | 2 +- ...start_option_pricing_models_quantlib.ipynb | 1345 ++++++++++ .../customer_churn_full_suite.py | 6 +- ...el_source_code_documentation_template.yaml | 142 ++ .../quickstart_code_explainer_demo.ipynb | 874 +++++++ .../credit_risk/CreditRiskData.xlsx | Bin .../application_scorecard_executive.ipynb | 4 +- .../application_scorecard_full_suite.ipynb | 4 +- .../application_scorecard_with_bias.ipynb | 4 +- .../application_scorecard_with_ml.ipynb | 4 +- .../ScoreBandDiscriminationMetrics.py | 0 ...document_excel_application_scorecard.ipynb | 6 +- .../validate_application_scorecard.ipynb | 10 +- .../model_validation/xgb_model_champion.pkl | Bin .../nlp_and_llm/datasets/bbc_text_cls.csv | 0 .../datasets/bbc_text_cls_reference.csv | 0 .../cnn_dailymail_100_with_predictions.csv | 0 .../cnn_dailymail_500_with_predictions.csv | 0 .../nlp_and_llm/datasets/sentiments.csv | 0 .../datasets/sentiments_with_predictions.csv | 0 .../foundation_models_integration_demo.ipynb | 2 - ...foundation_models_summarization_demo.ipynb | 0 .../nlp_and_llm/gen_ai_rag_template.yaml | 538 ++++ .../hugging_face_integration_demo.ipynb | 2 - .../hugging_face_summarization_demo.ipynb | 2 - .../nlp_and_llm/llm_summarization_demo.ipynb | 2 - .../nlp_and_llm/prompt_validation_demo.ipynb | 4 +- .../nlp_and_llm/rag_benchmark_demo.ipynb | 1869 ++++++++++++++ .../nlp_and_llm/rag_documentation_demo.ipynb | 1692 +++++++++++++ ...ication_scorecard_ongoing_monitoring.ipynb | 4 +- ...rt_customer_churn_ongoing_monitoring.ipynb | 6 +- .../ongoing_monitoring/xgboost_model.model | Bin .../quickstart_regression_full_suite.ipynb | 2 - .../quickstart_time_series_full_suite.ipynb | 4 +- .../quickstart_time_series_high_code.ipynb | 6 +- site/releases/2023/2023-nov-09/highlights.qmd | 2 +- site/releases/2023/2023-oct-25/highlights.qmd | 2 +- site/releases/2023/2023-sep-27/highlights.qmd | 8 +- .../2023/release-notes-2023-jun-22.qmd | 4 +- .../2023/release-notes-2023-may-30.qmd | 2 +- .../2024/2024-aug-13/release-notes.qmd | 2 +- .../2024/2024-dec-24/release-notes.qmd | 4 +- site/releases/2024/2024-feb-14/highlights.qmd | 4 +- site/releases/2024/2024-jan-26/highlights.qmd | 4 +- .../2024/2024-jul-22/release-notes.qmd | 4 +- .../2024/2024-may-22/release-notes.qmd | 2 +- .../2025/2025-jan-31/release-notes.qmd | 4 +- .../developer-sidebar/update_how_tos.py | 282 +++ .../developer-sidebar/update_use_cases.py | 228 ++ site/scripts/update_code_samples_sidebar.py | 94 - .../_validmind-test-repository.qmd | 2 +- .../finalizing-model-documentation.qmd | 2 +- site/training/program/training-faq.qmd | 6 +- .../what-is-validmind/what-is-validmind.qmd | 2 +- 164 files changed, 12462 insertions(+), 9967 deletions(-) create mode 100644 site/developer/how-to/feature-overview.qmd rename site/developer/{model-testing => how-to}/test-sandbox.qmd (94%) create mode 100644 site/developer/how-to/testing-overview.qmd delete mode 100644 site/developer/model-testing/testing-overview.qmd rename site/developer/{model-testing => }/test-descriptions.qmd (83%) delete mode 100644 site/notebooks/code_samples/agents/langgraph_agent_simple_banking_demo.ipynb delete mode 100644 site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb delete mode 100644 site/notebooks/code_samples/code_explainer/quickstart_code_explainer_demo.ipynb delete mode 100644 site/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb delete mode 100644 site/notebooks/code_samples/nlp_and_llm/rag_benchmark_demo.ipynb delete mode 100644 site/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb rename site/notebooks/how_to/{ => data_and_datasets}/dataset_image.png (100%) rename site/notebooks/how_to/{ => data_and_datasets/dataset_inputs}/configure_dataset_features.ipynb (98%) rename site/notebooks/how_to/{ => data_and_datasets/dataset_inputs}/load_datasets_predictions.ipynb (99%) rename site/notebooks/how_to/{ => data_and_datasets}/model_image.png (100%) create mode 100644 site/notebooks/how_to/data_and_datasets/use_dataset_model_objects.ipynb rename site/notebooks/{images => how_to/metrics}/add_metric_over_time_block.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_accuracy.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_attention.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_auc_1.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_auc_2.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_auc_3.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_auc_4.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_f1.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_precision.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_recall.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_satisfactory.png (100%) rename site/notebooks/{images => how_to/metrics}/log_metric_satisfactory_2.png (100%) rename site/notebooks/how_to/{ => metrics}/log_metrics_over_time.ipynb (97%) rename site/notebooks/how_to/{ => metrics}/run_unit_metrics.ipynb (99%) rename site/notebooks/how_to/{ => scoring}/assign_scores_complete_tutorial.ipynb (99%) rename site/notebooks/{images => how_to/tests/custom_tests}/external-data-custom-test.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/hyperparameters-custom-metric.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/image-in-custom-metric.png (100%) create mode 100644 site/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb rename site/notebooks/{images => how_to/tests/custom_tests}/insert-test-driven-block-custom.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/insert-test-driven-block-test-provider.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/insert-test-driven-block.png (100%) rename site/notebooks/{code_samples => how_to/tests}/custom_tests/integrate_external_test_providers.ipynb (99%) rename site/notebooks/{images => how_to/tests/custom_tests}/multiple-tables-plots-custom-metric.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/parameterized-custom-metric.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/pearson-correlation-matrix-test-output.png (100%) rename site/notebooks/{images => how_to/tests/custom_tests}/pearson-correlation-matrix.png (100%) rename site/notebooks/how_to/{ => tests/explore_tests}/explore_test_suites.ipynb (99%) rename site/notebooks/how_to/{ => tests/explore_tests}/explore_test_suites_output.png (100%) rename site/notebooks/how_to/{ => tests/explore_tests}/explore_tests.ipynb (99%) rename site/notebooks/how_to/{ => tests/explore_tests}/test_suite_output1.png (100%) rename site/notebooks/how_to/{ => tests/explore_tests}/test_suite_output2.png (100%) rename site/notebooks/how_to/{ => tests}/run_tests/1_run_dataset_based_tests.ipynb (97%) rename site/notebooks/how_to/{ => tests}/run_tests/2_run_comparison_tests.ipynb (97%) rename site/notebooks/how_to/{ => tests/run_tests/configure_tests}/customize_test_result_descriptions.ipynb (99%) rename site/notebooks/how_to/{ => tests/run_tests/configure_tests}/enable_pii_detection.ipynb (99%) rename site/notebooks/how_to/{ => tests/run_tests/configure_tests}/filter_input_columns.ipynb (100%) rename site/notebooks/how_to/{ => tests/run_tests/configure_tests}/run_tests_that_require_multiple_datasets.ipynb (98%) rename site/notebooks/how_to/{ => tests/run_tests/configure_tests}/understand_utilize_rawdata.ipynb (100%) rename site/notebooks/how_to/{ => tests/run_tests/documentation_tests}/document_multiple_results_for_the_same_test.ipynb (99%) rename site/notebooks/how_to/{ => tests/run_tests/documentation_tests}/run_documentation_sections.ipynb (98%) rename site/notebooks/how_to/{ => tests/run_tests/documentation_tests}/run_documentation_tests_with_config.ipynb (99%) delete mode 100644 site/notebooks/how_to/use_dataset_model_objects.ipynb delete mode 100644 site/notebooks/images/composite-metric-in-template-preview.png delete mode 100644 site/notebooks/images/high-pearson-correlation-block.png delete mode 100644 site/notebooks/images/insert-test-driven-block-correlations.png delete mode 100644 site/notebooks/images/insert-test-driven-block-custom-class-imbalance.jpg delete mode 100644 site/notebooks/images/insert-test-driven-block-custom-confusion-matrix.png delete mode 100644 site/notebooks/images/my_tests_directory.png delete mode 100644 site/notebooks/images/selecting-composite-metric.png delete mode 100644 site/notebooks/images/selecting-high-pearson-correlation-test.png create mode 100644 site/notebooks/use_cases/agents/agentic_ai_template.yaml rename site/notebooks/{code_samples => use_cases}/agents/banking_test_dataset.py (53%) rename site/notebooks/{code_samples => use_cases}/agents/banking_tools.py (99%) create mode 100644 site/notebooks/use_cases/agents/document_agentic_ai.ipynb rename site/notebooks/{code_samples => use_cases}/agents/utils.py (100%) create mode 100644 site/notebooks/use_cases/capital_markets/capital_markets_template.yaml rename site/notebooks/{code_samples => use_cases}/capital_markets/quickstart_option_pricing_models.ipynb (99%) create mode 100644 site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models_quantlib.ipynb rename site/notebooks/{code_samples => use_cases}/code_explainer/customer_churn_full_suite.py (98%) create mode 100644 site/notebooks/use_cases/code_explainer/model_source_code_documentation_template.yaml create mode 100644 site/notebooks/use_cases/code_explainer/quickstart_code_explainer_demo.ipynb rename site/notebooks/{code_samples => use_cases}/credit_risk/CreditRiskData.xlsx (100%) rename site/notebooks/{code_samples => use_cases}/credit_risk/application_scorecard_executive.ipynb (98%) rename site/notebooks/{code_samples => use_cases}/credit_risk/application_scorecard_full_suite.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/credit_risk/application_scorecard_with_bias.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/credit_risk/application_scorecard_with_ml.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/credit_risk/custom_tests/ScoreBandDiscriminationMetrics.py (100%) rename site/notebooks/{code_samples => use_cases}/credit_risk/document_excel_application_scorecard.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/model_validation/validate_application_scorecard.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/model_validation/xgb_model_champion.pkl (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/bbc_text_cls.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/bbc_text_cls_reference.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/cnn_dailymail_100_with_predictions.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/cnn_dailymail_500_with_predictions.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/sentiments.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/datasets/sentiments_with_predictions.csv (100%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/foundation_models_integration_demo.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/foundation_models_summarization_demo.ipynb (100%) create mode 100644 site/notebooks/use_cases/nlp_and_llm/gen_ai_rag_template.yaml rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/hugging_face_integration_demo.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/hugging_face_summarization_demo.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/llm_summarization_demo.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/nlp_and_llm/prompt_validation_demo.ipynb (98%) create mode 100644 site/notebooks/use_cases/nlp_and_llm/rag_benchmark_demo.ipynb create mode 100644 site/notebooks/use_cases/nlp_and_llm/rag_documentation_demo.ipynb rename site/notebooks/{code_samples => use_cases}/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/ongoing_monitoring/xgboost_model.model (100%) rename site/notebooks/{code_samples => use_cases}/regression/quickstart_regression_full_suite.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/time_series/quickstart_time_series_full_suite.ipynb (99%) rename site/notebooks/{code_samples => use_cases}/time_series/quickstart_time_series_high_code.ipynb (99%) create mode 100644 site/scripts/developer-sidebar/update_how_tos.py create mode 100644 site/scripts/developer-sidebar/update_use_cases.py delete mode 100644 site/scripts/update_code_samples_sidebar.py diff --git a/site/Makefile b/site/Makefile index d944db545f..150bda6156 100644 --- a/site/Makefile +++ b/site/Makefile @@ -326,8 +326,10 @@ notebooks: @rm -rf $(DEST_DIR_NB)/templates @echo "Copying _metadata.yml into notebooks/ ..." @cp developer/_metadata.yml $(DEST_DIR_NB)/_metadata.yml - @echo "Updating developer/_sidebar.yaml with code_samples directories ..." - @python scripts/update_code_samples_sidebar.py + @echo "Updating developer/_sidebar.yaml with use_cases directories ..." + @python scripts/developer-sidebar/update_use_cases.py + @echo "Updating developer/_sidebar.yaml with how-to directories ..." + @python scripts/developer-sidebar/update_how_tos.py @echo "Zip up notebooks.zip ..." @zip -r notebooks.zip $(DEST_DIR_NB) > /dev/null 2>&1 diff --git a/site/about/overview-model-documentation.qmd b/site/about/overview-model-documentation.qmd index 17c56cb014..a57598481b 100644 --- a/site/about/overview-model-documentation.qmd +++ b/site/about/overview-model-documentation.qmd @@ -150,8 +150,8 @@ How the {{< var validmind.developer >}} works: [^3]: [Customize document templates](/guide/templates/customize-document-templates.qmd) -[^4]: [Implement custom tests](/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb) +[^4]: [Implement custom tests](/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb) -[^5]: [Integrate external test providers](/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb) +[^5]: [Integrate external test providers](/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb) diff --git a/site/about/use-cases/e-23.qmd b/site/about/use-cases/e-23.qmd index 93e06c3c63..a288345a37 100644 --- a/site/about/use-cases/e-23.qmd +++ b/site/about/use-cases/e-23.qmd @@ -139,7 +139,7 @@ Use this checklist to track your E-23 implementation progress: [^3]: [Working with document templates](/guide/templates/working-with-document-templates.qmd) -[^4]: [Run tests and test suites](/developer/model-testing/testing-overview.qmd) +[^4]: [Run tests and test suites](/developer/how-to/testing-overview.qmd) [^5]: [Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd) @@ -149,7 +149,7 @@ Use this checklist to track your E-23 implementation progress: [^8]: [Ongoing monitoring](/guide/monitoring/ongoing-monitoring.qmd) -[^9]: [Run tests and test suites](/developer/model-testing/testing-overview.qmd) +[^9]: [Run tests and test suites](/developer/how-to/testing-overview.qmd) [^10]: [Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd) diff --git a/site/about/use-cases/eu-ai-act.qmd b/site/about/use-cases/eu-ai-act.qmd index af5607cda2..2f6504ffa5 100644 --- a/site/about/use-cases/eu-ai-act.qmd +++ b/site/about/use-cases/eu-ai-act.qmd @@ -233,11 +233,11 @@ Integrate all components into a complete compliance workflow addressing Articles [^4]: [Install and initialize the {{< var validmind.developer >}}](/developer/model-documentation/install-and-initialize-validmind-library.qmd) [^5]: -[Run tests and test suites](/developer/model-testing/testing-overview.qmd) +[Run tests and test suites](/developer/how-to/testing-overview.qmd) [^6]: [Work with document templates](/guide/templates/working-with-document-templates.qmd) -[^7]: [Test descriptions](/developer/model-testing/test-descriptions.qmd) +[^7]: [Test descriptions](/developer/test-descriptions.qmd) [^8]: [Work with content blocks](/guide/model-documentation/work-with-content-blocks.qmd) diff --git a/site/about/use-cases/sr-11-7.qmd b/site/about/use-cases/sr-11-7.qmd index c6f971561b..a5462a32c0 100644 --- a/site/about/use-cases/sr-11-7.qmd +++ b/site/about/use-cases/sr-11-7.qmd @@ -211,7 +211,7 @@ Establish governance framework aligned to SR 11-7. [^5]: [Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd) -[^6]: [Run tests and test suites](/developer/model-testing/testing-overview.qmd) +[^6]: [Run tests and test suites](/developer/how-to/testing-overview.qmd) [^7]: [Preparing validation reports](/guide/model-validation/preparing-validation-reports.qmd) diff --git a/site/developer/_sidebar.yaml b/site/developer/_sidebar.yaml index 3b1680ea69..1215f1cf41 100644 --- a/site/developer/_sidebar.yaml +++ b/site/developer/_sidebar.yaml @@ -19,65 +19,92 @@ website: file: developer/model-documentation/install-and-initialize-validmind-library.qmd - developer/model-documentation/store-credentials-in-env-file.qmd - text: "---" - - text: "Model Development" + - text: "End-to-End Tutorials" # USING THE VARIABLE IN THE LINK TEXT MESSES UP THE MOBILE VIEW & BREADCRUMB - - text: "1 — Set up ValidMind Library" - file: notebooks/tutorials/model_development/1-set_up_validmind.ipynb - - text: "2 — Start model development process" - file: notebooks/tutorials/model_development/2-start_development_process.ipynb - - text: "3 — Integrate custom tests" - file: notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb - - text: "4 — Finalize testing & documentation" - file: notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb - - text: "---" - - text: "Model Validation" - # USING THE VARIABLE IN THE LINK TEXT MESSES UP THE MOBILE VIEW & BREADCRUMB - - text: "1 — Set up ValidMind Library for validation" - file: notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb - - text: "2 — Start model validation process" - file: notebooks/tutorials/model_validation/2-start_validation_process.ipynb - - text: "3 — Developing a challenger model" - file: notebooks/tutorials/model_validation/3-developing_challenger_model.ipynb - - text: "4 — Finalize validation & reporting" - file: notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb + - section: "Model development" + contents: + - text: "1 — Set up ValidMind Library" + file: notebooks/tutorials/model_development/1-set_up_validmind.ipynb + - text: "2 — Start model development process" + file: notebooks/tutorials/model_development/2-start_development_process.ipynb + - text: "3 — Integrate custom tests" + file: notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb + - text: "4 — Finalize testing & documentation" + file: notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb + - section: "Model validation" + contents: + - text: "1 — Set up ValidMind Library for validation" + file: notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb + - text: "2 —\u00A0Start model validation process" + file: notebooks/tutorials/model_validation/2-start_validation_process.ipynb + - text: "3 — Developing a challenger model" + file: notebooks/tutorials/model_validation/3-developing_challenger_model.ipynb + - text: "4 — Finalize validation & reporting" + file: notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb - text: "---" - - text: "Model Testing" + - text: "How-To" - text: "Run tests & test suites" - file: developer/model-testing/testing-overview.qmd - contents: "notebooks/how_to/**" - - text: "Test descriptions" - file: developer/model-testing/test-descriptions.qmd - contents: tests/** - - developer/model-testing/test-sandbox.qmd + file: developer/how-to/testing-overview.qmd + contents: + - section: "Explore tests" + contents: + - notebooks/how_to/tests/explore_tests/explore_tests.ipynb + - notebooks/how_to/tests/explore_tests/explore_test_suites.ipynb + - developer/how-to/test-sandbox.qmd + - section: "Run tests" + contents: + - notebooks/how_to/tests/run_tests/1_run_dataset_based_tests.ipynb + - notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb + - section: "Configuring tests" + contents: "notebooks/how_to/tests/run_tests/configure_tests/*.ipynb" + - section: "Using tests in documentation" + contents: "notebooks/how_to/tests/run_tests/documentation_tests/*.ipynb" + - section: "Custom tests" + contents: + - notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb + - notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb + - text: "Use library features" + file: developer/how-to/feature-overview.qmd + contents: + - section: "Data and datasets" + contents: + - notebooks/how_to/data_and_datasets/use_dataset_model_objects.ipynb + - section: "Dataset inputs" + contents: "notebooks/how_to/data_and_datasets/dataset_inputs/**/*.ipynb" + - section: "Metrics" + contents: "notebooks/how_to/metrics/**/*.ipynb" + - section: "Scoring" + contents: "notebooks/how_to/scoring/**/*.ipynb" - text: "---" - text: "Notebooks" - text: "Code samples" file: developer/samples-jupyter-notebooks.qmd contents: - section: "Agents" - contents: "notebooks/code_samples/agents/**" + contents: "notebooks/use_cases/agents/**/*.ipynb" - section: "Capital markets" - contents: "notebooks/code_samples/capital_markets/**" + contents: "notebooks/use_cases/capital_markets/**/*.ipynb" - section: "Code explainer" - contents: "notebooks/code_samples/code_explainer/**" + contents: "notebooks/use_cases/code_explainer/**/*.ipynb" - section: "Credit risk" - contents: "notebooks/code_samples/credit_risk/**" - - section: "Custom tests" - contents: "notebooks/code_samples/custom_tests/**" + contents: "notebooks/use_cases/credit_risk/**/*.ipynb" - section: "Model validation" - contents: "notebooks/code_samples/model_validation/**" + contents: "notebooks/use_cases/model_validation/**/*.ipynb" - section: "NLP and LLM" - contents: "notebooks/code_samples/nlp_and_llm/**" + contents: "notebooks/use_cases/nlp_and_llm/**/*.ipynb" - section: "Ongoing monitoring" - contents: "notebooks/code_samples/ongoing_monitoring/**" + contents: "notebooks/use_cases/ongoing_monitoring/**/*.ipynb" - section: "Regression" - contents: "notebooks/code_samples/regression/**" + contents: "notebooks/use_cases/regression/**/*.ipynb" - section: "Time series" - contents: "notebooks/code_samples/time_series/**" + contents: "notebooks/use_cases/time_series/**/*.ipynb" - text: "---" - text: "Reference" + - text: "Test descriptions" + file: developer/test-descriptions.qmd + contents: tests/** - text: "{{< var validmind.api >}}" file: validmind/validmind.qmd - - reference/validmind-rest-api-vm.qmd - - + # USING THE VARIABLE IN THE LINK TEXT MESSES UP THE MOBILE VIEW & BREADCRUMB + - text: "ValidMind Public REST API" + file: reference/validmind-rest-api-vm.qmd diff --git a/site/developer/how-to/feature-overview.qmd b/site/developer/how-to/feature-overview.qmd new file mode 100644 index 0000000000..7f9536acfd --- /dev/null +++ b/site/developer/how-to/feature-overview.qmd @@ -0,0 +1,69 @@ +--- +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial +# USING THE VARIABLE MESSES UP THE SPACING ON THE PREVIEW TILE +title: "How to use {{< var validmind.developer >}} features" +date: last-modified +listing: + - id: tests + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../../notebooks/how_to/tests/**/*.ipynb" + - id: data-and-datasets + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../../notebooks/how_to/data_and_datasets/**/*.ipynb" + - id: metrics + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../../notebooks/how_to/metrics/**/*.ipynb" + - id: scoring + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../../notebooks/how_to/scoring/**/*.ipynb" +--- + +Browse our range of Jupyter Notebooks demonstrating how to use the core features of the {{< var validmind.developer >}}. Use these how-to notebooks to get familiar with the {{< var vm.developer >}}'s capabilities and apply them to your own use cases. + +## How-to by topic + +:::{.panel-tabset} + +## Testing + +:::{#tests} +::: + +## Data and datasets + +:::{#data-and-datasets} +::: + +## Metrics + +:::{#metrics} +::: + +## Scoring + +:::{#scoring} +::: + +::: diff --git a/site/developer/model-testing/test-sandbox.qmd b/site/developer/how-to/test-sandbox.qmd similarity index 94% rename from site/developer/model-testing/test-sandbox.qmd rename to site/developer/how-to/test-sandbox.qmd index 4e842d5db3..c9ea3a23c3 100644 --- a/site/developer/model-testing/test-sandbox.qmd +++ b/site/developer/how-to/test-sandbox.qmd @@ -6,6 +6,7 @@ title: "Test sandbox [beta]{.smallcaps}" date: last-modified aliases: - /guide/test-sandbox.html + - /developer/model-testing/test-sandbox.html --- - -{{< include /about/glossary/key_concepts/_key-concepts.qmd >}} - -::: - -## Getting started - -Start by running a pre-made test, then modify it, and finally create your own test: - -:::{#tests-beginner} -::: - -## Explore tests and test suites - -Next, find available tests and test suites using the {{< var vm.developer >}} or the interactive test sandbox: - -:::{#tests-explore} -::: - -::: {.grid} -::: {.g-col-8} -{{< video https://www.youtube.com/embed/5J7wysDIXgI?si=KPkFhx3O6NknijRf title='How do I add tests?' >}} -::: -::: - -## Intermediate - -Building on previous sections, add your own test provider, set up datasets, run tests on individual sections in your model documentation, and more: - -:::{#tests-intermediate} -::: - -## Advanced - -Need more? Try some of the advanced features provided by the {{< var vm.developer >}}: - -:::{#tests-advanced} -::: - -## When do I use tests and test suites? - -While you have the flexibility to decide when to use which {{< var vm.product >}} tests, we have identified a few typical scenarios with their own characteristics and needs: - -:::: {.flex .flex-wrap .justify-around} - -::: {.w-30-ns} - -#### Dataset testing - -To document and validate your dataset: - -- For generic tabular datasets: use the [`tabular_dataset`](/validmind/validmind/test_suites/tabular_datasets.qmd){target="_blank"} test suite. -- For time-series datasets: use the [`time_series_dataset`](/validmind/validmind/test_suites/time_series.qmd#timeseriesdataset){target="_blank"} test suite. - -::: - -::: {.w-30-ns} - -#### Model testing - -To document and validate your model: - -- For binary classification models: use the [`classifier`](/validmind/validmind/test_suites/classifier.qmd){target="_blank"} test suite. -- For time series models: use the [`timeseries`](/validmind/validmind/test_suites/time_series.qmd){target="_blank"} test suite. - -::: - -::: {.w-30-ns} - -#### End-to-end testing - -To document a binary classification model and the relevant dataset end-to-end: - -Use the [`classifier_full_suite`](/validmind/validmind/test_suites/classifier.qmd#classifierfullsuite){target="_blank"} test suite. - -::: - -:::: - -## Can I use my own tests? - -Absolutely! {{< var vm.product >}} supports custom tests that you develop yourself or that are provided by third-party test libraries, also referred to as _test providers_. We provide instructions with code examples that you can adapt: - -:::{#tests-custom} -::: - -## {{< var validmind.api >}} reference - -[{{< var validmind.developer >}} API Reference](/validmind/validmind.qmd){target="_blank" .button .button-green} diff --git a/site/developer/samples-jupyter-notebooks.qmd b/site/developer/samples-jupyter-notebooks.qmd index 7040dcc9ac..11a5aa88c7 100644 --- a/site/developer/samples-jupyter-notebooks.qmd +++ b/site/developer/samples-jupyter-notebooks.qmd @@ -5,20 +5,84 @@ title: "Code samples" date: last-modified listing: - type: grid - image-placeholder: "jupyter-logo-rectangle.svg" - max-description-length: 250 - image-height: "100%" - fields: [title, description, reading-time] - contents: - - "../notebooks/code_samples/*/*.ipynb" - - "../notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb" + - id: agents + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/agents/*.ipynb" + - id: capital-markets + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/capital_markets/*.ipynb" + - id: code-explainer + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/code_explainer/*.ipynb" + - id: credit-risk + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/credit_risk/*.ipynb" + - id: model-validation + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/model_validation/*.ipynb" + - id: nlp-and-llm + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/nlp_and_llm/*.ipynb" + - id: ongoing-monitoring + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/ongoing_monitoring/*.ipynb" + - id: regression + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/regression/*.ipynb" + - id: time-series + type: grid + grid-columns: 2 + image-placeholder: "jupyter-logo-rectangle.svg" + max-description-length: 350 + image-height: "100%" + fields: [title, description, reading-time] + contents: "../notebooks/use_cases/time_series/*.ipynb" fig-cap-location: top aliases: - /guide/samples-jupyter-notebooks.html --- -Our Jupyter Notebook code samples showcase the capabilities and features of the {{< var validmind.developer >}}, while also providing you with useful examples that you can build on and adapt for your own use cases. +Our Jupyter Notebook code samples showcase the capabilities and features of the {{< var validmind.developer >}}, while also providing you with useful examples that you can build on and adapt for your own use cases. :::: {.flex .flex-wrap .justify-around} @@ -43,31 +107,53 @@ Our Jupyter Notebook code samples showcase the capabilities and features of the :::: -## Quickstarts +## By use case -Learn how to develop or validate a simple customer churn model with {{< var vm.product >}}. The easiest way to try the quickstarts are on JupyterHub: +:::{.panel-tabset} -:::: {.flex .flex-wrap .justify-around} +## Agents -::: {.w-50-ns} +:::{#agents} +::: -::: {.tc} -[{{< fa code >}} Quickstart for model documentation]({{< var url.jupyterhub >}}/hub/user-redirect/lab/tree/quickstart/quickstart_model_documentation.ipynb){.button target="_blank"} +## Capital markets +:::{#capital-markets} ::: +## Code explainer + +:::{#code-explainer} ::: -::: {.w-40-ns} +## Credit risk + +:::{#credit-risk} +::: -::: {.tc} -[{{< fa code >}} Quickstart for model validation]({{< var url.jupyterhub >}}/hub/user-redirect/lab/tree/quickstart/quickstart_model_validation.ipynb){.button target="_blank"} +## Model validation +:::{#model-validation} ::: +## NLP and LLM + +:::{#nlp-and-llm} ::: -:::: +## Ongoing monitoring +:::{#ongoing-monitoring} +::: -## Code samples for your use case +## Regression + +:::{#regression} +::: + +## Time series + +:::{#time-series} +::: + +::: diff --git a/site/developer/supported-models.qmd b/site/developer/supported-models.qmd index 76e961af43..930db0533f 100644 --- a/site/developer/supported-models.qmd +++ b/site/developer/supported-models.qmd @@ -10,13 +10,15 @@ aliases: listing: - id: next-models type: grid + grid-columns: 2 max-description-length: 250 sort: false fields: [title, description] contents: - - ../model-testing/testing-overview.qmd - - ../model-testing/test-descriptions.qmd - - ../samples-jupyter-notebooks.qmd + - /how-to/testing-overview.qmd + - test-descriptions.qmd + - /how-to/feature-overview.qmd + - samples-jupyter-notebooks.qmd --- The {{< var validmind.developer >}} provides out-of-the-box support for testing and documentation for an array of model types and modeling packages. @@ -154,5 +156,5 @@ Analyzes data points collected or sequenced over time. [^1]: - - [Implement custom tests](/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb) - - [Integrate external test providers](/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb) \ No newline at end of file + - [Implement custom tests](/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb) + - [Integrate external test providers](/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb) \ No newline at end of file diff --git a/site/developer/model-testing/test-descriptions.qmd b/site/developer/test-descriptions.qmd similarity index 83% rename from site/developer/model-testing/test-descriptions.qmd rename to site/developer/test-descriptions.qmd index 4a248ee33d..fc82f7d8bc 100644 --- a/site/developer/model-testing/test-descriptions.qmd +++ b/site/developer/test-descriptions.qmd @@ -6,27 +6,28 @@ title: "Test descriptions" date: last-modified aliases: - /guide/test-descriptions.html + - /developer/model-testing/test-descriptions.html listing: - id: data-validation - contents: "../../tests/data_validation/*.md" + contents: "../tests/data_validation/*.md" type: grid max-description-length: 250 page-size: 150 fields: [title, description] - id: model-validation type: grid - contents: "../../tests/model_validation/*.md" + contents: "../tests/model_validation/*.md" max-description-length: 250 page-size: 150 fields: [title, description] - id: prompt-validation - contents: "../../tests/prompt_validation/*.md" + contents: "../tests/prompt_validation/*.md" type: grid max-description-length: 250 page-size: 150 fields: [title, description] - id: ongoing-monitoring - contents: "../../tests/ongoing_monitoring/*.md" + contents: "../tests/ongoing_monitoring/*.md" type: grid max-description-length: 250 page-size: 150 @@ -36,7 +37,7 @@ listing: Tests that are available as part of the {{< var validmind.developer >}}, grouped by type of validation or monitoring test. ::: {.callout} -## {{< fa flask >}} [Try the test sandbox [beta]{.smallcaps}](test-sandbox.qmd) +## {{< fa flask >}} [Try the test sandbox [beta]{.smallcaps}](how-to/test-sandbox.qmd) Explore our interactive sandbox to see what tests are available in the {{< var validmind.developer >}}. ::: @@ -63,4 +64,4 @@ Explore our interactive sandbox to see what tests are available in the {{< var v :::{#ongoing-monitoring} ::: -::: \ No newline at end of file +::: diff --git a/site/developer/validmind-library.qmd b/site/developer/validmind-library.qmd index 26d5a0c69b..8970eb565f 100644 --- a/site/developer/validmind-library.qmd +++ b/site/developer/validmind-library.qmd @@ -55,25 +55,25 @@ listing: - path: ../notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb title: "4 — Finalize validation and reporting" description: "Wrap up by learning how to include custom tests and verifying that all tests conducted during model development were run and reported accurately. By the end of this series, you will have a validation report complete with artifacts ready for review." - - id: run-tests + - id: library-features grid-columns: 2 type: grid max-description-length: 250 sort: false fields: [title, description] - contents: - - ../notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb -# - ../notebooks/archive/configure_test_parameters.ipynb - - ../notebooks/code_samples/custom_tests/implement_custom_tests.ipynb + contents: + - how-to/testing-overview.qmd + - how-to/feature-overview.qmd - id: code-samples + grid-columns: 3 type: grid - max-description-length: 250 + max-description-length: 350 sort: false fields: [title, description] - contents: - - ../notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb - - ../notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb - - ../notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb + contents: + - ../notebooks/use_cases/agents/document_agentic_ai.ipynb + - ../notebooks/use_cases/credit_risk/document_excel_application_scorecard.ipynb + - ../notebooks/use_cases/model_validation/validate_application_scorecard.ipynb - id: library-documentation type: grid grid-columns: 2 @@ -110,14 +110,18 @@ The {{< var validmind.developer >}} is designed to be model agnostic. If your mo ## Quickstart -After you [**sign up**](/guide/access/accessing-validmind.qmd) for {{< var vm.product >}} to get access, try our quickstarts for model documentation or validation: +After you [**sign up**](/guide/access/accessing-validmind.qmd) for {{< var vm.product >}} to get access, try our Jupyter Notebook quickstarts for model documentation or validation: :::{#library-quickstart} ::: +## End-to-end tutorials + +Learn how to use the {{< var validmind.developer >}} with our high-level Jupyter Notebook courses covering usage of ValidMind for specific roles or concepts. + -## {{< var vm.product >}} for model development +#### {{< var vm.product >}} for model development Learn how to use ValidMind for your end-to-end model documentation process based on common model development scenarios with our *ValidMind for model development* series of four introductory notebooks: @@ -126,7 +130,7 @@ Learn how to use ValidMind for your end-to-end model documentation process based -## {{< var vm.product >}} for model validation +#### {{< var vm.product >}} for model validation Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our *ValidMind for model validation* series of four introductory notebooks: @@ -134,66 +138,52 @@ Learn how to use ValidMind for your end-to-end model validation process based on ::: -## Learn how to run tests +## Learn how to use the {{< var validmind.developer >}} :::: {.flex .flex-wrap .justify-around} -::: {.w-70-ns} -The {{< var validmind.developer >}} provides many built-in tests and test suites which make it easy for developers to automate their model documentation. Start by running a pre-made test, then modify it, and finally create your own test: - -::: - -::: {.w-30-ns .tc} -[Run tests & test suites](model-testing/testing-overview.qmd){.button .button-green} +Learn how to use the comprehensive out-of-the-box tests and test suites, and other features in the {{< var validmind.developer >}} that make it easy for you to automate building, documenting, validating your models and more. +:::{#library-features} ::: :::: -:::{#run-tests} -::: - -## Try the code samples +## Try code samples by use case :::: {.flex .flex-wrap .justify-around} ::: {.w-70-ns} -Our code samples showcase the capabilities of the {{< var validmind.developer >}}. Examples that you can build on and adapt for your own use cases include: - +Try our Jupyter Notebook code samples that showcase the capabilities of the {{< var validmind.developer >}} and cover a variety of sample use cases. ::: -::: {.w-30-ns .tc} -[All code samples](samples-jupyter-notebooks.qmd){.button .button-green} +::: {.w-30-ns .tr} +[Code samples by use case](samples-jupyter-notebooks.qmd){.button .button-green} ::: :::: +Examples that you can build on and adapt for your own usage include: + :::{#code-samples} ::: ## Work with model documentation -:::: {.flex .flex-wrap .justify-around} - -::: {.w-60-ns .pr3} -After you have tried out the {{< var validmind.developer >}}, continue working with your model documentation in the {{< var validmind.platform >}}: - -::: - -::: {.w-40-ns .tc} -[Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd){.button .button-green} - -::: - -:::: +After you have tried out the {{< var validmind.developer >}}, continue working with your model documentation in the {{< var validmind.platform >}}:[^3] :::{#library-documentation} ::: +## {{< var validmind.api >}} reference + +[{{< var validmind.developer >}} API Reference](/validmind/validmind.qmd){target="_blank" .button .button-green} [^1]: [{{< var vm.product >}} for model development](#development) -[^2]: [{{< var vm.product >}} for model validation](#validation) \ No newline at end of file +[^2]: [{{< var vm.product >}} for model validation](#validation) + +[^3]: [Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd) \ No newline at end of file diff --git a/site/faq/_faq-images.qmd b/site/faq/_faq-images.qmd index 22b479f7b5..2b640cdd5f 100644 --- a/site/faq/_faq-images.qmd +++ b/site/faq/_faq-images.qmd @@ -4,7 +4,7 @@ SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial --> ## Do you support including images in model documentation? -Yes, as long as you can produce the image with Python or open the image from a file, you can include it in your documentation with {{< var vm.product >}}:^[[Implement custom tests](/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb#custom-test-images)] +Yes, as long as you can produce the image with Python or open the image from a file, you can include it in your documentation with {{< var vm.product >}}:^[[Implement custom tests](/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb#custom-test-images)] - If you want to log an image as a test result, you can do so by passing the path to the image as a parameter to the custom test and then opening the file in the test function. - If you are using a plotting library that isn’t directly supported by {{< var vm.product >}}, you can still return the image directly as a bytes-like object. \ No newline at end of file diff --git a/site/faq/_faq-synthetic-datasets.qmd b/site/faq/_faq-synthetic-datasets.qmd index 3851aa7635..e406354952 100644 --- a/site/faq/_faq-synthetic-datasets.qmd +++ b/site/faq/_faq-synthetic-datasets.qmd @@ -4,5 +4,5 @@ SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial --> ## Does {{< var vm.product >}} support using synthetic datasets? -- The {{< var validmind.developer >}} supports you bringing your own datasets, including synthetic datasets, for testing and benchmarking purposes such as for fair lending and bias testing.^[[Document a Credit Risk Model](/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb)] +- The {{< var validmind.developer >}} supports you bringing your own datasets, including synthetic datasets, for testing and benchmarking purposes such as for fair lending and bias testing.^[[Document a Credit Risk Model](/notebooks/use_cases/credit_risk/application_scorecard_with_bias.ipynb)] - If you are unable to share your real-world data with us, {{< var vm.product >}} is happy to work with you to generate custom synthetic datasets based on characteristics of your data, or provide scripts to assist with synthetic dataset generation if details cannot be shared. \ No newline at end of file diff --git a/site/faq/faq-integrations.qmd b/site/faq/faq-integrations.qmd index b7c3b0c3bf..8c3d7ac9ea 100644 --- a/site/faq/faq-integrations.qmd +++ b/site/faq/faq-integrations.qmd @@ -90,6 +90,6 @@ We will be implementing connector interfaces allowing extraction of relationship [^5]: [Do you support including images in model documentation?](#images) -[^6]: [Load dataset predictions](/notebooks/how_to/load_datasets_predictions.ipynb) +[^6]: [Load dataset predictions](/notebooks/how_to/data_and_datasets/dataset_inputs/load_datasets_predictions.ipynb) [^7]: [Do you include explainability-related testing and documentation?](#explanability) \ No newline at end of file diff --git a/site/faq/faq-testing.qmd b/site/faq/faq-testing.qmd index 6024603922..42ddf51e84 100644 --- a/site/faq/faq-testing.qmd +++ b/site/faq/faq-testing.qmd @@ -14,8 +14,8 @@ listing: sort: false fields: [title, description] contents: - - ../developer/model-testing/testing-overview.qmd - - ../developer/model-testing/test-descriptions.qmd + - ../developer/how-to/testing-overview.qmd + - ../developer/test-descriptions.qmd - ../guide/monitoring/ongoing-monitoring.qmd categories: ["testing", "model documentation", "customization", "custom data", "explainability", "ongoing monitoring", "validmind library"] --- @@ -59,7 +59,7 @@ To log tests as a developer with the {{< var validmind.developer >}}: ::: {.callout} ## Want to learn how to use {{< var vm.product >}} as a developer? -Check our our introductory series — [**{{< var vm.product >}} for model development**](/developer/validmind-library.qmd#for-model-development) +Check out our introductory series — [**{{< var vm.product >}} for model development**](/developer/validmind-library.qmd#for-model-development) ::: ## How do I log tests as a validator? @@ -74,7 +74,7 @@ To log tests as a validator with the {{< var validmind.developer >}}: ::: {.callout} ## Want to learn how to use {{< var vm.product >}} as a validator? -Check our our introductory series — [**{{< var vm.product >}} for model validation**](/developer/validmind-library.qmd#for-model-validation) +Check out our introductory series — [**{{< var vm.product >}} for model validation**](/developer/validmind-library.qmd#for-model-validation) ::: {{< include _faq-explainability.qmd >}} @@ -95,17 +95,17 @@ Check our our introductory series — [**{{< var vm.product >}} for model valida [^2]: [Accessing {{< var vm.product >}}](/guide/access/accessing-validmind.qmd) -[^3]: [When do I use tests and test suites?](/developer/model-testing/testing-overview.qmd#when-do-i-use-tests-and-test-suites) +[^3]: [When do I use tests and test suites?](/developer/how-to/testing-overview.qmd#when-do-i-use-tests-and-test-suites) [^4]: [`run_documentation_tests()`](/validmind/validmind.qmd#run_documentation_tests) [^5]: [`ClassImbalance()`](/validmind/validmind/tests/data_validation/ClassImbalance.qmd) -[^6]: [Can I use my own tests?](/developer/model-testing/testing-overview.qmd#can-i-use-my-own-tests) +[^6]: [Can I use my own tests?](/developer/how-to/testing-overview.qmd#can-i-use-my-own-tests) -[^7]: [Understand and utilize `RawData` in {{< var vm.product >}} tests](/notebooks/how_to/understand_utilize_rawdata.ipynb) +[^7]: [Understand and utilize `RawData` in {{< var vm.product >}} tests](/notebooks/how_to/tests/run_tests/configure_tests/understand_utilize_rawdata.ipynb) -[^8]: [Customize test result descriptions](/notebooks/how_to/customize_test_result_descriptions.ipynb) +[^8]: [Customize test result descriptions](/notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.ipynb) [^9]: [Work with test results](/guide/model-documentation/work-with-test-results.qmd) diff --git a/site/guide/access/accessing-validmind.qmd b/site/guide/access/accessing-validmind.qmd index 6762377cc3..777ad4fff4 100644 --- a/site/guide/access/accessing-validmind.qmd +++ b/site/guide/access/accessing-validmind.qmd @@ -9,7 +9,7 @@ listing: - id: guides-access type: grid max-description-length: 250 - grid-columns: 2 + grid-columns: 3 sort: false fields: [title, description] contents: diff --git a/site/guide/model-documentation/work-with-content-blocks.qmd b/site/guide/model-documentation/work-with-content-blocks.qmd index 8a29bcda8c..999828aa14 100644 --- a/site/guide/model-documentation/work-with-content-blocks.qmd +++ b/site/guide/model-documentation/work-with-content-blocks.qmd @@ -168,7 +168,7 @@ Test-driven or metric over time blocks can be re-added later on but **text block [^9]: [Collaborate with others](/guide/model-documentation/collaborate-with-others.qmd) -[^10]: [Run tests and test suites](/developer/model-testing/testing-overview.qmd) +[^10]: [Run tests and test suites](/developer/how-to/testing-overview.qmd) [^11]: [View model activity](/guide/model-inventory/view-model-activity.qmd) diff --git a/site/guide/monitoring/enable-monitoring.qmd b/site/guide/monitoring/enable-monitoring.qmd index 9ff0acb8f0..6bfbfa212f 100644 --- a/site/guide/monitoring/enable-monitoring.qmd +++ b/site/guide/monitoring/enable-monitoring.qmd @@ -14,7 +14,7 @@ Enable monitoring with two steps: 2. [Select a monitoring template](#select-a-monitoring-template) ::: {.callout title="To try out monitoring, check out the code sample for ongoing monitoring of models."} -[Quickstart for ongoing monitoring of models with {{< var vm.product >}} {{< fa hand-point-right >}}](/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) +[Quickstart for ongoing monitoring of models with {{< var vm.product >}} {{< fa hand-point-right >}}](/notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) ::: ::: {.attn} diff --git a/site/guide/monitoring/ongoing-monitoring.qmd b/site/guide/monitoring/ongoing-monitoring.qmd index 9242c53664..cb351f3d83 100644 --- a/site/guide/monitoring/ongoing-monitoring.qmd +++ b/site/guide/monitoring/ongoing-monitoring.qmd @@ -23,9 +23,9 @@ listing: sort: false fields: [title, description] contents: - - ../../notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb - - ../../notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb - # - ../../notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb + - ../../notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb + - ../../notebooks/use_cases/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb + # - ../../notebooks/use_cases/credit_risk/application_scorecard_full_suite.ipynb - id: ongoing-monitoring-tests contents: "../../tests/ongoing_monitoring/*.md" type: grid @@ -155,4 +155,4 @@ To try out monitoring, check out the code sample for ongoing monitoring of model [^2]: [Working with model documentation](/guide/model-documentation/working-with-model-documentation.qmd) -[^3]: [Quickstart for ongoing monitoring of models with {{< var vm.product >}}](../../notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) +[^3]: [Quickstart for ongoing monitoring of models with {{< var vm.product >}}](../../notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) diff --git a/site/guide/monitoring/review-monitoring-results.qmd b/site/guide/monitoring/review-monitoring-results.qmd index 6a83688a72..3ba3018b45 100644 --- a/site/guide/monitoring/review-monitoring-results.qmd +++ b/site/guide/monitoring/review-monitoring-results.qmd @@ -13,7 +13,7 @@ To ensure your model continues to perform as expected, it's important to regular ::: {.callout title="To try out monitoring, check out the code sample for ongoing monitoring of models."} -[Quickstart for ongoing monitoring of models with {{< var vm.product >}} {{< fa hand-point-right >}}](/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) +[Quickstart for ongoing monitoring of models with {{< var vm.product >}} {{< fa hand-point-right >}}](/notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb) ::: ::: {.attn} diff --git a/site/guide/monitoring/work-with-metrics-over-time.qmd b/site/guide/monitoring/work-with-metrics-over-time.qmd index a94057d6ba..e474ab0643 100644 --- a/site/guide/monitoring/work-with-metrics-over-time.qmd +++ b/site/guide/monitoring/work-with-metrics-over-time.qmd @@ -16,7 +16,7 @@ Metrics over time refers to the continued monitoring of a model's performance on ::: {.column-margin} ::: {.callout} -## **[Log metrics over time {{< fa hand-point-right >}}](/notebooks/how_to/log_metrics_over_time.ipynb)** +## **[Log metrics over time {{< fa hand-point-right >}}](/notebooks/how_to/metrics/log_metrics_over_time.ipynb)** Learn how to log metrics over time, set thresholds, and analyze model performance trends with our Jupyter Notebook sample. ::: @@ -94,7 +94,7 @@ After you have added metrics over time to your document, you can view the follow [^1]: [Register models in the inventory](/guide/model-inventory/register-models-in-inventory.qmd) -[^2]: [Log metrics over time](/notebooks/how_to/log_metrics_over_time.ipynb) +[^2]: [Log metrics over time](/notebooks/how_to/metrics/log_metrics_over_time.ipynb) [^3]: [Manage permissions](/guide/configuration/manage-permissions.qmd) diff --git a/site/notebooks.zip b/site/notebooks.zip index 2ec23a970da72c452217749fd2e0d7a680a0b198..683df4e4530cb08d6bf8f23ffa28829dbcb11d20 100644 GIT binary patch delta 523077 zcmYhi1CZ`Y_x3%u_Sm*Pdu-b__Sn`vwr$(CZQC~X*fZZbPyNqZUn;p$-M>|5BwXC3jO~wfk6Ln=vc$6KcKyaSLUEgF8>?ow~+;?-ha7zv;S!^e*O>4=~;rx|35&k zZwa~x@;~rtXbGzNABfd3vIPA{to{}2f5is0SHs8_6zcYW!N&hwh3Ic%xEbl<~2=4^nxX(6_S;&z>Qu@~Q_Kup-dIm6dv{)g)+kA?l7bO-1 zZS-V5{`HZY$hRLr@NvYPD~p!Hmoaw5C1aN;ZsS`j?CHLCJ3cs=oti&IruTJ^ z@iA5;icTqx-fI^g%e=CWlO04X#lj>hj24OGN$!+QUO@5vyb3@eP@#_~9~-S$`IbRs zNuO@hUp|B!O>kHFx|A8G{dM@Lc2rIorG@DXQ4%CIAeRGh{kzJHvPcV`#AHMn4Fir3 zc5V{TE0c$ue!di@ogE7v&qTdA5ylE)`dN|4BRxnwrp%a8)Q(+p-h3Xh2$nc6Cc?$y z@m3Ih^CF{yLSaeGGO6mbGcFbtYSS7e$&~kgu<_fU`KZ@Sw1} zmaqT$S7L;E68z^1?vS<3^=Cyg>HDulnm0NPE6&88XHqfv3a!K;D>t&X{>>twdz&H1 zi#clk43yfIu9^gJvvo;}jUD=ajLF$`nbHvIjuj!; zKAQysG|VgTt=qu909t#$f=S}X6);9zfA;4mPt9P`JQ_`EKG%o)EB73fVm7K8z*a=QlnKr z`aJ>M5*9Ajdz}}cmr)U=;mMPMmOYleF(z>f<=s>HQjN08p_K7Y-oxD%63gS++b22>y^&W@`zN2DBmqBrL*vhdr9?dy z^L5NAub{zRZ((5XB))Zqo@M@*yvp}`>w6l+ zHw>HyjB=}t(QuPI3MgGnrf%SEw-K})itOSNK<&lN4m$(d8@_c&kJ+y*Sn|=i;=V$? z5XSd91*O9=yYGpf@-Q!b*)6?ez6fBr=ak(9QK^1OGOuw{n~3&YN!%uMDSuPKlsSsC zrpp<$ay;Q4jL!;%@6(eV!^}Jj%d%@{8R^%Dnkd0epkR9_%B^ZS2+_ma=DND>-@id)i#2n-%DaJs2MGa5|kR=*q znK9f~rFr2R0;k{%YG;+>p!J^3@+P}IpXYr$m!{LrpLgl({TyU2A|5P{ucyD>pz`F^q6XRiKG7#<Ni9=nCltU0Hn9{%FQ4f(Kr`~Gxk1pY>Ih!cKMdK9m5$3X!^k{9Mm`;q z7H{#Bg>C@elph?Kge`@GYuW^k$&{HXfi2FL%rM9@VE|QblnaTJ9;`eXWfZ*_XnDrC z>VS$8BZOF8xWBTiB#9*1wnZv~Sed60&I?L8#e-t;q=TK`J`*7JE0c5qRG6n9KEV$J zc_caj`wY5lM*(Gbezr$;F?C-)cve!I?{usbIh)kUdSyo=B7!Mb7S0Q~paF$5r%>+5(WLRboUKiN`(fvF$oJal@IY z$#HlfCT*Kdg~S0%Z4IAy9uetW6jsj;Z{yCA052|8ZjooMmROWrQ}AoE>D;BtiC=M4 zd7HN`zG&TIsGWWz?dFvx5_fSKf47%7E|(4_2M8$5mLH%knOZO2&@yUM}WpTILs zR0MYA12({or!?(}-gFu~>Y6x5Q{0#UoEuhRIlk1OnF|QCaNvD<@?|I1r;O6-A565O zK_u|;BzyrxC=}rWJdL;Al64pLEHRm^tpH(U9y(84#)f=D$%X;0=>)cc5L?ybcL7dg zUix(7MJ0Na2T4?;^e7lg=~gBrm(RZ3Qc%amuvp(%DUA{o01x$-Odlc6hU?LUz!o!wAkq2wG7u3G8I>gdt{&kE~61 z>~CK;OEek9*LV5PrNh!D$X~<`Jn$Sa@0*{ zmv9nzaMqTKmOeUB+*;Kn4qZrDYe=-hSN>>-R{hcNtsb!7*R?;kT|cMlOXVrB#MFTP z8q-z(dL%p&-+X;%IS8PA-nRwzY%>!o5Wo;UYPR++5We`8(iO}rJ>xEssM+8Ac6sDz z>3-USn^~2;LC$OtUgX->b|6XYnh`?gEL>>F;aAa^H*FX`aWhg1(%@Fw^&v>qd`q=q zxXo$^)G1VDgfr`kkUEqW)0G;E3speeimMPT>$C^hVog>g%p{tt;Kep-^t6Ug6jv$9 zmNUQ>Y=Yl!XIWXRG=@g)Cg8-=CM?j~`6(4AKU% z3R@J81Ks_}5Yz^ri{(jtG!B3iJ=0IqO@PjIqLG?SWS8f7)DkSkm@#k!lVGkZc@eV( z-)K2vcO};BMBXHF>|5D!#3wTjz3zv1Svd7OS{Q&(rNI+)tyJm}dTi~di(DcOJQi0r zN30$Ar&>Owh#rgT*?VWC#^4n#5Gd#O9q&)jvHCGM4>*jkP;j5ekEn z^Rmw`hquWYZ_0XA#WlbUeZ#q0Bz%SG;%(_Prb{;ftukhWxe>wMEPrI<$jNKnZUv#K z#2_l=C7xX^@~kfz?G@cMh;?9tn^yYQX$OyRRl;aq51!h%4H?V#*U429Y6cdv{LBO@ znE%?xfCr{qOs~Xo&oZSIkn70#wQ0@`V?G|{p7aL;K}*GCFb5#Dh%|MEP48WhL0Wq} ze9iTN-G^daeh^6mD%j1=`e^_a+hqPo!XNIc_m1d2OV4En?{H8Y}wAGU3@|9vG8J|xb5G(vRH0ntAEYvz! zI_LRRzc9Bex(e;$UC<2^Z4eACm_qACeJU9=QKcF!EqS6QRfnqidzV5n9dUV_8B6X% zTop~zA_8C?>CCs%uBa4K5IkehG$3N} zmpo;)I=(kP7cerO?CAyU*pO@C(`BkZt)XCm(p)|m1@zTy z-KTz~&Ev?3Y;@q2%P{m{ud1Q2#HcU5$jefj*a3hn#GqB`d46EX^_WM@-~)$CsYoJg z268A--cItKtl=Xx)Y7_~7_H2B5E|L=?H<1w&)M+I=!&6}q&$H+|6MBN(xYdLX+}Rc zEBxVhC#P%lXn;1`b!@|7UpI|hqQz<8mMXqO^)=dFjIZtS4;!^{QC7fRkO`|D0=8(+ zE*hXVO9LZy49|QOeNZdCYJ)K|rwzB_>+9RzII(-i+%Eisx?e5LHiYWUoyWz#W*DCu z8|;tIdc&s!v4B|yXDtL+ycNSk8QR7f}NvhbAr_|4LAX&QTWTUKS67Sr1C4rj`l#medT+3m|SA_rig zQ@8(=%gRZ9-(0!WDZn=yO=Ih|3fIPc{gWwMuHO0>W<%o$hPr40WyE$um`+*O$3Da& z1}5;B9g_GOvpi@*oQHbC5aE!}W;i1JtPwkp{!~Hsqg}%hWN~wEqUR{K> zPI|W?)UKIcZl{ZzyXUt>^G43da;hwwfqJED(=Oxgz^ZUX86-AH0a}QVK%P*m+dw+a-%+wz_<|uE)R=g;213k*w7g z`*=BO7X0B;T_oY`i*EYfwhve??BDT4y|GI}ujWnDcN0DOhLN^Hbz+YbEdYp3_-OW0 zJC3JJ0;ifvD4#@+Euu8AMI}w4^SD*}RnnB@Db*S-gVKlLUGOxW&n3#(cb}y|9)gPU zP#d{9YRDX_grIXajyh?IeRA5*6uw9&8)jBt+20xf+BnwA%qF*cGy#WdIJLK*!iJJF znyMDAc$ZH0*j7V`Y}Ga$_A<{NZ@x#=@-Hry>vwM1uP4zk7(g}Z=gud^GZ*-m=Zwjk zZ39G`VJ9f(d$?UzlYLZCHNfcXg#pDq7e=Kz;ZPQ9gHos1`)Xqrj9Q?M>wQDCo zk0l3Nn_8pSE{WsJk|i?jMo8;srsiaKcg_WPCuF9^WPiqD2_RV=odaH}zslRS4Xe@$ zF&E!ewW0h4i$c2PsG9m_j|KTOLba`f)p%GiygEH)1!}e8M5K~-3wfi|ykh;GkB3t` z=kJe7&NDB;IZ{{7De4?0cKjS9I{bFiJ&Hz~%E@iwtCHwFy$bWmZCvvtaV|4vJt5?H zuv`nWQP1*$7eGO2$Q!3G0>?4oOkr6>jYbirG%=cPBhO8xjj377`FxXiy@z_zmA&FX zgR~9aPNTp#dFS&5?54Zh-j1>B_OC&=zB%ed$nk3@WOv}s&Q1xsfsgr1J4W^@5v4KL zphoxjxoy1CLqxI7wMoXOF`oQ)wvZHJ4sAc3!DJ3(3cz5NI6ztlxm6Wt~Ceq)yqWjkQGCCD4}&zw$M9KYgQ$lRNPcklLc+KMa6DP zEw=@qGtY|$^UJ-4g2b?4FPPAM;9=L{)2(zIDj}6gl$?M%=3JO<=e;&n2(QKgQwBf= zaR{6w42bMq`{T(<@1O2_JKHybiP>f1(ByamOGSrs~_=8a^S z?W0(eIc#g~fdWhlYoeB7O7G5KyG16{LVm68&Y=Bj=A$8MpNY6;^DF#mBz($X0*pXc zlY1=Kt%CN1M5B+B$L5L@i@8np<{-|>c=RLs0YLnuf)M!4>t;)E^pi5&v0sV+!uDGs za9h^d9w1xS^XL-&L(z7{lnSlV3ULX3Z4c4%ICx2X@0}fT(+G|NgA4shee2edsH0?U zUOs4xRo3cOSay|uFMlYOQ)uZ@xl?ufZ9mSU+PqH@afEVXCP!*#B?3}N!34Af)z zj8c}HfGY1=$Ha6 zGH9Hg;?ey`PiGdeG_Sij3^8$80~L~aHjavpv^^ZvvZ$^N2cJ(i%|N7@(ZLCRW z*dp@&)cFuV^w5f&MT*E-_Rkps4dv3+?=D-Imf&USg zxZ``e9+f5d+g{@vDygS-OYig449J;D*mtdLP8R|@cQ$iM`uZ#I9I*4v!uA{YO3a_f zy#3Y z_ox{Le21RhKEH6i-Zi!OmP*c{jXuW1SelNJE=JGSeR$tIzWKbJhcqs#11RkHv#d{E z$(1O$V+L9(9S*2xgZKHmyuM0j(h=e!DrPJ^*6c~YtzvF~`8q~CAz)(q{tCSgV6k@U zCBf0AT0asQ28Ef>gDX^mk#YGANX1=1G&vwsgG@c75BHI3V73c!$~>y&>B zSQ1jG2#^_ID{67>wS``W^Fgv*@39ZC=B0ty!1;~-eS(93go{ol{02%dN=gN@zy>Ln zjeENj__&YsD=5)v)&ct01p3#Zfcn=2%Kd$A532B=?a>?iPwSrv>0j{B;{dAjABf#V zt^@r?>i?C7f2HwXX#(xtL~aHx1^b^zlKJ510z(S~BsvKMB>8_HNhd}#DD{6?a&2dV zK(TxMhFn4CKmf1}tP_c#p00V=QA??Fk2veG-Hf`!iMCV)GZp~6pw%fzAL399FCJvV z&qvNQk#ImU0O8OA#5wVe8hOro;HwgAjDzwwe|W-G*frn*^@qpv6>U+T{1@wcglqyF z6krIym_vr#&#mR>FF$@RkB5;SJTcQWobi0-{EDF+)NcAPM`;Y!4>e`IV%9H|P-26ms78DrDjxeFB5p zKj=o}SUf?o7xUVA@rXe}L@pu~|UJ zubX*$E6^8+L*PlOAD8+A5xZ{w<*tS3i!&Rz3Ij03=vmvin`{A@_&-@fEom(nbg&=_^MvF(uU zfZ7??Q<^$N3+GSc2wXLZ)A{wuut#zHIV)NNkq>efP9DN+0at5|Kw{9(te%LN#v zCG#_)+xjUXZ^WzMV?;tDQRKjv8sG(e!3d>#B)~sZmGR~T)N9+prCcYsq0j%VdrsOd zE0w=0g}kw-@88oj{{92t{pyR6*XuUo&l(uzpt$-P?^K_SmDj6lwOd=?G=ARh#F(9x z<9R-y^LluW`uh9*cop>Z`rP?)bp<%G`*Zae4Gy}u>)#LRMF24fX_AT`XX7#tA)_;$ z(7pk5n@AF@RRl*y*Haxl=wosTiFB!GFT~qdUz*lH_LsA_-9#)bA9ghYPn0xjlLs5u zDaCCDyjWAj8_!HKN)3>YSYww& zTfHWLh!9Hz1DZ`ic@{tYOVZDf2k$LpBCbNAh1G5Yf+oD!MZ%GUX@GE9ai$6SyOTMt z08O?H&gY(d+8ua9aLz-diMZulV!|_;Ifh*oxsdg{NDF%f7evMzm#i^ejByXth&}^R zBuUBKzB;BdPzTA*JR#Lj(hFcUwueMP7tsuT#R(qHtQiN28k|QRcZ9p=iW>oTj8ca;LHvDMBe>*+a;Jps?<|hs5Lq{n&)6bdj9#r2 z%pD6sC{;*|mz+NX_CQUySO%=2j2${3OO z*z1%%Rgi@=)Y!1)4Q=y4k0pu$<4!B;J1La%6$iU!NAXY$0Cw5D*{hu$F@k{k+WXMm z7#N&xse@V7V#!cC>ssNVDgatk-y;CNej|<&%CS%dzUm+y^83;Z2>69{!BP#;{{^y7 z$e58S39cI|2|Qt+-UPVsa2st~zoNQov~j4Y+;xIBtK3s0*c$!xUt)%@QH2Y$P9@8J zSm+T5CXnJxnGvcC+*^W)r4*i7CzVd|o8#k#L=j@F@}4p+<>eRhee`#Pjf^_F0uu0J z9#!#OMrDE4{lj;yKyf~wo+H!*^p_mh4Ef>GKExkIrg)V*x)vblmteho84nG93lXn6 z&gqyI457j^M2_K=>Bg#%wu&M%lg{_N)*7BM)2c-+u^A>HKM^ZbXV$&rkWiBrHOPvV z6cKWIPP&iPqGyxdu1Bp!nZ36|XEE_RbCIMHjbkFk5PzKuEip-Ocs{SyqG%&KFOF9sH z1-I}ffb&JLJ0yRDMc`*(c2aq#pQjV3XHAaLazBK#657L*`wKz=!PJlDNjhcj80M~5Vr;sRPlC*`3J z;OspRvve5^r3{x2z+cgNOs5VUYU_l^JO8_{*7FCe5W=D$0vQ7ZMb|QI#9c%b@rx~wFu3U zZiq}J+e48c=a+L?-NKx7(+@9*sYCsrejR0fwhF<`#7rcfQ06g>xhlgmoG&^pCcq$qcrom&n0lpqK?(iUT5Tqp+ z+9@JKHnF(^Bef|LynQEvit=e!(EOif&2>QD3NUNV()dc~(~w{TrvGeW<_gFPJ&~m8@_~?+2bFyI5q9R?h4Yf(^7xt0cf~ zPRM5I*0sL>^bT~+I4R#smDzops>gc9_rB8pV9okI*La163LT23d(rMMIu#E|wILPr zg**s8CG#3?QdglF-g7qOIfznR-!7o}E-b?g1*0NY;NR|Nyx&K}jLr0toN>)Y)dZ=PUuaT=%uoKyCgP@En|Sac_+>m9)W7@Br_d57|Zmsy(@=VsNeIE(E^x56e{a%EbO$gP z-KV6gt|WO>tFS-Y=+k8;M^kJkSR#`P;4p4#uAzeqatuyj0Vqz_Guvvc~#0bms{ct zR4ReqX}z>+EK1$;+4t-gnj(O2a8V0sUr)eZx8+lsA8_O7dr87Ad;cM`HU zbe2ypo&xUzuIji@iseO$wg~}`6S;^}-w^Me-8nMCtgArF0!7&t)ODn#FyWh*`C3CB zj?L)?$5-Gf66!tJXD1)7xAI04l|lmKW|HVef+FFd*~^={*Mjz;#Q3O@qLe0qciO11 zESA4y?ple{O?k3w{g$zOhdUq{r+h+SoJBHa1{<})6|}1~5ca0Q5pMvTtHS6adWyx# z;#rJJnfz+LF==6_BVkvyo9HS5b^f502r!gJ=|Cv)8K5p3P=)_|BFEx;L?GabUfdP4 zGyz>sZ-UnUI$Oy{ZwfE!$!m z>_SjIQl4Wqt%C`x&pZsczHap55M^4Vqma>c&8#gr)@?)SowDRz_mMB|0j_WCSNhZL zY&*y9GeSR&JfhfNuGQFbmqtq-$v9hC$Z=7>L!2LQrt>#={5Qh6F2}=5=v&+7<#y|m4Tb+WmN9-T!k_*jT`(n&WR)R6UO-l*su)eCK^&Mt ziMUHhbOM%pI16i{`Y#({=MuN+eEC-Uvz7%Pio%5!0y6$&e07;f3w0CRrQT?~e_rpo za4E}T(q~%8VC|EKy6xIDyYBVqPK&i>9DF$8rik8c%8;h1rs$R5kQQh#P2=$)TB^7Sw?3 zC|<7xHUGesB;`bPQp1+WE0yS6@Lq$7O;0GNRhXm(Uh9kl6jE*oa^&fBMJ$Qi_{D8v z5+ah#E8<>uj;rPTl=uS+;QV=gs)9k-QxMWYMe@1WWbC>S-gg`-wx7YopvzQFHhu6{ zkD?g}biHCEonz-5a z^$d7@B+Q~>8(OE{&inb)Smc#8urL?C<&A9#7uw~39oKGRosjkk%J_Xl=?eXl0@@kU zUV2z)ONz6WAz)L^OID^+Ludb{^7n#GQu{7sa8<)VUVJ>4erB=153g*9B+bpD$7#W+ zK0-wES<|u)ao4uqJ@ZnMQ`2Af z*S8FSwd1DtkBJ~|ej8hw?rPB|9C_?c&+Y=h9^D*IwV#|jM3UaVq2{I3KGqnrM6WUO z3#gg;!?gu2GS{m7PfAf7p860Xa7IM)%e z>WZE`RwBLz7vC_+>Lh7r#dgYlAs=%3QNciA_2xx+fZ zpa?sP1~=uS>5zGuJ};bMbZlFc^oC<^9t${)y#^ZUyqHD{uOzF$Ot)4qV4UPExy@(M zPPCmPM2|4zGmJUZ)d*K#C@O!DNWS%2qvt4k@yUi5*Whg2eOeta*>@C*pue-9PE3$; zF|3wWU=(7`D&Rqg&0IuadOlbG=ZvVw-p6H0l5b^O z3o@#~Mz0#v8UTG<*8r_irq5W1$4x!S6WEL?eTAJmRJMp$m^YPZa9vZA05Qg)a$$_? zEkf!2u9>C;8pf6~A`)){gKQ6|M0*bRAs>)H9AQ=j$*a16tgvBRY3lllmJ<|xqyEfJYap6gnu?33Z&epKag%XIJtlT!IY~{N@^UlLeh6~pd zsd-aBx6t#Tx*TmDl9l>2RQ=3As>A0zz;l@R`zuv}Zg=C6jtv=#fBXz^PnHIyYb&ZS zZ`l7!0*mrRj0P7;I?De-y@O3v(gPeXWvFR)%{8M^rj)adk_JDp)G&Q0YatAK<{Wko zsIud5SK!^fQ`sqXWy7G=slL^;0HZ|^ddPWN!KG>BoJR7N{W+Na`R{Ccm*z6R(z(^y?U*Rx zsAJ6*WCzN?OJvkXCEtw9wW1&#`ClO^7u~T*ahZ2#-IJvq8zMb0!ZZ}(W=Q9n|ud-NxB83U*s|$#k8mMrNeJaC#D4(lAbageg&W$hmS&`61UH|{P9p$>}189?Q!ka z!ViBLvgrqe&Pxx+!vtM+i&E3}rBUnJCdnx~o^RQ7ohAf8fv+Y~sru$=Ys2?ko$0UX z?y|gHCDU^C7L}z$s>`s+YM|JQZ_m<3)_W?FD59&p2~($TZx(fKJk7^Hk4;iJ7~T}+ zwd~u0Ul+KNC(r8Jj$oW(Tcv%#-Jee&l-IqDxp6Otz)-cRZEKN9$cFl_(2V4pXX0S5 zcMzd9-BvEZi~%{2Jh-8TK&T+Gn6LDF8?$T%>{M0f^kcP2NfxL#<=;3MgSX_Kl#MO2 zl3S%yXGIFm_4oS-*_TGUF0|PXjO#zv?LO@Q>y; zH;mlpN*BcO)K3zFFA@VAkblq2AS?D-l9%%xiH~o0Rn_2grQw#AVElhyq&_Qzzkdn+ zkPDx6VALOz?(%IqhPPMdLN6>%>ne^cKZi=WS_KD|yCEHT^55g@)VoX5fl^bzg1o`F z2Lsr_*e2YVdOe0Li{$=PlfCAXse%@UK9p49+1tj(3xbkih?YQLeooYLjd9joLwQ3P zyENm~p}AIEOPj9fWn65Hme4$K``c^7SN>oG7>f;tgSfD^$hhb6TRWHQZDrh)d=*Yi z7X*48HXlI=6?T|>S*11knSSAXaIWBK0`Lp8t-|3c6XXzTB339er}cPdib;2 z3bKYI{lVHnY7V90h9oVy-E6W23WkzpHHckEO~u_LdgXh?_%%=K?dZvRQMm;I`@+c_ z2^t$&{@Huu-6@}G)gAguC|o!!eru$w44}nsw(h-|PfM>H4)4S9@qJOvjole9qxifMZ;oQN#<9Olgj3ZZdI5LHUJ&70;9zd zY2=b^RV)@I6;J&RhLdkkc6CNQhL+ggHHE58a;v^`$K8>37$G>z*YOc}@9n zQA>OPlOq)6w1W+-v8gjdKo^6P>SWfZ-UAX)O*Wxe*HC$=K(~c*rt(q#y%hmb0Rl4^ z$a=?IW%`R7h>LIZN@_^RZjgM%-@&q>5J+(BMN};pI<2Yzasn}VDaVaAAHbLr$w0o=}!1tm}T<_Ja;>4|eMl$9lQDS5YvZ7NwVgu30kyPI!+P zKMr?mk&0B?RqP}?{72VO?z|85xLH|_b2PlovD|rGV_bkOdyWQ|LbC1Nd|zuG7pk7T zed3D$yp1tlU)J_eDw!z73PAOCCK9R1*wr9Dy#E^??lQ0B-2hOCn70>zV@;V*n`(lR z?s?+>ODuviml6Kj7M?FdNFEvEu3)L@YOwn)H)?@a3xt&{cz{M%MRzI=$Vdh8%9dLu zRL?;yMO<1hVF+)j^86YO#7gL0nBk1UUoT6}dpE~6u86!edp30ZFF>Vjwm%@cgIJzM zMF5en+NIP<#;arpO22Y7@%G8nfRObX5fc&i7_v~_>XgAx1lx@5Y^+WC5iC! z3q}v*osl@?c=}9ea!2?R(RI^>+g0~Wcr{K8YG%ucUWYN^1b`Frf6K!VtMGtWakDJ8J3H z9N6_SO4DOd`<+GC8U45!;dm)V6Vn&c^hVSlnFF{cP?>w+C+buUVL)eu%~4g8|04~i z5U7!O$pF$0xc;xdV_L>&S{{@Vg-FFEHl|R&SW#zA1!nGpVeeoKen&%0M0r=|`DFd0 zi;4s3y``98Eug(`T0HOd6Un8l_5?Lq>q(n3ES(Nos=ur&&f-!`@_DsovUCU(Ue-Zc z6DS7rP8Ono2&XcFa^j(QP5K7xHI0&LjV=kPf#A+B#FYYn(b^;vl$7&UP5UEq^kZSw znzZzH#qz^ARdG~-#Kgmiq28sbJPNbLuO7?w4P?Ht0)UI|N0-qfc1?V6NwO_1#1)klO6?bEo5b64u`GB-JFwDgv>rkFpaR)C7f_hPjI9M)sW6q@kC zb_#EnG{dE2Ak(RZ5}6R<{et=G7xx)GX57Sv>8-%GKOVKKZ9{)|J>-o+`t65^U>FKa zUmEp5Zvh4u@7t+I1B$2k3tY>n{xTI49|}5cS;;5WD)HPkfhQuj~xj z=(-^_b@o{wH@K7IfL$_!`Us^XJwEB?2E?JT$!so=KE0 ziMTo@WD|?&^uR|6BfO4_+suiP<33xZ0m#;i4X<7Pu_CQxJ|tNZ$z}Ewux*=CjdNb; zV1R|cD&o4=-S2>n&-*5i)Gzw%-@ViH^I3GHwK|=X^K#$PDZ54>yk`R_QBfCEx7M@s z=&4i1>eWqe6Y4o?KZmjSxQoSP2eV*RT8h&Ij=F>6{(;qrlzUxpc-LCIOL@}`iCHnv zCw+DkplXftlsE$(4G1?4(a|K12ZggMCV-G~Qla$kQy6R+;z1tC>cbD+-!~bb66t%V zXR2rzEa1KqJ!fSSsQ$o@sOh(-$bL7-kKMteo$pdr%vcv7fxn-MA@swqx-Ucmv2yQ~ zo<*h`E>tQ%tD+v80vNpxqNKzc18i4yWdCr_+tC- ziVWSpb(K#g&*^Yr?N&vdz`R&ep~xhOG@|!zh3!4t>1R34puF$v5%2vv2Rp=&#P$glnXzNSLzK(0Wd5Y%q@-?kKnj z5WftI-!=p}?|)~E1v4`Zyb45izcsdiT12bRd!x~rCvG#87hTno3xJ*Vs+qDSM``3 zYh*!01C2uY@im6KL4I! z%+ZsSi}}w8Zh!vsXGDIfxIhF;MYo#4x#LXS{$dPSCj;aU_F9&Gv|P^9qj185F;fD zP?{98-I4=t<#0vRT^BXQgzf2_^Zg47S16zkcz=6FCJak4g3l)Uv{!6HI6s)nqhuk1 z$2Qb$Ep)5I^Yx|ukfdmUwdw?&s4WWUm&U9`j4V2hjSp{xiAi1daQAH_*eT(SWXdGS zzcuU_JV_3d^|6*RFe(77)9UID(dVXxSC?|)o(myh6 zqQZ-J_6y5Co*hCctNZ$9a?<5qM_ub*NEcZ4cd}w{09@ldF z^YCki?0ALEZ`twqu#(rLrV-d6w>@gS&nKym0(_b7H2cOgnJoYi*q5@8wrhtk%y0qR zYHNe;w}Gc|cXx97f0jP?yf?SPdTd)?^PjBSu{bE`c$1{S+qPdK{*+NZ@Av&MD5CGm zoB-bqe!iGrzN4&veDr^4MpN8hK(T#XoEpY{tta?Ctod+JqoVxHS zTU&O3twdf$r&suE1>}4v4qvXm?YJ+vf0JlHtiCX)AY{Y8 z*iQ4HJVHz^SCP?X*Y1uR1z!^FHljVq*~WHguPHHIAG_AyUPK6u%z~UN;s5kF@9El( zL=^`=*sOF6oKxGqs-~1N(Sy&*9Dd{}BHipmA@1B>O-(q}E4li|V#r;nNx+}Xo3Q)8 zSp&8>MVbH^+&K%$!CHsTmcCyd0hOsF9SS3ql0Kcl4|5n@&aKy5AX^Rszjh)N#)7aC zFeZ5D8oPWIyy{q8I1!7s_P28}a}00rH5W^RHfw$tzVYz?nTikjZK0>Kg7*Q?o3}Yz%eg&`5#s(!kO2BddkZz# z-P%X8J<_N3p*vTaoaXn07TN%TpO z+iz!bQ?;=oI0G{tx)3oXC;=R*nLXr6#s9<{yZ4jN)ybSbywJ*+m zZOVW|hwgo*`5J-xt(Bz&B-ztL#n5e@6PKm!^31w43zLN zgg!LKwzI}4^1t#= zUX6@NHvHk;AP{?}?;FiAOeu)g&DXQ@DSp&s!Cc)dJ@1G?noHgnH<7okR@dO zIS80!D$%@dQ1m8tR8;bIFxDWAOcDW<^h-Lc0?_!9H1y^ZC06yr%PnW#72>lojzyO0 zMP$wOA!R~?i4>JEOV|e(WD&m7d+jk<4}ZU$?92^2QUz!#e$T?ZLb7<9dg0*kMI()c zRi2!Thi_b3YOETrvPQSlX1(jS$d4Tm1B!i%=rI@HjUXsTp#4v1cT?${uZZbl- zr}mvj{j|moe$S|o^|AYHjtLPKK~cM&)|Ss>Sqddtdz>FTH?*-UGLEik(P|$q$&TM& zut+v&Bv|(gEmla1cdOD*gQTc$_4v5AdAYase%wF_Y9nQd zM5)FV1&7RX1d=cC(nVTCZEy!b&t`xGRb4S=ebd1>Sxjr^j;irrH*Hak9GWN~!D83( zXC0;gK8}JWt`0)6rnWFDGQlN5A~IWwLyCL90}1h%SYWScl=0ZC!V|xHmkjR1L&QgljNf2VxhNXI!b!657FNFK<;36IC=Cncc9Dq@A`? zeR@^D&M$857Q-46{T^;sPYp02j zT#;fMaxg1+jzY*q6sQs-(+8d+Il6SyTBTAspTel7hl$VC0X&afLd zfb9PQFyjoxS^)o~70_oW)&?lc`Y+NmDB2UC{lB-^?6d*mYCWkr?$JN{f#RRu51YE^EU6*Ebpk#Il1s^+#dr`e zA3`5gsQAFLc`Z3Ph9e4%i1tB?##svVxDmmxuj{>ck)seQbL$r2)GR4&?<+|3pN|H0 z$$wOtS>cqily!I(acp3mk!fPyLr#b4T8c+RFj)g)Y9?D62j%hweHN?|vf86zlNiOV zVc9v3<`9RaVO19~G9@8T-tveb`~WQC+hVNTGY*5zfx?)yek_M>iQ!}iQu?JNnh~9e zR>=R3EoI`G@aw1%8FPjmjbj-Q%}COPa}{@q`;cgG2TW|gt^}C` zsLcw}BMtbSWOd+rj%WXnP=C;1?3+1&7`v9BBpIQm*Jr!izrANbA|BZ5jVtMAx!q~Q z9f=UDQc}zK>HGbcer(M^!#lA}m`RzS6+ZlH-HIf0nAQ6nft7nkKo=$Y1N{D&5SW~ z;hvI*@#0{Km_Qm1s_X0jCgA^2X{X!=$MK~erHDiK@k!Q$voT!T8yM>2=k|MB04^#n z6rV7<=>9<*#$06==(}3a055VTA`s>Lct-m=oNw6hY^U7#4jSuW#E;M&?QGuo^f_PE z{L)69vqtbw<1 zELAja1$qRjW?*W#Iwu8H?ilbuC{OGdnZ-#()v1n~D)j~KgFC^=>NC=P$z5ySUn&~n zlGEU4tgy$ucVrF*D05FJ6G7*;Nsh(!y_`4fU&iHzdSK5Q63D(lfdQXw;jq0x1))Vh zDC+ljKzecgT^S7fzvRa9y+@2o=LDRjForR+KgYa83Hg37wg-*O6wL8>XI1Hbs#sU; zlIdOWQOP}6uKcum>YwWO;lJGDymFuAs z3M>Xg*f^Ay3?qaI-T@ZlV2OWDl=6+swAl& zBtG#0$3V|GaSR#b-8hb#1{EHs@X|Sghxn1_uP|;}TQg?f>kfPlHpsAo-kntYRtZep z$(f_6>vX?=^4thChfaj{86fdlW{`dp2&&1yv|Tw41^+tam@|w}QbqZiqrL=bw&Tvz zwAib|x6`k(VTuWv#4G7@RoD$jC>n<+a1e7xPj^<3Ui391c{Iq^0L!{L_A~g%4s=I; z?0JGj4bQ@Tpa;%53220SAX4VPs=cJh6=h5yMRnWq<@$fra6s04!eTnUwBM@f-_Mur z&VOE6MfiPd+kBO6=lU?}0*@>fk}^-El+ zHs{aW>HMu$XUOq7Lk?*x#yU0!-D{zoW@0Ly55;VnfXGP&5q~0ua+2bD3INY`jviW+ zU-mAD(+ie|;p-a&>uCtiD87?Xq|P-SzPgh3f_R%akiD5zmJKu)Ghf6(epK&=j5OXU z;0A85;Ux|?HJqioPCrZt`Iva{Y;!dfNtH%!78i;I6|Qgk8a0;GS4^{r2atH=#>D(7 z(4#a0XCL;Ou-*>SrirH{(LyD~o*n2AmL-+|%x;h>VMM$o6A51k(OsuN!gckePw2V)6<^vWwPMW{Qw#{L~M2<+4IG%tmTNi))b-~vw{=y((kM{ z%O$oPFC>d=V4!vMYGH@rW}* zge$0(274Zl{+qb`LQ$5ko$#u*`0j8S&wro-sQU3gbJ{k&$ zKQNTmkqs3J##of6i0B7M2ljoo)1Ub)p>B8e=IA z8<`XUSVmrq%9ibU(y6hpH)eg;(o(jX!ER)(Tz-F zQ)*C!fkM8y#W#CxC~9%fM))ude83hhdS&a!C7+u`R_NEwWJ#pC0BP4XPhxN2N^2;% zN*YKwZ%b{vJOn_Y3JcztSvO7X<7Zq~eexEAxi4U*?!(cQz))z$jov&}c3?k~@v?Ko zN|B?7D;^&cthD$f5>TT!(27=Wr_K(RrMF@PuGf5oHRFn|@c8U#01#^mW40C_^-wQK zAaPWf7x!}yc5*VyK(*3VtAjQ>^m_?!zEN-|2^jXfM1W`?l=nrpgKP@50Jq1`mMCjp zOj^AxTyf<2(wu&F^5Cq6*rHRpW@JjBC-L@6x0j00adl<*Hrk-1D@n zSE!(eZs%o;)HB&OJukXKZYU4MDz;TB@Ej5FC2cc@gt#*1W$@hzcv*x`9ph>mS|}Ao zLQtyH@oW`6Rn!&qpFr+gO!BF}{)*!{^Nf!5ASt5JXzcQq37<_!m=IxT06`Syh4chD zCn1{!$ybJWdwppz#X}hI%xscd`VT@FDz82mKo@o5@AxeUs_qCctqLj7`@+ za+2tN*hOlIett-^zv-k`9LvlNN#g6W(^S9(n2lG9+pnMc+y=epJS$+PMrHrO=&MmB zVHjLJUdZzgB=UhZI8rsP(nFH{Wbsc(2NSUCY`jnoSw9wWx2y!ZW~MwuBN;|9`SLC_ zlzsTLI7L*b4g`qLFmw(}qZSu$!FNrVltC9iw#y`eRYB<(K|PGH+!KC9|o3r9Kk^ z=qOmus@p4((dp>AIHOnp)ULk#t`=oih0V~v9h7-$qq22J)LO!%UmZ45V%w$&YfE6d ze&A&u&C}41qO^^bebU(jsOL4hzF3Q<<)7sxN6E`hsrNwKKAqKFssI@gGj}Jo9Fu0Y zKhmAYM$E{E64*I-F&{9L>dh|Q@HZcpL^?wWj3zb^CwKOict(Hv=QGmR+=a!9=t}L4 zN+4}570xhX_oy^bVcZ-}PGEWFtoS`&a)-x?0glH|yP>m0c_E8;8?{Ge;6_rrfKt_j zD0XM0IVm98J>I^3npL;?)UYYyo>3;^ka&!oq!ya%!9IcJnKw9o>uka%mXz`MYGKp1 z^w?oClR=6X%vJ5%jW&-C;<+xx_ZEha|T`nkfG1URL3Wi)}ueBbv-|y zHmv4`^ z2fJguqkytUZCBUa$5-Ne>0OwD&%p@Iyjq>i4QC$mC6~`+dZHb*q@=b>60__kC?R2X zd_$nBM2;;^LTNUb3)R!h@QfMq6jPs=znG4@@J*Z3Z&&_= z&M4RH9F0r4{e}b)#1-2F-B72Uuzty6@&lZcg%D$AL(pzhIxraZP~}wDP3U%N`i?pI zb=*{TTA*u##e&0~Vh0Of;LPZC9&UpbJQTRHWE0!3cU}us<;ATXi7WlWb3V$68DC9e zD(}Z7ll6Xb4hn`ly}i)nW`jjW!2>6dw`lk&f=3W3iOCLwW%!a|N}%sSP!fh75s)(> zm&4zE^5@1I1HMWPh_T<(AcetefsR0vUtxiss6#&MHMh~quce5S^*S54sT`(AxD5=E zaB?J?T1Hy$(v{2)nqR&Hz4-&OC2pXFc|dAzS~<+)dz53g0Fh(8LfL-?ynpCN`+4S8e$s++O`yK-IsDmF}P*jGi9f>h>fYqppA;RPoLr|F|q-X%rmu zT24Rgrkd$du&%x$Ouyzx%~j*#+z!0VcK_SalD36%`grBFOy$1cwtX4eor0LMXbk1F z)Y6Fqaqo>lqYAl)e&P(K@*GvvY5pAsu2x{s(Xj%fIE~Q8xvaU&i>0hNV_Vr=AZ6=- z3-6xeOu|m36CldR3r>nlwJ%RBD;@~iXv~*%?Pw2WJAx(JRgRY2ocU;+V+hoKmiVD^ zmt5s&1rDX!any}IN$Sr64<3vhz0=^`&z0+*kh5L^pIi;3Phe%gv6KM3M!ulOKP8icbzunzOUkFUEI!hac zuky?-tM0eG2x6J$hgOj&hCEOnRrdg9D(z?lYB?l=0ah*c-#N6J`RewDbJ1m!Pw%fs zz*6F}yxpV2O>xb>fTrNrR<8;E+?yIDqbV4#djXuJ&Tz;g&&&+f$~bb#-5iMCj`0Yz zfz(~>PxxVP28ZQ}sSk8IN=SjMeoLpnS{N8H?{{7A=(-H5qX}+-CC@;x)u{yw2X-p(F!P`%5c!1Yzoob6h~m+qn=YMt;{bCa)S5=?LUX1UOLuM)1uidcp|me*+)DL9WLB7jP=+eE3!pLF zF}|iOtiH)-U6cdw>7PM+y$9Nuc82nqxU%YtftL>M$P^S(mEAvx-gXD5T5Qi{KxSBe z8eTS|^CBkWhyR_vr#57LD)l10m5bgPD0lBm?6I2pkU%(zA>svEF1LA6wY9gy3k=!Y zK4PWw*!L?N$lmVdwjz>#WX{QHt&3o_)@42N(t@@kOwOgZOsYe2FMy302t8 zL3VZN=FKvcQk^ZV(VtUY@v}MM0#xziQu-|bizVh%WtF^gn8y_Bj%|&I%ae*0Kaedt z?8jEyYpRm4IP`%f_u9MZ*79vMg_iIyk`?71l8)n*o$<3qt|}|U)lZp*wOxdnuT)c? zxHsRsy)!&89X8)%P}EegN#vpO23U0Ivh@(lR4lff8Y5tEY#+j(+Z2dZ zg*W()N6vHQedidWtrZJrFhw^NFauFPYq4K!+-uYPu4`>d3+OSNu9K0d|*b4hXjJ+#=fW zB>d~TbH9vwB+xxL+e(q6BiZyIP0VG2VhDB8Wyh;n#c_E|x?g^thMF^VxQk9cNvIm%eg7aeK0Z$fr}Nw!RmhbdsWCgs}!+DvL%hXd47l>Kqkt1dX~ zzM{uF+OMSrcc9^F{qMBONEArosZ&_Uo}zRkhGgv6Y|dek9vCe*<~tT8_J7xX*cWB%BEh@4~3m>+lW-ZoOsM> zeC&V{Tve4^bT43poU$2A(ArRRyHfYEOr~eKN0}moYBDjqa%-|R##;CM??iU7U2W8f zlkOXJ=0BV0e^B;=5N2w;sor@*T~5p8Z-LmOti3n69X78VKAiA_gF#WH)~RWjCXSNud>t;+fHR$$q^ zS(`;M9a3`I_q%Iuvo(h3Lr}5dhmshF*jg&y7@{#+?IjOba#4dbS}d_L$|ZEGctA3c z5e@VW$2YiFg0Mt!E+-?0f4oEt>JhY9C`;miA#7uHbCtWdt`qM@@wOV|1^Ml7mc3-y z(>nqC3tM`hG>lWTP-*e^>8&=#c7kEaOwO%bKfNjk^Z-8|>DdO%QQUl`w}CJS!$JN-9|LsBUrk9AVVa7>(za8X|GZO z{G+3PIjiny=*tE!c|nwET?X9#sU*q?SiN9e;$j6M^%(u}ZBf z`*&Qn2@>k%VIxI+Kt`n}nBhRzC4h;azIB)|xqi=*Pi z-%DTP z+7>A^WUTGUlMI%6o9@f(ozgBai6Xx|VzcM28H8q2tD?LXcmK+pG5-a2>BA}!HZJBi z8;KQA+=Y}`yu}Wcawf!q=NQmilhFxqR>ju3F)?$qN~?rZ{k>xWDNe?fRfs>5jUC++ zLFa3Pr9p%a_cEi)Lnb$El{yX$Ffg->Sd+J*etpO24yg$c(c?|Am#%pccHu-KQQ`BY z#8bAMOW0zQia-7piVUfEJCcfS{{6?HJ>$;eL@K3;f&9%RnF}if=)uh+h^)=y51)n- zfS zx<$v{zv`t=xu4}1yZ}rtkL&5+@K)TLrVQUj?6e&O&xtu`kI7W>b)y)%-bR z3Y`R&tY6De{ltb!op^v`hh8LMmgmEZ1T(?!Skt2kFkCG1h34W<&r=c=%d%-_yv0#% zt#$|?kkPRsaQ^;ovu}6u?mH+_&pIw}fIV!=wk9gv)efcv%5xh>MO{~%(W3rTB?HQg z!UhD$nPQC$td1~gIQ{^mUpyNKA&lm*hZvqrk*|34vqc8de2A%7hj=0y^v-+Y&wj0r~1QJ~s!PBti4(`{)M(|tUjBLv$x!bGen z<_%7%|4xSyOWEjoMc45P9;eJCDZO^Kr^|!4)m|CVkR*@_Tig>*ei#Gk`YJ z1Gt{nz-e7Y3yB431A}qkLv|&YbzhoM$S5Vou6ppt>Jm>Go-&@)3Ocb{FVOJe z3;tRIsj-6O)ZkJoinxAz z9?K-gWz2%b8$Pb9pyld6$zGNd6HRnR&0F918}-nMs+lP|SEC8b)cC zYKa*&5PPNiOycc6(=GsL?$j*>ceX`Rz4rBh%3t`6clphj1EWHzC@J4)KgS(~-p5Q~A3=bDH$%rcc-{5ey8zCKS#G-U0(8M2_^&D|>tx%p%y$SRvpm zYDH%pRx?yjN(a?wP|Ac|W7?a?+|=bL&XC_Gm5;gltx%RQPuRWlx~Rf4UGrduYYR|; z%ev*GUJaiy^IxHBf?(4a`V>os%Jk={k2s$MeJl1))kN#FT!$dOfy{^YkiuoF%92 z7xa-zdHT^lx+h0bbbgjVthQR5ds}J~A{g0bR}_bpgy@ zl}e9$xDfGJuBUEbCP~(JiKPOuB=|+MGZuLS@K$ZtkP7ls#yc*~r^(b`fLxAfkf|MI z`*|MB6qSpL_4L?KEW8;R$WVdc^Ft5P6vvOxUf zd8UuS1XCak);pQ^7;-ao_6oy~KnhR%$(X$_)) z=h{|XTd)?y9o^{Z!yWMonCOQhTQ;FyEdZRpVaGlME(*4UYL+B3%{XX4=nw`Eb;_*T zq&~x|a?i+u*4`sD{;F413rloS%TFqhn;GE7S+l!-+x9rUgsPtu&6-|lz%Iu91k zwoMCrd3_}?Yqk(~f;LOz&0|4IwxBZb8zB!$j<79ljGxKSVL4AvS6QDxJydOh1R96G zl1uVP3pCE@^p7)%UvnKBzNuqQx7&WgPDT}kL$}?zg@BzUIZRdI|Al^n`MKQk$h~yYNigQXvAy* zyeebf>(j*z?W@;-*OSNO6(23H2kS}ya~l1Bf+7e7O98lf1QdRN|2KQ$zedYy2f%x{ zcu0`FNhlDhJmi0ysPY)W{sPGV@3@Guf6H-W6u|$8|9>RlKN9#K2?F$uQ3M0X?hXM!)|9@BO{Z|d=|EiWJ-4+bc0&5`4V}u(5-~jVHS;mhd2Ip%kAiF11J&a8+3WW64S`?+V1V#y4vl7p@|(gT0ZrX0Rl>L`$VQyD}x4u);K1Mkz{-coJ3c)Mp_ z*za$wVIGDy1j05Gf+IV2-Qap2nDeALD1aSm`H+G&T2Xh?kmQ5L0~t5$_FnG(_qt1SN-3uC z$alvg3AG{5r}&sK%L_VpArAiu zBZw893v2zp8Dz4`%j3mUo|C?1c-KNfCw<3w;-o4s1ZHK^3xewvstChh**TP(a7) zIr-&Vl}(@VOCK$fES~2qkm54)>C6(c)?yaku*>SH9y!h(+bW#38j63}8R9Yxk~;`k zIo7pI9|GlJ^B8}F)NavVbD_>j|IBM9r^`l@G5CW;)$BQC-VZKm%x!YWM?_LMvAp_1 z^Y`>GdS2qHF6hx;(%5g0cwR}B+CnIh-UOs=eO@)&s^2`cIAXN3z#=@SkPQX_3gV!7 z_;SiHL!Z)S)iOVX?b+il(`*zjh8pj_+Dr0To;(FO1g>(M2OZyF-x)RIWB%&0M?DHv z6H-DisHHfEPr};P54e71W{B^{oEtC^<2@O#@C>!!7-J;+Up#f*+2X7#Olr0_hghMT$2d+gSk*kG1<;IjG zOc)xCjXI4uX>&XKskf6BKrn z_6elAwa8eH2WqAoml;m5*suV9KL)IMu(8|`LnT`VMvy~#4o7DMl0e4$@D)+i7XAEW zpf`Af5j>UvBCB~k;Pwe34*^~vfcpVYAi041w0p8e!_z7MWK%=Ez<{g;7^SQg%QC;L z-h_%_ogEe?BwxXLZmf)0PYa#dhV?ifnlVnyW3^Aq!ZJ3@1sn!BW{MF~7hPMV-IBw- zS~HpwiV0e0YGZ{8r(SeqRWmHOIa>c?o*BG5IwR^DVs-38O?AtEZ<~1(Fgjo7-quKt z5|s03Ky4b^8Fs3XTYI7eY9{%CDk-Hua)nAMlC5{!=sk2qQsC*5yZ8p-wjE-4c*kC; zK7c~>5s#>7#TOU>N1U+@?KUBZ5qbKLgMo;W~ioL3`Zuh}3cqXrE{`Kif) zk!Yoq@(~u6QO`mN`j5cRm0~C61BTXf>*r_ z6@E6Au?Q{F@FcK#SE9@w_(2_+I@;(RgRN`Uf2`~38e*#!(Q#PevwNw2R}eBUH|SF> zSDAcm?s!nJeTDD7jysZIQz)g3pnIF4^Em95W6}zWudv<)Qdi&XS44RO#n1^JvtFp;EPI5%*Oi$Wwsl zQVecGcGFYuiR}!t_F#I@tkSfFM?GCue1v^YlM&Ei10q8Z*p}X2UY%I3@MGDP-rCF$ zo=?VWsM#+CV=s*YRp&rLy#}H9AsYz@*0WeU_HMTNcWMWotwk_~I)W@Y;zCHrCW80c z{~(wqkuSafX=yl0c2&Zb#B(exEZx=oYb55D6RXHE`ui#1qIo?}Nxz$_z_q-Y-deWj z>^!Q8QSY!84w0Iz&c)YxO#Z3g;*5mqvZZVt5mBWcc@e3JJXh&jSQ1+>Z zCBXEOtS7a7hqThV1MQIdanMrN^ASbWAzN6-I<|2fG48dzQZCf0(lCDkydGJ2^nE$G zlR+Wxt25UM$On&QhN(f38xx8O7nK$V4*iL7!Eh;#`l00NBE~+SZFca;fm2sI!&p1~ zh*Zb8W$fsOLn*KFprdiwHk@HL#$uC!-fLh)Lhwhvri0X_Nzd&7G@}a&H|Jlft8^3kWU8>DkFs!Szq#e}pT0>tn4+2B+b=^vtu^PN90aMP9--Q0C@ z>C6f>FXTb`-K(ph^BDAdI(T|IEZN*PvDyrlxrf@S+&WY@ul)l!#rLw%mp3qWOdHUr z%sXj%fXa%lAG@WUomX_`KNmE11@zJPjm75@`C$D(?GY+_py+sU5i!?vZH@i)#5T;E z22?RAI~If9Qu9&L4P!KwnFZ8Q+$tC%Tk`lpVDU-`n_PFXKG^1bi(BO>&60_5Qmn#Q9YByKU;V^Qh8nU^I zy?SRbI#WGkoMn768!_AdE#>vxP?=FyYaOJB=z(ae}21m-VYl+n=7Y&Szss!tnS=|4!CS&$E?C#Jq z8PRLXP2=t=aCh-b!divb?Xuu)8#&#S2;=?@xax05Tb)%Q;h-J-Am$yb`%9jM3EVn3 zo=$g(HS6Pl_<~1SX0e^Oo#@@T#^V>6-uJfmrTaS7j64fv;n)g>LM8u{y}YP zpj|HK8U9-}!xNQ+x(t;$Xv+y}2FQWHY6$%a=xb6g?Cld^Z{I6j=W*3&&6=3YB<1X- zY2x}bsS-d=bvR0oGoGFF1w+Li)&QyW8`S+}@TRA5Z?N7=P>_B);1N7lgUdkgZYE{1 z(wOKB!KtGEX%ddFwNNBEDU&A`V-ip#Msf9Y_o0DV)eRpa+CGPL={&I()hVgzU5m#K=CfxMK`cqpOz7LI}DL7tlSFBw7i~wLXhr z8UCz_18&XT_GMDP*Pxog00UQ5ZuFhSK9O_!*2#^2afwtOlI5*_V_&}a*dOoA?kBIt9+)SB>Ym~YhOXr;iH3yrJcHd>F6IsN=I}l8CMe9}Xy`QzE z)2*rL2%rZmyeF0WlC!og)kw?ZogdSI2uS`C_?{~QCwBins4kPJy|e{O_h!r6UcN}a z+6oSsRw*GD!~0*q*c0^1qc+EVL?d;lnc?XqE-&!$mzKf3%n@?r! zJ`Yof+j;tWxqY2Jz6{f~+aE7ack6wdbbjBWBW3fdwwG&~|B?o;ZZnE>K4ySX_Fd~X zdt<`l!D$K0K$J3-~Ks@P)QZo&BPQ?fqB zJ#3r*n(%K+eTdx1W@~*6=_^#nx67<5^SVWhvK`HiNLHFFX@0}ad;Xy!m2m@TCu5A^ zfU%#WJjxLTz9g*|BN3qZ>`CKZOdgz-Faja-@a1>GM-WO*8e=y5DI=@GL8GJDQKYqu$PWRoiU!)))LKz?+ z690WIFikuJp!oj{a^zpPUzbX~zDH+A#Atw*op$x!wYP|m&9OIe7G3LE4aJutZzNkx zCt0PNYvh1tB5WY6Ur}6D%wSw~t{{n0n`n#Or!)VH|C8k0>1?4RPxVj{h=Q9SgVNMM z8~aT*O!+dV7<$N}3A}MJVThXN`dN&-{$O7#gZ*1&uu3uSHDRKXM8ysK#ZU1gOU~s( zr>r5}G$h@kK}M{4S}pO=IVE*3;pKh%OZ7u#vJmZ14npTZ9D~|K;iB2N==DKnf&{(E zmUvQ5O~TduCm5SZiDTcNx0~NFIpXMubh0@m_AU*xhjNDAML^h^=y3`raGymhZBs|| z6zv8yD~(FW)SS!!E;&$LyL`{&f!V7m4h1hkX)$GCfSC*_S@+T@w$GBPHjOfsDtXE3 z=wYeID09ua06Iu3&wPUnX-HHdc{33~@Q+(uMIK1QBN@2^T}mB{ZYn%{rb43y84AR5 z)(TW^g&Xyt7N8w5C~E!T%X(o728*h*jwD;ssak|C28x>jIwWo>EHhQqq0oo53ySlsj4@O3xjzG8Re%oJmpzWyYJB`TKW0E&m zmVd=UrjehI_K&LRoSV9u0N%!s7VWd#bk;8f~ zDD{Kufc9kNikak`1&~(IdfXX{_j@(r#8*%^5wQq1^J`O43fPx8zb7l)3yU0u#v}xz zNwdQIRiKtc**r;=e@P+TM5LbM^;p@<_7K!I+`DPEP|@`W7p*ab0pZc{peaae`}2+g z0U@8~p8g{$JM}vpep-{>Et}xj@v$#!5|NSH2X1&n&24&lBo17ciXQO@A4(Nr1Tl|?QKZ1v2Bi07Bmc*X_`YCkh(wqf6$6cX1 z`{ax5(z9j-;Sw;cVIvKtx?^{(xAxBxKnU~=9u&foSA(brM4dyI<_=y!9wNwJN@$5@ zBJ?oj!yd?waHw;9RYl@m<<=4}95XGVT{5DdkJD({13^#!fHir(Kd?#V7*=Lj3)FR7 z<^k-s%AI$bR1V#Wub1dL!O4IH$e8M-e~!L3wrfZ#U%^=>RhfWS&m#t^^)f}hsu1~# z^AWjTzQ3C@Yv2a`^A$?;yjDK1;5#l;{3zo! zsu%0~t_FPnoeAp;$AqfWk3b5$*&hSke(h{l9Wx?6jjVI-om|4*o_7I1k0qf14c{8w zp59~v-GY2PKl*w;AA7iXUY>H1R-ip~cMzL{$I|pR%Ktd%?L-6hsj6L7L2rOG3#%`~g`ykzNVB z2p+xNiqUPCj-e zSix7+Ww{BL@JVebpJP9Z`eh56-sqJQ_M!^<1+B(r6|g?yJ6WLJ2J{_ zir+Cq8td#C#og&AStWQHjwws~#<}Cwh9)fz8p~J4G))qn_}}r^mOBzEF9MZu(MZPx zdVl}c5#~wZoL0WvsEtJ{2ok5TM``u=oz(zh|0HI1MR+etr1peurAWW&GLEX z+JTu#!#Dx(*$`~-Url7W&XD zf?~w7K`ds2^0Z|#HB){kpKrAM_}mtcna>CRQD@)nKJd?;Jk+vhR4Cyz<2jTtT=={K zX=2^HEproi>rSyC-=wI+h$8A?aj*2d=~XDGdfk>+;5qouqeCjT%O9wqfr(MK@_$aG z`Y&5lLQ8MN&@BONJwGwNFlmW@d=bY9EWB8js!-n0Pev3~GD{&keiX?mj=(vI_}lCM zsU#7nUFuXpNMgi@8mP!t2oNo@?cL9Q#b>hcubZa<=1!<$kMWvd)e)}Vz9Gjh|I%&B z4N9)KGw|%sFAGX#QTmGoJ|R->taHpEEgQLi1<4!0%RDrz!yH20KUaQK|MYi&JAMBa z`i~J4UcwGdNTU0LgV-*<#}P;=1vuN+7DHX6&! zYG@)lFdA_Xsrsi<4wKBP7?c|w;cvAGL| zn}4icQD+xw@qq}^u=8$=-~c35Y%vw4w%`&`_Fsdsod7~L<{c~vGkl)BARpFb0?>EN=0AcOgjMCsZ zaL-fZQ}QN$f$P8;MQvEtIq?@0@LRob%|bMn8O>^=KO``qI9!4xUS4OTJB!vW9O(oB zo(AJ#S?m-|WgK1@%C5KrdJnP#+xE#R%Bfzc2C8(cP16Ja#Y~2}e%?a6pm7cj&&CVb zKMAdCSV_j$>c~D~UZci^(D~G1*K0!E2ON$5{;kLQG1>7C@2&@{$|flXVu@gtN`e`= zf;qpLnfp6>;cfa`VX)#)&64ke=gI?s80nz$3)ok@CF+B?y%aQ!E$oHb63GKW4G0vm z$TMYJJhK)RX+pDq$L9MDWR{5$L6f@LsRzLy=6j^Jg|a9YU`Ns8v?xsv&5enf@X`5gO_y zravnV!BluUXe0}&!LRo8mw9M)3uZ=HbZmBTV7t=LNd-gA*u|5-$u~A@Oub%lN;P!` zA448~?W{}eupN~9r&FyG7(WT6oE~1;rJt}6k5^|UTcT;hf)BuDQn@vUqW8F8_eI{G{OJ!8UABw1`mMvz; z3E#pAeY71<{oSEdtKC%;VA4F}A8oz;%jP`XR)jZN4#VkG{U?lT=`0lpW4u^?(AX5Y z&2lKbmEdmBL2{G$nl$eW&_AY(wT=YdHq%kuKd2sxamB2%pJh4!mFYNStI=70kT?M6 z(Fy(5tz##T$DE76Xl2}!my=_375p*m%<-kL=Q}I^OM+jG0)3QCzth3NBq+g79A`1~ z59PlrtN=JEOZIaF)i!t`G74FyFOXWDCQi%U*L4d~=N)|*q2I4{$AKsn z{}`NhWG|I-53j?-fcAQ@os#ibZ0JI_158t*XcvDyb+>&HdnCTVuAj0uwgP0!>sOR! z?wC?1b!>5w0>z|+923w;{II(INL8yt3IWI zIu7$$bHS+?HKxHpUc|cz{J^~r$!l(oKA+L3>Sv`58s7Ex3uu9Y!$}o7i zxAsIKTFEPkT?0J!8eyAdukJ~Htj)?!%rELLKD^C|`5t@{f#U1wZE7UkgwmZWExP== zl0^q%q4-{uAeuL4ZsJMtAeVb7)?gtmW+UtjN0@B|h;v7n+9WoujR0_de@6jVgM9Ef zJS+2yKX|ZWE0DvsA@1;iHq5bA%wD@o?FxJw-rXT6Fq|805mFd@!NVbjHQf|qs8l|< zLMsf6RVm!O0J;pn@xNT^S1y>94vNI<`wlkI>pyg({Kk$MG$6VxNNs4fNp8W=f|~Kn-jY;7IVbV{;z|hob?ZCMROx!sC<5q}X*Bq+t*wqEWWAjxok!byjVq(e z{dWB{(NAvYG|}wxhhUe7KuvJ+yIT1sAGcvH#@I`L^VU#6`%Wpf!TBd|R3X+D3qIkW zp@~t2Vp8^*R52vNLIYwSm}zG|k%*101djz^HsT&=NvOuE)@w?gT?#Hr2K9Vjaipgh zJPpJrmld4XI@YXPZ}mNsV~njnJ9LTRQ5Kgwwx+53A2-vxpOs+I(=16qncd)s$KWoK zjyd=2voi$15I&9y)72T0I6}bcZ`drF3;cER1+CSiWsas)Ra=*LIQB6dP+^hZ#q$h| z*#s&|2;b7Z7oy4?%GR>H=Z7+>6K{eh0(ty>U85f^u-?t{Y2D6ePy?8*bn0nSg z|An{YnY$uPJhmpLS~r4rbrd7K^(q)do%rZSD2AT0L%E`l8*PwvX#~WM6h{3z(M}5d z@}_)4E1pA+V&=@W?rbH^qE#?ik#M<2LyyM*$eiS+yB0xVhSip1m{dnDUE7kYoN>uy zybOUd4>6_KCd{9I#o@(tNHq3!05Nj((dq|?b~ur0o8zZD6Ntj|NZG1Cu3#OeN53qN zyK~c7N@MYJY}fuxXMn_>9!wX>BxU!9d5(p*tal?rcYE^P}aN7V~2jQ20<5XZ-w z=^kR?li;kxz23#OBDvsZsaR`uicvWlwoN8$h}vUEXkhgl2m9Mh9uKDQeeZ>z@xYOo06D?U6vMl+G%5bDq=7gygdwzqqY1KMx zurf^?ajSh11lS;nUSl=9lSpz9l8OZF+a)&89G-tdtZCiaDZRSMB*wv=47+QBR<(Rk zVG&dlifeM{lrTTejl;yizlaBLIzVj~FBMFYy$`<@pzXd_gY+ASLjo+YvqeXWz2grTqSLC8gfm0wNa$@|YS_c9zHO|O5H+#pE#bHfh* zWK}T_-AZSbizqg*eShV88gEa|9f}?WT%#1`0n#IIeGB z9-I1mC{4d|5)zt;Rwum}86w206)ET9TCh?Mfsw5P4z(Q>#@`11hpVSIVGCj_)URf6 zEBt@XThA1lZn8ZvQ1<(Ho8-s+3~v2iogYQT(OMOUi1jyiA!fvyGVRuUo%2#QWZ^}V zi2KaRpkB@;`vbR(9>0kIquBLxv2=H<2&*oF;XSl)2QPM~Ash5e;%buaNT9UR)O3XsKNHr#ciAa9t>0Z~Joe=+pCf4CYYCySk?a*uUB?@wV??BUI`u zh$;^jd*0oMd~R-QStP%e_N?gdWZgwaqBal4{k5Zugr?u(4{`0fqxZ=fP{+fk<7oA4 z!l~++hUczXncmwyjj^sMDlYs9lDE2$$Ru%*Mp;;6@AMvpHsC3kw?^%Gd7eO~%G=AT zD#yAi0ZPuuV0GUNsJ@CgDV2c&s~33LBAz|Hj-y0gICAInJOlP^j*?5(*9*1S?Oq&i zdvxV`ft2$4`z@<^FJ|wx_S89qAH-k3WWPpf>TFX5fz=55tGK!e0`YjM+S_ zg@D84Kg1mY+6~UD3t8{b@P!^~iI-GBlw=DlJH_fV2SzU;ap1b?`K8_YmVxUdW{^jh zCSC1^EC<@w0%sh}-)}!i!}C_H^yLN#ayL)t65*j2%r&9xK#1jd_}Ef{1B#gxnVSPQ z9mtVdVWdhI`LL>!z8x;Y@Sy-Z8 zIj(hWv!j6ho>g^3WLQ~=Xhb^65V;?0=!pZNMmn@jAmK}p&P z>S`LAsung?MFxsX7x#8`R(4jM)1#Cn4wV72&{tOq>dT&Ojbjd5Dh%swP%W4nE4-a) z`q)oBpn?V9zpYx{3Ck_UQQVhsD_z-dA-g6A1COMBU^K{lhCY^~cMs6lA>f9S1x;L2 z8Xbs|@bHB&2Ik10lxLfHz26|Xga=qM+Z$J$oUthbC99s$#Z)%;Y&}UgFk8zosDA^L z-FXV_y9Y<&)6Yz>Fy1f-1p{(J=zhbjW|eS}2p95@#Tm#yP-q9OQNZu_X?Yd-dy=uW zyPgTOq~V~gFVv+n%CdADb^O^0yj=w=bNEEDn7b1FWHZuM$s%_)4JXY82tbOWyno0D zN68Fg{yo%y-?YwJ_lzP7iCX2IULWW}_Eg}%Il4cQ z@)IHo<@KnBF9`DFI@>sR!jS}=*$E5J9oE6PQF93PrgGML+H)NBz+no!ofX`7juSml zLPpHQ1?5w|0zoL6s`V+e;k=FlyG6TeL+FW&EY~R`zuhz@*QtO**PrNz%1!_-%X~m_ z${Rukmv`9gYM2DlTmKVS@q|&agJF~m+lI>cvaMW(BKPw30jxpWL#WeG&1NIqDCaIt zY|DjBOzE$8GjFMM4FWz|8`F1_nau>;52E0nK>$-#selZD6+!Sq_%qUmW)O8qRDK{M zM%oCO5z+X-A-m&WQ^m}BQ)NI)k4iiAz9j~CB6o-UTNR|S@Gh=WSvL!XT*E``jO10M zk^9!J^r?P-(_X5|(o()JoqJ`3;Fc|!$367hgr3;|;=ML!G&-<&0OZhbRXR6_H8tdZ zqD11sZ-CB+GxP(xGqg;*Hork9bEOxo*Ro2mmQ0=BL#}CT7XbApi zegTYenpA^0g7G7DkO%;!w#91WW`pyMXmrB4I?nPlenELk=xQrav`@hv=3224K5G)~ zZJ<2cyRovORn=54qwqjlm}01+if$uL z9JYaS`I%${7#FuLtf^%e-yhTX+#vjTqI)17k6Ycnh71eEw&#HII~;Ybd&_{i=>CY3 zZYax8n7JSBKcSzOp{bp;^}GM(3-=M|i(VfNoe>D>v~>I}_$|$kW)JsGg`hjJP65N- zh9MA1pdo}el|)(YFvbIi)KBr3-nMfllGfNjN=8;yBxa<+?#SxP32N9cCeLbuH41KS z@*(3`!0N{j-}z5KRcw~n$V)}op?b}=oZo4>^5BP(JPO>Ahb=C^9SNufEq)9|7z0x0gk5_JawPmUpp55@y)h2N3CQ zh8B`5i*r3@>b|<-n&X#UD)}zclOZW<{ndIg2}iO!`iW|+-tmCHq2AGt0kO>3gGHV~ z3x{`#lcER^*?sWRt43>|AUoxVxzG1`Ad*Y>+G%lcSx2W^zH;pnXl}Zt_T9~A`Z|iY zZ=u>e@BQ7pudt9oeey3s!ZD7cUc)%_J}cuilda&Ebwc4)*5EN{ukQ(6BXYJz1L2Cyhht)u3md9QS%{H|YO83cbZW*; z27ePBZ<*&7*`G@Z6;16k)R*$08mk6aA!y=Q=Cjf-$mng#w?`&5Ws9P@mPg~a7l|yp zy@CXx&?>q8v??nkq$edyV#m1C-DsVcgk+yu1-`8cd`>)DgZoFevtB2i7rK#UrX~ID znI8lMrr`v;ETFKpZGQc32AtaE_ap%eloA#cd5~B=RabCxx7E-XDd}xpCfNW@paq3e zi4TQNjzcfk;&zsSgyc$ZnbGb3l~F4EsL*;jPxwqzVbRv(vKfIdtirjA(Bep1{vX|H z3XOO758@Bp;CttG5TGNDjOZ$*MTK*(Z8_6+#i)1nTu*+49NNqR1i%Wv?*)e)0vobr zFz9TmvO(yl>4nVPPTu{u%51<@unCJhJwd)*D-isK;Tn5xz&jZOLS-@|Hvycw=NP(&r*iRr;)s+WbG?2+C<5Jo7VNjU38dtlL z0;){oVrdRlS|kIWQwJnhqQi+hqKGUPtpUd)(puD+&|_#oWgo$zG1w}~*)L*J`Q?&n zKpWAGisAr94D+sHIuC#)GRLlkM1$!osLihxs5_nfd&i0MTC|?>X6mEBMvYamlB@^_vnQIDQm+aItOB zqikUO=wr5M?=2fy1joDW+pGbEWA#I|YWd#JuTYZk@14)w%`*U+)R({M*|R;zJ?K}A z_^8UMg$e}^nP<=AN$g)|R(>8I4>xnQMO>6czstqo1 z(Z<0)+p7>m(*%G)yHxnWZ8x3j<9Y9xa<}%Q)MOGqzWB7J4@*fi4V(5HdRb-6r^^?N z)HB{+2;cJpm2e95j&33skp4oakDa;OM)4=U>bcQ$ui-m?b8EXm8^R^81I|CEwGF`k;(ZMN<7fS?+aNh=@pWR9Cq1cwu z*ICXf$_WMl`WlWma5J;3?F;gacIXL<{eJl}l-vy5jt0~hQR(o7S!<(3x@F`;<}=6& z7W*((^3csI+yVxV2^y9$A~#JtyR^s*7L>H`$>V)FUL52oLKMTf&;jY8*%1h+YJhMBe#1f4Wy_ph#!tlHql~@GzF01+|)iF%#6osSN z)KQZqtCVk;1#P&-eIoP>)Mpf~YqzVx8%evE-n2>tzr&`5KWi??-C9NC9~gD@<_a>> zas`@WowLCJWy47*BHFKx^0l?*%f*RjusndH_~DLe0O?uw@ncSPxG`}1Gu`v zg6&ftKJPHG9rwsKF$~|*rmm0f2Huo$JU(v$lMelV`tcLRUwQBB2~33P6R$%a_tQlK zU;gyTj5?*cg8^`ct(44C`@(-n(MOvE7_w{|QM^zg5igYQ!xjtm|BARek5WkIYTLDi zTK5=kc_QLj26tdcB7&+9E3WREbizMi+Uz7t{HlY!O~p9G-Bfni{dHl-JpB%1b(Kba1fiXYckASm>`fz_YB0 zhKkrK#sKwIsH~B#@6D*#qz5@Vi9?A0uCS3uwS$#;5VwXkPu@%`L21)%l%{exZn=Yj zt$7 zK9&d8ou<{8v2v8EC|_-Rwgm^1&2K{Y-uT&7b5xk0d0u~Je-JXk+h2g~gRs20uf#Ov)E>sLQD=9(=u=#$(RkN(3 zv|Wk%vcL3X_ig)ms8p&)Fj+#b7N;&2_TwAM>K31ccdY$7_!K&m$K`F9HWh^Q6lcCf ze(M>-f8=js6c36{nPY~{aM5PDu3PUuyP2PJtcJq0?E83C_eluMnLLmtijS;Nhx^tM z_^14*!6Y^g1Z>jbNMtl0fd0E9>+8^dt~;Ync1XZ@Xpvx{xnhIp>tQs1gdxxpdeBDJ zOsPWrZcXO)$555{W9LGUzKr`4J9-|$l?L2a+H1B%4aGdRm&6HcYl;N23K`0m!5ik$iN!$D zy8~q7%GIA_5HS-HU_*(_8N7UBR+#_U&-iY?$pf-T2fY-4UeN`6cohf6Ffm}I^UF6g z#;tC7Ib3#powt9mlUvp8@Wt#JXG5#k{M7I91P$VM?%YtkN;NXkQ?X2l)GznrG6A>o zZJ9j1`_d>rZyccvUM7UY%0LQcjZ5Vfw+Qx)nkRxl`A)Y+fVsq~Vz7ZV+cOIhK6Nrn zSrS($BGf<~h6%!id>mLIGUjoLz5R+$M0`=hPHCKIRkd^0u7FKN{>=Q-sf&!BZtATp z({L6$EZ0R`Z<*v$n36GMg^}v4BE|jlL9%2K?DxP7KVEGlqh9coT+m~ww&&a;vNzQl zdGe(!dcF!Q;M64YlZ_FTmRq3e%{#fX$;lx9t=WEI@dy3EWwONH8u;DqAIBwvArj_p zu`n{9$~PYWklD#MjhTC(HM>PQB*_$$tnbX6DVwEV>XkMnLbyukPuad+q8H=LKrC>a ztx1#TlWdhc{z(T;q8xF1!x)jsbXK*x^bpEUTH(0dR^sS7^W&%QfkDCKlMsb>6prtPzhf1?XG^_Pv8&@qvKOF z>&gj)C;FF;u@sPog>*lwXTc6+5cb%oyOtQ&osB)j@(MF4!MfpvBC_83Iounk9uEEy zGA6_s(BXdwoc=KK^i||y@GwMZ9WGejT1VU30&E#vc)h&2H0%H`t^|BmfY>IW+HV7H zxgOAE)VWQ`tAu5r$A4F&lZy`8;x2S3kIi_&kpH}wA^N!M>DB<>KA6i|mVr)tqrIx# zvd?%(a2S5@bv%v@eVCskllCeZr(i#NlpX|g@|uwi4T+ehP}(zHGfABxuFXb9MR-}a z07qUY*e~p-f%FWU=v&vHp)QWD{^!^1tvTQY>FW~s4s0^aw}a{OrHRJ(S-~t<I6e7LSy%|h>1Cfd)KKSV(&ArM0(DYwC5r{A8wRp@3{4VV6EA5BAQq- z2iw;-gJH}e?S9EX@PH|#Cf!n%0zMzd+naa=VxiVLsD=!z;F6fAH`Wij zF2L|f8#u>F!DRe8s_S6>=xy>WRkzQa#j%6q(#J@^m=}dyNtdB}o^KVK5npum@Ek() z4Ygu(A(TOciq@yY3$y zeaDt4=aE09p8`Ra-@6yKbjZ1CrTolMKM8Mcz`3e+ zp>&xG-1?9F6J0Xzri#Jr(5Ip$E?I&~u9_B$>Usq^n*Gd; z>rapntVamI*LH9;_7F%0d5 z7;dch;*+0lM>9ZnsO2y~_C2M{fqO#$i2s1NZ;$FyqU+*U{G(5IG zpxLnP(llz?Wiv6eO4YC!Ztl1IE@X?#b&&t!NY1nyfa+_EDm75=H}Jg&dO=HR;20}m zfl5(v&ES6qK*p9aQPo{6S89&x;~H`um}n%6GXk1!Te`I&PD9+Yovy$~e(+=g%HYtA zL{RkH-+Q4}P25g=>cu)Np+$o$%k~EKh&{G!(t_^$e zcbHRSD1tuZOgZHr$fs6UQu=l1L9P07g8wtq4)-ELKM%}8>?urG&eVBfTq%NVld#`H zk^aG78n4Tf&}=#epbE1yqbREZRv`4w3b##EE4K3jC7UZsLF%_(LDk!SKG^Jc&C-9f zXX+s8>_cE+255!x9{#W~s`x>6PI4n<< zx3f&^I0!z~e(V5XbdS*W%ZtLLS7wbl~v)wa^rXFxu!@ zE|`+hviuE>+4R7Fs|{Jraz5Ze7CXV@1yD7YdKS2edt^O0daj9;92ZCY0H&1oXENU` zkSj3b^K!k`tkChgg&7S4YKk2Ne*5`+EI0Nh2hTZJS5FfJ@cn^5`4~1kNzpNIuU&H- zL#nezVd+0bA0W`v#-kjB69tu&J#)<%=~FBJ%qSqTw3zzQtSnLiRSl9JwgTBn04%Uc1szb zxm4Osk^M63!aKo4)ut~pPZv;Qoc{5pP05M>O~W~;S*Nzan&c;}b^6HKktm@fHyW9Y zb4WoHsxj^hYw^4RtQmi-r?J^t4rNId_L|X<)E(1ZibP84p=rI)rEyIm@Jq_HsA%UK zq$jD}gMg&(Iw>QihMY9H;7?8DTP&mNx1vSS_pI4RB67*-4F){n724xudJI{1;z3x- zx{=ychsCoabM&y~VlA1e`4@A&Mz?Y--+ofEj9-;!fFt1mVhz9$k%FX%eK+N5==}xE z8NnX*inLtYs$L0FErTEZHZ>r@Lt0GH9do6p<8sYlPCM{clk0z0#sbY%>&u(rwu$mH zLCVY`;eY3egj4$N0qy0~)OTAeJBi;IromE&GeAKGE*}#@QDmWZ z0$Fm2O4cvyQG{DXyWa)MUulgaet`@LbN*;YWX@`mHX$V`Eg4-wh?C(n+2G5eu!AKl zl7>F)#nGUAR~o^%2z7k9aM^&ry0^=m@7!437oa`^M9hZlM8@5-Us$&ZBkHDaUhI>n z*~zuoX9#!j@%lk1l&o_EocSh6sZCdGv0CFha?Q-wzWpArQy0#RE=RD*GeTr~rS^aZ zV%;9-1?^xR3jmYM_2^R{I2L)Z7VM^jL3us@bi;vG+%r2x{fF0Qy z;$&9@7)^0z<89vWmuV`%QR!m(aH0_|O$KmlJ6 z-TF$^Q-^P3moaC{zQ!1{8?raecGb9$>RhHxdRYrvqb`v0P|p^e3$uX~H`zg6>*L~p zKL;kEc*Lg|#MXDw@k);?8+l4q!MnS=DDWLmm z&wdPkG5Y5^_)(87YW;#ka?WyKU^M~)a1gWOE02qEbt3-zbU~&*=lE|?s0DYS`qb2GKKug1Mq50m!Qvuq7ndH{3>EwbX&UOWJ8Y$vi7(Oa1 znsC|wdC|iPol|Ht$9J5_mGuHrxmT_@VGEPPfUbz2#(IV3o|@KId<}zF`lX3+tE6z1S9(kM5OVooShPp0N`*J*`wyJZIQaz0y(h_O@0ZSS|-RsLj>}y5=JT(}-w7a>9 z<0!qB4CNA}NRznPD-hJ>4k4X2CkW2gJh-?%W*{A8AvMsV{eUh`?N>Q9X&Wxp{e*LW zF>Z7(`-73Oy}q$NOsI({**VZoWt0j@pqW<|K=m7sJU)$;aBVG+XmQbM4xQM~b{uL7 zni_r*-;_R!Q^R41F|hmzP&kdD>+5a|ZlL0$r4rb9Vc%XSO&z4)j8tW{;g)hzvwB-w z2hq$%`TGfj{suz$lp*f|F54AWV%lCl~balS}JCbVUlkR_OEx_<9NnB6Hrs zS(V@*|48a;UwH$pBdxoNOELU4-AW(MciH&QI1%Q1hCy(jtCFT=K>DXS^ohg(^3S{SWXPMWXrdwE+qc&d40SUboNKXccoY&0X>&wHf^cKYgft z1HY(XYSHol7-fx^;9!YAop&l8^v5D&G6pYTw7xk^uFJPVDeQen5Z$;0x-DiOLzqX# zL$*Q}|3c3dte2`I>x$$1id%8PvSpdv>F1xU)}1-QxX8H^fY=y{6x)JD0uYsT!|^>0 z9dDv89*ozBL=sc1&_NmJYuP(%wRh6Iqpf{9LMUef>0qp*Hl!Cc)FxV5PxRvu|CG;~ z0?jz<*ET5XZy%y&p9r^l&p`w_@T*6hP;isT(W$Jq?CPqdb55+yNUOf?!|J(ZT~07% zj`#spm*`0dz!PGedytgcCeSc^354U;&ROyw1~>t7V#A%7VVL|_@eVTz6gN{ zAl8rEyixgc^mnm#k1pV#wp@z!7Ns5VsArGe=IKG~*Bk?@3I_9ywT}5iXfRpeDUo17 zv-|0@0q>><`NV9P3!y5i)cL=?Py>uYQO`E}sqv>RGWJ6$33Wb*s#kXKk=woNYN}?C zt4}bl-38iS`vXRE1z7_BnhL#oJcZsBt%haZMQ)UWh{&OgRG{h8k(C)vpH-*WLifoG zyyF>-p?@L%_{);&KeINnMrQh;k`FafwgNIK3W)UdNn9KKxCC&dhZ*}-Gm1+Op|#nK zH^t#mW#?r6p_^zu%sQCVi>b;AF}P~=HN`2Y8(kO(fL#3${bi`i9q_PIzSZF|Frty{ zV;i}ngRPftaC#4^?SDZBE5?0!=Os~6mad~|a z1Muxp*p!nfH%Fdz@vBTh!56 z9m4qja6Q>r2J?hh+*Bfe4O|`isulYJ?A?5;zJJuBH`GA5PxgfM+H>ZNU0+9m)jqWG z>RL1-u3}4YeI!2YMw98ZS}OnGK^aq*b`Z~-lgZ-$5_2Fqh2pN?%W(Az!Ph-WWg=Z^##|wo;`zPQ>rJ&xW;#sCzhXV*lat5iy-0 zHCw`ynj0=sWZga8gwe-AM%JSP$R#R=`h&#HRVMAS+OT-QXzj2OTixrgXhl3iR0Bo& z{^_t|;=02N;%^QC<+WY_;gH{--8Rz_q~?JtqtsB+!HKE!R3*3XmYlkEXWbXp3B>Qh zfu5__F8Wxbh&#ec9uqMKRK-hytr@ssi#?zmzvsp^2M-sG+1cFGt-Z$su-iLoX=*N< zE+M1r;J-pkv+kq3`PQeu0p1NoHFzu^X+SaZ%s3iJ(v3m*8~L2}GjOYQz^pX5WHU5JDh(vbiVk1%DeG;duOeQa*9VkO)u zqp0UBXqY_`;owqF4}^cN5rk)rx#>x>jem7UcY5EN$(xnK%bAOY9v+r1fUWgkPG;ay zEExBU`5O2nx(zb6i;~^T9n@vGD=e?z%|&z$MIO53aw`bT(=b#CESEH2rfHO9P`DRd z-V<1w{C*@`jq($tpG#gg;8?t3mG!)+D*FdkG6W7ckF^8_Dk*+qF6H!oz*8`mV~*QY~c>N4@jWL+2D z-xLo5=-;em?utZ!6;DJ_ZYpl93zO8TiLo75JEu-ac{NJicWSG~Fs-U#6HXkQ!ymu8 zS48;FyG2^omYPV1uoy%S>fD#7V>Bs8P*vHKC{GsIj5d^aK`l@88{Vpq;>t{K#WDRl;+|J0TsgNm6c?D!0<5! z`azt-4MjM>%^Q zs3JqmK7vXd-YGU3*U^mjg)x5)#kGrLl2yuK5tzix^+&E{gQOZ9UTlJZx4@fz9dxO| z@(%Qav3XB+?+&O7xnAi2^YfHotG|qp;LqU+<)|3I>=VtDdAr6$kqzB0Ov&vs+Fxkf z2g<$s{VZpERgzIR45RPz!?fF}Zm^=AHV{*PJ1gnUm2id;@il49Dcdj-?sI~+nW>sG{kwV3y(tj7bXh1;_t%91d(v?@VJ1s z7@Oj76spS175?(>6snj$EQZoX?aVF>n`trzddaIl6mOqVk6DDo!L8{fuT*LdAC5em zqD0*2C1{H_;WQ^b-Bx`?*p*N*ze~tq!{bv@VS!>RCd;KXeVt^?9k5zeS@Z-a9Co>Y zxqz@TO1wo=Q+l`|Ug8P@=6pKCwZTpp;ySH=p8#_4UDh8Awl19*Ga1bE6@#@%9L9K8atYdXb;O*Z?(?fFvc+QspRRj*<0Z$9z&ex|C`$QgQ#`ZFw>#n2Y_ z<&IGoQR)_1oVzh>7>Hbwy)c!If`<)+=2DMX$khZDM#fubrF~Hxa)K!bIR(DXDs!Rc z$&kn>M%*idEIs5Yw9}+)iz6FA9e7lw(KxYWXAzEF=S@%-M{T6SwevR9_H~uufNxXd z$LKy)nw&(RoE~lEbc1JfF4f38-WfdnrX~0M*MW$Xb)3~Fsx&D2Z zi(y%j9xrzCwwm?Jxk|Gtd&-gFR>etG13DKwipJ-@_eAD~i@1bS*-RavRM|V0_;Vn; zzfQ%Jh!ewQ@&wgt6<=Em*AQEmRrIv&A8?S-;{`BQPPp=iiM>nKk9;gI?O2U z{oh|ZRxXO5hGtdWgP9 z8NoZd3pAym@BodH%)$pS;MtaViXM}FRscE5+nH<5(r!-fG?Z6;QJrBqOri_rt-bwe z5^RCsO)c6i!@^JGwdBR0n^drL@=y-6ida$+HlQ?l^Z}O}lVl3CJg6C|hYf&L`0+2D zCz0g_QCErXv=!NFm8pk0Z?~DAR?X@+Q@5=-QU~S--07tgx|RZLzWpTb@n=~8)9BKJ zg7xwl(iLGmlvqO}O{SJS?Yi&m70^P9C7+Er(ouqSG@fTQv}f%7q{3KN%nH*L1cy(( zfk7TYW=_+ikO%W3$@UA)g0u4q)}8 z@IYym6POXsypz{AFZn0{RhtzEvu!# zzvEZy3UF>L6l;1}*Rkgre-Fb7{fPxEBiq!;AfgKjS03CBg;!G9;ie!*p5O?`zQ}I|6!DThMvUO=Lg=TeQu5p|Y1G-sh zOD#h@7f9#35T*!az75jP9(Bf%+kV+HHn1yW@otEFSv(6yQaQzMw>T?e!w=P#>6$Yc ztG+~BtyL{oES)a`p_UAf$r1jSeKgWdeq;N;IT-l4-d6dwXErz&r&a1$ZJ zq&mr*zvZxae7$($M+`JC<3e=fd6&A^4r?N@ck&YA^GFjNvQa{@8O)(sMvb_~q*09RNH~al?o7FHb#ETQ+jht|hM-ghdDoG?jO944I~P4tWMU z(c!Yy%j@Cg{?O^eFE~xr3n#@0ZVZScta625Ec4&ZhlrcR_7Ghu^3e;9*f+f2NQzCQ zFk2Lt&TZ2Ol&ACA9%Xw5kL&>Mpij0c?z~y^akQ%V0kskVw(!3c!pQRLZA}m8j^(CN zaX*i))|H1pPEUu(>-0c*)~#~l>ke2224#ruL*IWJe8v^BeP$EAGd@O~re~Q=9_9EH z;`AkZtD5%_8z5{VOXK#gH{~dwij@dkGZ~sxHWmIsz+!TZ9ha)vnB!A4dl>{ZD`aYr zV+(UnGCN8f_WpS+`oKl|VU zs3<7-hWwwh@c&Cl@0i8T0kc;MzmcR{yU5#NqvyWl*nYA`6uW1VqCP1cdqjiGcp!5&3)n9!2o~2R}R1 z`*hxrO!@}onoGC0t-CaI^|c9_03>Lx`KfQdP4kOUZk$m*q6J!@aj zU(3`AnFoM!V;pU%=K$qQq`;1g%E~@gi5G(v`_Og&!#;B(VFFP9W1kKF7yIlM;A+DZ zeZk!&)8_sE#Xj@X%_%lZjATP6R?&~*s$_6AZN0scdX4ViAcOzP)Ss3;AN9HUScdCqLALhDH? zuTG-}7-LTTDJP%}mpNjRC(>4p63V$MnbJg7RoYZd_xqbaTK#@3LR?X;IR=fRz3~E7 zcUqB}_HhAfTn#d*Ly2r!mSd*zBCnnFJ({7`i1O&?z`c>Y7j0uCjyDE!WJdRxD(`sjDGRVZ?k+EgRHo^Ul4ue!i*dv&!`KvZYt2ATzSD4V(p9xH7lkAmfSNp>JN zSZnc|8d-E?JKXlch@%f};Y7;19wu4Vp84*DG@i+ZIJxQwZ|bnhKd_jjqAP`yVTL+5 zs84w^L>P_f_++YAUb|%dA*YCV;DE%Q0&sZGJnKSSCM9%+fx)-Ei|E2ALu-I@p}PJE zz#YjXu6{0_E@uSj*?Uc*!R(-;(q>JkF&q=v{B-O32=mSngTIq3b*?<7V;r8ze&Fg7 z(D`L>%5zVF|E+%a70lf)tK{zlBto%G0p*q1b*(6i6JT(=<&9?Nj~vyol^|1yH*TUZ7avcqJr^w{6CgyzA z@v8o$u(0)x!e&_NFg(^QJ5{Qvqtui$aG4Iwr%T;Z5ti4W7~GpQOCB?(dqr|vCr6hod3UBW7Wljx~PSk zv*wKD2#f)+2GWsxlZPm!k^fchr7p0jV_uguwDi1Qvdfc7nD}<-;fVG86i4t1r)g#u z0Sff-DQp>Q)Wl05r!=jez$;k@1LZ19rIrg2n{s7|^+AH8F-#}R1VMArL}o!gFeApt z;YXJVLg))db!~-;;ssrB%pBtsD(G^y?n(%q39vk3egI)}!ZVz&+;t09-;?ZP7n)RF zh*3@vW28sKZfVLQU9c8_^pm`ugCtQwHU#1xG-gn1ky=}WYi=SIN$Zo?2Qpy+oRpIo z@tjyyLd8?LN!ja=Aq{nzP~6RVvQQi+$HCiG(7wj1txKhwHneZ`XWdAn!PTU5s-yzz zu=eOR)b9}XY;>G} zGtL4%@mXV=+CYZVl6t2wNEpJvfL)F=u*@rX(kAi;(VJQZGJ#XjArQ4b{JHHQ!{NQ3 z%?zKo5vjxjksGu4vcxeE)sQzLgFrDpt=`!fuSAYIb5m+5_#RIB;pXulB-J@dA-f}w zLK#Wr($J~o4rDPzB!=>8LQU+8#tWz!U%Y_QVo1`IW-~XZAy_;F7;9e*a_X?pU7*%ilj1{5;eVIvPV$TUnNXB?B@m3H#k=Y`& z*_nQ}l}A=f`b4eLa}gd|euJd`pQFX3q_L?*@ZvBhVyj6nGGU4;XYB*+HL20;XXp!K zDLZs@s+42f9(9AT?wXmwbHH-}Qt+EH!)rq@39Y0xg^0MQOF>>i_5C@EtE)akv!OKO z`wq`~dxuSTwlH9WURAv*dxiO0D&7#ui8GSAQ}z6P@ujJx1aKM(IOcd zIIm(_3cNm2C|5J3^C>bqW`h|1Za-NBRQx89DQoiJ@~7NNy6iTH@e_17h?#R1G^Tk*=W>U*;OcMML!_K z1q$lMUT8~B`^ka_5-2A&OMJPQmJ+U^)99S1Yv*U*R^QE_?*tXx`IfwSpS&gO}PcQ%22EsJ(s4F@LGX|va&h;F~UH{k)$Fq7^#mWLt4u;18Yolk&y~_jvoVc9T@ZKKMH?yn`TcIwuGKdGhnSmF>P@2yVUx&1qgf z#qQB|80RaxJ5UOrf&2x!JUW-rhCKL!NUWIncQx{YyX(h_6qzDb9}cD@k-i&GHA^N> zS8_z}j$Cl1sOFyb&xUmEd=z5mg7=wwA(1KQhIJ<-h(ftC)@BFYjMw#b4Z`cr$MYJ% zB4)LBCP@-L=M?RrQHF-;J557CJ`12_m?2#3C*>LWUupfA*pZd5)`JfX%cWN2Yo(x3 z4(~gzu84flhqaR=ENfh_}ks>OvlF z4M$V~e+?9|zUHr^drSRJmONpTF+Cm;>IHB|q;i-=0>yyn=uJ+11P;>gDD4c6-;( z>Bhh!t|r>*7U$}w>C)HZK4n?121vYE*PwpHE;82|02&L9vS}vE?i-?zz86QGA1MEF0CynB z=O4AJY8HK=`MyQaX5arhx8N-|zV3O=p3PT*pzEV-r z2gLe5oD$1O=e&q2%+0MeJVl}*}8 z#hO^@A}ohHvY3zz(on8eC}@D9IOr6Xlem^^dSt+K1**I_atG`DoLylweY2IpY`E)X zWRy~^%d`8zxJTii-)7FAAveI%q|GCoMAYLSr0niboP6V>xmsFBM}|P6;%#_mZy@(r z)afVZTgMf+;K-SKfd2cH#JqRM_a^n68(z=opcDbHzgxV9)cUZ&KjLL=g9E3z^dR4l zSN=$e_IVBgDxU53gfZS8_x%TcdscL}%q?{gkH|=iTZFopnqF=$L$=Dl;mW(HNFxu* zQC0a~32#uqRQ^t~ZZnIa&KWdCK?H&WWT8KSKWJR8%~ z)@0)H0H4p2RSl#M?|s`RkFLF(%_!?!2cyuTgw2XIbLD3cdN5(G=!*f=e+;_3KacNU zhx)O|7tB9{&A~mq3Gt0PgpLQS8Nc0nY~u_19Q|9km6M0j0&368KYAAE@Mccg&qEkd zHiz)V7c4mJO7B7V9P;6C!Ls>beSQd+j7G{_A1A`g&_mSwo_NRWY915@ z*=F6iGqR!78bAG==wgS=GMiQwpZzz#4nP3~6_V*`v;bBr;@Tz#&J!Bg{9plzCzg1} z<5W`A@8_;h%K`Zo8v61%^D!yKkrK<$uB%-qpDWKkaITGMTRpjia8{tb_$ z6I8KBwexC7XsT<<3HdQUNuG^LZ|2|kWHtsW3X&&kxFj4~=LeJ~Pap@#2SedkeKW^a zX|JQ&%v{~z{ob!>GUDzKzt|NCxl!NxoR9EE?Nrx_m2>d-bDENKLQswoG)&HhEvmv` z0R3RJsNK{h|GA#DvtrT6AeX^ z+Q#DeKC-UkhhcM28s&1dNb{!zucL}(tT7>TH;k$^8nnQt1qyy-IaU5a-!z*M!ib_j z78B1(fpV86PD`DPF#8%;b0n-7QskZ?@m}7vxq*fZou|S=o|id;Psp`WCi>#6g~VdY z;*Qbz;A(NRK{pSgn0nei%~4$~thiMXX;jm2q-L?WLA5~Q`pu0_F8}+hhm$=|ho8qI z*dNH1u(-e)6-%lTB9JRa`i+{HB#@myXEk1q_3!3InvnPF9%b^nsWH0)O$H#bVgA=7 zcF9TeQ&^kY?4Se6rnbo+Rr{!dV))RqDJKYN3#>E(FjvspC0F3u>t?l4tu>dR5!J*4 ziQkBJocMM655dKF2yYbW8&Sh*6OS61QjfH)fek?aoi}C7k|^GlL$x%f-y)HPku#&2 z^m}Q(!K%F_d)YF&DL;3gs&OazNQVHlW_77Z$U-rYb#P>LCy^Zn%;S_r!>fM5WmA3* zo^1Hc_C(=koq?kfp5%fSz5*^CMtbIvckZ*Ke(c20kgSXs#_dPlo)e)z@LmzwT& z??M5M=e*BoesbBY{;IP{#ysa%H)6|UmROGVDnB$95IWqj$`7#%zB${aXZ0N8#)hwL z@h*h?W$ktB!z>W0EeKxEapV3h4C(^VjM{P1e>vDtNA?;nFAix~Zl>Y8R4sKznAE^< z)yY17=w^IBQQgpVa(+O{^iQDCJ-$K2JPZZOcQ&xMuk)C_+d096z6t5-l{^@{_Wos| zI*~-vbOloPW|cwh%~Jt8gugq4zdzlR%|G~rE^Y9T?MPH%)!2?*3mAA6Lw;PDdhz( zA<(44g7a-%x)Sz@iSiV362vG90oM{{JCT@vww%0!%7(P92fYZ4$Lt!IV(8iVT|lH} zSY%XZaC8O|m^8r)P_P7y4Nejm`zatFy9>InGpK%>iQ#!inRNlOi$3`R5I5CW8#vG( zqoiX!z{CAnX`z=qn^`U_oD?RH0dbQqIxh9o=4ybFbU$vYde6;^@meg|$WPd{E z7#vh{&mauJch0#vgotymYpJhbB4qk7b52o?hk5Tvq~V6lU>rQN(cR#WTJ?u8>`tHM zYHBk>7F!K=9h+C;Y_-`r0ndbTN*-)u-vTYdtlU*JIAXYVxgWf* zF5(m$&9ydDG&V?S;VjE#G}GW_xb%E8|DdU8Q|j>`b7l7VZuw{^0~N5ths*037LIWo zZccumwhC4EO5wb55) zpxl%u9B4@&u}%gQf%#0rqe~ou=yreg#5@gO*E|KmQ7V;K)IGIqsxSU%g9RC` z)(j-L7v+8p*NH#(T=0jic97`W{WWW-DGhQbenDo=syNYZqw~3n}$;f zk0qJ$;Pz|AAZ$ox#{r=V>x&HM7}~a%{fb^*Kiydq*!opy0wfF}PWL*3OMEhC#oeTB z^-#Z@zFet!d}257V49njdme!V6T`7%F!*eiXC8M|Dj{*QzhRtuaU0`oOkGgvK?z2D zL2bnMEJHXhd7b;5w+;M$GSwHmRpiVo_uQ{d3w{Txa^n?rkLIgc=2PNnIH_@fiy2Hd zHmYS~uN->;0c4NxLZrrS7-iOHS+~kvy!<22xaDauSj+CrV=XtW2y#ydbB;83stL(k zbW z?(})pM)pWBNZ*bK?}^V+`||UqcSmSC$Jtw@HVlHA{G6%BY@#w!!$wH5J>y9zmoFXe&J@)NyD9J6hOS$>Z85~nJl;=Y!)ZC8EM`@9>cWGqXEIEMB45}2M^GXJCVQ#T@84*Ea|^9jn%txMbD zmVDKcTg~*ksWtkO!q&p1T_|%MiU8qDl(0B(;s|}Ml|2r#t`S!|&if{~NpO?8OTE%q zCrh9qbgXElS?A4lEcUuM^CFfZZU5-egGC)XI?P}Tk;==-fkO!JM-jA^z^jk5TPfHK z7P#5&8xzXQb}+hw*n{qKje&G{P^KsR%?aaW=J)Z0>%POQ;dK?*doP*5zQF})s73Q$ z{Jx$Yccxz9hVmF#fL%U3;a4_JD1rsa#-6}4?=SqjNv;*fYJJemqpjdYNr}%Jk9^{n zkvlHe5uWXZF}SxwpewRk{9i(W zZu=h{ljbD;icnAfOAcYe?`4G4AQiD})vcDr*E@+(P1b=6+xQMVWAq{wDeaq|b{!h|g`P&> z{DVphDao^S%~3mLCmOjSk6XKH{N?OmZn$HsWv$6|!Qm0|GWxxz(`|m6<%nj|7Jc3Dx18Z&Mxc0uqN6m zJ*tmu)11kqUEa2p2SS@0DSpq+Xi+>9T)7wQ8WZSzHv=Yf%I$(AsHP|JWXlI)G1E@l z`>RBv)FwZ%kiHn22!s^xgLc`W)*=u-mo$q0ZpgtPb3M&Xe&S1Bx_o(!w)P|xDEDH< z^4{>n0B<@26Mx$LdOG`#dmb7`7Z71^x%${SJzD$x#UjnDmB<15JGx>5W6wkB&iHk!H73X94 zw{tk-PsxWYlL+^?aM*FcaGr4A+wFISE*s+%mnbFgy|K`5(B2G+wk$#dm^?n5r8eUD z{?&)OUuXqGcrJgpg6Qf`VM+f`dwb#xbi9#V#VY_;ilMK5bz=qtt#Xw(r6pRkRDlptXr3=foX`0@}XcaL+g#OYaV;2#*tW2qd7}fVkp28$|*getn zLhgA9spMq?FC?nE3ym*4KWB2dEsmx*d)4DG7H9WzZkXt29E9eBQlW$KILX#dfyd*u z8u`GIls!7P9cR=^Kod-(rdP(mtlm~^@hDE7x$=bM78F)geo`EC9JBx8sN3bx5FlU0 z4<8=*peH&f%(wnBQ_(nxVLWs)_JdP&_Q>?o84~nVV}UCx)0Ehv7jygYMwSc%^WH_a ziw;!ZHdxru#|FSMVEv5R#6oE9h!vgF+zW*NfNwO%aiq(#a&cb-Jp58T=}`Fz82F)D z5{oMtjtvt6p{Ud~t25nat7*$j(i+&PcXQ1kuY zG)lS|Q-6XY3OK7X{cX7 zY*eX<3@*4{Chjl%y%#ea;?eGKwLIpb_Qz^%UC&Do94sAeXLckHfFkRUEF7cclDgO` zvPMQaqdhmSHA+TX8YszV!}*Jzf}GgAxZZn#U-cDxx0)0=}uGB;SHt(nI2@ICr4 zQnemS_owa-+|L=;R<^$f@3W?zb+XiQHYnVkDzHDFU&3;CRo%-&F0AkZTo+pv@{@;Z zbApC4&I7@XlSh8|@S>N|zRMQh1QbWN6xEoF;O&ix99Zw-)BZ$)D@gDGd8xwY*qVXK7oZn~98@|eyk8e&gFfAN= zlQ0uTLNy*HaA7?Q<{8?It$V)nJ#iy740Iit>DT}xK) z%c1R#E=20nC^1UsdAru0f++`!sV|N_r-P?1j-*Pey4}mBlV&ya%9|Qze$u)0xxKkE zVzY(Ed-H${H)>qNMTGKr!mA1Hyxr|bc%ZwRF>sUbq*CadmG1)XA6q!gbjIg*(JRjr zd}PKJC`c1srWz2sR+l{W_}^;Bs&9k?5%B@dygksC->DCo6)bHt2n$o=U@ezi;Jyi5 z5(@f0?d{HfzlP4K?eH~9?~o#8!)owxa=%*_WXHPYivFn8Yv{gqBh1hJb6K$KR?vN? zjZcQp)y-_=ta=ROnBesq6=hI9L6oYkC*BQ2`hIKed{P)r2f2DUaBU|OP8)_x^zDZh z;mKRhYHfw(n!$tO^17|#O+35^>6pRt!#nTDM~6l-3ZlY#2602k17!%CFxu@P?%($_ zKcE@IXk_y0jOc?MVr2+g)^5xO)?irg2bZ3=8B1?5PHDPmfV4vH4{NkT@tWpfb23Y3 zlMBT{5~w`Ko!g$bFGA=EcK_Pu`(HC2|8v$^w_^_7h_4C+yg)&i|2HiEZy4Txg~f~TkO6u{c*p@%BuM`{ z{|tj$f))%43bLyOn%_rP1i<`HLEmQ%0MUP-_4pruN4-CuIRUw#|CD^pesSoz0RcIb z1qGq~cS(x>VC%LH76$*mH!%Ok=xA^8-(Y+1>H#lc_2!=>77a)ehkyOW+8;qc82-0! z`(JYa?EmV!_i>pI@Rj*b7$n@ppm{3@NI5a+f5NE$KWZDfnHaGCKjpa{`2hi~|K!0A zY-F*X0Rfq&`@ebqGx<#_0RY~ACg1z{cN8-7PYe==lKzohXDhLO#}mB``<}JZ_ELR{%g|lGdpsCzZZEx@63(@ zAOhx}1jTRt8Csa2AOJg15XS#a;B}?|!2eI%@i~7-y*X!!02j=E0>9hPaSfA!fW#z# z{HOB&WjE047XnEBf7uOu8-Lji|FU-Dcb3yyNK*A3*Jn#CL)V%qUun^8>B^%S7v~71 zTtk5-wiH68O0D!A>bW@g*X}W$E$N&G0iEYm({04q01fQParv|P=KA#xRgY0Y{!Qm? z?;T0x9Uk~P-KKj8y9KmoFmuAO_;tDx(8HicW1=Pf@^H`eK4GecLcP zM}U*4S0R$(UBP&PY+U8cCGO5JA)68!!x$=S;#49>f%bt!F#++eHpxe4O1h+$ zrdl$ioH#vHq8(|5StWK*nPE1pT$*^-sDJKE@(DK*VQ|7m=mwnTNB{CVM=p)R6v1_R zd9ozV<42~PfMrMaDkQZ7i3kj!$|MJXX-OzS-2B|pC=@eWs+IRU@MueMS3N{UEK_|1Q451oimX;? z?~IX3i`iHlH35w30dJID7SXq|JWV%8S3dM1m##E2hiihdY+$LwWhN#?Y)g}7jEyjb z6-r?)3wV@O-@{eM;Kkz)k2I?H+-sX_OhQRdho#OOFJIE&>S9wXgePrF;bAC&YEg`A zA-55F#Uy?(6t?MI;kv+3{4PcAjyp*TDG>!)ih)REHVN#aO{EZI!AHgA0U2yw4KTt5 z5lfGVk+^_IFg_)NpVOLk1-A`CM^bcSyE^ESzcow~C#+ENsx1qpSAUM>$|qm*oyu+T zpuBFyGRw^V=ICHLdJBR?DTDrl6VOBVS7SLv48FhxEnFj#KCzt4sb#KBm7;+Tp+cEk&TtvQ z5#@dRCTNE(FRnidUWn915uAb}f>y;`CIeR8GLeEC%9quCYos z!oNr}#@pW3^Y$JPspqw^&Y={B?N4Kdxp?)?aqzoD2D}vImEZk!d>oOWdwy7CR_dS) zL3*^`t(|}HwL_RWi7>@=bogJqLLjVn*q64o7YF*y;=%EF)*V_A@n3j_@%gw1C6t$2 zACL$7vgyU|&%rRR%MWE8^HQ-_&?_h$s3gss_ajQWXW+mgL`51&PU5Sk@oer)POauT zOxQJV;#!O|r#@*Cz$uvHyFBbp)n5r&1K`9F#z{su?liyok+8I)M;2qbvN%cU>k`9&!+AQWE9ebnnxv-)NGX{X5r+75PN+jg{4@qrkVz ze|gYh$9rTkga)*1(`5vE8CD+gxW)Qp^{O72zSMv1tqWe*$Il3d&}XHOu=_QwpyT!^ zQJiMe9-H&;=N22rB^B!ocINTd<%1}ktdpEwZn3i?+~7ikr{U43=^jP?rDyrL$;o^m z-s6jPTS3ne0CH>3uDTaL3#5<6fd4?&UbPb)+dSNzb$-pXSf=!~e{Id6!|@7TxbrMn zApIJqM96)#y;-Mf2t5YxJHXDd+VE>`jc5NqH61f@b{J!XK_Xaww7j7dX7RhxweY1T9c%$Hx#*=(G0jq!v~5IW zfpwirM=>)&`GJTqb9A^g)>0!RCQ5GQoi&5z20s$EIbd=V%`Uv~ zQ#vscjYQX%IIA!Wd6)(syRRr`oMKK0Ge5wr@bB&mDi!N>N9jdcTm6lj*i}2EL18&p z)?I;I;Au9ZjoMDoN}p0`KiY!NQKalK2E7Fp1_KvuEU~39lQ0H^qV5oe*RqUVOlmsp zAUhUOE#47mP0zivt`C@T()p%eF1V~-xe?xEqGy_xczh(%RO8fKkoi`mh~>ovlV_5; zP|hN7R7rj6neftT8-~!tO*HOxM#A5_Ei!+FZ-PCF^tt%?n<;%tj0W-v*vNUI+sI44 z$BWwss8q}dF~CGG{BtnwuF3t_8KNPp z8Y^|Cp+5Qdj@7~gcDgEnrf$QMd?qMwKx`lIYZh6_9e$kV@Cnp}JJjptEJOxqS18c8zW-gtUB3Y%eR8xe;sNMfI$Du}x*a>A0vGs%WWykP+aNq1no_6KU zCp1<%tehll+IG3LZ%vYITJ|Sg);0Ye0{c*r94=`LN$2{-RgWg;VrS>u9&!Q`vxg3F zhqMHHE~W0r4FIAeV*dLmd4(M8#`{Eo>I+MH^-m_NC$qc~bem4!mN1J3^>->k${ij z3}Fn7oEtToBdu@G>0nib2_~rg`gL7E6S#9@(*iiyN0kz|JL!u#RB$5K`HtP1d(xAc zjF}LO#LHHLFQ^`o;^zwBg@`L)16}ZXll7`d?$qNY6b1lP`JC3_BQxE16}j*$h?n!+03&4t)oShh7AA1KRo|zHHHZb^+6_5q zuDCkIH*93p6IsvoX03$j0OlodIb%BzaM@!wl{@%RGmd93%PA62QMC2najbPk0|*e7 zL42U{!PRItf@-KMxVLS_ZMjO~m>iYpnc_I`I7Xp8SpNCyJS1`TRw4!Q`Cb7R(v4bI zM-XmeaML-tBX{JuQj}_Soe9-IWau4a!@H!$A5-LA^551EdrEGr)q$cX$^K^RxWW1p55l^Jmac=-M<1$Z@R3tSO%rw@|r zhP%7?eNHC)_PYcFMu#Bjr%@szmfXQY2x3b_p$rCV% zE~O|aJ`K);&n*ed*2g4V_K_LzeYoX;Y0rt>fxT|hLi!OOOi;HQFVDC(_Z^^ zpgkg|g@~~~wjWOGGLKO1At&%$*wSOK^rHW`i8eRRidrtDSJ4p2LeeLJ!WO=w+Qt~fZOR!LpI zWR}%ekC(QV(0~Hm&Euj!V@@iC8ZbHf{ z#&bIMyX$dy&~j6jx$t!@&axKQi*ZSHtV8mpjuS(KBW4 z@Wiws>T z7mk`Q{`vOFb-gF`EzXwJKCgLfMDX8I@67uS-ABZsi>H+W0p+-4KmMBd2IG9XK<|wc z%?$YcjF*`M_92VL;>p$=$a>s8kOT_lRGSyL8g^BHODQwYx-=1j!2WsADDWaq+x~s1 z2{UjdvocySiECd-Bk32lZ_b+3I9OA7M=lq}tr$Fh!}rxT-_z9OxoO|Y-v zYV1$d=wy=5CltggyVfV{g*@0JoeUpa%q{-oP98_ky{(f*B zxiNv%!INt|0DYgW5qr{vXam)kMeQASO8@r578+)g71P(b+%~JFo>M*D4TZEcc-2dq z86ssB7?52gZ=fZm-u4aLwQT4D)x}*Jx9Rvpc^i}asn0BrPOHg?z!SyEip&IJexW$^ z)}@ubbulHMR_XP+STq?7F*24T>>wGCXfP79 zIlq*ymeF5>(`7`H(|I<)I{%&Qe=#t0Rt~8rH4v3ZEq~lwLCUV}IP6%n;hXZAm2>I> z*VxAjBNT}gjn^b#k!j>5>R{38PFhEAAM*&ZV<3&5iT}{|mDGky4ub9@I%%HI0G^P= zC|MesfiFX)bUO%5_`4IIJ%kj5P%o;+?oQ+Rj&>uK+a^;t1`q}>q1fzOkOOq`Hh0Sh z>*T|@@kA-jR59F}Rc0D8*#aB(m;=9ofTQJoaq~KSXEt0tTZ9f=-OA;f?fw_^N+2|; z-RQ4eUiAz-&VG)xdp9@JZ>A=8rwV7Hg<4_+BmGG&og*2i)ptD&;L5lhQ>o(<%S?yP2}Ce1S7O zV>t?aCwaQ-ipq-J#RV05`a!+udb=4>xq&oAX7O$;q|1Ln5Dj(6L!%p&|FH*w2V zFpP(7|Cc;ZPAQJiK0}FFlG$QtkFm`eoiB6nu#pV#VeN{*8;rYDD?(J4aK}wVt6w;b zokRy7UR_YVYq4EA$)3vRYhF6?*#$7f!DX-3fxJ*p0>n+-i z(KW{xYl-+=H*>aE(b3Tcv3BUF|DG{|#;>L7@PFg+IV7+btZ#N1W^Ue9(}(H3%;L8_?IH zAb;g!e+iwV8J%FCiuRZ1Yhn--v=anERy_k+PUrTp!zOQ24l|+#KtdecwR0tjLH6-E z`($M_9Yxo*TZ8YT87lL^nPhcs15p)NOkTpPe?-?j2g}PJ;h5V1w~F+}(t2CZd}Rzl z5}=MZ#6AOw*Dt#qnmd(;*;vnZaMJR&>wijdXBMCig8Lv?!Y1TWKz6b&9NI^A-~Ymz zlHW{&kL4@`rHVyO>%fhZ&mje3SC#y^pw41MW~{E&j)cMBi*@7c-|M)@a`%U85Wydy zs`|sl7xmfPW0naQ4gGSkFU#iLRt z?^W}j#?H+&vf}1fD*hfnI~S-i`OS(&hCks!9Y-YD=W+wE;?J<+ZUdR9$b6*TwWNpz z)k<4zJ~~I;{1&?15y}UTF{)mag(R`M5v8AE7tIvu%Jg6g%ohOjY_SN2YO?HvrlIAKYP10km<(*7o5=FJU<9shWv-r0t&94eS#Okc z!BXI+43sExdP+c$3!cc2FS#CC?{rwYvjDJz1j=LRhCg}Io!~ut0}9(l0I>4+A*}7J zzg)%puCu*?upzcf!V8GzT#h;SyQnwZWZPB=Z=NOz=d*%OWrnqW8F=WIZOj*F4#y{K zJ@^Ajk5lvDSXnLb=YUTE;j0B?-@SE>$|2q3SxTgV$q^dME-x^d(7%)(t_v=4gZMis zfR2S;_(4nGGFw5Nn_Le#P3=##j6Dn(Y_)Ll314U+*t75id>@%5sx07ZhNS!+197Nb zj#Ci@TQF#ls5GF(BP1{IZ5^J8spUi(#d_h?@fRS7r)DGTBQ4OfG`+uBn5RY}BX~dpuz4qnu*H;UBiReOekAEZuu#N@WSra}k zoPieba(cL!_ms09s(CCq<+~~cy=Q+nQ!bc~`(zFrZHQ#E6eCw2(^6#dKtp7eTrna+ z-TYP(z*jpoGfv!ke)}T#79CcUA#-lyP>d?>E&DbILdg%VQDbORG6tq(b%T|M{ z_4mX$$0>QODc-|;YicS^w*WsbQ4fFCMbi<`f-{FwusRw_qn2gM4qc_uO#m4_a|8jg zBL)&~xBYTlWicztWL2jO48lJaO75&dKHNvhp+5Ie*NnSPdO0EeTYFcSKc(jOPG4mE zL8FQ1f$Hn;pYmiI#!R)NO%Tt%BOq%rNW@%B)Q0R<3UTN*5M^VkuF)`-k2Z~AnBVt- zQa2Q2O-k!I!D~J&K|2BulL11E7B(KDzvzBLeTreOt5)a3{bdKI2KUv{HJ>Ud8s{SY z-cZ?IQQlr&?BY0aFHtqkI!LBFq?3S3rH1<{r zE#PxQ5%F=+|4ZC*<~9=Q*I`*2@MX-t3eYNJd!- z{eSZ3QuMv zqy-G)b*sjBn=IXSm*&cz2vk%ou)DQ^+njklkPdv^|7Ch`6mSdjc>y1kbFD|NKGU#L z(;Yj>KFCLk)sWxGOR0m!?l0Bc-RpuvC-!m&U1b$kZWUH+Rj!Y^+WjoDK=|m)RWyk0 zG{pjX;O@CL8TEI`XfSSfjFp_UjiS&Y*2A0VEa1GJF9jB-J0oh+&H3;GKegkm{Ibmo zYnwWbys|V>7f^&IVBEHLe2j+|EhvRXCDASPX%Kxb9Ty0!N~4PLL6|b{v#4DX-c7{I z&OWBwR2?@9@C`B#;&r%&kKx8054=e3Ash$~v}W2VX@`peSl}XU{gHr`Q}Z6i%(HWf z9fOC9{m|mjIBE@rFjz>Hh zIvq|(_p)UhhAth>-)=Zj0#yoFE;nX+fp)j=E!*v(rq6BM?YkHG`xF)?>!P<1B<@7H zEj91M1nU{F*Kk%2^POyR!7?9eRJ*&;o)Dj)Y|X8Lomur?bl2KIj}3|1n=iiMB_W+P zy7?qC-Z&HtN5Aoo8*c#br8^21DuohaXNPdHbrl@o4c9N^^MdZ8=;6CJqYFhi>F?%b zwrQl%G%oVvshr824atgKT`YR>w9D7;uF5&i_pCD3XNOy3u_%tUJYg$>0pKsZm8Q%rD45#)WtI_W2g*QIC`Tg&D@$Uxk(RpuZd@!ePrC;qcl+lq-TiBiiwn zodsM)XssJM1aj}$Yvic~jD)J;k_(%Bo=^wzOItmQ#=+_FJV!hak-qtde^x3=;-pZL zf%Xw7vmscp6cRkyvt^dioj1q6DyJhz=}rVi;D|NMvCAI-jV7^BAZFHe9pv-qMnO%HvM@M@by$5_Z#S&a`ZO6$MR1jjKx zY*Ep34ixs~PN=rW+UWUWeZ7ZrxEtr7UT51bNs1eOw5>BnM-X!tQA$0VA!~Fk_&M%Vr4S}BP~IzHiQHYo~F^lY!DAGW=rP9z1GLn2;oli_P| z(U7+O9Xc{ANELI?Q*46qFb8ag9GQJ{p!~rpeg%S%I)^@LYVr#+pNd38O7vjiIkT=n zuPs@#3WnBX_y)6Dcr~+rKvlGI{veEU)!gfQ1bJ`*iu}%#DDiUpW%RI7YOZOVQ8Tq_z5ZQKF3T>qYMQ;N;C_9`p|Tk6vzGu70jQVZM8PCX4SW zn&Jr_5z1R07r`mU)1==kD49N-vbhQf>19Rw<9IU5p_&sH zDwl1>RTk0luk@7jZXyJUiN$f)f>ZvJ8gnkd+l?-1Ujo9q#Ep%2ZBq8V zce>9BD2I>qpflQP$AZ@UyARKm3fpCCzKhRFO|E~mSJX;^K$X3DjL~Q7l0KhM23swp zBl{P6Xer3!mKGC@T6u7yTH|FPiOgii3o!K5`U`}b@)UXaf&?q0!U5a-jYG^p?cR0Z z-V1?uJ-D-S3_7HN=3{sOBiQX|0k(f}x#C`%33er8Vj@m}lGO~+4az0$fh(*h!rH1o>RxR(X_eAayifEf8grxyJfPg+ZV_#={Wo*AqEF2PrZ`P)prGZrL< z`+UomKl53OyXPxg07uO3Pop|sZ)EoU#vrCC6Cw@=f&NHW&TD56bP4x}cNR`s6K~Tn zD_%0rnU#f1;p!w8vXt2LcJBwE(g3KBnZ-FOasH2c*Z3-4ZwaglXf-8AADdT_8he+I-zFsXXXI@OiM{6F>Vw|nqhEe zm?4aD1YvnI?(msT2*}rA6nt?dW@s)UwmF{*4qS6YGfIRwDnHkLbNVWsVGwekC4Y}7MfQ&D^V0u1Y83=Y9bRQ`j&Nwz8w!c%0 zPS65g?eMcJtrb6Tg|h)8f;E+^xUb^^awT-U#LbCUMFK-G`ovPZ(|DG|Hs=ZbqcqW< zxZRAh`t2~wf04KHo{_|5YgSDlA?pOjgH9Z*Uljwn;97p0Mz)ar$75NDf5L`w8?@vp z&lFAE^}<2@H1IZqWg+=6Bt~qlQKB|5T969{7=ip~RV|cXx9j+Ag4b*T^^sb#`l_#F zp@mth$St& zAt;f>*6cc}C^Ei02TfL6)U#E5jC~ZhN+SFX#&%+O3P>xJPh4(y#qf@Y*S`0S-^~g8 zC~WeJh74arP=?-fy&gh9>k{Q|Z^R(|9)rX4S2BZslmnKVG?^^S`kf!-L&r;Dw(3}- z_}Ws^k=L8yb-HCf{*M+9G=H!199?`v5Hq`KuGOQj&r>4e2UQ=2Rpwh;Y8}jkt{-Dz zBA=-}ltl6}#U|}5(41E|`jES!HYnsv>Z>!FM z?P3QiMKmNS*}f=v2$1A%xBmU_#V6BZtw{l3kSV`DHaycP^* zAk{b?zvxVTx(V^u1Tv6x`_f00om>3NSs!pt;!0^d>S5y3D)p;rK9KuyW5lIw%L+s4 zU1fWY@Wq6wkp^iIa=(Gm#PWN0<-~+{ej89m#*%I-c|k3yH>6B+uvvM?(-7^=m2dL zyb|X0tn!N;F{IyoM!2EvB77P+39~41)TmaC#^{hap#>|E>oiHYhPzPP`~!AhiF@^P z1z!6KK_9`_$`kB@;KlpbXBV7m_WqT-K5zHSe-T&E|3OrsLH?5zqe-A9BccEUajO9P zpGh%;t$A>~|95O@(^3^Ye$!G7yl>M|9h{ruzk+4~D%%86oM0f&jbM2L8UHJ17TRbA zPWJy~mg@gK#n(5QgZI@pT7VaG{>SB!4yz`Nn=k*5-vUKoYUFBUU}SG<@ShtOGuQtU z$t8n~+Xaja1Y|xR1Vs41LsBUJ+q?O(H~&I)!2cq_Sh;z^fq+6hf`WiR{O7|j`L6y8 z>vBVFUa|o@LR#^1(O9Cnqhd;4#l7h~;@tf&_=Ffp%tebV$OJ-dPowsX{0#C;_e}Ln z_G|k-5EMG7`7rsiM$TA^jERr$W1EjJ+2AX*p+`_Oaf8QL<2JCt;JgfY_0FhM2919@ z;q8J+HtYV?6n{!`N<7EK`|-%}Tq7Mthm45j*ETnM$SX5QGH!<2#A_sQCmF_QhSzus zA{3e9^zvbb8Wo`z^FTn%+-N#48eL|+FGU6rL6`yN%8@Y#f2;Pm=3KE9aSB`h%ONVqoTy%rSW8t z7Q_Rl@b}pu0&s9TW4V+eGp4mBn{*+?hFQrl;r66x{TM8it#%v}lE3{-Ix@~7<6v4B zHEo|p7L^xg5<&^Tl|$4O-T-AGw?d`o9~Kf+`n_(})D6Gp73Y%V7{p8GNs%o4N|Izg zAsohZ>KS?qk|qF%P@PCs|ACJvUZc}GprzHciBZM`&1U5fZ4i`+RF$A+KKhPks6r4} zEM7H`M?B>7wrROkrZHSZ>S=~dR^Q2aGiQobND%1x2%63{7hm^m{>t0A^IkKV_2K7U z2}S)^*Kf^2gYpn!x&Lm@&q6>U2x@45;B7Nz;C&Y+L*XY71D09J(h9B})-Bxe06&(0 zl@Xpj_801()(AZDWU}|O{LanuHFjKzH%hf{1MC@(bGE3wG?}qkeq({&hAWPC4EKuOs?oVTynrP#xM1MBY5OPRMCqEfq=&3SQtb7 zQsc~062>vWp6ot4Bw0V{M6;=2j6J?!UEhL4P7wDy!put}f1_6i(;t>Y7WY;}deTZl zChd6NZmX?zJDh@;AR+K3OtO2nNU6_Xy5dKyR}&}fe;*dPF|YP71Jvd1(7!kNs|j0$ z1{1WOJXBA}Tdy@kf9izd{8^GYrFhhQ2$_W{tay|Is=sY9Oou)>au(vv7YgGtgDwUf z?i->_m|X^FMz4K5EFwu=2AF47G*y|VIwgI_qIY&KOvxGD^AM$?HYCxwbecB$bl{nn zU<~jnh;=bw-^lqff08}J+AUccXNJsIPC^s@VU?^dvTo`e^u6!N+NUqvl;r$r6LF-hvSw(L}0cF>l|i^tnRmsSQgV{lQLl8LIC`;u1~{FT;sRB536p8UeKuFnws4 zW;*7jUd@*kaOFV2mVO5pqY+P7%eUY5HgWg{BoCRpYW00p8`-4;T)>e0oPM|A&Fub| zhsEVs(=SGO`MJENXpPH12Pr*%AAgVj*!Jfc)N5d-Uv9^rrdwQ|Mz;wxPBr!EH{NRy zz-JuB4OW!pOx(WE;$QeaUY-*JtD~9`h1P+}o0}yc30QBH!YACpQc04Dq}lfw3QBZ9 zOW;ItXW(B-#KNFLq zExbUVK$`-s_X2J)UV39msgPLrJ+?sput(sGW<#?|7DmLQ#pE;FFr6nN0sRcg6(Gu-TWv(I061P&7gZ$#raY`gMP_P~Qt7&Zo6ULnmg7BVxm~Db^9lRM5bHSv!JE zPR$o~b1^|P=p5F=YkX^y`6zM-cd!?TZa7fqFgnJ>cOLl*NjXoFkchD*FsvPEi{P|~ z;NpI!h)6o{o%(tr1GaQ&Y<8M{zMqW6c<0=|8+`65-)L@vIXThlG;kkO=XmIiUR! zMy^0y+Jke+Z}w}G+R}oq1YJ^^UT7nx3T1B{N6@X|QQNFVP~{4$!c@x}?Um^qCHDiS zL^Q;>Tn(P(m};fwt0q%auw{?GjXX|S=)9Uk&|u(*&|i7u+NQ!cK7|&sHxm~w^khVXSB0N!DR3B~KYu4o z-AkHEdcUj_N=P}H_^02Ib?kua>575rvNd9eu+sAr3W)d}1kNH2D1f?A z2Azt9p}Xa@{B5?D%=rtuGjfn2u;rq5OP>`pur$P64o#vQZ!n4t@CFdtys?9aDe;*l zzbLzUmxC7xzo!gMlRf%1Q-Az%(@WJN9lUP}|5rn@iaLG+ZIrM_z;Qza1lhFRwm4wp z-PvLNV@s~K`x36{7CuOP1l7>3KeUB6b!2kJQ#PWCBPqIR@J3%rFwli==`i;mT<{5% zCRJg`H;^yg8J&^<`{#&U&ap6=sg4-2@{uxPmfy_d%tS^@YW&TQ-E`@n%8vC=obQ9C z(xZzVM?6ba5EsHSqQ4^N8h*;;r9G(sV_EU(`9EDWsYbEMJle(2V6}*_m`M$z^pA(a z!_B9DCjniBzG&f$k7XBOE(i^7AoEy5&h98&z?giiAV*yTzGP(>z$1M}E9y;4>#7$; zis-nK_0;*HyJU-y>O(@0vWGSrs872xjQNze8@pY$zWofqy+>(nEO$1sMsE|d#}mOH~9%Y_X5V+e;CqQ@0h?#!+|4fQga zAn1xHZs)%a_;FU_)PJA~^KlDJBM6{8bGS`cwN%CrZ-cz*t>I$Hz1XxY6~`A(R&An+ zQ@<10XrGT`4glIu=G=*fGI6*NFvqmzSob88CA2qb4HX*W8?uwJ<7=mc@o~ZPA;>Nw8opjY&$F z5z@7@?K4E3-nT?5tguD0FtdRe(RnIs*!{GUtk|c&RA7npu|IleJYkkZWWovG+~gFi z$tLl>;geWp3L?T+G3zMvBX2Bq)JzHc>0yf5)2F&~vv(r!Dph& zVUV{23hots8Ta2aDJWG-#bK~I4>cNn;!X=b$+o5$9D~nP zGm`T>BEK8fPW;^vJUupEZa9>~Viv?ZyAuwPkvgMC=F4TLhm``}HN0m&bwN)j}=h zia-BG(+g0@1S)SyX9wyKs+xGp3^0ttV(51JQiZAbnSXjym4Jue^3`8(F13#Y5N4;+ zi>(N^J_CLl%)1t`IyZJj`VoI=>Hf)Io7K!A?RBKH&SbU*DpH9yqN9*tzAs`kPY{1Vk7R(0GmA2)Oz(n_%cfx% zRwmo7T>UO)OBz7%{i-9^GWd;G>eeU$?k?W4wU~K!O{5n11{*~31YcT6gdu_o=-nST zPe`A$yg>2k7aZo1qhNGGRl7XGUI^F9zMo!+k}(N#HuX1V6$N{cRrr*jd(F2Fk6}ql zWZWo*AP6HC=fLSjmAM7m^m+O}1?hGBxIK~5u}mj5$8XF_cT$48TaT+yawSs0^A!$a zviU^5^iZAFmYU_aJhnC?^!DQe1l71lKK{`Z#wWw+}JOZ}L znz$~{EZDpir)#c9Z+mlf@k|7}NvM!T3^mBEyEM_x%^qH#*SC?0W34t|v#DNucLIx+ zU3GaBIZiOH6&s3ne~rh{YR%{<8$z^=0V{1gTJlEQc-tbjN4@I`e{+!vRt06sp#KfQ zk4nV2Mrt!>o@WIr8JiSsoEinzeraSEb18ewIa=dcH#--o8w#vk{JBP@Dyy zk{9lmD|j<#E(VwL^(JBVVE}&<$Q3L*C+)Tbc;=EN2p+aj%7VT&4)-eO{;I8EArbC8dp~`#U3=3(Qo8w7$$4oi(hm!%I z=luR>z7aY8H1^lJGwW+EADJxdWAY9}9&;DsMe$f{Z1uz=-0bxAMX#>UTb|uT_I9?$ z4!F;VTTR#!vCe}qEK5PU!XkA-JRJGdHV)z5RKR3!qp zB<(vs2G65_T$3Em5cJR)WO(1kONb))QG6Zua3KNdZ-V88enPm!STypuzA^FBT@*H7 zoJXLs=nc5lN?9H4w${nMwkRsa+nz7=SuckQEp77qk$u8)j?&^=$jq+y{n)9^2`sQ; zonSPj+*w@-VJ))IoOjzyWK*kk;a9He_;0mO2A49xDxkF;g(GzfZOv-Pz`1K2GL(zZ@HdD*xfA*6)GK+Zk*R`n*5);C6^IJYaN|Q zvzE*NA%2Gc8NG$?lsJ9sv}YyFd%Y&s1O$7t*PcY!E@_v1YKon)BtGvt?W9RdP#ls3 zZ9GAHWCUnW>hoKL{+8a4-+_FZi)Y|@-4SWNCA^zq7T!iNInmnTF|{d90CQ}M-p0(c zf*dH=US^s8Txo3?IdgSjmI@3yft^AQf@i*X~;SL5%i2|Lf!_!T0QJ=xBI6mrnpN zctSa(rd@ z+eD_CxBr6?36u!@F7h@Z>5f&q>;Vdxo3<8Ne>KP+iT1uM^}&wqt;umI#Km|0)u97> z?v(J|l>-W_KY$m_(0B{^x=kc4cOu`$TaPxJI5(Bv-wVE6jdBVqx6*q{f}{m_XD;1;PeJ8dpaw`028 z%aHfQs5vq&vBR)I?n|`#X);V|%k0U%Ffe?If@)C|5~Y%X;VkuvC2b1gt}$~yDX$|e zI-yrFcpnmcKUmzh7XrI@`U6Qpb~>PMd?RQj*Y0aoA7a&A=kSZkL-`B<0Vj_~_%$9E zXpR(Essh(Qe?O;I9Hk$Jn;4_3QOOMaw3k7PeAFwZ5ti2T>Yc+aT!C!4k)8!bJuUNj zM2O}j?aZ6#DBH83?Mt6Mju{%!v@42$j1nC7SrH$GU@)qKJ(5PJFtRWdDocD-b<X=Em|w~cGbs#XAxtt^y{RvmMAY-Q=k<}SWH<3KBLM4bxAHDu@xc5Iqe zZ4p{`;qQ7tg1h7_rZW-d~iXbAxy++CpqSiL;vY z@EFQ~6SCHDkwv5LaJM4by0*vS{&5$YA}@`|li2!AN>AR2_g|gB0!4Pmdjd^TM(t&yH@L0s7zW=OrY9*5N;?3B$*j(e6HuPbMP ziobj$AgwNeey=Nvz#Bhb|E~DP%YT36{WZuofuZ@%rvIJ^BNP_4g}nmi_@5v&d?K!F zId_*M_wQR`?A{LJaC+k3cWs8=L2bs5Evxa%n0hJOmm6@JYE%hJL%z%2D2%5?ennjwL`)1;SYB?*J0?rD)tB z2lr*FoEZcZ7Xbh&C07 zP)bfwNbK)?SXmP+xKg_kT4txwUHk(GlXfZrN->x8$g45SS4A}r^EH&IwmxS6?3l<- z9x?NNk}DU?G~#_OOwLxd41=DGD$ddAZ|2K@KWZJ*!+|AycRH1-0Y@2nu@u5nYHvxv zDf^I=oCs%l2O=bu#>N~4CgfW%hx)v(rG_TGb+euPV)QXZVtcAd4^mu;8_KMt$Ie;v z-~9E(FQ3-ss&DCE(|Mnnd>=#R3fFBsk_D1mGp-VHUtJ(G>S!Z=qVmn`v=_$mPdY=u zV860m(h-u#8Tv#o8bhpShOgqBmjv^uHgR4~{;7Ws8o5OIyMcu3?7t(>(!3mHq4Mu0 z&9){=&RfmBh5Qh#1PX3TYzsjkg%$Xjt`p5E%07%j%WVvA^$arkXE%Xo9jwfet`a}1 zIal-I43#b%b-v}WZFgt09-dBqO(wXleT*iXIoThpOtQ&1;i)GmkcA&svXwvoli0wI_U~b4uZ> z@)A8vJsyWUU`Efzd2<})v#(6BU4WznKFNEn!PEz^`96W9 zJ`pu}F+)LKCWwre7(oOADw+H@v9f2BfcV$Al5^674P-Es(1tTvs$D8Pyg6iWW?tTP z-Mb0toSsXOf0JhPm%u~5K|93X^etlYwLbZfcTb{Eq^Q#LxxbQTt_ZKShcZ)K0T*D2 ztSpZDDeCR-&MXJg1n>j1P>!*HXql}B7~fcK>vhlxaYXWni5$45kjXu~4~W*Mt&Uc2 z=5b|Bn8C|<2r)sHf%!NS^jsy%jL=^_I4Ny73K2~atla=(TxChW78y&bB=jqA-Yhm# zEHUEY&Mh<>1{nEJMb1jqD&%vbEjsV!NOB_#sTkj0weIMa-s6t601SKp&&2r$8oTYj|!!V!irl zco!u3=)p7%`&f%KE>c%v+nmy zLGBl9Q=>}I;(z*N!+1&KL2Y(xN4yiOqq_M=6!P*i8({>lAq^N$Tbg1V(ovk>vC@4K z6IkagB*mF|4OLsrDpUIvX4w*daHX9G!ouzSP_v>cY-R;cUP_Hz(jKKX%%`71)3^c= za(@5bRk^a@<<$fBS(sFo-PoSNaD{LCk4{tf=c&Y(wis3kUHzYv;VAD2I`Y!m!&#W+*;G`4Mq%_0IYV#~t!A zTT$H`*B+z4>vIQ6@dY^qsgy8O_D2nq z!WJ#8)yP(f{zexV6WLRv4Ny3xVXi8j#8g>K$bLi`!w|nt`SAw6p-j3CcKa7e-bz#= zC*(0!>EdE=*(sU5FF(r)_u6+B9tT|ymmAYt-V>!+=rqAKkXXP%5!fUm@P{hk|GjUJ z%2^oYIFtjl+fkA(uvzV#E1{D%NhfA2Sh6~T5}s+jTw7ja8bhef9|C3<;EngtYX(VD zLGj(~59GK?@5bJCZWy1hIFOTSwzyKnwbfw$?k8ogbjbcmC~tA+*S{ukasFnqyNGS1 zqBp)F@@v(zfTDNVv!QiV6LUk6)JvdRzPG}IC8iH(&epYBnB#c#$4IrmK2OV+3#rR9 zKAJFw;gcgVPTF=ZHHs>rwn-w(4KmDo1%;M6GeSN|Nrd$WKOYHXa(K$&?LX2Wmw}(t zKpfcnXf8E4XJ0;Wno(pbpG%>f3z2KA5Clho_jFu2@gF%dy*1r;&|`0TYbYL*vXQ$a ztF8y?AffLnlKaWsIrm>Q3gWsfHQh2$4 znx0(@GOtzx62^Me+*1}@bti~QG6FWofZ705^l5rrdQaz_YJfM<4E>-cv{;if<*cK?WJU@cNZwcysbmm)9 z?3`4>fQ0_%oII#H?5@brN6uN~{Jo+jhZ^b!)(*d<&#m9~(MG6#N0I4zj?2|_VSQ`T zPLZFLtf6kGg_`gQx!hm!l{bAxizEynlA<&J4Wa=?O@v>`?jT@P7M>Uz?^l*!f&Wp= z-tQ4kHGTxN6$teoDw=h~7BM8jmJXiQxSbhj&ZQ=hz%;sM&Ryz}H9f8z!Wj9|caRb_g z5*cBRqqXBR+CJBGpKn2z91asWJAL{m zh=9)rW}GgznBrL|t2B$|_}Hv|Cj=kj+oA6+;D7setGMf&C^e}Ua=8Am=9R^8_qI4w z#vk*mp$RE|zfa^PT~7Q{{wtCc=z8yBEQa;mbj-Uy%;y%!(M1Esv#13o3aX)kniIZ0 z-*D-`lI=7CNx005VZJSLNj8dcL*O`n8WBt?V0&)H5IQ~1Mh-kItMz;s-DyR%SpQf#a zFth~X*zq@>k3(euB}CkKmKHU<{KTMS#6O{!#=Qf7S{_VnBW}g21A;^zNr&1c&vgBF$P}>Q z!1*ACxKK&^v26vRaZ4a?R4>8-Gu4>A7uUsz^F%DEw$Tw_$?bp&_xn04uX}p;kqDnd z%76dm=f?LNN@6ar@}r0JtLo_eM)%k2)WqxF<;UK^G}p)d@h(67PVYN3VTGE0_q4;l zN-M^z2v7WLg+Jq|z?b{e`#TiFUr%$p>tN0^?s8^DYT44%s!LY*-&0CYB5K}xE_-(? zuX_)VBMNmupS{w&JS{sDj!fBD1;^{s@w&nGJ(HqS?N(r*C?OA}1LNFwu>CGq;2qu0 z3A)O;X-3M6S^QWn3&$R%R6VzeemEtGH}N!kdu8m!KU+FwROuo^sSt#UGa?%LO{r^J zds+U$1r)wdP!<&57R==jrM;~%tV&5rqKh;&GfgW1&OXvLunWq$AtOz4RD#BnPK=u< zsJa7hFkl=fQ>bBQ(GI}8c!PFzu=IvpDeBQOlPg(V32~QV48o^s((lKaDE!I|zSi&S z@>pXwh~0tMIN(Ez!b@ZDWST>*cK_-E{GJ8=hrH^4n5kI)JDo)TzonBH|L=&Cft9_Z z+y61>L{~bk$@&lK6qyAAqV(TO5&eHHmH+el|3~hrHSOz0#qg>FZIgDwE{jc%V-|nR z$%pS^IB(6$du&!R0a~h>I*D9zvom-0f%s|eb8}r{UE)0X@)blf0gRj{GP}ozH4;A( z0ud2$mIE=t;FD}soF?jFBaZVc8M&_?KoU4Vy;mJrYRQ~t2tMN0znJgJoHQeHfvNL< zO(i@p9v7b!47j;Cph^tvSJyr!nNdZK`@SYeW4*x#q30sF=&pNd5uh?@D<0S$^YoRu)h5tv1e5s6rs zX+9xob)7bGfhq(uev%3Cw>Sf|T{0#)W*m7m8LKpqW`q_d3y3N7 zk_cfk%gM!%nlu%1jKBz}O01d(ULY2*C2NMHwmf$1Z4rmtBhFr&{gybcxW_$R=KcpCbZt;J$tGHt{jkU}W5PDh zbT%r3D)?}vKF|YF*#SZ05B^^g#1AUjo+j`c?;?m7lG&~$CTLZ<+A^AJ;NQ5GWw%E& z`=sg2`07_w^6IYW{vPaWGsDJ z_xv^Q-WRWX3t;+fh!P;zf$!S0ihGg7mHNF#6}$Rb;w~s;MaGa9NFOnOy=bbW2Y3knSlajr2O@Gr2uLvhJ8eRt^<cB=ypki2Y$B=(R;i@_O8&v` zC&!KXz^3(3GaM{_M5N9`P>?h#ff0mS*BCu}wjg`1UguIWDDXN%Ud?v0TlIxv1wO81vvIm6K-&(?MSUG8g}0INIdy z_ls<|pP#2Y(Wm?0lNL^%mpfU!US~s)zdnxY`O+wmT}W$1zPFd>t-dEZc()3;_T_bI zW_LskV2|1%T$%a2QZXV-$qhL~xXc3P1-+LUhLIU?y?_XGL9&u0Ma+xGO+vS4tS+24 zCS6X$-A=}(!`_>*;2O^NaS{dda;@L?=!j0CS2Y^nP$;0(L8zYs{$Wyt!wHOnkR_Zm zOrd{7`VFq4Sxkt6hExYTg5ZRB3k3?5Aqw>+$x~Skx+<@AkqT;sD`k*gDh0_Y5wcjZ zunUFGF(&} z=D97v%W4muvxGqY0QqC}n!S1S{WG>MyHqJS3%(+<&Xj2J>km(2%aEeyB@Ig!=+MQ$KN?Bj@dYa(eDfe=}= zH-C2q#*Ac=Q1hUPlkSQGLa)5oe-22D6cr27(yOL{-iawvO}d3><1SE>h+zb)@{l5I z*ACJ}W?BeS);C_kL<`}RK3(mo5M?jQv9$Nv7J6+^h)r+ zT>^L&9%r9^IQCtaJT#`8CSTRH4mG0r!Y|AH6JgGzRzEs^x1onHt7FHaxH>3^?Wv#6Q>bL8oIe=k`_-talwjDj%G)0 z^eXVtx6(j40v)9r;N3X6``ef;5c00hqhK zZntNNek-xdoHzR6Lj2n(@k_YA{!1$=Fsn`yydm~4_t5^06N>h}3n(WJH~S}TkPw_q zMg<-3BO6U++~l$SzJqa0OG%u6LaVo5iD|cm)z7uoLlCnC59&TxtS@Fm~LWi z*0Zmbp%Zi8=h&>>OJXfZ*+3TlgsNdfCY5^#DSwBuv2 zlpl&5>@3)#a3D=c_GcrlSap#x!qV_OAEdqzb*@O6;Ut($g_1iSDzbTUhPW~IJSrJx z%0wZG9NM&^u&9`Pl>Y595YwXc&!6={*)2Dqf!*+YXi$l{BZ^VG>QPdWb|zI=%W_Rd z?s}|i;vLMY?XHC)5`Ve^Iw*UaO;Sj{w`yvnoD!ANlEVC|b&Q>RMPJ}ZMdn7mM{fH3qXTnr0kETQ&S z3eZUEk)UVAHniRSGtII#w=t8G0zsRT;`=ltn-ms$d#VsFs5VIQVVpR&%PsdeOU=!+ zQr#7dsB27|0|Njna$iMi=;2t)-=M@JO7 zD7BzOu9ua(tp2CS!&D`H&Cgyti6wqQ-5Kw-kR>zR>|nE5KYHXDIRE)O@bGqrG+JD zC0pvbgGrncIf9+4jo3&p0GLWc^ScgLD8t`w6*HuxY&K_XP;W|{nF(w-awejP zeyB3^7fW6Iq$FAuI<^Q5f-p-$;oJE3vmwJF)NgDe-|mz%Oq+MJ%qPCvQ*sJg#@|yo zWX?Gy>gcJc2yf3A-h~NsaTq>Ly}!UDp-{L}3yi7yVy9c(C@+aqT)8aFGRZJLgL=4!>0) zFj$&@I|*j8?r$6nhj53vWwyAQO^BgFZQH=(0KofHD`D##@fXQKD#3XyE9MH44=^O|Neq zg6X}Vdzq8}E$0egkY{hRx@l0%tE(w*@%UG*0(zb{?7lVes-GaNTF7l_J8WU!ovonHV4o2u`FGiDq}FB7~^0!xPp3 zyfoAv0=Xe-uvgyCM-bKdwmz^KaORIyj4@ece}ln$pl19z3E$y{$l!4pv?OfCC_mSS52@}U1B-8xG&L*gtl_Fg07fQEGv(2`>V0}o+At3p-1 zj3fZBiaUJox*}ox>_Fpyo?Maa{9mj-eZ?#7*A@B=Di|c zIm}z*II>JilSF|)MA-+8)h7170c(*y`Zdx>LSay5B8z8K_@@%1M19{5N)t^?fi>4m zrx}4b@I&_f9G&H&)6O<_JrOCEFv^P9Jn;6CI$yrlyXu*$DWhf(BI*SaTOGUYe z&msM-bi4tIVM%qOI?|>L9VMH8Q5+dT4hNyZ1lz06b`~Qc;)XzoT#P&R!R@K;bF{p#r=3BMyJwJ{thW81-QGWAuy~L79 zRA9QL%qA2%BDq$nTDUdl)CMejgK?w{eNS6tre=V}k608o<(QmOhbcN~O=dH~rOb#z zkDlTT+haMP6Oompj*d;jY`xNMe|@Rl(sjOi9!27vF-xoZX{7g*>}pRb4G}x&nlVmS ze#ha+j)TaOx444aPl(l0U_dWkjzl$JJGE=fd^SJ>FN?ip_hWa>bQyr|k|USYgW|N_AI%M47eU6JK&hkmGhRBrh15p$FcYYeu$nyGjn3d_cxA1XBfRnn$o!-|MG!#l z*-(-29B~rsjnb_u=FBOPO4G<=I&rQu?@gl!ZB?o((}p@hcwZ7VQF5#M)unmy@(;rF zy7+A9%d&zPBzfW0ax*VIY{Vv>d$9#>tn4jEYPu0XSD-U{Fc9B7W$3&Hk^ARcv zcu(1%O<1F8a}%}f0sqdHB4B>X=nj#Bh#(RuI*KQA?bobqjJGNdNp~B%jWoS z#4F7tbajUxIKDpGzA!9UviosMAL2Z3qj_{0%i8?>-+>EorN<&uda8cnkU7B_ZN@CE z{M`@l{hz-PDp|d@kw%T%ek>4lAitQIG_5ytsIjzLXaUW_t2*&5FB|zOP=Crd@}t?9 zX7@TxOubyW%)AW2G(zB}&K}_Y=rJ$DUMkyKxAtI{)^b)5eu(6_4_o1catTnCsM^8@ z-5DmQ_5%lDYoh_PHXsd9_)H>Ze6#|B3wVn1*@m}DsX}(Im8=oET@E%XD(UEan2#!{ z(i;Qi*17X+Xy9PsJs=7JrwLtnGKpXsFWLTsvU{ZpDuRC<-k5?1O2OxUWNE~nY+1TJ zc;|PHD)*3Dv)w?8T$QG%9&H(}JBExPgLG<^^8wT1cFk~{)t7SR9(}F>)0hD{ZO_0d zA|hIVev4K2uNMN&DaCe;np@9MbxWT16HIkVgU`d~IkX7=2$A?Cn3XbgYWwCE zFN!epw@M1=0!BL2gE$^p6F)(;Vi!Yh;UxfG8t8bfg4(A~962=`{MOl?;Ls4;{xE$; zi5RU(^iD_qxKTQ)Egu%^R?NM~k9+{u(VkcA=t5-sfgrc{nsLZ$F`uL%3)k2%u9#LJ}ES{Q4tNZ%=VoAd?gvX|OpPM#rP;>ZsE6#*!C zHXA4@1&vc{yL|j=Y%L#Q#$U^f*ahC>Lv5%hFB9okXre>dn&8c166)DA(y}27b;z%B zm&waYoJb{#-dPH3a%_CZzb)ku7?k@L;sI*-(%`>FYrOE-ioRYMwYH~?tcZnmMosj& z)LuRFkaM%ZqHE`5*KtaA7q~dc696}bHh+rtumoI3=FmXwS7}7-J0{;LDeHe--+(s% z_7l@12$$YRADPqmqHW7maS~`Je6{^>YCOtorba%GeT)nEO5`r+jtaHnI-*n7^JD_oNuS*8+ z={rA^pCQ*=Gai3nAPJ6_vjj?Es)*9Z~r z=S^my#O4QZjE0P zGf&3JSI%HRjRkYqEShgJxCoqSaKAd^DpvfPX7#J!(J$is`DITmWT1BCA26Dct`4zU z1H6NmH>a!ZY2iIE5*StX8HP|ur(r8`RYFEGLh8u6Vo~dSM$S_Gg}m>jI69wa@weGo zIoHZ?EFP>akFR8DZ!)@tM%i$*`p*(yHi2FBL=fQ=^Vu~crcKFdJ1O3NKQHMo%KLj| z3ZO@2^7Y^G`4)eS-p|TSFS_KXJ_VB;_{5o^?6@hireG4WlEgCuHn*?Bl=>{*4v4EylZV5mFT8V;B&Y#yrbG7 z(@`UbV`YYa79bb+23LmhFg~*J?0F@0;?=-$$1b4av>$q}D{eW#>!AO=>F_SHBX8pV zvrhTO;++MK_XhEiK=0dMWuvf_r9F@bki1Vgd85C!N=gS1$x;+c=iNwgy@W%R6VI@9 zv6_e{Y~LWuTRcyf=scv}*j%`fHlN|rC_{HhFM|7Ubb;p0m(TF5fR(azdjP7$;IQhC zycmY4yg~Xu#nN=u#~q>p7+JK~8?qaUlY`SPqcx0567 z)(s|gVNV7OH{0(1{sO3hN=0udv9? zDgUn3oiFC8o$XVAXd}(Bc+a+Owr0EMVi>umnpUn2%a|0ECEN3Ms!3VFAWu)S(-jH8 zUun+HkBJhG5HNP zXOhPLAS)xxt006*l1KJ8Xunu~>HK2gl!cXtm1hdIO1`iam3eCRV>*-7V%9fbf8P*C z!UR^ZFqT;VA3}gcm+#$_{w;_j4|c*~1_jsEZIi;%>P&JCJ};aOdH$Pobp9ldrSniB zkpObTdcdS(c|w}I{)k>kRZkWjLdgKy8aD}2UVIt@+Z9KkESUIuBste zw}P~(tc4=GOb%AmJUmM}^9NyWKif90uxtPc)P2GEN140WeKo!gmcZ z1ku67A}j16&=DeNr8GsxPsMMxT`G>oXcqs5bIJh3AzQoKNV=^y5*J6D%!wq${mbr# zQ-mb_P3JfNaS6JhkkDlKC`5?S>5$+uYG?MPLWzhH@zhj#izE5D+8+)8uac6s-YQJ3 z3f#=lKmkE%Ha_G_ja@^zWnXRGBLu*YI|ua%zh&G<927wK`o9=^$LPw!Xk9b5ZQH5X zwko!58#{Kzwrx8V+qP{dl~i(5=bUl-oYCF4`_DV}{_(A~w#HuHobP;Ig(bsgm1IaL zeGXmz*TMTchp`gnk#yVXYU2_z0r@4_7so^cbn7tAx_FLtKT&3>d z30t$rzjn>WDJ0aYOnLfE7Y59a(i{a5m`%7urbqk*=0A z(`Q^nHACTOfKVQJ=+5vRIUwG_de{e7q>7psAmg+8)?=;&X@L;&Z2~@Y$_&Ka+OFCa zV`HCuq}2ggOsJrHamVjO`Adk3QQbH-xdt)bmb?dVRHuG;;D(jSbH^6R#wLJ4k8u^n zu32u_11+rDf$X0*DbDfWp#@FBBqZTkHiBdd@m14qe6AQM7t$p|x9bZabev1}kMG04 z%E^Ia*d~DblP&st*h5?ePbus=b!L+<&XcjD0x>u5PAG^6#ZGKNj zT~El6#yH~5mxf#3aL&IDcTMp^z~3B2u3C6l4s+SfGPMg2vr)G)5C#DOc>}j*ABhp( zJpP1Q?6JR7o`@4OwlOlu#1F7aQ_nkGtDCdi*1>oXq((nRnu}-OgedsE1o>6qb_26U zLKZ~3U1Iiq`D zmj%7xd*g~LPJ7s_ld{v_DvF{4m{n){mRY$6s6qg;0IaxkiwoEi=oJ=*-TjD zoeL^I*;Hq}8}csgW;i6!orP2^eC7PTkL2-=q?gL1dUqUnVrCAL^-YUD0 zvn4GL%eX-)zxY*>Ab&qnpQ&&z;x<^#j!a{`--pbI2j>7u9IO&fYw3JSKkYhA+&Bmw zYiDdo72j`~b-1O>>oUz3U6mvcQ+dKY4iK=;uEmcE@xxf})4ujhC?>TQSGt)BYLC(e zs#A3sNqAx>Y7zk;E(Rqesn2egQrfIMfjnN$oxtA-ym|43#M@4EC?GWYT9?rzlwxBH zgu0o86Gz$m7w(}*%bhZWHci!8jyh7;RUw92?}k)gF0#*4EuuEM2;x{rm65&eERm{- z#|6LBWO2Q%q8@B?|8BBw$LBdRrq>-6-40Cyc=u9j1VaE5D5QhvZ6?=0{HiVkC)XVc3I~dGli|Qf(;u~Hxu8gXlzoy#SF+C^-rem5E z`6VnaMHeA22_Nd9^eO(MCO_qrf+CXf;OU`dFX2I9&B&UE5yV1g^T zs&ac?f)xN_Sl8@=e;oFG;FrPl8`CN9EnC@K93TfopmAN^m@{5#n&8yOb$7gJjdo;Q7oh zeJ+VCDh^ZKa3Mi%pqrzhAcylH?9;#shbSmiI&1)=za-{=QD-LV!&3kB1Kb zE@~F{H#{WCmZnkSr8Rgl=t&Bd@Hk!c4i}ajp-Z^Z16nYzdWIkS;LfCrq?PZR@Yq2C zrXNEh1K{XZKXyMXM7wRg{IwlfETVLXuH>S%u$d?krw$#xBf}p|&4&sv0YcfFrjB$x z2`2@#K+X3YNI^>UfV;#MEJFSQhhbID%aE-BXFfeKM{$^XS3FUCM+t=lo@*!lv^AP= zc)XepoR8U$UhwH88Jw83QKWwKsYGpyGAlLj%_5Dw6Bc@ zLy}j{C_Ff#z_AUTp!I(duD$Pt8~cV=!BYYv+>#)SK4%C4@}=0P_!{t zq=nfU{dhM|BT--h7M&BMKM^zP>?M@}Ts&ihc@ zdglV=56h$^ndoQ=oE!Zh!93CYJrfr_be~zqj%lqi7_SD^qxrG0Yl(_>1qb=>4s!qx z?JTZ4ywMpt5b1JjDuhAo`p>>h(YlhZasj-(povAjLD*LIPhJI=dNaAT7x0U7l05^J zNk2BJ{LQhsk}pdYat+K~Dh=$kfeAe$wy z)$Tm<^eVZ${{GHrU)#K5*FVx@mg80!aD!hU0lOn?O&+p@>~&@+;8{f*gPaS%&B=9( zmDlUxYR`EzdR&Pa4WNz%cg$1W%v>4iWmTXIVv-&(Q{VG->l!exRmimMJ;E0Tm{xN} znm00G%Lbke;9nQo#Myh6iCuO!Fy$CrjYT!0H?8+WsZy2WQ{yGy`AA7{QUQCMa=|4K zt2XophNm{DQVd9O=R#Z%koW+M0(k`}qWaK^oS=V1-uN#$Aj?Q4Lc>=NqsLpGatbAf zdzr8cL0R8xs4Vgsq%Vlh5dJVT1@3LS3qXr>5V zH)J-3v0b%xuKi!w|CtAf6_p}8ga!meBnn= z6lXLSA1Tm{If~cwk;_CD;;boYj(aFIM1gb)p(r+XE}W(eN)^)})^j+wwm#rIPX{Z z4V_SHU0=6Uj(#My4^|TTHO0y*nFKf?-TLZYQr=0%M#)0sSgA0GOCP`5ZS!y%RsVUk zD=^A8&bG!r>pkH5bEg^QF-k!riK~|^W=;#Kk7xJbqgg@nB!MZs zoj;TgeXJO5hQtSmk&hBZL|WPm!V|}#JIaDOnK~_44?Or3{0dLhfloY?PLpyn2J?G6 z3hX}|TQpV#UU?JMnSX|`+nUJX%=KrrB8#HJ6k&^B;U0sUIp)m2(!=t6sxONw%%T_{ zB_olGo*R%vPFn&?By#7cj{e=qy8ukjA{V7|%pKvWf+7K6fs3g=s^=Q1?3%|w0w z?UU!#nn~QfGpL)xqXrx9QPB{oPlI3581cJJ*t1826@-3Q@cU4Ee-)_2F2jMOnfgW8 z(anSgI->%}K~;`gM9>DgffAMvj29`ODdt)pkTqiwaE88?&^uY&TG&W21F{z)I@O?v z&MJv(utcdS%UEimo7s?29##Bi;G=yQ-38u)1W&56`JOe@<^k<0KnRCQr)iXyk5_tD zAM(nZjCQ(|af>xHVwgA0QO*>Z@fq_4M0yYni)aFbPe9okP%|kCvRc{{F{(Cn{M*WL zTNV)BbrW1&N*P7uGl|;OSka9LsneM&1Y?sMbh^>U?B^XRMA&Mz6flO> za`QHpQq8~St|D*t8IE1LO|Kv>`K+8D+R_kIA)0bx4VG2@CP#k*ZTMyy3+_C?6t(bp zm+k=d9cptA)x*l3*16Y1ej6Ga@t2_di5kaVT$A>yy-rg>LR6Tre$JLwy9~bSMUdlnhuP+9Qxl;K9Cy3Q+ z-mm+b`aKrL2+^lm)i$`Q*f94j{b=0khYEmuvWRH^H#BAJ3Fe6S> z>G&sayCxVyIl>8}7f7-mTapNg(Z7wHq8j>|DUHFQG{bg*v@*yIk$+QT!w(cnG~xi{ z=(5qKuEXQuc2ArG0I7_-UXYhjn0dI zTTc{Mi8P@L3y=|;(>R^TGY4(55pF<2I-_3>x$n2tP+g};JX`n0y!-`=I zbulX93M1R0l_C21OAr&853a*-AyCUC{Kx_O_MEf<8GBb_%?D`XFQA zmCCUhu~%}76>*vvFWAKoH(6(s*C%7^K79Ih!E=5WBwJJ2-lkRD{WAIPRQCZ@U!r1` zkjCPIFln#5fpw0Ry)98b9;{;ziEf6!+)=O7%2CQ%j~?-Dr~FJs;gBC znvpEGgtSqLUJYZ2I7J|sP(wv?!&g}$B8msaWDXZr39X^BBvunU2oq{0m6D$(Mz%)f z8?~=kLrNCGnX{l>GsUHb`?vuZDJx1M5~Ql_+D7j3Fh(l7a4=-@stw}=>kTpu<-=H~_k&7>F}jR=N>wHBkITXi~>#WXN5o3P8QL`c=3C&86)ryu4?oh zES0SIz~>}Lo$_5E=EDR_@-1VTd4%i3|4sK7%0i$=U2)78SwhqipLW`r(b?Q4*Pg2m z-6*m$e1HtoR>&dfm`Y_9*hoil2ShatB8nEMhFy&JbW5*YSSoF#O--#tO}jfgt6u`Y z!+m0GLTT2BfJt!oW6qD#q5a(49J-VBRZB!&n=OfJS(COgr<%eHh<=v zi~ft~S(C&kCqV>}Nk$CCz6_&JjfW~-z50HeUm{-nz zXIx$)Ert)^Cqd2Jz)jZO$kH7OrGX!nF&c%jJcCQ@k5JXX*#ij!8_z1gfz5WT+7pzq zV2B?&tmi8Ma zl<2CmD}d^=IWBEn?xc=YC}|3i2>j_9$ayh;QP@02g&$Vpp;&>UnbGJ7k2wcg4{b_t z6;Y7Qc3nL?VC|Ph6D%MlLvGmH6stcXg}N!P?tR0B%LQXd??x|JKbS z%fM}Ti9AxPU&poGTgi9Oi}1~z`;DK>rm!CF(UsTXYM|g_)YY;GgP?#ApY$isaM9%o zJ5^0+C_A}Ai@n%N=Xm!o_dcjqS|qGV*{w!N*t$0#1ypRHzS}87?zNuSQ2p#(cJf3B z%9kBcy1-aSAm89N0z`@6a_@DNJMQxK8wL@&k0RU{9wVpTMWgaX;}LgD+Wr38Vuveh zs%swub5z47^$i(SasmlOQ0;!So-6ygYpbuLr{JxFanv5m&Ts=WJB$R*voA0vEge1$+Z!HUV$)7(xB!_r|3)E+CNCGhUsLC4hSg)!%0FwOnW2n+T{oghkmr23Ndqmcd8WVUBfGwQtW-~Ii))pmgy2O zHZofmHnPee0jMbM7VnvT`OL`3Xq_j-{51Qj@{I@Zc{WiJU^3AbUQxpl)@oQ~3|v+5 zp)mLDAIRUisGaPsafTCbXV7#got8(3@#7L`nSSjX4_J#q<7;GXKq_+?<0_9Um95$I z#~xgfl7m(R7u{wsY{Zku*nMY-8yq-7@$ms^_mKQ~1{e?nZ-RnDTQDD5`ss!jntFII z4gAr%Fu&Y3xbYzqqKJFpyis>n|FA^M8=a3)t!lGbJT5@mb$uwBK!-3;?=WL(#+7UY z8xWy=>SPb;KBR+NP+`YHS7bGq6j%bP*IF(A8G%yJEc1T!Qd{>M8-Bek%!bI{(FNO- z3?dq{4e$$zl@_tt4_@54uInB--~#NG-d%Txm`9Ps&ecs(T)Npy3)=yOC5)TG{5pS~ za}B=g)w>pb2R#{GF`ISk+cZ)c=1ySNoPv@l4n3J{=%#)hR6%Hmc&-MnBZcp9{i~Ik zv&#h#(d^dLhj_*FDN_AN(XkVyGW#_v`A(8sKXjI=F`RsD>k(HzyD zu*jGXTxV4is5t37ZC8u3gx4e_?JSwJfGRaA^8*B!&H;DjgHbSVKBRR){>)$f#7TkV`g>e5l^*A(zxM`EJ@mE(y;&A(@!pS{ z08E?W+*py70RNgH)Hi^RnH=3)zh_;mXHLzn6vb=sE2fYVbq&t(aJdVW)-e@(Q>cs> zHV~w0t*;O?Jy0Yue4C%QFSNDhK|@bSZP_CcMK8*v(catrYiHNP&&A1$L3y){*E4k7 zPbbw$woqNCM$~Map}l2$cp;G5@|JzF6(d z7P3I9!kjMqsKEx8I^ZihPH8P^jn(&HJ}dihb|_y3qoGe(v^Z~+e4G9KlBlNho-LIE zO7;ZiBFV~K8t49GqS;UH$Oi16Z^VXRMZTP*$Hyd#ggavry(^`R*yS%ShKy(iAguQL z_G%iDgM)`)_-LY~?fVxZ>`na_;`q9&sZs!Uy2IjXx9z=bO=6Y| zNvx_jj16be95GFfL5d2zRYfkXE)7b%Svu*2^NF-MCojmfyI*-(M;bo_p@CIyf7o@c z!hjWUdH6126HqS5P9l;iGHWFSNH7g97weu(z>K*{k$e~{m3F3XuGPWwNSUBh-^+4e zQo~Qq-XbX@Jx*a7zdIF*HHY<^Ub*=T4xH z-=mp2M`S~|W*!>X0iqBC>d*`k<$7x)=`sT6L(VP{QU9@K(#V-$4Ft8 zne-9h!9w&!pj74V;`W?7$8^6egpXIgTqEwv@pg}zd3Jh8^_ zuY^n-{CT2ZyW-uuY>DeoZ06n*_As{Sge1G3(u(_H#iWiUQ(xqG^&;nJb%chWs9=Yg zLOSsxc3BA`)Wq9Is4r|40cjoE3vmd*AybzXH1 zX7L(?+%|P4jC!;4T`t`V8T}(3c>hy58vv)ZC&_CMcV6EypDOV}A4wRW4=|B2!S&a2 zM#6-ABdVaO^>~qzzvFR8M4*}|(AIQr%tli7vR~-UXw=*0h!LFyj6`5gQC#Tj7u3e= zOPANodboJGUd zf27P^+E)Mv%#=cMccE?3bX2@y;R^g5v<*K39@*r0oXxS>;55sZo$AwYjUM^1u8I;g zjvk$2Z)ExW`XGz}I8s?UpUApJTYJ2WsKB0s8q`1z2(dI_B2FZQsX0%*$vN~^%4LfX zvhNn0RYAi^yqb7ITfyYbh0>bM-Voo z>vqY{*MS%FHMpxh?`xJv+vwW3PP@9Roye2&d-Lh5?(jF*Ccl2lzLIn}qPBg#IW0!@N@I+!z&L1UUoh$v za={F@=0tDXF|YYZ=G~|Mgx&RpbNm;Xlg5fQxbFzpbqUcowf`5UbKgl zZ@_@{h1bPYhDqmYE+)2WVPX0X%2PCt==oL47o=AXr$G0NT`6hi-22tWvC>d$T)gTA zx9KL9@X&~0$ETZ`ojZQ&SZYO^^~!EtxzZDJl_&Eh_*hdl*romZQ$E=MKXBDA01n*d zdXNxfwyn$S*L}7F-~03F z(#pw{_7~f`Wy!A8Mdj4gRLjUaf%V-CAiR2}rKR)r$&;R!FES`+weVt;HS&uQqvIC+ zIfw1LWJu3=MnHH)*u%y1VLH!~%ze!R)e^P!6gWli?&fv7a$S$0Vx#hsoAi4e6fVb%($i4qQAQ=v!?^laAWdG+Lk|O)^IB@bbvT6$xS@xYT zM=sNA9w-~~KVCe44f6A3EHF?%JqQr%e-4;N<@U)?LuE5ET7?Bjt6zpY`f^7J=H0 z|D(yrEB+TtKG)&9Pzr|de<#SJVZwZ~-+}%EdjB5}{!7UI^F{pUI$UQ)V^?Pvds}@M zQ|JHJ=}bkE9!^w2XdvlrI3S7tz8wC4EYGc*p9Q7-?+AS=d2G;FDtVlLg8NVK{t5m+ zA^0bR|AgqD5QFwo$&-N63;xF&RJECCA5a1dRAY(?$Lfl-c#aU3)|9;v*Z8Ff< zL2dGXLh(;1{|VJUq5dZ{|Ah9R(EStof5HITJE+YFnpyH67f|c8EbL4N1oXQd2uSe1 zUqEhY>nte!|Di5D$$Rx(mrnS)p++ii5w?_~al5?CLGhevIjp*-8#AZ9D&?cM81JK+UcP3|{b|=5nt5+r)`Xll7IaNQZbotOS560lpMm>umL6Tl2D}tys zT{B#EZBj3F)6;hz8wpC7)d5A^V9*+0!2#v^m4!KQ0y%=?}Fnh=8wop=@FKC z_V&?eD7m~e~3mQb9k~dEdUeor+v>>NYR@nO~?Yv8%7WjW&H?E(Ff86j>C_9 zkuIWPYSj2rCI(g-PXX03!#+yXE(5iGX=j zfdr>Pi2%?xq1P)3rss41LT)Qeze2SI#f~CU^~04^TnD}l6W9e(8~8g8G=^%)11U>> zl_ZEbRE*3HN)#nIQo$}alAciGsw|9c;60WZlMSQ># zIAU=yE%V%SU6wp(%*+1^vk@;Rk~&V>n?M|4z!p$yRbnU`xl?H~6|K5A!5V9f2lmQn z@bq&)nv3b8gE2*`dXv`QeDBrYGJKyL=#C-*+o0}rp@ADg7UnTd6P)5_$6@}kp&1NQ zSR>9gr?5}n7f<@%m8kbUVmNFFh*3QApmqr8sCc;pU^`(mi4~p)8)V&^4=sVLE#12W zN=?9r9)Jr;eTI@qo?Mgv(TxR_7w6YAy96xofj*T+EGKG_X^F8wzVUdw7V!dx^~cnV zXP56Zpq#DF7O#w~(eHJodcg3OCwZ>^2paLq-y3HGF0H5!w)TlE`D}gWdUe9GiUs~b z!`=q69v&Rze)+Dg4rML(tum|K)-&UsR|-(@grP?#gVBa7VLcsTJ&B@}cDHN#cBAPmK85@vbkinA~Ved8?91YYDYZye1; zQ}1t^v`y%v3oUI1I7#-hnmgXdlv@&Q(D$d71*GgxbA0rLuudvxLL;PE*5tn=?>Nb7kHGy$8oyCoW!^UC-=?nhsYDaXZiB zAHx{s{Ef=$!(|&Sh2>HP=f5g#;JVOQBU+O3%-V@wpMjtZM2U3LhWk6w5^%Ds5TERP z`r*p+<3`SF;MpeA^Nxrc;Z0-XHyr?79?~u+LKdU8S| z%?etlugzK&r>A$rxSkXCIjlQ)O@*`&LFV+{ogeRKJa_kpPu6#zXrEo2$o2q*?JmF1 z`$G+#tf+~o)cr7ykr_sX(EV58EPb8G7kdRpg+T>OfiCn^%C9!G(i3#7x?#|E_?_>9 ze<9;q3E=Ir6ZEa>!k|yvf)nh;d6TW(qkUS^-gMZ1;=6q+JnR%|Gsc@^<-H)GPdhHP zweJpmJ_Lp{sXqGS?js2;SPURlmyRIlT93MO7W@f@0NpWJt(NX{jE4rqz2dqNAw%!j z#ZjDBvqlak=?E)%!$C6nTQoi^Z>4Ou&j#7#anu|}B)aF}wT}|v0mvm+Y)4&T1$~{V zdvy4DNCg6mhuG%%S%_7~2VCBX0NLlFOPxq1#ThCrDM}qTJ8N~rN*Ko)_oXVj1Jfa#H5+km4SFXO%j%PS%tx0`p3}k zWQzy0>?!6!>UOyO3SNXhJ7V2i;JVYzs_|3tDnVkApX}tjD7R5-rcZS~9uO)JJIt=1 zVgc)5U}i(aM@9k6PpkliyHT0L&odAU3bwrnBogcyC(c{L(iT?_YA8jgZ`zQwGAADL zF&Z2L>xMaJ7hZMcv54#FKvS{2+$3)??R71o!fG^X_G{|_%vY(^?B?~)t{>Cqld4IA zWF;m3Xa@V;*7q15Pb_+eBKA3^5q!LyKxcIW!FmXC+&r&;1Tg`LyWiFRFszZwt;{Ea zozy4x{y6VhJZ$<;6ufOHl0yf2y2x*!evPP(gXaI}s(qtMb;64_Cz9C<&@7x0!1u%x%79^};7ZgdHYq+^-5z|N@ z05WkDjN`g7ru5hgGT0HH8jC%?Z;B3l4yrF0R1sn5I;!+_rm(gL{d$ZnP-q8c)-Mkd zVW{v>O#@UEvs3@?(?ues!bhc0QOTM?@qmpVRSre#>j%Ki-~{FJCf@^Y;kK9ZR-Zd_ zbX(nb?-QSY`+Ye8s8sqSrBwfh_kEh8hzsjqqX1(^At#@A3T#M-3`a2!oBhUkDQ%#pq4+r$C>NY2UMwS4Rcw@Lti_tle2XIefI|`DjA4VfhldI( zoLdpwE^m0e*?@X-@m)^nv}{>$kbT)hQV_4BoRW^g5QE8Vdw2UsFb~^LnYa`#)!;@>k4kSjl06shBPicO zV$!Gqd!Yq)!VVVoonfi#In-ad82dQ*YV%MroZ5DF*9YkIQN_97DDJ1WpkZ}C50&&9 zoSF9xU;Zmm(8BY^mc4<3=@VQSQS2esBLI`@KDIr$apW%@Ho#s^S6mmX^@;dIX;fg~ zB~eFWP{;(=rqBlX64#CqzDWfr~{iGLnx<-sI#~C0fTw%tJBZc$WHR$l5cBg2dSuuZqDSykiE$A z)Tiz#H`WzYqXn}gscmD1j#nQp5I~^bAYK9V>jz+piv*o^78=ihadtL5WQM5YjeodY zCA_1o6&?TUvz1gu&_YD2OF0r)uUz{+zBBxqkH;G*(km$)^!sZ%D@OWxm`%@S)rC#W z7@akQ@3|>1)|c21EkByq8;iiY9DUuAXF8qwogJ`#gk{1_D&@L98bSJp1(0LKZ&&uG zFx;j*!;v*5l5Y}5>yqZqfr3g+XO=L7DtxR-s(kkRJW7>Fg}Bq!NyJ$n)ARUAQEYCW zT?t$bv)_!tRl^_yms3Pu=O7`imEe!;vv7?RJbD$n6+=4mqde zdzEG#Y?`C=NOaS-^(x%g$e>9$sx!rEp8;&dvX-rgGicZ$Vu=}Jm=A*vX^92iNb#m1 zg(YP&HePhH693Wj3)8#5=ybOTkXNeR0z|>CkLe|&=)ss-{2tBasesa*5Jk|GR5>!2 zWT3BcU3GU!H_Qm_Wh7Yh$^;NPA9dcvh?qIV+n;MsZlDv+ez(ETe5e8iRR{nGw3}?K zD>_+sGbuUx%99_GVYzkcvU^GIDy=kmviIRmkW54nZtX#Ep-23C+%e=B+&IW$R?$h%-K@`6wh{b!U^YTDOdG#49l3cEo@r3W}bOBNVP+-Dr@&1Z%pF?n0 z_<6wr6DTf4QY-K8ctqmdZ2>0$hB2?tS=72f*(hgB`ns0y~e@CE@z)XP3*NWkQxDYAAZa zpOk8pyBU(K7l0~#XF+9dsr5a$*(MWcS5?TJ>BkOai#T(O4~3Jq7RWxy7xstc5Er=i z3s4IMhe@|(qV5osvw8f4tb@ijr2X?SW%)Ug0e5i?X4<35e8GxSD}-eveHif%sU+D< z&KKCbdryvw;I+Auc05!BdkS;eyV0C+Sa(#TDcCJVQV%TU`3R1_11eqe2*aZmy%Q!r+haqmb#a z2b8J3oT5qPvQQXEC-@`2TDQ@91=S;Q*V1T35*%Zr<5WhT%N>LHjZ^rJ$>t+V4 zm@M`a31F&D(Yjj%g@Ioc1t#We+jHpVN4gn}jYexKh2}dvsv&8%Y_#kE{2Gv@V4+y3 zj@lFAEy8dYR)_(>aTMDRQ(Y*FkDxyyx^LoCJ~ByFiCvwIgi!D_M;J&cP@)jr0TOB% zTtE;Qs*e==mTQ5MJrkWjpW7dbM!(9qtNDR`^XmMfx4VY{3qz=E{hvNeD5_pDTe%9iU;fI$F3rH`?=lYC z<$yw^tTDQ!)}kU4^^5>Jn^}1HomjROj>X?hE~avq(dK!D;K2B);53W+KVY1k95vp& zz&vc$uJ_Lkgw1uZvOuR{(|FjWaZr^ijJ5x4=8Ce#o+x&jt=F2zNY)c z=|-!6JrfdrNG5JSO&jqUkvMYuXgFZ8tUk$;Isj9uvtpO1Z|?3{P41OM>)-4Oq}t+% z!p9|%o7t_TSx&VDFdZHYgh` zhUI+J$wB}v#$T_`j;@e3Rn|kOt)uv^Y;Gp)kWfAe^Vdio*uVG?TDF>U?F3q7I?PPB}ZEDW< zHQ}NdS3y}Nz)=)$F8R3NS=+KR_W{#MaDlwZtDC@B5bH{%z&;6i43g?SUSu`~Z5;X0 z-Rb)6l4GuszcR8?^QTTOknbK86&v+he+fmZj?4xTz+UQ-*}(ArUNhtMik91{_TV@j z8r6aaP4ZyB|5d)=?u`$YYh`J zsXD*P4-z5FMzp1(i<2o7iN$h?y>`L7jis$sb#Nq?)9LFja^mMh5A6W2yDu#`1Laao z&HJpfSagcgL+mLEXD5VWV%fcdb_-|14@(xbQky~lzyA*!9B@#3#*;i5>k*8j2rSHkaXdPYt|!!k3gglBmc z1_r$_?-)f}341?VpJ$w(teUz(w8cc$qarFM zgu424H;W($%(XMYc&-sJ?s`f+Vgz$1j4V-$ug=MLObh-bj(3>4w8@s{P%3hjM!+xb z<1~X+e&o&$aYL;wXU+hBHMK7BXV#RL|hbz+8n{KEj~KQ!0>O%2IGV{|>k)f@H5Bew9`##*og_!TsvHpuD^ z1x;3G)-G#EY6llwaLLvhW%(m3H2J=WDg|sMHw1M{Zd+0PwE0>cqr%!n#CZssK0h*= z_D7Fqd5&ARK}rt**@v`9)%p{TAj_?!3PdUL5v&b+&(ia$BIvCb=%OlS5~7@_K@d}G z%hXPQ3~ZBvQ0eEd;v1+p6DT@8H|(e1$l=SRK$&C)7vd(nK-u4+uRw4CA>uA`8!e*H ztJFM3sr#qrUSZe%x;Kx+d6AQFyv`H)CImV+J{Bc}8)}(=mSefCfL0UE7401)EB}u< z)z}#51B%xjJyiIQMwK(L{&3P5J${9hM^UX?U!%mCWBqXF7B-VIlPCh$@#TvjAMwdi zShd%RMhm(ZQ1IuoQjWy5Rj->2a)h1ZGog+5BlK(rJ=DYdSJHc8g}Fk!FxtQQ{W7<- z8ZKC2=wUPfg;ySl#G9F)FZ+AwWzj(!)j#-QFz7G+B z|1q61YFF}Mvs*yk-@bggGzbm#B?<>#)xoUKv#+fI4El&EpSH1nqESX~xoWwoY&A2S zA>=vG#=>x$-I5(9MV&RqUW_t$o{Q+5(Fn$-ZqZoQm-oIm zChNlh#glVtUFxUaDe^aWs9T1u$lB`#IkhelB?S8QC)lYEsq9L`asVzrh5fu|QVtuB zuatZR9(s^=(hZ+9<%UZI-=E@Hx3GfOaDK^r<2z5?i_bslvY!gqHFOhV$1JL*TGoLt zRn(P7sg@pc=zmOLnRy5Sz-2f9W7r^V49bz$b#e|W>o&H$rwFE8 zt1N8d6KN;WU+_uEWH^$ci20hn8IQ+^69+MZbs^gW8#Sdbf5npCKiO3-9#)vFa)Mq*cYrS`Z}ti z?mWNAfZyc-c)wNj3JpNo7GD+rsD`A zUO`zboX5Idf}$h+Wa!$Gk-y125b_iNWaDgK2&Q)rq%}jgHoEVw^}i;k+8&iZx{L1J z!8E3nA@Tab%uxUg^%X*qgY5P;KI12QKi%kCv%Lf-R2_P zrS;=Kj%a6K)4`kPMdY}Y|D>INhl7mYq#y`zv*4Zln7;&^htRCQ-5!%`?XK$p*ju{0 z9SenKXg!qB^SXfV2~}MWTF{P*31po+cTCEis_Y}KNCi3?o450;I=`YA;(ezVg>N0wK=%$bnURBK%7(;A3F!zURqwu z^Z)++nik@5#9Uu&YALM(riX;<2xoI+rE7@~%Bt)p==Sbqy_RHG`L}C)PxDnsgaPc$F2@tqgA z*PEfj^9oCGtq(Q6?W#m#c&RsmVj2QN_JvQ7l?EF}A_>MxB*aEr2~RaqIYxI5onXObEs3!Fsedx?&!OOAq*AjG3X$$fr0~ zDU-}C1xrlE|7hVV17Zupgb11SwAC8Ya^LIiN`_B{MKmbY(o1C1S`cG|#MsjsJ zzzH}ifbsMnXj4G)75S{C(UM@L&wl#0g9Z7jVh_=qPLfw6va-$tvuI<76O_HB)7PR7NB&SwfRKcg1D;*PET+AFXb3nm{34Zw=YI}l9= z;+Z8$)mQHta1t}g2PoKe;JT!iK);7;6&-66UVlirtSQazv3)oUCe902uM3_q>@9Op z&8HSBr*8oU(WYc>U4HvGU3e#Nlu!(O4L7SKEJylq^Ai*1l5Q^AQv9ASXx<$jvR z-3mNjKR9-dNftNWSZb(@sI`(;)C+c+`L_PJ>=|DG7Mbv{KsY*#!^D?3jm*RG@>ijU z$1S_7bn-_HhF5hQHd(!vJC~-(NoJ_ugL}2S7+{WTUY%wZSOii#^V)ch?__Onvm`lp zd!eY|`@ssB^2b8(wP|(x%Y+L5s8jI4NMNqY?-~=q;3T8%4Thz{kSid81Z2Y-8@W(_ zZwp=ka7;pJahXW6TAeYhl&qRM7}&~X!@){Kg~sif*lsK;CscCPwh^4HSv zSfr5)E@5UcJ!yjzt2^7gjEBal-_MazPf_@e-I>YhvrqU_ZmY?J z+^0hU($#9w&Sf+<+L#x+UbTZBOHWZP#;M=RHF*#c$`;7(#L|Vy@qRVHw8IqRoc1Y` zS?xkm53o0a#|G6@+MXsZ1@A`pKM3-gL^7icw+bEJKVt}d2*%T;mq0_r#?zG!qno+{ z9B0~4Ck|A(?y~MlTr=FDy)7B5p6HPhoeydrYpMl zz|*<*lP&GL>$jJbTB~NK*OQ7*TA^P9jge5-7TyEMosod^Kec6s`g{GGtjJyZ^SYK< z=nSsEO!Z8-Ia&yHo%sQaQUE(2ephunKYl`6My70axk;zHPJ=`|M^MaeT@^XB8%Lic z9($)0OcE#*56#YD{HZKFpTGTp_jGQ}&Q)tyoKrI*LaPs*=}ugH(#G3u=p48kt1a zEpkb{c`G)Rl7Ls=;ffAh*0Thac1G%Iv7Muiz}-Q$j-|y5-6Ix_fQExvy&?8XbQYTM znGEcDz?K@Kdyf33k4iS63L@2qOF8Q2U6j7Xvwmr0uwtKy>9Cp`6UQ?p;{syQkkZqCZR-ur-@nzg^+KO@JkwkQ< z8Kogxwl-=?sD6Wo0h!4P!H&Jdd-u7G(?X2~6wYK0ug@6r`|Un}Q(f4IXOeOa%a9Fo zbI=O{Aqix=s5!$cJ8YtXi|pI+?OoNa2bz57Yu6yQmPIt2QgpFA`=Zd8*1J%QiC`W< zHxRs$&nJiZeP66Kl?$sP26RKHa>gf0K!Rz`qHaQXV7{d2nU$=UGMAehTB5caqbC0 zSRPUMx<5QDp8@v=#MpWDY*p*AcZFKDnYg-zY;oq{9`N|;m3As{Q{5L$oW+%$^=CTO zhJ(vcXtdi=2gtU=p$h?K;WB6(7P^qiklJD1ylIBf{y8>~s=7(M7_e35pu z4E#vTt+aM(G-27Vbme{Rz9r$aq6Izx-)f{9{fHr@TC zX`T}LGsGtJy4yuV(5?V)Gnu-adA^6@abLMeR(#F4lpF5f`N0ea@c^rIl{u z7jt|Q^lT}VoFfV2sa6;bWvO3Jne~@Je^rVFZ>>ayzq;NHn+`%9x1udBQl*}d+Bmci z1>uy~URqx&J$%F2xfyw+;WohMv41ap=fOEu__lfe)5l+}r371R6 zq;o7OmGZfuT6GRqmSMcq%zt2{aG z+bGS;4=hgDO^@B>8+rS~c}gk#D=Lp{)s@`$FP);z4I{atL>%M z+fJmwKWCqvXwdRpjB9^-Xb%fhyL%p*{9qUTT9~_(S~-o-0K@)~eOPA5Th(s>=Wk-n zFS@~T=9@@ck19-1$A+r%`e-Yl?W26Y@O$^Z2u}Rct&+hP#g608rVTB!h`nfclK8@s ztTSLyx)3uleQyJomSs(1k)6?Zl7AOc`0XLDtyMA~CP+Bk--f@DE!!rfe^$cnKH2wP zgn^_F7ix}{VJ-0csXUuV?}ohF(3|h_bRN|;`H?L*-G3=9?w;ns4}`Y|4*Pj@7wR@v zP}WOVA9m+_c%J6ip=6UTKH{ihZIHjFx+d`CT3sxMnW;1jA!FtiEj*bq`_O)f<3+&Pr182Qvm>UzcZR1N~vrbTQG11GHncVH6Wny+n^1%iR z+SRwSL#KBBF)I#FwF6g$1u0vQ}KgbiC- zSfZ~#JX156VQ$XCA*`bnfJP8)a=#nLvs`BLywY0AlVM5JSDu4WGp2)&)0do)6 z`&#?raCr`7F|)l3`Fd|kAXH<#jD*ecgANidC65JZ(k}uJqEs6d4Hm|ZaZVW3PbXQ( zr`l4)vtI3$Ox+aqL%sKIm6C?oUqpW3_$8$Vj~d)KFqe3AkE+BgiWu%UZqZr?2w|m= zkQ1@IkW*3mw2zJcX{iC@q7 z#ohKMJT*e>LXD{eeeF)dj6^%HoxG-1r!wiDW-1fkxc%h;hK65z+Y3dK$F5HeN6TM- z;>+4^NJxx-9vRE+=StX;eF$^o-AzZNNyeSs#?4gGNih?dX%Slz6kroDSX_K9B1=)s zK;OL!UgqcAP4fBV>3%h7(6lOD^#xaw^Oy)z@X|!)04dMhE<{dAvWS|#FsTmL7=S44 zflK&RPYK1p-4FV1uVg02pGES#BhY<}>EEj@J)j2l6h2z8qxb5k!eDJe8pU`ER?;y6 zDJ;3fg&bLl>1`6~$rL!9l^Ei%7LziCa*70Mjn%j2U$Q7o8PVzrgy??R2Pc=U>5OO| zM_6T`Lt+X92+!$+_wKI7?@G*|8!OjCUCU>@78@O=p~+ryE<#`PAKew=ijB`$*Zv^9 z(UFf;V%PJ@Oky}953qTN4N_8nV^f-uzdp|2(<5=;_ERTM&7e7Ipv*5_)V1kr#1Mmc z=Pw&uch_CtQX8p;Je=m1jOP-KB^NlrM-<5P8{P)Q7}kEA~gRA=Vsd#o*|+P3>c*6fqC>CrDjAq9)(cAPAEYTS{+^vOLA zm&w+SwX9j-afo%)^GDCu{=4>`;##Q)EQ)p1??n?B__%nYbI*lvF6dkV=)=KM1+g>Lg8+uDgmeq z`QPkw7yzIq3;JJZtM4LMQNu;KWt+LkD$#wdNUU)pK8ULG1!6irMFsA4OpfmXe@OZ` zY?<|vjxv}rJr2EHF^K}}?`l<*{MV}3*s8H__)P*BVcqkvTOD#2=W1tXu|Pf{T*2!8 z^~c%Y{t`zWLW?lJ6NmNXfWNzCO$Z0xBOYmaqN(W6z~_|Mnw3si=fR;EKj(4h6XB) z4d#$ltoVjuRH`%`OOLi!va|S1`io*_g3Xu{nXAWZ?fWj*`0d6S>b#j>eMRwR#EstP zeL7o7wfliLaJSCFK@-%_7am!_drMI#)T3e8n_zz6gG}H^{AZu+3SnBuCn(&$Yx_VK*m(l2-EI)MQ z4YDbB#j&@>nR!@sZl8SrY2D<Used&^ z|0f%G9Kw(N{9papRVc3@@l`0?{{WtU057Do3WX1{c=k`TwIVGtb0H@x(sDBD|LFFv zS)7D${;TflA8|K}=JfaE_w3evgyM;g-3`A^-w3$vB{83+m@#0Vss= z-;MTb2o}FKAo~xHgLJM9$U|CZ{;B6DSwPN@ii)J*gN7vgKj^h(DnOY3UGFD?#sADy z{0Aug1C$}1KQmPz>Ztz|-5uB1wfJfmOT?B;G1s7t(1J zuLn^J{--9+Aj4EpGcwX)Bnp!FziN{G2fw~2m;FE56*%l&^12Xjn(W_>oR=_A1tEzi z#M6g#65<&^LYx1onyn}OTf-C?Y3vvBfAqxvr;q<1T>zAMCH;s2CU6kPn2 z$b^LS3NbzEKS(9Jaftle*5q0aX7s2W>ILD(BpqzxG% zk`gM}6Tb|mj>%?zEQ!;ysi6{@s_gix&FgGlpFq#<^I1PQ~#o@8^dR?U~gcv!Ui25*FlK$uwW{uht>OGm9$mZ#Ac0+Lo*tHDUTRfs{(C@)un`Sa`2<}>52^4Y z^CxMXs7P$0ay=*`I{`+B8Q~&XA=f$K1Tx3&U_K86J^vimFC8ZCdl=ozPg-%L%-RsW zeh$_F%;*hv!Z(08M=TIetOCO(t7zjzB?O(=rw`VWWA@}zl>0f z?qx&JrT0+Y3?ucUbw!z_e-=nm;UUkmmCYcd;1NMXA9U-GLPExIafu_A5e{d>($0L* z6!*2!sBqiV*I9}xQSiNxRZjEIn*d!gRL1^(7;9b@H2h0_S__}%>#Y%CG`-aV@#f9N z!9n2))nWh(N$t#oNgg6E7gbOAqd$7vecPM&gv^UiZg-A&fRwW3#~kpc@|9b3nfs1g z?*5yB(9=%soRRI)#>^yf$qn_iBi$Eq%F%y!)>_^Yp; z!tj(PhpP@>jv0QGZ0GKijcX<0`_O3vFli_xz9jWDkJxN3$CBCDFMo3TX`Nin9O%KkJ~I@{AvH6X0*Ba2NI zM-;Sf0P0exTnR}djlj(&IgvYsaD^hi!uNLJIz5lQ=y_i}C&fv;z)U)hNDctp)0>a2hq0x0-=8 zH(t)g1h4*-WENUH$7iI&i=thL`YDo{tVPO#A25%5{_;zU6|#|FNbv40ic9r4|L9Fb zYrw5=>aw%iB3{a&N;nS?EbX^`HlW?B}lZ&Lm> z?$w*+g)3?=w+qkuu2^@`@H8a;d4Ha*?bwyDl1ij<%FaaNhz(py0dyQAc`BHW)b~iC zPohZrh8CFU%tcw;`|{sGr^(K|kb)VKXAAC`ELNj~BgQaWoa!F~->|Kb6wuvO(TDp5 z*mvw_(@rTf51-(kcl7|=dT#1*50N#s6$b7KxxgTJ!h54t7zaHTbY z2q!AmQ&D(}U&_;15vY_9CGGGNs%YV$-E z`*DO;eJx7&TUpswFIHf?D5+~amm0n)HEq;6ue;xaQhTS6aJwj8-&6}sNTidPb}iS4 zCZ;p56UN3F|G1}-ojr#bhlSxbFi|o&)h%#0`kQRxYCDE=H{{W|d8wK^Lgx*S1@o_u zO`|sMz`#`2GlNU&$uJm>c&~)A!WOB1GG5PUZZVtOrMOjZ5ubOnWU6SQ<449C{>9}wE0>r!pFKMM{r>A` zDJ4Nj^4Oo;Uk4t9fw8mBA&K9|WuQ^1X5t<4uMXXf7ZM5Qwsk`X8)VHk+aMUOmj!9bG> z{FBOl<~B%(8A{CzY{V-NZFCm%%(;+ghAy&?81uXMj7Z!hNM3?Q)lC%5@ zj^9&*?yBGWB3Q9kQ+~aZ8T&Bgs14yO%%!Ton)YqO$!!Q}NdH9W_6iA00y5GxQ6jMN zn7}`yThm;PI&U$Q8%<2{fx)QWue8@FA%0{LW2&d5v%CLU|6+7-TbpAbGea$-Aa0_0 zGUu%(btUyDRt?zeHy+pew7+*~Dh2o*OZ>LA@e9oR*_&8%(YM?fy*1oxzAW44j5S}X z;c?%UtI4!hPn|Eeuxw%GyxbzyR@3;NslP9oTIGqSfaVCTNFn+4bX3WmCx4ztktUcf zF6V6s4k4eDK_5id?0KoI(bR~;L1GoVdNtWEckZ8`S+{RtHC)Cz=;b)KEXuC{>ns5N zB#NT<&%m9f0a+!BB-^(}uv_oiQ6ja_Wbn1&&(urgB1=uEi4ck3Wecz=r z`TAn%>*ePsur19u4a^R_QE&2DVy!AO|KNxcvWqyZ?2?n#z46(f5nVLHf*MrIa5^#> zr3AmB@bi!VQ2*21xM(gwxyaF|VEE?pYA_^Gkk=aOWC;heV@ zsueuqWoHjx&llB5yF{%NlQo-_NP>Jt4B^LzbX*}ddPF5e&xyVzXjuoc&jzJ=+tY>T zyof%rA5xW*#U!+viYibnx7z`~&1Mf1#GGJesm1a6YsSH(ZlV`YQQpd$R-X4zI4IRw zk+uYZRxmYQHY@icqSSR}@txtt`{(DFMDlop2?ExdYw?NO>fZ2z@_`k0!{>s}P$IO? zSXaX@lAK*rP0yyRi{iRpU9}NmR0#CY@Sb2-)q~NjC2Npc{MFC5KI3tH5|3Ap*VmCe zQ?2YK3mWNC`ei=o5&9ARf6>_$p{>mxpU2Hp%S#dg*A(hdk)u%4b={+};j}r}d)yd`W69#uH z^}0Vt;ndjDp?&tdSYPT`^uNJ*6u&*f*sNBuAioGhDnR4yY=3jxOjUsp&pkf ze%S=;!caTkjEy_H>F`Zk+qzsX$WeY(aks^-(hF1HG!@7ieDHy6K48-v-)~4-5s(lg zI%+OWL?Z|!SQ0PETdhwYw$TRmc#1XI>bcA|OE+w>6D*#mYkc!^DqX~T8*2N`72fcOy$wB-=7 zW5E1zEht~xE#OTP>&u8&^!lM~ich~iyTV`;;qX9AfSdjLLt_4kk;QckbN|l!@&NS` zELeHKRl!o&4-T2RkGsAe4x>%JBl$Gpu$8EXdEE z$~f>Tmj)A`9I|7CAO(nzs*X7P@{+@eH!ga~{Rr_VEWr=C4|pyY(xg*e;}kl>yeUkuU5?pWBBnmJhyDTJaxi z;Ve06|I9Z)C7HADY-@8JkB?LDzlo^((e$~6nfHLex;0rV;i%nX+b(%tFh&b{QLYIC z_QTj_gIMtu$Ef$_jFZQ?k=`M)FDyK~aRia3pmTL9wk^m4SF$ZDoM;3o1cs(98%DM> zn>0fO$_CsKAIqWE1T|s5cfyyO^~fzm{eCMGs}gjm+Pn?-ETDVmT-@ngQ~WVPj`P`zxFnflUD;=-k~?-{n7qU z{HQ}IT{Y~uO7s|-y{vqAlu)|-IxNS+Vs^yl$O%Tq#Un*{U6@c`V})1#jsICEzk(lO zJ)OCcR)NLHF4dz`Hmr^JMx=QwO&}<~>{CmM@RNk@G5%;n3zp`~J>YjC2E&*Uj0%}Z zzgW-i2M?tnt&N`iJ20I(%`w9AS@F&N!ao9bVBYQ6$dac_ zhg8Lsd%}5&ciY>VXt*6dVX@m1EzQft=P8(rU@{G3|RbE?4(-KY@Ur*x~*@ma&z zOD-Ryi>#;i@u8VDdpLOb^=b8?;buShPU@Dh*=~RQ&At~=U@iCCD*g8rFnvu{B>u`y zvkj$1*0Y>zREDMT<8r^bTN0sppYhf?wh?NAS&po*1n_eF7SLyPsazn)O#Q}isfMoq zuIg5Xgr~?xj$a#-$;ND&BS)r+`is`!b>ddf6f`P|LhG4D z9viK^b{D)T&tV+7E+zNu4Jg)6N4c~C7a?&F`~mzqEIgI<`I7@FDfV<^PwL)K|I>?GGODe;B$>_+yq@gGgj zh@hfxPoOR~gEz=e7m&A~=i>SHCo+dQ<=%z;PGo5j6uzr9G_JYwzGX!zz{r1J!xV|>x|p7a5#9djrj8ziS6E` z=7%nF`?KB}vsl>S+Q}AVtDxBuE%b9nTxh38G$Wh;t8eLVt;k6~;^;#9*k_wx3YctJ zngs^`xlR21eEyG7gt{>)vciXXKlB3S$&zPdj?R}Tw8vWFi^YD(G{tu{FJ7ADX@8l4 zOwq@@E#@d@ucW+A+oC-Ud1z&_`{6#>fwX_J{g$1HwGo3Brih3BvMMhV$w$DPG+*(T z>k10_Fx`jewZD(^iHd7D1Ney@21}+Air!QLH$ zxG8L5tzy!!RG8#*vLAawxvlFasMmw=L#OT<&e~++kZW5D?wkCq*ao`{$C>`AHHoly z))~=r-|_xxrqBtZqjReiA5cA7hlZAbJB9zxnej zU56;62Ok^KD2OiOzEHiDt68!qn4B7qBUyaSomKny-9&P?#cK*|E61E@b{e`n-?I-K z-VKc63tL)Zeq;5e5uGezr85rW6VZe(#+APE74CaYO-ew<#-bMRxMRicu{C#E)FdYd zl9dZ`4Afxsem`rgkq$2kvg_hT1#t$+OMw{iY3UZxx*Vh1-R-N@DXz zT+D5*>_vV22X)?$ryngE)8gHO43C_iWW&N6c-ae)v;zI0B5t{Y$5+Q#>4v?p4fG{L z;fm@v*J%OU6Jo`#Ek@3jARlivZkM`Sgk-mYkhLRh<|&P!qa&ZjO8*QReA(Y=N#u~# z+?Ps;n(!ju0ui{+@x{*8EXtntbsq`>w(np)%JZI?a zy(bmy$!?HTGLx3XDBVGP2YtTNk9Qk(qj?(%a!qwZk?4AHB}Tt~o6t>Xc1H0V7UCp3 zoVR?~(_jexh)OGOj+k|+THAo$#1Q(o)Tn}2>+brdLib>qgn5tOpf=AA<=!Cp`5`h{ zI=NN-AA)_ODVs|zb_ws_;O%^_Qfc$sgd4G07~yBr4t@DE!H~4lbG^+6r4ZP}x6S^c zOzv~nQHmk0D?tkGv3Q=hI8g$H{u6j1(4amy&|_KC>mY zgR%8y9=w1yz}i0rT0iM1=%p>R*e{+uQW@m=!MnDitTy8$q^k$IioHttsdWPnaV(jx zuG7iE`OS*;G0^)1^`!yV&Z3oGGJKOhGYSK^8%=DzgQC9 zB9n7<*(KZ}`IUtkVeEt>Equ>@=7JbE1*7nIo!1)kx3@@q&e!keNni#Z3x?n05jZm( zXfhF0<3Ib0e%zICadNHAARYgEPW&P7i8@C9lMRXt((IU*p$QxXd3oqQs(y|ZnLXdB z!#yD)2fc^~l0~hQ4oEJ0YQEpF%YRVfMZ?x(`_PKrz56sDg06~v7E{@RQQQ46W?+?T zrs6~)&MVh{%8*QDQ!jl`e2TxXlP*KPP&*w5>Wk!Ez&gB4-;KE}218LU+#iR^E)$Kf z{W)n^A_?9d#jBd$UCsc2$7LxJT={teYwT)iza~P1S(216^88VH`h#6X$4G0$T)1+Xo@sDq=k(&3ApgGOWaDx~%RVtX&xU>Gt zQ)gSfB#4UfSdCIgDViYV-LQ`QW+5x{TV?i6fzi~CpZZjQuwJglYlasBS_!&#GvlUh zTgkOvS!?KVwL#a>(p38@GWYoH;&d($AAFsAfJRSJ!1A^-Q(rXrzOv16)!9<8=kc8Q z*(39|R?=}45fqK;wIc5Iq8!ZQyC+_@q=6!v^|RHA1`BN9eTUpXE$|13L7eRzEe0%6 zY9DsxosK<*p1-ETi{M?q_P%v|U^6I}8%=>rV2^DVtcy09s=h-d+l4%({BdK~Dw#B( zd8~E*7gZ_QorEi@qRW`+t26gh$sUzsgSA5oR_BLFlz?729V?FxB^UwMvz8ZI&Bp{U z4*9WVR1zn&-ntl1C4y4&KZsUV{|cBupX`#JR4bx{hs+wgsH2 z+@AG&Y;i8erV3w~G(K?&K+`uLszzc!6QtxN;>l3Y^cCME#*jY3aXS$sJ0Y-q=rEqgil}Ew-5~rPhU6@ z;|J!j_m{2Hi~itVFvnr^Sgq-JAah)t!*4xRttP)Dxcw9 zqL&7&A;kT4+bGNjsZ}u$PlD|y1ZB?5Xq?5))^DgJTB<2&B6W5B9Nk0^cV+>5*~bYF z{?6~^hf9g|_fxKHMUliNfch;ohLl)b!R zfS&r#v$h=))$wHa3sQf+V{(gw`RSC@g}U&SlE8I1m62!lUL#D>DGV)$-h8SQM7uAp zs%g4>KOV|y8KG!o@!W}456`z!Xv1|vjpEreJEV>b9PfxvPnAx$T!mV2TU#2Io#7dE z0w(lEt4y#(JFZkVd&YJ#BZSH7OG#JXd~IaZU^<;1N2XT zmJUjFjnQ2TBmx0Ng+*(QCS{*ZzRRKvxsr;ZK|BkOfZ-`pIn68*y zmIavvSFI6qB*)70K|fWg^xaCw zVphiFmT|BClMWQt?y!T#7w1UI4(W%#)kwxmjyU;rO7E~ozen+ka6q>38t39OqfE;N z$pxH}Ng`H_c|CeLcRg&96%)8c%1Zv~HZ3U1apV3faAJciKNa*nW`kQFIL!P`_Qa=)1s5 z)y6mnVPC!vkMLei%3}W(qD99wX7R<1=woL`5dTLj!f%gsp`VY?sS7dVG*2oij#&L+ zebPj%C3j>XN*k-t@J9Y~x>7aI;@T7kL7egV6nU(b{rqAbtC8=V5m)7Akt&iGW`}3% zWhx)4gFz~e@=%fYOjk<{#8bffofNmsX}g`2_YBJ~at@R%2%(cFxhFT}V_(4Z+yr5M zRe$iTT{QiQuYe3LQ`qLqWkmR zX>Qwxl}JCJn%Be+Op|2rK8-KXiEJ*xaWU_zj}Jhc~ijg^$_tNI+}O=EJF>T_6d=lV(9RhCqYR1uXLeSCSc zbV$-KJ4y1?yhKRd;9FKkNtagY2Gwh-FGbTxg}P#t9z#z;4h6aGcIUu5mV;o%jGbeY z@VcMR@!r2+$iUuPUXCH*`xC!ofxL@h!IqUSwv*w;#%O5Bn)*rVn{ttwv^@?wB&>Q@ zJv7RoXF~xN7s@f->?j>fa$8jYP1obyZaiK@`Se)&Iq>9V!`qve31M~DpUYzcZ@%;} zB-~+URjaIOTS+$nEoPCv{YT-4n&}jT>L(C4eh=xV*iI>JkSEXCCq4AYw;9kY8p7(ssP$HhtF!dt4kyZ0pwJdsjfo7GOA^Bc*+^ z4|m-T9XUO1&{pv?>sAZXTh^#)-LAZ}Ax887Th-r|8;$u74X#`s_B)cPlcTWmC0sI;Qhum+OQr}Aaaxu%DtBWD zo8drbfqB=k4PpLH&G$rcoA_=kdM#bIQJQIYenwuhkrKV|VbtlD`hM-rHj{sH{}L|$ zr{f1J!J-k%w#xdIWayvTSnR1JV$x4t97DYN?Lk<+nO6M;R4l^OfF6BIZiHt{V;tVj z^;exvHTzmG#x-7!uWHF#f=B42)#oK-md->+Gq5I|)nUJ?>#SO-%OWC=OqS&+%I;>K zk@KujeIviVYOC(aXJ;IHuBBakj8@OJd=t4op-`NevueA#2);Z_P3no7v_Qr?D0lpL z|72TriB8pV5uWehF(lvueRH5X5cS1=B3iSR@K!zv^j33DxHT`5b29jx8r85LP}*rc z2EZy8l>K##`+JY5JnZxErffHaX5U91UPiQ@h(nFxNAKl|j2sf14vxSH7cDa*C^lY0X>PJgkD{av=IQ+*oPJ#mP%d`h!m9Q9O-Ar{gn| z3WgTw2Q(Hs$7mfB7i6HASh#?wCt<&}FdjDaYtbM&CQd=m_vg5gS)(H7^$|Kg+HpMt z#mBFXTzt8O>%Jqb)~>5537<*`TN1_U_g4mK1wTa#j0k>XsP1C)@dYIb&dB1C6(_P$ zT#}H1gV*P#fw=6*P8%@5 zFa}*9u83tI$8x`L`#9Q3f$u%X?o>G?EQZ`OC2^T1OZdkrE=n->{^z!%!M(U~?3Q~^ zwrq+HU@ABLRpyWcEo=UG{dV?_UNVp1T8ii680spVT8Bif;k9R_zD$(V;3PT&drMXi zUXKg4;03a(GE;$2*|M+H$Fa1%*rb zMxrG}N6{-5%V=IwcFVHb>DQ68Q$9o7VL#I2FtTUZt~?auJ|fbN!!*#>R8z?ke}dLb zVb!d%&$VW_SaU?HO}n2(@MA$lW9}HuZ4{Uj^02&Vo_0s1L~v!r4@yEDROMiM$e9o; zPRmUw^WOgPdB!wDAxw8#8JnXj$tXn*UxuizBKce5Vo$MTifGFDIP+H)#%!l28RO() zltMDxYz^-CsI!+XUmg|?FSYk(RZ*U1==x|+uB36qYUu9G@Es-r$>5xxxiG(_+W-;X z@CK&-ur-zDPh$>oJI;gx`(YUIYSMHa^Tdk-qnEyqW>o!}`dXO10-2edIPPneeG z7hfdpDB3*YdqQ(BTrq{iJ+?Mkza-_Nl`C2B_M zZMH7c_$SPlf+LpBoNR`IKLc9}ME)QKdHU9=rHeDY!f?CDw=N?ZPJSIZlpV_4B_buw zc@wS&sXy&{PH_&D4BdwE{N_f)X%&+$B<-D^zJ2t4Y+gU4qM`BGg8f}RnwlvxnGT{k zewRyK_1u(p!C}f&eSAUF71;5sF*bvmIL?J2r(;Us))n zQ3K)4;XnK!)nxERXJs=R@n>G#Rh-=at4f0#i3JxMr^R-zR0ACiZuJh`lW-JD!KtpCRdVL3_(^{2Z?NH@3` zNJ9VW_T&1``~QO)r@fn>i^ac&DR-8IBthD+V6GeT(-Ya3kCfl!`QrnkmL-DVsy-j|}*bmax6D>FGRHf&Bo|}e0CTDO|$72--RwyGb z+GOO&wj0~L%dIg#x9qA3{dFr-U$ZUNvXUtMNXlc(dbrzlX zs4HxKb#dXBcP{5cDS@!{k-_B;ymAKG;>nehMJ`}OOTpun&~JRBUBQm{rP&g{gEm!n zqqQKdZQdrjT0;JaAvGGl%S%hwFwtud&fgaR@yG%v48H`Ca2ZdQ%I+qZZ2#TYv^*~8 z{rQ%>tMjqSqto~v|HQu-wuMGkblFfWLk3n6zY)%Uq(hIA+HrzG6k1N6)sP* z1H$gKNN(XwK~)k*wx1<#6*IDR_-ywMD}>Jz9~-C>TuJ|yPrnq<{dHHem$hIzg1goN zjrWOgN<#jfz_k}!i7w8`;!F`LU>SDPMvs^t`{lF+XKgG5wyk+AULZefeKhv&C2t%o zZJ;_~eN2fSo(T>ib}aSXoBHynE*DN87(PyrW4u@TlIM|` zOV@iQ_Pq0VBy5?;C7@sv(%ab)PbtzKJhX7dH9nl9-7z*|#CX)k#RbQjHwiU6FDO2D z&`WIPx29l3SDnggZz{fo8R7@cuMN)?P=(6370q=?cuQo9GcY(68q%5!6cee@CO z|FNE58^qLq8hnQoPxJVje{b^-PyF3Zf!cr$&+)pe&|R2vxzJxhrrYjp!?L9hwT-+p zV_uV&JBK6IF*DMFuD&;0L&+Olum|;ze4o;H)DJE0;c^G!=83GoG5niKsgC*!xgD9AC4$v)N1WG3J%lgItuXq_%mOkRuUfJ*sML| zIAuwme`We7p?l{6*!7lZzC5%D9(CJ`)(Ch<+ZRM;Wi(50q2Gav&bb$#9AXk9gxA%=k(kLo7*sQNqUqNvE+%v{q4U2CIU|s?*9r;xXTDU{|nd> zc%pOvXRvxAMV=V`6!w-p5qN4&o(O;HPM#R?G&GW)KFW^1Fp`x%W}m)CfkzH%2^EkF zw}cAI^;ts2zaWp$2k&)=(2UR2!&; z9MlGCBNuK1b&%_`fx5_{+Cn|#ptjI=a^bd6f4M$eXt0(XsvR^`4r&L5%Z1xPqviVS zpmB1j_Ru6bs68}IF5DiPDc5HY&6PuSfELO@9iXLh;SSIWxjqMIwH&G=v{nx42yK)L zcZ9ae^*KU2RCPs(-{U%1g`*%$83J_Px{&|mP-ugf8%tJ{^f-vW; zB_=`GDA6U}_Qz|T@@n6|A(7DTkJmo+)V?<(k%h>|1~@}Cn6O1?*R7SE&hMF21A^y zJ8yf#)#F6|^>roHy*}OpCk8AdtaG$GZ#UsH7FO2(b6~;BO8Wf22eAMBE&oS%z_w~k zPxorB?{x2x*&k3yVJI=3@B3-Hw`$B37gNGj{~H)nSo{$Fe+pp#2etmUrQC|JU@P`0 zO_^O^8QtozR-93qvbz21$8^VtCTKzw{AdEf0!()@1Va>B1S`aJXF@PU;X<$?Om{&9Llg-F zE5?Mm%Ohwprn@?VmSDQ;A{e5WB3LP=yDfqtiYtPZVY>Su7^1)s48(MgKrlp!N3e2C z_w?w;(xdUrAB{(RAJ!+zByZ#tQ%f!&M-=fDR%}WsUwA1;QnOBp5p&XOZT$-Z-i{XE zukImzO@>Y6IUzkYPWI^6sEe9e_H##S|Ldv#F*;Y6L=0f{XVtz)sn{>qp3?kOSY{xr z-V&YbPa=lK=Fh5gk@9K3Tz^WlzOc+eR(&Ztw-vDiaY6SYrFy^IcuKRqu*_dp{U$p1 z7O{ifpH=T7<@0{I`IKf~VOg}SIz>#bFsT>}KwQzkNU7aF=vZ!HzAB{p;K4mS#4K1u z73;yzYF4xo>mf9}!(3Lh66Ya0e9635Q%$Yq_0E1urRKO=i^$GN z%7~_#dW+c3b;_aUxO$7k4pJ(%mfD*ZsU6HzHm&hD> z0-R}srT~f#Xa*4KfaU4JVhp)TkT)arr(K#wjM2u$mOLBN(S2nzAIt+#48VMV&;Tp|=ncR^fX4tV z0;CPVVnD+HECDPGz*4}&04xJS4L}e`G62hgLIbb@s5Jm9fgS_!6EJN6RsmZEU^Q@N z0M-B~hTvy_&=3ZG0q70ET7bt8tOKMC!FoW$5NrS}48cag!w_r&LJh%YAjuGH0SXPl zR-o1pYy)}>!FFKU5c~>k8G;?anIYH-pcsK&0HG1s4bU5bJphjp*b7J-fqj665!eq{ z7=Z(ThY|P<2sHu+fg~ev2q-iHhhadi5jX<$7=fd}v=KN4Y#D*$z?l&^0iYOzlK`PH zI0eufgVO+yF*pNA8-ufehA}t?SQvxzfQK=-0E8NYi$IbwxC9g$gWrK#V{jShF$PzF zX=88|*fIv!fHPxo9Y8SwHvmEta1)?60k;4i6YvKh4Ko3^0Syyy2e2>!cL5I*a1RJI z0r!C<6Yu~iGy#7CwI<*p&|?Dr0;WyC-@uj$cm$l8fX4ueDR=@9nu4bQy(xGG@R)+< zfV3%i0ceb6CpIHf=-R;*ZYcc-+t7I$}dmyMK<;_mJrE+6hv z+}+*X-5%#p?#(&(Chz6#tUWSkR@ThU)~vn8924}x82lTAVGNcBks5;)KrF^!MbJ-U zuo6hY7_1C3Fb1oD9E`!LAb(@98YspXtPaXC25W$-jlr6rE@QA3Xwn$04casY>wqqe z!MdOiCSW}fh6z|7L}~&y0I>ijU_;PP6R;6T!31m!GB5#~fE-M~rXYV4uo)=E1Z)n< zF#%hEs!hO_pe_@z6=>1~Yz^8p0o#BsO~AIG52j!{5QZt(9z<#ib^x)Mf*nCWO~Fne z1yisy$iNir0&*|~yMp{p!ET@!Q?NTI#}w=Vsx}3Cg1St>fEQ@e6zmP!GzI&BF5i97 z2Q#oQ2*V8Q2O>2C`-51_zyYA2X5c>{1v79U$iNH?0y&t0gFyaf;J=_4GjK2{#|#_- zsx||Mg1XGWVW3Gfa5!kw3>*QvGy_M1KA3}}Kp5uWXb`D6I0nRG4vq!=GzZ7kwcM~h z^#PH&jBU?}Ru}Dz)mAF6e4aWekE*$ArU92*$GWaQuS272ZI3HXF1%E_r>k7F0w{x} zz=N%zPZhy=AEVCd*PW6DXLYUB*4id-hqvCxQ(k<&Z*iG}out7&&!o+`m{aDD=8s42 zwJWdtG@8p~nyXtoOLMv}HZCi6OMFcyhX4`dn*z!Zw&vKk6%Xgg#EPMVo;-l{C{m0b zp4-hS>hQf^NgBfWQu2^D@`c;WDXe=2S5PT49QzjmKacSHh_yt)k*C;87Ej)$$kk(o2>Vx^md9Ng<|ePi*BcwA zjC=%Y_sfn-;qF<-#~Dub+MCivuUU&H_nH?Em3BO=fTi}tDXZCU2@5r8*Hf8C%_D^~ zytftt__|?~u@1il7mi1~oaBMG4bB7T7mku|8qCQzp`9?U-l}VN6#S2DqxU_bIN_Q9 zgs#aCr|FIRYaTcvbF62J&UMF%mkM@{7XZ%HOZ%D~9{AS_sjW{L z!uWjZeA9LQOaY6|Q)sjS=<38oho zUEDpAjc>YVh@u@iT~t!+%;crXew%OjE|$wgj0|&&6fkF-!e^QTs}~DJUuWi+F_M-k zJ9L;y*JG~I9=s*bd@yDMx#%lxNQ6CrX+{Y+e6843?RH?3da9dy34Not&1_FbgWn*1zk6DX$q-Kik>3E09TSjv?3nzVJ{FKJ>FZa4Z#sC|h$ zdQE#?Yp?e;ZnGRK?u!Q*VpIHUrcH(3t0iZtd?$P<7Y`=Ah2bXjbrukTwc*E_8GZS7 z&w!;GKephfZ%;#*xP{Iyc8zGdt-G0nY(?#F)*KmH1rO;09|6~D{b(fJ_Ll-3xE9Gb z***lwH6+LU!?yf~?UOIWkr>3b=}_)*Q0{ZE{kgFHoJi+ZYR_Ie(xXT9TMK6pfG>4r zHSPdy>@nAaMknFD!mD0_4flP%yw>)6ds}WWdyP@!+m^#eLy+eqIR1*Z)@pHCgO=mQ zH6I>?ty${^a3OB$YoHE>F4%B*alR^j`|L$S*`quNI@7T>{iUv@WtutZqh~*Bj7di> zkoVR;JSjCrM}nI3hs!Jz9YOvlL5=L+04fsHCW?PTap(xH7zS!ypg&xHF*es#rT^2D z!9jwW`-f|pm1NbnEIXwIDo9?1;M7E<@I=#sU=@%FXY0(}@;~&JL3mM@|K@Z|WP^zl zYu{v@BoOOLc(Qz#@z}|juwH2o|*W|zfH-BiOB8eeLzKKAjT4jw>_+ZT(eoB(Jya6e zHsRVj>o9Akc^0GEh)$)}cnzo&k`~Urq+W;a9NQXm588yh8fl~-;*JCzT*V0xEsW%~ z9P8%&%Dh1~S!A+{$m!s=@J_g-8cfZ-H3g_6-~B^8E;0Faf?B>{d(8Z<2^Sj7Z@U3b zziTA_+-Nqn<#!_u9`PJW_o?lZco!nOg*eQr3y6pSdiw;!VR`LE7X9MZZu z_>s)~k;kpVm5WSa)>os*uWuvHZ!dCJq$jhhaXnTUL`R1qcdHfM5^QgBvsZ%QG4w$L zDqBV;@YQ@=?&~2vn=9oJG`#@MiwIZg?HP6OP|1Zs->dx%xL7fcnl@xWnR_X%U|wrh zTyE##u&Ac-uRTU!x5SejS8&%xo-=z#qH`tK>PR&BwtBl?1ab%`P1nQ_!&R;vLx2@8@pHXbdy$sRt_I9UAxNAoo zTIPz(q@~>AAVFFwl1GEOa|8sw0$N(w5}GM!VxevqOzw4UKPLe;mC+nEZp-#_u26@a zQs?qCkw%AEcGb}wEpAJ9B?oM{e+w|s`eM||f$kTHyK?=zuj+hvQ)Ze$Qe8w+3+j0Bp4m5<>GnsXk?jA~w`vOnk}bzBmNIAbqH z+!Qi%hC?g`M=Y2c}RT;nzGr5YP7Y?_1GNR%$w@2#`h zLTbgr8u6bZuaG?BYv@%lIA{;}Wn9n`s19DS(CHJn?ICPA+zTItVEMWxQ$j!LcCq$ycU74S54?_LmD7 zq;xjba*mD@3TknfulFZUq7{gks4w0bb3Y8Ts#Xop-o}E!$l1H2&i&aB1nf)zbf{2M7GOg+)~wbg6XYNbp?b+MjKIsDMzM5)JIyEFhKjZz*#P?Sk>biMXfdHtQVs=4Np z>F-b$3a=&9jor2yiLluY#>dG^%_iG&CDbPK9Hn9_3kGE6Ac7WlB!5iyIFW`@2>2@P z=GDoaxS)_UlL?*=X+f1m0|Lq;f??(GJKrdK_KVZJSY>&8MJuG#W{sw1OU3#Vei_mC zjRW&sE$tRyIZ-7>5A~Bh#@?}@EjeT5jTYTF4?Q7y!R@`0ATQ=s4aS-$7Pa~1ykE|- zmCX`QmlXP*v7&(Wm~Fl$|yy1 zdo|v+diOGNNihK@7r054!}6lb(uMHtZ~<#pd7-dE!+5BT>|h=z1Sri?df^AD%`4^qQjV`Ik$kGA(btS0byn_bqge`^ImCDPw7H)Zk$W-n zut@~)Y1U%{Yu3O$3+GAr z-7s1xqdZo;^24D_v7iwHYKU~;(W)lgi6&R^`$uT1blt8#GsIdbOX}T)oRzf zTL`nUemcimq3n|jz@t;TxEeM^mR?5T6st;!mRkEzpx#VGsfiL-`%QuyvMOt)wDx&5 z;_uInF3(6Xu2RrCU}6{?dt(YrV5 z_8+e-;RGV#LEGDXbvP*haQ>Jppwb0@uc{74$7R*9vQ&_DD~l#@~w7aVOlj_Q_ zv8#^+aN25ijdWw!y#7Y2f7P=6#{D)G?axQIyW8%;%o!;%;~moY_Im1_TW&=ZTAa#Z z(;R--lK7#Vd+;tT2BGbxJ|1ocrmWm;q`EkMfdSP(wl=g7FQAxEcqU2d#HP3xY_YUz zO(a$H^t;HC^1hp9bH68A@k6a=gj0Fh z7Gg1^ULs?OPOyyT&#NQfI@Vr~dT-aP@V8h(*J$lk&O>XAR;uW9vk)f!QL20Bi8JFx z{|o2Y_`i{E3Y*oo;M5?Q;oM@>H4?;@|->@yO`f#Vm4zPCgw>SU;M0T{N+6RFb zV9_?Tq~y}!sQ5T@5(O8VGh8RE(j?W^qAt|WMbf@3XwoRE?j&9DlS))`=ojU77B#-< zYlnEw9dHx4miz+3A+Uo>)eFzqv!sY1KgK08zSUne1Fg!B92 z&Mc%ZbrykwR93#Hy_U;)U+~LSK@4Q*u<KrBOe=Y$d(f&)J2||4+)uSCcLa}IjQDqE_4Fpgb4Q!WaE8T5EJw0t_C{8Ug9ZL69dOKFFX#Cbz?sI!| z>A4b=^Zfk+@Tk>i{m5aubA4Q=^MT5|K_B`}+$$c?E;yoQ@UDgiRiZ>fjvg^!6V1p}sz znQf=)ymNDnRE$?lUuJ^j-rCexiZZDrkwu1YJ7B{LyD?G&r8@ zLU>-2m21`2dA;AAS>hCIy&MI6?%sUfcF_dAu4Uy83L$FXtSGOewYQhGqQtQy9nZJ5 zx5&P+1}t5n)j(XJ)I~%FLK{v5&o=X&5}=?(3FiYw|q>zQ||!$I*)!~ajMLbU&bAD3Bi z@_(o+Vw=XoL}Hu9{TIal7bN@_B*L^~ntD~9?12y1&W4)gWD z%K-+WX23)SqGrOh2cl-d_)PyN9c5+`G4)s zZXcnbVQ-+JpkOhfprDSmUmWl`W8Q$w52kaVa=cZt4JgnH5xLS<%F=Pucs z8T2T`=g|$YJwN@D_2n<}p-J(M%SM%T!kp$Ut+FrGSdk8*K1ya%YE??5(k~b@z#3F) zoJAVKc%EQ#rte0{fEbl}G*x&?Z^{*&us{NZ* zi_UwpWS^JrSbcy88F0JQtW}cBtcIB*=^UHEp}&3cnKLr4{VOQonUX@(?j^7i#ccMd zabw>pPpZv6E%{PBY_K z*jyDxS-!OlHGYx~dS9gNy~zeWg6)xwhkjOQ1$}bBRt6H2dmF8x}i7UAuM zb4m;b@f;*u26FkbBgTi3DUGRd!%;tV{kX}P8O!3D2}*pMwWzYp!UfNcEx5%LH^8lE zR(|m22Y2#^SIevRFip>Sf2AmXjSutq^#w@pU6|eySodtZ5Enaxs@s?|I4o*$mg78> zn`=$0QUQ);hUpkqi83JpxvppPu6^foM`SK8PLDQk>-$mOZet7r1eLaOXm^Yei^u0M zo^Koe-mWfQ_WxclHkMdj?5}lkj_9}}p zahwiiz6cCiGG6RbOLW*aRD5@eH1^T9Mfd?22Qcwd{(6rx%g)Si{fQY*tPZe}q12;= z%iGUD(LT$Qxj08-f;QU`M<+crPDuHT*WYnn{TY`ixNZj3KKx`w%rb3G&E^l8|2FkL z#v*%Ou#d2dOqfwNBy}0{g`=4X1`czRJr8EOLjmEdN3?KcZ}zF21&h6d!n|krjj>$7 z4j{J3A0Q*t82?446keSw{q3g;yBcx2$*4Nh(Jv!{?W^Atb#@9N_xhUtQJBr=WG zevGMQSYH?QJJ;vd3;vKrur;VM{1Ve_N+~}?seowA)~~0%6qW*vMuvD$z(nG*V)YSw zVjFSL;U16=M8^gaBJ!efiC$TvUA1WK)~R zKHU>1X?R2jTUzZr8t4tv3iamqc59!rzobUmIk*jNDbtON-voFZJ3+H4M<1zb4*bxC z=pgEW*5yDnq8r_7Wj1CWst;p4hpFE(pcW~z{Y|tNn(_Ym`a$v0{xy!|_Xb*n-!2&2 z4IW}`5X0LH^m?H7J;Nex)eU|eA0H(0rssx-ZLOoy?kcNob*KuDT?4MraB2 zRj3HtOR6j*265(%bv+^2ZUDsc0j?$-)2&V8E@jNG&Wg!GR<8&L1QkMqkLaTms6w93`PH8^(wz&)vGjM#O%6wK|o%@;=)pzUnfV<2#c* z&Y`NXAIj+E4W1|QjMBsY^E)rBCS$H#{GiK*z(%{*5svjBQFR;fHmnAoeY`68S$3oUHLmh#TavGgg%6S>9!@_TsYkeMFcyB;)rwzNZQCa;Q1qU?qT;^XYvQO~( z;7iMkzf^r)TbgFgL{rz8Podrle(pooqCD3Ah2!d*pw*e9@0K=S^%Dle2EB_d?{lOH03CXrM)aL_um?DVPTP^Xe} zD4~lE1745l|14fImWJNb9`(m%dJ*qMx!|3s%-xrpxH3#WJ5Gl5d}!HWu5Z-NroE6tQkhjdiu*ikBzD+){2)@XdS z+ca=-{xjN=b-20RvN8QAV6^LM9}!5;H|Mr|V;hIR##ayHHl|21r`Ka&l(I5qv<;pQ0ATX$RWjv3}NF>Te)K&8W*jPdwV{9 z;Fk`A6tG*2t*EqeX_cEYA*=o!j)G`9cKQ!R1iVv$CVg&1x=8K*xLnPJ-ppGoo%FUZ zj%hl!*8h1>2mektt{Euz%SRatH33e^JpC_vcn0&W%vIMmqe;-9B$o;j$$fhrWKIrs zwQlnl4}Hn!D-_q5K;9aahyiAi*%S8`hfizbJc;PaUPz7~I}L{UxBjsT@f}GEZo%@) zj{gU~!14=?{|AoHuX$iwQH3ggjV}@JTccldAbQ}#w#xXgQV7|y#%9$^j4zSyTYtW; zL-f#uZMF7a&8n9fUn1E*LBDoE^x%bU)%IUy5VB3os+T&*^b*d7@kM&XpWKKd z3UcDkWERPWA9I#M$ z&zAw1rOJJ7L8%Js@FU{r>mkAu&>__xwrIqrv%NUHKi~1-(Te71kL9RyyCJ$o_yi8+ zEiw#M4Vr1o?Icd;Md-w?I8MftA<7snL#iX6Ju_Zp?=%Wj{#3BQ0Jx&C7g(zxwo@fm zC*(kovW9*P2^nF8e<60V==`zF8Kpf(qv$&C2D^bTglRA~<-=j@zH_Kp{1rQR)WEBL zN-H@uBJyMN{Y?4N0HpZLa*v=^P^+Y&VoCwk(kfVXjoLjvOeUpt_GH)zLLROZvrOV5 z$*QWrajM86ORX7?2Iz>qE*F}njYC+-S}qaeGA8g%`5l>^7YfknV!qr z_JbG@yw&8&>CCj*npO5h(k-jO-PC4E201WamUK0I#VA?AZ%%RqT(~2|w7(hz4*Yas zQH5k2Bm2@G}9`U3THLty1j!#0A)Zv-p&yyM&QjKEDns3IZ zc?aPo3cPg(V>?LyiVLGxSOX>AKLl+$cti<;mqZ8vn2aBD&619*Os$X^64`1hl2lra<>4 zoPTjru%6zVn9aq@?D+mp8mq4aY}jgON3e7pJl zh-)*YPN%gciTy1ktcLcAmZuzn15;snHMU9=LR$&;j0RiKe7-hZ2WO)(? zu3l4Lo5*Cd(v{gJT`yjnUzy0Hot$cR zmWILsYxON{IU#{G%4PzIXEztZ63^O2`FLjtg1HR=tY6`Vcz1HYJF z`5+$X+opv{e|O%m-PeOL51}aGGYb6ee(aGzjMwpYH&@`pOLa%<$`Y6LL)g@8n!QCO z^xCC*L7Do{G;vAg4n#`Rm3MxSjh!HoMMg8S;9!Eyn`r9#huPX~wznth$(D)ySsmKN z66CKfu8Qr`@Ag*m!v${>(M})Fuk)VE12AP*5w_{BgU%|s<;ay>B<8rzr}ruzg<1zd zu0xop-(G}mkt;TQ#>{XEYzA+e@#=$jsV95tqx-QBdH2Av{c%nkgv0zGsX(^-z%I5t zp2BPT-ZS3)HyLW;_unRI$HQOJvpAV^??fp>>7PjP_G0^mV<#qmrjeb8rz|INnJvG4 zSo230F!HKDg1|P%eK$k@bn-ad!JrQh;0e1o3y@VFLGGNNISC@W-cHG0zd^mBDO+xr zS#9aYgxI><;(oQWb;f^-ffDNO2=b&|Zm9ofJ&VN=?%hyd9?qyde{0r%Rx$B<*;wpj z^EWtX$UfP8gO%`SWB>A1IeGQN7airK7Ef1)wMl8)bR?XQt|&~`d*I>ggjBoYRU;14 z`{lqWc`kTf-AC_V31}=6*X0UnC`9yB>_uQHI{)0a+uacwg z@+lvX{Ym;T`?;b%CQLQo$T6&JJGuqSsD*}vx0cMSr4K5v?`Afl@oUx7@ZakyrJng* ziZ@5L1x-kINeJuVtJ+|^6 z?rfVavylZ-(j4q;y$K7EkMZzjE2wRZu8AaTTSR_)g(}exc{1G;`0ji^D+}J#jEmV& z3%dInGRHK2wa-X%KuviiT8 zVkbD-CZnxoPnT~b3~BCnyCQm9)~#J8Jf8Q|3P3z+HwA_?kGqQii*J#iJiJnWE?kej za0Cj9$3b#dxT|p-%H|KG(||#vXT8ty^~ZnD6cUQe)&1-5|DItK6q#=X*24yST1M?Y zue6$ozZ13(qhARQNn2)ka*AgqswWU_s~~(F9z%Zh(qc-+JbO3tDJ_Q`n)zfc$~=mO z93EqS^?YJ<${CJ;ez}qp_(Gtdd2mKIh^CqM)1^kw!AcqGGS;VL6m6fw?3y7>ZnEOR zYaq?Y}tN4E!R$X{KF&DCFS6WN#-f~lxN7Fl<=t9AW_hPGp=_FqdP z+Y2M5--m-QVJi{1^4ZXtVQw%#jy%>wPJUyqE&=FT1JvY3MdI`|=+@9=i{3Kxc zGUgf2!FO|j`dL*LI08t=6?)!3mbHvSN+^Bicp1&qwBpum2oszz?sT_*+CtdIupW`UI?(w8u9o<;EapHw0_TRi%hU0VXvoZ$(RzHsU?I}*djNtLcfffD zX{hxiusBRHtmi*`B}Fm z&KwhYs5Tln#|TCo8XH1aO0NM-!K4dn>GxS}O`>9TQ&#cQe57K++P5^O=SicfxB?ID zIiR)4h5G7uAd|3fBKjSrm6Z$)=*hg4^zRN%H7Jq3kkc*sUhQyRC39ip{DYC_%;Pbb zC1X@?#+E>(ehdBVz0TZtK%A<1tb-Jr4WQV#aqBWuSE2M>d6xvcW(yS1 zmc-0@fyvGJwn^b0q?25UALyz|QC=bu9fguSVEEb7)6qzjY3Gr@PaG{Dx_>5%fH>L! znGXCi+MYxt!}D_+JooA#HWjLX)`@z(CXms$0w2XO>9^)B9kWHMoZTj zPv&fOc%kS*CDGmXZY@>YA~A{Z?`GY=F#GS>LwM2D1U>s?J9ijxGapa4&kJ3hs#LRi z6HNK#q`q&zMenj^IY|Qb-2D+O#_E)7l@#7SorMu+YvGweBGcH~sv-I$*O?FYR@$Ij z*5qUszf^{ty~!DPtW(%3q|~{<%EqS3v43^GZOWxIXX~`yxZ9v&-P9Uy&=CX_;(|ww zv~eTkTj1S~%b#A~Zp)qP!3M@%&kT= zW)WFq2;UIL=l032*&nIAV=l?|%5y(cYkb6{a^X~uBqu`M^9mx+a2!R>6IA)`iXQe7 z^xJ|+!uJF5Li{h;*pQfDWrZQYw?(B8kGDuUb|l2UxF$6n3F*vgQ}8=pdudmr*=fNq z;k*uqW;R|tsS5s??3U~^kLNe+8_re{_Bji8EAAv2B65Mr&g=O#2#&L8jgdK58N9^U zcGQyuZ`k7EGn07|lM$;Hssp@tb&P+N2dr zjCD{fZ0l1+uui`U>D<9?oWsf5+?>a%SiHUCp%GrIzZTB&U=#1M)`%(>mcOIA@eK_#@d$J8~Sv;NqzH^`TZLJ%SHmzkKZx znIfQiKX-6_z!ihqLYyI{af+Zx&sEq1&vMhbsGU?U>$i6Mi(!q)4>&4a5n%C)DUZo7 zI4a!{U zKppVo|1=x0eM6xe@O+1$8;Jf}C-C3w|9@r{+HD&42BoLPi0;znZ(f3hx`eKhHQ4c9 zoRDhU`BfrSyq1i9&y|76t)G{DmKKRNwl}<&-ZsvjuJNnJ8OI}i2~1w}C_qJLOT23y zK2&h+j~sZXd`^4Lj>Gbv!(BD}a1C4fAB+fU{5rHQbLauG9T}Qr6{0iceG4n(^FH+b zHAR3j7R<|Fp5CO%tgE=M?#xKel9Qy&P|gzmBp@@Dvm`x1#w~z|x=K`RdIQcG8Kn;OUoyYT#%gL+AI+L4 z*)1-)m*5xAQkK@E<0{WKa5l`}AEcDOk-H_itnkrTcD$>H@>auyRm`XHqx*@2Ji3Fb ziqQ+)r<`wfKk0cP5&Ie=QvK7)+H!!MrM zsK4Z`1|}L)m8^nc)fR*g`k1Om|Ih;dX{h?B?M%%+$b0^nz{82o%Z&Y_VQQJ`S`Z@d z047royZn_7Hh5q4k;E!#`7Ip3qa4c4XE5v&e{e2fV|2GdyA63v~FgLtVT={TS zNLndt;U!fZI}PcukSvf?V_g^C7k%Y}`$vsu>L2mH$!H3Al7w=$s*qBx$2OhEq$YM^ zgMR(ds9lrvMtyY4{ozjmM!&>taEM?fZ}VWX0sf)-`dB z%48s&m*TTj`7*#fpz5$4p$&@B_KNftBMt}&U-=MN7HGd0EfE*O8!0`hq~&fDQ@Ghr z)|?k-I?Qf)3uwinLoU12V#pQxh<+JYXi?0*S zZIB~+fxH?_7;m_Am(sGiQB2OJBw6e|rWjr$C~Nw|+A&Z?;vAB0uyF3OU=ZWCIf@;N zf~i`UWR%5K-NSJ9kLO~njWLS9hrt2*p9A?U)Q&7=>^ib5Px7`(!{h*YZCAAH7^bWr z3-QSlbdi!3_ZP$|swG|IuW%Y%Q~gxqUD2YV?%4UPRSvl&*;Yh)Riv2HZ6D7)y?Yp( z76mNf?FxZN34=ghZ)pcwtph_0oekM<&Hs?Qz0{tFe%zjazkN=Iz(cv-CgssJK>Ema zJLCB&29QsWt3rh<|036&!5KxV8ASEYGlWW4e_mEM!iZQOB#m`~b;VvVCZ4AdTXKa{ zD4K_i+9i3wThhnyW=eO#Ek77xFs%<##zsNA3M>HMk*#7VY}N?-iwtly4dY!a7wVSH zLL-u4p&G6R|Glimt5U4K?eKCC{#r?EUWGNby09f+w_QW%2OE&_!s%&BCNlCEZeE{_ zl76WGRI!sAUIf2{9`n7bR__;fv`d-1bEU(?Xk9Bssshq~S!GI4Fr1!BvsBr-8pj0N zKe4`F|Iq^P8x9vXHFC$t)dLeHjmBJhN!~^7UJ5Oq2bmF)X0f1d?{-^5)W*!#noo}X zDLaNEBAjX5z|+hxeEt#95SVNF^D|=SFAFM~)V~X)Nb}=z{RxlEdYwwxBFAg+>J38J znLlhl6<+xo2oDtM;f(|cVXjc~p2`DBDDkSU1FXvN|H@CbXY1XvON-BHA1v)0zkquw zAktH#9MO(TRX||R`cKFx=0zFU91sa&rdlLdIT=F z^YD85j#JMbuhGt03=X0@h1(9W@P#j^q69A&a1w+apcXjOyHDLK5?i;Um7{^g>U_bc zpMC9a1M`ViTqZw#!d{OC-Zq!~-U>$(`SGIvu5Q9PKpEF;?U}YGHsYdtI{Q@nysmp;*DKiA}XXXN&1mL^UO5@wMknL(Z&OzJD@uo z7cN~7_iuOV>qNLM^Ver!^lHA3GZDvBP^*RCuu!n=>EWg44tVI`Lvv(pyrcYbKRG&z z%Hx%Clk)PUEA1pa_b#rVh`C|FV+G-iMN&M=2^Q)@s*$6SJQpAyoZLLW+`N81nQ^y= zw&{LwV>Uy}^{hd?rqFP>&0|8lJ>7Zvr?c24l}=hTSIyo~;%W*2R}9_RX$vrCq^`XJ zCbn&W>w)tU;jL(dsy%ib8s5L|@l0`bMMEX6ZJfjMEC=rGjahBt!XYI3Xx5Sg6}&|| znNWkKZK^pIe$lNgM1%xYztuft@prcHzLWH=7^S566JX=G@siT(s z%H=jIuD=IQcH2O)41$8jO26@StrkPnRf9*>Oop*{E80$QVT;P0lP=%<&S7UKQ;KA1 zqg8xcHTpj(+hqO~1!ju5FO5+VQrag4wp2}vhjs3SBhB>Q?BFR3p^Ar{q1MZ>IfrRd z3)zzbd=QL^*2y#)E=l)`9w;WnrNmGrwPr z!_fUV)>mf8Yt}n$OMFn}*hs31CBOeN%))BOcxX#Pf}t9TZ6M+Yln_i@Ly^)B*{$*F zt>&%b<#$hpm2h~Z_spnpSIR1KjQyd5sXZUrJ2U`YGg=Y@;!MTxUbPkNyju>m!=ifb zWXTYE#MLZO?{RHYF8Osxp{wMZ&$bvvNr?Eu=k=#LH^ux90RC6y5a#Cyks@jH@?&Cy zlGp1|D*4g4`&4PCXNAHlx9q+^x&*W^Tr6dte;H)VMU&uZb(%rx4n^j2Cxu&Qn(Cjz z@zfA*IcUQHg}6%%q+|wu^l^)f9-7aSi0Qk^+~_#0fr^TSMUu)SQs1kK|0K?XeBVV> zwT!-)IcwhP0n_$5(rEAzff++Z)esqdf4-p-Skb>#v~pOc^6aYS)SpjU2}nCZZ*_+9$$z(f+&Qq$FbZ4& zq45TEwrx_v#+duy?|UUjxQe92TcjD!wYKtWiNH;D^f%CQQH-ZXpq%0vV(tcV-gBr> z&|Kr%!HM@3IFLB|r&`Ecx_XCW_Nd9A54zWlA1E??MTWvon> zl_F;P0i6AGKNr#IaAPK!jiWh>O!R&pGMr!_UyS1-`zZ+RuL|dF>Ayl|^7kN|31i}j zqY>eYCUWOoL&HyR-GdLr<*M)0X2s7DFnLi;E?jJ}JYG^FF*1cW=mnBdGu0kZe)C;B zKEAr-SUxyH4sM8N@bd8eQbVuOeat@`VAijrnE_*Wr&@N}y@8q8NcN`s`W_@#rh9%P zzo9MWbJtH-LFp$fp+uj8nnx0T{#deg8;Z8;Lryrwb7ceaZSd~zV=aO4)z=COq*AE7 zSh5wzsj<0~pWXGvnbz}m)0grGW?J?AO7-|R8ySgHCUrt}*tmQ8GZv*X3 zs=!js^cmM!tg)i(hm}rrV)3KX^r=_%Xgz2+s4Jw{Lq1jGl?tqMMyPS$bQSkt&-zr~A_C3jgiHRCaXXXY45rz~E`-gTuuX$Me1`ra%4}Chb+{5U9FfeL z_N51_ul#9Qou@o~%PTvh{hdHQeoU@g!w8%ubI8~I1}FKAi6fAvcmE-W^V(CnU)`Dd z*=rG>*sf)%34G930NfvKU)t^&PJU&!#rAox_XTIbdk@v5L`v)u$ru?Z-6Q{drK61) z%(39hfI2DX%s9FZ)BtbqXI&_*UQYFGOue{N3b()AAV1>(+@7?uvbrtkXJhMgoIv5` z$iWd#A7$a#zXw`-dmqx(R-boT91r@~Xiv%2$BWmIg`T4O+vgR0d;7<`kM_4TxG=r# z2%le-kO{x>>v2D=9~C}b9@ULN7qyb69l5fvnJL7_i~jl)R!t{`QZ&#g{oS0BqXpTz zT>eT#LPXG|( zff35?xq5z78R3!?({UtK`~4v<(m}8gKcl*y;uxK(rj~rTR5WqcX*&v>S#el`VQ7Tv z@tqjeR}b~kYPLMWWcb_fUjZvs9su@_qly;VzlA^WiBK|w1-|$nzlkxStfODEJ~A|H z5!EoGzsTq@clX2Dn0+Ehwjlzdb#Tc4AI9D(I1?`H(tcvwwrzH7+qSKao;V%njcwa@ z(y?vZHm0lo`fI+UnK|55yYAYDtMawe~ns&?B$7$LBafo31+Wydm)=f3D zq!`jVStUiwuUk`B*BP7rG(ll3PrnQceWM=l))_->K_U5&s62*mXkxts>%bX z1BP%Ip<}YYC=~CFEj3cwc{X>T61al13=kVkI)2`Ip0jsNs~n)LAg>*YfTh9mk4+YT ztBujisr&%A93Za52C%_uv1PXRe!~~NcJ;-$m82GFgcr!Fp1>@#x#q8`1!<3F zh^%HgbauRnQ*RwW|41Gi2u`@kM<&(lPhq$r%!gT)P+=`NG_BFO=pA)01|^tT)K&P; z-P_s0$hEqla$*Cp@W91`I_I$U^|Qs%0(EgGpF6J2RZqFeKiRH2d9z9(lB==6>{&Bixv?&}-Rs@@* zJNrwC?iL-QfA{Pg8yf%u6BFSKg0!a@Y->>`d}!qUA6eF;G_mX}NJ5(<+R=kND(U0K`XCY0%F^TZc0PC5U#z@z zy2oxFT;+k-0W7mqQp~=W+W;Oot~l-*NmAb&*_4$P_7DCUm3v2dj4A2@h+Jt|3fX@= zpDUMng4<-q>tQ@U_fJs`!ePKM8icSfwxNi&+7AtiKp$1TUea1 zGO;%S2&O_mfn&9ip`_!g<>ky}s-x&q2*z!x@8aeX*l^&`otI! z!eFgMdaI$0fl)EbU_)|ucz+>{ui$3Sw2zLM;iOe#KCVGlFRB9q-!+I$mfo9+x;9Rx z)xkB4u$StUW!WYw$RQzuK(kR4K^lqsRJkIWL6B2uulo)mVkgY5$(pfSL4uCstt$66 z2jjDHOwjO_lQE5$r+aG2FF9&V`Ov~bW;}p zk_L+;3~H_(xUz|5#5n48xX5C>HA4JMDA3aW?hy~%Hz6-H65X549R8Yo%-*Qr%_#MI zo;o)6f$z|jWN}9B{f~eO{X{tb716dMZ4~%Crl!khk35>aIoGWYu!Z_6wgm3BJl9y> z^*#JPptGsqbsmAv;5osSbUt3po#U4a7)md2s|?jID>I zWkVoCCtF5{b){c|iZo6kIe_?PftXsAMd#`RUR^#;E1Un}ks`hdj%bcTCHGlpS%&%J z^K&75N7e*FwoUsN%%DP-^`B5^V%we+fZj4rzjEyvq%e(ct<<9&>c!drMyg)M=Uv_0 z#(qZUrVlFNUmwM$yIBsZ^;3#x#u{1oes4%a#nOQ>)kd&K31>Kg4Jo0dLd;70yPWq9 z{7M7iwmWuml^`FMo=d@@b82+ccvP}jqnGA}HV1c@h<(a(MRb(6BK-^nO@={q7z9v?$IdA3xb(5+*deb zq?<$L4@Hb)?8HT#2cQCbnVy0>9L3uP5@Hz>OuUD#K)Ot8BPSIManzhI%!h^JlF%mA z2^_4IWc1Ztq-y6Dpa8Np$Lh+zq%shp71Pg=l$nEfNlp+t%1yY-`-G@Q1vOrgb)7Vg zU<64giqL+(c;s#|@vg#3aEbf0y5wNV!GhpC@Z@F&SSB5Vtjy_0eE;ksgY=#ND+VzV z?!^w5rO_QpAW*m*t@bTq-6Lf-W5%EB?X_-hx;wP-2AB&Jes8g1;~hkKS}f@#c59Ov zSQK)FAp4dm|5Mzw%9Yy=u8PimoBOKfg5mmiysZUFPE&S5V}k^Muz!b;BD_6% zd0y27ZT-N_`Kg!Kk>zaHrF!!vR-lo%+VaH1V~Lrgu72f6_00pxQ;jtT7UE~BRCRdt zAz)TbHh>>Mfz`Xd!IGH@e14WrNfjCcls$~y^`fek#qxn;uN(Hd>!wYx%Ce2^&_%Fi znRKysv)(Vy204O!ek9X7y}W*4QBhI!H)Qgy1?OWzSz2yY1u04>BOs`IAX7SE-Z${g z5l5|m^=W=u6i%?22VReX1$)`c4)#b=XxcN(Ip(RU2#kRr@RY?SnrrOtrlqy9ZrExU zb$3Zd(A4#aTt3;q>Mm;ZtBQHR-Ael|@l8Esa2HRu^m@l#IAd~8WH2l1p82EN>i(;b zwsLMjJv}!P-F@;DOEI#}>GA#Af8BR;tUVdV_FJ&vitfN(SEkVbzUaTU;m%Q`07eQf zmpA#e9U{*iqLx(>T9%s=Trk;D#gfl&gmDoDw?)?}FUY+Ho~mF->BWZ{E(u4ntx6(R1rN@YQEkFILQAz}UJs~P7%toapfccB zHYoZ0d3_B}=zXgA2u4^ET9j_3M207qTITG?ok-i3Ae8RE_-PTbTjK_GTnROh6wFA; zwsqf!D(LbM7V-C{{dJ;Tbh}=cM(qj{W}FHg5L& zCckDH3#u55qxTq(?HTM}?Mk&xnh}O7!y6}!VD^oe+t$yUIodQP_;!NtY1CejL)KGT zJ6~v-xu9k1^Mh)tQyCW8lsSiISaOFlCX?=jfy+oDq=7}+RmXW*69XEZb;(mp_%TY^ zNS5*P;x3^;R%GIK)Qi3GhS^phy-hu+fw;dH)5Cg>X8jE`iFk|!J$4ru<~E=i%7u_M z=3z2WT<U=AyNjHgc+z5gmfafpwr9FLG-!uY6FvSj&bX_g~&U~&~? zMPaw|zAqKxsCL(JXjJ}ZsjH;2lDq<;fqf(HIIL0p@i|MiAEMo8)brb&E5x~{a@;}JZj|fN?<1E!hFdK{H?t}joHlJLzf?LAIV!aggr3+} z8|K4}dnY&S+GdB0b=}_bqG{LZYB4Wl+A<;m@dAS@!{`i*;vbMkk_!NQAY!SRS>%ZY zWDL0}CRJaSuIy=jNarzQR?Qa#i!_4Q^9`RZF|T~1R`G-k)tdBUZdS5Kv4p=ue!MFk z^}n#X41RkSKDudS6i$Q(OA={pvuuZJZC1q+5{0yw0uK5n%?Pm%Czha|O1(fU#|{F4 zik4S>u!Z#H$23CB?duIrh`?Dv1N%7%TX=O)VpXdNFhha&x({P6{KtD6INwehm+kEiCtGb7Zj9!E1 zAzM&j~8WUVrqwgDK4EIWNMCF0{qnf~{$jR_2D z{R}SB^57$V#vFGZ*eHO1IBSrO-Rq-XOlS5#1+de=rUQ5LjjazW30q4`;q&m_dq`buD*1U;98AQr7MJ{(? zOBx?8WPY=b(L&PDotZd0A1Vlt%A^Os0C?eDBM!jr^j3=+2kKDo-W}r`ypS}v5bo&c zNChI2%1J_&#`yvH1z9ST`CWp>?zT&}d}?A2{S50_G(F-NrdzJKBn9>C-}D91pG<&V zC*2g=9!*hvBxtIs5ADG_IH}RE2f-JNA$(#xjM_YMsxH-e2)CF2;eZ1z7PTm~EH^#e z_#hSWe2XIEls-Yr7r4>A=Cc9{xx6%j8kr8nrp}xMq*UeB+t9*)RctS!V7dhChhWk* zqcATu_bYWYK$J8`=2v+~hSPc2xfhSJIpIt)#)`UZFmCMTqckhgSoChPmh1J7UiN?T z#zUf2T8F7iRJ$Yx)29KY9mI?e)m&V#2i3HOm5BmMg|e$dgmEU!>A-JGk0=w3&488^ z%cP|imY!^HKaGWg&?_iuReBkz`2G))$siRmQ1hMq{{7MJIu_k)JP-Fyt-}nuX+IU$b6pk??X# zvSJ+ag9?J)uWL-KTFa+ce=WIZ`v+(EK@^pwNYHn=yT|*YjVwft>>@5eRybFDUjbDH z16aI~(*^YnEz;Xf4i`+MgZo8GUNsZ5cy3Mj+=?D;3PHe@aL&2NJn>aAfvpWAr10;` z|33H589y-z`wYn^%YIkZ{hfwF6682oY#B(DZ)LLUxGkKJauR4qYh;wf{ihQM3g0q2 z>B^SdV5~jmN4i=!?QDL>v?7n!^#80{`1-U5AZLU16`0I6e4l58vQrBA$LLP!EF%lij!Jc<=G;@E^H zy9|wB0j`|oF%)1-R!py%Mb5SRM^Om-#XV?cOZnMnMcc(84wuv`2ppE+s|y589$WMx zaGb!d_;LorD2%OGmDPBF5>qy7s_fl^;!h|R)#y4s{g?B)Ma1Shix^ND&b{OVlHt$| zOt`c7Zu@Vt^i>1%&;uSR<6%+P1vwKav9`a_os~%GtUL$R1HRY1(i;Q@EW4w7A3AzY zvHYg84u&n}P4!7=1fev**u3(LgE61Pv<%qOTYRFxSSUUZ!~GkL2XeN!nsD)uV)qFL z&*Adb#MRqkfkB0!nboUWmeAZ4Q_nRI0PcPz5qfJO(E@kzupkh5ZEErqfrnMWo^Mt` zP+MnYO`y#$=J&p>aE|!V%OiM3)XR4kaS{h!Meu^VAaWH*>Jirfv)#g4E9zsvf&xU9 z&vP(=>i$jIK^BU-sr-Di*IpD?p4J4j=l9VX*BPPKRxxo4+p=r(pCMVJRx?j7?h4xEP=bzqYKDBc=)Eo70|90W;nyu93ajze5&l-G18B$N+ zh>{Z;ApUJe@Cyukf$FUrlk?+Qu+QfJ=7vIh7scg6w zl)8i5yC&N(YJz_&PHB&K3t79Z1m{6AcuzwVSP7fC(=0m>{EiuaTlMf02xfT-wRxf3 z$mUVIb^m5oTOhkHH0{#DVb3mH%H!hsi0ZnrF%Ru$#P$3ekrl^ABR6Qdau^7vgl7NS z-cxizfO*V|2nt>YA>8A7&nceC)5pwsS0AN_;o@TKRijvggbz1ncz*I6YU;rEYuu$J zw`uN)@HTdsj?koJBW!dWDaEbY2uWxOWg~QDjF;IJ8Ny{eu=^Cq*PgKjX^icALPHuW z6)2)5UTT1>rcb`Ce#LH21r7|fkt(b_MZ7RZYFb%MoT7vg)a5AeBsPZHs3%C2ntqAi zBEbbI2(_hQs!dcUd3bZO@eEn)9N%JvzW3Dt|CR$!?=tMkU|BFAbs&|Ft35krI#^g+y8mCAjs5{PTSKt`03Ra&K>2@qXL$dIS-F@uvtVTZ!x`(X1y2Xd#rpqX z4VR)CtBe4^MKu7R{r^0%P%j;f@&5-m;oWWU14V8_w(KWIXXNcRfE5 zjHnhfPv2U{a{MMAnz;q~vsPe+29@f%YHybu3+$ZFb+%WS(bB=x`hSNKX{)A2xUlu@ zC<2F+?djN71Zk9MbR$&Ocq_#!>{QMR<|$iO90cl@KD~dbwri#7B)K#8B&RDFp7Iyd z*u_~!mLfN(i7eN}SGkck+L4sdY8F?i(O$cApp}*~iP^ggU@TQNV`NPWjveVeg$#x6 z{#Wh9+Pf-tlCeT7<65ekrgdU`s!FGlxC*QY!GPR|Nu4{29D%GP+n6u2io-yDF#s7_gisp2%4wE;*0=^6N~NQ@YIQ4VSosPrVQ9-5rlt-)ZQBd2g5ytV|=@$?Mpr)b1{&0csg#RigX9~cDlTtbp>S~C63ItwHdccVh zOId{OW4WSVs;8(z%-SR&dVz~%*(6g1g&awF9M zE#D+3%;5P9@}nUCB0L%TdI6SnLG<;qyyQ`i6V{VR2QJ8HY|a2|*#RkO>5FC-9f12o zMLNoC5JZyy$RU&^i;R&+V2|U=ddQXjK|-`hi2GL7_NgGc-jL=)4S<_tDuv%c+Eg;6 zU|OW$RfeSS{2ilm>O9KihwhK6zH&k9ySgn@yWrDLuTE?{(Z5RPkJtSYv1 zH)_@ZZmaoi3BKBFJP}e3UK~#K?0Xy|8 z@5e;_VE7c>J}27Wy}lXj1R)`-eP_!qA$Fnm^gG@PzNpsbn5k*?w@$l)@%qc(x4%ci4Hzfur1Lr2Y%1JmJgIoTx0mYy zUkbyxh`t6EB3n>7B(fEKu9GrMfC&-bQk_o+p$bUO*t1RjTAy$Oy@1KQC{R*oyKa05 zdOiZZB&RO_R7j;9@q5M57#TdpDymg+Tacgg-D;M6eP0Zakqo;ttj_OqoqFdL(c{`g zQl5wg0s-3f{$3}~h7em)0}Ygb1DfUCo7J~sY)^L5B@SNL_M$!@Px;^3R5OkbN4ZH`xwkK!26YJ2KqUoi&H;Yn1pU z7#lduJ|u(SzhrDgwl1GKgs9qA&CKrik34*Vf%zNnAzpF3Bq9H@f?sDABZ&iO z0_|dcHpWknv9RZkP2|*<%l_z?&zGfElcy#BGMbNd8m{*nG) z?;mfiyy&NAhL+y`^Ln~}y|v*+9y!^+1b)hN^>DZ2M&8UA>i?eygl046i3^i-0yYK0%-`lYihu>sh zw`561xk&9I)5(!opB#cKUy{fko!YvNk}I(kN}LqwtI#ek%omF@=cO%~Uk4%H1DS8E zO(KI?z;p1-fNL0?dm*ju?K5hyvHs{#LiA?mXOoNSaFHWUjHa;md2?uYd9ten58&_z zY0!9e%H@083%f1km&o_)a0mmYG!!sVNy*aHTCR3my_djG&LP zQ;)Yzu)iqTR8j2~cF(V#t14hSdt1XL+>ovK;=LVpF4V5-1*&(sZB6vY=6`3KZv<%b@i9#9F2@qq#LEA_>JbEZjO+R z#3fXas@lJt79F~6*KSVkGWPQT~%QzGgJpB+l5ru$e~Fmk>*K zrVMR-@(wPJmXqlU1rnCL#&1zWouLMGEyMae>_O$cEsI;2jr&vNlBZByC$;($<&qv; z>>AhX9Y5?$Oa{qQfCkldv!RT^(h$zvFV^z(r)uVO_@SBLWCI zv34|-A-N1OUzV`DYDsRw2f%Wrr79t%M+vc8vg-WV0!wXI1AbX~v*R;65|x^F2J-By z1QL^vckV>YY;ko3rjV6fjyzH4{#gLxrt9n7TNX`0U-`gJChdqjz6N49N}j zod&v@z|oUX`=95nk~$IlVe{Q;ySrDke~?z{yBxP3mn=6V+3)V*P8?I>gEOGg@)8?T z6o{^)p%#qqXl&{3VqCehEfCF8tsCn}JRWZ6x{HK2TfI zSaUMts=c?z3_!9H4L?+VGYyWT(wP6y-^ot*N#*L3vnKYb&c=bPVS2y5u#Cpav+-59 z8lcL4+Q^uM+e0irFA0CvSdqPr-VfzNvR8ELXt`}@Y`hRZU5nD z$n1%K2TKv?z{I`#aDvkNb5w36LAd8HPC0@c1NxmZjKWUl|1sg-v^7oLob*!YMv9oy zM-Z<*yybX959CjyNp}#eAz8;)h+NJ$^dwY28vVk)$CzJw;x4}781mY?=WH2e_sM!) zV)1>#$H{>RY=Z5&0J6LwXFdjC zqRKkOljwj}phHwwmo5LNwWtYYU)jJ;ktWVI72VuLrA@_mIO&>|$%&CRZD_k#~2kW2l9#UZvC+c+w%_vu`p>N zN|5Y3Mxy~ z2EVK1D21W-9Pllv1tgQ-t5N>+24-MudZYN}T8)QOgMfeJU*tZWsE5bCULf%L9J^!q zmnzz1Fryv?C=<#Ou(!P*h3w_~I6sf)vwlTEL@>f~6(y1P z1gy?zHPha1XLHKYLn(6!S=tlie5SJvH;#^rMN%?3jp`@~&D@qY+SAjKAHTD6zt!+Z z_&^v4e(Oj37Xvm!Nq}BC3X6lNG_VZxaz$5+W)laIZ0Nt*s)@IdS)ev%3fCKA3oiRg6(-^b#Y$b0pG7RO4 z=qZ~S>4y<#ZzQMMhUTghc>xADzG)MQXPwrN^=9pxub_}ei(~|Oa|NQts{a^-9H5zN zZujpM)GjA5byT@IC1%58Lnm3P3t4m46AA7EkH^Cl1YP8IH=iK$kAV&!Sh=PDu8XQk zf_8hn3=Q>~K1@D#FMFJ6uU_4(?jGQ-4JySYGHkD0*9#t25$fck`~o4N@{6%Ci?#-_ z{f#TSG_oDyoU%ccQ4rE;t!C<3&5vITeT4FE z^G_Jg0?iyb`13n|An`DHwRyu*y!7j?8y})Zv;1oY*{d-ZB%n9F3*sKIrv?< zo*n)kszU%I;#xsFn-zwV;Pcw+W1IeY?~o^_0&IwhW46K8X9yesDaOqgt>ki3_Z%y7+t=HNd;jSMFwbQM|gj z9$p(PgIz!$e}R*BLhCOcX{mzOI9Nf8oy#aNJBZD{wuF-_P1HG&s7Ke++*l2q{ajUd z+vm+XwMuLKvxgpTitAMxUi^dV<1AEm1L7)wvlV>X;V;DJDEt|7;};#3p`RXgby{@U z4#-58HU}Q0jO9dWJJVRVm(S6mypGcTWPv)TRt*Upxe@2MqUl;0JR)pjywIa=`>m8Z z`h$ZiK~MfJCagwV6G8?X$uZY8PC=ZPMQ9Y`u?uOoEnl}@dXyRXxkiQs@^$4`Uyq7VR4kR?bC`N zXqesxwGuaygXLf8H)ftg{Q1=IP+h}%`nrywFJpTbzGeHhGFfI1p2a!Su8yg{S)Y#x zx|5T6OOrXRPh^+}%_hKXkwQa0(s$P9nG6=qM%Z&=E`gA0DrZo6tP>=*EbLE88baBl zvH{eUNg`jphZ>!iuZBZ`5AA#C=o*+Zw2FW$PsLFtXPq?xFR4InvHQ2MKZl;r6ds*hz7tXB zCipkx>^-|)e-Wxih4wvN)r@=@;B@6X>OqcZ4=?LSH%XVa4NXJW;=L{}Br_D}RN(2{|w_bq4OE8Q_kw=?uyM2|8l^G0HYZ4zEA?rMmt+vo5d5+z9w zgISfu+>(r0&mAQG6qewwI^CWTL|W?eakqz~Z@`GABbCui(h9+!gMfHQ2)jE~j?&}R zTaaMbIqEH_U33j<5}^@CcrQtVqcU>rNEU9|BIZ$1;l1bz(1dguhiV`iEfVs6onj6V zG3{!(BxB;cuPdREU+DuFCp=wTI&4aKI3%CslY6uvGPcYwgzp^Bj`l#LOu00y=(w)CRVs2|ZjcDR8rfFhozcPpg?-q0af}2lS z0KL_y?52*n1o4qB*x&>K*ir4nHH^fMj6bolVa|Ysg8Wyo89IvKp`zD-Jq$9z6Qb<) zfNewlZ|RrLeGfm+z0tJnW%Dob-O$W_{Par|wSpOYeoCDG{(r1=86cBdLFoMn-dD28<#Ls-xv*uQPAUku9BRvx+?jdBs)(!r>K>TX?EWhc}* zUAAyPwOl)@jvf47>rSgEj85{9hfYW!sfeds&2?9Kbdk8 zmX|JvwOr8M`S^64nptl^GH7@jl!$6wQNbFJJ0E|=hdG-dtDWBi&tOWB*9+9C=QmT<~O{x=1uiTNhze_#b zyuZ4-_<#Geb|Eu`W6jJk8Q|v-AVBcr8yNIp=ji`E8}woX7WlgWyW+M=oZqCoOwjuE zK+3esoCh z$47tPTB$}vonjb4hDieWU3OxobK@ApQfn1z8Tq2zJw1N}J`MN2Hoe+lbzkUcw13~Rb%qqr>Yk2HeFVG822AQCyg5oys3BXINk+oHzy%~ zzF$iUHy!I&%Nv>m=j?6`Iv1F3C0nPT=*{_|m9BcxNUMi$>a}a{?Vj3#wwrp-jJPLQ zzkH*^pGPtkR{WdF=De%jVG51vFGDoTUG-~?MSq>*^Q-@|U-n7I^3mi zmHFn|vUw>1iXOCu1wPs^XrbG5trZgKTy68ENfAnS8eBEi8143led$Z+Xl_d6_BT0q zWN`Jy8pzUPeV5d{hhdZyl8>j>=<-#s8twv9JF2rO zERdsUxNG~S^Oj$U@4xdgN~q%7bVdHV^O?9S^o{zIq|$s2Lom$%e)s;oJBHoF^lKqz zRGeZCs$2c|V!bwkeoxK18D|42fBjRma$0 z;=UrKD>NIedW@N__pO{w@^N`%QgaVp4YuorwU)ORI7T%HzL$G{#2zs4>+v?N8`VE( zDkL^aFxVyum@;_x*wy_Uo~}Z3r+=NESaBply8^khs4l~sMO7M}zM1A|p5xFm#{8Us zXMcXzrcVMllq&P4aP(W4CTd>{fQH)}8#{B$t{BVX66Th4V!(gV9}H_I^iLA|>;W7Y zBvc(mdo$@e1G$rnbiuRnW6g=-Mo(OOA2FCHa29Tdm2bkkLJQgR-(OA)J-T?`quX_# z^GJv&McC6Q;i@XZx58kk9N%q0 zjL=`Lqp})RWv39ov#SEql_t$Nqa@MT6HMSiVaZU#@ok8x?$U}oIKtA0;zMzP`Vzx+ zuEfklRKYMfG0fqr8bXIeh3(=*=>fKVYP)!B&=A9ngR_6SqI1m6nwX?w(q_=K-p|(2 z_ zMeN=GYwZ$th+rxRUrFvh6;xLQMxexm4wN_rT07N2={VWnh9)$j|Lk|`m^OZ<6=x5f z81`2Em;&BwShJwgm24P;Xq6Hp9l&pLc@{*%9M63g)C)jRVS?e?4ZJKmTI!<*Q) ziic%MaIeB_Hu z0^smQPPNFjv*Fo;Kp2 zl>ObxY`}&Sokr&bj3(RHc2W`*>?SjmzVx`Ttl(XYLmYp$Kwm4+hHOL?(2uIX-7;}7 zaOc+R7x$G1VYpYlR`=&F>JXPZYK+@6YAPGK%}vS%y`7vhke+G~?aK5PV=>*7&KTG# zpeJ0T_ZKD4`LfNp92 zKe76w(~k>G|4TdSAW{${J{S+$?ON8Z}hI7%TGI!Z`{7~UUyh3IG1Pe z?$ms{bJYDnnihI&&O`EFla9-$%oIc<3uGYP^-{x{edNFVc6~`x9 ziit-l9hrOc6^I-!8X5Df%c_ND63Ib{&!(Q;9WzBR0M6_onJm9W|{Ga1t03x zQLH;rjOF6PHc<8QADEkw?H^X!%-PMlBie}kHf=Yh#UPD{rbx&RNVt^kkJk?^irTQ>wB zddfn0Rn6nGDHlrkrSMXqf0Oq?#R`B%ybs>J0`^=K(mf%@F+*s(dr8#hFJvklRBWG# z!AIt4tQ(P=GJp%~4k`c!Bb9z^pgs$hc;O5Dx9qwl^CZj^Si=-=t(z<3p*nqOe{Wf> z{J1Sxtr&B21jqjo(3&g&qm|M7`>LdlN7WfPwlqhkONc90VwcXb*~CnSo+4DD*4qn3 z2#kf6>pIq&2=tD+<@_SRn!fdfNGGVv+)4xr9nZ)SCkHUiSwEeZ4llK{*cI3Zqlw6g z!Q*ANtQvZ9YV%+J+4@F5_)MU}d~eBoCK_$Ep--d25di6=3groqSi6@F&TQ8g?6qzv zTpVw94E^);Fu3jPhQqd#(w(~wN|5z&2y{<;HJ^SFb3Q{3$ZXB!K3b632-EA`MN-gqITufV%Iygd|^bi9Bx8*w^G+2q|l-G ztSs%!(eSZjZVi9VIRr2b@7*HBwxE&Abk3}iI)N|>$P{2zcJB~=y#^#s1QO1BcGP)u zK4ze14IjctxiK2mTCk1@f4i{W`_<~#Tb@j_Y9h0-Y==nWP8zMI^c)F_DloC0_o{YI zG;@ee_D{lq1QM@kWf-_P1LPjtROU1{XWF&MbXH0c#BanUTHYg6!#n=rC@?`FX~%lK zoz>=gibYwx!XB^ClB#E<=~d6)EWVuMv{yS3kurj)B-{=}yz~Z>M{{dcUwT7D*3yu9 zrLnSvL!t*YcsXwJ6Y&$PHsEZt*_sYABItuQKbk;d+z5m7FcD(Ib6sovDH1|x<=P&4 zBMZk!x-z-!=|SRbiiL9jC9|#6NK$EY!PFX;zHU z_;b;o+a(4qO$Du^sIe@Gb=b@lh0tNvW6n~^;L)}xiKbcCVl0xV;ToI$0dSJ!;(m`Y z{Sg_=n+BcTdTz50MjKG7!_%icAmDvagb_#6YYj1MetUlA_&|@M5Ea58La;9a8CtUJqT-Mg=_3x{(5!Kd6ab!8TYJdnRI|pz z{?hTY3K$nHfGm$o)|i&iJ^Vg_EMuIi6(JGnBl?~^3vkeg2ddw6my|`DR|r8A%x-Z- zX(*}7O}UGk`a{quxEMQ?6yd)7&!}$Mb+A|(Yb$LVAV>(st9WZ*EsA>i+$FCD5HbQ? zQ+4rAa4(Ada^@FxL#?_#JB^BvDmzpsWY6T^RY0yE9^46r9@R;!&NzOy!C13e{mo}w z<9{G~3sjkQ(Ny08861+~b1dR^MCG3R%i1hiNy#FkUA}AKZ9TI%_k+iEHFsQwBMRy= zo&x}x!4=#$pNEnlULIIlKFy(MEo0XmlD|_v336G6r1FfhO$l7iF(rhlkuw3fimdlV z@j9?-gVcx*x)xQ38gG{}8r!xcCQs-%b6LOi0td9~BM(g`d>d&;aY=^7x`H9pAxx6x z45tY1H)oee{4RIduy4jDIiMu_vPxr~+ZClexTg2`RZ~LruTNPeYX=D4#NN{g*pEu) zphWjlm^MuW<{iH--C(`9-#kB#rjeunn8tf_0Ak=(MuXDy%2wWQf5X)2g0u~>v*#3vzw7H7opPgUSdaHFlX=!rqD)YpSy2w{P}^6N^8FkweV^D*h(PNm=g|rU<|yF#PhtapZ1nO49d+0cs1<4 z0s!y8aegrdks98077`^EgsT5V#qBO)Z_Q|IdHDV}dDgOq~oSA4|76gX{}!fyCX}CHXQuJ_msNUP%4kN7n>IFB>u(;|8|6_)o%gQ0AMs@oqhGkeV; zTr!|mw2F-UX7s6oEN?iEu^hBrrPja6lS5%3C3Ce`1|S59Mr-u2m4+aDBa^64M50ao zRC8LY^~;)ps7Q)0T>ZU9&xU6~hac_-Fel}KfsYeWj1&}?eP*i=_U4kF zcGNOXvV58g98~T=SH`3P$z(=&WUf_@P%tOjJc$a+bc`95&R9-qNi-r5cH&a=O^bh| zXVYlUA~5%?r8OxXZdlhi&Fe!o%Wx*P*b_%aUK;+3ht-ZxC3mNIC^tq@KT>N;JadV0 zP@V&4be{_n5jO$Tb6QO@X)aW-O6g88N;4Ht!7@o=DKi;KJsd{{TDgTxKc6d7Acd2K zq=@>aO$v8q0p)7(8I}JRW#<%JY1DP=y<^+9ZQHilv2E<2W83baV<#QkPA47PcGBVL z@2@&l|G7B2ShZ^1%=c!ls#WipV?3f1w@X!LgZ?4hYpB&^GWQ)K+*Cs9V9GhFVxl&ZOhKhHoYPbO{PV1Y%SmUK`)Zfc{& zsy9v6xU@XUkN-Et+K{yZtFS14`{`oF7ebs`0hUt461K{ow!^w25LJMFPB_KOIUT_i z3d>fmH*Sy5i5Guf!IRbiLZY#)^|B2|4=3nO?Z{mEE6)Vv#8q-;lTM)^dS#=cURYqA zX9XXU2m{(@Ow0OrdPFyvR)X)8j58Wd5#F=mD^O@8o-pKGdBR8db}rRZlLQXD`b2$~ ziLZ7((e^d?wfU+C4`TZ!z3@$O+4}Fz&CO{4&h&i3M}PnG$Na7d=z8>nTYB(i`V;ws zJG5k1+W_IS)F3{x{QA}mX^de43v8vf>2`Bnu)%l{5n zoL-(Hx}dz=8G7np4~k*yrDwdTutl*sa?3k!63)nRh6dv_a9pX}8dZIK@6Z*hxRBJ$ zWaf^>gVPJ4SxPa?CvO|umOSaXPX`a&dc~lAzQv$a35;qL(G9WHe@!Un8wwxID)c)=S`z zExRZa$`$8V&yO-M2HL1(SeCjHS;v*+fq_z~0DFi3^pY!Uy_5V(kkl6pB_rU=EGoK&0oDV*a0|2C&{r?m# zLMs6$`QKsz4bENHrP4ozmFMaI#8NWL-j>B+(npbx<4(V5wI`)W1dkuQYU7N*hFbkm!+p;q176Y1dIyN_ z%@e1}!15hIir)T&73X5k?x79CMy~K3Xb;bsE9zQ;H&Ad)}GkOy=G;~-*eH*Bb zlm^~wa-^1Ugg-{)_%xxp%6kQvymkUaed-vb-GLi5=v&I~&MDgNRGZ3Pq)*9cJj zhv~7R>rN0;q$>)gQ|RQlzeYHxx8cxI?Jw#K-W)yNdSh^-lbOq+JQe-m6^nMIlPG8y zPJeS$;a&?Yz(#&+A;HC^&!gtnnf9C^&yv9b9bu~q7M~ln;Nk7)Cy`u!TcME2ia&6t zOYZu!*F1O3ABSI~b3|rp9iO5T^qqa?U4z>8+jkz^`|rSPf_JI1nDB zvj`Y3K{U>HE2IFbdicxm%1w6?PBM{1?&|Ec?7duYzn#0I7IO9bJ>cw{25Q7>5YIsk z=yhowP(SOi;gBhE zLMugA9;y%DCq-_d>DfyX*1B~;D=!@di3jZskD%jrV{wxST&3sV@!vNa`*-MfGc4N& z^0s}(^yY)isjcdan}}z2>zrm*BATA-jPrBz8oiuWhda@qHs2Qi#qTkCzQ%V4c=ZRq zI0w3n1o}U3LVLPA{X-VMJD4KifY=`!Kqtt^ezVAbw27H{Qu=SzLvNLP{Tn7hz14d- z*0H?9Dob<`BCyF5?sP!k-zXx+VcEg8Cg?;isoG#MiNk8f+|~KRiLuSl7x^Ioi-~mYKcYhOFN)aRKIR8f>1MiXfgG z>q(z2XwV7PwO248jBwV5zNLl>L_opPDN)f10oZ`!#X+ONn-P&CzuAo$9Dk#a*90LJ zQ!S7YsHr!h$s=K&BTnNcT>JHJ8G+b+3`modSGqn_|3M598c&c|Ca~&366zVOy{J?B zdaok8uqv6OY%ZRC47gPlG}!`dX9?EP$Rf4#cZ6k0PA8zMy>l#n;)NZ8WlUVZ(YUaL z030YD2q~FZIm`oE{cPy+b1i^$@MzyNKq`KmWE$%kUMZ)6wkb-SC{7I}{cPXl4>_4} zsma|if(ATGD&)_;VZ|Sybpe&%+xX@{|+ z#~i(L>0e+D02&<_25i9Bf>?v;U5;>9xhdE8 zZ$AfK80KS@*eZGohYuZtvXvAiyK)2_!nxGV_>^rZd}f!s)2~AW$7~8-yRs z2Xl%xlcLJq%-NVR0B?dZVhlb9r_@Yx-J{pl%CHgA(+ors=}@?KNIpR{Qx) zi#;T*NSDSv!rWm0Y!}RWng@ZXWB5lALGU{?w<<$2t}mUZVH3Tr!tys-%xyb6Gh)DL zfh*&LJ{GV>U~1({q~Kpmq>r5>X`==iPEBxorn^l%H z;j;HyP~$I=mx|Xne+|_WcwvrF&&9ak7bApk$M)ARp_;xxEgk&EE2Q6u#x!!SGW3mm z=G4V-n{?X5_!)bidxqG@0!-4W!eBHALvjE}olMTZZ|4aO_jc5+@nKb_!>%CnV3biU z^_V>cdEnX;i%SoNZ%{tz1orF^5Xh)olnkV4GY2}E%0ygy;eC#2Ag5i z1^FJ&NHZ6pBjJ@EsLsZg97?wr@PEj`50Cn`?agPueoMfDJ|u}`M0@3Qla&(|VuICd z{EB>`3_>fEGf!a?SS><5ZeYJU5i~*{fGBf6oS@Wfg`|cx%e928*5+wkLlT2$?x*m` zAkD7TLhNqr1BI>5el-UUib*w?$$$*p3#Ls!gF+#QPxPxP*iF3xOcDJAMV^=lZpUWg zlBFnjZ0O_LR4zmiY&x=Wqd2^JwO{@SVYl6t<&B1J;Itv#`5M%ZH4*rjZeDl_$-ZPOiTZg@`~~h*Zc_et*LX3x&$G?U*46Bk-M>011btM3wqsbvKY!#=PLfvOlRi!GmKxnrNGUT(A~ z2tY1%H8o*{=`(ELVO)h@a#IdN_N-CIqDq(Uhjyk}WC7HK&^W;;n+`RMk=gsCh}h~3 zC_r$gcJ;Z$W%Ao0B*4OPICQRIS{IrzN+*G$_KNZbE<^fB-gyuE>^4T@9tUd5BvF^i zOeY+1oAEE5AJeNe*CV_6e8<%K>Q~gm!GT%PJ3Xyx~}UY6+eK3~Vea7?hOwd_Wjz$1os~q&SEw zT6!x>$%62kvNzV)oTFC|#;LWvXYuLxQK1)la?y#gexvCSWd(P;^PCg;TRNDpvc4vPZlnj&c! z$&JC1juw}ue}pMQ{r2)Zw#`b){%F)+zEZwJ_0+qcY~C0M3*?w|KlW$}TiZg`t0XjY zX*g=fFAi6urmS{0nS=Pr8j&}H`Yx>|!Hb5mqu)o19+X+wXolXW#Ov`s{=p0C^(Htw z(dxu`4!l1!C)Z$(bU3~pEcthXJ|@Fd(_IEdF*vu_ZZ!U2W{MFI7 zY%*YOHX-&jm1!q298O7RNiheayNBp&(qJuZD{JWJPjofKfnVV4yPr9LrmM$zMuAuP zLPalX$P=-w1xle-a!zB>k>GT)AXE8EWv0k#k-$p!V+FUZhrHp8XNO=l7?D^tbZnA- zK;kxYqV4`5q=Z2ab5@44uskw&XH7z2tv zv#XAFakWaB4PD2$f8Vb|L7tAaG%uqj*SK`lFgsrl=LK=!|BHdRMh5Taz9rZGUZj)H zu*o91dHyUxy)rg>SsQ=4#Vz|Ltx{ev%xT|;q%t|u?BTJk2kc>)lCG)Oy94CQ-K3K#*9hIvx9oPaVnG~sfhXE;XbS}fJWcxDy^ek^( zx>DlZBEW(K5StLFy|S5V7nG@S-EUlAc7Yvl2Q#wRmgbt#%ttr1qYITPLt%`Cw4wHi z(U!Zc*ty;M%Vxz+SMLK~3))Wp#-Fx(VfiO$S=YgPrw2~2tq zG{`c07g)slX^JVQV%cA#Jj6k>n>rj`>h;3jiGLvf(#5-!TApU5btPh~s>olweFhV| zb06N)F$50K(Q8!dd5zs4Pl0zTnnmBMS4bU1SAJuh{{V$_u|O1Z;Hg;j+LkZao>}%o zOBg$=gYUZN$U<=4z#wwyz;!Hg_wi%S<_oeHVFHruygv&V;Ng2&dmIlby0Q3681%3% zn~F3H@er%3%v0fT!W=-p^fJB<6X#V+=7qtg=bk@(n%8+eWKu5<(dvF|f7_;Ox^${p z_{zNDK?U`194Y^HKWdkUW=K_6zU8SX8qgQf`NoPIy^Cf;{~jbDM4*Gt*g{24Q&)vW zFhgHIE#6?{Igwc8zUA4E8r9!A@sG*d*RDvgyyg)^xT6?U%O#y3IB=5+|93HYmnH4T zYf=kDeeUyz_!p_(9;{zCYgpCFla=`jV9*sf(g<>bbD9EQrW0fg0{Ueva+L4VX(Xon z1Y4&!V}^Qiq|#(`=(FfFfP4IY`;#CxZG>(_tIFfoQiZKEqow?I=ql2ctphrl!|;0E zw2F_@m&rEr6L+PLP7?4N<`M`W?{@Pku$TEt)k@FcK#n}Hrg9L^j|~OAUI+c|C#`@D zTM8&MWhNzIiM(}%9`}5x^}!C?Gz7bYZEMw>IR?`!^Gqar-9Lo<%X5gjo`My4;#5zI zGySbxAUr?u^ep7h@C;fKwF0%O1zG}})>PSwME18h*O}qrBuq$(=+4P+#t_$SI#ysR zm~UwI?#e}ZDNEMa&pcK=_vY357j-7X+l3&yjKObD^(qeWKD`Wv8#Gy`;wxR2C+bPG zKWZf*l;#oFoFawSj&)cb(%h4d7 zJSslC!$1mi7T_q}!e2UdavrX8%XW!x@-_rzVX(JkoyNU^p~9A|^~1iJ2_^NFYL5b8 zntGO}hZd_9dmPplyFI*VL0|73FJp6H$cL17*RJPI0%#Ze5E_^!d|j_Oc+wXdu=nfo zEdCIKR(MgLzf`%&2%Xu@E#Mc=s53#;|9+g+2@)B$eTUs)JjwoL#1cTSvS$aEF5Sf0 z>0;n_PTllW$9f&tTX&}4D+$Id-*V|#-Z}X6^prh|jQoUH)B2VtcwfDE`FPp%@!8`8 zwRTmX9n_mf>`6l4p`R=~q{-e%wj_tN65`&o6P<|^LvC%Ng3LMuUZ8{ch zzDa7dqVSk8o+6)+F+}iCSLLJ8jH{Z}F+Olw>eIpKc3;@De{@xQ^S z?V72Hn^uEs)P5GR0H-y%M>B_CgWnXn4B*4OMnh}_(s3nxDfG*WG%*QR&tX{E9;q`| zD;SrY6QouPaTmh0=&=6LAC6vB{UsNUewDw3ib<1JW=Jx!4EJgjB8zMd|gt}dTfmzu14N@>9({6p2TT+t(%=ssi` zZBuwzOS|kt8Knf8^Rtq7~i?R&3HZFjqNT zN?%n~9Z#OowKxEYD-3rB&e{<}$rtxdDh@AY;YR#J%La<9rSEa94L#Nw)K!I-a;0@+ zRq0>}ykPkmiUeNL9UvrbcM|PevekP%k7Y|Rd?eC_6QX^}2H9ktiaJZ}h0L zWpHT$ldtOldnS(>!8WH|PlZTf*lyyLJjT16{)zwHd7QHyQL;;B2|CIrstFmxgMTWh z`G75|z^)zCu=BR5MADJp_H(DMgUTGTmqso}2GjsLg_)Doq;7e%>Nhad(O+p&bvyX# zJ`AK0=-GB){);p#e$dmr4h;{A>O_Y0!?A05!Pa+RcyqY1$w%(Qr;FX4d&f8ju|l(?qch*@`l7_y*Z+ zKCORtFTg?a8u+NX-KMZ=GJ=}1i_`4#p#MeKZgzCZkxxhLy^tvlW5U{#CaZb3xmi@h zdod3}G~Gq3pujOi6&tu$(=105GvI_68(+?5sNXU*62w&MFx{$FMU6cd2LwNU*e;@- z1UP|a&!Gm7b6l-k;9D-p1vT?3T1^o74rtWFanx0G=3LBFrBKX?1AME=qW`|x^m|_4 z^r3G;FSRZ2O&aTnntv@>@0)Y7KirFAm2AY!pLVc3zt_tTR3Rf93X|A>RFIx$J3n6I zR0euyc0lP&#@TxQu2@0d?3@6qQXs<`)S-i#KeeH&eaaVwPLQ}C?6r<%j#>Tkm`j!_ zxWr7<(KMr4;e)9ZwE2v<5E{3Io8u=6G4w=AMGDrO3vdTS*^_X?{U1(zCb_P!h~J6d zo)Vz%6HZU9`eOY-_VuHzUm~y2yL-faUtgFn>v^HO_0(MMKh&`C9y-f6t<#f4EeeBgdY`hMnLKnZ)*- zY~*g!D|K^DQ-Qua``3p}F}OP;ZBtPCHn`~alK*>DS%ZB`hACa!Z>NT9>tk#yV&?2} z;+bAyzw`DqL#zqMp+t3Uw-1ZDumWWwx%Y>XZmg%&v&-1oh1HnRvm3P{_nVWFq)|$Ub_XS%(rA-+|MVG5`G$y7T=ki7& z9(4`;iSU1}SDp9~*+suCTq1&2WIv(n=K^TxcaW8%ZFX*My17Z}HxnEanTc+&9rCnY zLU2f{ub?p!Wv>L3-=P@=<#Z+Byg>Pa-G4g&OvxB{f8a8^PzKQ5Ty$T>Q9mQA$ToAO6YEDqjEj6{MhVgUk2kWFetj%|&#zq!V{HWP~5e(Nc_v$X* zbVy}%U3*C}Ke@yevt_4Y1@y66)|$G@@utkwlE}e(ES_6uTCFUW?oz z-VjE~|5j8^jZ&o}UGNpdMvjG9Tt$$IfGSlTs8CqIrrR5}VTuMB*d7=v8<5LK=I1B& z;h0l(7gDOSM{rL%VnTnADWjv*U6gp#$2G$=I`QTmcnkG^wrWCZ}&Oqi- zv&WIszbJAcoixG>R(C9D%JEdvP;UclQkBh2mo0KF){DJ&hx&fR#~5eA+ITLwnDX@GYLMb1l^rM7WLkA}vT!rZ*nh-C0 zsTi>zPw9&AT{8N!b0vJ!&>Q?(eN5|+R2|Y78K(GjrsJU8t(miV{u(&mlu+b^5t>{TV7s_m@q<;2USU%28UDgv< zw)!=>*vd5OMZ@mp_Y^#N&A8u$oK9$kZSZxdDberabU!fYzpMm-1g^!hKw4WuI*ECH z5Z$e9k+X?IsP4iu#A#Rh($XOrQN|`ee|%{ckaP5OPEly@hre*t;f3N>p_mymXo`SN z56VIx18&vtyKU-6y;J2Ju&KqQU&WNRJ}R5o=FCaHCq5J@Z~rY5(y405_YR2pksGIh zWPTZEVW0X1C(_3;{T$t8FF+LP%gCL7dH@mJzooSQv+J;2`MrGo(mV>vne27;%lBQ2 zfFLh7(kWrL4qJL}dVY2MvwyybCjhbi0wm(^=GhJk#2^vz3-D#?32^fdr05a;IEv#D ze!Vgv=zSUqy1nl27t9co_CsEQ7koMU+xlMV;L=yl*k0SJ)pAAp1H!)=*a0}9TlkHK z+Gpu1HWzWg`sb+9iWM3b`yHup`9~2HJx;^>6>$n5ZmJ%OXi;p9u5B3Cc4((W7ijfN zdQNnrQN{{F0>2DfSk+8r7PU8Fo6|$?_O;3y(?nB z&pIfx7>$8Eq>EvB|JNw`(SPwY(u<&8BF(^mgqMAHisJm0O(ImS5D!IM9h1z;=*kdf zPPFP%VJ;_V_syjkK<)7Aq!!>Jy3& zDnnyAlffedH?$VvfDv`Mq+bC15x)W77)6l=_KsM@%j=^*d%fF(h8HGaMaDR3 zO-)0!;H^)8h9adM_xYoy!A0fRm{p&pHWcdwEAXQuZq|C_Q!%iwft%F|sXX%{_Yq~6la^Y+OD3Q)dIiJq!UdXSx$|RGrHZL6#1JTVN*kta3pn>m}JY=6o0a{jX-L6v=&FuLPYQHZnG+-CNYr8J6=a+HWj~ttVS6SNGdA>fTUWO zrU-lpsL8RK@eC2YZzViog?XUmah`4iaLSxfxa>G{Z_aVi2Xg11eX)iOpL?{Q&nLSX z146jNb0_k^W%vg_ZGj0#QY2{0n8%0$VIkkUq3*L$TZ0EDcMS!v;l&HAg6*eijS#O5 zO#V5)sw5@ZmUN~79S`pC26ftL z5@){YE26Q%{2wdbY!!E>31>N8o|h~EB?0QM(0qDwe4arp~PUZlI~QaS$r-4w!{=W08xePQk}zrG3E0nn@rx!0R4e zR$ad`02YCfx+*0w#=zeUF%4{PQ=c`fTT(&(93uZNgi}5K^$NIi zt2NMvPG6s7`?txE6w9Aqoz#TiYNZ>aA*;+H)<&RtY|0yg7NdwBe7^iPTVf3W>xSv! zd`B>9!Rkc=y#xW7UABo7Axe-opYlD$FW1H_1Ze%CBd24RJIc9u*;OomS?W2l6J@Q% z1kVei^H4w&Z_RJ4UCLm+A2)7Vn1(;gJbI(W_F}i28)U6A;WCkR*Z8)K6$O9=k}}`$ zGac=CM9(#NGAg!I5)uRq+gdA1^A_VKd_cykHSfkM!n3s1>$!Zd%v?F3joe`LvR(75 z+SSZnZ(j{ArD08oWpDVayMdNT<4Q1d;ei73SHXj^-`J7yx`(Y%Mhv!rizLid@RQ%W zWtiR+L)l^ZL~&HC-h~b{wp%6i7FE}KpU_|!RU=gAr(%y%=R34dCY(Vf;QW4q6kN_b zMYzkBb${0PLv6zmShVR-0=OsD*B?TmN1j`-z|Bfg*QrMh8RsieH2+O8*UTLLrwIVc zL8!PMIl5tz(EVuKbRRx{gb@m~WV>`FEpp0ID6c1$uDJZjk!dkwK%ab_QHX1y#TyyW z4HX(gwBx5md!KdXIIP#)2rAm6@t4vk)jXSd@(NZvITy#?=8O9IWG$3U0Lmy2Be~UD zjmwAkkO#aoh4Z_8Ywfo~h<49m80P}vi~+w667%pFx^N2#5USUSdCs7gxZsBAa%=ic zz*PAhJ={BQ*|X#`!A&6d3|oFeVOfdc+0?{o@9XQ zefCYy%7l{Q;Zifn@W@`w&)GD8jA&T?WhPZ_vCA+Mf;!B50QW8gvB4Tjm6!&i*WqcN z5AVS$?~`NYAs8q$svbA0dwNEHbZZl?Q3?eppwkrvQIV(9j8V;IVWY9mOpR- zq@zK3g@WPVn4y?=T!*OK6?^$a-NVV=Ui4k@;%NA^Nvf!y8bUZ3E&?PI!oN|HY@M}LVr(T^m7$le znRmgR8s^IqxPyTscVVrdEG8X?zS&>u+1nEH13F5;q<$grz**flb_%QLsha10y3Aur zaQf!h7~Ta~J`ZlXQ2+IMqisJ$?K&CwDBnMK`QtiSuU&3t%_Yr4v)iCmf`zLIACt3@ z8PUzk#DzX^IfA}*2(5OvK-zWw+?~!R+^^M1{OHewVzef|qpOLF?aG>czjCLg*SeVH z>C@I_YPfUD5xkHqI88H!eGl@-!D*P7;_kj;%nqSqZ;5DB9?Z}jWNa5cQvcl795>>0 zsu0pwJgEQ%lF~0% zpYkWht=Gt^|OAKeIbLtDo3s_cXBYlH?o)wk9Lp>og$t@xomK zYdc+GOO8-#Cu~9CQPPx%ulJ5M#u~i(I++%7M={M$beJx`n9VTf@61W{{}hE6o^L1W zfGr64dU!U10-IZw+{_S+n|H%aGsO65KcJ>IebNXha4CU1-?I&~{jeuBf1Y~G>mj;7 z3;hzGqd~DK20_?y-1s3|XW1vJz&e&@n+%|t+Z!IMI*5V(z6A&O8M<^2sQWSvQx`!e zWlOn~{BX_V(c56JYSP7dNXK1nApp+y4M2_yA> z=rsH5dz|>sj2haojsiLY{5;>Ccj;>nAXpyo$rOZ0YC!ODIHmS9%(pK*zR`YKB%APX zxtpJ|Pe~TuUqhJdm2Ovbx$NGBCbJfJOx4a^zNlJlk&AhZt?aEz8_KG zhyEk#{b-VnS51Z(vg2CwXL(0%`r03PhFW`59K{Z#zvZ~q)Nk-BrL6<55uePsiU8iS zl7Q=Qr-oHS)eEkpBqXAWgkrSYL=_-(*j_`~S}90dM98NyAJTBjAl}ixhsyPxu(fgR zw4rsoh;cb0X^23__Bn=rCm0mWMr}}#L}IeCCRQCSOI4ZM0PMn($Jto_S$T94j1wcx za~cD>X1UF|T|+SFjGH!1d;Y3mxY-NJF^Q;hm+L0u`c3&+C|~$zt~65{=X?>bMr&)M zimGOgd|nlC!^pU0CAz7)ojLZ-AuPCcTJH~J%nD6b)Vk`EK?GAfL|7&EPF73wrxdBf zDp^*?4&e%2gO92Om)*}jZBee#D-FKyuC0ckr5yXvdAGQ;p$f+Oq~dnjw~_!-p1PU)a5$cfxxHN9f*zxm%bZ zWrbJ||Eh$ys1(H6%Ds$PI4$>WL&f{!M2Av!I%O=$XM5LUF^2odO35OmMHg zJrfsVnVi&f^|`XviLVtQNdK+rOa*DcSl=|n1qxfSWQ_cvd8wo@t)0vOQ~L0e#;g4j zF_5+a^0)@KtcS;r=u0%)b9kL$8B!Qj(X3&QP;%>>64!Uvx-yce1P5%25OfyTWMvW- zSt3wV${E=~*BBhRTRKh*=P9bJ=epNLefiPrQNC~{#^^px5kz7d5d`6l6vo_uK?|rl z3mj0|@f1-Pd3Azj3EeH@F0|vQ!=L^&e|GXHgY}qjSot&MAUyTpSZ`6Sj z!REFNTCIq4gp5e*Ub4JS;vVz2NsPJrynas@gecTD!y|tDNO!p++cl~gSj5ySI{%?$ zhijh){&IDWO*=$HQ9ynAUGALqPj*RmkX;4$$P{&&(w+HH^n^b6RO-0D$-bngQT$8x zQz_1+C)>cU*6_G3s!Ap2_f{|0m)XP=c`;GG{nZA3t)@f8_Ja{`Q4=h_64oh_2 zdQFW4pgl8j$;1GD?+WK!W37oEaLUE4`9(=vf7UzSt-@ippC9PzV}n2k&nvy1=k9CN zNYXl(Pz(jrYsHYJnXy>u$>@ zPG)qvjM3H*GgPSS=QmA-UU5PV!yl_?5KbZxNbSK14(`O;wc#{0(YG#!nqG^*m2gL@ zy+6jEcBf;WX__u!3Ywt5vE2?qY_wx5BbiFM;l;+(P2H*apemqFVPY$f;|-~{sN0(N zPqs+cZI5y4fcD$XaOyJLkNaMW0Nc%5F>CFuy(`NK`ODix623Py`k%87s@SEasLg9= z{ZwGT7!b3}R4yMU_D7uK_VX@ixNig4m0m zUMXafw-~72OauIuEZ{v}$3wSb4^UFHNc>S<*$?6%_yiQ637q8=-E4pT6dF03{lGvA zJY$&aV;lF@`ZHa(II$+AWsarZ^7PjzY$T@+|67k2bgly=j0|hdegvw5A)T)snl6R! zwc>B$LtP>LVDv4}SF=~u9&Mv)%I|zs)>@<9N+PMI#2qXNmz1@;XC(A4Wk~2a_I;Sm zhK9_xI%*;AflPznC zEazv1g3hXDG1&}R#!#ew!iKe)RDg<=8?7$tGs=yj$k_At!n9s{;V)B`*L%#7+8&x> zyU^(PIA4A#NA3wi-P=i%(-87*XO!3QRwQe3;|_J^q+;<>2$Iv6zE{^LXE@qzsGyEZq{?aPght8xl%-ezfz+XVCf-5q2Gwug-QEq`C)nihe*jJYB3B zlA+mrL!>J&(-ogM^~I(ti%+$8^yzy!tyMFuB=$GZycSKZ=)DNX)zwN>+Y1W5j=m7z zZ(p?pu97IfOb9TCXJ!wQhKg(PlV9wi?1t{>@kn`|DXcj%qKKO@Qr3bf*wl}~mn1M- zUWz#_q>Sflzv1-OH7gqyFsnP@%fJ~+-3iwaW_4Q(xuooZ!KRPrvaJC8&Wn-(+S|Qv zk{84vH6ye8lJ%!;p&0~iP?GLV3-~q1`&+;O*VUJbrCar5HPw1IMmeus0Z>&;wnL}= zNBLVxjyvW0Uc=YK!iJOV1~z@y>;j@ljo8cAb#~B)i)fLHhiJIh6XwwSzncb>s+*{; zFepbgvH`b@yTBye@N7$70dI4EYIkmpKO+Y5dIU#@ac6LF)v57SnlbwoF{sQ1JbfaT zE9y`i;Pb{MC5huK1^W=%_t_0-E^BhSe<_HmktGlnFeh205 z$>AxoIAO~9k)@@sZ@r3NOlm9~JE7QDY6gHUcM(9E^4ZBpw98Ajipol@^A!HEWvK^k z6Cs4y_e5p1;#B*aGiw^x4q^v4H-_}3JZ+ENq|3vKVJY2kSj~ahH zcKQ6uv=W|1SpM<2w7jgfz7HE06^fq2Ix5%Ybg8?o*{++@EK=`z!HEi`)MG-`kGopX%Gc|~ zk1w8o%6_o%BLmV+e4fC2uTZTeq$v*6_8Z1RXO0>&k$*-piCU)P6!AvC*Y@+FRSpbM zPOGd;Pw8cL89|PC4Oey^fmSS;@fFO2cg$u!XXbW@VSgN{#z1^1xlu>C!3QeP$XGr# z7EVIQEjLqVDST>sELSD?aB>c`Q(7qVhfQ{QH#9ACk7CWreB@LXihQ)?W=3VEP^sW= zDy0`e1*bqEb&`pQxf-LJ6{mxo`z6$0N7jr0Hgn2kNE#mui#2J z3!M3X=S%r5xc(ns%Jn^Fx0<5vLUQWmN{WP;gupB}jq_fpWR2 zRd=XwjZXS+v~RS3lLa#z3q`rcRvNyQqw%A?dd{4>T`%Spd{S_MVUGHmDI0vsMi%TK zgHEyUK??nU;(kum4L(qu$%iSL9b4>@rQX@Yb{`Y>THc197TKkcOoRJYb{^@Y7cD}yWFRLY zt2H3T1}jPB20K@25TlwcE{`)^T%XziwB(6k0_p7^EKHG5Ak;uV`Ku6`WZ2+U>U_5? z+v8C8DGvcy>Fd!h6#ur>7@4E6WJ_33yT@3Y39f)8tbow>P!=GM^dLWt&;{=Y} z4Sz0;{Ky>9#=o|Mb8?IpCn~&s?kJ*XLJqgl{Y}%mzQ(YyBih@$c|i;D5Kw7?2$IYZ z*61s0b(4SAK7)Gxnf>O}`Xvtf>ZSiSC}I)TYS^oH21hun8Jt5k5!=}^I)a(gus(@|SO_lP;cDq6JA@j6an6nAp;w zenvno92PGccAkW8S+ws48?Vi8T~Mth>J^sU;zVip+wFi{asSd7e!^`$4_nL}O$0}R ze;%-7ihfSusKIGUGUvyAX79^LZnX+f;NrYxN)7oS&8MC4qiD+VrdrF%)V+DB@M$2D zqpPHIo-013XOYW9kv}=OF~yg=V)*?&(@Dz>m0onc1@f31mn1@L%y`-mb=L-)Gu*P3*80D(kcv_fTJpJ7VD|7f>@NIV_(YL3HS{N+j&&M5pKR zE|HO)J6_Hn9(WRkdE%Bkq3Es}9F2@Loq#TFPu1?gKfQgiC1yu0PS))fuuOAh8hq%sYeAGNvi6adJWxPu8p@Iugwf1~kEaLG@6-+}XD#h9nizy@A;OKZCjno%l#djrNn)|t(qGZ^( z>R?HUh{|@=Ad*Vhnmtqx3mF97g$-q!43A}&*)pAfd2)tF6N*q=fV&2H;iMfA@oH2^ zkvLiv@DS9S*(j)ga!xW86SnAvCKag!Mx<~c04w(gw`M&6G{a~;n$Lqfe}J|lxAL>e zZS@AR0yZCO7=S(+N*qG6QMiuu2K2)inBsMGK-}n6Gf+9+T zu!%ZSs1ShOi>cRu+2GjAMByrr#e~AHUdOe@;flHAFwKu5pq_uCrh(E;U0n_y-2QdO zR6O25^z$nU6y1dMfD2RsL_Lb4;x)a>8y##e|Da*(ZtH~KsPn_+d|L4DdRubTKnx)k z_gEKC2=rsL6EqmrlSldsw?;-z`3whx?M6BHYi{gSW~hu!CJQZt2>Mqru`CQuLOpfK zCf&(ajgXfa(MW}7wu1UN7}Wb2SByH`f{Q6{3hJ0OsM^*qX5Fwb$QlzT42+=M)`${) zdYw{MQ$Ll?qspOAI(@^I#{_|e^ko~BY=Vf(r)y)|l(k9{c}6J{H%Rki{yK#vJBkDf zH+)QS1m!pUYbSr~;?CMT0cFX9E%;m*xxO5b&}SVuCHnB(FCH^GRJBu$!99KnDKn?+|9k`_1b)y`hR^- ziZCdgu%j$QG*R1+M{s#`A>eQa|{qFqivV zXlH|uaG?%V0mPVdy{VSjJlRrVA7P3CK)HW0B(oBsjG=`hhwPHoC1{GYbp}*@^faYz zI2cMgCbTuf@#1g;_Q}dJ4D2ls>JrV(6K;5t6M9M+ zgS8)k&vtBThkjGO#BK3v5XJV1-ha10^Y9;{Z_UB+lQe zGc+kYk<6<4uw-l=<|dysv|)QZ7LKAzuZUnm%!hD%~(CC*S{ z5|>&GWuKpp<3M9c=PIQKCY`vzq^o4A)!`<#&!2p+XUxqHNyx7w7tf6Do(UYI|AZac z)1Bl5leOs=U61e5&G3U`6eQ+CR8bsjBawuB>gh&YI4*gH5TM3Z^c8&n3|l2=g8g#T zN51gPOb011`w9_e>Hk65IW}nmb=|pb+tapf+qP|6ciWiOwC!ozwr$(yG}BL#_rsg2 zR4PB>RPD3ZwXbEhEd>APhzbktNZzE;1@n56^sf}r6UcIHaynro2Doi#oM(;J^(RFK{R@SY&*qLsc;WmaP=taWP~HBO;S(QYImFKLc`hvT@VUrs{|qr)C2LAuGE7YOCxr=cxW4+ za!CwKX^F|?ia}vLcUS*0%8hsfZdmoY$mvMW7Z!OSOzh5GjsmMV5Hr~CJ;=iwTQCr1 zx3KUQt6(VMH%kYB8_8=2s$PpLZ;<*t{A0c^58=gLB$Vpb@KrqQo*lLlp0fv9?7ule zUuFTDEAOGy*G4!9LD%CI!)e3V;``y@$iva@Ah0zTZvp$rRAGjytOWeJ!)Iw92a9~! z@3pMRvU9PwXtqRRhYH91mBPfbDPl^>m63)aF%3wr35oV6Kti$$}0eRg2_e&M1m*U-1wErw=~R<16FF#hh2A9 z9|xC@XeMXW&Tf0EC~)LJLzp9$*4Un95<=8S792Lh*GV4X=A;`IpekXCGO6Wfkh+no zatQEykV+j4@OvN*`+29@M;G*!`K}{HMB>EeQK_I}A0L!)tDl-SwIaVnnu@+nALFg~Fg=2f=^ zaf}n@**nsKk+->DI@G2Znq!rDD0mKGvf#;l^UhFR z|C8yAVvL%`aju{oyHwSse`8&EaN$^WOtaDVdid85hPbNK?L7|tcs;4}X(|1R{$BP- zlTqcUmI~4#IUi<2WtI+|R~MX!>l!>32|*RmnrEv$@!P)EGX&iz zy1G){JLbA&8)HI5O;>@t&6)R>0qHp8f(6AD%iVoeV~7-0Z8{A|KbBPXI!jfJlUqLK z?mq#KUE1rEr^8!wu8#G$lGg4Jbc+z8##yqrglw72qfnCI6k}1E2LJMjnzsJ>W;tn9 z0?C|m@nubo_9b9QOPv=}{EK~3aKwDuuaB5r+I-&$?Us5tU#X9bQnd?cCq~^B5$0?Z zqb~-AKd=fVGgZyJ;h5U~+{^_}zF1_`qC>4~jGmHjqg*~F0|b{UDzgz+oX|Q0wTryN z9AP`N0LZ_&;F1r*4~wwD;rH;-qYKM=31>NUTdh^f?jYcbi}YLt)+j|sS8^?VYk#wC z{Jjs1WkY%4D(MSG9|Sy|7N>D;X&Bc1ZL@=>Nc(wkNlo}0qvzwJXJ7c(#L`ImCelZs zZQ=>W`pV_v-KT&+VoX~xKm;l`Fe6qeL@Re=+kh+2`#<5%<3Pfy1K~OeCJ10{UH@`D zlM;Ko4p1qeO>!~u0_%&9c>dtn0#+G!#s^ueCKNz3UU7}b)fWOWy~iV|>uV*eP|5k|b)+M>EbmH`Bex+pKiw1v>I(B;`_&Y)f6@cNdkmMn^rZv89fUuyuq zlJLup>OIJr5_hX`!e|gTkPK32IzPbBSm&~7rFQkSw^~=!*!nIq_B{^g{k4T#B0S+nu?Cs zh6p{O8PL2xm_GJ_V(WTv0Xv}@pF9?r{d)*VKVSnA4q{_Eq=scL+<3k0d9Oj6S}{^cKS0p;8o}eBfGWT2+urXH&?fe zh|O9j36AVGa#$Cpoaj`KjxE;pGU&|#H%5%GW64Yh z*DzljKBCfNR>U^fzc>`x+_zaaSy&!Rj<)dhyDE1JP}U8*qL*Ks%YbgyF|+qq55V<) zuL9%WKDA#>-4;0pkU^l6o=_@S%+EL(*3ro-JJhB8G^L-AFD~5-}U95+jNhabr|b_ zFpi(=Cfm^!n-I?>^eG;QON}i38bBNPIqvD^bN-LxzFjdiwh6opgZVO&oc;t^${%RBh|XKzW`cp4^2jeT-*_Q2AUHVVIza*C9Lo~G`MlmkeDAdm|r zNT_v7q#^CF-qH(B%<<6K{sP`{f!fN^OBgAiSXqjb8{za6lheCuYlP#HTH9$I=h^jp z(1J;6Y5(>sMEPBG*hzUec=|%WZ)hdB0p*yy;E{AtC+orKayi0L2@c>F8)BKqzt&b# z@vo20Y$b~Wt~=dxM~m^Rtm|x+3RwGVC-(c2{&mk}*3PaP-hQHA-vTm7bD&@496~

%qaA#5B_=n#P3L9078WZPr#*8v0Lz=rH>9ukvQ;3Ovwz1w*S9M z`Je)NZ0aY^t)VLY7j%E1tlzi1-vNi}hu}QmFy2CeJ))DLfQY|V?Wr|TatakCMRI+1 zmz`Pi#p)D8!m1#elV(?8shJe>edgP*H0i09W*Nr)uHwWybT@#gkZ0odERms3GQh1Y48bp|t}Ck5!u`goYyhh@|IO zcD3-6o<#H)4CesWzBO)IY2@QJ9^s0(;hZjfwNWc_maosFpXyQ?Hj4bz7#f)G%6p$1 zmP=H(#9`MWl?9OV(CMXd))iweKGu%(I|E@;*?1raAi01nd5TT+BOfAr>fFssXmX=g}QcSwss==1_Nv5>FUw;POJ1j9h83Cgwds~7(NpcrXbM2)br}6 zAx@J5uA?|Eqn_3;8KhwF? zoJDmgB}Y|~GRIVR?_S2UP9?wY$p~$+V7_>vSWeF8RRhEdKMIXD)Hw~c3YN-wI?*f4 z?Tn#9JAWH+#Y7tbLkp(Rd{&!G4ozIHMoUW(uSwmeYWmipP)J8o8e_teeIHZtt6?4i zu!?kI+k!FLBbMPi#dwH|!3!J-CRcM%awV5eqhl8F+Z|6-iYWk|F<=r9HjgY%RjE$k zNx%h+f-ie~20Jp~od0;9>`QAP=%+N33q}cjaT?+$QX`a`8gDdC5DQU@P<8vES8nw% zG$b48f8jI;-QTThAS^NLO)v7i&?2@6AP>@SmU^1)8*o145z~Li_HNrHdmz@!Mf4!B} zF?`TRo9R5VX1A-8LLt%Q(05H8+ot*wX)DIp^6-z1=5IlU-)(>qs|^Bl%l2=4=ai{U=sQiXT8ecL^{Xq7lWo-?0Syk={x9q{ zPZ>2zd8_R5sq+h!sDwy6{bhZgSxS;u6d~vz=CK%MK1o*d3+-N8wWcxZR=K3Pm9AN7 z!SuW}s*q#)JZ6~fWgd4E@>(j*C zUwoH+EWgRcuAhReK|3h|2 zl1r@8fH83%nlS@}eIl#f7=7J&P80%*473sT1lBY;KWNgjse|;R?@Stb)87deS(!P6 zNxE)4CWtFJnl%NUNlgJcfGt!3Ix&X(CO2KROXfA#%M=#+B0CGE&Yf9x5!M>%?Ydx_ zMtZre4sPzQ-)4<#SwoA-vTXY5<<1S;jN5(7dI=J4jdpi9_8q&rp?#(;gb?rPSBXx= z!rs*tRqs}362RMYNgA?yyX$>U8>W=3e+bH~{D;^+w#$O_<^!CF0OpX)*P6@#byd%{ zZ`KX?Z0n*hi*vjyx~A`A;*T7j;Hs&fKbw8<6I@pL!35du)*hR}w;zEm`Gg*uy3+&( z9BRUa4yrs=XZxm=IEuA_^B7LWAoPsw2l?I|vPdS&i|4Dl3_KACd5dR>S}oBJ=ffs} z?_QM!5{}*&#&0dVfW`dYEpN0dn-q*n-V{9-(St7-X$v$5_88GTfY_LqW;c!fXwo=v zvay8nQRK)xY8_id!UP77OQlcwuaX?4YJ){k`Vjmxo`%zzcp2NylO)J}P*ENlLl=7u znSGTY49@yt2TidLPTQ&cXX!+P^vVm{Ykfcq$4ZIG_-2jNtr~F;TOnk9a-g z2=#Oauft-zhdQDP7=u0EudwUPu=r0Xl-WwZ)G^Mk+K7si`GMJ?K(nyXWf?R=>uv{V z`}d0;hDf;;Kws+OZ0!jb_r#W;hezSlV+wBx_k_-_sWu=z>GHWjKRMwtSfv=C?D3Z{ zq@1S64WoptA)NsY1xutKX#qqL8|`0`(wLQ;rXw7qot<5`eN0R3qaKj}ZHpXh;Ro*W-n;#I^t4WfMbqEa-zxDmD{p*-ZqT zIUYP%pRQGuqy}|6fkdu-_uF7dzR@0MSM8&Hf`6~QBFo=66nUW!%e zgVvUAsK8{fMrujM^sWrn8{|Sw6qo9*4B9UyUK*md>4+;<$f1uzp%eOJV1zoF+#`W5 zWwb}68axHZF4lwx-${dQyTu3~Y`+!!H)S1d z0kSn+4^ENalr0xb$c_28&5c+ej-*Dh`Gf0QguOZ%;{ zOPXEsOD92!j!*)n}julwHum0xI()&U=g{BUbTUD3eM)WvwqkmA} zctnYG2Dt~}vWpg7I_FBIxSZj{@M-2b{aWdQbX>>zGVi0c~wKQY=-()DAiB#kTA@c#c))K};T0{<;hw5z2C7 zP^BG9S(?T7_Z?~)DJ$sD37A5s`{Ys2)U?oxAh%wHWi=PNZyp_lAF1r}`^_I5ta=d`!GDWO-10qM z4$BhWv{rcsOX{j!(|bKO0pj3Ta_Ac{AE!k*2x2i$F zSMbU8(=*r0Z9|iHvE&T;@Ix$&x$y|ueB^A+yXV!ztJmvkQ2o3*fYOFP!}92bLWz<) zs;{ZsZjWXLBI*j5uWiU30w${GM(Cvvo3&jx0gg7= z@`2bOAjF6sT%jC{oXe+AD&`EL!48E6Wa2(`u!l?=8&L}Y8D>9`2Zj<^qx@UIoQP6I zfLtF(QHyh@CHOp)50dS2hrN$ACk4a`&S&`M5f1(VE;5nm3n;Z9AsNgJ2c%Fo=Ji(K z{Vv=mujo(XAE5uMJe%FlQw^&4KV^^Z=u=zJp9SeZ1soTY8c?185u(?BMn9qUKcVhF zq5eOi0knG^r4h6k4E{e=m-KskCm32FAkpk-%*OvM)?`~x1%Tpo`wTdP&VT@5>sZI) zLEW8ma3U6xXC81@qdOUO2IH-%^QO!I_yNlkkX|Igm>xVRL>~{FDI%eOLIA?P8Hi*2 zD-FtwW#4Bx_6P^nZ~oA@i;zpeJ=!;q`wRNKJOwiATbOJd928)HppZj`!pEiQ`-Yz& zo5#&i7oLP^5^|hzHA$-bKz>{ayb9=uu28WI6% zK&<*4-uX{&9BpMvW@$nJG5nfxI^hdK7YZfkYZ)^BpB|xp^*;ut^AAH*1B@`<0T90HoG=_pS)qAE$XOdX3r7A9kr_!BOjzmmT>|cP=@v-)hW5zc> zmHy?7N#9KU(R&{G!`WV7+xiDn6Ij#B+Z&>3SFHXp#IaWdfhf?Zj(*bBJJ*;{O)9~4 z*LzQ7%?N$l_q~IgQ$0CW>uwIl6=~R@KBizmkL~mmQf~5%V9>jI?W4H(&D4E7R)PRu zN^lObVN*u9s}I&5=Pm&J-*+j=$rm5D8a|nK^W*kwyG+bkJTcU~@jrg~5Apmft$IWf z%H&)L)%1Q7+@YLBGdQtw=2?risRtcg!AIQ~#i6XY_(p+0KAp^)8~)xv90HG0y?8Y5 zh&VN~&$ms)pPUh)7vkD$ZKvbC|}e!yw%-oYPYxs7F2sL~3<86&^aW+~8wP*Zw)GTKDEh z0KvjWc#v_)8#k5o=eBR(7PODY?e^*yXsg(ofQj*PQ|#lIg_v8K$V}teYSS6wf$I>x z2zP1#Ar`ae(Km;2ElC6@27@1R>JVeP9n$-iX+=WJh2` za7A?k=H#mao%Koa@B*n&rGENRmWmWV!OE8F(!bfUYO(=_DT(}y7}h>YC~L7Q1eoE_ zNR(Od#rpUGpD;qnZgKGU6(zh`ezn>*a7mYmEf}*mHBSlKCB^a=#gJEKwY@u4IVdkaN88n>qvdsLnr&8A*Z)3kwqs6D%kezz(RtiIMSR|TJzNBQ zzC5)*Ut9nVZ2nz5M1q6vZ2R_tdJsbNLmDL$#8^4aLdg7?jB8y3x{fCe)GB}@r|YT= z?DsM{heSG8v=!p*sVz>aBfsJ7ZZ#4M$%S1G!xtrsSm(jPb4apdoB|$=MVn2q%Qn=w zgrsJ~N>#m5ci5QMI5PDEY2Kt07BAEk`46NtLJg3MSY?wyTe%{Dh!9N-1DZ)mbrL&? zEa_vwgZ~;d7E`9s#A>q!@hh~@Ny46#sgGz;ajF6OtDQL}4_&qe&g+h1(iM13aK=rf zfu!kFV$3~~If`8sC7<=HKnrIH4@AZjkGwurjBy9lkUkAjBtgm4wlb>R{|}OlXjO6RMN^rpg^;QYR*HIk7F1)5Mm$6B>5TjBlkUJWJNGhKMKQVU-%GE+8 zRtdaseJ@EIP1KhkvCA{OsfL6X$U+F`xdMOynd4Tskv1gpzSAyytRM?%ps{An8{FcC z5ltKg#+_2obCfUTEe>|cj_Ri92kf+dwNpJkWC#KCx$~~G*4IDTR0FfD#ge9U(y_!t zod-0pzC#Fn`AQNYlx3z0e9=ZW;Pa{J=ZB1a#!?B<`w6m3#F&;Y39b_?2|Q+-+5os~ za~W<~y`a9Rx3ViM-*$jDDc?~f+!+4wU0{Zf^vUu8{~^BQ@G3>Sq%{MiL+chkA()mhKN-i<+M);gizrbpuqGpeB=x0Sw2n#wsCELYrZ~z%%3{{wJ z#8;1B$dZIA>kDS11_!x_BR{#Ut%clL#%iHNNvA4UOdUahI_ojAM6c{&8UqaUfJ7&N zwt#|_mkgOIlY%RkyJmQ&DFdx|Rlb{+>l<3oCG3g4fSY*|!g(Xu?309Pfvefz z<_zdIUHuB6zI3J z>QqEbQTO<)3vZ<_HBEA<>2Wu98O>Y zLEy^TM?e-S6qKN-LwP>5Kh?Qig)>@|$AF}e;sRR6AmgF;vDTb6K29ai#KGi1)S?Ph4ho}sUZ0iS=Rh)WB^M~$JxW^Ewxz7xHq^zsKcDucnJ=?_M=*%@ zCcs<3K1$H-CQmuXwrhiMe9)&@hswue0JWN1X0B^liu_v836d#iyDt#r{B$g-nVXSr z_~r#MwyXWu`$t)itxRw|J{^fCSo^HDU8L;DDMTDp?)|xRehILP<7I1$=0O$VDmB41 zAMxHVuCuVH)EM>DWa=@-@Tn!yvLqu94DOpIV%8b?vBShM&x6_$gEv8QwSI)6m z@rdJ7{ONH1?k1Wl&5ltsPoRr~H`tF70BO#JevHVFNn)zNNMpznRFFZp7%p*JL=>hlm)RILthVvrV$91syYQhLoVDFwz+|!XQ;r-= zV;N+bu1LaUXrV<5JIeatk_Zl#Fkw z!sM<+)onHHYgcKvziM@tYqZQvg$`BI^~Ym~PQ^`9Z9v6zE(d~7$+U`_%vorP_mmA~ z2BH|xy8~#p1KS`?!LYy?__yl`@7DndV-4*omJ(6IvUspZ9|VR~dvIU@2<}Iv1x5H&-b}I(?W7^0w@mR|x%H#4Gdj228GK zNuf*y->h);9$ubg57k)&X@ftdgvI(M>@H9Of&LLDD>wJ*BZRs$<^ za4z94D3qjpsMeI^BXJG+5n*^uhW)2PyR?u7yecK@i%l{5D&;_Lv>sYjX2q_#>^nAd z4Pn4nSvZx-oENE3Fx`95YWsfMErph(JC_xQ4C8XVTXC6de-@9<9{q3qE^4??3*|)$ zHi-ZaW7&ukUl4EYomn!%tjj=)0tK07G&N+!Frn+`xmp8m_Km6fhZo>U66#$zCr9tj z*YbvAmx8vzBm`*TqEtryx7uj1Eau2Ex6LG}#ypwTK8x7i zgKdzE6J9|ujw0zY{qD76MIxVQ(6Xv|DUh;)qz_s|T>AItrT6K{!`X0wtkz%J3-+USmzqYKO zZ#FMjQTmQz|0Re<_}4q21E%Dbs4@V^@yjSz75kO09|NXeB<>Ux8Heo}%EB72j%+3D zSmZL9E8lE;(lqBqnLpP=NY0-~pe_?`rf!6{&>e~Yv%TlSqbi9?ooXV7wM`u8v}w`o zxYMOOF4UT`^Wub`Scqs3Db&z^5_3odDcHecBym`G?zB?~rL@`{HbMP36^;R@^B4Su zbfdT+S+FcyL3SPI)fYw_CUxY)baXU(c`y*JXIDB|f9T%Lv)?+uzbo1?G1beRZusoB z*!sS|9Lq9?l>g&ZGa`{K%$69Ts-ZhXQb(Ie0nPL{rv_w0`En_!`46rrAuFPt2DV6E zsp!uc?C~<`bA0bcxE*x);InL1fr^3C!}c!RS?yaf%5p z3RBx4W}>1S{cQnb`59{o{l`iS&JVlgQp9>6%_D;-iZ%paRr&9|Nb?pzy|uXc%6z`v zi8$z@CE>Z5^i$4)*{Jg~ZY;~C%#@?pTH~CFSN*-7GrAf_oY%b3yW($VqG8tYr5c+C zAA!hQ4yPLUmHaJZH4Rk--mW#~uk;mk6VEY+Ypbg|O}tFoS_b?cQf5)HHLYV$$K6~S zY>M(KSeP^K()yOTGwo8qmUAb`pP<$;s@Pov=`y{dJlZL;ZhBZ~bIOyZ0bpa!b5^Ef z14rM6(zm>IGTROma8(0;UIKilUS_eI`xiDu(#A&7!<0ZYFQEkbIx!t7X+}+T)XNrW zO+Ypzv-b{@ow`aWaJCOc)#%dYv~kJ1xN}SQj%hLJu`%-98yvGQ|8C1eJp4FnQicD$7J7c+7 zHBI6r#x$^pF86)8btlo`)Z>*@5H)mSlS)h!O|%YT-hQsD6lmX`ZN&UiAZw6 z2@ATbx_Gwnu4NV=dHSneLS3WpZfHZ>AK#QJ0nceQlh=Fz||e7Qf;1 z$7e_%4Ze19GmA)3e2t^dqf%|kj=d{$`>{7Rva?dE)NU1^Uw{+w3oq%tVV`-CJ|~o6 zcw|$Q?22P&78^K)y$TxYwD6Y}eo;oAiB7ey-ze!xVvE!dnP83K(BF+OOhi(~Ao;g#V$pysFi`md)dx9cu)IL?cPDZJLcdDh>+Vhxbutb27Kl%9lS4Bm&T4p z3n(P!VZut9w?#X*d4sGR-9^Aw@-eA7%(0u5nH8WM@M zf3_#g_fDPFJlc9#hs~Poed=rJzlq0eLiLNlzt|vkAY(2BXX+Go2B^Gke_PBF{u)u(6*jC!!bmc>!u8%|Ysz|9~7MA^+MxFINZCX}s!_J>rpNRQL_I1WbMi zqpH<*k)V<@&EpH5X9?AWWJrn(p=4hV+;lCrDs`s7FPUFvG z5oH~w9#PAOf35AD4G3^EbwFNM5w5Gx9`?kWtxVQlt+YWzD6Xr zI9Q$OthnVkT+vc%RodpwvUe|qd?jwLQTfAxjOg*Q8SBI9jk$0B5)|Wih(FI=9QR{% znN3O!o9BkjE9<1k?D*a#lQo(U00q9PaHYzt$Bi}bQ+1{ry}`Q zVoSNL-|fE-Ak>$gwV6>5JO5y{iA_t9amc#b&)~Ggt0$5`k2esZ6`f`-z?41(kUY47 zhCr|&iI}(aYzwn&8tg4R$yGt~gYSzf@NKQ5V+BFUF-40Yus+6WxkfmvE}=Z3{yH_{*PuI>T}m4->871+ z3>W>n=k~SLhA;od^!qE;9}42c+9cze!*A(Ws=JYPRrHxZHks$|vEO(AC6wQ0^l6bY zF`FD>Jw~>T^sE8sO8E#{L$AN%{odYcd~S-wyZydauIb{>XvxbMkn{y>1*zJXh8vJH z=XSBm6v!J$l+_?{A~P0uk?5B17US1EuC<{j??&Sm@b3vFw}ldJA{_l+%|@)PH?Hbb;a9~w&xzCF@hmb)|(1vh`;tG^Nm`$j_#xxFTQ_9pP$R5X;`Cvu!*x!Ri@YSgJE_IrgUJzza@xT9S6S7V zAYh2WNwqWU(d+;Ts3sawu4<^^)~i2cbo_U1bk^X>bE=XEO7WbKr% zZW>pTBCO+j*z%15Z3}ko6~lVIy;GJqn-Za$hM`$C22ON`7&{7gZI+Bw-BIWuJNQq> zUhcF9^srG`j&nG)#lF;WRby0uEpz4<9;IZtd~o+S0o-{`(VIRXA2Dag57&|^t~%KWHP!vd7g;QfDw`4h(i)yG zO-LRE^EPjx;iA9uH9KOCRttocJh1PVj*8Ai43MD;;)ON0Ot7wJZ$oV+Agae#OA_Jd845g$Yr4+Pjj6v~ zsYoL2|9tGJQ}A4H8Yb*RH?qYhxm{%xsov-QC3MOX7#%(AfdZu@eqmGsU+VGZCN29$nO7$Db$70S$gM=jo%1G_#(X?hI)d1uyf zMn7ssI9iO+#Po$UxfbXXK^7W`MlCHSvmv?FY6$!2^52QCks(Ph*KFs zG4W8mDt!(1nnppnN}Ken;rE?ih$}h%g0*obC<*6{+Mkce(U18RYm!oqilv8f%HpU3 ziHV0}1KkS~c@!oKq#nz)b!5J=0)UIoN0;Fvc1?VrJVsQZ8xN*_8!aQI(Krs z+)3+5OaZC1N;8ynM|mGKBp{Gv7bbtWf-#>stJfZQU2n`WzCj@>5exU2V*x?cp>|dd z$cZ+DZ(s0k*y-<_E9>Jd->+k8zJ`C^qcdy$5`m9R9_n5M&m;<$L|koS(usw1I^e^E z5njiIEvCfCai7i70A%ZhhSx6tSdmszAL1;Dqj!pKE{m3=R=aa@PVQSeWycVN_jCXyD(al_#(HKBJ@s#~dUey=gnEwJ z&p|9c?m{u?{tQ@^mg3ZJN1efO|G?@*irp?ayelo<#k{G8#H^U-<376yP_@Q63Y-Ct z288Q|=xAcc{lb}LV?anbiBS6Y2@JLj(IAgx^}&Y@$92Z1MEdT@sVW)9R(Kw0?DBGZC#M_o+D!{ zn2B-VRUoqat+55v;;%P@LFR0BZ3;FXGlo9mFfR)xaU|`|7WfJ5gRa5>n~MHU80S@9 zrTiQ2s_x(cQea0T`9i70N^a~U$%6oErO01uWSL?@2!K0TY0Mv`<1ISVBf8AdqO-%R zrn(WRqAF8Re9F@es&sK5_Cuu+1}vUdzEPF+K?S+Klr@Euq%_`N{)C-htY-F-Xv)}K zKVpx_vBLiM`*yC*$Ot>4%Fb)AREx(TL!}UHllBopEDoINsva|AjV!;y*xg#!=A{n` zH=#AKzW^D)tgsfIURZ{yvAJ*fjRP6dF6>q=fP(BE8>j`t&)ZXsSvul!G5=}7tp8?9Yv7>JxMoJW*G|6YWB?sKf;fkob z&TEPZ+S5Dd`sWocQ9vE=ZhJ;13`#MAPbd2REZc@~elV3s$wCB=t*hHw=v0a4>q-0l zlA;FIsuOggvM8Wi9J3ZNwCFT4I=B`lB5~Qp-LsKkqkuP*DU%@k*07`ZBtA&i!&=P1 zr~t4`y~FnMTSl&c_G+u=@zEE!bYkwQ+X39@__0}|oE*YR&16{1uN?(GumnNB2F4}W ze@h5R)i^nmceC|l6IK-b^1gq4Cnm?d?+&M7zPAbCL+k7GxRTqOgI_gZ!z*ll%Z|r~ zmAoP`iNFTA=~3f-K2CiU;LCKU-ZPrcYyp74zLb5mT{(PVh6~_UTkChf4LpszyOYuV zD?InSH@CuiY*}CNAFtUlJ1A&-6Q{u2wqGFrlu= z^nYkZli#01v3{JN7{q?9CHQ+qcsY19lQP3!$aIe5Ol_3Qxq-^URQh37rQ{1-IMe@#l48Tt-nxKK>~AVq!dX<5CN0;P)x0^No=H zC0Zab(WJAs)F$CvN=R3oG(8P%vB9(WjY>cXdJZP^C45_uJ!TIR16 zkn z?u;7-Ul8szqCLpj#&-N!Rbsq4a;?8Pj}RJ}0Xb8`|LJkw)v+CkDh__IS?(A(qq2Kd zO(|of1D}yO_{dZIb-f3LxP5c^cfzS&$<;p=L+)Hn0{(c;n9cvq8nDSJ(geuh&Y4dR z);e&u^!@4xs7xj9P#B?*^yvhCn8oOFZoS$B*>n&<+Kx~d3&KjknBbvp?DAFcs$+5C zL@e6e+sehvF}TLpTqq6NsNpDlP+ao_$<0DYsmfg0>??IYeA>C^hqnJrCD z^Ls)IZGgZ~FWeK}Z7uLl%)9^Psg7#*TcM8s&V-`GrAhk7SlgChk@03~uC#fpfu_n^ zdD9E9s3(Z5;_@cIqpg@y*>0w*4USp=yJ8KDD=idXArIDjvT8p8ix@RvF;0T?UPxL1 zy(CM!1sk2;M}Isw=gbup3y=ql#k<(1$qrO9Rq1l$nR9=YqSG3B`mg1l`56)4W6&}Q z$~MO*WL2u$ziVYHd}yJK!T4#!!!(ukE3JCGkS=m7jUG)zPG?GlP8sUa`@6@SKWIj78*lt4`MCTMQJ)K|ei!)!FG9b~Rdp-=gO^AS; zVtr={G?_6JyKpvu^isc#(k$13KwA^eum4tT9Dkk=#Xo==+}|hfR{=(XDZGugO-m-Pd|c#hT@D zID9kAkktT#-7h>G5wr9TaNeN`$fT@>(9r&Wvd{%zAqonb33rr8mPUQ6)f1b|UXj)< zdgTNt0rX4c1D~a)H2nk~0(=%#NlQu3-!KN-5(ak02QV3=30Y?Q0h5d+npX{q-b9Xy zO5P4e8YGcPB7l;9NoQ368efuz?p&h8ie7lR<@CEkd^X0B$YQ;SthpYfOlUBnq7r5a z+W@^R!dH5)JtoTm$IJ2d?64zcfTrU249qJev$u&C4h~=Rud%SoANMvd_qN`TYbe1#zp_N4RO5<*LuS|m$>w=! zBQ2uVxdWhQGC+c=E*Y}EXS$4i9VVH8bU3)cb-+&^5OhDtmBwd1G#heu>StgtU?t4 zI{~)da07rOF57ySl{K*xv+4awCAj zK)EO`F10_9^H_Y(U&Sy`%_V$3;DGWIT<7|zhf*_Di0GOaOP`HV&SL~zCOlu65Q#S6 z4AH^4i`pVBE0DBOdMlyJ@Z$L)(AEur^34_gm~h|SXSZd|H3Ny5E%$uSfu^h>ijAYB z>%n`72TfPqgj1m(I&r39!*{*+v$O@%?RFL;p7J{$@kg^5;Yqr~M`$Y}wPvxjMoJ#< z&)aIhoK6zRQw40m;8#u7T}r@8H;8PIz_s;vqOy#qXxSFQicM@Zv5G?};=?ZB$|BjC z%Ew`k_k*i)ha-*D27f6Q=DP3Sm(v06vw`;z1~EfTc3+esps`JoN!lQ2g=t>PDU(3< zXWOpim^8_N^+(>-#}ll-8QvMAF?ODaGF9?AfVn;}&@Q8=fss z-mj1465=yC5vBWY`mZcL-Zj$2PH(U=Fpw;8Esvb#_W@9TwsE>T>*#qjkGBXZ{X-hf@`xVUtzoN~e`-c^rl zl?>7_CY1A-RAHf)3Oz@)gclb7&N-j&qEA=z z$U|Ss@;z`UdH;(1)s7d|te)aPsl4y&`{2+2)9Qp5X$F<~@A9Vpzfr8dNDFAMzDO%* zY36_RJ7IQPLACz-6sv!uSgYN(|3v$Lq64(oYPS<~)b+orovB!#*z1HqK;yam>RtcS z7fmG`0zu2I`gceK5N)H_2W)R|KYiov_xqcqg(#WC^+F))OEL`cHZ~V7(xhbfY+3vG zA8MfJkJsatHX2h(a8NCu_o4W5l1C9P`0J;@Cj|-~pmae)MuzT~Of9@^z^q}G3@vt; z|NHx9|3m0Fn8MVuSuiD2LeuLS0`1R}UTxA}WkzOLMNCC4t|e?MXeT79=#SvD!P@4c zVId4=|LE$8=7s^8Jbv#*i}=j8D40ZgQA-$Bw&Qt(AxRkJCG-q&h|~960&rhYCea-c zX3iP=fu;aK3~FDdBiDp5l0ymI5+e2Rjsy#&e~Xn;QFXWtl<@R<{q_bErC%r!>A)q9 zPFb@whqoId=xu(ygnNm(1e#r-LVz7$l)N2pg6(9BiiPsRU#XnF8u8Gp%M|p)v?#sC ztm$QCv+7NVk_GeS_X!6OsIZ3&EOyuY41AO(`Dqb)ypGaZu-zxKe~BnRsn8Eh?12nj zijflyP}1tM+-%=J(jgEIZFR;JbTeJ=HDQm12$jjns0@r20_hBW9I~`brQjG7!6RwL ziE>Dh#XfVd{@oIn)-Z`bZ{tu%&G7YgPmri2Bjw$!`7f2U9?Hm`^nW!jfOSoJs#j;H zgzQmU_t;*eGU|oK8pR5hR8;prSh=Vv;`56U${O(aJT8d=!BfwMSn>@@?t6^W$7^S< zzg8D01HcTBm3BJ+RlN|-xr5xIK?ai%ZQ{&1B@6Az#uPq|I22gh+xLUd`>E7MegKB; zLpeegi{|Z}qyuZEzrNo;*vrf5`@RTRl3y%3rFYi;i!g+-#>&@wy^#)9=tO`o%>Ma; z`hB!ezvHBXs=eroY?Z+TO zfK)~(h?U;OpIToqxW5=az6V7$FVwsa*37nC-mo3u9;lp-q2}V07+A5Z#|5r9zH49- zD-l_%JZ7xW8*l*T2rI2iPxCEvqyBIuua84Yg_pj{8vD_nG2pMrIj%?mmD4Id8r%DN zQGak1n-k)WHLH&={R#;Jc(I1X@&x9G68^%VZi^NYSMaCjzvp35t}ZgR;$OREiqUV zm!u2uv*=Hz<<%}V@3cwJqfe)pyNR((H;q6*5eWR|k)(L&Z|I<1KoK^k$OwOG3jyqq zPp+#xnk-fT<~6w_PQ2k`w;EZyN4T!&$MqI2i1HS2x_54}$v)yrW!wQ1aF;;$pYD#* zceyEe8q1PW<~sWLabPC@YT(9FW5weNqPhXXQ*SVIwDePl;8E_)lgMdc!7(yVtz$Uw zU%7sAW5zYrqozG>fR`Y>bPK4x36&qEfP~%bd8*nDw})qs%>YxV1Ss!*BJUMC$#=fM z>bxuK)sqmg?<2N({cr_k(5;$9cIy)gK~F*tmC z5jV6nCppO_9}}V{y*xFL%-a)Rz0a%wH>9WTX9$$AOq@qrz`P@$T9`WmdET4KYqCsX z`Z!`_mo-n0-)A)&MD-U8hQn*yor>~QBnYi&+12OqMXZsY?S1T+7k3$guno;l9`D7?vV>JiiwjFt)l2a&FGx; zALQ21&8EPKLxtI+c)9PQ@dEa4dQEPq8G(yf_2k55)lmg$0e2Jjw4S6X9%wpd%*F-E zZyd-qh4t1v)+=EhWSIB5h2xZwV}&-LFn^vpI};TRfaA^~X^397lLw4&1`#?c1ghMX z@(JKL+dH&E4NI45Shq+vVqq*Q`hIJ9z0ZE(iUWd8P3@lOH+psdmDerq^J~c<9iF;~ z9gGuu8R1&MY|_~q#oB8~5i3f}*h+*9s6*g+8cC-b4Bt)-gcYLmK+{@H3~@-7p|)~I zdh>)A5MO4v5ztgh>0W=)+Sgtr7aOP3Ghfn_H*>EwSEWjq?RkzA+**WrVhX(9Og_!P zP_z(&(KrsCodP8KObqEL!SU=5mgN*RxFoycl^?4UBn!>cI{?yMACz8nFC|ZzV>onu zE$IpVK7J^DJF6%iU@Bs|gpKs1+6NI~xSP)j*jdL-7;3CPPjQ)k92f95a_8FNs4tW# ziP$PC5D6^U*zz%GD6T7?W)k%$a?gp5{+q8uZUDwQi2cqu@0lMJ3y;VXdehH&~EZ;e+11O=s)CYt%3kq$;nea4f)giG@a zU|=7<)q!Zs6T7;eE$UKTfPBIX_M4k_ccV!rq4{JnNmLCTwY^6JBkWsFcR>4>TW!5R zCWn)0gO&U@-iS6 zTS;YRPNB2{Ju>jfQdi~Cl6Riq(iO|s?|MifQ%??Nbf2q!efyr~m8D%JT6SMMGzKeP zXhWOP+$}g?&^tilgu`dZ%GS84v%{`?tN{XSevKr^%UIN0!pbXIX_{vIo66pMqGQf= z_}m?9o8n#A!2O|C5==U$Crg0@HqS$Xh^$1$Y^^_ZEPhN-zbIXkYT&a#WyQ*ofPI!I zLZ#U*wR9drCKMfX74Kc?gfxQe#U*VNBml2}Ftt4kG6aOaFjpSI7c>pf`_)E!?n`g~ zOcZYq(Z_E2T#yX^)){-!ui~ZiFmq&xDL-Ujln=@@{Ay6TV#}3AiFLDC{p8WH_P&XG zt78{drR_Nkl~NGFz=BpmJL;r^%5Fk4JdH)JMiB}O@$MSecUTRW2}mg)?k-cUb{xrHsMCgUisCt|-VvF>!|1BNWfe`jRgGPaGIaar@u zT?*p7gr0f`LsI}jrW!MN_fXn}`AWjg$`&a>iW(|^dWyHu;1P>QiDW}9T)mq*KU|UA zj^?{r_ZC!-Exg9%v7rJ5UzZ!T{^!Pkd`$$3rNFp+n18g9ky-($l(bkJw%VXQhKU1r%!CsXoqx({v(Gv`L9)=9$_RrEI`n-ftO$A&79yeD9^I+_~d zxoJK9#$}%t^J@v82Wngves?N`^c7BTVc0t-tmFd&Fw$}vvi9pHTT~&wjXF7&zSJokK_W?@q&IO_Ul3WX8i)xJgRSYv0UdK zQ8DgBg;Z({onBI5v+?ocLUi>&2!h-Y9zYjFB(p$yis0{Wul2^5$N+24EINn~_a@a2 zra&Ve_{lWjxhUxj4hpZ^3~yIHw$mUin!e%F1h=D3Vso^MBdUWfN1dV=MotJKALrf1 zd=AjrIF;Cgx~Z=n;0N~ed`3zX)?f5KYL#O8K~-Y~T>t7r-Y|N{%7&FXh|*t7e(`A_ zd^Q~omx{p~Cqi!K6#$ov!p!D`HQJ}TQ!Z0fBTSeG+;1O?b;<9%&&+P$bm~pbHw16i(Q-ACv#9(RIH}YzV zJTK2NgGM#o(MXuGE5Bo@_tN^}*%Z*_UL$;{NEr6YyKAA5>8RRRgEzmF&fdJvW>DpsT4MF1Xb}I>y9b@?QWe{y&uu}J?X($I_>V}G5$*g{qi)w8j%q8Qp zud?Ffq!q`M2LMj5*4iG0zm$-Po1;p$Q4`Bw$&M2PMx-M#tZdxqPiS)0Cg(1=+fQ=> zt-*MDBP;OJds}l{gTH+X>1pe3f}({qCAJ2|5Z2~$=jbu}6ly5YuJ)&=FkJH%ydJMP zL!(8Y4kwU%A+rRz!Atj>HOHl31`>OqCCc%UtWJpY5&+Z(+ymQGi!RftA!CGngA9a0 z(P$ZQ4OExI1AO&!FEG58+4wC?3B$3qg2o-m(W4{=y<|`5>zemlO)f2j3n%($Gvb)3 zD6E}tF~(kqshmmdb*weL97t@ALuxy);~vYU1Do&PLz*^Br`{JRtskFqRl`;)$~3`c zzd2~MbpYiTblmVjgI^6P4$np{$A67$yZ=0ITFej1YL|9xj%O|4SPo}Bg!6y)Ie;>~ zLelo#*DnR5{dV;S4h6a-o_!I%g-IHABq6yy+#BT{0WE#fbaBafdLw+0+=I^l8VFa< ztI-Ol-X(GRbO$6cA*^(+4PvWm{v%mt=uBQ#`*8&6prfG4zV~ ziDJ~30 zJ;FMg3D9TM2kted0fJDDl+LtW1@5M%?-`Tc#*C$>`8qe5%-Gz>b}{kz&JErcU^kh; zLIA7FRxy1#7d4QTo}8KyIFhei7bEN#aaBadvc4QrnIESYz#ur&JBy93R+uDYT(Eq( zOZs0zxcCv`7_89$6zA#2_`2@+#i3~7{@LR)*}Pq+e{a3e;VM-C=m*{P66oA!Xz)~d z#C{xLI!vaWIseL&l51$S z3w3rh0!LlRZQPS$hcfi{+1!8j21EC#GPRQVD9+q5Yya2^A(X0rY7`2i%K~Ikv=5=D zQjdorSAxUoU{qrMtwXCAukUWzmz=kFbmrXsmlIZGZJz9Ji>eR!)cL=+dyMer-&M%z zjX}8G@?pibhJu&4W@ac>$B>He=Ye!~4TquhB<^GWzzuoP*{_sOeWKBjL-1wxnLEyD zprc2B+;@JUY11i>#JdI*zW_jzZeY?|-N*M2PX+YDg=ancdIbu&JGx1Mb*=^jHr9Z% zcWMw`y9x&>wZdYM#vE5Wf9cd+7UWKS=Z z?A|5tU0vQmYF<{mmgoXWtuhE84_55tL!rB&e@~fN{E*H%D+WB!z5w@l^|vzY4(2g% zWY!e{uI$~A$jBrrx_%M7@AXqOTVF^4%`pAZziL9`Mo7X7o11>1)MtJ!@g%;JiQ4Ti zbL&m$wwU=8gFlTX;0Bnlw0csswza|W4cb~iVWx4}_9^N~-|gqLAP|3M%*$wO2w}F= zWlv~k2baraz&eD}ooG-4^UQk@~vN+;^D&xf_ z_nCn%6`N9&R&dK;oRDoev^F5DOvs!4LNaT&9bIdyu1v&a(*+biXzrz1%C=G!n8Uq_ zmzTMVJB(Fy#LXJGD6JA!J!k0GbpFnGqnP@_x&7Jenc;$NxB3}{q@;jJBn^?(!=y=* zt^;2wQ>|^Hy?6+wV6tXc8wP=8`4s%xAw#GvxW#igcA78iy+B`?>!N4V)7E~WYsRT4 zZmGQuFs$Hkpa3YR(^MZzlfpJ;=+rIwTpq&)-kG*FacQbGm+Is6ua0ySPO$1D>|8E?9jCI|B?cb^-dqo(HR)n(k^$0H>3?jx!L8^P5>U8bGdMi~ z_LWvuG{yQ=5d$A@c(ovWtj;%V;4EP|h16k*csI2dzUg&{!27V)6+*|y(rJO}7%TWi z;Htzc4%adAW3m`DcD^3^>NB-COOD=&F2zR0LX=8b7nH$zgZn$#ikMB8?WuR~0ydx% zEE#N%Q-Csc^7lj3nUvIad&uQT+mpyQZ7`SvdG`-gyX8eUfc{$DWNJl)uRJuXdb1l+ zFIIFV@Y}E`F+$5uMOPesAE7fD<~%E}H@Jb$j~meLXHl|Ub>t5O^%my1tW)2Ujt*W{ zsgmcW;u1~j?Z&jCv8Uu?L5J%O0)?F;mnoIE4M2>evVw!=734R&bUFi+CM3A!tHLV_kYd0AXA@=Q`huKX)vT?VD`yh;c8PT-ED%9#-jCPJA`elz=($wIC zWdjbYHJ>V8Lbf~fR;67gi>NLr@mM&I4X?V?3~|fi6;2um#_b&Ai^fH zo`O4yV2E0C#RZa7Snq@yL#T*+1=S+zp9G*s1%Ai&3F;C1T`WJJou17*RxASf1Y9JL zDRxK~y1BNs#@SQbfqN@|R|WKn^gfqqD<1m%fzSHNlGZB;?bswxQuK3nr-{B3uU|Zq zeJ9gLtIP(~&x=cZz6pICyHMe!r`Pi;cy3Ar*Qj&;nqK z4=BGXu}H9NC7AB!FWEViB9OiF?7F?ORVoJj)zZD1Rdvw!VF44rB*?I=fujOIg`Axt zvw?q-4U$Zc$OC6k%A%c(Jwe$B!Ft|Wq*RvvTP|A#ig=O3iY}e+f>MqvE2jv8QPfH5 z5F&-UKuvN?ZZ({vF~j{vKLTFnqX*2#Qn2Igs(k&Z2r~xed$J7_34_+}&AuiDWQV*c zEPq(rYTjsHKHobRQ^WGl_aKG{4DW@5bJ_85+n+>>j4n-~fr#`^SG3RAUJ`*rqSVGocMxElm=Wd)BE93DO&uw;3b7IVFs9=_8x_u?Fv{-zeL=AgT}UHgLNZ4A(0a3cG^4b zAC(IJVlhct0b=G1t3KD@zy+u}x743aCh5^R;+CcvT|7O9zY`MG8jX;UsDAVOF3$LM zR7JwKt_spP!$7n6nq((E@7Lai{`tBjGHH@rS@9?-!W}WkidR3Vuiu+J{+FCRk#d+L zeAFR=k4=no0n9C|ar{6D;_QohR9t8Ca(=6UP!8EmK#|Eys;~B2xIUiH|^*m{G^GLHBbsFa1aq z=}^Ob^+84himOL#B4jSTJPX2~;obWBrDWi38rHrwPhIkZOy5A?RVq6zjQ4Ss3=r>P zQ3$u|L_gq*)!O5qoM6KpfsqsOY0b++L$Ql3*k$=FnlhG@j$^=koTrLO8C0vT?-x)k zpMOmy^x{D!T*d_x$yxbM4PDhU3kqezA)*ahp_oaoH#Z{m zIIlx>w>qf)QlSqN2X9)gg0M)YRU7>swsK3Ay+6K`mIc1k&rhpuo1<6nVWDc~NxnVI zQDc@RLBXD85II1W(=amfru>{5<&QE6KxzcW-(SWUbGUzPm_g0q7YOard4KTlC^lR0 zp^0SK@+V(wBmmW?h>~Tn2ZCPDf(PF0cg*WCgukK5YnQ1tMVGRVASuT=xOj0EYFMJ;j zOqwAwWE=MBCiz;LHPup@x5rC3e=A$4kOkR--WlcGbSR;Om5ygrEjRy3@=T(_TSr@( zERfv_T?e3W3nsoJJv2>@5oL0oJ&P~GFFjFX;(QnZS;*tKG`C1m?}%04h!Mz@z`9qC z>FN!I!HaBah4w^Ayf`#^>dmBwyYF2O%0vg$`J#}7Oursh<2p(}Bv2C&gbf$GC&sA# z+JsC(E;f4IjW=4Ga7Oo>{;ZPUfz@(}iU*fJX9=Lh43trgE7@JEBw)%Q?(XJ@*J6G1 zEWEXO=}S#}#$P5>&%oxz3WO!1z;-Fu;41f{xHHRs>m5lZnq z&;clZ<2BspHDUCR2&5n<|DgUEdn5*5fWM#7$^4{;s6t3X6lqK0?E=nj)S;O^Wi^M_ z)B7HmyXb!p@D~$0*1f3c_NFrlXEk94hbylUp0Qs`S3WHnP@zID6?BPiYZ`S`l_5Jv zdY@1{;pnqKUdA|O^~&v}2+MHEg&wNOM**y8myLMVe?>2Rhph91Orz_LD|mHG1nZ}1 z-4~uL5aQ>DC>Rlv`}XK`TQ@AwD(SEnpS4}mMkwU!zD$s!tEu{j)rhYhM=X7o%#Mw? zJ8(f3ZyB#ux?Hi;bN3h-U8Kmw)A2ich{RPTYj1_SjnA+7Y6_9-t|n=p9!JvnngcLf zYp@@zZ3&}PHwxCvNa%S;P(k>#*$Y+qFzhN6p7wFT<1k&$TtSQyE$vzxA|YCiF#dOLCNbpOzh^0Xevx)> z;O(WH$QSo2`)K91 z#^>VDzsKR*XVlc#_oJ7NlD|i8-CaCZ*Ud$z%uNo~JS&@-q6(Y;PypS|05;E?+;m%a z{_$v9Z+rne+g>ia4U2z>v$H4N?sc^u&77q+n}Bm(I0hbqWZQ~>wuey6Vmd}=C{L*#Q(hq z{oi7U)EqnzoZM~TDNtNERh9p7F8s$I@xK)jmmXdJBY}t=CG!RS7l8kS-+#jYKN0Ys z2n6jNB?|&2jr%VZgVdRjw=+5rkh>fZ5MS=xzY9?2_JGWRV*ejvhg`|lAW#jEdXijv z*g;TiK&}VV*m3y4LUp;!&F3rM`udTZzdYgw&`SNZ3Fq&A zWWtt--kKyCV#e?rQS?ad%3_OS0^zbTJ5vV~;ND!`wxEBTlD+r|WxQzPEMd)Uhz;R+ z_03YC#iBd|&aapvn%E!hsl#X5A`T@X4&J4I_Dzw^8DW&N(d37_;ci$I!T|t|PJU;40*LbG$qOQFA8TU)XYLfa&2GV8 zy-Qnf6A1BA{A0~$-v$>)PT78DF&g6e*(uwIdf3Id)A{CtEWSPL|MUp4|MUn9rfK_x zzDcn)`TY1vP)d`=eg>J!XVhv&l|fWXL`qfX#bN?~2n{7DiSekP>K8%PmMvuFnR^4u z&MLtWa~Il!Ky?9-!Cvj6a9_~;m{B<}mLFSzMr+($o?K#0pITWC$;O;?`cz>K)}D zE^2HM>RCV`u4C{f9UmEC-~wD3d8od3Nt1G^FZ|BzNvClZG6!9?S8vS~=`2^S94tIX znbo6~Pms@yis1=wRq2xsnX(b_Z%@eOSo^QvH7%d8eTt0WKXcX*xx{T0x(ohnGP+`!4(~+L55c3v61xBcqkkPw#jk?STNPgx`H!ViuP5Vh z(}yGEFLpquH~BPOP0bKu>g(N8u)EjY@%iKBzP5HRI{S**ekc1$DLV$OgD zZQX6{u-~=4Ueva9-wwoddg>#6ubBq!rCT-GV zR`-B&f3Y$B3X4dx&Dg&;6k}s-EQvg_hM4am?7+*Y2%3Re5VvV&`B(7u*fYPl1MnrR zX|TraYe2hUo8=NXnrV}7jfp}9p-@@y1lP)^GM{0hH*_K2=~`y7^rKX-lu#sdP&rfprL@fPP&WgkHZ?ru34qIbuR~S z{kEisM=k*-I%fQll+ID)i6?$lK;eYlR|)9^u~jJ|y} zlR|h5MWqK2LBx6U6q3~x{`sP#)q98LKM?~UsklGl^ztJOg1&+S4*Z{iazGzaA4rz; z&!+s6jP-Q_{4*D!6*5=NOMNqY;>(A$c9|FuefaA*G1Fr_%(Q0fH)6f1M%gh=RKCm$ zN?9-#vFT(O$%cuYwXG5NijVedOsI;<#;Khs4du!md(e=SO)z2Ssr^p4W^nIm3@ED! zRk4oLRn7lBZKjccs64F)YXcc_VD{sFm1!&|n5hO%&GBN$nWRUG#N@t+RSJa&mYyAh zkC0(;zUM2>qFea8Ht?aLU0a1Ze=_0E^W$P-u#_7rBL`&;6;qc!{u6jKG21ovG3$F< zLVt-dZngORCcW(RYE)#!=SF*ag4Gi8Cm0xd8IP%O=qo$GR(OyUWFamRl=64E-%w|t z^DqL17z>Asn)f#ZN>`y6w`wa2+-wScA!>yFX+YDSSg9@GlQJS@q`@m1OWUOHMBBwB z*jgpL{ixh~?@IMPKX^fAz`IJOBI(A|;V^&a8qaM5XE@%fKtd5+%~~uIVBZEvZMv{a~O#>HL&s#wp2(Q2WqpV#mTj7PD&<}~f&sn)}k^LxUO2uL* z^|6__H>gEZD4R8l`l#UN%E58S2el%(>Zo?dbc9;EGd!wSs$0XMoUO<|!MvnO@oBLD zkihY+OYW|&Pt8|(F>OljtQH0?CgRjoY?p(umPY`}^FSe<1CYEB4fyyQnau6`x7&TY zHT^G^Lg<6-f#z(n!NjBELHlig;f)hXmp}eC*B>XjC}4@>IusO??5WQgh`46Q$g_=1 zKKoxbZR9HGc2VTJlr_;>N_U@ML^jgv9M!-gP_ooI8$Ae~gtHk|4UIMvYJ?72Qij+^ zuw(;%_aGiHj(RvW=bSAtm0^OENojp{ zzLizTgjiJQ=PiQOAqkFrtR!{N$>n}`Yw5QQ9elCLWtASa)UH~G(oII0tkThX^bCmb|H@Xk z6FWERxE=zhcS7Q1AFl_Duy$V74~%ultt8DAD~GQMk(J2oT?*-0@LLXgw^PA~l(hoN z=nDP`m|@REL+wnMqLUgyy2*&3D8+Hs4-+KoFNrkToyKS1slk<{DyYlHC3WDzCqHI` zQkW!v3DYlh{JF$QU0HH<)5@VSDNw(Z1?qFFs)Wj=)9r5O>TWk@aoxgf)mz~lY^`){ zSKhku^Jf>`&qQ08N1HP3py~!F%Da5-m2`Am)0qBQRNLdzMLRGQoloF_@ddVp zujq!P;l@G0Sl6~T^wSa9G;Qox#vpHB3Vct=Lr&9=R##-?Q$=8+7r^#Mutw-HA4!^(J zm^(A14Q1H+HfdEO{^@jqzh3;Ddtp#6*Wv|hY&pJ~4pC*Q=mvN*te1_|^^$Sc-)9V% z79sHd+}3Sv#Vt0sM?9?%tBFIY^z&q-k}G;Am?G)ERnEk#EfQp8P+I`z)%+ygV=3|E zBE=jUXh2RNDFM)eR{m*KW4y@~Sk8LN&sT66PHa=YbTtwU`IOo;@4!J3GOQXRHHwA( z!y`&T+NEX>c9tZJ?nhni8b-r9jX9~Do%wFgzKNJ?aN3<_+^xfBTVkP{lc3jqZK!Lr zNa!24t{%u#=AYcG zhpd<}cq-}D%3?#Hc7ER<47MQZa)Ca~AwC@d^huHg$nw$iJ!tT53tzPK6O$YI5G9&& znKU@RGhsf>V{w87-+ar95T^d0%2>em{9SJbY(x z;)o=7?FZ}XqucgmcP1Beu2t=Gt&wPn7uPkG@2Ki@ULe{iiLy$;I~b;JV9fxFmL(rs zKz?0|zQOn=@m?CM;)Hr=1UYXM=*fHae%}I*sS^jtYgKNL7rSoWlA+EC5u)0-N4S-Z zMbFw)SsSoj!>Ibvj6ZjEMH}ERNlK!}QDk+D4jLhe(P8UO)5MQ^B}^cX0ilUZMYD(O zMu%2Y^<6P*>1p6}bV8A_c%&MU;$Z|t+rrFG(3mnYM1Mq34Mqu%g;s{ru7$0n}!$zRV)TB)HJTz2GnsaSnh zAWx+cj_!dl(Bv)Gy!QXpluWmzq`?Cpu5zDN9Ei_aH&-F9jCFiY2f!oxiQ##y@}1iJ zd!af{p!87ZFF%;9?0EVh`e@48V^|~yUk)97+p)&$ltpfhc?(BqPcy>Nh{rxhoOgg{ z^#Mw~cUX6{pOR`rPT4$W1&@yUdaS|6d9bf}`fsYxN-)SDSTN-qOy!$9&+D#SK6`ff z&s#$aiGMVr-+h*z#I~MG+q@sA5O#8Pb#i(;ynX1VYj!?gpYJz%w`hESgojHPl&!DU z)&D63ukX?ewLWJ6k+z*1R{NuZqCu(gD*)tD=l%q(c&8b4;m_rV;QiFr;pV~0$;QZ) z8#=beL}e@yFV`Tvk11(y!)}(Xf0OVZb6xP9h$c&2bjfQJhxe<@O4HgU^wM4R_6TOG zYe`=Htq0!0LZvZ1C`Uu|A^*`oBV3B%`94G~m&4(}c&v$IzlD;J3Ofa*Z2n$JeaH^D zXxm3goB{U|6ea7?}qJ$M9R$r8Y=>NJ)HEki`Sm5SzJuEt&x-P zdiPogo&;$F$x<588qItI8x+Iu`rmb{@@w+x^lMJ#M3E}vtuY5QrgM0Ih%Ow@7u$1{ zkK_T!IPp@*jrG(qKO{rsucPuIM@;H~TW2G>$OVo+MK~Lewlz{%lS%`X^0{yE;}t|o zu3)de@}HS9&YxPP^=ZbzX=e3OB308W2}e%JDf{uSA3NX5pGp%2s7Es3T8EnEtLZN6u6|J?O7hrE}dw3w1GxTy>o zdDrq8uFtZH7L5|M3PthS$We*N2y^v@047Ka&tkm{Sx8g?MH4Y$Fv=Z3fd|U)Smx)U zE|oS`7d1WsQ-RSR8A_xw)^c=i`CGN1W}qzzD0qbEe7K@6rwj^89nQDYC7Mhy@ zCKR9qj+r{@Na(MX3yC1=Bw? zgD#}06PlWc)(UZJ*AWSjj*W+s)%TZV)5f+NqDY3h+)yuqg90p3%~UY!(a7l25$HD4 zXJc;*+J>&S+kkpADtU`zF)tP}gZgrOa9l;_+}PO&?rjWZ-ZsZgXGQ!JTdbvaiBjZ? zj}EfczvByx@2|?H;@><&s~gw=Z%bAxpH0sB1JVLo2beYgcu*Bid;@h85sP3myD<@^ zgnNzid$t5zn&-$jBq172m=@%(0W~E`7fCDqiwo$+BlR3_MoV9J2BCN0KTNWP3U7wF zXpJEah>lMNOh8)NUUm%#iTFJC^`Fq$e|*3bq&4c@u?dczocOXP5gWM;ll>o3wJ}d> z7II6-C-AoHrJJB2zP<**U2x6gIcVHe8~lQ}X>tGKjx`nMfZ`To_Z%3ObJs}ZR=2{6 z%zOYwI1qO{uR|OPLk&j$6V@jQ^Rqn~b3EM(J03aQ2oD!xifj1?3_9(PY{T+l7}irh zaMHBDvZX>a+NN_)JFvp8<-}Fdj(|zR!9-x1i3ANT8J`E{jDVItSuy=jJ+7y!4 zC#L`N#fM~q4VDF-j*$T&!ZtpGBo%sj>)N*5e}$)Q<3#_0{r2C@wIvo296(maF%TQQ zydLNTmD)q2yqtAG7*L@TYI#N#Z6NO=e$2~4>~>e^+C_ru=ima=%DlQ$5O#OpH-*&k?IGCQGj2jG zwh|&49cl9#RbrsF*sM;GA#ii1a=}$o7?f?Nbz-6afQ|M=E1(Aoen}@$ZWtfWTC5d% zJ&8U4?NjL9r6~(|0a&Fz|KyA9)U#p*;Sw;cW+MxwzGru>v-1Df^c(0KJRpQ4rwUmI z9(4gzk~?q-b%ZE)C7~&riP+7Q4|gax%%R5dRT+tYom)fl=Y(kqJA@dm*% zp~3{bei=4Usgo(}QGv`?Sd7T^@||qTj1x93TI_}mS32rVv}Bu%>~1L2j-$%QR0jn8 z0|q5}-Y8v^^PQ9`{4M1*suSz|t^)s_pAG8`$A+%ek3bH)Js1VQ``X>AJYht79^T;G zKfQv#yXXXdo=8H2*MF;bd3uuzbP4kDp!9ZsK6P{PyuOe*zK#IEM|vTrK_Zn#y^Z)&X;n^a(^(;N)Uh(DxK9-ccH&l-^@Yb<14oV6@vJ6{^v`R2}{S=J!9ZLi<3_ z_`a#Pb&3)6velwRFI;jGgwSP;+1WvZ$ge@vY2twpLQ7i_NGjlXd`ga*?!PaDq9B7% z57T63T@s4_IUG=>6X_Kp3JH+mlTbaiZZ$20*2KoqetrhIxRivJKpY-m6{FIAb4VL#6by;Zy5IqYIB3JsXu1511lpIXb^@Lq^RiahpGsSFn zJgoq4Ukjr=yXsf$a3iBkr}-T-q;bw)(A=GL$SWYy@Jv|RHZL5nH#KPSFj&4aW@wV= z#OKFio9{`fy$F@ZL?ay&=>7d$hM6aXb6WUvqc)c;AxWLWo}|^{ch>?){F9jZ7TNS} z>IE8M94!5xPuQ##k_Ei<*Oo?6nG|U%ga%825?E|e9)AE&$oO;<_-RC z{Q>jxB%70~8f=->eu!N}zmpXb285=ta7S#1EIVD z{cXPZEfJO&a4ynAO>%i=T7j8KLwErQ*^q1qUyXV%*To1C)%2_t&X{U2vncrYO~I{} zDdg7wc(0I=mVUa}1wQl&pjfeNkW1O1Jgu2bO;q2>7n{u}pF83)i}?^JwRTk|5I#^%Wv?M5BByj?NUM)(Lscz{fA_^;* zrH~x|7XDNihIbP2x6_}hAQh)w?odWdV#JE-FVB__5G}OnImmt^U^4fwU8Di#j;r8~ z^8Ui9C0e_CM~z*v*KN!VN-n=Q@a)Sk4N7HEoX3F}7b$brKH-p-jXb!9;tk+s9-Px= z4*AiyPF&+wnyM`3{pY~!L~jmlvD^!mH^rl8Ny*T zmt&DK{6H!SPk5q*!7{rRnurOEMjAk_(oxJIcXmX^XMm|XaAzX&Sq}?iWTx5P6)5El z@`)gT(qg7a=oBR}bHQ@+kJT&e=tM6%6hR(x-is0JhoX)xqQ=$|Tqe$*HyGUwAW~)C z#gQ;2;K>W}VNE6kz5M-dhA3vi5yXqxBPJh%Kupec?x`r-XyQ^?KmUo^()$^{$c`tYnQHBNMR0OiO5G}x^uT|u zk)f`icQ7s(oP$Gi@d9?wLL2IqlJPa#vQOAI=y4%*KDD@Y8qf~`$0L)IdaQpZ+W$GY z>%c0rNy~sZA~+?IU`DQB&hMsX{*GSwTmF_1^^idLQfw)ya1&h}lNs>fgkX4P`ilTXXnr&@O|3cvQ{!u6kp58#ezT*$%EPGrV``K|$7Tx;wkHjf^n0)g zw`gLLVsoq7#On>OL_>SvDdf?9cSB;A?Xb*0oqCPX_*p3B?C9Dy{gj1dtST$n0z(T9 zq8~n!+N~)ReV%Wc1M28nHm@Z|X!p+lBZ3H(awpZ11!N6a@Cy?~0TB6npL#ziTAc^6T->Yzhfk`&+N%ex)fW-V`+NU705%2M zK_vIEA%?!WLOb2KmMBrYw`}>9$)kYV5)MJtD7H~-X$J!5>Q;>A4}RSr@3XXelawW)motlc{%U0V znQZ>AWKor1FrvIlwum7od>b$H$!08dvR$!8tFtiRSJSM2wAGHi^+mXi2ygUHET=Oy z9az_rIcgBbc(MGT(P;|nl~4pr!M&ivX;`lC&$Pd#8cS0<7+|pcUJzF1ivaJ z<_Md9hl9hfpaffSyrs}7s(%+(0q{~5>=%eCtq4Nol(I};AT`>JoECd;8|I>@WfiWO z8jn@Li}AU>N4K(HocGT0DDN*aZ=m7~IieOu_VDq2Fl@Lff3gs7;MlpvRkLY}P1r6L z{#FcMwD)3#e!tb81frGyv)QzvdMRIccpW7MwAF#_7LUc^!W6h2Vw(_0yZGy=xgChu zA@c=x>d4;O2#~LATvM64V@sXZzHC6~-?-QU^NqrP=($0(5eeGHbWzy|E%BcXv0pCo zBSE;37Q|`_Ie7o7fsoh9g0tm@c)=tq!7R)P7Y%wAjRw9;_0cQ~`D4KQryEBTsouNzKD=FBZW4~GjnN+j zpy-t(!CpzW4J&@h=O8JP`5gpmWZU3prngcr6J>eVDzzHx-Khic@HG=+Zef3+abFUA zJ|mV7XEB@~P?V}`eEG(iRCTAR*hL)_fu#9wtKg+NEciY*bhI&m4@h>kxRI?Q~S%X5!NC8bvqkU9vqnXrEs4$kt6 z!?IT~zn0x>)=Xe3T9siifCuqz0uZqGA#u&c-s?RYS@o>ALCv$)b{v|w85!sS3{s9f zLm37S^U|6qKr4PFwylT9UL|O?=+QaJi?LqbiT*{^$&0r+G2aa!<}bRQ-ljsrjW5}` z(xl0&Enc)I5{&CX38a2=;v$+93v|AhWDXM4WHiK1cYxWJhd6hDsYztf-0%nI^K;;L z(a!^q#j`ZK7{G%SU4|UC33h`AwqlO0VD{KvYLx@5d3J}Oz;JG~gh^lk0*6C%t2)U< zP$|4{1(p~XE0Vam{xs>n<9|8TuADI|>=lUC_U)~s)>w5SeaDXJ)FC>}i-V^gAesr< z$66NJ6}#v-hzoJOW!*4$f{W@rNqq-1Do(n-x5`VIPOL%AhXbcxA)J4&vj7y zkec-r>z%91(qgkO2-qQiy6F&ysu6y~7|n4X+?(L5CnO08Ll&WZEKykVUj*;4DQ>VM zB+~Ib_YiUVjY`(HQW5&ug!;pVa^fHNONJG?)DjvSnz3k%A`3t*)*?X;@*ro8dMuD9 z9vwUKL#1%CDZYtTBh7gN7q(k$|79D@AAW<**F5L)bs$RL|NbEgMX|nhM45fa*>mj(T!Er;cKiI|REtT-N|6yQ`6F^mZNQq>s7uGiwO}w(XQq>7Re{L>6Fe zF#!k$h9*Yki%8gJQbdsm3iOG*VWypUg~K<{i$l~`G+&czZIf|P(y8Wq ziy}Nk;i(}$IW6Hl*RWN{{vx_y``YWmg|D!ev2UDvgMZ~hH zT`JN<{qt(@F_e;&QA5eDsgKk|)q~RRHU0Z9I)$h%%i>1B$UM+KcNr{7SpznvZDM&; z!m-RLc7bJ>5OEzB56t%-iG{9L>MZaB%LAE!csw}W(<_r^+;Vz}AxRCaxVEQ3MwiJAyjKnQHh2u$@xMH4ra4D;$j6;>sKM&ajEYXQ?pqRFf|yB=Ufdy< zO+2ch|H4~v&s`BD99t1lt{FnRIEWJ5dgc$JPJHws6hTkfqFm9&j@HXKHvnTs3L<}< zXe9=Gc~QKf70n?>GIC^Cb+nLV{*X6b7I(fzLyyA%N}uGUxfDWShSro}7*|F7xwauw zKI4>5e;ER09AZeeiJw3Jip7g=7jNioU&qMSL#yj2+~Gi~X^NZfh$jroC1I)hxPrBx z9{n;u?#fAHDv1GP+pPYpU4X=%9!wL?AYt=^d|p4ZBO{h)U!0hS!c*YGHJy=DIo(x+ z8o|E>2BKyyvu$A(PM@eSp{#YD#|*JD2SOKaZ-w?Tth2T%fhSRU)WAG`OCXQqVy6wd zug1g>>KtMLh;f!8L@6B%S|^j#MeMM{)v@}Fg8Zx}j|Y=^zxTpa zGtmC}TW^SrY730=s$>Q!CnmW)vdpEO3i9)TK1|vr$oWJ5A=+&&!Ew zOzYu_HDnOW@fTL2q1@`-o%ydZ1sqSAHL6?-nYTLSq@D_(lN}|AbhDWOtJ764nN=oC z(Us&a&0ISTn76>qLTyQ|)+@DkO(Zp*L+O#og}9$n6c(}0$^GW*s)!rEL(W?Pl}-Ub zSK3(Mjk=_tr;{c{$VzZVR*^#QITj~sUlhFmaEJA8h99Jz$TDOz%EJ+o851U!Y1WSNXp_r-Y&7%&EWYaL>t$v9Mh^Ajic?|NU=L7 zeyEfU$}fUyKygkEof725y0RPV`xSBnPy4BC<0ON~v-aUv{k7cos*!%L;}Am&WcEKX z%l!ojyJ9v&75}Bow~;_Qg#re z?74o2Z?dw8n`XJA(pdz1y={N_dKzy})(wgl1zf!Z<^j?@U~Lk^;*K^AYT8sA=-3f` zjW!*9?iQ2sdni?}ViFRX;fHow5i&%uWeZaF#kD|%ECM}CI~-~oDvX~sJgbX`7eO;( z3)HVBa7(~HbFD`*brSoRSu>2!so0yP^bdh322os+4 z%-zZ86&|qkWr0oXm&=eNsj2))ElD1dwxP>i(bxLr;NGk2{usnApL=yr1+;s$`NPw; zdyP<`CoiHjSmbecBmB9!t!bY0R?@w!x087n6@l6`82i_jCIXsvi!a!v^N!X#yI&0t zK*QeR(TG#oJq^!Qy*$0Qdm3$3URYG{6C`(KA%Q{sB9)?`+RpJk60P4uAa|9@v6W;yE#fGQCBC} zY`c4Lxb5DV;|Ws2b01)*V0{UA9fIT{gU+>siD1rxjCYv_2(oX3qz1O!K-L) z+l7@a1*%=nZi%Nw_W%tN%eC26w}BJkBv&#i2TDC+H^XSXvrCDY3J4(TEr52@LL3=> zIfBw-j&yE(Zcqcq{&DWm;G5J0%cZ~s)5%rohCkeMrIoCIaZ7-iT?=Aedt(nEzTi)N z+TRGNQm1%aPNf|lOzGT+71C+rI|x^p?4-6`v1pe3sUZ(pumET6rL6V!x-2&cNLWb6 z6!%>pjS*ZQCSw(I0BY4ctt@1|L&F!it0r7huA?NGTiPmAo!QfS3W}|-nVetRo^R>9 zJYojAcWThoe8{k)Z7p!bQvdz-g)}&C(M(&a7bkP|fG!pqdcj;3%nE>5ii3|S!9O6M zNtV7jaMgw!sS!e|aFz?LJn7xxBnT_)p+fgXBbY~JQ!Whq0D?O%b)v+AlzdD_uzh)x zQA1+W_=GOFUo**r$+}?Cj!_8S_akd`_eR}F5c-5n!oK!A!?PdoJAO{?2H_2dlxctB zI$ncdZlfF*S^J7tb+h>=Sjf7Uod&v~d**L0qH+Qbv{#!F?SMo1M1t*_04 zeF#0^J(%cd7?`LOG*(6gh)EUobaqs9RG!nKlqL+70yEK9mha0=C z85_#IoTz)*PTiq`_~E}To8Jk_%*RpO{@_-)u-!s-P7VefNwQ+pOMiwumZ5j`)7B#3 zhLHwNTvHexh!Au02Ga**%bk>EnR>q8AUKElTQJ%gm7kokD6LCWKB0>$Z|>Q6kZfSK zlwwf*1}eF67uaM&%8Dn9*VGs!T=LFOIhFQrh<|Gy>;3kdLmwO=B3S1?J-|yA* zEcEjrWodIc<8MyIL0em>O`(@z>N0E}*a^5@0V}otL@}Sc68U5?)KbnQb214d$pZ32 zilDrINDD{%5vALUV_;Fs@YW0zD}lH6C{CN8NCk{BLLZ z_Z{Pe4-}B$GqHhr6tC+bXU1f>qFO{=!Z&< zKu(K1U{UfLLOQ2c=2AIg>`hvuXj@~$uxC*-XGQ`@5VEm@wgv^LEVEuhRPCtX?#okpoOq!qz#Qg zs^G}H0D6qn5mH0K@%}?LhrcEY8FeN~z~*k{Ht2l|46X#OcDc7oNF$+LT*cBZCUV*O zhnN|Os|Z84tzD^8y}ri16y-mE@_cCAD#8V}Y)IYjq2DHSP5Tk=wK$^C*Ngf=4t-ao zasruCg6}7a#UFh8X}mc?KA<~7O0{a@9ae2@s9rB1)K3MCs|xSoxAMnJ-{ z(iv=PhO3LFD?8Pgjz~k3IRJy>Inoq8z<0p-C7Y!yl|{C7C`Kw8vsCyhH8SY`Jl>!SO~oI^k3mYw;Pkpfn|TwG|-JD{lvL zte`)CB@S)EZ_%B z5maG$mmvoZOMjW%OcDZ&vuh{T)RMCg>okB1gbz<-55)a(tE2Wf1~1$LjVI`gsYO%27+N>u;V=FTS40_2JMNzMyt<```TEQhaE3aNm>& zIumQ;Fl?xI%Xn%80kw&%cuxPkJQ^9S$;V}4f{st zT23%W!p%)Sq#yHJva;&85M(Z3fl7 zZs)b_Uu06D`K#DUvzj&jl#n8xFFtHVZe;{x`hE2bD;u@E6S@#%Pe8TS#T&2BKc;7{ zl@k|RPLD_hI}7`2vlTrxsttGC*&_)S2Icjf*6ZcQoLyO(a#yAvxuq)_k6@1Xjs<@z z0qy7hggToc1!PKMoR1kguP(S|fYM7vpCwvSBqhzinlHv-NVZ2mQH|8v9{?L_?R^*! zOZ44XWXV6^@J?}(6@bFK51zVJXl)aur|i-9dEO6%vT2?>&Gyb~=rl`LE}i^Mjki=j zyLk*>M{#z|l$+;0znk{u7t*P2o|R5C{T~uXseMpyr?xQcofd2t?p-6`xl==eFdL(8 zG&n83^9kB2eRl0DoWZArY|r{Uo6R0kDv)UlK2pW4-cnM!{)W%HFyhToq# zdBNg$q-dtuVhW`f{P?TMtkJ)(-t(~Tl3^zvoamlzNx5#NiNAAhUk)5J8@C zkkckymw&(V&MB0G7v%a{2FqH=p8WkwA53(Z&vx&`HVf#J(q#ntQr!x$P|WJJpOjUb z=Dw!&Y|mbA<9UbNiM&;O7B}ngidr1 z$tE`!)YFgz6WhzC;M2x#=%3N9PWL559LfX5N&H;&}e?%Q#s98J`kG&o-{3XtV z_sp}_)d3X!xp{s;r|~n?cfzD;KdA*IQ9j#}R|!TCk7!0aE;h`D3$vsk06b68O3`+d z%3bF@HUYv`^-sh4D<&kIOgWc~BMOg_I=6mXU3bVTp_3IF2xmkd90Rj#=uma?LbP;b zYk3=mV-s!?_?yUh^E{XE{#B2t+T6gnC!U`Q%H$uEofYzh3V9 zyCSqHzNyT+dD_e#QmP@YWB zi3g`D{JUTTAut5m&HfzD*!x-0pf1_$Sy`Me^D)6Kpaqi{2lFn(1=}b^cy4W2OyYc1 zL3v)_+sJ4v-u^yzAg#;r5QpcZz;EH==)k1izAb*d5A8sYhrP{Mnc#MMlW6FN)d5ud zGaQ!ODeP{TjP5+}vxM&PX{KMT^hCxPaQ#Dd14zb*Czbgf(v)XUdF4(Mwa#FeeyL>h zFepS+g{xIT4ppjt@n;TIN;n;!LmMPVyxoy2ypS{ptscic!b-%6zPA;>b) z$v1pb>E)84UklNdlKcQh6!WfpIv0o~ICw_#iPofT0F}Kp*^AVBjXa?G$#QH!Zf@A2 z&3F)`e)0PwO}`PXQM>U&^$+m9;M)g9&Vc+*iShg!AJI=f_c_U&Gw9XA;g5G;%Qpo= zP~0f?;9~2bdujjp(Z_7z-dh&3FpgL2w`n~H`^tw(<C$q2n z_?3TF>m(4h(T2f+?G=cjX?)NI${T$BvwB!?=?G`YZZh3rb3^0uA4Sp@VLwSkpD<1 zc<<$X7&w#pf5(~Fxh4kyq#z&$=^!9K|8Ja04`Lpi`u~eF^@eGOgV*B#H~e#O+_IZs zzycH1S&Sl8J#pLuKt6Y<&a5Am-pO-*AuDS~CX#VWtK>WU(%Sr0lSAOgS{Mu{{&&M* z^rhi4nt)YRc4HlzZM2k^lfD}NJa0osyVE~`F5Sr!85NXFfAI%j0q$Er zN?XYSt(v0NL$VR25xG4wS7Uh(FQ$XzTYQTijtF#+g^|QA|e&GFl%MF zNV9}|$an@>&SV$LOct_vg_}?3K0(biM(CvK+>bd2_yU&&w+itxu*d)ex0)8*pCGgz2EHnMseThoUH_XBj{D~=aMHeSpbwBB_R4c-hi@!I zn{XZcxSu8x@G{UVJ?fb11_s0#wp27t=?!C*q>VE6H(=T}B!8hqB3dZhhb zIQi z9W!|tm{YR$(rP_K90J*D?I249VL=OHncvx8o&mO>c_W3>mq0F|7CDgw*AN-&ha$M$ z##+Mz!7aE5{MMkn>iG5XlxYyBbF+w4hxbFK8`VJvAm;Ddb6kD){NlpGw={-iy+yq7 zWt{Oyd{vc;1t!w-U*A-(dE?pXW-|XWhNni^$hGAuFfB`jB2_n(N~O87UrpL&o<*=viMHu+#5an&@CY>$y>S(dK!Dl_X^Ev zb7$EY+2DoElm}(9H%}I~xV0H8D}YxrPQapJpxbeBFwvd3y^6hO$dkXS#D<8%W+sL! z0yp1}IIEYG6}Kx;U-thz*?!x69x9gT;!hTnsm7{_hO&M`S>6Jec*feUgHE9{xSijI zf24qroZ`$E%WXYl_>KHch~!4mE_Fz^9xmK0({b(DXEXJ6icy!JmU$nq>^cdCIg?wb zj^rgR(B`^zSRW|+sXvL00|A?OI1&-X3#9#S&-^;Ho9jxil^NnU8d}6(Xe!?z{CXJ8 z8=(uZfF87#F;y%VyIYmMWgV&%+swn#B=HFqNZl)y^kNrtsZdYPq~kar6Q)&}0UIs+ zQyGfMpC2vCewH`~vb~qzN|J1{iath%1)9;QzRAm-D`q2nQorgst*0AM)^!byNUwvG zW;fAH@BNW2-%p+@QT3}eNqj8jlq==B#EzOraG?gbk@B1^Rz)$3=^=K++L|JUtVD+L zq4R=ybY#+(@M;IyxN`9$9z@K51X@!da|A8jm=@%H_R+uFZE}Mw(m?<5M=$S$J-mtq zqnqfr)c)m@5$#&Lv=k;YzQ)rx*ukaZdiY{`jkBTIV|MEMc!CD;J7;bvPPq!1@To{T zSn`+KaVfv+__lPe?tMuluNRJBIu8TFVMPErqxz-NvTHcoM)ebZfLw=b1JF!-MIlJv zishM!5I~j0RGP>cf(X@Li(!l~Ar}i)fQ)&ZY-hLZ9UfO$zf%$`Qd#Acxyx@|o;Nf9 zbm}avtCMmo!!Vr52FrO7+fyp>Csff0vfNN*R)PHf`5;N65ca!&h7YeMf?hXhN;dGZ zM9X7t5!s7!l`QE}20c$17IkZx#IR zcEI5e{tz)^muM)dcf}jGU-0bYoBGWCx)qyw86@!(gN)D2oC%ACZ_1Sx1wxo|$WNKx z9>N!+i~uZfoUKXY=aVeuJHAPK4#I3PJA-K9h&7D}Ii3ocG_dK^8lag~j-nc)Ox6aW zIEv|5N^X|7vFh@~X`t&iDM(p45&c2n`H6V;y&h$vMT5&IUy@7)dM;i;bY8>EPAW+Q zX@qO<@B8FnYOw?j68k05m4n?5J8OoY**dW%JD#aSpSqo8o){*|Sdywl3T0rl=oj&M zXr;~-nRK)%{E4wkEm5q=7m`s5DF^3aEI``8F3{ zuP1Pb`qA+z>NTZ!f)l+D8Rbl1;aDn z`PkhWrXCLdK`D&zvT88tNw${+Lwt!pu7oIO~&ho*tOA)=@bakrN-#!>i zo0rxd_eOhEx@4a5kl-+U;cK}a>U%LiM<(r*(@()zyA>Y|yFXkhOy6B#D!QpeZ!JbQfF78ISiI5=i$LW zsiL_koIYadH$~|4g89YG_Zpf(m4t)mJ4k7X1P`1-YSbxF&gb=ZxV?#!ClYL_g{n`- z3M!6{d}IEg=>!g+w1Tsr)5NMuR4N-g|6;avt$h>L~zZ>Ah=VTN@qyz~&&Yqzc#-;R#yED8=qGxrVkl`00eGIcaacd7`?7JVv8{E{iK_^NTS zu(n5ly~)?isO|&_LGBllV4s5mkJKFPEAisrPR)mdUmn>0DIgSz2*+q_Vgsqp2?+5b zUBn&4&qNaEkiNJ6Ij2Z)y`Aqj`*IBbK2Z*hI1o9KjSv5cBeNS_JOHj}$|bAti6dIL zvJrv~i*9Hq*kEI=2S9eZ9YqJ(u9{5;+540{2kr#{BpLv5+aA@SK-U43{{v&tq2?*U zP`3HEO5=`Jqgk`;Qa5PWW-%}_OI9-(Z0$f}#;zycb)xMS(85JeT~}ADaGSP<{cx`==2w9|7&wPbbg;3By5aFx1t?CkFj$8F3pq6_|8b;xp-;{gw{$Pp&jpR%dMBi~ibJ@diAV^y^HxG0=; zJ-MVWgYjm8Or8HF0 ztgY`>v+6L0RBMI8)OU*BkFTqRM==N|1|ma`jt#A$Z)RtP?9fFwK#zxVe=QDKIE$y* zp3ebV#c#==Lv)-Zc4iDQBTNz+7w2XVY#Aa!US0OZ0g}Lu5bdTF;oO9*V8{S%0t&B! zX>*Djvi7uHqLUicVfI~p^3K76R}VYS<4f#K{eD=u3vx<3H|NqM+X?g#&hMAFhX}MX zYTI(}me4xjCp_qq4@Dtm(cZE&?0HOSM4GYIi+<6{h#m(>~J#N zhb%hqAS`5DNvx|v+{swXiSUeo`=vUzMeUBjEsJ^uZ930wsxj zHf5`6{rJu3!5;PsHCV8(ltedIS zl{LX_6Xs=rl$td#m&sSiHi1d@k0#4@lBuw{kZ9K^X*SN5opYMZ95QU~F!Ht;^nBwR z;xQgJ{OalJL{~O}>wy4@N3Obfzu;WM|IQT-qwv{Vx06*>+ij`nAbO*l21^-E2LtV`F8XpXv-4J{!Cf5qr;eVbv;x zsFSvNu}_w2E8A?BF4WG;;|n2QyvFW-=94I?I$gfSYz1)OoSCnA`#oN(CX^9XhG3m* zh{*6t=k3V(J??ByyeMWb(<^$O^bxWIS=I=dkUPdC>M8w5sugca;L0bVD*uFc@ zzu?xXm+gLvlfGq8>2yI|bu2I#AG;B;EUu?QN=*C(IHLy&P|3>rN@qlmf!^Kq+66t9 zxHaise0JE%7G?Q+;u1ngLhJKBrX}PCgjiK)7LMWerdWBLx)|8HI3?*L?v~kyVZD3= z*ts}@0>2=-^c1V64&TNuqtBLnj4)<5WNw;lt8gLJI87RLGZ!>Rogw9*p3ONHW&_A? zvI0HV#>IdG2gV_IM5h=;R(DZxijT`1xr&uRySuw6@Xn9_wgo%jf9Hvyv-s8UcJjF{ zgb&3OdbmfRzjb>@J%3mb%#6!EPJJGTL4o#=c~A+yg^G#9BJBAa+dW(#gSmS9E*Ck? zBGaS?CgUodu?~LG`{mgCQjINY{(?ht%Cx6rHe3hdAZEo?92e&3oK=sn+KQiFwQH^| zJO7E-9w+DB)qJ`%*cGm74r_?@jcp5NM90N_$d4IRv1w_}|D2x+J@7P2(utoBt zasdEkO`>f1jne#-X2W~jW740u$~D;wo|X~I_P2OfblmQU>ZX%OdP-{1RiJ-OWjLM3 zUTzEQ_?xt{UCozGKv%4OKFItg6Ac*gJ2c!`O2KxE@D`Bok!;OjxZ9dX-|&ha1<0#; zmvzA|@~vh7hlv~+&vi<-ie*P^Au<@yWwFy3&ybu`le+S+Veksy zR8cN9)!8pyGuq7~FH>6aI&j1gmqjWc_NSBR|5za7cJ+| ziF|Fwp{Agz;3oklw3!_0_CxdmWlz9@X$&17HzRO;Wp7R8fQAdZwpuBwK)oiUO3Mw` zkIH)ZD{_-mTwUYzgJ@t?86jQ4bd;NDlojZ46^Pcw=K z@ZdRCH3zgS{@jQ9FrSKfCncKo^1^rMmev=U3nsovTV$vZ=iDhVH8%TXhAzJCjfabm z5lCc0ttD)^-He!bnh3`~;e=z^(cluz-yS-UfbYi!T*JSQ4XDPu@+P76M={|V0h53r z2?Ibp8;r3$KX(s5H3eV&Xa{he>}o9vpOSX5Z9%*r4u!_h482aL*T`@Mb1~IT;v}UB z{REIURJH*q%%57cI9QLgLX5Y!08HneiUkgsr%y)X`H$8$h01pNlq-h54+)?f6+^d1 z?_&sYOS{XI>i{lvUBG%M+cU4&zpuCy7A%^V$Q*wTWVYG`N0wEthV>+B_8kGG)`D?9AzJ`P0iuI(rR)dhFk{y zTw_Sjts&=l?J_$jk-oTJU?Iz8_56jSDDQTqe!u9Hk;=(_4P~dyjTSL&MQ`?X$MV}X zt73yIdE!`=9qEQTqq7E9u;C>J6?*0PBR)!izs|=#&0i%%aRXBqMTSaOe}Qj$i@^v& znGCN1y0~GRC@*q*&j>Eynk%}JTU3P@SU1K161;Qz5xAUl`pqJ;Hif+T359#aFFxqp z4ljaW{D^fUH*b`F?0ucgU84&)sLhw6J%y9eY2 z8|Yr?{&zgXG4wA)*1t?Celx2htE46m%6U*DrOP0bBESd_?}XLSk4qqXTBwn46}_0$ z5L&D4cw;OcWma~^0L?_pVdlZ4Zggd4u>Mtxj|onG?dU?kKjg}X$S(sGF8_y}vaNRa z{t@*gZ<~l6ZEW2%{nL9$JrCjD0|DJe!a+*1-lD`Q=deOIQ4y;CZ@_rzBHXw860OeS z#pU%yG|;D8ep6Pw%nW(fg@=iHEc~c>W-9;-xjB zUDVcI8N&Gfa5>pn0`q`Z*iFMgx9fQt-7WgF?!iaNxQXyIfP|U14vw)rBco-^@|7eR`we)RXu(RmP8|j z)lekwpY}V(E;~FRer6C*o@@CK_IZ6-tuxJms_v-LiuJ|q9GKcqm9qP;Nhw=*R=uGe z>wr!i=(+Ok!jDz**dwf@F<~=cWt=3~s=h0>=mW~}droXq&~V|Ht@Taq>U$gzyRE&3 zy86QL5;D>j{wt&;^FGpxcWwF`=v7}>jmPwnx-LqV5lbyWvM~sMBbVK_l+8CpYv;r_ zf^#5WR%Fe7)278;kp8U7zZkgF>5=sc>(}ia#k9~%?;l3!ZptR^=aRbvP3j8 zS=)*CH`yH@`Zsf_n*t$l*#l98i;@fL!Z>AWVr<97*0BRpPL)FEoyxKyRI_r}m;(pr zko8yBvM}FymvHmypGJ})EIN^cTDPU?Xbp-HR23FQijzea!wscfQ1g>1RqR(gMmQdb zak{s3y!_7!U?jjl@Ga+3>svw0iiR->zh;lP>hT#^m=D#o#rMER3f(H- zTj#J|X>*P=@G*Or&uf7BZQHhO+qUgD zZQHiZO1silrEOLkx4!N>&h74V`uC0%d;i%HV?1lkIXCoAg3y-CQ(H@`_*L*@R##K> z0i!2qsE6^6w`36jcY~BZ8r@UMAa7*j8p`!DlQn|&N!;a}A2A6d`I$%|>2p(tZLHv^ zLXm9}5kYWb4$U5)kp(n~yxJCJ8kdoVY}OAO1J} z2FP-Qm0ie3V~gILzFlBfQoZs)#^-5)_5f)?0lCrN%F(faxxZA?7M&WuOKfTOpvvx6 zP(C5;9?AC~4zgWvRf)#k(Tsj7kJ9g^e}R&$TCJI zPsC+dL=IszP#AE0et{L-)L^1siqGMsNiiWclM=2+@R?U3^e+~iKf6s#wy`&u3NU?9 zxEvz79*r_5;)9gINxyN1;o3@n9DTbw+;t2jIG7XyG{m@VijPAF7JnCWCftXU2q0kL z;Bao*q5TnqAyZXmtO}6#AX7#6Wipg9>ST0n`jakgpqH}7s(AN|e8MCs25LhqajjB! z^my#m5-sdbD^6YV2S#(s%YDsHm`w>8{ilosIwB!84SG{_)nuids=u3ru?t$OCYu)T zl-)iLFdrCRL5{OzW=0D$!b4bv$5=pPxZXTXN|0`ykSu9RD#m;Abd!J|rJtPi>l8|& z>3Y6OB48P=d;F|C~s4T)SQidF{#xz#$TDZ37)?Z6P}4l{xaqvzfzlpvx@W zM55N>#pp|W=TaS{;&}I088`5Fry1p(=3skn;v(#<(fNA^Lk}w|cqGohTObYNk6!6Tmy;x*c zfYbyLXPqeirQ?!!Y8TD-U~Sgo=4}q!!9QazI3O)Uui_(Dz7AfAeqpSjN zca4Q$>r`k|G(GmULAD;^G|E|Wj^(i}paC?x+GvtcqPqmkzWX+~hrK>Z;l^c$Vdths zV9>9n`D6TmB3)LzUsjL0dZx)MCXZt59p@Yt_Ky|U!uO@asp_rP_1_-@y#_khF563> z%S!!+9#_MP5NoPln=M*$}$Pm8>UDoy#O#qGkuyFI*Tt~)Mq|8(QJQ(Q9w_9Y1?hHECKx`OCS zwwZ9F`1gOecARW9UJ2v+@^%%9JaxSrH%j`MF?#hX*gLY;RQbEwZ%X|08kT1aA+L?P z1IiKFUS&9+oF3rR!s0^|auQ2lz@S%0(iv)O&UqpDI8S$;1yiR5sq;vF?PYDI)hLlJ zgpc;lmq~~voDZc)t27fIf%mdEUtV(I^66vcrgh}9ijV=h$&)Xr?1ThEkkw(`SR-^G zv;yltoF|^`4%SeO>bxD*XPu>ozF_|+BfXZ{f3{&qbF2Zxe{;8wM({=w@aIjAus48d z5k#X$4+7fTcSKj1{z!Zsg*b&$;;iSPyN_QBC6071@>oX++R1o<+0cQ$?~4L$LoqvC zR{#_??HclQO7}jX9m1V2`zqq-5f>g(mH4Q9yFsX6IvR9;;!ShRc0ZBl4y9&ub5sWdaoMV|q0K@X;A+7e(I)5Fo88McqZ1g9mbG98x76+zO4wt-( zh;<09AB_V{t(?dRckYw2zKbTHy5*w%D8?=iE~ggkrn-;$D>>@$JN4mohK;akNS;J_ zXgP{~9L|>p(W{&&GsX-5X}i{Wa)hn2`m1~_PGd=Fb}V~kelJxYz$V3~>M>qZzxU5F zPo$!r4ELT-ttZf>xmdL2Wkbh-d-6RTBTNnh#$b$GHJj2RHc>O0H$*I9sm8OPZ6c2M zZ^kLFLrBm^>h|A%9DN~Eh~-gH+Tce}+F*{8(p#BCiW=I?cFAVDZ>$MgP4i|ke&uA0 zW)8c_(_6+fHVvZ%paMNAa0xu&gusy^x+$DVd`Z~jmqWpALjaTA;>OaWxg3`L3w@pa zlo0TXnY!F6)N7G=p$BdnU;4)&!>AkC^?cMn;AeP-k_D)wL62l&>~`oNGi~ zjDw8n*<(}&b{$^Y_fPTC0_JEGz$c5RpZEsa)0&A+4x|pu8|;I4-DaboR5XKx(Q*F-g;&w9`(@3H2{+-HoIoYX8Y!Nj&4s zf~6xdo79_RL`WNNT+5pd6Ppb8Syo92oXDXi^}3rr#2-=WGzqCl2!Z?-C<3Z-l~YgK zC+pi?$ePiJ%sG*^EV#EChk*c^1zkQvSxzO!x;o!Ic?#Kh62D|R5p3d4h-aHQcEnms zUlcEZ`{`HV#n2#znGPF;WlOfhJ6>3bw{+)rF`^3#dS75 zua(8Abvod&fuzdJs)t}rz9dyuw3pFqp7q@&23JssOj@DkyI}T^wQuAo9?2Jv)49gS zI>JQIFu6e@_p=fjhqn($?3j+~r7|^J8Xg5exk%)bWL%7Dyx`jK+GRr^`c7I#coAiy zLo)8KomswY_Dvx>ux!It|6HYH|Dgi-vyG_gT}`Rjb)3_!b=^0f9=9IWsKqMjP%(#S z2!~Kmrubrb)h~s3@O-y774kKFdbe-n<>i#H^sS2pv5TMQ!Mp1*$1`}pbmSnG>{;=cgIR_$K~ngn#FA+G zM=1bO7r~m5*6;!o21A zK~QWVhT0~(a_N|XBR^Zn@vPW4cwz%^1%I*BaOKZgOrq2z462p!vqXH72_edFbhJF8 zI#rrQ$IBgGZzzv`oSlu3Hs}F!Z&>FhG#oPX4@(m~guNFTe8m^DeB}_l(?3O?Wn`OA z9q0NMWA&%_s9N+B8o>QQl)~=YXvtMR6D!)gUYlmASYqkR>xaW-jTMy0G5^)>__G_L zpL2M*sVE@uhWMYh@PC=07c2;rgW_ViyP6xC8oQdiGX8&^(9xj(&)S`@LQq`%|FhQE z(%ixAzn9!w`MZWKLjeu3yo*;nj=gAJw^|cccE^nlUcU9c5 zyMsXg&)jjEgWB!#gK3RV1E&tWw*u)P5W1Mcr#6K~&DC74L-x_BQPc#y5%ywE*HXf&i+@)n2Y_$xSNq zGXNpJdc~g$gzrzWRb1yM)t^D{75crxIBeZ*$FS%4RTL9J1Sgd?r1~b`S3zi!%bl7& z^*XFPlsGpS1Og8PmGfORd*>P$3RHbC{( zZRt|~0)=5OM5HYsK|pl+*n+ zRRqTO&IRx0TnNACYO4*0amhb2Ku;#t`T{nHVQ3CIh5B@qj*-(Tpm-2Jb z99Eitqejj17y-}cWBl@Do|^v#Bp}9>Iay~w8#37dr|6xRf1M!1c{HSBzpHaC0Mhon z@XeR^k3(!Pr8AMhN=DlC%zoY7aHUD$0rbjF!eVW1m}&8fY;50@;wZeYPb&r&3J?4l z=O}&swl>s>I?Sq9X6FN?A4Nb_f&4H426z&$kbd5U%Ow=(#EW}6#>ewn0KiC9p4fzK zR-vR>sMp0oBUz8Xz5Ie{hmA331t|@q<>PZ3JNaYKWifw=LN*#isFOaXz!#~idZ?22^$R=sUdoY7>HOH z()REo^Osnh%$~gRQn!9q7|_nseR@0&_;5pqL%iu~nmMB~c@GXBFC1T0NfdNyfPR=M z+BG6B2zpTB0BVh`!Y$iuvvUY*^0+O>ReO!T$yFh7EyQ#5d*s0{RF#$LBTk0)C=pMN zZ>8-5XSvsL2E74pM+;}*Nzx8}%))@*U%`a2HD&f&D4aG#DLg=K2C%=NTpepNe=3Z* z;wLL)-F@rH?Z^?#LYf~h(`5<~+=bhBQ0)VcZ-8*T>qAO{iXVHq%8}0?h_qt@)GW7H zKkILVE8<8nIb;xsXPK^s`|Ilk=+_eKyX^O2dBxwrO93+ZfPUUkCSZ2qonSg3;QSV| z-*2u(Ni)(HK&C+p00>&*q_L+{g0sa;TcXu}t=EBUisgxAsW?51Jyk11rNVJLQ8Q}l zb#whV32^mLaG}Y-z~6YCaEAmz7`v1E*l5dGPBs%6+R86LK!wgwa6ESb^NIDeTuTVW zI0P?{O}w>``f4T3N{Lx(*|E0>+nYE}@dt=!>wIKEHgNk+06muI`^@LyNh4jyo(Y>y zDelYLx^aLsDhogh@~JG{qs1B8rlf(qin}#E)p={n7s{@1s89}tuBGVgZ9#M(;)Y!a zcnrkL@Hi>7g7ezvVOlaMLUYx3{>P=u2&m-`0 zFelKf6rrIUBkVsQ4Y;rs>PkqtxcKN0i^G}zmXBospgh9=c2yc1l;z0t;`C^)5IbS2 zSXJI@O1~2Q3#39-jVi@QgY{z@3b(mR7E!w^oR|O_WGAB;HXPbbJu0eUro0M7~ojFaSiAyi#&e$vp{HLor4tx@x4 z)d@$w3P(s-Y}9e}HMAQLu1M41GepiCxTZ?2x-|7d1sW8)#@lTZ@ZO*GBxU?TLE1P< zWhDbL=K6SpT*NS1DuGE1tQL1!+pPMIga}FBmRl9F+iyqmi=s*8#IoMeq!V5t3nrZusok2M|VGm49wX zSS}a1AY}l51RRxJjG*}TcI<^e z%u2D_4$xSQzLiYK#rP_x8Kg^<@MUaMZ8GLE3zlS?lqec<*aJ4WE8%0!^iNjpbKHhC ziQbL#3lf16_j=?nE2}Wx=g#|R01;?rpEk(6i9L@p*;j)6NJMT8!X2Vga%|4(;pB9G zz|#{hTx~-+Xs4(Ne5S5WlYrc&{XdaI;i{A!l*`ZhV?WZISq2{&`!16iF*|TLyk-x8Q9hoP&q3$3{X>4a6hR;iHU*Af6n|hD5svcz^2rdwxG1$1B*Kn zLYe~-5CN|l%dq+jE=XdKIl&_@xVhckdZaR&507M4IHg<80He5L6Q9B@(ly*Vl^jxW zT%pTMRnj_<72|M%eStD?LG9}=TDT(YwJUFzV%g`K?h^N>m95+uK)^>r%9;wVk^!WO!-5Wq!?$@s4n)^3Igy>hPym^0&YO{TID@~P!DWe zY|00=okn%gYl&tbn_gL=#N5x=v|cdsa8iV|JcKd<*7Cr?!x6K|*kM4Iff1K2tT!x* zlx!(fG9A&TZSQe@pz)pe(N7&0AnIoJ!cnkkFAAz{^e>nLj0_^$-fCsby&F{-Ctajl ziz^fqxcp#ODD-Iy=z&|c3h`5B8u}Ud@%WqqyJj^*;<_x}k znJcU~1q*WmTf?A65w!)@-Z(T>TTvx7>O!%WBySQXzwmC4Hp5V1`B~fdmtG-)uVPqh zVmGunk7y|ZBG)%g3tfW&ug~kJ3w|V9dwn3|UQfOe*CZSgHIEFaYf^Ca@&$ARbo`LLl?pEb@ z+z1aZHTP6A_bsj|5+~K;iTj|sLydvS`<6o8r&wT0%_)bfhNi_UZeG}i)Uyq_^sQ>< z2P>Na2C}wV&f3Es3X+mP_ahi_9g3fRb^HxU4``6lyq0YEVudfdT^>JB5%h`4S=XYL z0f1d01@T1jWQ;%%K=DM?-|uBiOjUj6Hnu?xAQ2J!f`_xZ8_>DQ`cC3m9FS3Tuj9r}tjZbr82NYQX zc9$%u>~9uLp?K3`XkaK)_R^a84J0-0a^_v!SL}Epe*L}-qp-aZlZz5MZ@S0-ZH@tE z-}g;kB+}+{ZuPRhv2>Oc=xIlz2}zDf(7y!FEGo6{<9H#U7hT*fo5)LKmrg3;qmRnR zHODxwhAxxbgFsvyzycWN9pf?ul>^4BR;t=@)bqvK-jow zalCWP)~0L9EoKx>AL1*~$K2p-%NxLrKv1Kve3$k<;?k;7c-JE_nQpleU%-b3{3=wJ zxJ&X~1jB#Cy=9|p<`MmO{u;^U81h33z=GqqT{S(F=D@30hh+}akKp^PAvND0Sty%q z=E@=sBK~s+dcr#d5Z?_IWT`*}FcMN03?gvz@Rt(i7C%{lW};AdlYxeDWF{cfTzhJu2#aD|i)#lj}Bdezf#&GVV@VZX{r8 zgcZ(@#$fQuB5Bcf>u8NXvH#2ryAyL*b-&#EF86-E$MSG0$}C8jqSwl$=#F}@eS1g^ zBiP@abItI%V+#vO;EjDLi)RN}fZj>uO-yrPlHhY~m&Mk{yg-mYvsUT>j(f8gG8bEx z5mcHmePKWS?N4BwB!IB%)z8~B33QFl`zAxvJQAT)jmCTnR?c0-l>@GveJX>@P8p_K zIrhrO*Ls;|C@vCq$0UBDhH~_`W&N4pPDh^jih>I(>YUKv>M)-zmaxyR< z{yxS$9@oe(F&MU=<{ak$)-L?su4ZJrk}m$TmIY+1ImAi2R%Cs+uMGOk&!%JvzgFJ7 zVFIsQa?a#vVlrkl9G$#9l-Xn-zyHfHI6ircIej?zqJ~@o7SB_(hg6^L{W2p#7Iwkuwv6?OeqZ zIO2}YpckpQ4D+)V>384F4_IMqNIz|02+-;C^Z)jtF7xU8Wo1w|$LVNL{fOxpFaG^V zBq_#32!_JBRQS38QnAa$t(ZCGU|%=1S4hhl@Ti!t*BSSar7!x$=Kq$%AMppF!8z=u zujz9%1{y@hC?%iw{^jX1Z*kAkQn9}Tc1!Vn;yz#vZh++v ztR@{~)-6(EeA5>tq|_ylvKJ|yZbO-natXE1^z_a$W|AX z>J{0z@Rzc7I=bL^Q*4^I4dS$XIJ9TybnOuOsMn8Q$xP`ZU|@>OqE0DF0ig6|)Ed|J zj)UrIf^;@C3Z*EsGP_$GCwSjMz(*&)t8M&P6Cd4N_^L+`Q1WN5Ke}{PsYD?@(&9!B zLt+Z|Q&R98d74p}5dd9O`Z@hN=HAV}xC9g$o;?a~}koQL>DTA^B z^<=QK=Vrhca8t|*PqU4g)rj=ZX{bdpR_M8AX;G-~U8-t%g-)g+0EnBqx^Q_ISLLst z6%XIn1wDv`o?{&`?uLSfAg>MTVnseQR;cTN>cSm-2jf+DNG!fZy0Ta?VkiP{6;w>EloQx_f2zGc{uK8F$dP4y!(|6zshw->)&y@@*#qtqP0=rcAx=?W%&81B30n{rVx5VvRhYTB|RR!ywH~O zzljN$d1Y21U7|EIPVBi|qA zUu_T+5D?4%RGU(iPVxU2qcEy|tGLC4^tID}fW}jmCbHs@3bH7GMy10U;AYN597@93 z#X%5-zkyJe^W1-%TS_ioSSW3!oYvmtf4kl5iAg}If-RK^l?m}qCLDJ*qNQTGr9U0oeIM$IgGbur?vib&;3aI6Ki}lq5Sm z<${#aqwAe@!210yJ-v7g$;(`jdvGp$!EjnE+6ZtLz3p2L9mK+A(~S#+zSiiUiYR2{dy>QTbzhlHu0{ftdsjs=~#M zWGo*74AG0DMx4P5D!60C`r^SKLRnssKNl8#^4r~8lkd)8ObV? z8eWIs2XR|Xp&Aw5iJXfu%} zOVt_=hc&{gF7!DCEqj%NivfR0Owx3J6YnfAL+lT-bTB@esR`n5XUU+^6n_ zx6fDp@hVkkr?@s^WB%hDw0YbEtVq;1s zxG_%onq{xQBs|9;*Odal&P2-*teOJs&SioX(WYPZ>K0l@z+ahM(+GY+zo8W2#D zComA(|MaW<>kRyF+HdA)>Tds!`y2k>)E@93Y|mir$AEU_j@`~)Cxd}>jOmr;bPTZ%~_4l$` znaL-;Ashk2&wJ+NK!mcu<7K(g3k!ae8`(M|U*dOHD6nO7!JXNr#kg|iBH5!674P9b z;BGv0yhNjbcE3dc@z#uMnUw=uQwBjaS5wX_@m@`WI3$juG4r8Vj4EBi4#wAw)ToRM zG?9-{wv*Goxn?#Y&(>e`K1iy94xL6_9*ogUm@Sks3B@gKFjhNpvDW-d5RwmHY_d~7 zB2Qt$2Z%Ok(NqQ5p_vV)#5t8ZCdX|Aph_{UM&?XnM2|g&Jyaq_#P`ZCRm^;d$^g29 z2vko6gBD{HLEvC`UKa_uZw|zQ3;L@W9Cf^wQNoX;By`|(M=CnXs1IubiHSGXt8nv* zE%YyNGn0(CVDt?XJdTMTkWjcFH#X1@&9PHL;0ZBedH-;677FyPU<%Ji5w?g%M{w=qP;z z2Q!qANE^-yxjLPV-Xmmd6&IZig^6C2fCLGsRK`3TIT+g4hgYP5mH;pGMcIWDs@Qk* zMv?^h>GvSecbIgq++Ze&!Ced;fM4}bOF6B>2}A4+=sqI2C~@UY$!Iqd-ZSh-E|H)O zbtFWKNrdcsmStj|lzYV=*8yr3?)u?6tv+0!3hN_K%`zgks<1zn%!*tSIZ}JvM#6;z z(+!XzujC+MTENpbai!NYzb}zzQ#gQ2Ss&Fkpg6Kw^ca{jg&iT{ibfDP6{1w$ zw5G2nw2N~!A;Osv0kIK1N3gd%d%mx&H;!kC_x+5$Aij7Om|!Ht?^ z0I3kPE8Rb$NpZ-G4|cQxQtw;);wk9Kc?JRM=UMpnTMRw_l0b4Md7qy z$B|M+;El^p5}mPaiV!L%|NdCNa#$zQEnDmOKyv}eG)9;)*Tfr`I zgP=%nXl+wJP;`o)@qDvtREJunZ|iSX9HxSlCL~Gujc?{cfGb9cLq?LMkg6nq2o-Cz zVEkWD;o2~M`N1J*!AFW@Rt{j}v&;Z*Gh;W@x;{^Vd=4=mrlzdagT8%S_6D1NesWZ~@-)n%9X zjjsG59-3J>z)=d_LAIWs@YbPRX9)N@Z>4F08e9qp85XSC@~**`g2Lb(Ie4BnQ`p<9 z4o|8xupbbH3k16Cx=|4YiTzOV_wHcL|lA$)JkPxeEi;^{&U%D0Qx5JqjZE1FRbVR2WsEBmONt3 z#z|Tf@bAN`MjUhJ5a0-B#S*9QyEndr=ZL*kuu7lwHTHvp5=_Kad7xP6F&#>g5&cYs zxx`Ltn`dJsc)13W;L2ZK^f{tdxXrsPg^VL(Yz%}Nw5Du{AEYxAD@E)K!9f}kFz8CQ zWswu+0FVTv-mS%KSv3(#?2B0O8L+9K)u0!;KW&S?!z+2vRta@+-xqnKMcvdh5sq z!gtP{L3n0b5*Rm@Bq!Cs&qU~*g7!?mNBR6Z0hw(5!SbsoiMgWo5;6k$dBwB+MSLOZ zQ{<1+L{9UnmO?Oyt*vln+F+29`=P0Xdg?BIc`GK;UaD!`IL%%a5J6I$ZrQDvb)%U z0E|B0g-vjCBVu5;o|ZRvkp7naK9;ruowX~~V;h|vIRy$08_q9Rv0pwggoHg;)gN!Q zu7Cb1m@uzS9x}f^-%4GV>|0V~J)FIV1A2LS&zl)D1Pz;VZUDQDOEE)ujbpH4Geo|B zc6qV*$ZZ0~vz~VQ7m2TH5O3RGBY1*E0gZZ55tsS1c?i?n%xTh2=EXsidBhzF-V+fM z4LX@jrz;~zYKh99LH*34=-y}Z?>etc;xQXWXx!q|viv%2w%&hX!H%djT}u{K78~)} z_jSTX@+Jii)qe}kDOkJCa8+P}BSDBCDU{eU*RX6}QQ*L25*O()OqaK~QDbpN0i;!# zz`0%6hDI`zk3nyw8{Zk`sJSl2gZe<@(Up)&eoVclTrp(0KY{De+i568vrq5p%3wjBOB)AAy~N}+=cPbnw=Jl z5gZemBPk_;ZOjI(LKgiRcjIXUX^cQ*rJPXbcIrQDj?bsq>cSTvK5?T-!@?LiN2L6vb5y&1CF3 z{J_8LzB~4wy1d*}|KU%O(+jGdnA)eEo)?8WFa?I1kAI@A8InTXzsxN)fNm=#9KiUn zXXx-_>&e{TammmPqNYGe2jrOdxtmPBt2(V663N)n(9F?rzl@cwro{sb=a~ql#K3xs zWvg;&GxU&Kx~JChBp;gn<-H?I#QecOgM+XDp~kH$V07weBhD;yjS2!_N}gayJ~TsA z2L|nB5~+g!){rHSsV4blQd;onr|4;Yfm8LJy9ttM-frkLtF>jeFE) ztX$wS$>TG%@j(MJ!X7|WLx`FaiWmr+N8{~E-=Wwjb)Iem?k1(u+n-5iBKM>Gz?G9X z=EEH?CqVSV78jVe0Z@T|ilH%==x5jehOt$}hW;YUBMOf3UW5t&bD>i;sJu6AMofQN z5Fc6O56vRUI?IRDjga+%x{8J|-K0A**@HeD6rWs9cde4Jy*Mr86Ka_0Msp4gqoa$u zd~jaz6TsA3k`M^tSB+((FJq+h zKvL?eXM9zJb-*Y*5i>9+N9Jw-#pr3SzUsIEYtKQ*Wd2TJk)e;)#p&)0U}|d*zY1hX ztyWG27r-KCL&*j@>sC(MVaq*AcvS(0Qr%}9q)L}gweZ!19y=hf@Ib_#O;8Qw2R~`5 z1Qz!EjWsX`5BQ=Vo))~ZGp?wP6q@Eo@{ZE<1r3rfZ)AxBDhR;l5xPyz0VGEKz9| zxb{;bt0XM)ULf#;!2qhnd~NTMK*DCC4GsBSJ|WfT0g$&AXpx;K+HflMEOqltgcM#; zN|bN0afaUGVK>7Rex1^cH1Hk}(iP?$fA9$s1IGjDgEyYwR2ikFi;I<~v5D#;X5o|l z!!(>v`&W}4r=nEe*)l`TU)e$h*0YPVP6EM-Ruxtdu_9Zmku%b`#CPwwghz_Wb=3+f z04`?$3ZQGb8i@~#|8WK(t0_h7tD;j7vQsQ36YEapv;ABeBr)_mZ#+>ojDvJs`JB=qPe=Pa+7(*utW!BwAh42DVm- zF|OUJm|m%s|7(V)rdYFUNz~#T+q2jbvn^DO2XCtubARri&$83`$0Au;XVo0msF@h* z7fXyFt_ixnXDBTv^DA4Yu!90zP@`u=t6-O3=uEjVgT=x^OG{o#`y{+w9F>A8+B?dz z8z8GpgIE2D;f5S}Pb5?hY!eZbpc^!;1{c|Rg$M+!8Q0%t3>O%>2=T!WMC~oK(;>toPKA@Y5EL^6X|o=7y866#df*S&soX7! zTiM^*+@RWb%&JL+be7V}vIr8i)U{m8YQW&VeVKC1gA!1gt8(Bxy`w^g*_-A|5Xd!Y zo_%MT5O#P>Ddz;yQxQI&hI3NtSrHMjgGAk2{PwM4;{dxrtG1gtF!Jxch)rL1Jt2Rm z1yowWx!_8cj3jXpH)LSOuFC_2v>Pcnm^{^4S30tQJkofv>nWsd5Z)Tg>eDA91;A;B zRMDG%K;z?kR?$^7G+X-1?_Qv@uMqwMLIMHSkB%N6hn{x-+g<7^8<~sW<@mz;S#~>M za<~t#rYX`EuLd9VzAyRwd*cSamXefi7fb+(MNJw-?~|%C@Bl7vxI)@+@0%T|?XiuG zHwq}X*lSJ61S7*~hZDLBA2Od*4L}H_$UGeS><+hI&y=S}dcILqtGf8=x53m6fZFUw z`;zn7e#W|~L73?q!SBT$o|A8+ecC`sNo&|w;r`Vw`6~<^fUOg z+nvYH_4g(I{U?82weRZ2pZ4wJTERrhB9J=Q0iB_b9wRa?5G{CIuD&)9~ zvGZTreDq}FDh?G(`BU-$W0?7%vyD{HVjDQM@3u=8*vdnq)hF=+P%&bzI?;=p7_6}| z)W&XQ7JD|gZ?eQfPu#el-wQtVpsSms&zrTf;5-=@tlnzs*=d^w0s|^&Q)h?3M+eF| zXd8fxQqZJdsF@_dGf+X(YXI2kk)F*h!wk}-a^R6H_>(tR{BdHL?HTmo#4ipk>nA<; zLhxzCN<7#+_%eS^oYM@-ZB~!BOYs*Nj&n-oKO)ps;@iSG__$AM$_kbDUmz;igKcC#l8->i!AMxJPn+kgV-;!dI*NZ_t7 zg_-iy85MIwS7QYv?@ckk%z#y4=|Jk!fuj*OiArearM$MPV`?KV^t3T8S|q(9z%RWP zwb6~-swh*a#~LQiy!8#YHWm>FCspZ2Mr7xCN?K`g$R`R{%=r3CKPV|eSycWs}~zfA}i zm@4U$i4hN;qe_M5DOF1@3PSS7eVfl2;ObZ)Z^VoZ*(LG(48HMs z<$m5xPhehDk)5}hDgNiV1;=eFTBQyf<{p!2G3|N+Rp#!ZQ-E!~s@i|YT=(z>hJ`xh zzih`7gIbVJpO3aRfWv7N_g!l-bC$1*O&f-Aj-m*Ae%J%&outpGwoj>={Ka$@u4U|C z6tuF$!?)g=mx`UB4eJ$Xs_f@&aEmrr(d3OJN*tKF<0Fh*n<51xqH#1=Oj1~z*7-xS zfOn0sXV!5R2B3nRo-ZM9HEW}h$?afP9nxU;yiu)-733aTygkUm7pT{rY}CNcDjo+r5) zRS%iq6oUAPE$=+X!wl9h_}et;H^61@s{i>{9~K?@6d(=s)ZOFR>}lwdj%1{eHV(|Z+1w* zkoe+tbaVECV;gYUi2!0 zjq&5}HtzbAh6tUtC-X6hEZ0Dzj+Nt#$(xC6^TD{=={lg!o(9O}e69_gZe z#Vbfvs*Yh-0Ivp7mraWyIEPh8jI?p)jd-IyTp)JXc5)C%B<53bntedH7plIq#_=;WO|Uzq93rr9C)kQh}~os{n02voD48fE0ZX1|zVWrL~59x`lomj;*#l!<}HWabX>fV;(?^ab$d&_Gc zDLEJkycefOwzS$_?iL98xQguaJ;U1qt1PTTW>~cTzCviJTrZ~&BG%Ph7QNV@kgd>z zJ>rT0!F!H%GZ$gD9~GdNcXo#lT!=r>DiYc`gldQz?p}kmhEL=LF^1K?mV0 zYqOYX+$q6NVyg=W11mf7Z!4qb+5Uq3)!OUZZMiy+4cymA0ca_}0lv;hM!gckvNFyb z8&5F@%F%femrzVL?buP&0)w;Wqonh~m}{1YT|B)1aJOiL`geEpS*mxGIDiBfu?;Zn zzLmN9tiE}zUT^Cel~nKiZepE>q(dl{HDnPAOa7$iM|a{$f+$8RJwhyMozGnP0f= zmLFJiFlHmfKC{dhsQ0OGV|8q5y*NI-;uO3AAA zR?@d=$f3bN?mL|ccfT0OGiiFT?I3K_1Xy8sTd6&npc?+1 zOt|=nFGym7iI2cIJ58Gs!5@@P-3Ne<);vc9xXOT|(+jPMa1NVH0iU}Xr9$UzHzx8oL z4_B&CfYLBkG7?{}o;r2iH{_r_CfAG@Fbi-=59s`c` zZ}JXQa9eLl0R)Yh;~~x+zIa!&_690u#%SHi^6CN-wjP$-`n;P(>%6TwOphA{Z{%|@ zpC%4b7u)Mq+i@*!J6zWbbtCBvwQdT2E7!f`O`!Bxy9Rd9C^~)k z0Cq}&_@PvORvUI|T0-HN-2Zko32NGW^b+Az!x#d_02H7MP+2Db6B!lMG`7I*eJWE%zK``~ZW z=1hr~^lzY-K4LB0lQQ4wk%E-06p~kE2BVT((T=P_oPSxfz}CRv8dJRUAS7l|7X=eG zq77t(O1So+ETQIfj8l&1NjLa@nvHMT7GJQDXSje)H6HU(cgHI9QFl9%h4^0t0pz;G zI&lNB{H&N&FzAq}CYAzy)DigwrjeS}(L-s%hxsof;3-&AR?>s`wEEIG5s`;Ru7;5W z5BVlFy8NYrh5mxDEY3G5KQ3o~PdHzFJzrh_@BlvABF39V>m40r<(7Nz6v48f0cGVli}C7LRf9*}^Z4u~@KR^G^2A2SG{)EoLE6vt zH_md4iM>j)6&GdSw*6llsGx)&VbyOt*w4dnQr)+ebq>EYe#dm^fu^XOu-VF4!qHB~ z6oZx~y|}Qs^u>l>H(O!#k5Gt8DK@jwLrmqi-BDZ~j*(9P%NG2dy)kwr2H3F)uy2Xk zP{wH&p}@AU-Vbkt#toM5C2Ot8&hNd}>jyJcy)w_Gw-=Kgk40{;K)r&mW8NV_lq7b+ zS8V9Td0gS2YmZxROxuRY@xJKiYjB~q9CG08DHVVyZi&-c&z84?=jqMZ-zELwKYne} zC8U09&o=gDIh`aaw{j{$1E5)1h1>0{!ax2Sk*)F1wS`>z_liHyvjU3?a=-bxHk7?9 zpv#X73!*vQw=U|V4&%6*OpF2Lc z4woI6v+P5G&9}3)uAkt4@%2u@fp@{*XKdTa#P-CtZQB!Dzu2~&2`4rtwkNi2JK5*I zTf6(dRa<*;F3v??oT_uWtNV0+KV!>dPru%Wv@)-pLO(V-%o{hNgLk*O9^Zw>shkfH zRR!h(HoT&|Y;2d=Cp?v;QCO%Cng2ZYH8cIe09D%3;;+_sYt$`RQhUQc@GsR!&U)!q zwulKoYRgSkU*3a#=HCa)eKF```~PxXgh zvaIFDA$p{WpH$}Rz-AC`N7LP8Kv-bh6qVqoyrEClm1?aASrruTb>YHkf(~}{z%sTr zI|?ui))dciR&F>5hpi!;vT!lK#{I||Sf%!|ojRIz8hNre9OIb@15+#Xzr_E&C0+e8t=V<=`80eBQDLW?;v}tZjk) zwtGuX;699>nlq(h|IYsKX49@aNCVP)E^?hUJeuvFJ)3EAH*RyNRR(!#Dnl52IsX1Z z__|MxE(K38S_DqT5WON25dvo%=(*PX5Uc*sBbN zESfW?&jaldeqv}rncan1pb={PHaH;T2#I{`vePiiJ};m3TdP6)HX(@dFV-twqY!!! zFVZ1!Xjk8`pdg1TMZ7)R)Hoj(U|o`vofe4gW6JFrelJ8$CQc;^ zRs8G)LZUskDZEFKQV@fK)5B~|*viqBQTV7}IV1pF# z!@%W__q#Godk*h^ttLlCa3MRxUb>hUyOVnw6XQ-r<9`R@6`r|D(I57l70Z5IaF?)r z_(D>(+Zp*KQRVS&i+=4i_(&{ONg5oWCk3j z6XbCZb{SIvO;1!ZUaHk|D_^Qv)TD0BHOVg3kv6Z^PnOx%8e{HUJ``_MZRZ#hol(IU zxcCa9bvQ9-cTjKDd$07Va^3!~H`@n_GY}y5OylR>;SoG~PkU_{cAB2pbB^axx;^q; zJa4U~9qxRnSVFK9#69hg`(Sw~Rjg5^Gq>yEyRWhA`|ELX=-$YnbgYDD5QGb-jGh=D zN2%{|3-i-;T4PMb?dcLc6**S408)#lG13_^dyAa~)vXy$X4tuc*toz9)nZ0?5E?M- ze;wr9A(!NotaF04^DJ4zH^JZCZ-kS|iK%#B@2{z8JDc&2F!WrqGKDIBJ52a7-8#aK zY08o4g@BO4H)_Ryh?($fc=Jqo^FP0lE_f@gN zQt~4wYg*wX&o4O)?|0dQ)CilW>NP;Cc|}d|ytcWER2~1t-!&f*7#SU8LFRhNFw1wy zZ1WYAp^eguk*W($zc@%5vb+a(J(4;sF#|=I`)E;JF3B*0<*uORm|Ka#Gpj$#g+4+9 zvTs#oJ43$4Uny*4?rCx?RJ%=#JW$svdK|DXR{zs;Oj zLz$bFAr;tL@yjb1t04}J)1jCPF)LL1nmg`ra{onJ8K6f0#i))sq-s;CxZAIj95=~_ zgb}(93Gw3T%XiVLNHm6sOlRr?!|W8ZI$w}$l+AB9xFO^ZVJE4;%mxV7ggF&m(5okB z>ktdJd55zeM%CeIhP5YDBIp?pWHCfTC?zXU8$;3*GpaRqHAxItciUe213G^$MZ34Ix0w-rXAg&Ck&5GoOf650zL|Mj2Dr@j%Zrrx)668~wv_g?t^ z?${BjoGx+I3)qk}nl}*Pfz;%+3$^?*cDXL6Jy4bRhtO`>%8|fL@hx7VY(=23=mecf zYGDc=YqS0l&1(QFr23d*Nhk!QNk0IMOh#6|r_!vUHNI#UuJ-WLloVTt0Czg5NENmT zE#WIuRB?&Oa#jR%IZC2uIOOF}Scxa9rv&VP<7(V-@Kz(3zXFiu$IkYgF|afn7tRw1 z>9zgF?P18XM1wuF50leaB59!4P8U-gna-8?%lVY3aWQr+)KW1lx)UzU1bff7!s z(pt=28DCnKigQ_q{qVVaCGx#fHbJvgY*imWQMPuj|Mjs^j*NM3S^2OC`NtT#p|Von z<1g`t)hH5^G(J$iUq^U=Y|u{DaoRK;?A>k|6tpwcE6B+B{1U9F+o{21_r`GP&&c7S zv|X?|{O7adhHQoeONOA4U%$_q559hEg1`Lr0^05Ky?zgr$BC^K zX^0{(yB^BLIPZwJnPmje|E63Of07^q9Zk(uVJ>bpN&%4hdXW($oXfx}&ApI&p8B`} zsZtI4sXL$fIc_&Gz5-i{a*9{zevB7lF7zFYkso$>|8;TRh+Goa`5#^oH{Gk*fl#M7 zyw5AuXCB_UI~F3Y^SNR;Ki6lcR-t_=+OjJ=PrdEf)_cYqUadB8R7o6M<0eAlYcJDR z?NLt52?t;mjFl_*=cYqbLWgKe|2E_)VruDV@Pog($0YXM;sPb8y#dfx1p0XPePNah za{q9~vfYkcp8K=c=`XYFztJcd!n0S18)bERLScx`2&jilUJ#dJXhKR%fP)GorZcB~ ztjRZJ8gK30;x+L=&p!tnf>|fo46TZS2c`Q+Z~|bc`UWc~*EY9H9acbxClvH>@JlZg-S{FoJ>@?21RU{Ao?*36ZZP0S~pXW!PkCb7CbJc??qS3gTow$T!K`A~Vi znrf)QNUvkHIj;?sy*MRaZa|AWHsJU_u*ET6iRT#5CFw zn&cH{rF(n!ePRRlzLlQxClZbQwrgdw#03wSD_W1C?MO?NVGmYaM0v8RDt=z6SFP}a zYg`#QO~HNpf9@}x+Ec|1$KJ(AeHv5a0QU1e!9{V^iKGAemO6>|E)6#)C11uC1L2+F zVjOYWF!%FgjB-uQ;q>a6FeL0v2~!ck{&~F8$8t?P$DF0>IiV7t?SJ+8P2{Q8qs{mC zX-r4c%Tm@eIkqQ#i>`CM-J$I0M{_( z3NjzS(#H#ebRO5bcXz=arUYpf4OS@6b-v}ea9Q?7G?(kc;LBFac5cWBzBD><9=qu; zRLMk>X70hi#q9!*ckm7L8DsZ z(mU?4t!6c+(eU9K#%&@A_%zLXbbgoJ`XDEMHBD- z)0HzvR2|IpM&0&f*O`9R4(j+*Z_xG@cb!uzRu0Uoz7x1&*rmT^TOx$RJkNRu*1v0- zB*L`Cq^am%cr`GUgLxMyfL#~)T>--AuZX#5HA}1`^*k>e(qKP~Ev$wa?a2PB{!p|j z;|O!)VGk59$vLUk#0FSx(2(4qwl3JQn`ITH8L=4ko>VFEV?}P`uId7jsxZw=-R2!l zHBCkRpFy=mdNSAC9CAFB?CvC)*Pwg}XV0e=H*a(ak>}bgcU!M!!0GT(_At7JYy~rV z%>vXi%wQ`FDC_Tg63@xRDtgu0aMe|7_qEEOqo`zFBBk?*5u40LPfUKzLtOVy9oH%n_&M zxslPb=hpFO<=cpqyPh6rd8~O-2a!!KI7;b07BI48%O-O>O~o)Rs`A3USS#?go_V}$ zZ}nOwtv_XdKUdzrqAc~g&t4M|5`N|kuS@mxOtDHj=TnyyfPsnh1@)AU`Ly1u7jbO zx07;V)&EFltZvxh1Cl7Uk06~w^7xZ8668h7Uo-jK0C#C0=Z(CV%}q?1@pP$135BEH z$4$l@{^dNmo7kdQgIMh!f5W(EN8+(ge7UfF3H3Cvys{D>P(@ivqG$rc_v9jJK zOa~O53`p$Q0j16))MhT=ziJ;nwr4_jQHc92fPKO0S#;~_!j6>2wp`gv4C070wo~lz zYdu?!#V{mi0+qS6Wsxgune>i`eb#tmY&kSbTowIaM^gc558shlIUM> zNpU^wem@?tw?{m(D?NMaFzM70PMegTL#a2CZ--sixb8?x{;^qc4mEPP#KXPF1-I-5 zJmrXr)4L50?Q2^Fgo_`R+QVag?rFh>Qq-0TKyZ8d$BOE0k2BtGoZ9;ga&F8F#5bbw#7hNSAqF#1iA~o&GX3yqA+dbu|WsvcYU_9b*`y4q?Ze1SKZvzo6pD z_qY@dqZsstKna?Yh>A$di;<)*4wN+p$PQXCb=G{32B5~iQIXlbH%7fTzWOkCqHB|P z{`xzER@BvYZ8W(}Vy422HBIgfS-N!%VK6cv1okLdetHXpWe6YK29GmgT5kB*F|aj) zqH>vmv0XMSU|yPDkd^t#4FwCcfNjL`T}?Nd8l+hq&T!u5%-N>r=2>Fw1UG?3k5FWnU$2@HTXtxoRj}}ad;nlE}*i%S2Uo%s_d0(vkh=Q-w4lKtw zZA-#87uiLmW>ac$orh5V;Uq@9@%z>WlaZ?e_jH}wdx};Eibt=lEppzB#S#VGI!8%? zR7_p9d4Ryi6J5+?Aq!OC=JZhpcoJNm;JtzFlGoVfk`~xUe4*Y6B5zItNbAsg3W+2F zj6rG9V+Cmb?&l}{%;AeR!^Cq0L#F>(XdFt?^kp3j4mx;P@qDGg4;bqg?ndVDjbtuX zKhWY%dpM%H;vmmnG@`6*xvThEqab`H-;R+*Gd;FyVuj3-^oVi<=eu70=3m0dG#!Z| zb5Oa2%$KcQ2}^KX)FpTXhCW*UY17mRBmKa?*Skbdn2bS67=dr#*&mudE&L+w3?1Au zxo0Y!j9~SXt6dpnZ3U376RdJauR>eFv=KoS{WWdU9n%~!oqEHCw@ii3kuYH>!1KUU z2u-`sov{Jz>9iG5np==_mouhj{y*H{ZHD>SW0we4{zjtKCm(`9Evm@Vpv8)Gr_aJK zWOr<^vG}l28@KZ5Yl$XONi$X3UMwbuk2cr6!+-pjbkAdBbX8S>dY;APCh)YKiX7C` zf8Hk3H^V5CS6N#%PH?DhYJpb|U0JL?HG4uM^a4XKVKB#@-$++SSTslYA(nYlH5@0kvPO}Q3zUMQJuOx~Hqs7KT=ivZx=S7>yOXKo6$iaA zFm>}-#c93TesPgDv|vbyow<28ZaW?1OWa6DX!#1(v0S32$mqw31%bO_M>k3Cr3CgX zHp0g0R-c1!s6>x5?X}GP>3>v)fnlo{hkOnm#(BSa`;>nH;EuN$Bb&_{utyNw4K!Cu zG%ZZS$X4itY8xkKXV;4uR5Tq(cCROQK4vP)TvelbG4xK$%$>s0V?bp3`@#TJkEQw& zp;D@`eM-9_q0tQoOuKwD-um zQD5vT$+_@_tOt2G$J_M;2l5|&uWOQ~JKhu&( zmZLr#zX7?bgah??wIh5%vF7E5FVM-dNZG4o(#`I6)+&Vn870um4B2i4DLZ>NcOrX3 zo4G&`hY7MlPx*H&Tc(`vQr+zze-a6w$7NSpbBsVDmi!Z)Kt7b;Cwsl_I>w(H(#TfoK9V@rp?pbh%(SteXz1Va>d`NC1SB zD|R+SX~ELnr!Woa3-iqg&FMY(GI8<8DRG84;w`)&#+BWXzkm6c=B)&wWhl-r@|H)^ zN=YQg^=MH^kdUBf?PA|>D9z&WdFUC{Kg>u+%q8caRBF~ze2Ex#O4)H>O)SHPEF!!B zna6J;84Egdj;?BTZ|#%D9M?e#g6M^?JZ>E(pghxc;sG?hEDYK|N#^H08+LzX zQXm;VHxoQ{QO7Ez175RepK2O)0&D=V116ZGH{OjBAP{Qj%w~m|TQmz}Cq(BZ601Gd zk9iS-k~*Kz;OO+XQhYPi;kCO_CGHg0u(pPet^j1TL|rlG^lNu}?Mh4Z*8mAYZJBDw z*C^NlWhU#ybnX4J5)(QYQ)84HEakJ>YV5*mI=87hL~1o3O4b>v;}b~sEt;7IQY^VwM=OVpH3~5k&^ZyuH^0H-GbYaJidM=fOZ`R9%<>aJXjgA9 z$j-#NDFtIaj~0|HYcAot+xr=JTl8F6Smg>5)n@$LtwD!m&e1cyJbw0-jw80=lxq3x zcD=_jt}J-#UDvDlX5wJYP>uq4pAcY^S}12}Fc&tUHK1x;^6xd11Bz6~>H?vDXjHQn zwiaTBh#r|e{8@38^^C*|Rr)oslgA{}Me>kE?CdLD;#k}vbN7#i=su4~6?J&)U?pd( zszW0hD_`am$rnOL`ILuDh{3AtzB#H5$~(Rh*A9HVrx)u%DivbC#DEbv3wgD|@=tW$ zb7EAMyOecjGybYhX#^wdmAn#(0mW>4ucP7=cv)a(8GPT3yax@NbHce`|0X5oO_?iY z52yGwE#pa4dP5OI`dpu|^2P`LaG%HyRS-DgXd+NDT`jsOc|bjs4T0D!;X0zqyQ3-~ ztnF_V%&)d}>*F6GW+(++z_!hjElDvtS+en~X^8Y{6OY`EId*lLVMJ&~E{{)!Pw#cW36#wb&8*DH5^`sW0C`9(+3D=i9 z2Q84r_e95NGkMICCG^yYupmbyWapW(Gu3#AR|!Pb@okjxLu7F zIi@2hoT+<$%VUs6%)gx|=Y6TIgx4H{v*CBrTzHI2XHo<(Lp9gbW5>%Yj}8j6u6%0zN$iQiZd}j{gkrH+<=FC_*TWf z{`DNRUp#$PQ1DX-{fRy*2R$Gh-A2Xf@~=eZN>|p!AEvy+k6S+w>RJ~l$;GaU;BC%F z^(@&4Xx$b7q+ZDZZWsRm`g{$^D!NqPc3IzIv2dn|ioS(cNm*P8A0|y#hxL(QHj2}k zyPY9+UtnHc4}N&;uam_PxB4?kVrcao)Nr}8TIa$s!n%&mh zWb4s_pZOcwOeC5yfdtyCVHI9BbLE9(jNe(_1mfMm3fJ;=lVar_^ZaX0&v#nlJv7rz z?jbzF@{>3o=IKV56J^guY61gvSasCn)VX*G;&HRyqA4SMRt*qz!xb{Xqf1au{H-_ zYNLt(Q$B4QMM+o_zbOg z&8L&_g(9{zlzl&nKn1N^c#g#hAtom~uin>NI)Kh3$VH3Swo3K&FF%3vcmD=R{v|7- zruf9xZs**mQBUNvA5=5jN_OyOcx^K==rRr<8w7V~N@OWMi;vS|`siLRY$V#0GS*sW z-86IPr?<*=2TeM8=~6szFB6g3zxr6Wz3QZMYXQ)XEkdjksv;{k#sy%@tvStQ)8Z5yndG` zc#vPE?s@uWhERwakZu(?ER0Q4FkpetsBCVc!SNk5iYf=7_3$u(X zRHRSPXuCJ$uD0S2tTs>Hbyx+5^lgX z(?$C%0@|&meLPTw7fEQk5cDPO;OyKrm*DuWjP07hd=TTInRv+^&e16db`c3MRFIrJ z@Z^47@nyt|S*z>lD#p%x!(5El3zvVnxM_7aOg4dL9dV#Fr+<%AX!@&&X}a>oc;`Nm zo!UU@{(e(lW24l}S$EO`o2sK!ULG#&h<7ddx4HUb&9fZGS%!DLc~rGf%_P}nOfqCR zUij*qG?VHa;cUzqQ;MfV``r%koXD*8*L92ICR9wZruXC19XiWFu=J3)6gEeNJcR;>&K;!xOXg&PN_ta;%W1wp51w~{c~WL6pk#Qv5u zXFX4~6Agj1s%=bY*);sVtF4xz|Ky@j<%Bm^b5asT#KsRtH?(DzD3t|BrBvjMq6g+e zyBN;QODuRbPE(;$N#!>FsBI*TNmPPpPhi=4lXZkOj!FbwtlQx5~pY3_`UNgnrijS!7(U!Di#sH^`&7sKT(>e|!Y>fmeWvLj?y&N||LszV>um z!n0c~z8%b-&A(j!f-BiaY&e`y&Ru%#=~LtAv7ayof%T|Sab{^Y9-&nx*enNR-)5{A z#QaIAK%Jk=S-aRjH?*{BwRrgYC7D}fS48QH269Ut5{%TW^Zqr}Obmz!RXPT?H@RW>X z@!Rrxw{ag!+|_Tc&C<-R0^i>?ebusi{cn)$)=IVyYcy4=deZUo&+zaj~o+JO3b*f9yU)4Wf;*)N`Vq0DV~P6|L%|CTN>9iITy61Du2S zhS5y2bh4B@FP0ai-pZgt2#7)34cmBadTP5I&;)xc%CXZLSM`=gVfeUPAURa!nm^Fzg5Yt@^onty&( zg-{2)aXasGQn-N4U&1S-b59VnfQRx_hE#;Lo()_5D?*|6et83`99Fh8_E|2M*+3Z& zCBXQC$rV~4bE~(Li&Tltmt1kmeP&{2|C1PM@3Azkw2OHPM1!xniL1re(aF3gYO&*jYEG4LaQZFjXKaJ7|Ky{#HdjlA}Sqkd9DPd4oY zT1v^VoUe}8^Bw5 z=(%>0;Zw`t3(t;hpBbvvvkHfTKhF#5VD}q^kIxs{pKxhA@$N9hUhE!JX z-4$+w{)8|}|KQKcUt&E~eJ=lgVi{_jJ8E+rOQ=p*e1s($b>&^mI(X5){26i|b)AXu zZue>qw&TFzs;L~Cz6X5iEgX-aCU8ZQA4rEn$XqkVk1e4U?$FSmQp0x5lh=TWbXosv z(xurcDnA3UVmmetYV-EWMnQuaqk&sJjl~2$_ekqryQ?WhVEv!JKt?Z6@>+#d0}}+f zR&|tDx)BcscicpZV1iBQe!HS_0e&l+oi(1??*2Fdid1P+MapxZj7ANW3>dKz2(ZG! zwcNI_X1CU3nXlU)!1h-`K5K3bk-knY={a?%&8k>jAX4>tS;;k&5qPj0L*OC>ab-HS zbx;Wjm)KPAemiuL~#r$WwCM(={q}G0djOf7qpA z1_d$7U&-+_O@YVQC&{`jfez;$JU#~9*7PJlbSW7z5*hFj6)Hj5I<+m^ib}5GkUOx1OxLtjs=R*w?0Qg9sSBw+H{HxXh zAb4zLf@Swc>0q0PZ31pW7GFyW!RXu9ts|F&IZ*4i(Tq;*f0|}3MVi;jxkV7SRJaJ5 z5>mOU3Uig_*YBpWo#Wfl#@-hl2N+PD?ns5hC0gNeZX>JsAB&RR9X#)36YyC3zG2Bn zn7r>5->8gxzCG0)f%D4kYB;=+!(n~kMQ6Z8u2KsIRzR8!LTe&z(5f69>}@<`rb=C? z)3LKHKnfCcbe*)fjX70)E3k&+1aa+htoW8~T{GZyoGxatTfNz4SC`FwaZrohXf0%4Ph(j<)>r3pbCT9lWsfJScpw_!e-nDpE_ zmhRDL7P>76n^sw6OK8Vnu$tuOQx87vvjr% zuUGiBuv}mJ8lSFWU$1iivF&9r#HGfV@XN>S`p7Ynl>hHALDrN-aw|X4<{+I7W$^=Xn=RF;f)Q!2 z`$a1?+bU?ug#5h;apXolIOj{e%~0n=)<%9}d^K3Rku8?o{NT}CXWTg3e46Wx=YT%q z1f=UEOL**PrdJHWF+uMBq0zQfl}@wilFvDvD;x9r!D3_eBREr=Bugb4y6J>|`Gz&k zs2Q)z@M@#Flw?Y&C{=XDQ#Hq_GCD<$lVe6DZ`d!9nlnMR%mbn=3;7Gdy^e!LbDCDB z&ib-Ki3WiRHgo$8;5U2Yy`2a}=R19mKrGWKm{GA+=|)p;GW+&h46Nvo6qny@%yAn;oLSaBf34GT<4VjANcZ2mdkd51 zSJ$DIXWbby=1^o}e0GHS?*?BEcCT<>xfPuV^O<=$wRc;d$EK>@gM_C=Ll5VOI-5)h zkw|8>Q5b^DO%pW?NA{k+haS8EE1?r9)mbCBqbnGJ3_=J@xbildgYX6Lx0YGy3uen% zQ6^}9FXx6|h-3ea_RIgH4RzIB;=2FPQ`DQTUB7yKlkB9g_Yo{k_~6Ha-@KH{a+Fqf)Gt*T#7=`Mv6nr>sF1K(l(+UeuFwGc5y+t8 zZru<=1sRCemnWGW&2H|LVk(iR0%Z@k*X>OlzWuo5+{kjmso_0Z@3|M@aKqZ zi{7bzi(Nw}yZbcoiWCx#!Q}bpeHtbp-I81WWhX+_d%49MpQk|u_LFyS+cvk_3eL=j zZt}J)Y0o-W4U46FIXnw0x(9sPQU$g6XSe<`=v+im03^t zJWlq=uy$f)v!To56upZc3Bncqri=`p6fpH9z~AC}s?We6B0LL3#pDKu#|#;VT055X z%3Yo~H^U0Jp9@cicf_UX<`BOixnDo|8eHYOb~!4|ZeIo%89#G^y|AZEzT395uZIkC4xcm80G-|QA z#23bJp^THUWT!;R<*Hi58x|101}+dfmhp@tUUUg0Qh2Lerj|YoluB!X;{Ek(PP@Zx z3Fo9NKZUXuqljJ3I+d1eNZ6l&6Ea>xfabPv9js!iJ^A^E!S``tw$OmKPJr`NaF+3)4OXKjVtk!*!yZ)WVr1EJ|; zp+gP85s577Y7>7_rRMp2crNf?39`OlnaQf6C&&&?cgQ>Xrjh6KLutU>S1xEC@eosx zCFKg)nKCA+bA@R{w4&XXeFf3sKnSG@v3%>M)TB@po6(5-E+T3*dlP_x5<5006`09a zPU<2S!BR`mnEk2)hh%-xpImcfy?XhS?4Jh|TA)2XjQ4F9n`f=P93ZZJJU#!mDRWkU zpWM`4z$et(dCrSDB0sQ$AuIN=%Ei{K-90gFh3()4l=Ev9ek+9s37vm%bF9Ra63-a_ zTX?eVhv$GdB<0&HNW&|KKLuTx*Tw-wqEiZoOb;ul>^!Jv7A>PkG?($!2)-RB68#B~ zcp~`RogUr614bE9tSCL z;G%T*uxJPRJ1*Zb{#6G5TYzobtdk9B2Ae+J<6~(GZHf^zccpXXpWxkKdjuJCVy>XW z%YrAk1H!g~(@hEEfwBPzZObZ%Jp<^HXZmxb!W%eobm`6#jM9vV#rdGAG*`#(HEQ)f z#tVdYHLvZ`8+P*54``T5!nN3$9G+CtxzC7_V~zM+(DvQ8Iu)9EJ$14U=pw*Br9A8$ zIy{i;+K_Ip#%6uJ?z3p-?nt)Ta%16L&Q&}sGq&pipW1nAqWy-io5K4ppM`=q&TTg zf#ot5t>utJZ#wBeCG{;e&sTu{CY`Px<_>aiQPn9d$@PpW{$X#qrPr z5oTg@n>phwa~Jx@1suYih$$0#d2EqS#mK>Q%b?u_OSN@CM{0)~QZO7LWlt@o-LVA1 zGsALN7~YdijhGNzV>CFxd=b)}EZGmI&_dPkNfFb=h02Cf0kQv^$_}XWZwBGGX>a&V z8OTFA!yq$&SalsF7J}2|redHonP*7VTCCphMO}1jIkVV~Jv8rzH{K)umS7ova6T#l z-w_8fPi1!q%HL2G#20VgcNiVoMr;R_eqVC+==z)6KToy-Px`^F1i|mITX5YM>FY-c z&n@hra!OtSl87FpmM;L>@WUZSCN3Mh7^*;XG;qRbT|W8SzkA2!s^@;oaOA8fd&YbC zgTvT*095B^qBM&0Yaj#OA$<0ClfhO6_OCO|4L+0Y(gG6mohMDSQ_?bz1%4dij>bKK z`3WQJ@Gqx#&wsecL|!KzYr0us+e~B9HQR+(Z`%;JvEk}_3QWNBhjvD#DuzGjNTXPV z@dM{zJL7XoyB8m($0DpXbpQ&+A3dieKw6L5J;wV&B>vV1lvpNjR>e@@@y zsa^E8@3%-q)P41MP(?}oCS#1CvCP!uH*8x@ok&~mP`FCe%AI6K=^?l)M z;lCFyD&OXgKdJdRN?b195@bW%k&q2j1k-3U=m*4;qrv(E@U7X^b*{Z*)Kv+jA*XN& zuV)rD2n55S{!~ej`9E1L=xh-h&3YV`*_~TEAVTFXKZ4q-Rd>uBWs`>aet&d*(;%PNJ=XJ)f? zz?!3bJKxXc>umkLwYDpLyZe3q{k|4>R+cyT^B)jWo9D)&`x)rA z|F0cGtp_SC%lmIeQx{uvoC@5#=(LP?G1MdH{=;>&3d#fF&j;-7oF3oLr-AeEUs%0T ze8*VjO|X`QTtu(Cn^l4wJ&~!#DWOv?w?kEI=NIQ5+wAc=zpmwC%e=DF*X%x~T<|*k z*8rve3UY^($USStp)go+UvhrqO|uy=$nZm5YRAOO$`2zBz_#8NVKdLta7M+?4h=di zM$O89ZYF;?tiuEZ*BUje|2jcP#%#hS$lY_vCu8^>FKKEdJMC)kJX5ih{N>+6p5lq! z)8>NfD>;&VN~vzZ@~D=G;6xmh%m=KOdK{u3Uo-w8X_CQ|mD(6~*63^*2ib9CRTMKc z8}tTQW_(@up|mPIJ<#t(mrKrRSJ#{z3gA@G$P_GoZGXOad=vlgLahIuJx<|KK|m1s zK|tt02te!|T+RPCYTAFA$LPrw*$B}8BeMG6YV69sfd$%)+9Z*-AOkxIMP0zdT2sPx zS-RBN6N1l4V36S?yf}+*7tKASTwDL)cXi_#!1;V9Y8(6=K!FlXx3k5*_Z$ABS@Dps zRTLL;#Ee32s;&jynz>p47n6DUXNyJK*t_;;`v+sbLhp!PSmeU2>t-RTjn3U!Y>h8+ z;~Dp3A`S3(!`cR}BjwYqf;`=RoM`)2FycE~6_B3!C-5Q5vf`~tjj1&M2xb09on!!b=qp5Ig1oIYrt(wAQ%JpaWZ|W zC|{^76afsy^Ttq`$p7%HAKLf%arMw_=kNMQQ2|EvBrxA`E7uo1Y>x1=LpLYF@>sF? zDyg79b>|1L_VWEn&Vm;HYfNUy_pFy?ZsX!^`~{mO5-E?%gNX|Al6^D_pr>Bhg<7#J zg;*h{Bfi7x)rvoBeyrc2F9gwH72Pact!>+s7=alP=$_CUlJU%zg_uu+*pe$r1QFmz zDFI3uR~7_^!-YDf!}W@RL$GITII)F1Q}_eFP5VgK>!DzO448GCQY(f0_x(%3NOYp2 zluRZ$cT{9cqs_-;ho(6Uz7V3~-Jqi2GnFbWU6|}n{=8s=jSGN&22s*9;&tYM{lFc$ z%raw%n`(35DeRedGw7QY$ZW}Ii7e)&wg3J|E* zGb2Bv>pzw^ejMVo zE(9JTVk?9U32?zwpr@|v@ZL54T`;RcS z-D~Swv!sUBHIE{xowlm<~`L-|mj5;~%E66#uE3`mjeTKof$1vxs z+`t>mP5o7JWk0cj^wB&(Q3JMmCvYC{6aFn%CgW9$I|La~D@IS^|9z&HV?5!K@TV*N z8{_Hb*O~GnTfVhCfp&*}_J=-W)p!Exb4M5TX&b}6wC3-frmnFT^EiI-f1Wfe7c?Aa zqM~iCD=ba=gEL^GlY}^c@wOiNeRq5MEv}M`r5fcUJc@bZ29%Hz76T~KlF!Gs(_={P zy=83gHI}{wJxzEctTdfw?0ZW0t3eDaZ|Gccz$? z68rX1j*3h*(E3c|cM_l&OZ{7kqrjDJdTM82=WM>0rg`PpTgPqBTOVg4e{4)$=ijv1 z?7NP^s8%EF47p!N!@gq(02L-50{@U)y8-K7aNx*lQhb;ShFLVed)G##Mu6jKqCLwS2du1bEFlwNZmy7 z_pC$sCH_avcpKeJ(_^U$XzMz}T1+^$4WE|MNA#;}%b>z=eYWL_+x|^Qfj;ZV*qJ~a zVNHXF^O-acgc6WT^+1Ph?vm&Z#EI> zNK(1elVxjc`OFk_qZk&I=A|Pv4z=1ouMw|z`@eT>=f>35yphq@nRzn2 z^vC_At^56X_v`k-7}>^!1Yqq$&l?jnI~jwpoKKF-nv3|2dqya&�$@NcBHM?p{qU?~POa`13xM^7NfrPuJAqy;V2-2PlB=TkVsaFluevK8vVzECD}T&2It@OwNyr^2r21Lo$r<;B1P50B=D%a3T%`AGWau~69< z^~k0Qtqe8uV}GUk#Si!DlBuB$dU5NJX;pRVm&Ei{8L8tG7mpYkK*jA3t7^HLn!+)g z@-*iBJwps-+~c53v?(l2O{ScwDfY{lfZZXNd*`Yy(^I!*tfHAf1L+CJ5TsBi56H}6#>J_J=HwC zeuuKH_Wmh`)T~f^`mYqE1DBpUu0m(cJ`Cn=dRxZcI1qeiZdI_SbD8o}HL5I3C+=n3 zlOTxygRXN5&Ma!sb!^+VZQHgwwv!G!`C{9)ZQJSCwry+rKXq!VrfO#2?5kbtcI_9> zyH3U^8c`006W=lacN2YnpQE}_1gI7TjcnJZ@vg=t(K6KXDM1sE6nH2f)jl9x^E+Y^ zh+Apcu>ofvFkBT-%JnF`;vss+FSYH8-WD@Y>|5g}e-ts|Bz$E!C|rR1JN8Pb{E&p; zq0B75>R$HA#czlaMvoKzu}>IO)2w)BWF%+|jRS4nf3^Ub*k-p{-xLOd>KQc}tnpBn zwOlVe13U3iIhhS$sXlAiH&_6pYM+L`2{p^G$|x;|;6_VuE1JAOAxR>xhS2uJ_&UuP zb@q4CR$`)|IA`FRyB<_0eE-tT$*P%qdw|+Ei11MQn|k_W7ic@$uhp_7#3g%RIj`qf z)Y#`-Slq*`gIDH}#@G$!ww0qB0hvZOFb6iAV72+RN5dru&m1 z|FhHluZ2%g4`hp%4eq3VNe$Asf_}y2*{w=wgAZ#q%&p0xjfi;v_Zq zO+7Mzj6`y)GMK0M^?Y$kPvlEnn9m$FNK!eUmK{9W$k`sERFan8#8N+ntLmJ^`--#$ zT%P{zRX(5;(7%n$5bkYz&U{1h7J63Et2^^=oxMds1*^(#ey zHktc^{m|p@H97y_`yeoV1g4SjZ-@Cp%;N-5LdGSThq&ux{2N*N=V8P4`#r`ja_~Vx zHx?)LTaEZLcCl+Mfzj)?V=T}QL;@2&{CWZ0{+YM zflx5>f@+Wlg5Mcq?qkk8Jg*x34KL$;>AvL?+@zuKr>Za4$^o1>Rdej)5LJSMP~oG1 zO)%fgHD1n>PGfVk3J#{rk}A$#;0XI~k2}V}{8ws1ky0bBAb8>Z5-*G=`Uh)f;;dU| z7$K$MTEoWUksdm`B`-ekw$?Q)g^|(I*Ixqm{K=f3k5-?qy_s~(cs)9WKaV@H0ud_* z2PSuCP`-coF$hO)ViAAE?t~?}?>`;_nqRNKG6xs#q`J6~Zxj@UOA#eYP3{KA-3f=z zUKvaGPxaX#q5y~%shBFKk4K;ahI}4R#=m;b*YUV&sAJb&lK2=iW#ZO~1IRDk)$c54kK$##L^uyo4qBT3X}h-1c>3t?kg^u9x>ItBUN& z6sDr~1wgZKpn%9iC_N~%{uw&9Y*~-}s98=#yGT(U^mL00VsPL>T4s2Z1xr?3o2L!; zb3nT;IWn2=#g%`qr(9CT!|UY$J7USQ;C#Xd`)VA;C0B-JzYD!nR(Q$WbgKwsyiEjt zXF)z!=}<}PAKzz?MC{U*I0H|D83C_~(EaJ9IVnMB11(Zg#)4f7MKqjXF5;_ph6pl- zrtl<-DPd~^>u?r`3a!6o4(T&w-7SKS@hhsPunQ0t0?&lpfY!|#7)ucYv{1wtx^$C@ zQ3k)53RDU)NMxI@EEL`ksbRgL{M9kZmZA+FqHl6B_*E&VOc##_g9^V@)6C8tnH7UD zp3krBq{LJ0HLJ(cEyo5*v!`ZlVCc&Z1d~5<i@3o!(`s5m*MR4XK$$LG^Lc}nE4TxKuWtsEA#du;-k8Tt-3Mei z;=lU+j{TjQD+SC4ctQJKTNAtH!-M+ynLF63{Y3KXa6M}Kcw4T-tRV^<7=nQLYI^m( zhgp)Rz-t6Pfzo@*TM_5Qd1khxz9OQibp|)!IVa$j^oYPPdo%`sRQLrAy;>UY5r2je zNHd4R7Z_UQwyCuF-JwGE1?X?C<1LFe&N>N2rdlKJ-@&?|T^rwE4r9GVW zrK8IMpMb(-q1=c{Wva&At_&o}As*Fn@nYP^g~K2UUGE2#VUFfx^A0yREJO530b@X4 z%B_96;aNanxd2xOP#FF}mI40ilY+x2^y;fg&N3-n4UmKqz@dpk`~(W60uGBq+fk&+ zMh+kc^h(00&G7B=`DQi829rIn5udI%?mK8JdrmoekVKRaV}@JpE#~zaVIh1JZf=@V z!{9KI4&_o&EZ@s-L)@m!dW5l4aKYHK;PU032CHfdsN${MbF_r+uHa2-v$?kBBK_br zDas0ECDQa!RSdI;%nEebf)+|C>GK7Z@^G~Q90{)Z*A&$B&E|X*uC&*X&1{^+De*3! z1Hru(mI-!9Vw-~0^K_3Zk?OJr_U7n>OL!Izu^iVo4p>kyAT6;~0w>3~FIT{vYHLg$ z#Uw;Q>W?q)y_g3mFc;k|UhX+=b${x2B;YrupgIeGvwZ!dtC+C_3Vd;h_}u(pl>0^j zr2SI0_WuRzo)y@M?REt5pcA$5-(Qzy z*v7DWop(R!XBm5pEis^5EC7j#{>$N=B{*p??b4KG50^S*FPVF z1RoA`jzlt@iQCbhO*U^-m)wxQTQPwFCI?G=Mc1{5jVMrrYkk2<_-0xieRX0}IFIpk zN7;#^47x}q3}^K4JW7mX;E*@P{5_XP5N6}vNYXv99E!Z*W~9|2CK}CUo!~WqRi=XX z9XmkN+d)_(O4jFW(}{1U5iy*_IZ9LLazhg_`odS0S^C>B`;1QNqtKS=+?|L3(2^qJTu*{zVP(_YAfZTB8}v*JQ+wq?f;fV zg$SPX8M{TwfI>=YM%8|I>t+iBV3Ck*C$|ku&7Pl!)+A$KKh9ZakLzz)E(Lw^Fx3$cL-&HvN{&(yFi@{A;&`D$}j^@xnZ#Y1P9S-rgh_6 zI@5O$82gY!qp@?FX9w&{<-PR=RpE}I__P$rloA4yvy`JNk)hi7D20VA*U}YgcHeqA ztRLwM^QWbAqK48$nw`3d@X+~g5ek(nZqen0r%|d2LR~#i2F7zg20d?Ohj3Y1;nm_r;eVL`QbHKZj1I^frKELM*eOj35&_*H%k*fW%U0cv&Rm|X0vMFJ z+tGr3`zkgO#5f#xFTG!oe*bD-v*aV|cOFe@x0nzx+}f9_R9es%Z7m##*q0-Ct_`1S zQeDrs`a2}XRv8kR{Td@2FfKa^``2n?Ep62SHKQ`BQplQt)4@Kwg6U zwFMGFrrjM~T}!GEOd3;N;;cnP)sQ7g8o8!2nUjv0-GRbp)Rs){J3t9`@VJ^oh6S%F zga6R$jR8SI2RGj=oTpvcVNS3wlJnqZbd>VuNAm$RibCj^-Fz|O z6OrlQWALDmGqjZQYch-;t9px3O?(7q(&0%_>#Q3$rISV5Q0uy^`s&;xAYZ5U@xn`? zfeGkGTcB?Uxn5647Oi(fF7*|D~K1{ zjF$xOUmL!Qx1ea@uba$fp#t432Y}tqcd6gK_ilE^Gidj1wu?x`>lV1|+x?L7T8u=7Z7EQI&^EHY_j2#mR6)(tfA zD~1$S^XT2e@B5$Jb)yNlC1;7KfLGN%m?#zOa*r5~YMUPH}#h7{zA8DWEC zDh}Xa>w%L0O5E{8S?q|B$#=<|~Iatzbl#+$M)3=c_VKDUYTL87vntw$g` z`EUU|ZeFQ|wBVyoj-ZA1pyj}byL(m+i^duc#{jr*JBycDcaGu#OJ7kyIaysju zJaF0!T$>-QGO{_;b|7{O8IMQywtD6)U3&oRr`C?e8m&E$VbkW~r`j<(Aq zXEi-Dx^zfG-D!Js)ag(CAnv;USl+D}_m^j1P4P!M;Qgrn2_yu-)n@al%ue(AD+WNM zo97Ni66@>eYNht8n$^tRjTa|I%8Yx$t7mB{`=Pofr?>G42X)o3s$Q+9qi={@R^Ylg z^McSL>}BNgx9)v`NT_+aU+WgQG0F@oO_}4Q&P+Ep5z{)W>CM=Ti}1r{qqAVAY`ZrD zz>Qsyg1lHuu_#NN9axcmItCq6q!%z6vkTNf$dY?1Wp;i1ewXOe zH6S>IIXH?)=nE`Jz+ddv`MV;!Wq*cE1}I+k?A)|O^AAG1l6|2a6OYk=(w_n_=+rk* zo$`lw3N?_sWFHiKPEPd{iFE2;9KlZp>iVb`{&y16TRKPfpZSPYsNrHwp8fz>F=-mm z*C~y{wD(#~n>FTN7@ z7(vfH9`uzM?H@i)c;(8V78f*NJxN}|NAWpJ*t8GgAb&TT0cU8Jx|?D6USXGdjZ zS)VdA?QHV!h43rMXQ~YT-yR4G++2bRZS67WJ-vdqR7rgA5gd0tfu+uM|^B{r5)7|B8D zc&`EC<`8IA%9_r(e$3`0l@4vSBGzBhR=@UnN8k`CGTaOg^&ef(m&*^>)>ryCF=dj; zCehC=Rs^($YWb#;8lqh&!y3C_b1bu+V_#be5>cgB<#)|Yb~~5rryLJN;^OhyDP0^qifS| zfJ&z1i7Sn{Z(GsZbb230?d3ILaX|RzF8uyT0ZVkrs05m9GUp|ogNSON zc73}R{qzT7>wyR@KhB}&u1?FwNbrI@T9JO=TUEW0L1R9)HDE}klVMSQM`dkw7bq?5 z*>}*XObCi2B|yC93dhgf@bCksA;Y{xOTeV_@v53=+z4ebLhhsd+N8k{Zu%haA4+W|%*p@4Gz9&;cg;15w6z`rO%?t$%~39*2= z02``3R=Dj;?C~VAXm!5;NU=x1!Dxs5&l+8V&X60K>?tayP7FK=?!6rJi7_Gmd4A(< z`9!OcpbEV^Q=G-1RpOg)>XR2&O76TfV8z;A+3K+A`|E}M-v#%yCu3;|Vvo&fjrb3G zcR=Rw;b2&%En_e!px_7?|Mn33Jib4k51uf1S<{ynURP-caybA{?!BjhNEHozCGat4 zMLW%+^wXJV>&|#zWcJ~ulO##&?3`+3UWwR0b=|r84$>FbOuWX9nx#sdHfL?9e%u1&}QM>!=SzcAw zsM}g9umko9aVyAn4ha#Z8bYq^n3$u)pUs$XxvUm^AcEPH_5K=*ULAWqo_th}+&iD- zZ+#qwww0tAgIGB9_$3MA?)1n|rK(1-S`u`Xg_5afH1a6gir-}R!?Z8~P%{n`K1l-H zoV#pQ?n9g)Am55n9fN`>q#Ojq#2>tHAsN_LWFEd_Z$HZM2rrONV0<3jLKyx81qUgY z)QE(YMaSF;$DT6#)YI30B&a1CX2>5?qMHnylio!`e^|2CnE{y>bD_cL|Ba$^d)~*HhY?}P(@IK zCcG2%7u)NJXq^VW8uqP|ZC%6UxJoX^~pb@jQV+q`U>t!QDeUTWokj9U$ zOf71L=hcdzSLahEcAT_=giNVJL7Pv_v&;BqKb);oIsF#OE{4TuaiNh_YcH?P^fFy{ z?gEypkQ+Vm{*}9%@eLyJx`LH zDQ98TmPQ?&fiGdoly7KV?WmG6Z-br+5dZLSAFJISJy^FnQ_|1TS>f0^2jm-C@D{-P3iz&9^+6qyMA}xPw92KtdA{T3TOo8ZAm@W812S6$^ljQ` zr2Y2Bm^dSNATbGa_?L>#71iGtaWmU99KIaotiWvoK3Ug4nsQV%@@;tkevYrS`NuRd z#wuUH6|mOSVp8U6ONMvNMFFL=kdxc^>&o@DIB)G8B9!G!8e+{gpK)O!9h|4 zHD$yOP8nce4p-mqIYiTlYxX$ws4Q~%4a&%TufEMe`s-uKPQ(7fh9k&eF9`-B39Bx% zheT*CW^{K+&B$xhY-3Ezsc5f9iR0|bmEhSri%qgnrk7|lS~@geZ6fLyz`?sm39V&} z9$~<=iFs-QZ3?cY^oU(YkYS$1%0w>T7r1CPnhtyPbKpuR?WECO0O$9V^A+fiua-A* zja;}#WXE&LwInV<@w-4pltm+IY8w)6-BKnnjQT>2XrG3|a|iWGgEsBF>$vnvtMx*s zP5ON?)mB5fwT|0(JA>5NUO*R(p(}LtvGjP?6?YSxuUAp)GtJKl}g-s4DDcjNK=Z8hLud;-p zHs^)Vhab$Sh3;Jk0Q7o~PSKmZUH}NO`gEE({z)8CN3^hKI2RG8aRaEpG)0RhV4rbi z{Ncz4IkvQr1Z^7}u5G4+>)MTagfpGD?L0w3?LMOv8*NBoi?Lv01-<{qX-pNst z7ADWBU>t1g0o~)={wn#}>+xo(SyiY>8iF_0h@6~X>LhajLu85ZV#*m}w&p|*`t42A z->&MG5hzGXHGIPOGHSe%p8F%E>1l8v?h$AWj^WtOVS@?K^PnqSz3fnOkU=CRMpVYn zEL3M9TzM0`KI!0(JgKWjKn8J4W0=R56OiYY(mshfYY{nA9cu%kzw{e5Kjvt(xS!Aw zMQ^0W?>$ri#}t^1bYyFG@nr)Lc&=M&!q&`HfyVG~uk3>Y&Sr+N;~N7^oWxgs_C5z9 zc6LxnG2~Dn{ofLFu8!h5qoz;PsDQf3#Qjyu@*EP|yXpb&WV#8!nHbC&%JlDqf zRbue;A5JcH!L&P38W{;*w&8}PihtHlZS*^*^OS!B49pgdy{ZVC>`x)?$O+g*3Dqmh z8J-BjxaWwbsg^=_A;9}GGKwl7+WnMQnPp>0jp9`2ADNEmY?iyjX4SIuN`&;5TxoVw z(5Nc=S&ke3ur9IFDC_P=!*H4Z08bCHMp@a`lqm`4jhYP4N#=xbQMLSxdola(nt4q%+fqCjGyXa>__b_XF5YMA zI7-Lf;bZ}`Mr-OZ)vEbF$D5MNT_E0v$WRh#_0pY^w3C(k26JC#5_|8d+5D^S@T2mN z^OqA|6lxAp2A%>f{T=C_0Drr8>rUOl?sEyR9|-b?W*m`pLan z$!n!u)$tDPwe2bypuJk%?W6hfikw32#<*?qI;5o0{H+xES7i^^pVMfIv<4n+s{BwA zJxU(8ZKI~DFlQq;IPL?9;Q)LD7FD5yplWd)3`uh+q!+sfNp0!@+Ix*LN9^>HRB#T! zuKSVhOeXfdM$K~66fDQr91wl`kYUh0SXNjiBKd4c&qyj(Li<<5B^bdzzSeEdf- zKjuh&RerK42m+lkGaEEkdRDOa)tDf7rV-52YLqt=hbBSvHXSS0*W6zdCtRoLzH?G$ zU9%8l0bNfz1J-L|Eygq3K8DD;5>;@3b*X_n87CRVT&(EXoZc43n9MW+zTLX@TC(z5 z4Ca>@nwp@986L8Imt47F1cR;SbY7_oL&E9cc}9}?tOQqfIxQ>iY;bV#bpdEqu3_su zgjWq5NNOO0xjf(8H;-YaxGyp?UN;N1iIM&nkXLZCj70&%@Qfr}KMkDL@qixy%>xsA z*2AZcN)@x_F0=tQF>q(3IGWK}yEqA#JeX~k1w2$0n+k#TW-pXqhq1SbkNS!KPq(C@ z$kVjaIY@RePejJ?t3L>^1crd|3?5I)ztF-%cS04TEY1$o{Tz zbU%r7OF0SR1_Bv(bZao}AzC<~9BM%`4^!_pGOA=fTY9|{!Umc?Mv-^g;5BgI%{^zppZRaIC1>DK zsG71%2N)Q=zxGrQRCwI=3)mR4N9ps2BEsfKNqE3LbCmDg4tm~mH0Z-NGTbl^HxS=C z<#n&B$ONa+xO9};QO;5#CR)4vASRd_HOp>{1Oq! z0n8?v>{&KcBa2CwnP$D&c4YrNW~ftegue`cEn5L$G^$`gp2HAxfW7i$<_uJMZ%TY$ zBLmf!_!U>P7TGZ!ZCOfP$|Mf;UV~YVNCy~Yazuk8Oj87>#+*1c$AgU`l04IgYVL=C zQBMWe0k%3yaz1bYPs=I*L-Fs>$m92Kc#Cx+ zGL2;Yb(L%YT?8d8p{a-cHkX}0>)boJg!L~CiDcQ!%Ln>Fi(?f=ScI$F75>y84rq88 zM@vn?8cdKH5-5)Wpl_{0A;b9I5yQfWxwQI{d6hHpC!ILFR# z*{&C$Fs+v`AA>AY1w>Y2&zVB$!tvK9$J{FUaZHYNY3mATbClR~*QrWh4?$b z$_hhqLEJTSpZ4?_LX4e1HqwR+xY8n@KtEl)T0T4B9}?H0El`t_HI$dZNJ)FKPNBC?kn-MNOdW^;tg zYfv+I3ZB#af(ss8vbq*lW!F`xfeKX)Bb1lNY&Nn4B^uXzj4F!;-QS$<^hFrJGYI@b zeu^bnq0S@ftF>6Nq(x5+=`^3JYb6^8f4T5R?Fc)Bq|e2lT3JyY+LsGox=!=`Bd@f* zX4m;4+}g%O9t}@=pGD_snG4QQDNX9<_as$Av1zdzhUNO@s)3kl{ZLL=d0NhkBF38rL_pY9(Rw@h&W$q7~CtUb4w^j_&EhM(%{QLIz!ZjhS0FAVflzbEjD35>K+Jp$#+o7oE@@!Sw4 zuOczb2sS3}zuJ-Ceb{Zzbpmye1jYFtL<93rr-Gd0gi~noXMrB7 zdF_E$=ze9rvULkvyAZ;46V&4YtRu));ZSr2gHD9)L7Kxx9S3H*;lCS`IBkX3W3Ige3fF1u|Tq6{m70fR(;Tf*XJF*5}G3 zku>~I^z!h%2$mPls()R3!3aJID>Z&^1;1Fx*6 zsxJ2H6#D=KUq}bq@PfcrdY1_=*DftD-z0)h`kR$^S67ed<}Z$dkTrUsuvH%*PV#!w zkw~z~TA&sId(WAmX+?kyt)RraTtXY;35EQn8u6f!7W4Y}#=(hYeO7gZG>FVrpyz)T z29D%Ccfs{etj5GyW}G)Q&8EY!!(unjs*4@^C2j}!9XEL5Mt>;3H=Cz5MeZP>X3-&} z7W<`;4yvgw!PpVY)jLU6Xuge2v$@Y_G7mJ zIp5cLIay$}9?@If49whma3_=mbFGtt-1J`ZE4QT+MX;u^5Kv2TFx(FT?EMw79aeks zZ6E^dC0#+7l{NPK7F+CFmy?*PVp~ddo?y8Xq#T?b=p^+nsXK3T3SGSbEe~t{D&8-Z zO+f-KuW~tNrTQ5z#6DN~Hzb`9!5z%A3nPa$8kcGo5x_R+pq7e{Dpc2D!;?umIt=pk zQV-OP54_*~m^B#8Zq!_5=*xIrsfc})S{@3xMBHQ;(ePJ|VE$8QyPa_5RE}dHn^@ zk0zP?mPNS7SyZ#{CmgKwSJl&D8e{%*y8+h`V9;|yTRjo=ZAEo>32Q_IqTej+wHhQEq0{z0eePkBE6$-C6 zl!#cNB-8h2@q;RZnZ|2(6dAk|1LnYz1-_bIY4d>W?>gD)8d@gBQuWvq<+kiaa z8@(`(}va^QeIj~;VZGzSHK%+>k zMu4iSnEV|dr;FMlrh5zE)^SjZK>kRnz(3ig%Ney))YDR?LsP;H6c8@B0ZP1q+;2Z% z!0p)6ly|t?9STEP2yKeEplBx-PQ)@RTh&75F)bo)k`)PQwj!A#%S)|6E6$bc(Y=&p#ySwL#7v?KjxlY_!B9oY}OAUI3jf{NHSU)V8ex+SVHa0cY) z$w2A$EJ{=zNTV7yUo#8w0hnUr3(iO>4!t?4BL*S7#*{xhs?>l>H_YRs!R<#11D zfoT#{D{f#d?-4)Zt z4@iAMpl1dgiwg)Wdo8GQ!zzC*sKQ>h+1;fVN12xs52o7K2mHE4FYTnMzs>WvEMTAs zr?}8{8{yC9j%%P_)3Lv+mE zDxn$Fi5|bBZ@AM+IaE|oj6SiO zZ?lME5XY#hs_`*;do93mDUyV<9lJVb^twtQ4?=);9ibP{gugIz`NgBC>joC->BBT7 z<#OSWf$OV$|C?qX&o}u8DIc0x6A9^&v(W~dAxvgskClZ=N6(7t2<~+e!u}=hz`m&U zVC4yLTlD)&F7GOjs@Z?+<>oqJ<9u06z_6tdF^-B+x*x}^HC)lWI?onA7_5Y`n~MsnaJ zF}9~@MJfs06aDHF1iEk#mvk7*9FIT1EccWgreqN4BmhKw(c*_Am!Xbpmlhl?oY4}N z`lRUA6?{=l_@PwJT+;G_$4bq zAM!w9XvN~;;SA)3JTv1d?xtj{l&ag6r^=Um+#*ptlti$zg2pk*pmG+bvU$r7w)6Bp z9_^1KOF{NwZK!BmjIG*`>Cg!hD1`HCOJzVO2b&)*fbl3v_fb>UXFYTdl+>fCb0LO+ ze}Hi3x@+fgEQ|0yRuG{sQ;@Ql+`$o0mWlyc78u53_=6e6u-=E&!ymOrs@h3^eh^Ez zr|at|xV{?QqOPKhiFje|y7RVyTl9b#fHEkLL@hk5}+VSSUp76g;}Pna;xkSRSF@h>yv%RmPt*R5lf zFGq0-h`UwoF;}6M#>ShB?XFd^NV z*aHl)JZMwvzid}Ko7Tcd52F263l)M@(FrWwVQ;YG+Ay4e0js&GF}i@GdlwU~iiKrV z3Z6M;IfzXDaf`bl2Mq&0J7jp?!<@a}U+rDsVs@f{0f((CUWYyH+9y(DB znX?J^NgNBARXuZ$aqIxQpU;LYnbmz)pG~zfz#^)?Fnsl^FUt(Ka5DW^$x+N4AJqa> z`}8*;^x4b`)&j$0Y1d>B*wk1B!S#(F)va_%#~W;|pz%N=thz)0e1F2F4|va11MZQl zdQ9@7Kp@QSG_nIS+y)R2XzRZ40j}cT{#?4Z3SZlgA*>B)TlI$baM~r?1_6aC2wG13lEK@^Wy#g*rNW88=M43{=}yeYy40>g zhHJ;zyS)GEh#bJ^#W&Dsxf9-zd~U{=iK23tFAL)?gXo62IxYgKq}aqN#=@W(ycb3p zejSBW7)JZ}9w+TvqK;k7jAcH~2PbF75G&bqpm;ru%(%`vAF2`T(@I1UO)ve;^_tDW z)N+Iaiq;~Qu1k8ry-#i)G2D=v`tlt5Jps%JWiGXi%o$Lx6doHz$2G=k7}%}Q)GVHe z=FUTB7E5gtt5Ix0=gVL2IG`N!$?Z#nz{`H8r@c1pD;6$TX6_PEUKI{eD`W(PDa&G- zu(!N?<;-mJ+b{WR81rX#s1iQ`%7Ks$dbXlg^)V>7wIS*mjnCU=<|cN1|ErHtcOLt5 zyMbv-R~7&gkReeRa>wi$xwc~o#^B&0#JGSG@9SIQT5p3T<%ihHH4tIovR}EIxK$aO zrPLKOO%C^B9)eJD3G^1g8l26@h2Vlf@Pn)C#ny;FGK|D6#-m&1b7>SPPMewQ&0CLZ z8M*;uyQyH&dhPUu6)%KW%{E)nG|fVDckS|7%})sG7d*}C=b6~)x`Df^9k|tI4!p7nKt?S8xKKDU?k^y-fMyi z5gecmJn|;0*Oh+Q$jc$amS4E$5nX7*h4Z&{w1mLcG+|U56Pf(PY+;j+r;ya*K(3g- za<_r8kRyd*qm8Pl+Y{DhJt2?dtq7h4B7OYu=EWvYz};gq=d$NJemlU{0b0i0-PeYA z(EW9{JImzyKzW3T)Do@@)%zek_*WhMQUYKyfX^qClUz^c9pN&nl>MJcY>fNX6uHdE zKYO1LRDfdsg(B!dJ<_;jK8}3BHEI?n#bpLw=9ms&F>^i0xP7`g#>=k(PT_ks3L&87 zB(u_Rn56{W$o+a2_vzNklP`T%1yBT{LAmp;xBpgXI+XK!Sv594oFc|M#E|P_ccZy@8PBhRf4@^3gM8|F4}C z2Y1-r=0Dg?naHBO&}aRjKYyY*cL5;5Ph5#%p&a{CCo@4?81ezA(*Dv{QpR+!8NPr- zyF^@IEp0rd{t6us`%AURyfZb|>8%(Ef~lMAI@ux_%95H0XKu1>v-C&8FmxCt>OVL`%QEW`5r@wY7Y=IzU3~8 zi=&y-&ZfXqi0r?bLn>5Cm7zMCVVo?^Gi!0vucl9&{4-_1#Ky>k6*Cg@KwN)2bgR}>5g=lo z`T+7lm7WdFOC2J_l(K0uwHYRyux5;CvJN}!$lDdvU6w;P=J71rYXBeb?8}tC2`OM! z6T)0%ru&aZnPqE+d}F{Xd%KSb2{1z`FoiZjM$+e0HfhI}X(WC~{(mdtz+I+)6@7t- zlKIM6%$)e>+^`JGDywEcJW{HM4PTC9QH^!&_@g4raINJyV!l|2gM5J9*MfI(F(xuJ zA1@dG$@yhFI8{*HodbyB8leiKeebX|NwVc3U3+w;N{v*QYzS<{jw+a!?c7E=>4Yyf zEzB)B00FLi^bnc(!k6g7k1_|7(U|rl-@99Z6>6^jc@g}9wU3io15c)z3RUrjuEZAu z6AU3ESl}dk0WmrC|1%&@Po4wcKPofwXU4DXOw2&0G2bJXj29rfKBkrtI3y=%=ZX^5pSj~Gz3%y>-}i>k_j36 z9_RpQ{n&E6HN4&JtnIx*?DgjA0Q31Oy}FVTm6Z&w-2r^wy`GMqP}ZIIUwpLjh4MrD zHls7FJC#Y`|67hPfaFCaupf+dpbLSU*QFF;_{G%0P4{nyDoEW6FQ${<$I12eacQGe zlec?_ycnF}fhR~^hO&5NYlBnqyY7nwpG-U5meA?mH=^ikca)Q{h0x8d091>SHBz+7 zPZ@6B;~wxTP#eE?xFaSdo^|CHdvhl|U@7%R*1;(^aYkm~%B+Uc+C|bbHcgY~^2fX* zE&J%as0cA2 zz~?q|ruR+of3O7ffd2o3{hv4G|JLzORrUw|Z#KaHBMXrKhb%D6KGrNv00i{Q4+u#6 z|N0Iib;2K*6YviJ^dAV2)o%}2AYia3;D4oDw13B4>RmW*j3s~PmYzx_HxH4Q-_ftl zM`b0@dlp34ahpsjl+a-=fn}IV5&;zgD^D(}9T7jzd~aB+NscF8w!8i##S*ROIP$GU z*o~9Tt}L9~XD^g~g#HT8mLd;d<3@7;kPji=4k}6jj#xBrT`d?QFSvAzczwJ$HH)Ln zsP~o-1@B1{)y0c!VQ+mjKg;LmnyJl{BcwIL=(5duvLifG&RJ-0O*s4=>Sdy|PMs;f zRVjFU%NJhfpGn=$6?e+X{vMyN?9-ti$>HAh*!$|7X*~_IR9=&ZOC`~%Fi&6`C z=TMmd5{fUdBGE}08`>gdvWxAZlKY91S=D1jyA({JP5xCNC4;)@O`W6VMW#q9OIWRe zbz<{Wv+_a|*x6oJo|b=|IuS>j{US?`x@$(0n0^q0E~-{8j#G*VSt*SbPcG48ATlnO zqn0a{4Spc6=rb*)Q4bEXaKcb`Nlr;PrWw@+sI!2pH0SxAE9h8w`Zp;wszf(l>Zlcm zlv9badVZ*ttn6!>$!MjC@MWqCy{rUBup>WjZCBB`xOh0$Y0-tOIP=IaA}!w%1xzE< zoCt;98^X!Mk!%bR4NyR|JM05Nw%v>~Q#Z7~lClT6(8HFil4Vd7h=;RDD7unen2}Ed zBGKa}gsi;&i>`C*&P3_9Xl&cIZQHhOTb(DiZFg+jwv&!+cI?~d{eFK#jZw8m&AsQ! z(-hCcHV@7)$#OaH^p~^DCZ2d25{P(314EtS3P^DpM!Zn~7`4=gw5|)O5Ve z!g+;&GLzZUrga>qb+2i4sMvAN3$|kyuT+}{lhV=eY1*tyP32DkGFHhwt*pMRNp7-l ztxG?K>r_9FWoSej+t)cOM?~>1c5bn(tl-Q#^j~hxU=(6i9ubiTs1+K$)pXv?X#HvC zTVMv;ovz$)ZxQe))bzbJV=*+;2~CLoDJoiAxx~6@LoFutc|h&>Xzx(i&c(WwID+knj;%=kp{SJ=o|m6`xk1Jy-!{ShfMCCEQyTT zsI%6pHhqItc6|R`se=T)Pj>@yhnQ`J7wVFKWGmM#UGb}S+K@!e#6s@{>6~snD5t9& zJ%pB8_Sx@HFf7MS#pAc)OKtmnRDOLd_@!mSi1S&|t-tZ+lAlc7dK>F&?J7#Ywm8G9;8kHZv9p))=i=o^IEakTW=aG+3`&#ff3;W^CCLq;KPrIL(rI(MhUc4s!pPfe4xaR<*2U#cm)~@Zd$Q?sCDHu9vDBVXvo1oiWDO3>Yo{LS!DlqnL?ez?5(N^$co1du zq=##2UJkMbO)*_yj_>stiHsNJdTfoR*)hAlSryO+%dVAsqQo~>jh#_JcFVfjpU9OE zQQXkeccfsm3(~#|=jy{f9d$l11bR(7Fs8K)KtOi6bu_Di%+|7nXWkXoH7j`}(7Gv6 zVlfE9g>jW6;b<@&P4U_cGQxczg$AZYLHsQcv0!Cpl#lp%C_K30Rry7+4FW931S`P6A*+7 zFo%-6%btRyJ6S%zzih(sYGpL4c3i&)U6Y}x6iOPp9pIWMvi;C5OTKb}46)0kqT)zf zctZ1YC&Dc+ujWGQQY1Xw{V?>SzDjlm3GGEI=n?8g9Qpdx$bgHW_vm|#4=;-+=+F!n zlb;+A89TmG;+u86f4lCVCd-%*y$7lU?9Y+DQkR-`yoejF1BM_!3^yo%Ce9|Eo~Cjs z4Ks88NnclE=RKlt#}t(e%*2o(nz@A3qY>omiyZKCGu~dMD*GS7S3q?^6_v^1`gpK{ zsrH0K=9wAxh(B${=io3Rwl)|%e(y~K7iUDp4hS26)d62b3G3(N9=1&6=LmQKKD$(T zy>GkNO(poupauI;YFO}KK8EB62|By%&3$^(FGt(fc+tj~-a)SDt0m<8@Ihr3q5Xs- z=}f>@?*hb0T^*$JoA3G3t9|M!UcB>uqhT7^@IzPzZ`&hSKxiT=BR7f;Q1`hWFe*e?VcE zs^|0jUk+s`$yh_Yt(FE}wj^8Rd$h6l&^^Uq@3W%qn5~h{nJbJqjFjE=fM>N1hyp@g zw%m!aEf@~Sqmhj~DyaJU@zIX$!3>zl>|S$Absm`LDF>Cus0XfsZkYehxst%y4pMAA z(xoX;T{Lc1gcpecBprVVI7OGQ@sg5D&ppLDIGyFI+-NV~OOS`UR=9Cq#b}(b8rw;No|p}} zoF)iDZ*h}X0brj%kJAx?+>*kv8apuwkP;Z;aa-$K?^qtcxM3oCsd;|2AFsy37*2_R z7yz9~r$n0q><_gYGc@4%BCe~)i8(x%D>+SUSq_P8-J^d4!+dZA?=xMU9>_^;H^;Tz;I2W`NG9eP5 zMhxjCrhJTlgNg*K33moRinz$S3)ZAS^0!SgT9~ z?2&fXaEZB|i0Mbm+t{+yM%+t!@oS16(bN>~VYX&rHNavOYWR=Q{Qju75!~pk@OEXy zK?i9RQ2%DLOw#O|`Lyl77x`2b)Qm{WEJbdnlFTDhN-*_a8A%$L|`_yx}^lJGpy21Gh#EPdfw?B z6v~db{qbjAD90vB$<@NAFl7rMdt}V?|FBo}4dLJhuAdmi(L$Hv)Wowad<**V_IG&~ z$pO4vJmO~P>6yI&d3UC&#fKkcl8@0?N1{37At4he)jbKdF(_q=`D<>awk>F!<8Au@ z;L>;4{6aQ5&OmL8&ag|HuKv0MC3sWfH9?OXL*pGqPD$E6WKe|-PyzLLD4fg8t14D+ z4z2egYCLUtch6Ahk!;|RQpw=dM6Htd?$Wr#0L=1ewX5S=ZpDYe8)DGKWRp|tb;GfO zssXZg&;n2AQ~v@Mqk?ZwukGPQg~9>=+_uz1-yOR7aM4kCTmO{{=AMcfND}O7(Jq&s zb0ou4+*V+In|GI?4iURZ=E-+nqPYG!B5gv1ZLZybO*r(olokuba?7r~4Ok7>jq4?R z8VeKkKGJ(YB70+YbfvcEFZTyoo%!jTU#X@TWWknK^pjB_l9Qq@0+Smj#*mzVd;lG% zZ|&E{K|g1PPUQjmWObIs!?JP zOS&twEwvdqrpH2Hje)6-V1^~wyJTqp&ZTTuV^4?v)(>$Ma%*6gMC{-0xbW7KlpdQ) z{&QQkBInJzylOo|HD;&=syg0>%EpbdDOKfU*CH|3pA1z%rFkRUhTp$h zGt|!9pdv6x;C6Tce=gpQm({|ta>wpxmYS0S#C`p&x*k@TCV+~ctxD7LKjk*u1G#?@ zLy}RZRGe+kSnJ@bhjF3C074KPk9c*n`Ho+6BqwLdUaR^C*WP*u_#>-{R0tgGw5f^e zj`8)E1?)8zvezBzh@lriDTFmh1rNPb4J)!pu4V7IGW_zKZgcwGmE&S4Y{Wn8#I1pT3^6mGiOqmzq!hR69;*XJ@KRtmceH2 zPe|nf-|y6!4wrp^KVetBcmIafK6g$Tv4cVqHkLNVaV2Zs`BnFzst^Nd=IA)9NX!Ix z2q=KTpGz6c)9NozHlYS@;lMqB$$RHgy% zu{OPDqS5n_UZ7PeSfTc_I*@#JwS`6!$&A^QjK#I;iup1dWbRmx7Rn;$vE!TN{+2x3YxY{YM>L@E-tmj3XZ@?qt; z7J4z5m1Sy-D&6jRWrjf`**7&a1dAquV8rgGzM8Xq>~=ianPsPc0Z^M-djCQv~v)9b(DtexYJzKx!>%!@T@ z)qSZ@5H%_Cxf#isOolN2w%n~InhwXavFcg@dO3guf8WzT{2a=E_MJO@;?K^33;`OQ zlm*B7v`Y?P5G$O?`pB)y9bCnk*v6qK8Afwi(p|6s2CKt`u~esVYy?^AK4RRfz8{Z7 z8mQXjR*ufog%8muWW?(?S=@FkVhCsqFG==vf~UkCJ=)JA)SVVRPWFH0_hCi2}y(o`Ac3QMx-W#hyJVj_L;A{J`FX*;?YZu%|l zAHM5moZg@A%Mf%4u2Bv6KV&W26~Ithn^G>$5_D;3we4Hb>3a^~xDd1pA;x}#ixoSJ z4ob{f)O-m|1@KAl2v20TVODINLuQ7pT|#jlp==!b->S(1W!H(rjnN^A*$k$C!+ zX*yjz{xMs54FjE4dsen|-nlyr=;IiM>IY0ggBv*Q^kDoAh$cg`LC0pEUubz-sI~Bb zqY1{e*x&Db1+LKw7pS&9dGdvi*p55Bq1FY#Cua|rm)g+hqZ^3*|6sEp@58nw( zj6*|ir<|1qnSz0ZyMzKx;tE`VvX(Vi+<>+-)Cj>p6+hWrg&%=&%tDq@il(fHVe_i> z#X{~1Hv8vKt@T015)d>s1o;vR6ixd#c~gu{+4MrJz9yl=GKzE{oZpj)PHT?GHa#ud z(%ph1dDE^y6j(sD)R4pzN}WK!gwIm?F8nng(Mb^LK>ZI|2Uup!XbHRqF)0Vu&W`1`}c+4OU-p9cpbWgH?y{-I)! znV;58yyQ-}>U-Pir>R9tcUnes?XykMXJAJ-%Eaod^RBGV;Jn$@?4IHQPkUGys}p$t0LRbEOk%7{7&c}bTw?(scW|UGj2VDzVkK60 z@hy+aYRmC7kc$UKeOEx!oT;FG6-TfZIjgKJ?$rVw3b3x(5~z+>9;d1_5}J{P>}1LH z?SZ90O*}!$xa2*KcM3Zlv%Xz8$G>u|n_b!W)d-$q#|c}w!9JkIprE|%&V$5Q1_?}`=0R-o;gHa8x1&;=51bv)v{)0aT2a@+xtZ8M&5nZl}A%sKhh+jvpY z-Ol@J0av*&&4z8nEg?PxM$BOSZy0Lpb!$x#_wOIqUxvRGafKC(aEz^3g?tc7x zi8h>n)dltOMsH^#oZcYaJp(YsIV}!R0bN+eV&4B=$(ukUuuUOZdeR_alui zRWz+>JJcZ$FQGr1q_TDw%U|{QySDCNDv}Y z%QKx7^wk2a5{B5bB;7&v;W_Ib=TKxmQm;j;b78%+DHK5};Fv^X#4_xqfkN0m2qZYW{DzC29DEqkg*Ev42{ic)MQ_*-V zZDmo0>Y67Au+C99A9tum%Ns~C5Cs^tU7EhmIvoDfCMDuFs#hqVN(X@4@dN;LW}A3Y z+tPU9Cpgwwb-k{`zklJMYTXkJgazC@hNjtN{7wo4SM+A}-WE92!{0Q#X9F~vAaEHF>IvyM1Jum6~%(NLcZ)xvd zpRodx575Dc=RHOa$82{3x@wwk>^Zqjm!Uh)5v&$d$=&YVAy@5;#0RszSrbm;YB3+8a1N2BG z$Jt#{@=AVexmcIy&2nI@Rp{*^jsIl0mJRjk^g*)TCYZnN70ju=+F%*@><mEAbF@61}gfx?`_0FGL4F1g$n#YCs72{bdFn3 z?$$Ij^%mdYIPsaYfOpra-iw^S^dM951co#^j(TNF)Y(>fn4iQ^wNRV2S*Tyh(63z$ zamDsMoeA7i`^pV^bl>sNK(wDV8uS6^a6^gO7o0XLo^$Ph2=zBhG>B}lf$}m%S~pVd z6#CH9Loz;C=ywE+&>8UJxTl)-ot2r|;ZhCz7T2^v$l4H!TptX+&(~_NBqoes>|HoqdD3w3*S~Q?DKt=bGl7u78Tb-TK>-8)`dS_X`R@e2 z>$d0(%0ITTnpPoNVu#ff0&OI{5;jy|N$uoup~p5drGvcdTI68g0bj>SadHdA+59b7 zM7RoicBkbaU}ehUv2g*IOiciL1ah8U?Ds-`cr5|>Mx7O6T!$#&l0+jlaAGZNHyKbD@t};TWd1*Ud;d zN|meZH>B&m*5Us3(P}=fLxzv`_|Sl1m0N5;q_=r=MYiJ{rfiAq|sMQm<3*8_ZeLxtzjb|(4oTXynEQ@YLx--uB`T7P`ev;{LyAw&+NhX6` z$gqlIa2_LL36hf~XQ^=EV^^R%x_5~Y!rP);wxkoE%Curg_)XFqO&&Mk^>~g2v92|w zK~`KYScw9%EGukhcMJFd8fLuFZ-u$R{mU(t}bJ}?ys$4t6#J}@f4#X;wEInd*|}S zlBD~*z!})x>HLOUSd+5qlagCtdCWwBKI`H#Vm^`k|g)%(LGtkqznP;ALjtlT)?apAu| zaT|~;$Qp2c4-Pm*KH+{|C+fA#U;udz)8_u@+1YNzlD8+72@P~)-S2OXpW|69@hQPT z0LQ2%Hybw|bVvD0*yO`H)a2-3I$1X|SbLRwVwR35GXg`Nag*ZjYcBK=Io-r8k=WRd zi^x#wK3SARs?0JI3kNad5cPjN+BP{f<3WUr4pZCVfTc+NghC?{{kJ$;6d)5ta8hi{ z(H$q+Gm;Ld;;jHaP!UG6D{%rnC?vCkx{NglXr5F{EhBQiz8oUprv>IqEZ#Mf(T9YY zWpqtNi+c`*Ji%K7U1X3-q)xK$#60V7yZ219$RCk)6-=1(uZrabbR=H&ib-<5XnD0+ zbsZ+jn6?Rt#wZQ{eG+>SK{oo(~s@&_<+qQ{m;c(xni(3T>>qBM$CyK+ex|HWB zm^Un+CMg5vneg+K^7`blh$})Q4^Jy%NZ#-HAY4^ zqk|~hMfY_?QDG-a>kri$Hi_^ z%TumE#^iROpB+=glD;rF_x^?7Bvy^w(V<`-u?uKb6fT1{Q&+h!j|QfK<_VS017}$# z77N!=P&I|c2TwAPzep%}v4GP}Eh4qcT>hO+Lpc`^1%881q0Z;iO@sO5xMy`CXz;1q z$M~J~Y^<|827vs9lo6Q+;TqkX6OT^PBA6JdzE7(Ku=x8$!iXDBX+)?2IW(J7Zku`~ zX#GFVto`cNhEKn6+ow19^BF%zIDA>0jCy~L<=@SNavQ&si`Ol{gJxNF@Ojnp`g(J> z_i06hFgJYur{neU`}I@m^Z4q}*V+H^)YaJ!rymILinzC9C{=j)b-NA;c=-VRX zdv75f&vUbUqghysQeO^~Sj`x2*b)mbNNb$>f^L(U4WB<@pNe{dC4m|^lsOTCl^_~d ze|mU;`Mgs6Gz5eW74yCY!W|H3RM?DffMy>{H}F~wroqn^$K%n*C~lcn%n6j)+W*W< z0TK_8OA?BZ2+WR`%Z7y??f`yZRrN;P*E%NErmco}`Ar!ez&*_r+T?3OLdv!d#mmB# z;uhnTQe~Zq(le;jDqs?Yzs=h@CD18h8@4er_=(vs(jTLPwv`5SatYX;vEz{t!?cp_D^rr-@UQ zb2bJ^6rG_VH_2vVu$}Q~Llj2|rUzm`J(fU#ArJ=y=Y$Hv3$<%~OS?Q;fXs;{-S6K| z5GYFRy0wABJ&8k;hLQ(GoOs?G5{uuQB@&J}Oi?PwT&n7gFpcRfO+Bcmd+AmlvJ3@) zIzh5ykGV)Bdd*mlr=*j}oc-KXCfe<|U`*yla#V76-n5{Z@WJCIrVuz1x7)gK^iG`9 zZ7LeS+PPtn>ccmCbTRsSezT;nFCS!R>dxB_X5w38#jj1NLwo;&-z_dI(-%xdG?5yE znP35OJw9wfW|x7ANq_GZEU<`lM{*Aky4c3b4VzHw)#bmTvsT(f=-y-yNNmo4r@d7Bdv)}HKLA1YZ2 zsE-VVmmiVIP7XmgZ|%_8GknZbO9T^CkeD(Bjbz%;+&35VLh6Tp8LDdsuuLmJ3Jnf- z>dXeFgR351E@k3n_@wIewQnrG7xQL!$58(57l?hl; zTrN4=J!{Xnjy(NCFVJ_rnF+?(L7CUj_W1($eZ17L<+F@>^Ak+Dxyy^hP4@Ep_&gn# z(QrNHUm4*gv)k0c$Gz+_CKM8Y*-0cGRL?!J6y8J4b45@Eyp}P8;1$O!io*EL#1pPE zARpGFlV%M;{&V|;s3z}*p>-vm7#~(@Eg}QLp%07|L{e|SyK3@rZnD1=x0am$jcQ8cs6Z^$LX`)!FPr|DUSSFGH0Wk^;KW*SJ z;=Yp-h2I!U*(B-I<1jXWa}bIV*7TP_C`>M)1cbhCwTm1Rcu1X~lOSeo_oB>!0|c_) zXeH~Z2{ke?Kld*`fo<;M^(zurS9q;ygidWcYPC8hQrOFrNCRRF)k3J@uo|0DSgm3P z{6^DgKPtm)pvLmG^4i!ctuC#2FJGWg)4Ir@iq-uRl`G~Gh;4O%+zKcYUdQ!`vq49c z@ytw?$N|s-af?ya>tq?URyn2KO5y}*5Exi=Pp70%0XmE7NkJHg+WtSfziRaC^sPfw zK?=lcm$io0Vc3Pp>F87c>2;J;kwIKIQAO!6rMNm-nWE%5r(NuDgXEBETUOLYw3PDp zM{amGwU$$DKwN?VSQW{M*K~GLr;tjcVk%V20_EYe9j?5U#=>In>?S+lKelB4^xH2e z-@Huk?vHzT`;G0sJ1PUvaJf+m=9;P}czl!F*> z5?~TA9+i58Cbavs5iBE*%yOGxUat47DvwZ0yI0Ocohc@J%CKSs6_AUdV_A?wh7b|;^lfQ=IC24vozpx?Azt1wvyy}YR+>s3$XQZ6&gVmfKg2i~1kRZn7}g-x ztsQ}1yS9KKAX^Jw-eDxn+QVr#9z#vd5BKq2`3Kr@i#Im zm4Y7G3)>d(#Bvi@@@@Kd?@;oLABN%O*`2Qq7Ch> z3UL;C(8!zx4|?mF#k2<*Q)~weEnTgIxeW3wj8L+c4yBti90A-+ng? z79V!QH$nIK7#8VhewbI&34n*lARLAs>>@ZgeKhXJ_P4KwoOc&=Bv)0r_dB2lUK&42 zSLl0?ki6jzuGQK+@(1gQK*-2kRqmJ>{Z+j&mw6Av5f=A!s1L8^2(QXLlk@_TWgI5HFg(J;MdtG65!I0BVL56VW@Fuv^c8?F zkeK0N$Er$ASr5%eh!T;s1}fAR@QxM;1vS=K#gPC;Om*$RXNF%x(-5+4Q8BI8_bvO! zj?4N|=lD_}SJ8oI>Pm-9ti2HL2rzB+@xrX680E;xq} zuQ2?lY^nG~NG~V5I-X^Fvho!4ext!Ovt>An3*ldtsMQM20;FcSJ53+;&4q|{wp*Ar zf0}F^F1JXo-lGtk@)O;Iw#P9KQi8fPmxqkfAPU(y>aOEz8gqvPObiHgM#tc%z%yDjt{#Drg{0u&q_T zcCFefX2+=2o6;`D-e0=}Fk*nri~N2Z_@O_d8x)vD>Or~{6dCns*s`+$X|)Q;IO4Q{ zr&~U;WpH{j5h92tX0Zn+RW=Ozh60~S3|1$Pz0>tmiNTh7^eo%=pb zhBPy&%intk{fME{#KeYw8S3O)??O;SQFCKsrw_yYg4K!-iB73SyMO_I1DPMd9uwZP z+o$2zkC19ba|Ceu8Djd2cXP#8Bm=HkaALMcO0iZD`-=~O+JAvv{v|6Q5c!Wiw>{!^ z)e3RUyN2+7*pp7mJO5powo6XhJ=lR;=o|Ceal-H3hcub@!lA0;s`{hZWSfg|>22}% zUz2G=k_W#d?l}yWt}&pJk(ECO8|v{{{MQx)BBYk5Llh+6c8`Cz?h%VfpHs(zE=0(& zdk8BQoX$>=Z%{~sDNs}IrT)LG4)Zj2r~DeSwar%Xag`&qRo;x=!tk~=zB*o{E{Y<- zmHv9RzClghLjM}Y%vDdeMGiw_)3Hh^u3(cIOC^CeLE_b2To171fgoNqj|StsD%UmXhTHbyzfFwg22|#co#K zMXR+T2Crvc&=(-em`y^SbA57TK~Az?HJ5Xt5PL)sz3>$j&v8;-nc5^&n;he?T-*LF zA@1LTer=6(4FYXFK_2r*xc~at5)lPLf=7yjrAwEFPR+_g3-6Y7i_}`h|;0BEc1e>SQqyVaoEuFt% zm!%ITj7rV?5RJX19$D!Ml9jP*RaE-6q9}ncVx@mz)a2A)pKCGg%NMEd^4?F|a55m# zkOMwc{>}lYZgGCd(j2y$TRk{%Nh_vYKp=Kt1U@Z`r(*}!&sg>;7<=W%)6HsA1qbT} zo<}IvUo01uyYj=9-|YUqWwLBg_3M zhqq!I5(2MGe=M_Owh^F7gzgg{wX(Nu`svQXkde|;T(8VJRy=;3fgBn@j*5K0PZ8MA z%(DP-m35*;zy}Oyz4s>stMHP5?|t-R5m^bKzeR;}zWXo0a1SpnYCGeLp;F=HR<50M z^*^7=j^`?*w4+FM+2voBifM+lnUt_zwvBL^;`WpM>ypAG+d->9%T^it-mUtB5Fc`S zdG!G24&mJIgl5ra1ZXjn>SIAOY<&m6n-)MV$}0a37ya46oKd4%3({cZBTJ!zlrQ}m z2=fP~E?qGHKv46SxAjrWqYhekYcI5*e!`$(NSkoc*jT=Fn*&Ml5Q>{nCQL&>a4~>w z9g^4IT7m8S2^cGmu+vW-u!Z2k{0JLsdb#%5SX)d*voLg059pA7Y>Wdw zcY)OijH|W%9_e<$b1-@_gJH#n1?K0oV2-ejgA1))%}ZI+D#do-AWceg`ELqj>_2ZD znX9fxy;$Bd!-C)6yUXQSQ`r`hn(F{}wWltvo2%c!KxS73z$rWM?tCb}ocH2|f<%}w z-Yawnk>!70`XGC&_-osVABS^o+BFB% zVY}%s9HhGh%;dfr@*>8XYV+gQkh|j?As=I7>~l();JfH5Ip{lG)ppmVXXF8gbY+v@ zm2~ls9=vjzwnn<{ED_8mq=Y2}$hNo36A`_zZT+e2DJQ5QrTrg>)_ij5J8iYnaG(Z- z!g{3*=R;ddUo=s6z0XAmk#<07r*!B$p;i&6-}r?aIEK({%M;(2$OFy>obbf!h6xdn z9wdf~ip!t=gSp@$5jt?Y!!!W7^fIWx#GaD`5VqRijL83Xz;SihxlIF?^Vz`4YJ_#U z^B+_|+WxIscW_OwL9F`hOw2r7mm1IH>|AzN{g4)mf1ZY z>CC#-wp`gqJN^N$bM<(%Z8@gUpF3rBDC=7MOzu=DHiCCk+46X;N>~7=dx5q-r9o_B zXpRyrUaRkGJ4^_Kh?4Ty3D5i@qMUnnkuaQZF|UUdO>a@Bvs&t@3`*Ew{1p zO2;j-Dn7)hmhI`KcMazse?Qz1nQF&i3W)MpKEZ|aRQtP79n+hxo_5!IymZLQ!w5iZ z&`r_*XLt}R z6PbM<{>0bB?B!{4lH)4$EG8PAZL;hEt$08FMj{jERDE2OM|%MG_e^wDBvpU{gZ_Hk zh{3A;kAxzfYWJ`7gBtKvvX4UYpEsSkFY^2*8~MG4T7{r(jv>(_zJ>6joUCBnkNdr& z*SqWc^L-==1jKlOZhye%&FAaY()3f6 zXHArO+kv%vh_`flYS2L08F6jdq_#c=rE{b(I zQ^>du%yitB-K445x#hyASC?XWCcaB2sINMohm)(X&0tsHALLU2A4zH0A5Qm4(M;&R8?9iF}Z)|{{aF)G78S%1($nlYS64nlbjx9kC%BVX<{(86-KT06ecxg0axOL=4isRTeY0c1-`Oz;8?ps@w zDc6$wa>h$@;{@-B9Beu0_{H?BXFs@~#A5_0TTZw{@7X(?StX_~djp@Hjx@EvzI&&O z5Y$^^5(MxanQ}wWk&h@_dQcv1kISK%Dx`7GRmx0i(3TZw?ZPz5E*qKNPE6277ou~V zlT);W&QsAiZsSHr3Z(b)){+#mMQ|3NdM!DUrCE`;!lmxEg=pNBUYD%;d)M3Ao!E-)L70#VhIiU54n$G&an?2nQ4OSD z^d;iIy35L1fbxjU^gWFE9O3q0V+ko};d0e`)YjJ2q(SQkKq(QIp}>EMdi#40`w}Xo zd`2RSTTf}gCJ2O#0u>TKB#Mzz_{SYPS@-&w5lF984|E?qjmV_S29mt)bAsn#$#{8a zWy^Gx7nftUB?~NR>8YOk5Mb~@L7$&inDh=CeVNMu2O1161#Ki(589ZF zB*Uz1voTA2k&%Ldc~>H;t7wa#a7nca7{-2piay_=4xMHX5L|SiNjr{`qjHn6>onjQ z(`A($hXv6-g6zD;b0`lC`5y3?v|mvp1?}UU&vxj1PCt@QFj=zzg7Rtu6g`5w!AO)2u8B z#8_VA1H<-xk42Fd!_JZ=A@GS*=JxGFBQ8d!WMOeT2%P|QXhuH!DP56LZZ?-AeLjpd zk95fsv_VC@j3yxWIE!50g(AeWxVR4pG&D`K+Sik?2zOJ~=rcx^k+G{oOy90vIMOqB z0}A1d#OfX1`50FiiCwc?hUyO*Cs3GTjknf`flzPrDFSAG%pbd`H3N(BYYj>t)=i=J zV#jOm+Os3ybMaN7gRA%ojC*Q;l9&!2z1AXr2`Y)M*A%FV=v(;7@-kY~`Hc?nlUL7E z_o}G?SgJnW^Qk_Asgup}Q10p+{+Wq=MgCK6PcPN;T~Fh5t24Y$e8AOKPwj zB1&}wq>W@nEDx_zDO|%mhh`3ly=PLT-kLv3U?sN@yUOjHi!&oLoBrd$9o57|Ne@9K z(Q!+D)BiC`$}0mPO)om*Ng@)UoXZMJ3~H!Q(%rYTT9)qjjgS_i@#&ECw5~LV-2*lJn^YdeLLy{1>rP^)|}JA1YuG ztV%Lex=W27uj^~xV-Fu7eVKgZqC9~J!U!^O*Wb2pzb+0rkLxPiD@OJMta3uoYA$W< z72|$2VYO=wStmNrB=AKX+m{r;I6y;%>I{4wR8hX)uWrEZv*cT9Ky5gvDBiJX^9;1Ay9@EBmuFN|P@>R=#gdRLOz5?&+hx!!q?S?HylK_bhBkVX5 zOEK(Dmk^aDCpBfVI201HNk)n1O8)0MEH_w)7{(F>D^;3~6RtVNg&oIXFZEOeX7$2Q zn%KyqKuk2Vw1)>w4ZwjEhX8&nQT48%xvc2rQVTTEG&I0sa-i}J?2hH zP!etu)t0y+@0!_PcWg~XMQ&cOHU<58;Ei=g_ef??KN0CWG%+{GwUEDEPM#fMYypRU zwi9ca=comZdJ%KR1#z5;rYHw(?1Xd8-U@_T*-zOF#+n7ZB9ujC{-`M1ls#%mAqF^m z@QeeZ@U;9TZFr`#Ewg!5qze_IgTC2 zuH|ZT4x<4JXTRH}hPYgjQ3N4towyl3o$K~^#vDaLe}Y#W5?As}@o+>*%YC5KNBT*K zr{N&;KFN-PMmfh12#6buE8)L=J36j!LskXV-V9R6y=&bI{r0(Ud85dHQ$rI9wsr!# zpYmYhp;fWcmR_GE;)$=Lv|yhck|$)52A3lgQh@-IvN{x189|8BJ$fP%OP_9QCuc8- z7S;qG4&kxB#>&<$0=d~<2HOcO7x#u}>VxY}2I|`|8r>0j%-aBQ7kns1QT(wjz^krfkJSZWgiE4kR!3mN_5jJKQdH zBGQIy0%55Mb>6swV|Q+!?2+Xv)RH+Am}$U)_4X31AJ$*jG>%uInektwAzT^vV+ahp z2t4P#f)fk`5h$#K2R!%?6@L44FWoKX9;JZ=#@g3L0ablyp8biXbNh>!`w@P7I_ zqlBgC-_bvj#;>-5#v03aDA?(l-W-X4_BhaTMmCpz`J&FPZ-ehFcWE$f)_CKbSc^?% zrQQe5>Q_voyOG}z;dcs5rRkf0>ZM;w$Y%gLu3nR{ zk2DOKTq(l;MHaX*;1xnZu|vA1o959Ic&~ZyZPB(Q68MPFv(uB^g$ z`WT`tC5y}>=NqK&?Q@;{2{v1$fXQwnxhz$-CCmHyRL>zM80{lGoIijuWlXR(X@yXj zJ(%4#05{dWopwgO_SU&njdFtL3>2>EBHy7|#^iPO_njWyS}+%042s(!46KQ@_ai2{ z|MZes&ZYSXA|=1IvUD5~&X<=TYY+gw!NJtFmK0tF=sTxxX`}PDi4XJjne1HQ?|sBZ zEB%kQN0oNo5MO+o%QOHq=;dh$s^RFtxo$}n6aQ*JYKiU#;>VhO6TRus@w1ubW#(ks zvDL!AN&mYji)qjvm^Nb2`n>%$v&++8@8rc2tu`I;Ol+J+b-SBF>#RN4-9W;4);m|P z9`%=*8$N8fF%KF=L{$+(KN8!7%z9)i*wdcM3f40;KYzi-YB1o`-UF-{M z)TpG;qyKz5zp8V8#T_a`GhyuLYy?bAhXXVyqC?5VI{j{ex_gWYDMKYJmc&VsVW@gc zSSL6+IJaH`x@et=&-^B-2a93swYz>48n>(zE4zk@6ZO^Wil_zo3MB4dv#kE7`m7=E zQxY6T9J>))kQ0EYlY}~x3X(A=#!gxn5CKHgX*1qCazeM5<%Hk#m2no4%>HYXd8|L- z(2GqN$RUImR0qT`WsO+)R^B;d_HR@6K;-#b$(kw&IkNCl>G}Y!rv_NO<#46Dpu&Or zz7)a|n=IFy{!p`WQz?)~{$H5+uBhuWXyj@wPYK`QCri^O8j?tL}@7DJn z+qP}Y#C9fjGO=yzj-5)j)#$?0qjR4b54H`NWK?&}%8>R3PDh7Adxa4e=u_LBL(+^3 zK00We1uuB7$dsY}P{Ot-UQmINQccg1n|eLJ!zcw$49&Bb-$Si_`!3ZcJy&nyk82Ts zIiC{Q`rkt}F*~T5qSew9+Tu0`rCrqY%FAw&&lzr+^i8euGN`-@x?mTv??y^mT}px^ zF7d2?1bhqtKh^d3=L$Ac+$)vpNQ@_3W)1Sp2babDM6zp(s4B@8KY?FY!+O=*zl14@ z?qM4hER(@_Ws|B%md=9Q1-P?c`^@rI0|RCE3S+E38j>0s;;_LHBc5xR%$69eK(*E@ zj1>STrq_s82dF|3Jt(Ifk8G89L2amYZuy95HIw~WZ+owt1Yv1#(Ts+=o0SArc?jT*B>mdL(K2fI<*J{%}vrjxLOk)(@57s^#T{&*pW{dJrPrfriHtak5(E?!M2Nk)CY)F87z8 z=Ipc|Rx^&fMFxM>t(NTH-%OYLuD<#zOd3FdKwiu)kU!#zY^!*N1EA7CVe+SQ;fM5@ z_B8i7n0Ajj#oN!NxzD|f1(B6HI{;LA$BRnSsg(kP{7;=L#Dq(3Zp!5c=nCiAHf@$C zT{lnf>zUySPNf3vA2HU-%USwYLpK$}rVSY9AYGfZk0=}bBD_UrXEJ0hV}C$9`{6c# zAVg=^ZmfsIJ8T}y1VR&jw}U}P=TxjsC3~@6j*Mr2(`sK{mDN;egFhz| zNv7OCI-w1>?K$=ZYk-#8ECzQ59Ma(pn`TF*L68d=D{|{~9R0e$Hn=$~9-3sGRh4G+VoJ{nL44iU$yJx}BVPODZ|EJymkJthJmv#~LoX(U8 z3IO23000#J)A%M%Oh$IbCVD3B_SOa#wkD2DHs6QVdQNsOjz-_3|JOBRJ0lkx6I*8k zXA3)9J!cb}@21Wsj2;Fy)~YK1eb@iN<^cjOzCpWq9LQa7>d2!*AjTAueuR!}^cda!v@R9kmBS$4tf z=kyG9v|MfL?H*y>5I51VbDRJ?8QSdKBWJ0S2A%w5pp8{VO zKAJRWWQEh7r#10{boP4ovzi`gHCoW<&P1hzXfM{LKfG^8hGZ~|d39udu?kh8w;}D{ z9(cRn_`ZMdqNKxnx7VwRRVygvWa&Vw`XUE2SQrrPrRA5fkp#e|^XnT({W_7_e8^dj zrj?=-`2@=jYDGr74OadY!PtNjZV&Xamqo#q7z3hi|E$h+pC?XqXpZaJ4woi<98Zp> z6cEmBCl3`swY{TpxxPro1C*29ufi%=R=@*HtpfrvOqR(2Fkxfpi88YVi_!4o>C59j zu_PXZ^3_>7JesPj^o{Ii;?-j#Rdk)jRArGkB3!>7oT}0~ z@ZQKXy~j&D?b{v_#zhU{wd3qTP1+~{Fqb*ghtO+~At-O22G+zqRFviAE{78LFA7S< zUV1`jt17aztXv1V z#oYOKr%0itO~YuSbaixaCVKB=!2S`t?jMQ$yR9Y3%^mG&2^4{2hUvgc&WCr;9-4aj zW*CJUXW9~09$;XmS7TZZ@*j|RnBzQOXPr0b@rB3V&WD7Ky<9nU(RPB9Fj0NZhvMi` zLVgX+^^=MYdVlV~5*t*?5|PAQU;o+4PcR)VusdDbPi603DTNpYI#9H`rjl5=f|(2l zaHIwcf2AJNIePIC>=F)CSGUJyOcXo7d`RQUK+81l-2?eF;yiWL_jZnMvTf!zFpuh_ zCH5_ULZmJ8$H3)wp8aSiI+oTN6Nj2RlXeaWr3@|jLoFsy&dd74W|=9Ryxy{M{(iP> zNPcqG+PJSKbIGd^VY4F-@7j4T#S!WBO1uch7S`!3YBr_uWFt^};ZiA3$6F=f(1Hx? z&J$_|*#ihS!XUrkK|;oYt?Qw#9XV-iVt%yNB_8Y6Y?*Tg8t73O2`5o+Y$hKXFR#(~ z4+(02^QY&%>4}*07v>2<*`;0O)~8?m+9Jr01N`L}L?)cD>wAcF!aulM27LV@r%5j6nC`lZ{(Y))4-CgY%`S~R z8N5Mn{~#pC38x5|F+|&Njcv=K!rjY92apjK(&;r`6xPk}8HNh;>ZY~P$r8>c*_zSW zoDn{)%5=I;uDrBE!i z`*gX@9mldM;W@2JCWV%f{x~V)d|LBt{3GHbGw~9*0@$hZcsFe<{BC#d0|F5{n`;uw zzG`*UG7eSdRn4NWxOnfa+de~In7EsKe!#hLlzJHiM*9SA8 zY&Cl6w`O~`^kaEPL}yd41J4Da&Fd()W2Lo1FatU$YawjGQ?-~G;?|h=_HjB9654gn zZF%0r&h1e=6FFg=H&E%w$o~~<|=`~yU+5zVf`0V@4r0< zhCvDdR~P``9v1)*_)m2^hrVYPjQRhq`8tH43}VH)g_@ipg1;71o5C?>^jo1hc-zWWA`X!QGh85h~$wIZN7$$ zn8Gq}B)f=X=f7sRaGbZYVdj^ueLLU~fwTFE5aAgwN)}7CJ=YKIm<4*&rYXO0vM9+1 zr5Y6NQ?>jqh|`zQnQ3Imx273h*K=z&Yd4C4dfj?@G!Webq~7Q1R!zDCCl7xtW9$FQ z62t}TSJ&4254!MtG;D0ydm&KXXAteq*{ohbF7(&*gdo$eDt5SJ1V*PJg^ZcXd3EqsGgdcre z7(%Ods(`Tb>`r`SDoD%>G}b~0Qphf8Og`koO+7jxXXLkVhx?b`A4YoCGGngEB;`#* zn|!1Z;gOB4s1B!wtQu3&9t7#Hm zwkM!bcD*)}&VpkOo|$RgcR~U^ zx(?M!TMv7L*$cQh`#}FymR`U62NUy*a3Bds7v`3>Y~Eb_`kK?z_PGk0;0nnQ$aTk` z)e1z^W_!x}bME}9sdwr6DvX$@(;$Wa5CFk1BfsR6OV5jg_cmBRTH)*w?(suPQW8v^ zCU6pJ9-~8b^3ENmjMD}Z-XVnv%3>vI#`ya7G zU={63xz#Un506&puw3=Oy;UY4z~5MPvsx$aeTJnYS~!739_oWBgSW@3xQfnN>lZ2{ zJEQO7)%w*%%N}=%Zv9e!2+bTO`zK+@Ql}A(juNA7GvFt7 zTz)lP;`X;iG|=DG-C@q8^zx1F+b20c!4={|?o^?>*?P+jmt!zP8ARm10;3DY-!C1^O$OMi{~C;o`#&FQgRWaAa29Q z^hz8;{g`MenA=Xt?ve=Q+I<8<=6(Qtip2<-6fH!jHhC5b5JnXaZ zk1>3Y0sNf}RMonKp4ss6_Dn1=IH2do#KfQ_5a?M@hX8Rbi z3FJLbn)9agXvoy;JZ!*!;6cyecpOpz+M|R5-XbgA4RZ2HNWUy{Kj{f?uVqg`5ZaP- z>pi}?Ac(}`?3wUzz!Nj*hJTQsl;F~iE)D&zIn%WuD4{|?7L&w?)c6+n#^eY2F zjADg@{Ft$Ay{910FG+lt%R9)Y1YLvwNOeA0)U46nbrC&IKTaRaksL)z_%|@W!>reO zwfkI`6WxUhZ5cfxfhpV4(!ymmbAL|>1@p{H6{{40+oZ4nAfb5j2xl1Oj}7MZ1ZdLj zpjLi}v^zmv8?RyJ5IHJCra(V}X|^9yy0h1BtWS>uSoqzrwKBiB@ZBaNNSDVBfdVB!>fbWHtHCrcu^*-G! z$bphKGz!gwFcvC8t&63h*9Tz*?omsn9^;Z=_1;x{fB0>a-5Q%sbk;-gul?%|cc@T# zb@j*ZAEM=grp1!r&}E!QSRjFD>_Fki%{t{u7bp}M8mWdGtT|PEan5(4gjstBN5B}m zw9&W!<3mdR`n|PeuHC89D8*T}?C8ni>D;|K#%qO`>wv>B9s${3jnkb_4rS+4yh*bL zf@XngeD8V*<4cyZGw_pLt2gXMdRYOq|D6)8vpS6g+lqxlf6vOHu{hw-PM7Kt|E~Sm z9`?CgE+Cs@@I;uQE5MnbEJQSHI8kFtskg~z zXcD7~V|_>G&8g^zdoE)zSv>2D;ma09{Np}!va0jJ+jLJ}3KG?KG2C}~zY;qiw2(wh zF?|wvYd-ZKOLX`4h~YT@4ik!l;zt#^mSLWKq6c=*oo70{b=B*yDw+fjGa;2)t=*V> z1$5)mxKD(n;Ay2+L&_Xq869B102Bw*r-(7G_qF1vIY6}0<<@^6+do$!xzH;{26X=d z!SH-87BWn6Wb#~t{YJrhZ0`<^9ux_ve+}<$`rwdUip?pyV8Qw{6b-sOYuwpr#6giq^Q2*~vyaud|i! zt2l*XvqBHrij$3KdXifK)l3yl+yd793l(Yq+T`8(V4g?En1+wNyzknwV7d}^D!8fN z0Jw|@GN@A`Xh6J}B}044PVC5?RSr5_EL!c9)H`0O(8g%&47VUt;;Dww%yivJ9w0eb zT1xBOh7L!8z&vq(le!p}jQ2SuG~l8_QSa9T`9o9@(8Oi=XOA??xIQcnrVKPo)e(g0 zw~T((x~yKm%CGF|(p!Tx1S6G2!c9O(36RvRbb8roA*M|-<^vKR=M*h=bMO`Dk8ULOwe^;nSK~-6U-{#06tk4cQ`Tovo$l;Sc8NL_mc&DLb>wBHv0^_AB8Lm$}}8YZC0J-}WN8 zr`{Kb>asbtO!=L5lFJU^rB3~KPTAsHXsv6NT>TK}lueTOYTSl^=E`eZ`Zz7kP>Lnv zm9uH^%iY5u&i2qy7Oj2Z<|oqGdJg1n4DmIu=vS8n$m1WFH=+x+K7nciCS3mp=MX=* z2GR81hILx|CxNQ7&-n8p%>`XvGp|0Kug#r@^>suC3xuGhPx)oX-VAJ9df*rvP*it) z+p6zi5|8}c8^a`M;%H6gd#%tyJg)>n8G%Cq2ZIc&*8iwOC!!aKs+7Xk+n+|^FeN_y zeMXVFHH~eCiH&&B=o5BR_Gw%9rk)^f?aME(KjSQ=I;nxq%&G3qA*f$aP}f-PLg2ec z$rU!4u6kLWHe(~yko*cTVtz-tR(5|fb)1`(iPkY%u(kIW#f>9-yGi*A487U}#K)FB z^=DeZSWL*Y9#Q?yT)uKK#=5Wm!a|wMdiTn2`<;9bHC~7#o4<4D#6=wo!X-ziauo7J zHN0_G&tnOX2?LjV5qsB~Bb(Jd+#RS?B}Yg58Enrtf6Lr`JUx7Y-j1)@+MTE?SXBFG z6GtN}C1f}6!kz7{m^+>P9sC$Oy51jJ+Bu(3O`KjBvwm*g#5?&h2!sOEKd*}`Yg}xu zs7aZVAngz`t(;N570$@uu>o~gzs;R^03$!7X&s?&*hKj$BiS_~H8#&28DnNv*~EI; z&}HoH7VyA*;21}NmdxfC_Q(*;vKZZ`)x%uQF$f*Qf;1Pz9xzH_zBuzy_2h90QT9$MY=nDTL|jN)_1bJ8kf# z@&L@`mDg~)m_vfQjvF4O&T@Qbefy$R<(=xxLhK*n6br;a;kxtD2;Z2{{3es#HoA4h zgJJsti7#TxUNu;7h*Tg&Z{rzsYa@VBs$Y43>Ti9fQL6H=oFIuc(H(ufnVBNuFk z9E=uZD5%nP|Vi5Y|&KLcbivl9ts&`@b*3dk}-^?XqD9CmbhX#V?me2v^V z5asB}8Ib#a-5_$o(^pgH$W!mzf$ zkm5KU%Sr%dQ#mX!u`Ark9}J>gX(m)1j(Z3xH={L~&0&*Pa9A{zSVMd;yz4r_aunoH zt*6!e26IIr0-$$5Q6vk?O}G^6B|UH`gt{~0L#>7}HGFSL2v`Gp{T%v%CHy}I+-xnAT9)r>+8`J2Casb7aO#hd_J-+23jv!x{63-o$1EWOz*U#;pjgYfWi z334bk}qTwuF!I2U5mnE^g8o3s% zlMSqLJk2PoMJUkWQ@pHEjRU{|JV2o7 zvlPl5YtCFt??kD2`Pi%v3+fiwKo)4PO4F-W{n6Idv?@4%AsaaOd}b8KU8OWLTm4qQ=)#nr#x}Il7lB=%9Nat+# z+KL8_*u*G(;2L26xHW$YUetWr{;`l1Ip-RbG;m;s{qOqk zqQ7EnmxC0kCd}~I^;vjpzG(6G3yrC8%B>5IKIh=F;W@f^>H~QQ3p_wF(f)89nNKM^ z9hqeWUfW9fKlF3j$+n=ZI!e7T3?<0h#!tjZT^pjQt-PGXkec@)gp4QSnOOtE^0PAw z{2jjr%0h*$-JP6TuniJe9=Pwp7)VPeLvIMnkDpQl zP3-MZljTE`vW+x&o+m(bJLucN2r+(y_ng00(v(Y7_ZMyuv_%@5c29o!vH&Uv&+}iOI{u)XhA;41`(xCk=}fkuYZ{gKWzDM3h1M% zh6p<9+FV%iY0OBPP%sdE&;$&R6;kDJz^$wE#V6Jrec&n>Yxhs>&)+48 z92a-M#5$lIon}XVLw|0~N*gVafiF_NGZC5oESfWwe(x3v9vXOBnJ(1(0*XDFxoj^y z?k0vuf-&3eV?*ubp@9b#G0S3W*2*~7gIOHWf`X*kBKoePQ5ztN8Uo5FC~ks4 z-&lRE$Qb3D-_Z@#`J^KtycXFAT8o z1XwlTMGLE2=nR1;n1##A+H3Oz>>ig*%F@ZL`QU?8cIoIWX06tuh;5phFiJhDjpqpO zdMi?Xb>dV^CUaD&>_MtRZsuw!AG`Xz-U4smpI`5H^>@v;?DL-^&UDHo3x(cI(PTI^ zzi@jW){p3==A~3PXUwp!L|~r2CVbM+!o7g5I%9)e-CqxRJhVHgq#?-3Qu~9!NxBE#ju3+=9oP+n1jHIIiWdJxtxCs2XMbgmeparb6inHvqYM^TM?a8iLia9z_E=_l5V9pdbjivL5PFHj(Me4ie_!u`W_J3T<0@ijgZ37kG~x3X8q-C?(7I@t9l zDB?GBOC9$eq16#~{7P=~PHMB-D3@TRlg2pF-9P)?aF}-xg?L9zt%$tgJ`nNY)Wnj0 zp^V+eP^i;r7l9Sx_qsTEyYxM_P0k6As?L(e79bkFru!eA#-DvYwLozGE$y19ho>+B z^UPxwXPOj#?zbXgQ>qpIR(}39R*_Bakam1iUxjM5<%nQ7J4%-%c`Djxz2a%gX0UYm!9U!G}l6mXX<&#)SwSHK= zkf6iy6TDN;MrS>xi15?n8Ki4Vwsex@Om zBbOvdvti}46=3eYf``sexQSUcSdO4)$CJs2UN$CPpzTUWT){aud=52cifWu$hymgs z(EaHIIzZt;zYbpyco7wWc;5)^T)Qzc=JOzymFq*-{Xsng7;ZLb1UeI0`{5vU=!-S@d<7Za)3;u;>~#;ZWr zKDNwye8T63#jkLfy;|VqPj17Ou%fb6XUn>hojO`77aiSL<8L==NW5}}Zt&ps>!WNH za6>SM@!2PlcmwND%l5NhAgEg&cTz|Ui@h~9=`=#~a2_z^G zubl)T1oBk@FV2ONcsB0_xHHtCw#E*pbj>abeY=Yi*7-l){VAW>^l#qwuHOu=y>_o2 zS2rAUnuYf=FuGb#`VDUFt;ja{BD1{%GL9-~|`> ztaXXqXh$?~3K&Z}9k$^zfTVy!bh@VGrETZa1ad-8H}3kUpn890KFs`FNo{R}()?EV zlHEl2lgYBO;FJaMASjlbXYU(eEn;>qK*GqdX!RCkxC#K2KheaJbc-I7W>>ph(&m|T z+Hj|z%fqTA?-pF9AwSz>64}cclQ00zaPR}`E_a5YPKrgp^XX-i3A#x>aRL}v74xV% zft+bWx5IH5+ykiOJg^X}HD$y`1aoY1u0U80hBeWi1DJ=KG~wE_QIs?uI7B=PwRIRKzt!-P%7T{#*P9RB^Pzu*Q+?HReS zJZ6RJ9SXVZExeZ%K2 zAl_FH@w3jJLS)RyYZ=s-WvvZP?2GKnQfiXG=N|BGDeRhu>)bZ41{s+{%*WiC`zyV> zSPHs`1S~>NX6W3TZiSkF{%yQueOZkM@JR!aqVxlET%i$n!r1zR481}MRR{jW5Zo+z zR=d{ZfQz8V&(iUENOu*+9gpoQhmb#92|c=2@+(b9?Z`%lXo-Dt<)w_}^+Ijw_5&U4 z#|U$2YaqhdvIlJJ;mA45z`|PNADF(5K(ff z$%2#gThO*z?JE$x9K;{UwE<;JHS;E3@+TJHwC9#L>d=wdmp#0({rcLaVRsCFl$qoC z?Xdp+ec^0hp`-J5#OP=XTHb{QVmhTB-8LckoDQ;BA^KWDz+CI}H1rt~auppCeGUSn zZU4(XOKi0uz)2dUwbQO;#kG3jcTW-?V=*5?jK|W|MG{q~hw&A18aeR73MX}!j=s4L zsx8n<^O)Q|;HIBlMwI7lELgE(r5J+u-G$3vn`@flAcBoagW%4=VWG%@a36j@x&+B@ zI>}3!!%R9uvFYq8tC6xWcsBg`z?O{0S{*|~1*B6~0g+8o(I4XS8tQCiErA_cS()3J z4XPWfyHYq2gJ^7IogauZ1=Z)T=XfLOaPr9nFiNuwRPZ6#Bvn%%1##3*pJHZi&oD{F z)L}5KF!*RQbk9+9k3;@m0}8Ze9&^z9*wMQX>ZB06(-?}3f1L!a&FexgB1iw|E&;Rb zSmQU32%kEgMY#4$^4Lbr>W1Ffff~VFkKWYnA4l=|N8y?!@dm62Qsr1%dJK{EqKw{zFK zL6(PvgpQ-t8P;ciPiUs^RW&}w|30m}ImIq9@n-q0lO(|yRX2)lsEeji#-0AyMIBUP z0af6bG-pnxfCH@3$t}Rv4!7~ydCeZ$VSb=`E2$DQAIHR#r&DNUlq=;vIK+K92F2;V zEN75aQQ<g;7>iwyWXoelM%BRA0qr@Ja@_=lq4UpCC{5~o$K>H|u4DO74%eGC=KglsvE|t|7#C7x`eR38 z!Oz4X8Sm%&0Ar%;V9}y5JoXtUHRW|rT5gKp5gAo(jqxEUZd`#pAo6=L?nAMdD7K>h)b4K5w-uqY;)2CW)x1deg;8_}Sb z*Tw_f=?I9ts1X;fYop&LtDUb zcN{_;8d*{ogY{syAF$@vug?w^2(avhzqDcQRb{9tFM(EvUBfHq#j-1{Y`0mxtp|5! z2bqg&9WRwbcJ;wodQV+uMvTnoKyOB!{# z-uX&A&NSAEJUcA}c!)93spBBqY}lS4AEEnk(y0ecz=t`uEooJ~7jFG1A%?}4a4o)! zwkfIEeey(oXDzvoN{B=>Jgtme!O#LCjFr}45+4ny`1DaY2bdlWdS10yKkaIh47s)D zG{F1SA0a56-%24KTIdDa<%yAJw|@pxh~UEMauH^#(Gp?^4rS}-Oa7X!Xpv+uPU5qo z4+>2>OaTfv3(E_kzG{hqMle+k7U4@ZX4~IUiMt^_~HXuT>g9 ztKxsqq{->gUTjycA-P86yI4tPNe1FjPq6&ZtvBGFyr+rlU0g&tpuTyR9lK_iIYPq< zsa#J?$N8bM5Icv@gS(Gk_=v+mqx8agx{JFK%YbMU;D$2NMXXP>l$^$)Z;DpmTw8Ol zq*#_w{A(GOdw6R;tXGybgHjX7I&n=+I8S|p0a&ZQF`|59ZVE~;MmG{)Zm4yXrU|OV zM5)PmR9=?s8mA*i+b3k9#YtU|y+9P%ex!_shKsvZkp zl$3kU#sjxmC-LIx2?DRKHTRTG=Xb#(yP&9l!w^KPt6DQywtU0|Na}5F`G`~srz4Hy%+(NR z5$vD#9WGy*e}gbC$hg2iWsyC%lB$Dv9p9Ri0M&jMy4HJOa}zaB4s*-BVKLExcg=_+ z99b*2Swf=7+UxOGOH6!z7R%Y~=jQ~^ zs(|_1^-**=1HwKK^!Q(Cq!TWZYpKkYf5$e%10#tF>Bz?%0{4}GujiGeuHMK(##$~& zyrp8Dl^(1H2ULOa?BL*3aa8Y=B!y+;ZgP$%Rn7;a7ucoB#A;uq5pb9` z;kr@paJ@H@kqG2iIH9}{b5BBj>lacdFg&oan<1Ftg^tvz~UI22eo1PI5Tc`?9 z^Z9zR|5h9|^7oPw${|Y8{{8C-{L|{tXyg`$3EHwRsPH^wa2bf)Ox>r5r03)Hfb^ep zj=#as0P_E&Zpf+YodrWnA(8}z_`i;g{|-Ce{rYqm{s{ooGJ*i4|JU9;!2i`x@tt+- z<%Gu({{<9)P%E!efR&O;G}e?pWEfGar~RF5L~~<9j)gGoN3J%IcEd}}Ql@gT(puw| zbNC7}88{gG^WuD4Cks{c5Ma)Z`|<1oc4!Ag#mj~ba6S;-e3uH_*-gp0Z_or*?w$;b zQo8v%6WKPF$-X^b-^^X^k;&8C+2^yKOvnf?*!A2Mu#cn3%T`^7>$A=pv zhIKICUgqMS7k4{k0PLelorH>7XznoUVcADpZ((9I=SExj( zP?{<-hIQW#bxjYG;?c3hmOTRf+z-pxG3IJ>wI`Dtk8(_MKi~*nbyl3^v7VR8Vg>o6 z8l}%0tm)bIXwu46jN=rLxTk^$QOP`xE67v#8d+uix9X;2MKh`OOlZ@Nk)OqBN`K0^ z5(x|7dUO~RpKRNPow3MRmmn;vK-l;%9ulY1f;bYC61SH)=w_4GD)<9=p!wiW3|Tdv zG=zGBxHB*|&xrkCEb2hj;Mfq+ty54(t31P8gnyIjcc@~uD$bJcadi9vFPHHI9k^>H z6X_W^*u&rz#;8YfAULY+R7fN^`nyr>D0T-rAf8Asp!G+R9X3OK=1n%=Vqo7Hpgq8z z&+UeXSs%5n5b&cGxnmzFmPMt~pg=Ke1ewkbJ|+e7vCCw=FM5Qg;)m?Qd~dRkRiH%R zIYq9Ne7p^)F_@*&K)JCZNR7j7`(N$8El(!X2TpHL#sPP6eAn_62c`4}ZimH(od0}|(@ ztz_Oqt(bLGNRw%GJ&snZo6wTE^$PT4%cZ*00X;WWAMF;Mupm~6FB~j&QpBqn)&(Uv z4RyvCRdkgEiToE3mq9*sPILnBtRx8cxEczwTjVz$D9ML0E-FzbQZ?B)z?&OxkRj2^ z?|vd?v8s2N}#7Ae61-5@y^q#t8^$hMviWDFNgxPFJOxe{Y+|UxUStVA|z)` z%VPtD!W_~^t)`xB_Jr`InZxm9*TC1nsbu*vjd^aPu*HT^%zUzFi6sd$1tT#_OLhY? zfJ`4FixajijNpkHiz>Ub)jQM`2&69dv9{*rUgfzBwY)koIv9g#bm3d9WIl^(x;lt3 zSR>Gw7c~WzwBQ|Bqi)*1>q>h&Df1qH&~!85aqs`5smeos>RA3ZZ+Lxts&$nh&pz=y zUSjh=<;vz=Y|efp3g3fI2xpiw&`1E=ii;fN53tdztIg6tQFDF&UJqLU z^{oS^d(}x1T<|L@@mr|nRxI^q>RHwvRj3rhyft7E4)mxj%Q48p|5g&UKHdcw;BUBb zWjS5t1bRTHG9}|vx1Nz|?w?s=g8sts<8Gw28ctHoX@8T11l6+~;V>s;c`;&Sv0tlZ zoKplIP4L3?niK~$^Tun***?CHaDejmbKd^KjrMfP7RT2ee9|ufP{7r%TvwCN_OjUZ zvF*Amq|8c>&1=ztS=|vZW?aA<+1S)e$X2|`1#e^XQ7c02C6eJ%II22a!4;$f2(x| zlxlRpJkDgrJWH(lD`qi|FA_&wawBVmc&gxJ1iu*PqNbjr+k2On%lfsBC_1K7cw7K) zkm;@K*bXI_=3KHKE#?1$3GkKjZ#gtjaA#3Dz3Y8lyi4=)c)cif@`VDsdC=^pb$h&D z&(GUhzpZjiIp2Rd1%RGD%NR3ELlqNNH;+><4{6VDf9354AJ%}{k$37CxPm}Fzk0V> zufwNDPa-6R?j+&25sG!JIxr<`)H9}p;gCw#{u%Lb^;M@ka1LeedG2)2{%C8wp@sKpK$&NtQ{LqF7 z84WpDi53#O2j&vt^+X5rab_v@a9j)v*!pr1krVoFmN-R5Rk}-QtkXigruMO*$rCJQ z*d?)QUu-)ol{kK^6DNg^Y8J+L1I@p_(h4FB!OE&UT)$@}T-IUl&@(;}E)WaB_M|Bd zNO)kmR?rd@{n+d)%?7U0HY;#a+Ou{lYg-Y+;Juf!{(e+h>q7e0Ns zCck6NXjYL{$wuXGgT@&koO<>fgC!#;>AxZFq0txITlsrX0aTRtsa#Yrl-|B_Cy$~f zWh`0vY+6$p|9XKoO5kwCHo*OTpGu>R*Y5nC@Y6}AhRD^Rt#=Iw2Rz`bz5zIAod!W0 zOM~7u`IXiIedIzy7QyUiBc8I+xATt#|t!!C%zh1PtRB}%p zr*J7e{u4f4T7w6+?h>9fc`%wRuC(e8y>UR2SeY96lHbQ?dK^$*zKM5yR(1_qYcOhWl8cJY^}c}vuHk#Q!i_K9f~RlXAed!H zv)F-0k^9O&sEIrrzWRQN{I#+1A)ZPFR(ax0*6Zc(lM~kUECd9^FL$qQ`g8(a?e#Ob z{c%mRye`ZR&>XM@cJD&vxRa4t@+{hMOSHL2zUq#LTf2$jm$7C+c;rfB}Qm2~(SGwPRm8(;E(W^OI2uGXZ?6vi>MR zGe|euTbpstpN#BgYhqkf(mhN$KW9*!Q1MQ+V8+6#Qt&VMUjE-&Fwv#uuosMnWX2-T|8}t2hH&ruL zb9Zje^?7RV^Q?7w?5&=d8#3tr0DZrb8TdRl(!18kHdM_n{-9Ar6dh32T~d@{p3sas z0Mx3)j-OG_gIAc!pmoes!*uZM3;REorA%@~RYhW|wxp%7(w=AXAkd#U-XI2_LM1l9 zy*GWT=W{oPq?>4m{P*>dky#9bf*zqZq_#^2j7b;Ub1i^I%8~XqNj5lH7}Bu7*gwVR z-uL&ZVJcs=jZK#xh0iDEjfR+7i=ram&09wgu40W&aMR=ur( za<(jkBhFi}ZAjws+V_Fsnenky&cN960XuPwu+>kwX@7kG zh(t=tdOc6o&Tj(p)_z$nl~G7A@o4Q3m~zlthz@up);h?04ZO+O`r%NpYSh^Xa z*nG;3JIY!C$z)GWP|i^AB9OqgqH)euXWD1CF6Nt?+x@Ps*{#{%COhbd7z)`()g$XW zC#@Xbq_NBUHYQmPr&RR~Cji!Bgh{c!Q@x{uIt7!JJmPr4U#pyYXLEQ#%<{;mr_Zfl zkLF`T{$W{U)o)o^a6p;zmB?`*WuvnU{^~PSZ0CNT=X-5EWy_g}FK_YU*1lM1;K*0I zbd6$;%x)+2&e>!tYA6oM0a&xqmjbq^YYY?=WU*QVpR9G~a#4qqApkN#E->R-H&#UiijVz>dg25y`$6dH}69Kw=P0TRUl@xrK`> zkmnmA8_KOU|<|dYZQ%^WzFeD`vJ;S?}U$se-w7MVKR)U>QFMQ z;0F;|ZNAyruc@if4t$+k<{_uYH2SEupWVNvM)05=-H3lJHB(!?dmp3`+@&Y73iz%=EWctrW)mpAdq_mw1sYPKs?AKgx&j9Z zw(MK0ldBvNG7MLnu$ObCXtfN|oi+c(|2t@he~WlZtG_W(ss<2Y!hI%}okE>;nlaLG zX&_58At8(u#y7Brgp~E?e5ZWY{iv_4he~Aow^XzB59P;_s=wIV13b8mKN$Aa$BeU= zT@5&z-IH<7S((#mTG&>|IsJ0P)j*H4bvk`tAKt9JI&3UD0dk|PT~OLhRGDp#)fSbt zoZfdGOc!274Pec)IyhZMLr#hiF$NJa=3uQ(U!}}LiPO%Ej{mvYp5XBp;v>|J`sdjR z-{s#iby!~d3Cs=cFh_a@bF@&&#rv-6UZpD)H%1zRlJS5pg@1qc>OI*2NOBnji>q28 zOHv(EpEieLlL^9JrnY|jEAOMb!NRs}- z<^n}FvLghls=5h%0#zIbeHR>h%;^1dmuB0W$S3%&cnJu~1|GG~$5}d)pVf50SM|BZ zGh~Rx3K+BVTlObpGL)=n=Z+laAz&OC$hlMdXo1C3%;qekLqr5m$FV<^vEnb(O;q?K zgX;29N*n#CNI)Lb1jP#cQbG~d1prRCUL&5#+n}fJ8u4=z9Dpg+x#Y+M)0#f| zC@}hdd9erZGqjK)2m42$`Q#GbyFB}B+&Ne9Y0H?aQ+_3z4w2~`FcJ=MmqF>uTg+b1 z*J<|zbNr`mw``%)Z5(pZBGq#0x`pC)sc{{ma<(@20gYzBS*y;wY|o?fZyxkH0LC+A zZyJlQcaPWK4v*L2wrke*`iJaE*b41y@-OhtHNDBbLzYprT5vJBTBAUNA=ZN^t}OV7 zX3HL22Q*0Q;ATsZ-#>0^`Lm-LEKBZI_;L8QzAImJs}tCw`D)Dhs>O$A%TKpD)l+B) zjio(V<7M$Dh*>ObTq65avTWq(0TM`9QJ^Qz@)aA!xM>R+GZJ5&-!MH|=>&UtQ%?n@ zs0srGO)s*G4yq-YiG#wX?tN-%x1QszlNLiA6+1@RYI@FvF_OQB5p!lWmR!fHqn&V1 z&~nD}aA&wrAp~!d9~HmuS=RcO%Og^Y=5Q{zVT%`Z>_57dtt*>8aHP#;U@J_{Bdv9k5uT8c1OQ>06n|!ZakT5JD&@c1PK;G(pxVIl@;!MG-ff z2jxdPYl~fqJ@P5gpQ0l=fSv1h+j~An!D~LrclrL3*qus7Iq}KD8|x3kcVVJg=#og2 z%o#J@>aw{l`j1A44+jl9A3`PhK#Y%M2@XnF+FNnY2;Sv48NZ;@c1Wo!i0N2-)6Q$_z-b^^PG^UvBbzv z?4JORXn#ff2(epW?;fkGvtAdFT3S@2-iun6YFjwh8NA@5c{}Jg?re-N&q*dOSM@<` z$jQIvoaW15BJ&x312KYY-j1QYcck4rBFsQ8_Xyjqefiy(_WTt2N{%zW4d9Y{mz}@T zi;L**q7H|=5rh`2=Smn$I~a6%i89a1J$*+uoWL90XL{)8YcQAl zbIp44kJZd~pug&8%N|GAi&2~`3)6=KhKzdo)tp3;O`YEyh~#aFpmwRO)R0nnKBbF8 za@vx}(n|xHj#8+u&TGoJIck!`Ww$mYjI!7PW{w>Ml2|9@C!Up~oiV+a@i zP1>G@sYszO^AWKla*jwvPe;v27E1O|v3ECvCzl>&z+~q&(8Nq)GE|=?CTr;9*{I4R z^?i0X?;4+>t*jwxAldB*B6o{Br)IW14pf ztnTQinySprBdE(Uyr@#=w^0}N0=dM7(kPmKK;a)b8I|%oSP#F?jO7)~QSJi?$Ej1> zTu@f}&awJmZR7dS50}o*r}vL-V_Fssvvbc%G>L`Od0N{d@DEL$S2k=D3#cAr8Fa#G z9i|28Z-0Na{?3fz>Y6YT_*e=!!RujY##spa3J6;&As7|}F}7a|W?KyjUOg+X;NMDY z2ZR`-{?6JOOP29|sJ7n@vCs*fD#oTN5UJwTG%xSllM)-LB4-L@-%ezcMpC#be?;pm zkYy^k3QXfg8o0DZ{LZOaA4LFT^6g9+H1oe5@X+P(BCFI+FE&P8#L#Is9P0IXR)EJK=6H^L>DC|c@WBDKU#$n-w)-wLk# z3V36*^O%Nyf0?koABqgtQ$J&^ocM9NtY--V90e5zwzypMV2~xqZTN%~dh!_emh9u! zi_X1%3!8A?u4xu5CYaK*CZe_7TLfg!VsoOLcEXql6t6;#oavM5Hk|3~Ex1!DKwj3! z-~=GnTDMxrnt|47^@Q2$`HA@{gG8dC{#opJKN)>z2t| z$hyizaC*0*N+c={nTX_2P+~ubTY>AC&<&5y1E$rDG!zobaO$vec`evb&luSdfn|OF z-A%n+opcAd=$x>n=Lyk`;?2fkpbM;}&5`{-lChN5>4L6@7n+m&8kxiBIAoM`!bcst zeo~-VIcdBD?oMAxNRNdrUhH7D$SL1Ij^g4+_M6u(9i_N>@acchjh0wgCFIn1rSJLx zvUAZy$c*equT8LuJKPM`2v-O0;wTb_jp0*ANZuBUHHdh4TkqJ@A6xk30KG@!rAJNj z)1xT)ZR2_ix9c;zG|9G8izV#Or>(wXxmeS`SnBUO2!h+6bs;hJBYGciYQU!l&V!bN@&-X0wCVLeNllS8hB^F%h3(UvAn>-L6*}Xy{6W(rV9nPpe+< zbwlBmoLrl9#CDpodVD`hY2Q^WIgrlIdASAOhk>-nuBKuNfkv%^tNw@^<0j>%rvRU} zr_e#hY#@iAL>`GJIhk_0Ybs0D*oq3Jbn_j0H}ZWXq+z2(iD?lx7l_f=uP0vgwnTM* ztUiGMadfI_&blfOgnAAaP*TikmgC=*PKJl(u~7;_(-k}amHs086Z`|&OSTMZ}l;hgRWz7UBZy^;X0Q8^PlC?PRFZxGr{dn7SC^sUvufZiv{%oCjv?wf-!TMtpcr? zpqEP{<9~2z!@(9M=v4hA&X>h3WQD&`wiWhbPo8q#%R?x z$w*FkqMx0~6TLX)z8AOD{Uftrwxm|x-8(*e^NvP;u`JiicbG} zDXPNN$zxPq2h5XHUMa5ChwmkdF+R3%TJ6(vWqgStH!^GcW(qHC+T@|OK8>PTws4?fln~+fmh=*&8Z6dLsQs(5~eH-t^c}G_6Ztx1m30`LI?X1>dT5T+uC39w<;&{ zt1FqEvb2z&!D&k)bxXDJ-LPp(WBxmAHmNBuH>tIeIRBw7{ST!ETOsL?5mn4qq&M^( z1e3gPYgIZ!&|x}aVN>>TuF8Q$+GX~8)0i*2*pGuf%qFD^;}y>koa~Pz8mPq*2`f)_ zH@1O4AT(MSbkNup*l22wSo-DI zkfOa}hSEWPQz||mHvv{y%$`qYA!C3z<)MiU#H;j=q%|SX9_j0(5+K}RBfdm_*{8)5 z=_~0nOJ?$kDOb4e-nmc43EIW+mf*`?b(I7d; zw|zu@h~;pb1)Bi94At=S@U6$dGpAlcqO<8+s+uV8jWOOJ7L-%?v_KPY2CNfE^6%N?yYKid-^{_%+ zsogQbXkk^;uZ_Lh5f+nS)rNbbC)hNsf5wRLxFlhmdBcs^7ot*y9Yi{N5tCKWbdw^L zjf=EkhcZkB!>?v7tgutVl*P+{SSqQ7qC{21(JP8&or2C#s<$v-($Rqd_QE{}IHqHV z#~GG31F~B&m$GHZNa;>WxR%i0Q2m+a{4c){h}ecB6y#n|v#}5fury;^T4OHvQ<8=P zQ9tKJ0}95ZbO}b3QsC_3kqqV&#JaTu&nGJ!-241M83_W#l?&53Xm2V2dB`D%ap9-Z zA7Vn~>o@)XV33t(hoV&x=qB31i=mof_<0x(tiz>56kcqmRHUB!l{kiLW? zgYCX{@@Si9HImJp5}YrX`F;I_)kDs`4HBqU~`;?a#&S5 z?X0zP9o*+zSdXJ?)RC3WPeL9uCTxVA)=9n$#%srVtZYld{-v&mqfWJX14K_$xDLhy zCLdjaPCN=b_5fUqSA6)3c4~13M$sor{hM&To&mL+628H8tz1Fr}Py zgyz?n_>I9w6oc!~mEWK zm&1DYBIh2v> zj5bful>wW~9j+y2cD=q_O*tLKhGz*V^RRxZi% zjXR5x()+Bi73;(@^Yk^ATqPJb+$~1wDTTmi>jF9{yEg?wOiU2@SOoDkCmB`rg~pQ9 zuoWfyq&y(xIkfQ@r)l55g^g9&q}b{w(`vN29rB zwE&)3BQ%nODI#l7!3Gxt;w(J=bAae0J>uvt?s-0ky$}5L{uSbV98QAcC{>_3X4-Rm zOqv`S8$A;a)MrtjXq!Hn)I7Dl21J7hsLZwsTE~dKDgg#1f;e*!0neYc%&9_H{;$rY#*gKlwo{_(4mX`+oLtQJmU;(1zkdzopr zMU3I}RqAs9^Wi4=%K%O5MhKmC7f_z-sn-9F42nzF3fN{g&Tr33e@qE~If#9!C^Gvo zcl4pH$96563$*TG+?y5@0rV3EyUCfDL48gN_~wJs(-~a&BJ0NPpu*=x+v42IMT7GO zns2M0h9)fEBepeWk8cp<^rGm4TBFa~eCiOMz?2`+M& zPbo}|6$60+sfGD#E_d-|;F3e@6C4VW$}}LBOsREJ+umrGcz}KWgTB4AvY+#MMIQlo z;xGVeFZ42HwXxFX^VX}#AMYA_8#1a(4bqX%fEaZf=j1Ebkz(g(Ow>jxW1L@gfxhJu z^FG^684_FGXsw8$y=P8qR1cDOm|nv42YOv@oDhj)EebcPlsLFFF!Sv<0rSo@fK99Nv_3Nl#33&bgbBt$i`Q5G7Vjk2Yu2Y zS^qgR?nILW0oFtI25Qgcw^p$g334VxB&276m;Ky|Gs&I-ix%v^$Wnbg0npgEbDDUt ztr81iJFI=ej55)A;Bpk<$~~=da&XE#Cf(}V6jm$uejnFwT2k{EQ$sHX&%3*7Tn>=? zr^J^Lquo@jZV1rOljnm;_ugqexC8x+AJEQqUP$ zGgSJIB5EA1bfM_vxRqc1Fkv*z5n;i!%Zs-m4{MB8;=<(u5DK$k-kAa}Uto`fnr8n~ zj9$>giJM4Qo-aPSjRqa0-IBLP#)HU^CWQ<8$#71ShHeT?Pb3PL zvBgwiJwn4#zvyh(b z)*&e@F~hSv?-jb9hH1woXCN4v?`xj>G$$07p{>j^KzQg)YAx#h=zZL&>NnrG(*YwY zmDv*d$5cqk(X9oSzw@#-s)q>~-R~qpai_et>PNkUPf4S-K-F46A)3V=2v{};6bzfh6wq^d5U|c2Zo=NuGd&53_nqO(dUSLJQcomKv zsc617FBr(|)r0w9e0^4q>B|U4%!?p=2{QWC8QAhe&R*&r|AKSByc@|JMd1L6ycF9K zp8)jsQ6tLBvVTASX$(YO#Koqm-c7~Ef&Z5_fDX^|Ct^KdS{VVKS-Mv985{~-flU8W z@y-IGvWb8c@^ggMGsApFiz{UwI%bbEroOOqi7Jk3cvQDeyhw2%_aDUg}bO%}Bt zXs`^U{4qMdXK%MViDkJcL?{QtNhco&wVlU4sd-(D6F?1iLWrs<_=-{4+1qjfvtKuz z8$^*$nQw5D?T|eN`Dtpv*UG!Y!^if@8T&MM9D1KQRQp=HdG1(v92XDktf$@};E`7A z@`gnT=E@UzCE9zXO}Yu8AIf#WB*k+E1Y-O+%tV8w5q9*3Mj4-Fj=jlrz(cGG2bYq< z%OL4=vb@Xu-Sx9Nadzw3e|Upzb1+HuPu3Pf+|VX3t#U+&VqLnsc&LDs;{xYlzFBMM zoxP06FW%bS*cJhHWQE4ZyOJ_Xqju#`4u|h~;We~e5qw#A141IQKAVpnaMA4vV1Vp7 zD)lRXm)Tb&MkIx1pRN8UtQxqZzZ+W-E9WEm1+=kh=bc|(4iYG5 zA2P9xS^Y3{zBqGnq=M|i`j}Z82w>`0xa_GAQ|&IVd_4SjG4vWwfdZw;OYOrSJl(d2 z75HuM0T!UwxIn!jY!)GhT!nEQ5E+y-%Hj1R5Il_*kP}JZzW9}Ljv3=KI7b0|8x;Q66r0i$9?#89I%&?9YIh33keMRMhR3>b_M^nH zqpelUD@07j%=+2$a-h7=JmGG37M6ez3=ba#7k|SUA-o=2(UaV%V=-78_zrW08j=0J z xs_nvAg@=6(H3+c;VuvCqZ*s}a4_h<#*E7Fy%E*Ch{P2pr~)PP7g%*33k5h*Jk z(hNgiKT98X^jqegtc1ZC9?(t?HrjxzbkLdFa?$!5a;Ad2Cf$(4TqNe zHOC}xJ#NMh1D+Kx5SLUi995t8&zoEigxZq1W8HW^;3Xosw3Ie4`qnZ(=*Jo=<(+&{ zbauIWiwomNqPfYoZ3$fq3n#vtTLg*Z zK|B%UuRg(AaYMGNQiS{|$DF}HjW@EPKL++qRUW$K^`5#W9qGrX?cPWoT$}a?MjR!5 z7!W@#q}=e<(FKz2F2{=H6oedulx3X-h0OdjAlk(P{ow+IkI6)+$?>F^gLu6 zY;66#^&g5aUBqLiWUB16$1W)P!6?N|WcZ;M5fzRc^o&1H4;Ve+n_f9H_|byDaYhXI z#c-+^3o%4&XrfrAyP7`gPmiCkC7D~gtTmdy=oR@*@su6t3jMAS%#1doflFnH`D7Yz zaEIQwXg8f$Ws7}}+6`Tj(FD#_%fn-tD_I%|AY5&WNY|}4rP@IA7gNWDME1i{oUUu6 z2;x-qd#$#X0cVQA$Ruj2AdS2Y5t4DIKz<(C5@em@Bu&heacQP)f96=lyE<|p@xqGE zvPTn~-^ezpw?mfiOG5&L{V!M~uWw>Z>STkL&My*4@{0ssD)HN?=zo@kZ0KG0Z$6y(scd&npFDSmtH#Zr3 zg_es>jy|dywZ6Y#Kf0aHDmu+~UzErUIloG&YWOfZ-BS0@h8?vxAvCsmM?JqA7YS=+ z=GreP_+z0j@M2(X}lza_8|E2!bDFcQ$trk?()QxZokZPscqevufnYzpE#|uK%_1SGo(QmbGk6 zRJC68SK-+w>y~%XA$Ap+lx9t(xm@V_>+O=){>Q>K!!s&Z(=WFW=sLc+FJ8of^MrsY z;Ul(k-`~6q)|$o~uRh^8!qvmjJ87RdeR{*?Ur5?@z=R2nP{qugtDEr5QI zSQR(+QlnM_n=uP2FiyZp+xt^ns6WjmVaXo9S1zv3*7jXeuGc`r4NH^+fGs-C|O1c7R+~(2-&ifH%@qFX% z#Ck#jZNjd5A^eY1F*g%hWU1Hp?V$1urxfK@(8(cM^a#%-8WsjxJr)?_X zliS~_LBvY($nE(uUexocm)vIQo0ra$g8m?$_PQk-v71coTH`0|Jp@*1ItRglwy*fB zRi>`ht5@m0x6Q_`{$~vsz=+0}-~te5%%e^-PfL)~#fQuxRF~lJZpE@y7;{|igV!(w zHSE4k1sS$i9mIsH77+;CHsLB5Z;#fSxZ%6Cq(yYlz+JiT9#8#jnMG%x`N+?EHfp8I ztX+2gU8!ZDEoE5a7y5D_I?c?$pZ1;bC@w)N;6(T;qQEz6@<3iSQO6XiI0IfMeZT!Q%R2HZOfRT&X~>3YF@ha)bny=;HK zFmx7`!0PV*`S^+0x{TP{%Bc5?*}}MJV+QjK{zsyVcLpo?2BQ#%c(Qs?^$dSXa~{pA zHP$p3AnF6$_|eWqAK#;+-2#D+flvv?s|JfvQhK7Rhc?Lgt7MhNM2ZI&vy3qey39v6 zXep)1FUM{pi9>tT-VZm#&J5OZSB8)-9JJTFhjUIV!sWGBW4=5hZ=n2@`!{E0q0wW8 z?7Xvo(VB3E6U)r*X)_sP=McmY9;nFKo@(L*7#HGJ+gNV*e=K1kFapNX?aS@<>R7V+ zZN_VG z%=40y>K+$?f(zgT0o0i5IkHuCa?f>*v_UbWh(g)NMA|WCTO>XE?xRXSj97mMSAhxl zbsYFLy*O#kQ*B!N;{;P$UmTa4(Sh(7>T(3>? zfoNBB^yNJwKDOV!onL43``msUD(#Ow^M5>?y|#RN+Af^0u{~4V)h_dS-t72aom12= z4{v|tB7CTJAOPQwXwC25wj5D4{<}YT{xOcv_eF3;^GAah<*C7?(#|~YbsWoedw4v_ z@PMY)j->2|Tuu{+&LaWka$OGkpcVLXpSJ_)hm(X6II>~LoC0KaIK_Xt-*-PG^4Wq*Qz{2>Ja5&XY( z@JyU?8vg6>Ap=GR_7=_t)_OJuj#ehlPXFK1BakM|b;pfXWS?iX+)l*s3e%)6Wil1t zL^e$^DvPQ)cMgw?0PzVp9wI^Tos|Y4X?ryFkQF2WWUMZ^DL+&v{ReJzcP8H^FynS1 z$wBeU&1>ZA-tKAtng6SC5cnGR-Iq*ptEXR;(b2AzDoWEcNk>*qE1+c1^&TM@RK+@C zb@J-7wX3cMH1ry(lmo`r_*pc$i(6iF=2D}Z=PaBIwQSU>@KVPUG9N$B^!{mSgsnv! z%1~KSv?!)6U))`OUfz{NUauYk3>6|JG&(B%XbBaV3@OqnH5x@ydF<#SDI2$#QCz1< z?vqz1Tpf|q*drrWBZ!7IDpXNJLZhfT9PO$l%M>iM(3sgi_?Zl64tX}KT&wj{s+Q@DJt-5akedbm1 zkJXrvmrol^eUZGnzteDs-9j(}`1{$61HnoXmv-siNjC{Z?fVz-WHEx-$ugPEn{7pA zZB{9-b~+-GP?|8^NNenW*Q9pNizO&y25|M!SXZJBt^KXMGS5RD>P~@pqW$yhrjx#K zl>Ms9^toMy%q!?h^2!nm7@?sPKOGjQN#osRT)CLuGjyv9(*@$5;x2qfth!Orvpg-9erT4HwbFiHYUKKtB#aANctmrf(xIwl1HZai=XsM#jjAky@M z_e9%Ki(GqRmFN1b&=5XKO-_rbwu0n$qKD3m8FbOwtMH$`W6NNVN2aOcX=ni)~*f*K@Q zcq5SQ>vXYDk zw`zNTa~}%D8SS{=O7!?1)V1gbhhOG`E-f`nTHtMc5DXoh7_~SWbWQ=9ZTPEqx9d9i zuhx3K2fn=|^bN%yMCchy8idW(u>vZ5>d`Ar_Gl3j_0cbI04O~v*^n_6^#|mM6H-Zq zs03Ismp$&5(A|hC4kDTa$kxY4t!^#Hj6_{V25v;{;{YQ6fBMTIYn(m!1GJzhD(YTa z6D`h;K4Nx($^mHHt9>tEH1yz;qs{)%3TFLWE_}dv(SEH=DO@&SJF+jl?`N4#TNBRW z5~xSl9sZOmuz~kOzlbqE>4jh9R21tPgqpmOU3j&6{1!_!T22o{ju1D?Gf(nSHV@=) zlOrZHXs8miA3jnTi?V)=RMo@P!LzzcMRr;$VQaf&sxKGEGJUyxzC*KaXmZPvyv@Mc3>3@Qg`|NsevKQD`L|D@*}StRmA84k zGyEdp&*M`~BrgZ_mjKw9-2t)&Do8?E_|Res7PK_nRElb8SmJ$%EEUH7I$ajd9O{)Q zB@tXW0TDAZPF_I#5pBnn?wipUc*=>+YgLG@ivqT~G;w$|9gQ#1WBYy1jtCjr?6alHRg_DiZZF$F8l?nWJbi zBoCly*tJj!d;y*{LDOM_Ug3{858-_s#t2A{5(Hi89JHi?hJOe&i{+oCOIr$oR-f~Jco&A06+wZ~b_hf<`X+sB zF9p1t{2vk!R+fn(xr72Zaiv?Yis7;}aWXl)g9;`gmWA2qw!*o7f1!)wbcRY7Dj_|u z@xu_iCqvJhg)~xoB5eWoHGa+k!=V5`)AKLX!qkY>B@441gEgoqW27hele5KwS19V# zz-vF&9cwzQ4@7%QTwZ!I$ha8S z0l$C9dBxe`C%lO5t}j1hT}`J5vSwB5VQIT)B&cfh7HM5D16aJE-D1B?2_6eTDJ$%0 zQhVDCe25(6MyOfc0VK@u17AKQ@12)U*mCJ%>H#r=k#T zv6@g9hrgm$Ig_VlFO>HIoSx=_;r~&?KqD>Skg3OzkBenzpB6&0k!)gu(gim+YB{f5lG5;kTMfhQ>t>M8*#Qc|uj#v_;#YQzOh_j`o~dmkug zFfX|IgDOk!1uHR18zEb?fSDN8RF%rT) zhbMoD~NP(|EsyjB6{IVl#=tWFq_1tdkj?B61_g zOB$tiUww)Qevkc?2O_}ylQx!13v%UiO4qLl*YruDF%EWYo9LO3Jz z?w!bw3OHFY<98NF7S6mHEUud(llR?PMFpK7+$oeSGY3Ma%@-wGiG0U*IaU#yEo#sA zouRhqu6ZSf(J8Au(b`X=RbmS`X0yY`eFKkgKJ9^K5}WX1!yevA!R+-pxUv-BuT;nP zzuXShaxVZTH%!30^sLL3czCWmSzeDSD9Ix6MAZT>^Q6&YR1FiT&KeWqO@M+aE3; zJAiv)qjcqscC*<9S@SKXhBdF3G~_c5fzGt9W7b)T+ymg?i8lWe?3F=P`)5+JfPqWJ zHg0_^@|bb7^<2jIrD0#sarqDNV!~$FN3~yq4Oah?Q?oHzLW(GVgFJJ`c(?-NY;$D5 ze-}8Nb4(UUa4N;2LelN#3vsKdSeB4Rc7zdC=6%@sZth(Rk~f_P=2pw?mw}z+)(Apg z-eBjhc>vYV6O%ypiBoH$r~1*utGl30Qs&p2AVRX=gf~m3b$q29E=Id7aW6t^0PNuY zD7d$PdOKp)WE)D{X~-~j2Cm6`vb))=4t4Pbf5MPW8piCh1gM0tD+>JVjorv}&hmmo zuisSG-+0;H{09l>Iv3cSMy^b3E}&OXG1WO0K6L0DtzA(^aguIC%k?(9 ztefd!8J{s(ynL8i2l?x(K zBG7_49$XK5SVcogG~x(o^t-z}NeCIain%iMh7JLn`XnO$U>ZjIjf*q?Zx;mLF(!vU zQU<@23*Qy)cmG5DR_-ng%bGZCI|6G+{@l_XBl#!8|C#Co(nQXb7(qbpcR@gu{%@)N ze{(rla@_jjQP7Ped=Z!_p zo0uAk3kiLRiB3;;7u)N4^(u>qV+pj}L8}ZIP%DaVE-o%z6@@k5!z+I&0tBk?>tk<% z>uMh6wXaGV)O)LdCKx%}&l7rWtM&D{k@JrHk>?y9uWY@Xu%vgML;L4*4mXuQ8hbD; z&^lUX-=l?Y6^hw~)>vd09C>q;#tNB|FX$9=iMwGkBziw+m0n&6u@UE;xLU`MK?Ux& zw<3EZ$@ds8vEtaSobDB=qHZ?sFHQ&qyIrU$VsF(^Cc0# zC=d$)V&R_k#tSw->oq^+s0Pl620scv>4WGtNlRne=B&0XuoMfC`C$K|_6rf| zKS)FNc8o@9RNg-?J(-O=rgN^FEVOW-)nK3UcVOjCnYD==(Nb_l1(7z1HBtKpWWk6~ z*E-)4Zbh7d1WHLdpS7rc{Z1OC!KrL`~pZzF|WIv*K ziPT`N=O9~KAPuHMym80`cnAh!S~*1Fx`{Z((2nGQU_r4N*rx-(*gb&_7Q&hA|2#~i z;60$C-}ido@T9?);A|y|8^ri?nH(3ljN!TQn;=xZLyoQ)tp~ECt?L2^DA^PLxtxs= zGvZ!@Sx$RaKxEf1ub92i#(5p>Kvs~*Xx+OG7{@0crYvt4mgSaR2`KlYOw z>$F*_hzh(h2_hgue*MUfD3^?4f4df)g6mIEd$1h_>w+&qm_--FjiJEQhlLozr0G-1 z#@aNLP_dzrpqcPs8Q8!IHNdz;BKke+`+dED);u@GP1mh3S{Z>9SySSiUTlqn;}wKb zL|A+>bW&Jf5-(!>dL&M~(k!CF83C|4n%)NZ0CGXyR#lAAh;C^lHKZaC{rFm3ABYSa zv^WyEC)GoV28hYOcK)b+gS`f7i#6*9(7_A~X{%(jr$mU?M6*+X z5D}|2A=kZXj;G6sv+YR|MS^oE@*~)Qa>*9hD(6&p1cW}cQrrmkZ7Z`Us=s=!3bZ{3 z8T7FN=UJ~QIgW@QM!>>Vf0<+qIcrnz^}T01M9yT<ff9v-j##&nSEl3;B2ZCu|!LL}BtH$(ia#@*+K0D~KCd zAx2VZo$i{6z^A!4J0xjcH*7-r(&@P>6N7aH#-mX|Yy1HLV$&j1j3l|e*yxu5J9e*w zKi;8ajVbO3Bg=e6yh_|oTo5ea0~a)gYO{joK@;v03GP9&QsL*-l0{>O2O+gV;&|?d zpb@pflFXNoLN*J6C&VstaFu`HI(08J?$m4k1m)ZCLq}S)gtL7rkwk35L*uXZ4s!q6 z4pM+kK?CM9lRF!y0QfW6%I=Sz-{pD2TuBAeO4=;JO!MOA=~(MB$6XNs@(0N5Xck;= z7b*|TDrdW}bRj$W9pWW+Yc6l-Zp#yp{Ihnb zuReN=uLo^6o*mP#!U1EzfH2lvZ~@qp47btgyEt^ND3t5(-P6d7CWqfJwoBc~i?CrJ zGKw!<)I@}8j^X@GOH~D3Z6H%%9K%x=blGj{m#br5O^FWFS#d_>yxj(6YKCE+@Zd-O zSYtD?<}1s%f*POpsn1Ky#DChLOQlCce0~SoWnphm~GYyGx#jqNt{82$x`fsyaJSNj4z^!@AJ7zln9KSlyHi6lcBqqaPUX~xYBrU)n7K%u+4SM4Gu#z zHh*XmG0s{z_V^&yX$+k}*+5?mE8LOD z)Q?vY-}rt4h2e11@mzWj)NU_b-3@x(4LY&(=gzuI7ng4>4K%Lap1xCCIeoEeB^R5g z>-Pow^Qf+`Av$^l4cpti>lX;jiAo~ zhldaW;bXG|(IN9rzF@Y~a>aF}_ z&PhY*HbO%8Na(ouoK1I7^WUF+?>9z=AeY*-wf8xP=k{de0SDTK$BINC{}*595Tpsv zq~W%0+qP}nwrziH+cu^-ZDZQDZQJhI+0({G{JV#`R@5;wE8pjdZL@&;IuCBr2Y<(C zN6k}+rY`}KlNRf3l9ib1ziqsITuHk&w=lVkO06K zs{;LUB^E0di)}1Ks+V$#HOoF{LZrt0;Jcp|5aTM(;%*!sLa@Oh4XKv;ad*;g`eiYRiyW&^s zC(-QfQ-y%ua@ZJ$U}knw!OlOTNrOa~E&`)grXn}ME#jyU+QfydQb;rqG~+9-`Nf;t z-3Bn+36>wqZeACw7S~Q<;Wcr@VwOR#rWC8EwX7uY;F?aR{?MrryWvb zTVGqta=-}vop*<&9G0I{9G}Hp$&Af0$|16eoKMrFN~xb3D2g*T zIi**?Y~%DCa5-hkHg8-;Q$yLHHy-7V%^)6k4g?U2hcy1dkE}tHa=gt#)GRIdN+OwX zGanlj6z7;Pg!;;bqV0f26#)pL%dvl$jGFIWkvOXckzm5)9>^^DtNjyp@~tU=F)*ef zFDI!VK@!~~&%B9H$oRjm;6SEUFK1c){Xf|f#Wo53sRw>_kc5s5M=ZyWbQxPQ#NXT9 z&MV03VEJRWG7HyCcornuio46|)KE6N;+cP)cXPX`n06ShcXT7gp+yB@Jk!@vUEi5{&82l8QEa2g2#}sK}+#(tj^EoF> z=>PNabc*&fTUWwR+RiRU6XzX4#KJesDPuTmjZSOiT_DE>aZ)^iC~k#JISx@%i}K#hjd<{nQOgX7&Z^B}!s3 z(*`EEPfePaH9ff7rq`~he_L1ZEOt35i055U-} z2Ea`V$_1{$Gr(Y)^k)#wa5qQ`YxT=7O>v9RMX#XWw>Zm=UT3Uv9qHGt&!vnN;%s}!$naaxthlyKU>ce!R zzgsTtr;G=3McRQbqis*vuUeY6>l4b@$8hP^3@wVRYh*i+#RJq-J!u}%yyn+2-C-y`=ipk^UuSh4rbhD?Z(hb zEY-{9(FiaHr;J;6vKoJxks{q6VSxNw>=F%Oi~M>zQxTM5#mL$POfX|dzy6%Yh7rgt z-q<1z#`$)W%Aep2qh3#OK`%DC1K<)t5W$zZl7WK1D`}@B?1(yBX;R%R{Pk(w zEI-f(-}&pepthy#ldR{*7_7gk^Ct$UuYybjDi3kUj{wh{7Q3#4@S_NZ^4x3LIP;yE zSTcae&{qsFqB1=KFAhLUnpS6xhj{Jprz>Bx!nYFfbU^YC4yl76Y$i*UR7cja63jMV}SW@^uN;3py2YOi;V~`eO#{j1K?rx_6w(rKOsYv!UYDw)&aqH z!Hc63>a%Mp2tFbEEDvHj%%uJT>7=wz2#u4Vp8n(Z9=e)hhhj@!LHVhsqaZSI?#LQd-VR#OIMv;g=9^1CTT z_i_p8m$L+t(>z*cY;aYwEzT(|!&!iHj7WvxtYH_WuT8Hf#r(p$O-;R8HUZ!?pf$i; z+%UEVgdtX94SH&Jsvo^IB!OpAbO-4y(bMd~1Vth^3UK{zz+1qOu_8`{l4sLEGFTZ3 z{U8$KcDAyke}kSJ_CI!A8V$~j8k`$?7Ta#@IXW-*Lq)IE4-haPC(pf#@ z-H#$|^cPb4?1bdI7tIzY;UVzzV`$a`*$bBiEHgbAzRKQ=w9{_5fwC}~&sh}dR zEul(6704^S`y7oGV7)u%E~`O9fmF+bq@~X7Vtu8mHyAAnZnKQ)be+O_FI=WoFf8%? zAGk31s1LV1Oe_~`C!jit;io&&V#=bbqzXT$=eo*nTxOOBmBB2)=To)N3Li~k3q3Dp^=bC%79^k zAaRk1*}{xm47UjUrmCdtmm`%O6?YPCAdcxp@_nh`g2I`5OSvPD5e&8o0S zpav$75P*qNmPWCVHr_|`6Kg>L4;qIm2Gbq&B4XHmBNLL9e7S(GW6*M^{?+a8_#t^4 z3_1q3$ysACRc0}~2k2W1=ztvv4&F_K^YypiyaV5$A{Z?dJ{~pf@%TUb%PFw?_aqR| z`@n6>oaKu=CXgFP4gP^kGSiW)Qj?tncr!ZtBfzfXo>^z3=etLr*dsvA?nM(l=NOTJ zel}BHJotK}?mM_MF>aUiYj_(K_G8pNq1HbLwzr*3q5eF|zC(eW!k)-6|9mQL%4W9E zX4f*v_vj-qw#$+f!F~t=Z?ajLi|~W#3z&mZrH=2HbrGBssvd*p_N3xSZvyyvh(V$MC%gc& z!vYDQG2E9X|N8d!DW5+(yt?d!bnG_1jdZA~A2GWJy(}Y*5Zf=+WxIFx@g;pDStkN< zw^Y@u^G>EsziuVRCSOO7IiFVWU8^Ep4uHAl6+qR>97tS{td&DKF|LJr2-rKMGWp$7 z`Ii5*0GjD@_XRDuw(i}Xjb?|n`)RI%qbB88hQw#wBfK@L!U}DG>8~W1la@ipTm_Ip861wQapc2<|AyMA0i<>uVzr(K&WqK=e%~CqG+^J!xtM#BZtP&9 zCnG#T1~{n5S}aqatFmg!!=Bb^#F_G!NsMe1+Z*#OxP-7mdI<7uv{;;B7|o#|{4Eyu z)xkmM>|0`t#3}oD!R&D102Bsvz)~^W$w}?a(J?Uy;7`>ea=3d^sMGshJrDK^A(9&G2&B_n@ z8Ho&=`m+)`N*{%2x%}t4iYrjdr=+Q&qa%t-Z{PC;_{WWbfrpZE*}dj~2&;Z1qc8@l zh2|1)wdaukpwoqX}=_1dVg2Onki_avwc1~pf z@kjjwgVD}U1jUaDZVD6~8PgH2v~S)5S8w~^H?%(}tXj#DEbp~RGV_xzx~s6IMk6-6 z$sQu`=-O@P%Af|Y7w#qNI++-P`=-QKwC!Hl0S#_}O*(UJ>dzfBpsHE1K)_qTtMe5p z!QNc*9I)B7a@-n6_YKGgsaBJx3JZnWA*(9ha5v*UjJ~!aiDRW67Y_lVGJ!*Z5uP~d z)N4U(K&%J6vY#lwNSDu^^UlEwp<(lCSL?As7c!m+_gKWXcdF zAERViPJm>yS;r3O|+Rs^zLLK(|FN`v{zy!yTQOln@B^Mm{ z>m*;3yvY&$jca#M5rwk-^XGib>6#PSUTBC(KoPN9S>#Q1(|+~xD#Z@Fq@y0kCb6q1goFx1*tDM@ zvN3FmiIXYZk484aqZYxwfnZ@GvPHxud$;pQL{uYy3fHB{nopaf``aJ$O8q2YaE|`C zrM}e;nl!9vqRHBD>g66nHxq0%_5ForT42Z4E~6tTF(3>7GGgarUt9$E0>JqtzCM zcw&F!qo{JhzEAUz0en~io{CHXe#B@CY=10-eI48a$P|!M`p4-|x;fPF5NN!HAB^)5 zK==A(Po+)imaO0KM-N3-VW{3*WX~Z?&a=Dkc{5J|fa0KdK5B52iZmP(Q<3rh=&6uO zIhzE@d#&*M*j>lq)dw+?;@O`nU*>dXRprlLoYp>-up9=UoS>vd6u%=jPkY=qqq$pz z(ZPUYfR|>{Rk=+=afGHexQr5t8<0*E7n759xHH|njz)v-U5vexpSX;RrlT65gWhXP zx^gklkaK5idgrRCoq*RsDStH&F2qLk)R$YA>hbQ^;?n4$K$c9)zm3sR&SjbCtZc|itUF9< z0B1_~)=WCKbD8KB8byiIU@%@dznPYv=bh}a8Rz|!=0NgXGo^Ss-lMjmV;*^H~p79zw!OH5Y~zLoY^_??vuF zV!4bI8Jryvthrrhww4$Ra#65sS&DAj0SM06{8XZ6ltT+I^N{YGE!S*4zaa@FE(j&P zHc&RBlI`6=f>U-#8o_iAp1|mtd$*JADg>_(e0pTaz3r3L`hu)6`^p>pEZ{cqU2%U43f#DL*UIDjDAvFJ;jD>NIRMhRj+LMSNx(tk0MwtS z;vHn00N+*1qjQb#<8rSJRQ)BY?K~)z?MvmeUWm!V;nh`<{$ZDbf0rdMbGOqNka|J! zhEtaezdDAA)J8`!%QtrFi3aV6Eo!rCrLV$w*SOuNU|jSLEfUWV_>4anU-O(I8|Y;$ zrw(YE=HMAp&7njlo+^F~>oY{n0C3M)GVnmnO3%LPYrimqmdQ8ih<`N;S|(n?PL0-U zDr&?$VX*8gaTqbUUH)Di53S;#MHS5NdpLDYemnK*5w2TUxWP`cf5XBBQhp=NH9-!9 zA+$Z57Btqx-RNGcG74M!TFBPQu#&`S9+jWhDf7vVpJpsXabn_6WNIkk0HAX(VI9&i z7q_laCxR%BKW1Py1GJvWE8l}Pzc5v#QTosL$y20;4n2QJ?69~LoXy#aMkGry1kVvZ z6c1mO*VZ^I$(+UO0~{(Txyv<6NL%OY6>4$J(9e4k?4iT1$%kn3)4QbYvd@O+|yG7RfYCT zeo?IQa?8nDX-dm#SJv-4-F$#FO)YHb0Ty_IqQL|Al;P_`g{?!^__sQMWx+J z@viKq88-xk-1v?v6_*i?i@Am|t|^cM4HFFyVn)k?Iv$gHfLtP0Rf|=zbzOaYSZGJY z>!y~Y2Ag|T@^!C`tEq14klSyl$=qWSc+p9P%*Z+eQ zGSg8Yu`TTDX0$1m7aL|_H3_~$45y42|EwIlKjxRQ$gep}l^FYV zeCAUN{x~Pqd7_c5trey~I8LNhIy>(<)2-A@^SBjC;~MhZ4%;r+Y`46vfd_em!hgZi z-Xa!Qfc~W;QxcR_@v`-vZwS=xpx-9(0LLla!VT&mbMm53w;K^296U;@v4^==mVQol z7${?|T8`Y(>M**x+?I3|E72y@mEg4k<7l(okKvLLYIu%?=e~fl!#w51&W&OAWro4( zj9&MXg&GSkrFN?@8O1YLI6qr^az^^I;peEs+ z+rsm{k7=OJRwS=pjb|EAH?+?=h*+IP9&K1fpcoz9Y;ygZ(3K@y3;$|yi5%xEozmVO zOf^@|I@<$eg3Klik5SQtoIoE5NA+os=G+>HYN?w!S!KFT4lf}_Wq{LuK#lhf$b$}| zyNun}8mBW948+A~u*K!kG7KP}|zFO2WAGYi@5C?%F%@DuxL{SnZ;@7jR-T z212Kvb{Vh?y=D{A=*=dp?DOjZ01o&pja8r4gS2$PU9En^F|%>yEGXQWt3^U^)1jP& z9#OP?sN`pG*82R%sKpeco>(5c=2*E;1{oe0J?%B%G4Z}a)y9w|#^|N==Lr|zblw)~ zmFSs4ooaJz{PJfUb$*5Pc;=0<%DMMOGWnwrPs8C~?xU%;shA0r%NR-nSja)WZ)buv z29++c!Gh$U)YiCY>4?3iHKOes=LFKpL5fM);;~VUKQVx@%pCZW#^NK4sg%u|b7A#> zDX#zxC`^0u+8vlSrE#c~gc%SR)hK~w&5#3{1}MEEm*hScI{7)xzmhMbTxSwf)L7bi z^7bcqZOxB)1MO{5d4wMTI`{f$@oQW;pz%=rrRE;b^ zr=??9s=(Ap3|la0ZbBnUODJ3U<-dnGX^1j;f^1mu%9OaqjJgNm1&f=A21|D~0VE2w1xpdul^=Hh`A6+uQi_9m;y z5`1+;GG1@V_|19c)@g8j+J+Oar#4=rmQ$ z=zWD_{}B5bQ2P&rK5zB+U(~)s=GC}QQ4>3Y=9I1b&o4TE#=nj8&BbKnG5$`T(n7UiP#c5~->>lHn5# z3jzrln^u7oJF}~-m-3$(LAQLO(>fK?l7GN(u8u#_#+Ip3E3QEyo#a5Wx1oQX#)bA% z%XeQ#jy9!7%pmf;HtpNqm7|t8{HL60!d5~jYu}?F=jA}Negy#_h3lfm&cQ}M0ftb> z*+h&1Y5u`V1VvXYj%Ct*i$QaSZQ~1N-kZ2beX%v&i{X+pnxMGtAmD|@3w!9i82Yz8 zeLKG2o#($YR-RN25=xHL^o9$L6r_)D@3X>dx+JpP>!v!3V|ysz^oVH(3Xbh5=iv)w z$Q8Hf4(gpdwAN$4_OoIm@!?}JnWr%D(Auqi)AW?|JO}Z}8VNmBjj&Al81K%2PC-H~&Z<45tSdNSWruK8wK3n`9 za@-P`?V79Z92K;lFNM zUavD2q`d&};Hc zEY|@~1_53AZ>u_Dn{p{bZk_d!5lrtMht194G=p!s=A*oR?mxj`q6VxVLewyD7;G_C zFr*T#EOL7i9HOLGF>~?t;wh6s*cY;}m|OUj@9>k09UczE85G4(KrrSkeXwn6>-R65 z=W-kQhV)r>gA=g|a8-+PUJN^cZ;)dLnB#m&SfOEG=%gKMeXm^Pl8qKrW3$*`UpKzsfFoFspU5S5+W`rFTQPSo{0{vt^Xb#+r)Z z@!(>o=^-c*qUSveX&MH0oQZia%Rdmu_@yj@lJdgl(B>m@f8m9ae^I%F_^D z>qh2n%F7U?>vpzn%Gcm??^96htJQa~-B(`?d&hO0#1-HB63YG5Xpg9C?lc#$(LQ=A zu9`}Rb3me=){VcDGkzQZHXG5uaky=hY>M?-dFd#LfG>gg;JE@>(fR7l~t5}_7nz61 z(G!p)IH*p0l395GufTO}BhTF>b-TB4oHA*TX)sJU2cY&5`^t=zVXM%zth-jLAqaRB zLGZd8?5r@~R$fQX0hdo$Wm9L6SpAGsXRjl%Fx}$NQbfHl%GQ_z8x_XVhX!Y1gTm$r zTQccpk*=-y>13Jnf`21A#Fg`#BORm%iJ;aMvdB-zYd_}zIZK}*hfWLdqXnp*3yMjG z?Te>Ak$3?D5o3yj8J&4&zqT(OQv)OqWy9XWH}_>E_F!k1rfKM^o$qZ5tSQK9THdXb zIjl{hg8H{{>rUx&iGQ+^1=p)aqAgyzYXt6SOo7v3<0Y_1mG(CVeRATMg)^Sr+In;qpoIER*x&`=1Bvfl zz&9sF#A*|C0V)HOkM`c1RYr)1Md=Rs;hHf{jhGOiQUw&;9m#f-^t+>Rx)FzO)q5hu zfdFq#if#79v(&0nixBUwh_2URu|t|UF*Y(j>JWDu=^>tIf*b5SswC`^tTX6z$p7~)*U@~mMFmx=Pig}tOu=V-*4OM!pZ#U>YW-Af0;|}e3u@9d~8GUGmv18 z!v6r6z@CQD=1~s6>Hs%E=ntJ0~TH)jRnv31Yw>0c8JD|MTU9OpTj_e z!GfJ+$DThd>pRN41pS7U5?o(8RvoC-9Qg%sy5`ZG9S@y7kLjq>^^9+!kbA88O^|!p zDf`Q#zI)Ue!>Ewz${&OgBp7v{b?jUQdui54lR2pI$4{Oj1F0_|sRI17pBrRgwDRb7 zaIbFRDV+ZFf->K9`@L!4w6SsDoSL`<9M{Bx-f8T|u|EgIJrg3O*-HY3I}6eH&V~ap zNZ{~o>fS~1dl!Y#fL~vbfeD)pY>HgXvrOJY`|y*gJgSsSWG&PlmZ2psbW7N$Yvj5V zUE+H#ru%TFJ&s#{)c7nRWX8FV|2)*#9SrU;;N7t{{qbV{(K5l1&-Z}$(MwRX-Ir?TJ7?s8IBGv@e+_;RtN-QjUjdrAb?HPl}U51qj!n9W;$}n z&`)*aiAu2&D`nwYKOptJxz+&?-(9WZ_tnGhlymwKpjsnL)%U40F4p+Dv5B7{oM8=bs0D+%8~X$TgT{(CrSEjgdt~@pjBP}JwGh?0sp@Zr9}TJ z31RMv&FEm~WN7SQW@u+; UKXHivQI%#zfx5?|0s-BN{fkQfr)vuI2IoNW3d{%Q zKym(`K*h1{x5K7X@?Sy4c}i1cSxV)fsUtpT1cJ>JEGIisXAa&-&<^p|(s=6RK?-tP zNZ=RZn*!fZ-#Fh?|Kv|~Kw_$x_P7zB_MXQ|^wL#A)kT$|x+W&TcL1jsN_3gah_Ha^ z1@{LaC(D>a4pJM(tauyjg~Pt_v94vtw}(AivfG7Xt9&siTaW~B{iF??TzB)BUHjizJwDgXS z5+Q(h6+oktHZ`Q;Kcuj*&bGe70UQ^>?Ti7ajI^gx$z|2fqK4F3vBpX0Us$IE3``&!lBvX3PPC z$LBS1+-49Xk+K)aogT%cu^q%CEbA5_TyTffUXb_J1ma|@7e;1WMd#BL7QtS@HIyt& z+qSEe9|6;407!!f5)e*ojixS3i-T{jSr{5NeINGU!pU!;h_#x4NE!mXy{yyGMB7w% z2^Oe&#l@sZo)9kl$E&+?=82Jg`9pwg2P!V^)f$}AQ1-`TmQgWq|F?K%;G!_WJDXJ2jIX=?EGFE9+xG8E8K4UEpH;ktS&Nb4 zsX)tTzwYT0^zgIr<*J>q_E{_r+J^0bUzbk;*yVgkF`Zw2m5%T5BWgch_m2OzjPcx= z$`1>xKt*xKjLq#Zjt_gKinaj}pae9^{$M}^!{^57C$7_=SF3a7^)->bi8cNz{CA6;Me;Rpru8rv!~PbEhyD6bxZ|9i>!Rg+S3wSs`R z!wfTH^=fdl=~DG&^7ymu>f_0A_3Ema1@Z&{2IbqI1&2_8lk5HQAP&grtJ%aWTmj)) zex@N1P`?XDS%c%u7`EJ#;e6*oaaVI1)>Q0V5Y=JTZ6N>C6`GG!M{r`BIGK$-LJMZu^okC4sN%u|iE z#O#{1fw530-WUnCUjPIj`VHmHj?-Tj6SP(&9njf}5e1=#3paEg8r+w0^h&5EO13nyF0`zt!rK=ZPyMFZ(18n7lPKaV6 zAPJKtU9F4|ZxWD&a(`I!siS!W zz#SGzT($dHd~hfYqdB>A<VM~SU zW{ARfh_@>?Z02DN6Z269zscuHiYpbL0qQWb^C5y8W7B%a@Zh9Eheb*IEbwi83Fhe0 zdgY0Q)QNe-8lY(D`f68lCQ}w&J>K^Zp6D}S7r4$76;`8dc1=e$@M4OWi=l#}k>&w~ zn)J~AZXr6B4_w3oiE)*_j_&HTAotpK^Ci`%*-~H^+Bw(0OmWIb znbOR5F1?!g+l!mz?SfS_PR7N7jgeqAc>{Jok?PQt{4cNrDjg%H8hWy6V0eH;$>b#l z6`r5f39}U)71tE=)4&P7BD0>kSB&Ot!U}3-8>tGkR)n++kri|u|MyxQeXCnqT%umq z(MxbfS3FAA(O(!=R4D;di$)5wMcYS0&8WxfUryuXYow3Rm?X(ezoZYDSB%>TJoYj2 z`OyqL4>;;f73%fD-_Y|Wo{j)QOpX!dEtC(L@k>l#yH~Y87J6FZoMR#XJZqof*|=9< zp0G^{vOglei=v(y4?}EBj&&_2-B*ohhSYWL8g!%5xfV5R(qO1|EOnRM*uG|d zIghWPxJm}=xWmzP&U~|7*3FkZTT$yFv-_bc4L&gR8`%}pk`PnwPUC=;U6&e!G)j`P zsnQSqo0u=5gIWj8gGe-4tn&Ykt$pil6*NQ z0>=(A1hoCJak9-DfKq^E+uX{~mHh1Urri$4t2838_I(L2Kxj=Xw;mdA#e=_(^aZZw zSd>9na3s-AV@PKgv)RV{@mG3t8wGHq*Jj&>mE2Ni;v&OH(#`yPxP0k^e5w&6c&$q) zb69!kZ=o}$w;5dH{*1G%a=yw#oy*lMX-DrMwY6I8ro*dQ4qt%(_lcje)W~NC2SQu; zw|<+%JN$wWDppnzRmy@;JicnCX(jYwoD3@k0^H6!JLlbijf=|%c z{ukS;m9N}eNR1!?tXOsG!fGxf$)^T;tiI9{M3QJQ)CLeReW4@bdUR&GbdtQnVXHnB z`wcTp#ZD6A8~RgZi zSX{~8v)|~_P}PJkQnY)#*Q)zC_p`~tia28KJ)^mnB2LzY9myiYE#hLocib-7(LET$ zg+dF?O#!&oh9qkx1-I3*Jo2#mp}U4Cv`HF;^&dc0^wILPU6dI&cb2%QBSI(FR*bhztt<jvk*}isKb_fk+F`NI<&py&DzHw}cB9%zZ|s;QiOP9sJLd z=Hs$^D?5s+?k~8eMr#okRf3t+(hD>m!#4nzg^Dywcl&C=f8! z2}@wxk_=3ilk2A2WxOBHOC^C$Ltrg13V65x`|{{u>dM0Tr5?yxrN}w2VWmy&{8Un` zuMsgzHq4B#t2OYpFg;Z*ItZGzvXByGNyo(<=c5|!H83sR!DQR7P1A`#xz#E}G(`aD zVM~N%Gm@$IBcH!Ky{O1DdDP!Rg#mbyL2T@%zvZ%XGwYP+r{F{NXcxVjX`ta?O{0J! zGp{|Du=@Yv&1IQQ?t{hzT+1D|6{`tecbwx_=fpS_-PCuuGB`l1%Aw4bwql0!Iv~b4 z#+aHf!HTmDojcmzBlmeG%tx&&0^un?(4F@&(8rO0Wc#F<=V}x1?AJjiCn{FgO9=0iX zX^SE!moDJWRRZ;HiY*;(tCgq9Nn7=^>QfzC4bS}^&jh^er{^wU-aC;kkIVrnSaOTb zP$DF|h{`;ePeZ%vh-0m>>E z*7|w+7;e!X@!Cz{Gj2ddsj?(0`0^6HzciX>j0r)}D;d16XYEF|x1)A1-hEzLW36G;k-TWBrz?l*xrZ8At4M_tV!!E#pU+G6((5T~R zP{U(wqPQ4oh|Dl1nTS525lZ-ut&}Hq*N~VUFfA{BS?{J=ue;d4<@TFOfV-#Sky&fN zb0T@}x;Q4KUrR2Gx+|aB(}c2&Jnf6n~x8b#an>#t3T3Cmbo%c7m&@guq z!yF{kS##Q?oK|R+CO9CND_iZ**K)H*I!(8 z?i0Kve;AEH(Yv&oNaV`i?2XxxHyuK_1z!XG5~D3lCKZJKmNdgwye8%&WLMT}3#ahZ z*2(%juDLVdFXAo*7Y-6ZdduOmremS@OXn{g2AN&m{nSgl`1Xz0%)FbV5 zBNKE~#>`L5Ux($Kzb7A3C7I$`L9OHcT_r)hfe_AX?evS$r`>_^)t)8&N#bA$uVrO* z{GuXxEa0!5WC2WAAV813kVw_E%uZBuVaU&2Y{q4pzb#j9p1aLfN;8O_ax@@>U^5h> zgV$uM7LyIH87C zepV29|K7vj$^P#6c|QBPI+|V|hUV?<{y4r~uU=h!WE|AEQ>QeyA#Bm9?L0A+WdGd# zYs{<5+lBAPR~;&EnRMGw!o=24Oky14214DF>8pHhfA28;zJ7l`+_kt{(PPR`QkPd` zy<59@duUa;nSaiGPMMS#7 z`Fv5pX;QgRDV5bvURUbMFo5Dyq%FPKKp?M-)4%O#vp2gJTZ{#dZR%g_xws`rB`aU| z$oKPp0Q_&Z3N)VP!aV^H5PBF85Z`|q1)67E4uUX3Y7fO&|2rAPLm>3PqZx`rAhdui zIlL?|9HEl=?@Pq9!CNuLRA)P%&x7mR&^I8!`#6_j8A|>OhOh_fx%mR_Tvyms2-O&< z*YiH9AWP?%h&A<}Tit{Dv1gL8GO42lC{?<^$PePgVBZ=@_3_pmks*b*BLu6vPwV z7`?eQnS{MFMtBCrRW%FowMU#c*%ph~B>zgO1RTQ_3V z1tN^iU8-5~_W1C^;(su+y44%6?KmrXd)4Rg6KJ1n3p8#Xn`}87Kbg>60ncyVOdqS*>>`iU zB$%Wu2=^#eG=U(5cR>+F8=I^STqF? z2^jXW$`6lx*eibJAY$;WOJ4LUzvZj#Dy665F-fba$_}3LOQC{C+=MWZaV>Fy{#m?^sjuix*6VWzW3` zd4<()1cK#!2+yylsK0J){yHXHemj|ej|&tGVlYQ5Dh{Thzutbdx_NrJ`2C#y$Yy4T zj^APL4~OUXkM99eczH2RkFxhC%j3rlZ|7g5ngd#)0v2#c5H*@8?mXqmj zN`FQ;cbwVi8lirWQamL76P6}hua*#AkOJ-!{kN+Nq{qicTm&N1ECwS5LE_2^c!-s- zeN!x27Q#$bXbwpth-d_9XF(d^?Z|R9)%CuC_#P+h5lL_?qR^OGFO10gCXbB&m?Rmt_I+Q$Sur*AeQ0J``H;^7Z z^8GQ>^k-{VcC$`{bbot90DpS9QSn#JELQc7RctTlSMKXc6c)@ckFc}6gdLLfth^&Q z=$|{V&JsSZ@&?@W0}GPGIA4KaSrvx22XTfnB6|X0`EbYH$i{;f3}(?380oEKHPI3l z>b<-;2mminMmqFVbyj+y8+FPmpbX0@Qwwz}F>wht_;`ysL?X2pnMP)qsa-q51vY!L zG-n+43f<2OCWaNB^+_7DgN8VG8*U|J6!oMcyG;8A>rq_y!JIp;*4yyZu$M@LmYS@Y zdUjvX*_m5xHB77M6wngR*-+bg>s1=P+7;?nV?1}&1D0opdu9e~E7e@X+M3JS>%I00 zti_haX59jOPQflFuUng8x3aq*IZ16eP=!LLNk;X?1&L~2R64eQ9kwWQ@G~Me0=aZU zv>3%MXN29HIA}7B)QKUD}H!) z>~dn^mJ6CBJyXa5fmJbLEpaPukhl`I)IkKVd}_9vUN&4M8pD0Iba!NzD)$%h2A?AZ zyA`n)So;?t{=5QCWQt5-LNmq^al%238M0G*ZVGYjzc8yz@V>%3`6ILwe1HD1!U~!E zk;G$@tsbob5|M*PD~V*L7dITS9-{61UYehru@M0983pdwy`$-@wA_#vJ;#+j-4N`lvH3#q5L z64-nlb7f&nJ(#&@G{Aub7s?vj-Ow`YZ=S?g%r!J1@N*rQkOan7+@GnrHTq>&CWE^` zaB@<5>ykt}aP5aS3C$)3EtF@QFK=@}=kS3t0N|Q5PS02b1)a)CBn$@4Bb9BqikQv= zGf?<%_Ngn+*XFVhPkyNV5R!+-dJ1DjUKR<`h9Uq6rMrae+2`#bs_pQ^U*EGUqY zoEz=K+%e1yxI0kA)=m+7;h%ra5u%sD!gP&1q-H61NcjE_^Z2EOojmuv%C3#dMU1Ce|t~!QP9pCdQn1}#OWqg?dTT#nyc4;+Ho7?Tk zziZr)Zso&FjYi0vuGy*g+Fkq%A;6%4D&kjquTj7&skE=Zun+q>EH3Q52k1QAiq12nV zKbNEYFq_YwbYQ%A(MO84CR8B<;1PW@aM_}S1;U9rv0!k6*c=6w{NcZIkH#xrh)?%r z!Q3;WuRF}ldZtLXIG zXBSTfxr79HQFj~`BXx#ON~H@x;&E5ns@%9igo!u`o5(Lp2TW++X3G~vuigdixNmpJ z{YiM1mrn5c;6sSo8hJJ9)0!Ez{{TcW;=he>?X)*9Kzr1_$>I>h)$&8b6i*@U?pBw# zayqcz#mW=EJ~O=#G7Zl$LW}bc1N>0Z3>{|X7@{2M$vgqfHk*6lH4<^afdZk-aUz>( zH^pygX}?(>iL`T787yfCMo_0;;>l9qKMrpuq8A1x_Wgl7mbw}bIBA}!2CrZdXSFUD zAzxVF(o$d5Ri%(QUrarYw@O@s&j9nQnQ!l=WJNJg=`_Fj9)04i+Q>p@X#dhFP+2xY z{*_@jQu8S1z2-z00E1Nn;vtq=XuVOl*Qd)O7WgHRf3uMrr=}OWsU(SYLv=!oA864E zV{r?mK3|At3s@umebl2?F0=by6dGK&nqmJn1Dji~BNSc4Bt_meL5z(FW>@@Gh@JI% zF+7@~&k2*bG8B@_Hn?a*2~t$! zrW|m_b`$G4KI-XNz~5#VJ$0R3b_9RFHcqv^Q9cmQaTg++AMwD?jbQuJv?ZC55jk47L0>q72R=~0k7{<*hbH) zO8u!CILIH+=k&YQKmmyre4D?VrF(Jx`jhZUch^kt& zd_SmUUqyHSDicGE&9owG^G0xL9(rt+T~E~yruYW7#=C1NAJp5hEs+OxhE*TbPY
dADdUUA4w8-wYXIc}~Ck!XB7;+VLBg5%keqIy%M^t1mG z(^8acoel%C)9^vJsF7!%=sWpXn)g>~PV2BcSuaNFJzy`*%dk?fZoJdjH^d(+QMv7! zzE(mICW=grHNQ8aQPp{59V&}7)OpFAPT5PPT zPH750?|MEjdIrg{x;Md6srX9N==g3(9;i;V>ziMAxN5tF}bV|7y}xF!yilrvnjFMiQ!W%^cWbU)D%+=Q))MHr%Vf z8tquWYM_Qg_?J!g4dl$}LeD);zgKQKu$ON=Fdw&uZ4JfU?gb*sX8ROTk8G;FtJ<7NW8NQF;8_&ob6Shm;tc+*%*a!1M1I9=&e8}C^#6Ph zGX(xx&35?9X7=a8Tq&i1*VsA02|2{*SCiq?rWZ}}07gK5jfpI3S^FD?+~M_J2>Sh~ z5Imlmd4K8$4U=>Ctr6@W@zBI_)q7z1+^=J^7*RV4xFU6^`9Zs$^eXz z1x?EOV8&xvO6MMX=^fU-Ey8W1SSUxI8p#pP*+P>&b3`-CSMga;JdO~_GCCiM5iIXN4bD!>2&9vhbDvvB%_m*R)6GtF|+LNXN?jdG|cCvwGr%--m)IT zCPb&j>gfg9rBsh`~&1C$>8{|V)8g@9oFyKZ=T2%F%R=a4rk4<$YT*iQAh#bU9D=E z2*uRGY7Mh_;o#{oHElG5cC2yg6?QmJG-uS$ajVwi+6*>Yl0%90^krI059;dbstouw zlL&dXH1g<;BcBZ0!#EosR`r0s)uI@)>=E722GiT=CgXdaQuxaTZ1CANiX6v3dl6e{ zO4fJ;s4jA^U*8G^NIA7=>^NA}Sselcb*FrYdI#47`&ze~wBKsL&K?j#?FhPqgdvw8 z^Ey1S3F+Wn$3~f|0&m&Ykn?@gmg^cuR(|z-ROe?a7?lMT(>ZlqoUH?g9^z_v&e{uu zN^vjp*l>q3|6iik+e7S24SfjZ=#)oLqaehQ|G#Z$7g^%wa`N`HmA z+bh&u8I_08U40nI(=C|*d(>sHux^7l37|F-#Rki2{MvC&uSF#W-~KXEa7#F=LsarNNo8qM^30vd{IXEqwy}Z<`P;<;L?<3n{pmCW zk+r>{SP5Ktcl}ymSx{DuFZGD`#_g~a-@-}**^zHMRD*^RVOrY?I1@0KbLknF-t=zn zGH!pD^Q>M&DzmvqU+K7-gT2PGZm4I0dV%X>2tqkk0a3d4Q1YJI<>D>2H$X7k~q*Hjt zm}tzta3rt^^{0l>46#|HkD|^(TaVz&^9aXpfMLL_tjEweZT>-uV=;l>dIs{|lF-Is zRKI?;gkZ0i<8Vme6B|21Bs(``9TKIgev9!FhYDvnAe*MYGzxerfKL*yvd5wD!HX5Hz?ZfvOwmX29fej!?V(t2GK>=j`It!# zNSB2>d4MV8CAT(TPVoevR96MeG6pTtmv||ry4$t|z*T|q?>;s8S%^#_^kKqT*2M~u z^P$6Pe!^ zBurtB-^)B_f3P8l@P7!bptC+1kiRso9udh#8vbg$X_`zHvo`2>(z0FzUl<^}H9FBS z_fHje1-#}Z_=(#U3ob`{HRc-iBwmV5T2 zk|^LEj-ZVmHkHFE57Q158#nzG(1CD-(Ch8UXv4@Xc8>A~*Ab27%g7Vk*umct5qLCb zD)8q%+3UH;of~g$PLM(G1dQ#f-sjvB9a?DW-{3NTF@Z+PeEPi-Uqo;{&A4?UPG3to zufGEKuuFF}Vj8L*ExxlBl`dN)T7dNJ>fWWctJsmuL4aVCSfeXD=dlCsuP*0?0Pn;>L@9ZqY zsegt{Nb$xph*z}4y#$dk+fCTr7UH}jeryBN71Fmy%MN6eJ*uE^_XiiyQS}oxSz9)K zy4iF*<+7xY-NaBjmnYhd$y#T)>}g z9z+%CcP6^)N@TG>kEVor$0(Jo2DsavqwbUO@q(hpVu3W0BZ@%?l4a^h)m0XWDH!D` zD}KbsEL-gTtJTvX+-woNa_aUqVO|+{|KV+RBQyf5G-XV<=d`GF=#n5QQ(1^8=ivkK8?8vjElW^2QL7kjM;n9eaWcyy!2D@~_^2UBqcM|ds_TxmO5gki4&i4AV#^_ zG6gI1he>wSci#lh`QGRy4c za#y_i_I{WS`$ik`DhX0(s^dza=d%DlbA3z}vvDW2QK#mtb_2Gner0vqbT`crwb7co zZCi}NPoMz`?q5hZwMQ*!sa8UfQA(>3E}cF~iJ3Vkt8cQzRXAPXswj|3(rox1=Ewl& zAB~jXIcl9B`17KZ2@X)KByW5|?pPvHki7S;$pdX7m5tGwM(IAK9nL6$5Z#i&FeK8- zS(nYe={tF3v&B%$LNZ7JU}D;5kmmGo;N3Ksq|{k{qpraL4(%`T=pPKA)nUifZpgke zHGLCp-%lmNwmFsq4_s#?2y@HC0#>Hf&id}s#8vyw`V(aeqz!wVp25f3jyCx37Y0iR zd0g9*p|luOtW2Gp(vVC*bW%3W1CvnZK;-_2u%zZS6-mp`ilAffVw)1>>p)F?xMO(Z zZHtH_!s!-YsL;5D-`ElAvID{h2aV`w%Qh#>^BWk7o2vqvVMk_|`r?7?`Q4bE4vc1s zG+0cjDmVH8<`*VO;CT`Hoe?s``%ppqqv*An6scK`q5CEZYS0g|e}? zMio}`v0!$a$S3XUhNuWu%ZO?itk&81cj|$Zn-F0Ky%H6(Ic(kV_1D~@jgO&h^dvZ$ z>e&^-J?7R0aAPRo29hLud9zeb+E{zc#RS*xb~+oy_QYng+uM)mA4OVM7evbwTEH-_ zj?Fbw=270j$O$H(!Wd*1VUPrVIGk;mp3&(QffG3zE z=#eYM$5CtW6s!$+E3v)aZwRbJe~{4~sNljoi!BmaqUDZ&>;YmBRB|qdGnNU9NtYOy zC;^(s?~ex_;~s=Qb&~Gr;Cj7XpDENuioPmr^vVS7jca54v+VR2fEWe!Z>^KJjk#nw z%$Ubri}o&?7w(`801_tasEZi;eD%dKr1VIYBP+L&vjySvaof-ji`I(YuU5#;;hU3G z*?1V2ZYTP{JeuYcB&W5=9>jsU_6!v}_ye-vix7~I&CSSwNx$OU#Dk{UgK~)%-kg2+ z;+4AGZ2Ua;Dxby@pMWtDKY%Si=Lp;aBXK8mpq4FPeeV#eXF=023sc3_!WhLx##b~5p#)6q&X;Wv$ zEvj(S&0V$(M2Dh70bF;#WSJX-eNh9x+=Ph;&?k`e8KvhzLP3;%jZ;LCMzMbDEuyBh z^^`9L%Dshl$)Z$2&C%XPv#)5Wz)Zd=K{l&-!Y3ZJW{=YGQC;6JZL6Yy({ob+tbOnx zkpE5u<-QksBwo_kFzZgG5=JYqYJPK10K>W_Ap}zp8j`7#J%Z5HUlXyLF)6Dkyl$-s zwLYm=<|Z@3H!N~PYF<08+y43421X1U#l!u~yuU;Ef9Jqw_O$k3&;fu&Apk(+Kjpyx zWrtyf|2Hem;6C_wh*s6cw7Y%qB#4VcwzP-?khI+%owZ1FPQNzod#3sV7cIsdl|=2f zFKd_2gXyA1t{<5}#O%w38yPaZ*qThitIFCb`D4AgXdO-+wn!DR=s5S0G5MFGX7(7P z9Erf4GEOL>%F6q*CS!yk3oODas15+D@dx%{^aQS^=Ha+m=*VmJgTV6Zqx4GjEMk`+V?+F(k$p~sZ0d#MVvuo2w$>p zM8z|VD-cjAOF zLJtdW7NKFE5N?;Z|DgK~KvG`JmdAhL_lX~`d^6dVI}(@kK+mumj^NvM*^ssMaCoA1m95up)}~=`{{Hay#BRpyNaOrs zk?z&e#cAC{K(D*UsI9wavoX3Yg^TeMi|CEqVqU<@+VSzEDoyXB^29M9#J^WUK&VDVsZzSL&#xT$2J5#DRJmOa!GSf)ar*R?( zAy)UQ*oK?w(LEobKgmhC`mWo$SQlw&4nH#F(1a^gy%}f!RJ8W_t|p&YCMQ*}3Ccs> zeUB|ga>wpYWgDxG&FBxKt|N{N5nwHYp@a<6$#Oxgow0q(P@RU10RD;{2EQX)Q&U`X zAsRt^OYK#@Zs=#45E)oNdm>XzQoklSg@;13rC)$^^vH&K6pb*1Q=wXj!R@85r@!~w zTi=XuER;wk!-y40oQ7(Zs;?Dgq&>3|1DV+P)uQkTeWYGO_{GN z$Ccmb(QzJh5qF{fewh^9H1Y8_{?oWk({D88-WuJD0c+UFVI<@@ z)14F@Rt``Y7069LMf%T1l@3ubXqH{NWcX@3XX)QU( zT;7DpaK+$fMH(SHO8G#R53&I2KNdmHj42eVP%&nvKZ^r7+`AT4^~}h< zM_+B6p&MfE%7UZ#BtwF)Yq736&V2~9MFuHJB6!5lB9eGHiLK919VSU+dQs29>0@l4 z@@b=mfX}DVFu~C45fCz8_*e4#%@|wFnAts9c+TfM>zMe^7v)!l&u@nPP+QmdB4Bo{ zn%0)l!^=RGt8JcO2CywEro-Mhy7qfii=(-E>V=(JCpVEK;p{ly(;|4HCOFV7jm6(+^1aWLtJSl9r4T#n1^_btpA zUXy*0Lgu++{n+Ee2ECbVz_`!vekdV0DPiUyy-PG<-rgXDWk=`7_cgm%J$)i0vGofe z!|GF5n&=t05|GCg%slYgmIk5_eyPtY6x--&-M*)T`%15|3dO;PVcdxt(dU^fk z4y&wp6tYC%s8aiAS8SE|xSZ_RBSDlC&F=7D#_I5XsEJM_-v&~c#o`JtZ`(B3wmJAX z)G{bI_*}SC=9yeS-mVGlfNyOlk>n;meV2$xu^R8=9-1zB;4{2hYTO>L#DZM2a1zzQ zoYDkrd$RNg{PF3`%9v$(q-EVdBt$#aGN2azz8z79y|tBPJ#M19zS}xQ=iC z3yNE;s)1jvD)#(q@B_?thNbt|c(&LL$#%K9B@uAnXb_`^;wZtMnP@HImd5{}gZ!BE z4i;N@4`j_Kq}H=R(#`G%r1GKmF!ZvQ77ee|g1^C+G4tV|PM*y+vjgI(g2G5V|JXio znDo;$3Skv{HySKAZ5X8i(=+6hZ?U^E1$UE7s8nTDc zg77}~$aZDxSDP!2Y)_MJ`KroBJ&3v*1tKvwjng(RzzKl|T30kPqQa4*_)#k%Gor|0 z{RVkjddY-rOe`E}nSz+@N|9tb+tX53Dvz^<@PF!pXVowv;d>c zD|tS8;)F3nF}BJjj)KP!(v~nG>IjbLj`jZIhR|Q;|qTib! zN@0iAJ-7)~C35I26%WW|-oHBVESnP8ss;UV7f}MAs#i^6#8x0yVCVRjQ3#X7|MUHN zv{an^#6E535*!z(ECmH6FAt{o$O8q1tW^K&7uhynsF&{imswo{py`65+e&`BEP*BU zLJ##7ucI4u=fHPGx0*m2)3^6|x$fs1szdM5HJ@ zhv_#oo<7pFkix5qRhIOSq}QQVwY-FU1)(@r_@lzR&D1HBA!sy>B=wFAvC99`u=t!$~UB(_!B#uc~+!rBiy zl@=z2kH3cuY5LdTtjrxZWxq)Pl0*FqV{t@`OYMDw(!n$-NpqI)`wwn81pE1FWT9F* zciLnpcCHMO-(aRRch9Hgj0jE9O<6uS`-?d-nfd^zl0bI9rB}$*2PX@ zGiRIE3Jf{h>tV55JQFf-+u}Y3Do2W0FJP)OAf0{nP~;3SWS~uYUg}{3Yb_^W90FI3 zB|y;e6RkBsGgg@tSrkc&xdF!qa+yzQrI@fMD<5hmzjgj`ufdh&ov=Dr1+`_ucyxee zt%+H2KQM|`Y^Jsv8I^5n;g%Im z(XJ3Jd?e^uV5ob(jiws=VrU49 zC`^i|bQ2XS>wGbAroKzg2%ZqVjIP;zjs%^Vj`CDVXs$g1=T2Gd+`+Oc9V%sqy|Am| zBI{0Z8y}i;efE5$i%5+IQbCkHN-jFB&Ros;x$3EEN!NFGe)D20t7j!ZhL*vJ_2Ct$ z>F86rJ@kEovTE@Nn2-+q*zX~Z>@eNEH$7N?>-%9UYC-EV*61uG4{=Ylj8Buq_|VF$ zt2q5jxlU$iS)0kWq#K&v!|%%(Qjiv$(EJ%{(aBWjo0weey_RqtX%KHGdAXukX!O)O z!Vl*8tXv&B<1WU86bAt`LvFJ`fX6*nIlT_7jh-OPAyc^mO!>oyiS93iG}?P`YTVuz zGYE);P$^2Gz^g^e?=ZoA2~|>uN!{vHY4-QvxXAepC5#MB<6|*kQ1{HJX{q<%yf}ZB z4HOs{rnhyw_$(|mmIXF-->&!V-rwuf&hEh(fGsZ^VqSj_hd_G^om+&-`r{$)7l@sY z59oS(9nBk^XLmhZY4aEb=+k0Kvd)w8!1iCR$QwuJm~S*5PU3aXWIupDt8Bb9D1YHd zF)IjQGIhigsETZKG$+#F(bfwUS~6S@E@eYld@}#~ocy+@Il2h$0@X*hJakrzImVdA z0@@_}Q6TT;$%lyT>dux>w~#=z|Ba`BoC`YRy+?xsOOpTwoE*Sv7D(fXtkIS#tyX^O z^9tGNa)QmX69?nLf29g^u5pdTHgT!>3YbZTuYhS4k+MGPeP7yS>GCYJ8U`O;**P(K z@E{aMpR;{c;OF}*@b4>JQfb}p;GwxCCFa9E)C zG1#Q=;kqcK=$Go?9%PvDOAn8ysY9(RVcn1G)+gkwe>p~FIT}wpDCOyR)U_t6VS{in zZUQuFVEtS}y(JIP@V1i!K*B@{4b#hjMeX#G_~q$)YG>ziH-`o`<#PAt zW^0y-qti`ka5_>~4}}KQY=t`;8^jGarEn?xKRVX>{6G6!W%4us(>MrO=*d| zu+Ai9<{EB4UOYB_87rvKYE+7X_f1fvMKolOU1p{S<7GZi|ed^O{n zpmVzE%vdEO2j(}OV01yEmRhRYFrkFB-8YwZ?KK)34Aq7Mn>7`ZIj3mj)9r`M7fN%R zDkR3AT>na&BJ@|iCUX{L(azl2YhRQiM@U@RaU|Xi(@(C2E#-h*qv1QsyuJ-ht7^Fw zKb-54qh#HCg9_l;DN{I#!{jEvPw7#iLfNZUjxAkrE4x!WSu;Z2G5nWiVf#n3;MFH3 zm5BjiL89fVi@peD*m#&dPnDbK9?UFGmX$DZzrIt3pH?sDuB2`iwhJ?IYVk8d7e9{JUYDZ42EbJ>f2<0wugYw2f=Z6Zx=OTCmt z6S=ip^(uZlzWI2#mId?G#gc^tQ0*}~@DF72{cO*uZo*)8>6hDJT=BQ{VRfk?KRc_oupZc)bNoo}_0xEMbZVD5HsWZTS< zLXAjg8tLj~NVOzhI)+O@mn-E!Eh`!z>U!;@LnmFUUcolI#*5v$a*v(E{G^QPb>3|PBa!4^C{ip*ACLT{WnoMKBX73XC0T43omxZKJT30 zNH$eJ#+i?iv!W2jY|7TRztfgRg3aMlzl{VJ)g+q>wq>#%Rc18U48`i^YXo&(U8m!#m@vNPXtntFl&ZLd|n9xHRRJjgRlqp)FXj4;3cMIeG z`d)61-*Bq>4%nM4BCSglanCfsXPZt+# z64HXO^E35`fKc()vD;1qs@9W2%iBTQy-ahn5NY9ZZsPK0cfA=nf@4$)jJG$pHJpC` zDrA}5OImRJisR+1KlJXru9b)ROF_%|LdStL%P%hcQ6TXe8%6RtI3-;yhMtcnXumpmlmD^3?y6!DsC|2f z?pfAMGQnuFzuw7#0D5%o&DuHE$yB*^S(j$MU8Y!6_1M6K01AQE8|Mc`{`Qd=zhYYE z^&n8O*<6Mv2|VBNt5v=Ju(!qWDQr@=*kGuS*ilz!euF8SJnZY{6wb(6{HcCamrfKX z6qMw0yk^jkvISHRhu@dRt-p?DC3iX4y0M&Y$<#^YM8v7q1;Ryj&x|#$EF*qDpR~H2 zXb*~22$4zPNLNuqGmA=!qNheX18GSK)k{ za@t`-2{(;U4IC6J_U0Yab5%?IGL6QZp1G*X%^(n9D|Oep#7o~FAfhJlG{^x}7X+3` zuMetm`cu3EK)aa|61%41+DI0&L6+f+*l@GAJ>TxX>!>g!e;4qeda`ZNqr%$b7x_EDI6m?c$N9yVZCDTT+qp|a6<1&rJx=*V8d_?0(@XQfdAuCO_6 zOhdtWEgKod$)Zur+CmLh-zE{G+7fIvSU|YJhZd<-Drx)Y&v6EXsv?^6U(O(&+T9%D zx1$IL*ZA*>$HK`Us)k;8U=;uo>|slzV%Eg(aMJE|#g4O4IC?4dP8YVr@=#BUsnaE6 zT12<8T?$ythA#$^CArFr4{@b`kK&QpG{Utl!2LNMDx4`1>t52hv z*Rcpd9nE%4&ofx$-=pmhgrs9KgtD^zfUT3jBv~nO&bd9ZiE$`UT>ywOu7hvsv?$}p zih6EpGzuBS>WHsN2&8Fk9RY7S<0?o(dv2$2I6xTprJ>z>5(hXMRh3<5mrodJs}ra_ z3A#c)j)cQjhgxVVvdR1v7kd0===p!#n~`Sy;`6U4k106j28D%jTETSICFdppU8Vzk z*@2e2gmZevrSCu|*glIN$Ww=4<+KLWCRFjARmN?_5kN0>Hg%0?5`zz~-u;hG7*e zvclk=Ysk)eM_n2X!z-HFJg5!>*8~(97I%01wloH?89tt!uMVRvW}XZX4PyKuy5?m- z9Sb@*1$5W&i=qK zAN{GKvnEmsj?>sw2ya8ENSd~CL4-G^SUSs|WS-c6H2#L%Q>_b1-{meUKi<+f6YU|R z+Rr6&tr$6VX2{$!Hf0psaK)(KOMz`|NAC@0D+;V8U?#S+-kP7AWmVrY?LNoq6h83& z5&7_de?D6N(z^fp@n2um|54*R!9x)Ix5ocJ)!@$asCBb>AOSkkZ~%q>bX{KJ@GJyH z+WWs<9Zrk@0Rj>P6bR@xVg#609`b(;W6Hka2#623N|gg}{ikqgpkaK?BLKxjM5Q-z zasT%kb>S=$Fd$$-z=41VfdB##1QG~j5GWu}L7;&^2Y~?s6T}w~SRk-LdFy{Jg9M+64=NP+mzS-}4@Zv1~s1%_vikhW!x zka-aOD~chN+K(pHFMzRq`v1?(6#Vx9!N+9b!N=qvC_qqxpaMY+f(8UFi0>fiK+uC= z0Ko`?2?R3;77(l;*g&v@-~hqd7JSS#rTVV{C7#+f_2)qVK7tYchn3$y(~*|I$4&O% zk4AVWA2$dd5WH=je0&>u|BAGV%Y#D_2m#0uf(LN?j}!Y(B>4Y#UT-$}!f!VDK?t2W&H_JbyB!ozxIA6viTVXI06*FehK14$Qnsj(D9? zvw-_zQCvPEOA*+VpFx3SI4?*fyP05tqNw)2@H}UzfHKyFmnmt>=D)oFRZSDAM(vot=KENK?A!mOTwFy~X<5OTmYZnCWB-{&GEWF# z5IgI=$ftI+4sw@4;6m;v(%Iu-kXJc0V>H?!rJU|_CmkA~VKwLGw%0co0(K-J$+vXf@+o=r>>e|D#aXQ1rt?W0UC z^KX1^3TbQOhn;f2;NAKAutaxv?>d&PDZ^U6wghouCf;hw1&^}ZTlv>UY;3Lo8!w)+ z)<3oAz;*>QrAqA<%4p>Zc6n0FB$d7Tr?iK8VqKvJ=ViTGEfCJj)aY$5wpiJauCY*y zfkL>jQ@_Or)$Xk`Q(p+k;gZ@CZPnA6g0VMs%HE;BVmwkpiN5qQDY+ydv$>OZdWqNJ4P=CI%MEYD^zR&alqXt+Lp}jNb^kR7Xc^beI_ql#KcLTx zJU+_w^x#w3_1RgVCr>*s5BIn1h-eolH>dBvI(fN%;@%IAg>v?^qi^@{b>l`5(*N?I zt%U}5qn90(Fd4o{piVH5TSg4frRM0==|3J;l}>?4>0-3Y)xcXpu!pjqqWm_(8X?vp zb1{sePC{ES!98A&n{yP^y(ajb90%;P921m6fB%+~G`=L+!Y#sQ{k=vX15qF#LBgd- z>4i9Nf;ale(oGc#T3)GbfO|%*fo} zo&RN^`2rX$+V5E?`L7D_KEWl#e80~)Td_@4i0p>Sr50IHtbFD2q4T+dzHeqW@79Y275FTFpTd_pHSIY1LtHmC6I=bXO4=c+qR9u z^Xwgk(!o@#msSz{b?p+`I|WgLbaqf6T$Cfm1A(-88N?cwZXT2bXS9xiSwtpzzOEGc zOJ%W!p@~v|MEqq&HQhCT3Nip(vN|+qHhp#bn(V60LM!z+*q^Z6;0Yb*y&An>^=T#G zn6pz?JdS5`$S2%@ zV}Sey-Kif&Blct9>XI|2Tj!1|g5qWhnsJp`b=LNK_qtV-{nv6;!QsG_w->IDO`qEo zMulNw))(o3d+-l_RG+c#StcSKTmHYdCp)Vwua$U}QcHS}{_H$J)R|--U;phz7Pz$= z!E+IgH!o;`6W8x;d9ERUm)E!djw6CW%O==d`ZwL2mdHzzFxF!Uv(Wt_>=E`){AQa< zAqJtD6A+zI79rJE?)$i7fHuFGb|T{1H?&9~e*;Ld z-a2Ni=GB@!Hr6fvHV19qR^D!I@}Wuq{NP`(+y{<`5Wh+Y!5Soxsu%$F`ytGc2KgIg z!;5ct6Tf>qSUq~7OB^JhZe$VfLtbaF%D0of2WQd}%PpV*4TC|4S5euKFn!67HtSiV zPRJxEWLL0>9uNuUdxOK_uk&PBWWP=ZPx>3Fq4+n{QMDEdr3EVF&+~{C7m*-MVD5}d z2-1$0d=0;`n5;iIuir(5U@1E-4J(!t!6_9s!!>-%^x@J+IPEGYM)xOLV+P3DkESzVbV0NN1?K7>Jtb~V)iJNBQE*F zyl_qD5SdYz-8n7u1AMwb<$DL=O^da2TBn9i(|2a(;oXt1%vdsvJn2yUW&u=*I5(IU zS&6eVBE<~LgD9*4^{V~|6m^ce6fQQ0L|H^aE+O?m@~FdprC~67s1yHj@^d%wKH@Dv z(U+Pwye#ryibAylg3zlGFI4oQgzDu8LJk+0u@rNb-!+wBb4j_(3tuxukw(gc76ZuO z-m>?ssGM2=Br}|s!=nAKq75CbHBOytL38`M!Fj!SVG(q|vJ$K%GCRsjBbjd{iz%fr zic^_DNsqHzRCh1&Fz{oKK-(vyeCpjfr_D;8`NnFMB+mBy=$m->O_Xob;O82T`(SGY zY~Sg2n6;CNI8-PY;a`EXkYHh0?jCT&;xr;Mc;?2k(}p4=mT6zm%)pBPfvJ9cKeL|d z#}Z!buVTJ2Pj968t5w-yk^YcVf?lx227dwl|28*&S_NcoePV&OX4DKb+`d*ACJ1OjNFmucgh@h(6!C>)v3Ck&%K0OE=gY+b zhv|nq)?0Q-5P&MfS*aBSPQZQEwYw#_F_I!?#7ZQHhOt7A`pn3{TLYW~6ba;o;d z*R|KhF|8HYi#9}^4NqnTr=pn*nQ!^+x6+Uk#n;$SLc)jqtrmCzmdkG130ii9!2y45 zBJUE)Y^bDrAKb3{C)%m$2$GBgARITiX-Cc<9YU7J9@9!Hj>G6K#o0GNO8EK+)^f~# z&DM|4|79!L_uqP1F?4*jhLqC{m?N~40t zVXdNUxaZc`CW{OaFjH>6wp2su1{(re@(${5BzY7n%$bZp>4k*={dboJ5D-D%Mq3wx zn{d2P|M87O;Tay|K0$#?B(xbk&6b1PX4u*fCQm>3se`mU(l8Yc#X!3EqU+`rS|oW1oZR&An`SPMn+G9gPTfkN0ZOXJFfk?=QYy(F=l8lrF zJ&0GO2Pq*qg|iiS6YuK)@Wky$Xsuy1xB3UY?RFgqrieHcd?Y%WK98;G;R@+^n7^Lg zRd4>t&wJ9WFTKl|>0ph6^e$^Jo?#?)Ab%L~?P|>yYsnRkcR=p2drO!o(cm1utnN)E z>fxo%w)*Q7frp(zc*d(mHL#6}zCeQ&n6d-sgC53^3!wyh=M+2xfRyH1dO^oQ1-@<= zKIHcNcLyGRXy#tR;BQBybDoC+24-erIGIuOuw!Nj4f@ec)AI+8@7N~ltxR1wqK77r zsxyX(N07aZKi=^~br}jiOP{0)jewNRjsDU1*UDn~KI5-8a$R0w8pPIC9GHP>z-}(X z$Oyk~Z*DUGb%~_|+FrkX?=UbkYv90kZ z9U{YHFEe@c4ELZwlReNGn;ji^g)+vxGw$g&yX+jHb$1ZcY_C`@E9`Bb))Nq7@FR_g zIdVkcz(Ysw&2Jk)-M6b_+SS4XZqSUfO|BHAW8S;6FE8l;qZn;+DCCPgRrsSM9KJ>H zq&QhiAo}#SP+i~n3FKNC!JMfUlA6KW+p+j2uI|qzyeXE;2E?QUnpfeNMU)QeS)c2Y2EM_U9-#tks6K`5KWMX zIVP~|&$vtgByIQuA*cFrVzgE0&5(HZs==k)(Q@fQB%cl`t+qH3*pBy-rE>kCCv*roN2!ToWfQ|%Qw71}NnlKMT- z@LBiUx-Tv*{nRCI-4kw)Io7)lmUQk4@#A5b7_TxG0Au{I7g}QY3v;j#B+#jotBU#3 z=xI+bY~irGSg~qArYR7osH-t|wP6d*A~$-9g656%X8NDqH4gcwa{Ft{MEX5Q{fhzv$CrAvKm1Y=~ zr2aD2L(H%mED=Iy&dx?+7UY3R3#nyH#>3Lsw70ns zTdzr*c}+~3Y9v7u6*@Q-!EXykb|ku9mg_7EaLaU;dA9>+*Lj_t5cL62L2DD(Ll`r{ z)Sz}BYU>IiRD;ec;!|uWZfq8)owFXAE_7bN5kNsK3Fe?ku3M`h@qgSf`;#ckR6vjA zH1i$-Dj-Dex#4iun{ju;@=EXIzhg(TKqEDm19~@MBYT}T^(nbFe3%YG8_d-ql^`?( z&?BSa;TcX!nvJ3iPH0O|nwsq3pclB$Bskam?%F;ZgFsEo=Tu)hFh4wScJz1U2^mKG z;d|XTH-R*}^uj9-uDFpg0}A`uBUq#`oTlYh7I6e0nwGR|l%(XW+a@#0g+~MqE3)rxZwT>E%-@jGUm`N;MQ<8EDc7J>G8U_};-~IOW2LU2FNsM%NrIib8%ZuvX&1U4!VKBT8m&S8m&xd$RpYSryldM9P{aMZF1Z#@sEM=WY@3`k z5hPGirAoJiVY|Y{ksftKn*VVy5AA1+;*wVzcr2tWV*f?>FS%j@k)HvkC}Ce2e?=wQ7`i1^hU1S((XH+4(c!?td`Ez5 zI3cEj1s}PmT!r*e+<9U8BF0rQ|KiBe$I~xv-_Sn4FWb8Jh4!b`9w-V*b6(Vz4- zeIzV3;wu8}kA{4Ki* zCaxwR1V@zpXH{X~5A*1Qgh8Buiax2@YB03L${doTZvPws6yI*>I{h|xI;^^1}}ExTWroO_@o9ScPoC52Vi~#f3K|&4H0T ze2<3y{zn)J@E1W#!imipalYrJXi%c^s8PS=invtc=5zA`tk1!hrEtBg`g=kF90!B3Yp=40Uhicx>iCpIc--1AV*g(cNH$THCr~ z(5ny{-a3(>Jx4TOp1Zut&8eio!E4II>Pfx(4Jax)ptvITp(_5L!g&xy5}w+-LT|lz zs*E9Y4XV;N!+WF1_)3 zwW$ef#9ZA4j2HZ#Nh9Y7S7i#P+Ac(Yjvn7K&Pw08>kR81iP}y|0I4pfJm= zVy#~&HuY&}Rc<)#TfVc5g(f~x_WqI)V4$iB=+W{C9L!i)$Gm)?Sv5eJ|08y{LBO@{ z-RXR@E;AmXrvcfeW+Qf|8!}3Q3u?(d6q2<|st$J#IW%4z-JFV86nVXAifvr@F;qd> zIftsiBm~y4V#jR1D7Cvus$OmRLp=tLBmP>RWr~;_Q~w|oH40Z8FGJuaQ2MUibr!(~ zz}N`3hQGrbbPzmdX>!vs3~>6bXySx}kAvC2q;O0CN)mMd6A(+E2r62F?+D|#SI+2q zw3Gydm`?pZkGg6h9^+iQ`mx4eed5S=Ws^*6>9ATqvQf0=wwlX!(`MWsU9;10uR^~8 zOJ-SvS((z!Zk~IP%-7zm_0yQ;`mM77Fs#EJahdwj)5X5=$Jb`sbSek>JS<^~=@dK} zM{!}@Hr{MKAQ8V<6q1pWUTyZV+MwLdhtlj`*p|*1T^2B?nirr=-C;x7|LL80Erhmi zKJX5AZl{cYtL^;+4auQD0SSUO-rWk5_#0_F{Y-*zn5f)tXKseSuI+mf6&;!ia6=1y z!os7Zo!fF4NN9X^erVM)sJ_v9N-tLN#rTDyVS*zx5o-GbSpj_)*&|s^ z%4IgV#gMBgiN1Xs+`6lLVyI>nwA`K(6M*cJ8D$UU_dBwK4>DaIyi!p(EBJ2URs*4 ziYn$`bKz4pxK6u8z@szN44(+3fQy z=%?EcGr$Db{X#2qd%5HdK)Pq)i`eTIq@vaV@l4ab)=(ukDw-}&&q1TWk6PrEW<*HL6a9_n0j;+zWe(BduK_@t*@LdMZF&k&I~m%z2|s?up^@G?1b^`2RA#3V zRbB?~{p(U6itQ=^`nMj~joUAl@4BQZ)GHJ5KgmHE-MO-*S<`Ye`6{%$(o!~E*He&e~t`ms#z(lHBA8O8bI_cy!T?euZHQ^KQg9mmc<(_&ZO z*W>Th9TMdx;L2MpTZ#%$NZ&m# zwn5KVot25uG~8A$KZl%yP8rY}T^;5x1`f2YkmiW>OO@8Y{gwq7I5cZxejX~gCGQuc zp-)S%L+D>WKyEjTZ}x>#Lxw&CF&ponW@M;bi5OffiPSlx58SorSS^qbCzD^S1nInW zZW020$>pXWPGVn1k7)1{10b~k@nsD+Ou*T+{nVqBK&QU?^wIF;Jy$@=!PRdO+|Qc% z>8GPdb~-5b<$1zpXCBQvkH5B0i6#tw^s|eQr}uB*OvoZQAJcm=})1|ua$}5j_|>!;5#6oYH0<{!A&%{PAqO*M1DR1-Q_Q#-{SaS zHvJ1_nea&3t*VECGl|ipzfPYhWOSJI5%`f^6r#-NDxjiaBEeZ%^76B&>q3l_^8~Gz zI!RTOr#68=@#MUnm0G1bh@kW0i7rWaaxgM&pX@#`e-dD%9V(N}gKK!r=LJTko zeFLijq8M2~47P3TH!F4kl!3}5M;1bUx?PN`zjJg*kXq}Y4nQiEK1{sWmc<}_bD%yN zaFTajnubvJO%P1hI8jp$&Qp3q@6MoDhQC~xNE10M_&b4i=>`6x$YDmcj-hD21S_yZ zyinx(tKOYM^h;5NJr#Iq=){hNR!u6=e|e4afr~A) z51vpU#p60|BB#p&;>p+a|5A+gzub`BdB)Xnx5e!p!D>LN-Px1T+6c63D!k+^rg=vp z>#D4h1!Mn}0*W-O&)%3Xz49o54P*jAcMZIa=85A;8k-3L`wNMgJr%Ay$v>n{KcaDG z8_9447Zf_bZNJr+-?@6xLaJ{@K&?QcK>c?&u+E1y0}+Z4(3ybAq!qHN9-h1lnTIWG z0C8z3y5pJ_xUp{0m*;!A;E!<0mHS@lwl=LJkl;*d%{B!wv8qJoHq|@ zpFIVWRJ2FY5Gp%`^~_A55nly2t8!1*MYq))U=N*#T!@N9bAMHzM@amVp(N>kdEaKx zVnIb>^=b8}Ay3~0Jlj+AU3y0C>?Y7f5Jc}x?>_0g|o@ph}Sl4Uh?BgBGrK#JPlE4`s$+4RpRu)KNQ8|lM3v8>MTsdzlden3CTe4v_RQHZhsB1&BzC^kp0Zr@fWfW>M@ z!dX;I_@ay9jc(q6JD{H3ZMT2`)zL0uR|vCxQ9zLDTd^4EIjW8;%#8aO(OVnk*YphuRWb2s5nt|L;;n5|s`md0HzmI#TZ&l(INbgKd zU&^JCP=A0S5Od9OHVI5kqiUkhmWBNycEiZjJWYYS4pa((kMCF_7AF=SWDo*`?EU$zbh{I<*Jygn~*=(BfcyjrF*x7poBd+iT?`fisrz!8#oOd!ed?!8k~tY z(QLB2v?*VMB4e51nXFL70NOOJN4eY%piI3%kT3WC#N*lYLK6Kb)E>hxyv{K)t}3S6 zuu3kGSeqkjOfx3pCc@nWw@{q`tr~hl$^q!x|Z~nNfZS>N%lE zdAtl~HIg?_lxml;0pml>X~*pcwU^$Jzw0W2H4Jg{oZ^zEP0pT z)b|Hp$v`mu2Aeu-+AJjwCb(RjRkT1?&A-rhDgSnpZp!OBTG)Kp z5+t>%^HzrOqvzKn2=U!DH(zg@&{t!E`OE4-X{|%p6Pwl5!L&J25@1K$<4k1)h{nx; zl+3&Fs(FkU{dsPsOF-x>ZPjt#;+=WsBsyO*^FEpphNa8nM^1$#cOU=9h4g#jF~*1n z+MK_OnAN3#0NB9mBip7xxUAZ_c7+=?>bMpYAU#<&j&c77WJKP9g%f%oKISrgEZ1Ll9`=7htM-nX8tU8}!A|?fb~C-hB*k<$1BCEgl?Dj~nSr#% zmU}+0Fi$PkbQV3RwlOExe2^MpLoP(*=`%#ts@d7Q9}%a1G3AKMN9`Xb0`gLRHLJE6 znes4<1FC+IS-Zh43E}`-zL+~6Z=#eR>xQ2^3&QBxCk0~POBI1_d>_xDrZb42lro#i z5I+1FJRn0+3NJB1c54cD$m#+=VyPRz7hHbSZmFqyE83?w&HSSk`ush68#^zp9iU)@ z`p%_2_D`;%2}7_kfOSRQuRo~1wUXPLsaeAA0Q}05LvGv9as%7JcAY)(7tvdyqCITi zz1{tDKG%$Y4u~5vFG3Q)54+dfZb=+pdj*;~)-U!+x{!OugH%OntiDAw&ByH;i&Wm| zN1>ziun28z3bvR1P2{a|ay>?ghb1uHNnFRLNW_*Hp9s%7Pf~EJ#!aKQ$bx)G**5`P zfXIHDn4>?eG2U?{<`VKpl2rAdPf=&vc--4IVZN-5v~;20y=1^f_O?|Kg{-4iNy!yh z)y2?ynExtSY(0eXQ%vQGd8g!$oC>;Iq#?tV<2}tY{rr0&F3EIbR{+u}jqw z$_w9?;uKz9y-fV1bR~sY;wD{gQEmlEM_pinw}%~f9@aqTo2CX)pV$sLkJt^a_61z(u7NsVZ$YHKz2$kCpm;Ngtqet*o{>~oo>~-a!?}~clm?L2FyS4 zbEXpC5BUv9b9zs`IWNZ4DCo&j;0(vVYkDu_3S+LDD`F{0Hj86Wea%6A5B@UTF3-)U z7yM>HDxj0!P4c~j296yHwkxcX)cijce2pfSmpI>Ky39*8znru zP0baQ2KybBS3635JS#1xyykqcI@}%l%|hHF)Il-kqv-O|#OecR&HI&qZF>W~q?P$8 z!4qV#N+jz}*$swx@#W|&D|ZQ>zZYBsmPLRC9Ir0x@%=UtqFJqSwVDqIN0ptgs}VJbScr@zsdp~m|j`M=Bxhbq2XmP1t^Ftja$p$y4(Jl0Fw2OiEH6J z?gTnA$qJ>I=!foGJ<+cf+5wVStpjR!7(~8fyC+ujugt!Rb0b&kOg#u@QHgHqN_rhp z0EnP1fXV}T__CyTvd+M%k-2mDRMpN|BuRn7TW~q6E^y{Hz1zz24=y>j><1bNB^0)G=jLiyu+j;X z8HCUXsKcS~6_ea+TgE((Wi=tsDncJ3uc5>Jjz7O$i3gR*YCQBTFFB!aAgWu;ZaMa{ z6nMtIw1Ayq6oGpVs`jX+lbnZTX4<*PA3Up^zw0a>%|XHh#QDXc3wwZ#-u4yq$A2Ij z32Ll@!Z)TdKQK+Yzbf&$f4O&Qo{iqu6RpI6MV}?l5WRjfp>Aa8>o!e7r-)q_=}h zZ##}l*9eG;4w_;?EAd+P5qk3ej&|)USuEH3!OyfM*2g#SfJ?MBSobFsam955*%r#I zPg15ob;+ps(ibM(vn-xTK)KUuvOgp$jg1$34Z6YrV`f$iFZ#^kPsO-L)#8*ifM`C4 z%wMKDrg#rg>NxmZTOjm=;1t&j?;I7H>!H!ZeM#}#D6RD9ds`=8q0O=;uPS<|y*k2- zJah+rs8$drC?uSx4eK5<)s|wX#UPjmJ0Q(I%wa^n5fhT zq@>VDehK~Xlm+ULx@CnI{c^kY=I8NSKIS{pzpaM|Jm$9Hf4LzA)X_qSEvtJQ=j$;)+S{7TfFa*}`qs z>00C2_r7Njno;D9tc+isqQ99r;oq^TWYxo(DHN$B$6B>YdXZ%nLtM1pUkG>U^lr`; zV}<#i(T)>SeIgWQh*qrUV|_wb z#y@RadS?G3j47w;vJ^$o#GG)xs-xoF(W=_&|8ttfDHX_~um+m^0SlyB&gIvr7#5a9k zfk)eX9+yw6)btUeWt=Ma2$(BCl#ZG+@9}f|xKEyEYcF6y{#mpj&NdZclQ zL_9pM(XkEFpClfOsl!sSTzb z3%6wYfPqOZBuFa)kbq3D>3medLGTbr^Nq#4!)5MOU}2tK>(QRF7Bk74&B399~3HL{_m z+tzO9cdMyNgqa4vjV)fPMhG9DlvMZY5Tu||Sd4)(+rIt%ROm(KuU z9I$-u9QqjoaMAoumYvQV4TZg7RNQ0~>fa^bMqjM3&}ts4;mhTZ=;sjW)7#yl-=>We zr0$c@s1q7iqDK97fbuMtx2y4&rkv9v-JmzB)jMDe_B>fQ)frn{w%>c%_N#Y;YDHqE zyf^Ic5Agp^g6#hQc7Qp7fY_0Mfzbb7RbVQ!moO9HzfFPvH{-eDw{h8&__KSfIVIg} zrcEglOPDG4yPI6LsZeKpLHh5zN=g(Sh^R9OQUG%R6-DZg|298k02nciw01*xW-}cM zYfR2fj{ody>yM!=J(^S1iu;<+Yu}o~)PX`T4s_D4FSuqbw{& z%Da8IBi^4y`c0jp6oC;E9rmx4}?Qahv17%he|StyH= zO^zbur+2S-J|Uzu+&TWiH%x3y))VS%unD`m5(1^r4M z?gpGf$%0xh2SzMhH`>+r(nmx=0IQWnV-jq-*y)8_g-zj~*Bj&e`Bc!{2r3%HBoCAK z%Yv>@ip-V=LmJW((H7ue&%r=(KfH14*8j`E!pHm1!0b~v{trpCEz*-kBO)ur^R(bV5}r)6~BZJ zrJdlj$&RU0ByyqQ@yueKWsppj&Ht&=ToYY&4VH%VG6DI^qkr%M8- zCsl!>FAV4LM18?g&4io~TInJbs#4ZY`^w9kUQ~N(O&jWCRN>WU5@#XplndBxR-5}3 z+`tV0Zm`zA+kyluCR^aUUcXCE*jbr`M}9?ynE+>l_Tm3lq*@7!e@Z}4`1xxSWWPF` zBsdiJj!)9Xkpiz|w`r_TxIj!_t)IxQi$4*%0y_^CXcur!!~qNQhYONbt&U;LRvUTU zykxJ!!{9cDnVGXI84(pB3u-+*OKMtyx6}-fUQ}BP%$NrADTq$I#=%=FesRrbXhx+D z-m5X6n>n}>2rZH5NE1eanHNZ(l!AXBIap+bO}Xc|womV~uSQk(ZTO%l4!0yHBAJ}> zWyoBeSQz@+dGr+UG-+ffj^4bo9p7)*0$KGZAAh$2d8Y_v1^7dL{Nm=*eLd`M6W9vy zRS#xob=qX;mZKlEvXe?vIX35E$KvY!`RvwOVbf!in3Wp7-TKr|Px8><A)MS?B4 z{cv+{<>%RGN@P0zl+rSxGHXc;NYwc1NoFt}dMFd{E%*1#jB&2>f!$GNwYkjgGpGdI z9|GKYu2PFYfp}IXZv&!OlM1=H#8m)6VWNPqxA;++DZwy!R)vRCaeW3poFLqJJ}Z($ zQ77_)(>TnWJDabg^UuUv1?g9W)L%|-EEeW!ZS9T3$6E+q{* zh8iytErVp9Ho{C;Wcp*&pbs#G_Xpb4gEZ5Dl2ks!NhSuPLpz<=uM)wPXM_o9U5gJ# zL8BTNiC{^oV-KNFW{@ze;ujJ@2j$RCdIeX4rAigcgPP(IV<|}|(AGT@LQ@5X6*JKJ z?I|V7P7#J@wWDxfVOdS3dub~$V2umVpT^Eh1JXv9$A6&hhtLgtUfOO2DZ}N4pGl#@ms=nB0!W4waS>2snnUrcL`kWyKQDFjK15zHG2L&Yb_~G4^Dqg&MmymPls-Z- z(_#q6r5XYKUIztK&+H($!1z|j%n&JeZTqkOUrAU3F_?|ePH;WN9_16~_a%_%S2%6d z&Y`-v1F?!h;0cxGcyQ1?SNN-QSbAy*vD2jt*)p2@o)&@Nm8c{OK^qtaRHlO(e>ji?*RNZsnc zE5CazP?Gm`>>+@arsIvRo`=rI|E@zy-;Yp@2 zT(G}k-j@*KRh>hFO^FH-j}G8*JMm_*^jVn_W3YfVS($mqXtu%T^Sh&P1n=q=!3S>f zZdUa*F%H7_a2|v82a!PVvp)c+RwhM8u~*CIp7gc(w$QgbiTG_TWb$OG8N&#FaPGxZN5lM`|Zn|n2U2>-TU%+gQ= z#2bKkLr}o&DRkS}G(`nEU-}2;58VI(8fuxRBhJ`#$p*w>x&oWEB!nAZS<7$3qrw3y zgMoM|l_BI2(xok)Qy{{aN5dYe;t~ZQ1AGF4bHtK8U=}Oq$01{j7c=2;Z^nC+HV6uG zTG+daZ_9hh-vv|Ao}B2K0_sl2_z#lpuUY^Z@O$~n6fQJy+eYL$kLYvD%M6{k%3C($ zTx{9dqxdrLB`1ZjbdmxhqRQ-7XLazEQ7Ewdvygf)V^2>ib<3by*AgXk(>>q>0iI0a zg}PesADJUoI;j|}h?=!_aQ68VADKuSWRhTx-Wxh5F%_UG#cX`7`6%-q95+a+8a%TdHKL#LBaO65gq@xjPXSrK%tj~TE6L?sGj z3_k>F5V%0vv@ktgc=&Zg%aq#f0!cG1)`d(%B4R530Gqh(|(*+ORK>2_h_;J<0$VBq}GT?WQ~aV+MW$pPQX0K=cDfZx3qpHQAS zRfry}cq?>US*vaOH4m7QT*jlT`-I97t%WHrJ!B(@tvngTR(9k>*@8WqLxy1})RljS zgp)P-3lZ0LfT5urVCYAud`=m;kWe&W5e4m-nWT)io{Ak=u_N%>qU1rnW6$spe3S!Uj-FVaHlS0c?L5^EQJxvSI`~w!@{qMvV*BKfJNhAe?IKeu(8r=Q58F zxmIsFG^kk9psBW6RW99R7iovv+5@zlEj)u15(xCj25nuj&{s>LOy{yP`e{At_UM-5 zx&(GvnH+&fKGOPWSN}~#%cB@%kzFtadHyN2>#2avj|cd0HJaPsR}Qb70i@1LSr%W( zY$`_zs6Ej5njw7*szSk+jWEf$c7zA!_|;l?!l$I`M+lwP1ui2|0j?O=Q$@+`{n@TA$~nN!@KOia%pN=)>Q(P?SYWWd!W?I7yVz>QW1H(x(; zHSACV<@Av`*lNm^--@c&ZCT5tB{L+Yb#+^W=Tt#eTHjw{+M~flphyn4?D>FDQ|0Sb z--JKfs^l1=^!p(n4o}owD?komyL+{G1EBsLnzlYZGr zH~>(GBpDUnf<;<3O3k4^L-l697BwmTco$qT*HhTz8tuPo$Ta2c7H0Pp) zG;ibb;s{K(g`}$C@VRzXN^!_fOL>HE1_2t!&V3P1+Tkqf9VM^&MmUx5#o!7Atc+2> z_A-|Dh7@!bCi>D@(kV3iIYyJYRJG~GWt6;O5KWZCTk|k{aQ&H#4-Cpua=+Yo-h?XU z4w7dkb<*q1r%KQ+Hh<^5Q}zG0##0Q+ic3=-eq*xPOybUcE+9VR#j<#02Fh3+5d+9$ zwTdr2qVidAu;p!@l`C4k*HG{`LV%ayhPPW-uIzDjf^+hL?7}-nPfCOM*KE_q+%SnH z6BYT_?R4>PbVcnrbw=$jCK(|6O&5QJ`ys)Dsx{)~?&R@zh6-1@ez(NsX~@`a3E` zUSu6n0>F=C)az>z(QuPkM%0+6rnrUO7Pl(dD$+2pqwW?s!smdCwe(?1KHi+{k7B#* z+Dp{fum;#jN|zweE*>nu#hDt=c$`A~kJ_vNAxtGQjI!QUH(|sm$s}jjHQo7|d zeIFSt8I38w1gd`1p)M#^yDvm*m)Hj8W7<^C)hI7dfN9=;wnnjwj==Wy7&rRouC^Ff zNO}~AYTi^lOKR${Q<=%14g=5&`6c#ua*;H}0(O3|WiE4*TF7QT|89cY??Zn(q+@Xa zbqM}xceZhO%6hyuW38O}MqSVr7lA#)q>EyY1dt!J8GnN@R@|N5IWWeK*t|DRsE`J7 z`Q8tlrM@qZ9Yp5O^5}FOM3}qK>}()W@rHfAAwr-Vfcm|Lq_Ikw5_*7F=BDAE=&?!W<0<9(SRM2_5i(cpw*7vCxvCH?+N$zB2*FK7^Y&hBiyYyiAs`4V15Xj%JCkMuq zPL*(TQ1fuM)sAu>l=s-5tq5&dNviU-w90`nfbP#Y>Q^DtU;=DHx)}*o)v=J{T?^9~ z@!8aU^setkFn{VgQzte{++6mth?4{Hiz@>D-K4K|j4$F*jeTC*Y~0pjM&6#UlrFca zdh;*Jo2ZOGT-l)dNLJ@?gT4A1y@%PA+N9J)JvVIX&cA`}gYnk#Gy@{~iYU7Qm(_7hJU zD<;0In=mm%7<14rzl2XbeSB0cAM#L#$220uZ#I=$&~~v~lU)ud5M8cYOin=O>)x&K zm(oMOLE|V_St*_WlHmnczkMxcK-z~QG0shgkG$qMnF0LT)>x>mThPZf>}2X1!s^d? z5T(_`&cODg1@rSL>VcLO2bUi}9}ClbdWFDs!4E3fgm^RxHTR3r%-!=sV9wzF{cN${ ze%sr+`ue_t*)=b5EyuZjtVA~AH&JfmEQqkWDtW*)^I+g#3qY#y*W`^BKV zn+|?4ZTA|p#{>r)VF5itb%C);Yl@#n>Tv0gu&GF_= zP61?#w>YSsu~{@W;^_{-3w|&Z=yg>%H-8=4Kk8BQm2X80zgX!*j9j^CE z)c>OE9-=diwsir=wry5yCl%YaZ99Khv27<6+jdg1ZQIt#-iv(}u? z`^I&l^$YA?Cw)Wv{DYppsg!I=7(bIHb$>ZSscC?={@2$2GbEL!F&Ft4~dDx%9A7*a4r&9B9&3 zN0Yw2%<85hPgDJ4|L&aBbQHFaJ5? z4o@q3ozTNwYuS$%07YW<^#L1h-QxcjXq>_}N!7z%Y{I#fR1@p8X41y**ArLuLg+P} z)0W*>)>@{n9+nlF?AQCoO+{Tw$xw6zXZk>Hc;#Rn;`Qf@EzCW>MC~Mw=`|034ILq zcY)XKb5(mGdG$@T~q}`^?z4 zU`4LtZ+<5;!Z^KM`f8ATQ3)F|(=Vb0Fwh0UlxZ^eOu|-VlsU}Pg=~Pu<8Afe(eR`& zW72S90M~9;`uUe0GdCU3B=X|#JF`sUOOB`m5BWm)c!%n>_P8*Iw^qDRU@GrD=*|3# zKrH-UM$TOLu@f+5wDLbwIh;X!0~!>SqxxYPl8&H=0Bq8hyixP>r?3+tvXv*)q@3b_ z@v*>j;y>G+O%Y1hqTqm621Ik~7NjM=K3QF9_SIi;8p!cD8nPK)BJ;C(tsx8U^v8XZ ziRk|ZhQk2};E!Vb7VEI0N2Fm=!+Bwit6)O^(;8E zM7O5jhIzPe+uEMwUkV@TU4GOWny2UcHhsySy$Y1o6MtIEzhW=@eB)J^$g3ryzPnn% ze~{gag1-Q2Z#n?QE{6y!t}-GsT%#f#$X#(o>Bj_ncbIo)=hnWgATWC^YW>w0pYyVOj7$lO#e2T{`J$ z^;PUkL#W;vvQ9QhlAQ^*cb5)o>SayiooR`w{572UG+V$iI_0y)z&h%I zgwm(WSON%s_8i1JK!G~+p7DvXpG5M5LVFOTS( z1ubBW4{Q8*U-UbdYro}B?~g|gl#rnb$s#Q~d=p|R|Ar%f{RMjCROxnS#=SsOYpOL& zCHy=-xUYs$ry9(r!P|z+nx}k(5bJsWX>Pyn3P7OIyVSde38}iT`}^weG5*?B9Wi(4 z(ZiCfdSVjdHJLbDz3fvcJh0DU1e;^!mBWtyT6>M35l0h?!#$m*;~ab_avl|SD~K@J z=1!<8J9kI99TB=etjm>YrgdG@y^{|FM7uG*cMld~r?@F-&5QO(RYCe}_KOwWq@e-m z5dfQhon<4(LG`3=zsp(lbo$UzClB^|Eb}?=Dowu{M#4$|Tt}~-R;Np(L5&u1NZL9) zDGJdx)#}ga(yvT4e>4uM>vf7G=G4)L5}tZm8gtzkE(uh&mrNeWzfR1=j{Jl8nnXZ} zj>`WE8cTKf$!*hB391itRzFw^2CIN0ivW$5XVyT+iGmqZFZb`CpX&tXYg9)v7>v<5 zc9(o1-Qmyrc^YO~6uG9<&xIe6BI7IBzn|=uYUJ9i@Vey%F5*oq+km(!>DqqT=7nPa zv|>Ot2Bx=T&hQ!9U5Wn-`XuwLkq>M<`gM}8)B`{p zbe{W&Bbcz^EfuJ#%>cV)D z-1@qoU_HX*pVPPy z!H|*9jPY{bioauvWB%EeI`DCI8lF4F=hM&5(XR*&51!|#@8;Gd4e%Knd*wE;ap2S@ zeM}7}{UZh~GI`<2{^t&=C#3%+=9$)*O9W+~0vbcW1Xs91dH{0r1WN9{F$Zv=N#IpL zc%7d3;ZdWjT)Tv|1*9vjzo$=^XqFYX-e+CtmDC&8c&Wgw#k!*w=_y-K;j?6ZIC?Q5 zUd>*U)b+;E{r&#TAuRsHPXK7t%fVQ4&+=wUg6gi}TJ5AN{JQLI^*q09qhTY+wCo6K z)OsHcaN{U7d{=VNBC?T^$OVY1zfv<-)?mv%7K+{)N?t`*n6MW9g&Wwv);X2o8%$85 z>zlcgSroDZ)aW0qNFIuNIBzPr46uVYt{VNl)?UeQJ~$WGdp{Xo^MqpWP*u1qL|?zl zW-B;1RA^GeO9+x#-hm}WWBH+Kx zHg5J_AaF~yHsLBki7!+j(_hDa?#F@D?vVJRer?xi=ZR=Uz^ADw2J6~sucL^%^R2~S3~A|msE6eq!BvLK(-G&_!`#Zfc9 zir^@5skPyLKSk(iJvQf-n))d|Qh1v&j?EZOeccb=>J2OYYng30-``x9} zt8KYVF7;7!f(V@F;z~-KWr9+o7^EzP6qTz^eoE8=ykX#YO#nh!>`>WU#+{Gvu23yX zn>D&yQ|jUu(e_A(b6#g25|Gwer%{OaYa+kyba62-kXnRTUAvcjy0XPt zVme>v%G}M78h&robRi++QQ=&h`e25~1*NZeEP7F6js#G7`bVR*svk@B4U6K>kZVXR ztcF>$4vW-OHstq8OYTVL4IuGqi^KX1BeK&tg-$8hc+Lx6Se9tx)Pxzgpdx%*VogAV z%8{PCW{|7k>U;hPUg4tP)Fkc{rFCn(fW^`fRU8!KRtWGS9g0Eigrr$9^4D3&OZ4RyY6hN==!t z^lOvTU88>E;380m@NfRB)^4seVGG6q@=jMx`8=;po+@>{fNKn37;3*T?W&<{Hgu)e z8u_g=C|!fVv(bhhQ#T;)Qqyn-B3$`cdlRGJFb$w5*GFbIA=#qRp2Liu2tsZs&TLsp z>EWAjRH~jU!NJzzIQB~U0mw)!_RM0%VIJq!3v%CMHA@J`dh zsAg!iPcIr%R!>-*8i7TlghS|*5kl-j^Z1gQMt$9qU_)9ZNs`Edur8?lxHmJ19Ft=pFvXe zAoM2a{b7^!7#u_lDMw!i1Om`7Eq3{5<3N+`Ji*mi@>}(Os1{;~F{_#750^5tP9{R2 zRN%P6vT8E7eZc3h7yq0H2P(f}pIud%)(XG?ftwxMBT=z0<+(pFA?7V`lPqf?B|ei? zOjxW4JZ77~`c?{*302D$(f3%qgx|f!UoK^784Df#C+9U(2~v=pc)kJ-yXPnShosiJE=w(^V^Ge&!D^E1w`)@tDQ5RnW5zi#*)>5lg%;t5w55Pd z)in?vs@uk^R>Nq7T{4n#Kl091rg>1 zCoE75A)AOdW|HEFjZdX)j+gZmm+v+mWH!Ox!|ue_43N9>$u?wXX7I5$9o7wb((jR7 za%d=@#_zH0X`O_9zR2zQ2Qajvhb;o|uiCdUZg(@yJnjcf)Uv$YIZJ~EDCsX|@Md$k zOxupA+Je2+U~V#=BGG8sfeY`RH%?6}@(&92QtS3e3agOeTI2ZKk&?S9hzyUW&yc_f zh2TZ*-5FRI(ju5Mok+@KdmSB1cDe`VQQbUWqq=jvVg}vdc$^EtR6xM<(SD+#hB?mp zhw+ylEshi%{acIA(+}lB{aNd%t=K(odN}Vq6T!=R1LFVq(yWH48S-V`$3fI-_Vc)% z)pn3JFPGAgb@CkWMXC2CrqeIJQ}LjqqS$G%RHdIkKa-)~ksBz# zi_^fNOUn+EAnHZBLMzQ>ojW+i!8CU`Yh0G2C|x%8Ifp4SWU#7o0PDVsK_E2kib54` zhc=T(+bhKh+W{_RvRW4wjlVkF)=|9AY!F<^<=SX*Po?p`BPG#GE*80SnPC0s{xq)6 zT!OmNHPP(f3eI&a*ZBeZel1N!b5BaUI-(}tm`iZxO0}%aA6wba|^{kRgCIyCLEp3so~4ys^H6HuU}NBtfe7gue1Fn=~K)J zdWbBITkeEpwG5k%%OuQ0(!~i#J+@1t@=ms*evQLao7YfkFG}B=?FXn&Hax>Vzg%zdHdnpqnP@Ju z%|5-maOYsM7|@rTT3Lx@{}tBm^y4AMIA5Vq0^aY4$Eq8k*~SFQ(8S?XYH) z#vTjzNNA6Q*=x~-rE5U@|2!lkrzcKG68nQ}(b_V|@0hJy-1T}V4rmKhD&3kc;ZJ2j zCD7CNc(IIj(fzw53`!M>Pxjx@i+lchP;}b2lvKMXtycg5fwg;5_Q9;w8lBXkEk5L= zQZ!rapW4wECG&my^M4X$t1>)Z$|?1w1ZS@ji>&K^uXWp{d!&APb)FohW?J{8)N9xQ zCb9v*cQ5$0x6GSFeSC@D6a5TPWpypocdJr=1e_jxb>O^0`UocjfBbw|GXtNQBIQryCo)8_$TEUF1^U~+6mFJA~^!F(xrhlNK1 z9mL|3!wZqu4` zRkO*H=TtF%5I(M1+`$M-712D@Ux+Y7|EHonMF|5_s|m~;-0a}Pcf8$_ls+b9L#xPW z25kobg3>88wCfjD<%KWJY+T8@N}I_ylzcP&L1!?c8F&3g0Kt|s+*vRnv+_p;?*TD z^X?0*gOpiZSQ9g*RRZKzFgLKjM+A=_vEK#IdHc)G*o8wG;itO@(z~?13E7udRlxxd zwoe^xR#B(Y8$l~@j@W}clz5SG=h!`eoBWl0bLXxJ^wB^B<@T^KWZJ|==dcbdp0T7Y zsSGjYRR-<9vx$PG8ysm{ip^XB74O+aQnE>+a~ZxZudWOyZ7;sqY9Hs)7QS;$o0tPY z%kqh~!9Ta4qX(|DpSfw*Q04!0%$jsm%rm!}2o?e7H{+~KZ{9TQ60riQl6sl6xka(j z%u1oIct7LC2`*Y4%{60TTM%4;DiveRFP;78?^Nj0w-R3G42v$OrNNjn@oLHLycIIU z2+||VpHduC<68K;s7kP0VN{93t`rKO9WKY9%YT$&-C1VeJ+?&IK`A0YH>H!~lcO4m zpODW3yt8j(#4~_yFtq~R6{!5{32W)OM#^rm{>*G(T}PBmEg#-gN($#Umk@=H$?f7( zbnct(;k62B{q^iEORD@LDyMKX=tU(FbD)u0CX3ES1~8w|%OAz6rW<5wTi!?jbE>N8 zA8Vh0$6rsK^1$#TqgdWCVzP%wQ^RZ*^m8q0s0$HJ1We4oKh%Er0*&O|SYFZxSS=X^CFM&#*VIlG~QNeHQ=U_*)tewqwF9EZ`n{id2TKsb_?&q zbjb+T=xS^Kn|FL=sG--f4Cr(Qq}=V>cYp5r-I=%uj*M5crknD0eH+OI5>EK011BT< zc;9`YaqAxV+eKqi0}Mm&P!JQ^L=utC)J`wB4%#R1cahSh%#= zFp5IfT#;u(t&xxo<}DmE6slZowJS;H`+NfN9&^te4IClFLdd};rJM(={7WRg)8hi^ ztm@18Gixz2mxpJfdnHIXwBR=>J9LwN$45YVtgre)z0{Sah^d6YgYIQak-)ag5jA}q zK@o$AU^_|eNBk3lgMuUng!ujQ@ZHso9^ELfb?`q!OTvPCr-G~`cMD$Nu9e_J)Ye1$ zMe2duxpWBbNZLu$z>Hb9QQ{eqSAj`uZf84!0Gm6*3*S=SZXo*&Qi2TLQIp>4DKjzw z9opGm00Gc3;f#E%DWSkv-0s z;}VALY`|DP1f>iG6a$ObG^1h{BTL5wqZW|lu$|+13&Al_iAZE&4uYQOPtCi!iYryO zxgP{wUL!TVuR5<_Ji$sXg!xcm${LL3fn7DM?)`Z#VP-hHpEZH&6mw}g{Q5zcE}K)F1Oy(>`Hm6WZ0N9apYB( zhTP{=QGTUrM>@Wu;Jp9FfC?nh(~h;VR}CH&lAmUM-Eu~UfX|`H>eaYbC?_2$M{_qv zAtJ;iEDMbUkmBF|oh%^}2!cLFJ;obHo{F0Dl2mFL#q=dbZygVh7jQpEIEL%^^}&tb z;+_qLN$` z2DDRmQ)^b+I0l!4F&ExFiEXBQ?3#je#x~6jxLDOf7uI%VIwR$K92-B;ouzf0dMQb>{n zRC-|O;VmrhTcmLIiAGOl85A#k3MNF6&Fc(}Q$K?;n!!2%DHp<=QgxXq(u!7vSdbvI zD~Tg-(g?2Ua%{aMYquBpjNo^)?Z?b0UgZx1)Xyw$g)$s>IeDwMMt}=S2*3YsQh{aS^{X5>1Evl z>~4ZtO_E-PS7r{yv#!(IU)lA#sDcFVDYn&$!eE!$a;LDfaUm<&<s>L)8OCR`g5nyZ&g<|$`XL# zcFb>+_c=-VN552XQ5`*oY09emgz53g5Qw4+Fj*Cj zf6N~fYeUqZDo~5B9QqFCbo>(JcyAD*tSDw7*k>xvhvRK%Je$b27At{guzxU8?N;q2 z8EwTe$DQv^Ql@0N<3y9^oo>QuP0VSB)!;%%a76{chM_!4tnV72^)$ihVzkxhqDXqK zP;u9t4{sgFbsO6~#&B8u=F-*$r~s`d>7&WvU<_$aw?J1Qsi3f|x|4WIfZT1Bv0@X5 zUrzXVGq(E=sIuce2Y!W9j+l7D&L>h>_9u3u9PnnpnxePQ!WF=XD=);>5)2&uHTYmJ z=i+mJ!ibIDhxZtgaLs1yf@ts2qU-(i0cEjEi&^hZ3S<@~c)MUNi7na&m}tR4d^lyx z?RvifE3jS9gEtm2@$mUCYh&D)B*#9qbk<6izl4T8$SCDH_8NY2y)?ST!%E*}XY@F{ z?joDx^rqMHMXrs~bp+L50ke7Foujma#rsA7Fg|5;H95Fq<&j74ICH!MN4%2g_qdIp2* zITn5^%|e*RZSfBOvcKt@6gWd_6{?WZ6WDs6-!&O4j;SbiYrhRXvAktbk(DbBgirga z9@nzHe}G$cbb&YlaPbFvZN$0ALn4HnRAy{ikc9$t1xWLYlgH&9J2m?e`MX42w%Q$I z(GzQAXvxAn$SqV;3gl1b{14}#XOo{JJDo)jpL>u(g_1saZ~NciH;E^$nT?WTnqfF{ zs_Tl^bzsv=9Y}_h3fN@^0Y5>f3}Y@F?KlDhx04V zkN4wb$Kbmd>%;#2^I#j1kihqAadtVE^10jdVR=Mc-{&=n-IAu~rBR4wxhl5v`|Djh zvBFk&?E6dpuI~~N;q77MnpE_ViV!|QIH@nmPAkkFkYW;pz>l8-uySIcgM8f_DD-~r z@pJkc6`eD9j~6Hb|+KzjK*_q@uN|Kw`vu9D`U;oC85U(WR^bv+jJSSnWKB69ht(OY_01Xf>oS=AkBD7p&$E9cFJ<%F6J ze!ULR3i=G<>xu3}c}sy(^g0+=LzD~q2ma@ieTU=wW%B>Ud)tE*Q#9y6KqrzwKvMr3 z1>StOCD~?DOTTGW@RB^hWiu!U#dFJ+4P+gP4V4c>w;q?xzanG#$eXF(%3vkWS zFmDeaW=^txslI09zXGuHKLSvt*pvW<1B*}F{MbA6f-G8OLN9_0Z`XhsyB?HDuBMzg z|Ma4zFc_V*Od|oKAh%?hjvYDbI1pa>La#_Bra>c0REcp4&G+^Qr-*CcG(N^ta8f+b z)Ds7$ScBVuR}nRSKk{DP{6d&og?L^BP^YxpARK>z4}MW)fQmgL1KlTf3U-X>*>6Gu zJ7)Y!MdHDY_8><+5lt9dQJ@BXZ!=oDknCR8T^#e~lAzq7=z+Y<*h2094QwaAK?DLkK*hmL z8|Za`o#cu1{fmkoZ;cCM4irP~e*5~m34|tyN&EGpKs1E}HfT@5f;=u#%dVUbb@=-a zZwzpnBzG+QVnM0%&mVOej|V3bvz~gJ1u7bjx3<4mH+_2>ZrkPP;kN3B$6M_F@+#R5 zoj6c*3RWsK0}bF!{UwXEBy#|`ahXU&VyB)^UU{@?XOx9-(g0C%m{8Y}k_Xp!xNT7v89DFwh&e!j}I zHh{{4Xt)EP+sB%qg$K)ukyxUpL~K?xDKf>>EcxXg_C2~b*YHq(@V^7V$(5tJyKRNIE5nNK0p8vmWR}}+LiH#LO4?< zQ%vHU+{>HiKWs%zP|7F%_2&gYr3A{Gy`byQ?+_-ymXbvb>rbn9&E8UDhW%4c$YTx= z#sak0E_SA6>uuL59Q+Gb>MG@u=iDZ}<9KY>Ar5i;HYR%Hcm0GAj>_o>%z{kd9)W#` zdhN*P(BO($LBoJw8ju#v4-(CO$_|U(q28D>6KwVykTZY~xYZCm>8|W%`2l9hEdW?+ z9?ivf+wViWcQ%EtOGs^Osl&cyi8p886>Ua1h2kvOxyAh2LRN=C8iZ^J(=2~m^GuGL z{x$^zivUkodZS)56z)pao1hRE41D;}ZBkIwMsoyf8Xv$6P8;w=S>WY>+Sp99y+QDD zS+FZm5A~_@@$)!Vff3~3V<<{d@#Q?wqNx|W();i7iNbtySymUnx1-wq@eFg>r2D(p zd$VRo(6P>oEd42VyS%8xDbwQzpJq=!)1A=->y+kxrDu!ZfNYh|d}Icr(R+*LSEbh* z3f!Z-XAD4P#amVOC&bg%MbPp0s$GMA0z2GHGlHu2dY|(?nj|y|f^Oj^aBOlK03MD- z>VA`i0aFjYH|p;-Ysy$%EAs7CVF<@Zq2Im9*Q6U?Yvrb&Vng#d08h8}bLy7K+Z*hM zC$4xNUwGK}IQPowH;m8wJ{?^cT-T%41}Q(iJpdBA_ypQ7YIEhcvl_;y0Q&WX%Uf8C zBj)uJxy}@Yn~-ZZem8rs?10(EI)k|B)eYGy8c@^^i<@|9O$&K?+2|2SoIu3vj^W;) zFdA(NL!AWF&~0q>U-0CAXC&Tj4SI+(sTv6dpF^gqU^5I}v43X13rjrp!S))Yg`88kJJp}X@8j6iaw?R))-F#@=wIh`yG*RdRX0XsuAs8pd zM~A{&QKnD~+#7{JprN|i*C=w&%-{kYxN!|_f3RQNKJ`?$zy`34sUk%GrHKVO=(Vjg z-Bc7H7-7xLD7XF@RzHnvbDh}x{FYS!i65J|C2Va4I1M$AlL4xVXpeb`Ht?e@pV)TSc3eMo+p6t4(5&7H_G*Ti_og?s1<4zh$#`&oigE z!($H>E&fe3qC`DlWJ)`uuz-*RLD0TA$}7t%D>*&TcnO`MN5S&md^`)FU;D31gkc8G z5F?q=nt_O-4j$yLg^Eu96p_F4Ggd)MUt8UR_>W3t_)&>=0$VB7s6J7HBS6C7nKX%8 zg{qSFee?Rj#}fZli2zQ>Z>Na(%wS?C^AL%0ribd`{TxjWM+~Y8svRdDp{^t|l5Mxx z^}4GP#hDE9Un4A)0;YS$KPoYr_YM~U!$Y7X4apLi1T@0_*X2+E5l93YI^9U%arl;) z9LY9)Sw!{fsFKh<))*8R=s)Oh{2igDOu2lJ%`H+s8sAy`X+YJ`W@lY|r=cNSXSa9L zu(32A|2ZK9x86T}2(G&t+|{f$=|sI^F5KQJ|q&F7)=$NfzQVY=sBZ2oL4 zToK`D0R9u>`=|=1gCG#&ao(sz-GVq7njt4hH=uNo3a z_|MvE-Ar4%CaOCR%ztfd?Upgy=msb{aLs~&SnBr{mhgbs+hjZS zPW1#2gFZ06A?qf`TsgE!d{}eUAc_UBCvK7Svt9CM?f>>BwA zsNHYvG{vSaq)A*y`;Z1O@9%kIswV115aT@(0Ik)9mNk-hw|m&iAQH5o5nLaC87N}Q_A zbpp5>=d8~oUP?9xn7}Y5xI1kFWu_?nb{>PLivbqqNga*BaQSz!i7Bw!BH?5MwPI2_ zfcUzdcj;KluG08}3X@Q(thdK931y@Jh#eQiPaG?zxumVB)k}`#DQ)&a1_4UjAK%<0 zjh+MsWNu9|8T(=Q4GEYS?UFIYKgx=?=$A?#2e;B+7^NE^3B>40B*&HcU;ybI_0iWK zI0^NO4dBeb_cSsQu_R>6h0^~ZaL7#Y0Na5UviM0@e%oQ6J)VE4tpPyDCvh+IF(cPC zOJ5}Z(-)Y|nTozvK`N51VC~`xpKEIEDk^Q;fT%nqFA#|G_2ao4(8 z2xIEfpQCS^8&p^zCa@XB4fHC`3J39ZP5$P;-TcYX!}%e@Q2a@;7?Ln zbM0*7HG$rs4px_z0IC+``+n9H0LS_fV6z($y;n@y!01#83j=ZAXow@&w1~iBEbSR4 z8aYqR`qy5O#m|7y>LIFVCMC?;X!n}g8dt-TTsv_Rc+fBw*vd{t+iX&)P3h?A$r5sY z=-*>QX^e#qI7nQ1C~(#hEdl{ zlQcy?JzQNNA9`uds&#vyWrT*+p&)k21izhhx?MOox*YQ7ZhO0-~vo>gS@Ytv(gd1$Sw|2{JwQ(v-<1LQf` zDDS#e53CPQ-@Rv=c1#xlVPfJRV?&?pW8m-hkYqu-PU3{Ewf2jwEc|O-x8`Oox5}-2 z`ZD0yO7~2WT9Z$RgOlP_P~im;^j3)$`P|5u9!X)lT;qi`8)8eA$miTEtIu*2%2a(Nlys8^K{qUf81kVLH*%R@LsW>UZ7BK6vT<$Uni@po=6G^o&bu3s8P`%@ zWya2?-XI?K(sKcx8hmqg!`fc=BJ+eQR74O&3Sn=|$y=d|s{$xZ1mELs5)~F9^c_gL zJ6ND}&kJ_FzaarYbIXX(TdV+4C{k7oCB(0Q%1xfcr#+3X9$7}zu$Ea^Y4IYcA*8)% zM9neQCB}{T)FV@ltE)oJv3-BW;|UXd8WwqSHT>O@y$P0LaG3eLls6Wm5$*(?^?Q`; zsG9$A+D#@v@Qxk4du{8+;;XOK7EetZ zK(y$u=D#Y`HT@uon@5nHw|RVhmwFant&D$8DC~?%_X+P_6Y1syR+PFi(Y1)ES$ks^ z^SNpnasks;@ei1*1TSnIFk!P1mFcw+1#0rc8kwnc#;&Nq#T`;r69@<(G*X$Q4Q^cC zk9OHl+K>RYT@_7gn+D59QpVtdwMF1*Ic-!pIH#9R*XH{6es1aA1-$02n=xZ*@msa@i+B6u19GNE;F1^^SA77)D5zUEVah%S(zi(6Im}~^GSFP zd!h#Z`lAhVZ7lnzL~s(uUgr`CK>^~-U@W>P_CtBrsB!H?F{`23SWU%h8A%d zLI{uv!$cD9MV~VlPV5t%*%J^Rw*f0mViyn~_3I{}nCfPi{d$krM-N*DiaEe%$5%(W zicA{!UZO-XyO*sdp|#ZxN6;N!!tHz-*SQY?k2oxss{6fWY8s$!eJ#0GcJ3D4o2h;6?_?Y)Oa!m)d@(Vu zGVBzyIBw9HTo44VjC*g)h<_mb*3k~h^ll=`#N#bfTy{%b@AcyzXG^w3u);yJV0Egi z^s>nj$&;bdMU1V(v$e7NxzjBxE$F&wga=YSC!n3AcA9POoG56@SHX5?Bb65|B>hoYvIc_t46ahX1*u=k?R@dAnS+&u^!AZyG%JcFi4dCiL3kRP#JC? zn(?WOo%|zZc-)S}PzC}+sqO7T*Y%QJ`x!s)w3QOvmm|(MFcNX6i70l%%>=ooI{jpN zmdS28=(8nbOJ~m5%?krO01Gz^Wx!D*zJ?P_K~S9F3ZFi?jXIj3!@Y`uv=dU|bg z1wIobJlBtUg!TT7T32RG4a2=*cLQF>0|4C=mGhNb zbZ5EBV50)s)Y#SyY5=FnVrtpRP&fDo9OcL*=T@^sTeGk-IfE9>TT^GYMbC6C6e<=H za)WxL4yJ!SmkRmA9zoE4*rR?vUssUOz13BwR!@$)DTUfzS&SEW&@Rfapa?TcmVi_p z1aYEuimVOzka&a%NoZY3z$K=?X1KK+3l|^E0K|oml4N(a`CPJ87L);StCHv2V}Q}8 z(lD0$pB%1IMO_Y#WC(O)T2f@a^?F)v9ISEcz6VL13*5GW5;}E+7#QI+QXp#dosc&& z3pV6N+K?Ebe0;Ea2y6M#+VUIe4D8fi<771tT_8Dy-G~KOoxkW8fKzfpi-CDq@x8UEv01%)kIUE z$4bTcEZ1mJn``s+F4aM=tVFA19MC&@bgQsg^?)C8rrvz@)&k23wj$wl+ae0NS+}SJ zslj;Ut4}VOoSyju06NG(8Q%;NRk;$jl3A+b_7trl9536<%$3+DC5k^f9D~EP*2olB3ZD$#YP>6UZC}ulsEky-JMv~=1XCu;lcWE7Ez6x7VQ6EsJr+s zt?oVkiXJxO&gGf<$19zzFjW_3frqMsJC*f|ZR97+dwZC71F$;}Vbr9+P1NI+p{sG* zBGd-ZPj)i5FKivMY%y}_qwEQ}aiF;n87?o{i)Pxvch;CHQ-KPW?)-F(o?JhfBiYv` z9=TR;W=h>8a+QYgqDZI%VO#TjN*K*Gjpw9}xdJWjJn}<2I44^m_jpq~d0PFa*8!iv z2AN)tOL|(W5A8?OG#5;<=~&-uKVXcV$nllHQtIkFkz<+Ki*7Q)yr4#Zv*LlU{Q ziNd{H%E7355;~STGqulep1Uk$=YX`0xHFCD2eJF@28p?vfQcY z3b581j&~z5oWM}Vv(Ga8rV8~~+@+~>;4p`k4WS_*LuV5QhkPPn;|2VAF3Ri}d|*Ox zb_1tUEnMj&=wI+#gax<;z5)^*TP zk+&$==e3cHLQqu@%d9rGAxRkat<7`{0W6wt9By^sHwDQi9-q;>rnIcKYdkbKWfVv+B`a3m1D=zzz7aGSE1WO( zQIFp>+GE7B_+awt!HjC$6jf|oL!V$2>QM1omGOMpdoQ=+wKx`kzf&WX0GfrY%*BeR z$PtW9BTL9KRTE^Te<)u_1D7~)UI;(Jx3kbK<;Yq6zyw3geO6ha%=P3RJ=BT^3dx3h4 z(ekwsgYbk%i*luig=wB=Pe88ey50oNi>%AxS0JtVvJ$_lZn7*{Q?0%j;X+-uM%NJ2 zuUmgvv{hzE=yZ&tOw>Cf(OGbSmAt0lgS45l=|+~}=Qzo>z6*B-K#-4UDaOchVK3fh zE50HiX5f6})W&;5OR^K_*W^%#25%E!o3Bixfw<%GC73pca*w5P+QH7Gm+aPlfH&jv z44E4Jd0ZA~O>oV;~q^OWN5*Ys%L@QGdmCI z+Dz1~iGE?iBehNiC|50F*+-kuX(Pt0M2=_0xe(R5LxXIkIR`&ro-}w^*<7n0-#d)q zPqB3|D&@2E07-5r?Cw{+Gk5bHXD15sM|fX>D70XnmV4ym;j~c~idH{7*8D3V!!}Ks zzuJh+%xGU6o$m#S(1d9}^rw_}ZYd6!&^p)dasYC5AW)tHC>cWbMeLcc?0d3_{rTN^ zBGEtM*kp{C_r(aH@Ywk1EumSS7LE#aRu8VOzlL}pYcEQ86yJ4vJRz_8J2sKJVoZT?vSfUVGHJhh_*OF8S-RVuLV%4Xw+Z(GeV1 zE!uRRZ3$TWHJF7e1L=o@izi9tnJssHEw}BtxfLV}5R!2*y-PVU#ba4BYGpc=W+38N zIn`Ah;VnXsmOA!~H}p_?TdW141fttm*_3q#(}LoKz8^>daAsfd3Bgbbaxu{CT%eS+ zN!jl9+1ErQ8u!QBE$-QnT9O6%Tuvg6Q-OYCp`MViSBoo-Mp^H?F1~;e?za&r$BV0* zDKc;Zd_VqL|9VNyagvV^AsByd&t0PDj19V7kG?j@e0U6)Uj2T3vRz9iJdC}$W*zJn z2)6sq+T9Do-JGm@oNwR!R>5NbIF_`{m2f^4;#eJj~Yf(t3~YwoQ@I z;-CEbWJL6PmTukPS$A*o9_vtd>2Pme?%d)5tS`Iu)G$N_p(^7({G5%cM&3En)8!r! zU#r|dA69s4+J3Mt4s4r_7{p?E1UW+)I+b-iZIPcp@}f_aX1+Un#9TY@=f9**WHS#} zP5mfwj2Ml6tNNQ_RK)%`mLgz_zoV+&w!2Hq6rr0+(z==t%?vh&5e>j_qWDD#nPUpr zDxA*kp~7vaP|Hvw&Zmeuzp1%v<1d?C;Eg68api83qp!@FLtT)L-INaDVzQKUD06L! zFBj2jFI25AWK5l@rb-%pb0!kGubClnZ>d=zoKdZ+yGY>J2{u_jd?Yl++0v{Sk<= z>M;=8EmyoS;(mdA?il3ZBjE_pCKH&!I&(;nVsR8%J)3*~pZX0wRegilonEQYiGx&cd7k5kD>R(yJ#Q?xNM zuV(L|0T*At&0Uav*CXUM5I8S3n`PUm^kl7CmE|vh)UdCi2zf4buHBB6Jw9SJqin3k zQ%0J`>mGr-hm%Dt01HdXsecsSIykP-DR|D%}o$hzd zboPX1gj7GDT;4~$;@p!$$qjAsD))kBYAbKV`AWh)szbS{gMon@G?oX$VCb|* zM;9;UdBZ{T_S!MlxWe;RgjWlL24xI~_#Ssy$u#nriAb&%w&77X98#t) z|Aj*dXaesJM+((vWq=fI7?7L3}7gfyM20XvtGU?X5UOw zZSde=nfm7EaJ{80RSq4wGtNnSy}1tiHJHAZ1YnUaZ`J=FzRoGQv*`QsaXKB_wr$(C zZL{NVY}-l4wr$(C?WDuW?>{p&Gc{9lA5NW@TXi2!-P&ib{aGtV0r}Sl-$yNa6qBHQ z9H%5CY1R`B?(m@?uE!G*hqB}r<#z$EKRH==NnLYcwNLKTU*~@cW%~KI5iqODCzmub z-kni+Zg+=YXk?5@TYv@PNmO=w-)5u2aHI{VP+-@uya*eZ>CsH!BXi^5jBGXiJVfqg z#-xxq(<7qZu;{}2cYK;^i@yU1Ap7grDwBst=P5$o!OsI@K$Ro+lL+|eSQZI4BaDU} zAV&)gHcSAP+Uw*R5M}8K1AJD%WH)RDa2$`+U@92MAZ6J_AAnJO^~}l*$!&${9-W!HhPPAE$k=Y@w4>rLo0r1;W1_yj|nGt~5t>`ZW&vK7Fn&b?U8kJ7FmIV5?hS zdV61<^Y6-QD-{%-gO?@5W4`S&?cEnCzc;XYV#Q0H@_QuD<#y~W(Bh<5U$~2`GWayf z=8fQx^Gg##cL9y{P!(p3mjCim~rv{Bs=#X z2f0n$y3-NnJRJ{f7lFVgqv=+#7*r}jbh6jPDNVBNd>(9}y>YCmT^t;UKb(^!>iAlX zY8kMOWeI%N?`L~qRWMaey=JtIiZEYxOgu&_-}3IR}CW^_+lPnFo_iC~l6B2>cG zX|!q!JWMfQ@{d9H*&Jdkr41da!FN`9JxzDCyF6L@`s|%|bnm#_=wX*rDlg>c z<1v9;hyadGk6{F=717Nzxs2a#uQJ5eYk^uGlI%+N)F@Bunnsm5`jqKN1K(C70 z75#=rtJLD{AuDfx9`Po?zz0(1a#2H?um8I7dVoiAu<@l7>>Q?g;2xccIR`UdO^(=` zZ#&~}Bh#qTBXDEzf^r2x0NHxd?4?Lu6$}j@_2>Bg_8Rq>@>>Senq0dK;dGF}#w3b> zvH`%Bb~Y?V`ML^+s}VVw8^%{#q2|rN3_F4diGDjjT#3nyjjWJ|`OIr$2OT7*hghPD z`wP2kJ&#+rQ&S3>_wuftWd4w?DMR%?-~W3@=Zby+*8Gt6l8P6XOae>5Xepi1@&$Hr zW4+Tw1JULmq+XBKee-uCZh-h@-FcrW0iX}V1_5L!Jia?Err$-V(iD58j``PfV4}N_ zi{y|iMIVAbmpTeT3TV_bYzX`+Hwz&wYi$bH(ZBDwfn5EycL)P?L=)zFq(u8 z6dV|KW-1(gzp#JR%)&8KBOAsk94wiRvx-E@j&gGsKPU(PyKFJ@<(t=|dOirt?+v%{XtYAMD`KZ|#~Vh6b#a4q?kQiU=Fh z*yQ`YC(W}4(kM6D=Dx=R5p&R~569mni zLz>jlD#a4XEu;|FAEwm5*NIR(|th_vtXD(Rjhu=qNeRIBw0xt%eQb?#siIXo+ z@Q6fqp}cG2+|kl7)O{p)%W8f8D!DwJ>glilN0wXc1HVSgKNG2LC7k0`y)Cr{FrsGOq@|BYIrzK%;`PXrxVfD&Q zN4JxkuXEZ(2B1hIOqYgiR63ft#z^pref0d)9}!tmUNwp~5j?;R3{CuKaY@e$-i;9= zngTi83U%2x&cvK;Q|u>z&e@o++@Ee@XAixSNKRv6Pf0s2>Y+F?hDZvRuRJAW!_3{0h<|J%f@(s9p;{q6gz* z-j*bE9sz$iE<5DLN3GK(PcI!Y1|86#PT`G%2;cvWpy$jY9d8ly>ZA(oheLMJ)N zND?YIs6g`OcPPV_K55JKY-D3Hlj?@NGgh(-wZB%jg5K9S5rxqo$&C^VexPWi!>Wu+ zATwaGfs2M>e~RnL?rtY4^+cG7W?_!Pk1Q9fy98*87MZ6V2j2d|ad-x$$T?03nzO{D zICDml-m_Z6`3#If7*0}Moi{waX8dO~L|P1j3~eN*9GLICu|F!a0cebgq#{RjLcXD% z&A1PO<9arJtpsVCJFoTGD!gXVS1Sk6XrfHuY;zJLJ1iHcA<`@S5<}cqYL?fY(7mvM zdjXX3{1xNOvrzQ)?@_fISHo`<@EWaIqb&ZaeGYCF|aIdetnMNc@pz)EBHJHH*1gc+=yD}9gS~k?>*2)fAioeP0xzE?3 zWfcH1%qxHy@>#$MBx=x5E2mkSX zcDpww%(#%}YK=tpTC6XjylL zsh^)PkX^(eqI^?3gWJnr(^^{MzO_Df4YpK)?q2cP$ZUZy{ay(csK%Iy4lO{;pcUyt zH2jFkeBk+>!_jRmho=F~{$*L;oDk5VeZl8QRR;4AS)(Tluzre}b-nl?OdyUK0;6^? z*TV)e_OZk(0Q=_|<%mh>pI|6_-$ zRb?^fCb9Lr(lb|Mo%}A^b4}m=@922lzUbEM{H5)IoVfQUp^=`Z3DDELNCwEk#z8pH z$D|>#PdiaQ{$1h3mNjgOx$?%etEO#F^WZ8LNfnq_n?`m^+-sDelJNMbt;|h$t%T{mE5GI)AQ3bNR%j-lG<5u8ghdhHx$*qlpenv1p5s;s?|8& zf}Le~=Vts@j~kX6!MS0xE-)sweKK$W{JAoE%KbpYJaa0Obm~w1`#3!Qo(4wZD3&EL zk_M51xZ&(1=~|2_OgdonWnl@r;8L55s!uu3-kDDFYZY{ocSb&ZwF7eV3$G*(2oTai zAG4s=;sscPUI*$p^gU5w)kwsh zh9BV!*V_hkG-kF&O{gK9D@9`ozdi%K?b!qj^l#{KkIEv*Xk#GA8`-_c`JroEaSWjo zyXw`-KH-6)G|Z=D#qLwiW1KfbSIi(*q`IS84%m+9(i@ocYWm~x>GF)k-zO50APaBm zfI_P5dVIMQX8^(tGh8*`%s!M@$A> z_VY8@^9o&IVeAH_b=|XQN;r-5(@`)xan40Co>UQg*9lN+8k7?u;PtmdqsvyhPZK4F?ht1|-7+9t-m(m!55YC!rj7eRDuc)mFVLBLETTUTE zIc3P2+W`x9`#)#=(AJoGEVk524$d55z02=XUB-RRrVzY#w?)>O5X*_AQNdXAudjRi zBzDOpP0DU{tROj#CjBWGc*k0Rmn}W=UC(!wt4I?}}*jXIsi(qDBoXPl3|MWNm> z8bT`RcVq=>#%N}0F|=@HrKwZ%&Fv-gAfCe42!IY(h*JiHzR3^vy&)`|Z-BxJm+{g= zYorF#J>t?(pAmihcu`ZhQf_@W>nxKg#NG^}8AJ`Yf?R?lJ{n@!AEtR@?7V#5FpK6- z6+1B;Q1~x~rvKf6RvO~r&A<^vhwy!!Ekg5|IML~T*|%K$?>rbL7G0mELBP;opv`jB zTR;%WP9fRCd4r?%oCht8m?K&A$$o{Y*uCWst-f3T$8&AVC#Gz}t%XEv@)BzyaA1&m z{sNnXtT@*4jB!0l+fIP>(85VphEZht3z`VKrTI8pX;B`yf~c zK1z!$O~03cc^sG9Ez!?E8ln#oE_)n=628Cx&Yex&!?|x~`x--^%m3+%YiZs|UZZi% zc%vQPqEOMLLAB_)p0Ej_Tr4#G=gV7Rmrvy*ncy(gh0P<=X zI4Atndwa5iADm(@_I^o)Us_0&r>8~TC?XI^3kgWj`8TW`)+3rQ8(d%;QML;6;4s%pbiC_>?hC97y4i^e78UNa&;0v^( zgVlvLTbu8R!AU#lm2dYf4_lr@`rK}BA5yol`t9bU=|0c5xD!Iqgn5GyKv8DS(qa4o z&xYcc!NE$lT_;fzg-dCtx(3c%H6tRh`%G z{!-QbvYx;^JJDV4T0gJ@s~p}q)GMhfxB-IvXD=pWYqU)+2&;(m7=K?1P+e@j{h@^w z`5qq+Q<_HE4pv&XBA>EzXiAh*$~vh1(YIkJPj`WlH@>390xtPQfRkRu*z!LM`4y_Q zQB(8-x)aNk=_6bts^wI$Z-a4;XgTXLw0|ZTsuAufM{@h8P~xQM`=0Ij6rIp*2Sluj z7xp?1U_FEd1A2Xj7(zxN#`6k`v54#jT++6?=WP|t^x>jq7k$}i+;i5|s288Jfy3iwv+7);Gb^T~Y=5(T0d66BI*&b0S9tU|q-OJ==r7^fIRP`!SKe80fmpS1IkoNM~Ks z(`^N-;0LR|t7)$pa3nck`wn2v1=Zm6_T$MQEIUrH^F0kXoi{1qP|0E5LYHjj(Mq;->m?) zzQ!X?VxhJh_GFjb=mp_+nWbc6{VDRJ4C$y<$%R|uGU-q-ywC`rMW7h05xv>t^Yy(SLbftf3*Go53Q^YVVR z<4T?xRSc70cHMhr4WD`GY^@_jfuYq94eU62wE02^^OgO?PEOxwarR7=+DuK;B=ogM z?G>vafQJ6hKd(?tE%b!lv)ghOgwe6Sb{JUPfo^E8vCTF9%az=0$Pr+G*#C>IbsaXmfAcQ-;Z?fq8M#IMm5}BJ$%pPUB{S*+1 z=*Lz*?<$7vZlQr31S6PA*F_A6Bo!`xSo0#blV2AP!;npav^6GMqsX8*?R1#XoKAg; zJ4fjh9MKnAzxd>BErUPN94N_#F+1_(23avH5n^P43k%Gns zfFs=**zjCe(jFAFgi$^`nO9t)m8@r-W@L8%j*DQQ4X^8j(`(6mUkrXk8>lcXG^_s( z@tV79CN@8eE@>_2M3x@S{Ke|nz0s-TGLujn_tFK? zFUdnQ$~@)Y#?v>$2!v^S5X4FCeU@wszz}B&JZ7B`j}j0FS<+-+BkcJ4v~xy3qyNKt zKHeTI-1ov+X>YO_t^NFgPB7S7^l;FEZD{oSQK~QZ(=b7__#8=4U1>6>(ZpNNZF5nTCg3n}u$P{ct((1c#i$aaM2iNQD0)Es^Xal0d0Y5Ct`YZU2$QoW20oC@2ZSbcc-6dyScKzfA ztks|2CPzh8?|Q3;KfEGw%>l^fxm&m3y@ju#=${Wk)t(Q|tk?OgMz@bQfM@hX`;;M6 z7{V$``9d+iXFb)0@_wk}&rOJ-htfCURku!7-cBx5Y%Q_{tfQ>?gE;|{@Cf!~_|-`B&TJ$CtkK>OIQkNew- z8vOx6+9P!Rq}5g5Gvb%)Ep4sS>WklBH-JFM4S2Wr+Yx(S(IbXvh(11{VkdL)s2((f zX1Hz)G=Pm0y+54K!^iFJ8GB8RUJM0;$60q&t!PPkYsQs)H<#xzc^F({q>Q`j}$ zzjE#S{L};Rrxy2ez_)z~$7F#LW7}Zg$1(S%+L4`aAO?uOB zh^{FA%gR?iVk|BD2%<4q;XOF+spDQPlw)XDN^ud4FD@YvKn zToA1P;%l)4V(L-xfPi>ofq=yRUvzG6-}o#jJJoeZK;bA;N=ZtO zzidiOa%-ZjA+^^b*QV1%j7Lo(;gOkfNiNua*MdrzsR=k~sjro-VI$)>G1u5`-^DohB;s8ri%*0oa#DrwSC{4ndH z(AIiUxBKA}I%kL~-`{WT5KJwBD_%;tVpM-eDO%zex{^f(Mn+pQ0;K8Kd@KL7wo&HX zfdvBIwfO;3rRwj#9Zi}ha|3%09h^$PFy5nuq6xw=%fjWhc4PKxP{NhK-{I{(3T|Hk z>d?U0k&B%ScBkml!iKxps&R5D&UnkfH=fR)tcZ*-ouXMgX?!v#eN?ke-C;D0 z(p1b_XPZ;*?-98Cbjd5MtAvL0qTdb>_!WCB?UcJbQtOHjGu9)0Ice0|9VrGtH zox?|Tl^n#xNFaTt%?@Er3!$MROClFJk%Xl}Amz+z6V{qva(a^J(O>RW~M z0WS&@RLL(KniLVc<4szxQ0`UWZOTB7o^tKoNL@W~cSU0hj)7rEk5VgEEqqx?#|P3@ zF>W3*Zk4Pp;SNNJV3ixG&Ju>Hc+i6YFl~Do%L0}q&J*X!E~XzprP{5{d%Xb@)v|8= zTeb)~t$In5gpQ`B2m)7D<;pI83!ExUF`kuzz_C0Ot_9qU+tL!rDocEF41toC?zfldQMP);@3tDzrB0gM|0CPPLgo-i=;MKP8SpnmU= z8(ABIpjabPWUwJAsE~xq#}AYSH@b|Rvd`K5#=G{Ryi8OC3G5DHV21qMMu$IX={FIu zscU_MjRf?pEBNt(bBPFG5s0&|MFxlUn;McGPVw z;WEG<=iZ*KUU^_SlW@Fmw*B+(!1UdyLv{ZWmJAt#)JS5I9s8HoG14J){rHp;-+wQ?~QPOw!sPRm{tsubUR%HDKpR1RE`QF z%5B_LUUZ0`{8?NSlxC-;(j0)|+79k_@CUU>$tTf;vIkgg#r|&t8~%i1&;jC<51;8C zddlD+ODg=B2@*`CV|=SH308NiQc>{utA&R1o0HmJY@EickxT&~<1?mdK?Dr?UADoQ zHsuWY{!;J>N&=)+^4^oprXFFZVu{UQK=n3XdsTs{$+1Mq=_0NQ+&$o`H>1azR=tJ- zI{Zz+Wy+$G#%J5(8mltUEcz|Gitwst6Eel_5^u38q7;UwuphWn)_)-TePVTc4`L_e z&xdu59mU&?2&tKM@%8xmPI=;4(?LtbcP5ehGcJ}-Wdx)~NMK<{MEM0Byc)dZXnD9zyQmh0cND08`_I8hFj0^?e{A+JBG&vkdME`(Nqe29Q?wGG!iHK%g@DisY z9!y5bVHSA|mK5VaKk#VmerOtw+&es`diS5b)qU3?5KrqUI;f)v>%-e?Dl|_qa5S8J zPY5XmpK$0zldEKCugQBk70BpYa(@;aNT&ZP7a;hltsLhobO|6r%pgF~QP`0kV z>pAwg0cU%~C>C;B#AyMh2tGfBl4@@8lcSP@gNGZLc%zS--G98#llh`p99ZbdgfK(H zxn4-BG*~At;x16J99FISSFP-$xV3_kgyjsZLQliA%g&_2HDd#GPGpH$#(Yhc%5Sju zQw|n1i7T_%{t`gLk%zM`fnu$N!$YbuB4q4B-Bn80|uglx#)I7Lfzivud#z|)POq-*wvq}tp)rBN!E%bp@(QHATmsNV8DJ~^CpWD_1530(=83yj zS{U#ghher5EEPme3T~W26n+sCd(JV=kBRe(2Y-dyyZA)M8^v?=E-XO&=%_ttk&n0H zLM!bnB`$R|tpTb!@`%&u+>^p~xzpusm2ia|pK2xf!`GU)pFMaI8nxwPvcaa?0h=^{ z%#&~21P{PYf_sRq`>RdyOdBB%esYJQ9}Q6pqC9aU3)pFR0@z^>$@_gv)5j-dbhFPi?yFCDlN2Yu?ZK;jTFOL9`~yIJo3MSzrfB_?F4xOuxs7E;O4 z@sGpdoNiY1&ZJ5(zAH6<;nyJbFY$J4{kR;9upEf`6bgw8EfZI?zKawV{EfOkNft$5{YB8UMfT#)^T~P4&C@%FK9iDwfzmC z*ty!Ek@oJ(kt(k$a0JsFst==UC636j+)Y2h+9X;O^s_h*zy>MjdiPMZs40rjhdqYQ zlFPQCFz--(U!Zf>V=t{<*zl#NjL59S7hX=p;`>uOq_+V%uL3Udj)F^O2n%w-ij!w1 z$=P8qNPJY9TuP<7!3LOlI}oiQ)T0N~w?qnMg#vklV8Y(W;544VVBH7<_2Ly@5mV7r81NtwN_vsi zx7Y1r@nn*tRw}covvhoFYRYedAUxF%KG6Ex$rgQ{5bc~*XJGoSsVEFA-w{XoHX z+fCi2*Wb6pn)p?%>8eK!Cu$413BuzgvZcF-pEE zC*C=Wfh?=|I~_9|T*nSpL1lpXsSxiNuboqNzPnPU~3=UmfK z33BGHUVb*pK7hQ{6aG%K-vLT6?SvR|84usq7GA-3K7@TYdJ)-Of9t@1hgP<&H=;&&qYUVZw1=S^1UbS`5_*+H6}AAlbdv7yaD2lBSygNJR2_(=FEi2ah}^ zvc@gh95}Oe{et~h`rd^e^)-@RClx_lBNaiLeg~({mgj0CA{emB_4)p716{6`c6*h< z+%pv$b9>t_w`HDH(SGuaZS&9H$R1q$Bhx5ux)=Q)=n-v=q3?H1+a`Jm6-ZnXa2lD< zYDewidy*Z38zBKp!Gt~#{QaC0TZ7_yWvF;6XI&I^u@OjMc?^HZj)PvRo${uN`oqeBDWNie-b1A$-`*z(1ysCYK0NJpcuhOyNp(V&{ z>L=a+A(!nmBsObvOcIRq{X+S@v7qJ76Chcg&}u|c00X!%xVp_X0mY6YEWh({k$?SaCgyPexmeYxH0uIX ze;ow6?Q0nt?XlKEdCev(8_KAo94Fa~(YbWM<3oz^;&Y>1E3A8_$}YJ&YERt3yCjI7 zc-doX7c*^j?^wnBP=g%D+FfeI2KTmNi>o&pWyQWF7M^`8K6_Yg-QIJT8fgbUL^GP@r4#uZV-5bD%UR^V82)bG zP*mIb>J@K5YMvPSD~9R`)m$OrJU^Za_E909yLYeR(*R(0P-%1Z;O*FK-tqi}2duoT z?@w`BIaY(Gp6c2@uhD{h_hWEgr!rG9-Mw{Hay}isr#)?Q5myVF_GndlAU#*}zTJ6^ ze|_0sxRv)NzKHG-2>q~!LH#mk`>(|xw;Hvue zr#HVmsyX1945KN}z`42t|4m%>T5BIy_3NfgAjcuvCrUg#B=Ys?58nJ`afg3Rx&4N{ zzNUxHjBPiaU|SAs8Cj$%oEKCZ*ZwJRqOIxBa?JSwiI=FvkZe8ddhe&z!8}%SqD8+Q z%DB%d?_ayEP~&qchv{6OMsYCLF?hjLz_^>d6f}HyMGE8|2-! z4)&l?LI}G6(Zp246-cm96(!KS>@mk7z6^Lp-j9gEZ$kcgogDE7s_n~!)=j~-D zsF%^4_ZovFcCV_Adzc8d~Nbb2# z-w^?Yx}r<5BF6l$g><9rQ%ay|NYG+8+>GUdYl17;!M)TCTp0lOj(btcfTm&>&~MIMYHOPw!tGW~`AsMRB4e(Hq~njcoe znvF3fuk+l#o=j~gUadJdm|&&o2rZ!Uf?7`MGl%!loEA%ec6NFyp)hOcK7jT91;zt3 z=SS_vSCyu*C?9-B-A#+&nP9(r=@k94RbHUa`Ah>11uvw#m+ZGM1OTKVD63m%!fRGZ zWhrQD_rCfqdmmR+18e`vPQ6SsX)OZ|_PZrsiqcZj^lVtL{X(C}Al&N`Z(Q-HkdHFD zeOpij>T@#>1#0m#>@m8jqX?|@kVpY34JdUwCVDF5eADus3qON>7Qi#6G2~`7_(P)qcFymibk? z)yE&{)3(W9n;?^SsKf6d-STE++{=0tE0$~Dib<>gwSr%x57hDpNSVYX&+q}ljc<(4 z_=8Q)BuXIEJzHB|dakn9>=p!2LT=tsgz4i-twyBL1vZ0{hLk82BQyp@&pGTENuo{3 z8`}i-S>3_(0>$?vIVPe~O_bPewZKPZE8NqsCZk3UoWE|q>ytKd^C&>O!jD-rn^hgR z5M38Sy4E^|zf`wbAKlY?rVargS~|6yWvE>VZS*{JMrGQB#tqE_eB9O*Rn<%Ps-gAE3uB0pDp5{L7&Z-T+o$N|--&%`O%uz{+Ve3=!3(jYbp!3td=$S71#e};$0&i2 z#4j0C2!$pUS%Q~%$d3T@NP$nN!e!`4`Ef+hzd}&;1`QdQhLnc^bt-AG7|KR7|B!#n zx8kgL>Wzt2=P6yQMZT*%q`LlDFLtl~$UzyeE{);?Wm8&OM=KFN3L5fI7cWRUlT2Dm5HHwIx!uht$I99dQ zT1H=9ZQ`LCZB_*c7br~^M$l9@afB=q_`3Oh`2>kQfYu8`2nlH$gqoFWrcR6^0C}j%2^&9Uma(BKU2&U*#LVMiOaAea*YE^D_AWm<=45Al|vtgBmu0GatRe7vjSS-FKun$W!VUZ>5alocO1gf@|ksV3ed!Us`D&k z{28fK0|8^eu5>pctld~#$#LWWbKpassWRW9ydfL@pD6>5S&s*55m#6U$(9f(wCZyg zlkMynOnE#4kgVqWSIXpqbxi82Y?Ij7*bd(}N`3`0C`@C{`UV;4mTjSq10~ddmQ+y% zDn;~!t9kU^SZ zKJ2GgmzQq%oaE%oIUV?|++Oegrt}P5+srU`&D)P~dLE&W`FHJ4f3LdS|3PT!!W+p! z8X#YlRaMb$?S)V^y*??=LePAyQPaqGHT0r9`R1yBv}zGbInH=1?W?vh*LyEm^KkQc zx^-;>xc&aLWXSjr*ph%+P<0e6_jwgyf|pj>xobAw87l(jtgxyr!?_yO^&%mVHWCT$ z-{q|~#oFT8!KR5woqP9WuO%?%sA9up$8N6-ycsgQ(Nu@bE`@8bgz*QgjS|XJ$rq!G zG{bFjw-)*|*(VTB`~`#Xtzy3=njAp&B5dRVQOg(^r#D4@nL_2$z z|5GRYFNKdR*KJ}J6zTtCp?k=CWrxG*^nF8#0R=7Ho*sYIL=RbY6lgn!P1tNYx_*gE zVmz3BJbAmgz-#9wok4@P8akT$0e%0=BgiYsCt(qC%wRwdLA1Ui9v3<-)wXR$9 z^7HJEk#XhxqRnM^!wwYS4*BDVe88zq;nu~NE^>mTXPV9P(XLu5eN?HxBG>m=oUA6I z?nerIk#&rVt!NvUA`~qeq?_EnzAjW4>rJxMvt;(ie=utAppD9+S0Y6qERa$>&ZhwsP8;YJoXNKG&{CSD zgbnVgPn%fg>R&jCL4Rk;xZZ0G)T(=`j2dx<{rxu?7EB`*FK?lgl4_N7lFC~*N$rv=+HHk1n20$GZD@ z^NUPMgyT(>4jBhPvGEaDRPB>EkPt zZbaF*3gHCXm{XEEX#)sD0d0XOn?vx+ExhD5I^uE{uRA-&i6&IVi#eC13IujRJ)gvd zPRjcUc7sz+j{S#@@{bg#cMBHS6Iz(dOT5Utb<@H&;r#(H;MlZ_K>-=&Q!hm%VI*c< zO2^)0{D>w}ys!N}9OjRT{Z?;65m?orl{yIV?O;NI=vF2@Gilm2#J!DSQkFuvEcI?n z$8GfjR&Du-%a#R_^Ajvq2}agT2=!F6(ad zgBBa%7|pI|f6cZqMv+XyAgR$bbGY`U!POr~P4d|he`>)X)yuO6K?q0W-j}8QIF+-UL%6A82b7y~F zf9`SQ`;WIXCHtqcK%l!j{>0uy!E($ZRTOV3;J0Bn4@0U>=ZyA$y8DlA4 zSOFAps-OC>U`k+c>`@~EQ7wdQ$Zl#1EH zZ#FlCCO{jD$wU>VT=p#dK_m9!#23Qd8TVYb$;zUzx=z zAmYobru-*|Y>O(e(Uw~@aLNuyOe-ylbQ;1`;Pv=$S>rfn@@XPq>p;GF1P8n^f%7de zAJLzdWi?>4r3@wUM+b~v0S8QWoVy^Pr>H9h#uS&kwX|-L{s@7_)(QgqDzp_@?(=Sr zL`M?${}BQW7g+wM5GWs~5aUi6aCqk*#!3Bld4&iK8de@*Z;)Rz$N`QUA|q8?E|43a~tMHDPcU~4n~5X zzT~*qwjz>2nng~%13sQ_=ZIg*bIT>ydX^iOtFH9%U3C-L>$mKqj3i7m$0y zb$wqg#oY`9)TM*OqRYU0!}5qGGPu`sgQyCK^lKQa^a{V0{N-m}B^WpA7#Mi6@Ay=WoB$&82#Ud`9Q8vg8Mb2Hq8J^Oim+1F71Cvf z;qR=md#T@da-)F~?@G~ou&xi37}#v+=%61xbse2^k%WtinV$^Jwg~}sX_0hV(xir= zV-PG3z)9g)$fwjrur36gR?xXGt&(d_G-e52B8@_VkvcwO=CAMP5&s_T;Vwc|d|ib7 zk|4|2(3a+|gSBApJ-+!w*s*ax&^qvH-)+sPO$!D9Jj| zZ|P(ghkKO0`m_!T$R#lFjM13T0fLy{4H_5$Op1y{u?Qgo7s8~5HSsaj!y6u=>TE8O zT<%2Er_jffrSp`MFl(NHA9vI*3Obgo)LTC_KI})5bU`}n>0X+u0IUX5wu*s%YFJPu zz`CC9WY;G@z-(ptRrhN5)LKp0nyVa9kq3&MVQenM*+wg%EqF@`Sw++g85P~$rmWTQ zrYgkwR`iy})qt}6EQ*q`aBm31=(&t&WN%UrvY(y`G&~wK#+^NP_G^)lA>OYpF!9Vl z8YBl8a3!YDYbb8o$KmhIjLP+0$v7~48h!Y#y&PZvpuX=5|);|xj+LD(X? z6&aW(q-!XmK9naQAFmO#C`fjz486QW5U%9umC&Cdmzy>i5W|qM(pMs4u|f&EY>2G* z2NVipdAYNvt!$~%`dh(PA47P zwr$($*mh3-J;plg>^;^#S2gOUZmLFI%;$aP6cO8D{_~Y}*nrg306;Cr7^eBr-zEWz zs+sjhbXn{t!JD4_5BHYQ65!PFx)`3IAwPpDit#+Zn# zLL>;|5?RHM7zRM^ zp0I84=?~1~T*KtIg#QukAXGj-%TFxXjEm$8Y;~HjEvE`N@C(H*3*Sq8PA?}sEzbuQ z;mQQJE1A%MB9=Jz*;Rpn>ie8iV5BT`9 z-Wyqv15RxCz->|@_J=mZ1<3pJ@dl?Oain_5Kq2{4HnD@4ztAii zAj4}0Z{r;h6arcce+J2tPpK*h5C>r=;y7#~(WOwBkO^kX+4BJ&)r#k;{wkZxTgaKN zF=~vJ_4@4fOFUDTftLAa*hdW)H0VOcyZ*&V0O`MS3!Ls4c!~&iWKrz&2-(V&CPf^; z$8?R5PFqE^SVNipN1C3h^SOf8%Wuu&PG}66u;kGk)%-Oi)lX)XB-+`Hp~3!fSTAz1;!ctqltxKQqQFqGD+6i zUEjIeqH~pq%4GpSHf(Sfh*p`liM!w{6%^^UaM(J}UN54hH++snH>$cdx5H~;7b^1RMejbYLVsfyo{0b&}33A!2z zdvnrqeD^p;+z|s|;z9gPM^sqdS8xE->IY^s;FU{|II0@w<#0;rt`}&Vdkp3AbS{yc zxtQ?gH3^838slu`6n(gcCQ#7@cX$cJV#|gYr;W3DTDs6K8S?HPjjoRisWbOskB@CE zYOFHP*`t7}A((a!-YO36+x0EL#C0r7ZRv(QRw=bd8eAXUxa8$Xw^6uTc9$8ZsK=>q zX#SwQ&zq&shd(ft+Gwz;^j43q=746w;h0J}od;OnyRip>ZRySn3yhX4x$rQWvnZ7M z-I2Q~WVbSm{b9}#Xrb(GxlEg}d;Z5$TMc2JH=QyWb|eaZMDT#1qvoCO$;D=PriKZI zHcY%>{(x-{TPN%~`WPi%Ucmb;r+6`sEg|bc*BieKee>qNb%l2=jlO#O57FoZY0TRt z#TAf~=h~R-gd1;a29ySM3kW=}Hy$A@hq>(6NonjfvmnIRvouuL-n;lzDuUP7ZSRHf zBdd$Z+z5jtG3=j%0!wUExWhPig5VL^inn5WNULWJEDTTWGJw z(TuNEcPfoPyC$6z&trF9*5zARncmI1y`tV)3%;>(@`{(>T&{JP0*$TLs(T!p8UPqY zTzo`9v#A#b5wgWgd#a~7_qX$+^kMWy^MGLOm*rlYjPH&fSW^mrfRn=fo<%({;(Px8M#hqI%S_ zBa9cnECuo)wdGJsOY=qnBrACmFF<{oUB0?|{+L#zF0ikvwf3~On#66V-sWr@M4>-l zL)P2q&HAowKR0m0uZpppS0bpc-l<~1LnePZ7jrNYi0mTdvOCSGMLbOMJRXI|VRw+C z^Knv)VGV>fGU;s{i2gul zaA(w%ALz{Y1z&MzArSyROghDd45I@d4|Vh)k@*P;tUjSbl*LZb^KD$Hc~z4b$cRTN z7#^|oN{@22s#M93{t2(aa1iL>cN@?$%@*j<+YGPCM9s$g%QC(rX^lKIjp`O7V8XTQ zo5>`D#wU90>7y+g@M(1JDxQ7S1Om8?J4e|^I!7@+f9w!#j-~X|sYBAC1ia_i2`|B* z_Qbp3a6ruG5$fNT@g~8lc~ArQ`C+z)9EMQU7;ie&eD~j*YaR`tRSX;%D`b^22Nc5f zvb_#$hms2(x(3(D0Ow)rVS;)l#H-)8#m*n`^IgzcPPDEAXo{IepP^Vj!tUE0#hh^W z#dr0vTNHLeC17dLS%wc_RJS>;I9_CNK|h&>lrC$sQkYxq(@V-hYMOvGG`$I} z%|41{U^bt2>ehHzdgA2i_P?*-k>~4Yx?bY#0PzehaE^2#Rn)hW?4M=}Nyi%;lMKZ& z+u-kZBnq(0!6lRbFy~00*{1(u&O<-U84!;IDbbbAg=Vj`{%5@Sti84B-wPzq)QH9N z4(BflYe9SVk+fxFy!$GH3S_3x=HztW8Me)TlJ@@_bLRbF&XfOP&iV#m6U1t`*2(rD zIJIojf|$M($GCVfRqdp+b_#1V$3|<#?2m;G$9-@k0oBmI#Ge7mH&4xxm!TIo0HliL zn5%trJDwo^3qJKXxCIXU&w_N0-a{8T9n&+t%;wXRYpZnu`=;-P;;F=Mak&ke$?vq{ z%a)?Cob}*Z?=?ye=9gsfEr_y@6_J~{P7voduEkQ3nqGHwD|zi#ZDCKXGg#f0^6Map zIGPrQ>LbjJr*i<(9I4^>8-GON^q>B-x}NQYlwls}!J3|r5o!LJ9KVit<*(GZJ)gNB zB3<{th;$O{T#e6B&@}NC%)oz$bbMLT9S-50**TA$gRW3MM#q~uHT|aSm?%Ga)YsDw zknrdM5*|A#H?%7{#{u@94G)#c>KOirTLCmB~oY`7U(Mw8xa4| zuaLC-x3)fx&o`bLG}H;MkG>4ja&@P|Rg;Ej!n>L;;<>zQ$C~SnZ6qI^61!9j6Dc+z z>~@WsoXp^_$^O*hhTb80E0p3smu0l0=-yxwULp(Y-G$>5S>e*#L|mf6iqtVsR2!d^ z-x~lBdCh-_H0-^--Vc%fGX$L*e8(l~DE?I+Z+#zR{n}&(Ydf)01>$7{QL>|+Z~oZ7 zR{!ZDaK}#j>EpiL&0lBPZJS>BX@2QF<>jhFDFD7SJSZQ73;x9z?U_rR6XL;CgL|1< z48Xkj!jrEtyVJXQX4UM7tH_Bw!iS8s`H2|p-+!4l;`Z;b$m?9f;{zKN2m=qW=&0;e zXMWph_#EP}2TZ+gn0)<*NXISb(hI)$LK;BjBF@m?8ZIT3$f0KPzFGFITm|AnsI5}i z{hvhI;)h5J_nGoTEd0MjnwjiMGn}AiA@9*ZCKfQOPVIwLB+as-0v8UFt=I|8q< z#ac0uTRw&Ui%73IEO`iS_iQ4w2SGbI+Q)`Ks}QO0pfUpR=Jsen>=S=PKs}J4Hv%8i%@3rT{Q<8Od7$E#IKFar&T`J9J}av$^IW`gmNkhX|Q z3ss-4zTt|@{_K3b)xm;dQ>{hm3}Q@itty8`sLn3_uCXgaQUl(}PWR-KL1tZNv3Dd$ z8zfDh$?&Oam>Wf$-sza7m=4tB^hF_B(;kBArPblkBR=DM_i@0N#DNvU2jd`Iq$#DJXZ;$>B8|~D-c}|Qv<%W9toSAE-Jk2 zI@gXeI3o+=d09z|vMofnn`I)ty5b5fPE~P+0S0f0!00J>6j6_LdkekDrN2PG{KN;u z*(Akjy*r(=I1;c(`9$T@0N$?^Hxl&wLgdO8Kx95%i_m$l+6-wQeKZ%O6tVnnKyb|C&?-v)oS zQoyIh<_Tt=fh6nJ{3nqLc!-U3(mh|8patLL1@nz z!1)YT7b>^mNHjaVV5+s{v>wQ8j-`JnGY5vQ!@~)_o!LE<0@-!Y%{C>fnLNwhLle>L zCR4C{-o;=h7aB`Flu0GqR%bq`$2Crb57$pI4}$$?&-q?)2%Ym!)>e32YP@b`S9or~ z$!ufs*X6-z!rRbT&~kej_+^g)YJ2@RzzBT3TRYvBLyMhSpCRF$#gP24kTXGWzkEg4X^p=B!rhf05rVKG9_GUt_-)8)Wi+-xm%CI9!fxPWINMS|4Q}U z5d_w^X1XAO-?BrJvz~P1W~mT>KRR+}#AM_g+ssFO{%YE_;IbB1ki2}lS-YK6dbfI^ zvzFZ3t3@Xe6zVi~Z`k4e+#ajxx<{>fzuSK_yS_Nyx#OP2d(Dswq$rAi0z5xYM0L#D zT-Pw_dtylle7GEzmUOLA-SDJmLEGdti<~kJ_ucD~B+De#m+qWrXZP@YT5R&hCTXbv z0v*iQTTS!YG%zUvFQ29%Kz#l)>GJ*$JQkRKf-neO2$N~};+QJH^b%5hUfYM<^Mq?tOZfnaWgrktH zOx92}8JrMQK5`vY@Ws&Hf%E(!$s5;XS1Wl2oril;A>e=7kY-?wIir>|S79H!ok+9l)*glI^ zmVq8bo}2F9f&o$rw=R3zTiY7{cLkR4e{xrUtXF3<3uiMI7b^$*|BC?IcMTc!f&m1y zuLcAp{y#k==hx&cCC}JWg#!;qF%4rY$ zA}jzN!+j8Q=pxI8IkLNz0zlOZdU_pVuwhbxk zaumdqnWy8ts3e_mq^7&0UC0ZVJ7kL(cT1oObH=!g%dkpBs6X6rAW1q|qwCaPH=dKqR`(IG<3TVq zY0OZwGma5iGG_$<0inU1C}DpmjK&w*G$d0+=zgmRv!q5sk zoJ8ff^squQVY;9Lfd*}^;P>No!Nz1=sg3vaeq-@}0R__SBT%K7hXJHDoTUPayd4FN%%I)`2NMeoo2r`YQzID&fN{@cL6`j z8#}K6zo3O+c3=ET$;H{j-N(txLx^6d87Jn*#rs53L8eOlU)|akOwNmuiiS*;r-h+s zjf1lAq z*}Rmz?Bb({G*%H2O1ard6s^q@MN>o&jtC6#NJrInf>b=Ga_hf+tgawM#y5l+1;1tc zw#E+8s8XhH*=XTP;)jog2>cl%a))E_fJJUW<$6xVYYw-M5_x8 zj3#{2Hh=E5<>sp9uYfW%g2_lU;XwD}#ApmrZPY}0X%}bn5j8n4%!n)|5!;k^qwf@j z289JX2AH0%r>(M;|>$exLjsY%`p@3QS z!$Ze%MH&t8bjzqV+*JLG|H~P!YJ9i5(N7FW+SkA*Zk6tYoeu($T?qpwwS>vUfr8Y> zwSX@Nt`NwO@m$Xtk|ZpqE5j+fNBjYBByi#YHqYbARZ4VDn9An(+on$5wfHiiw{Q1M`Oc)b%tp#R#uv{ zCvhw!q^>agZV_EyfIxNU%VYgzYP^GhSC9~8Z{5k{LqITKLgtkf!E|a_CMpZCtrWoq z>w3fO8Vxf3bwXZ*=@ zRBl3^sW(90Qu)@``t5Uh;s!!)w+hw>rcHIhQ#3AIX+?8Z1v$ZE>^VGy!2FZrh+weCC6rl{^7G8w&3_Ng$O_XQr za5??55WtZP*NB9)W^;6!7*-MTgtFi;qVsy#4g;hHEu1Ia*2}g8qRrC;#6|Fc$Lpe15 zqSHxSLs2e`4|yqM=EUEo5UqBO5x}_#D2*U&%I=VTI z*+*X?&=Ejl+CGCDO(Z2QXM9p%3YbopXe7oe+T&39c~Y-IidIOX36Iy2^c$C%3!G`k zlAI-%?j!wJovFwWt%vLr$YA48*0W@@sWNqwP9~%Fp2=81&l-4B>ub)T1HiUgnfJad zp-P2ofZxP-zrH}$ORuM|7MP*06m*B_#ZXl>Lqf<+m}(@H<5o`Ua(dsBY3Yhgly%W~ zG~tBZqtV)42gSUr$e)Tszz%Xr?#W`fmuY3YXX6*pn4A8g<4-n_&ol;RG76+-!d8DS zzXY!EMCsQuj5x8h{78LaEZD<*@H8Lry0KxwfEY@T$1pj%Ks*-0N8u1O_4YgPyv5#N z6|4Yp#G!ET<#kh4s121|ckvbA_%P62cjQXwPLr4{VI*I_=7&rUJp_l`e~t=I_uk34 zsuyo;#>-A)1~;~EmOG$^#g<2%bBO60UT08@9aLi)i8d){6ILjqvj$mYT*$T@Rk+-l z0B?{mx|#bc?Z557&&FSeK1>2-mJq^)RbCxU=AmuUdJH^Q6Lt1m zV)S9_nz(OCZ%CFHO5vx5QZQz=8W4Xhx|o&;n$p(J zo1{|p-_!MJ#*{OlqqJqI;R`7B`m#J>0LX$=$FzEQaebd(7)gRcp)1UeM3qXN<>=Ma zGeQNyB`jdPL0{v4)+`**-zym&*(Ehue;dBPq!g6l7ZgQ$Mr?q~1Ta^Ku+^1GSj}d^0Vu9VYU2@oH*P?^%){El zW$P5Z;)Yk_Yfr~^ro&9MpuNP32U^CA1J`i{lAOn=b`eC|VtiHE6@67mA`3(( z&IBTB>9i6Sfk{#O$-y8h!^mH2ILJpDjW*cfuMaM)IO!{i32t^;^2T*nl>jSQ>JV|z z^aWj0dFwW8xX)Z6*@G|CPB0jjaSln3Di?h6-4_sHRSL+9D~9~`>y;p zGt(~Nudg#zf0J+dt&oW*NXR*S9$2iC^DvZ6^kHAKG!SP`7(ULGWZ|)?aOnWEcv)6i zQM1>2Z`+6k;0}3_j8n+aY?I7jL$=jLpzEYHZpb~m^OYb>P^uD0gDExVktr)%`tOB z;D~@7@GanC00C$YB^RHK5pwD=7e-#*RQa5D14D)mi2xNNYJXw94|L+S9Wy9Qu>>kF z2G^@wC?N*AmwKfdK;Y;|;}<$Y?Hd!D;wL7XZ#IFA*#2ByR1xh#zy792yA^mdqj!LN zN|g-i9z3WR*AB!5PW)J>6P2Mdk|$3HV{ck3^#I`)^zC4?TSq9y6Enw^*d&5ToB(47 zsMU^J`3YX*Jv2iHcqeb3@*}A-JXCPQwC{~iz?0~G5f^R%nNwjfWM@j{19p&`zR;s~ z_a7ct`?culYHPEURR*BP)M(4O2|c17cfy*l5aG6EMcBq=tuYcY=0(lK3zY?HSeKKE z&!o4>`|WJgBT4Jq$1@X&Fb7__@0?57!raCW!%FWj!P!MjoqrG<`LRBfoC~veV@<>2 zNkR|T*=o`N6jlh9=|TB=BcFeE!yn(ARcyKhthxq{bzm19k)z?u}vfHLnoy`G2;_%u{iw{`*I)ad`gwqeb8~B0Ir>7hba!p;2 zk%p0v*1}hdE5KGdeq<6RXXb(Pno`~ukR)Ri;Xcs-(`$zs62702lC+0#YFdR}`Sr4l zheajTR>#O7_mN5aSCi>d7~JogRjtRA;H5KBy-?+YR}8duZ4cYq(<#(t^2*E;wvoIr za)2IZ{(gZK!uv8kTnd#Tbn|1hbo{=10BdCtwy$#WXlSh(+26B7@yy^gv`*12QMJsF zj=oNS7BESH<<8@FAIh1uRj`ggEpJ};XlO*A{QRDLN#Xevv^YHoMkfI_inZM?~q=1rTM9KJuRDtx)M08Rn zQjG350?WX)2(~ugP+vY=DB0h|7#fUj2r!#9)#{H<^C)gby2hQK_w*(L+fU{F4<~Dap<93CmPV3cE2xXs zLo|_o`SlR#jykfhl40hOEX5qsbmccOhQ)3)un(5D?gSPPDyqW>`84_Jw?g^a`_*lipzN0?yoJRsI>}Ycspdi zP{WA#3@i0P&-|BCaTh{K(ohT$p~+wVjwc>M&wHibSxUFK$YOfx_E^*C_;9=hiDiVA zjAEj|v3Q2s%coNWe>AI=rFB560zCnh6@Ap8j9^8&+m2N)KW&5MdhwxdxTkCfWrgcs zKt&QcUqXaEfpdW=`X69uFHGXeZb-SYqbvGN(T(m8e}}G`j-R@y{0QqNQyFr=|gAw*Nb_IyWlinte(*AZ)LGx;p6aMWe=-1p#kD+`7gqcDcJ8F*q zT+w<{j7F?5?x{^GJE-e~+?F}OgU|gf9qaG)W(R}4)=5zHLahX+wGH&h4K3OwM}XK& zGR!|d_6IGTrt?!DrVFuW6VmD{rV;+;Hr_)?@ z)G=4J(LMb`ePi1Ka;u>n<{oQleM`v`g!bhH9x!nTRRw1vYVTU~Cy{DEGpVbc!LGOq z8$#+aAs8a^L&Tn%`J@3RQmu%=(<@`sL(;M8xJkIR``h9NlX^gzizhzCsCC|geEp?# zihVOl-5h2_!=!J5r3;ADec+X3nIg;}tb(5op$}1V&iG&iI6dyT4~)$b4w*33?(&Wa zYnrEByE;HDDoz`KS*8l0x8#luSee^VP2xqT(C=0>QU(Xy;eS3p&30<$?WGBoCaMx*{51I{6$6DJlbWI=ERtLPC80*MhjZ*_K{kH{&Z?#=B)w}S?Vjn3H*9ep8gYN@{< z$GB~)3oLk#*bAZyxyH;1`33FGqi2+|fKAW|43XrLDPB2DLGncSMn<%I!#m4K8xaP_ z&4d%2oVL$SNP8iMn68QjOqcraRBm1SjQPScek^E+H^x$Bk)mcSY0vi=G96kg3zTj{ zT^~u*M~l;{tOx--Tp>$*_0wy|Q?J0YfI}w9+?$6(Eez*mh0u3P00{+tT)E8RXN!Nc z$abOpgyQEx##v@RvL?p{&hHF;ELtzGqX6$Ts~cOGAGA9=0AQnUONYi81K(HVGxVPO zhk?1*wXw{G7Rrm4{kxKi28W)GGJp8-xg2X43+4B6y=%Y>o=)Z3_4kl__VtDP-bh+p zgBQjt0WXZ`WtVX*JlqRh=|laUQe4G8FUqmSNQpy@f?%vq2@=R7TnPO=K6d*=SMBEG zzcGW3#;t~Y2=Csw?k>LTEAN>#xq<&^R_K!=l# zqEK1spa%OcpONtQ{Rqp59y&ww?h=HLu5}RBab<%XLQ$_5_b73$c@3#bh&Se#gjf=T z74Dt|@*V@I)nl7?Rba~4Qlo~l*LCk;{ZD&Hn??XlTF}x}G2;0m;xS*@qcXxE{$Kd2 z9#a7*OUu||JE`7@0Qh17H#5e zW@_bX;B4h$^Iu^!c(j$)VDx|2F~a<%k>p|l0*Hg@JiY6(rA9stu^Y9D+i5E46%=;MdOZP)K zBIkC74?DF&ckgss1Vcl~{qNPo%a%wAOJdPrMb5Vo!D%H8#dviz!`|K?1%br+1MF@oXw1%19Peli;vf<{m3> zz--v}y$&^najbYGJ4q&to^b*LGe=zwZ50V}iLr7i6@wQ?9!x-hgz^|bnfdSJ`FSPP zu~T_t@z7l*Se*$S`*ajZhk7W**_3KIjBT)5VW30|`6bA1YJEh5bo7`mhO=^cP~hH@ z^>A?D_uKI|jGo+)h&ALIo@j&d5@;<^%NaB{j~VR|XP(+A0jw~DBdk(s$f{zLRo1wx zh!9E@WPWUrpHtiteWeQ^`Y-;mI z$DiaNPXd;Z-uZnT$dE5wYN9S6=#26QUQz|Xtk^I^ndbUFT1^TActjc7fw|Q0!RXdS zVg|q4P93mD#WFn6Vm##lA7T}56_AG>6lNnE0nq-k`M_{^xdewOCk_^5Wp|?xhJ|I zU+=f=Gvo;1!r!NUix^%YT1&Ikjngoz_in;pVSTCVagN>sEI7Z6-d}o3Q-F+7Q_|Mq z{Ti9kKm|R!{}|z)cgg1oI|}L02W74SzuQofoIRb|>D+@{3lj-QFH_WdE;p<<|MPO) z*VsPF7|=>o4rwnf@vwg8{$z7$Z<&l8-zESf`k9~dWXZ?Xg=DkOi&b|^2osngwZl6a0Rd1e-|(0& zs>_ifmyT6OLfF^8K#W?m;4jj>UuY%*X{Nw>cexf>7pJ;s-dG;dLbiHnuolW2a;c)O zn^78R7xvR<41QJY(ff$f8CIZ+Wx=deZ)gAwko>)c(c?p#KzOf3OxkqiYb)*jiQAd; z7GLtcd>!3W;`h_=DcBIN!XInVx5-k4VY2)WCZa#d=X z%lj2uS&sdkr`f;72uGSP6aIXoRPkNUqRkf1_bL(pOl8nRyiI_~=X zzw>#$w#XTJbp&da;zT@a_;J@qs{BezpZiK?=c*kv!`l}H+6d3F7?8$%n2H$^V~ANFkpAri0p2!m|RHNr2&(HCM`kB#nqb*__f0KoYD84 zPyS+KI23LqeA^vEt?=tBlLq0_MAIkR9O;6cd?=cqd~H;?E`!IGE{#Q=Yv|yzY}YfS zUVE4{_$na1Oj~FT43{FKGF5cUM!f*RSABq{fhUcfTukm^bS@p65GqwAeP*YNg9HfJ z4s*9esS^#=`6iCh!!#cTK*H_+G$O^(R^Kyn%GR;jH>OPvGcfGz%uY*mgKSF?;Y8*R z@D%xv%;)nX^P$4MW#1)nElTbS`;)pt7)Nit1H4`DUg~_76c!sV3`Uf=1yL04aJ~H( zR$QR(J;(@*aXx)w07JHP&sWbOpWnNCU(fE)o>WdnQx6k=EffH5)7HH}=R}Pg^Y6E; zJmiih+-#w8z-lGKQY*keot|5J#t7R3nuA43T>MP5p3+f05!mrcom1_<9)sezrnfW*~3}tsfMO z7}V>aif^`xV44FgavbgMP8DYeaB-o`1o{F`=qgnw*q{c`$=^WYpMV(AldYSRWqA4S zdsp^jwz0jvMR^z;YK=p4ry<4P^6HeC4sk?R>e;JI5%uy#3teO9|05U)?Y9oep0PG| zkOxjA9tk1I{*kIXcM+$V-XGX`!O6+vN$-r?plS=oI2EF56Zi zZp|c*F-wQ@@?P^1Kc%8OWoX9QuUiS=b z+9GpFPJGb4V%3dA@MDpbBkDo5Ly`r$cQIf4^)QYelu-1>Vb+hP)l;jw>-(ij3j#nB|u7w@A20z!mFWf-)yh?#w( zMj1TnhFzM1^Npkfpm4vYN|ynoQR9I+41EHuos1W12FUTe)xkj@%;-7_@ZrpO znpNfWfEEImZI-!0W;4nXSO4MlPPJds*AH-2;_b5NwUw`-1^#k}{~~M>p2dWOEIzbx zWf>M}gX_pUz`MXpQaPsOhls!dl6W(2)P=JgO*p=~jJLnP_#h0IOKR1rLGrEvD>Oq7 zeC-1|l^HMqFaG;k0n}tu8JuQlB!UMhUZ;cXf@Q%r28Ks_Ka<4I2g|5&k8%Uq-v;ng z`G;~C4nh`924S9sZS8D8U1o9mK^5`RHHZqVhR-sJ-e=xdEIM9JCA zCfj4c3N+HnZQe>uwgf$zhD{ry<)r@b%VMq2tTjTe$79z;{U?`aOQz^Eph}koYmL0+ zBqG-^#A#dl^DNCmQ;?Gh3EO78wQn|tL2B4@>#i^c{ZHCiQwEnzO38HP-4->#6K5}` z8%#$Z*$;9yMst%M24&V9>UtBQczkfj~H$4dmDml)VzrcxAc z%@f7xyu9W)At)x(gDoS>E?N{@?ZAFfE;Vru^!;9cC<@`{qF1P6X<-()-QUk%X%Wd$o+f0LDee?>=@wo9$Q^NaedP5Q0{Z z@;zPNN1xDzUT4u{x`N;jx}WpJMj9tkKW>WQlqP#%g#UcnQ1p@2C)Tn$U=UFiZVzS2 zfFFMT`4O&7B&j$(onDUl5MiQgZBu60lP@I&dK!(@$!k-Hqk)SJg8r-_JWL}lHk0~XSzy2la2!%7;xvT@Ry&6{3FZDkNU z;3H24k+$u%%b(U?rCQVNcavhEvM4E}@Y(h|TZzxQ?KR7$)v6p=zn2^=if!bNNTsle ztd28@X!!d5b3*B8EYqW8W<~)4!P`D|4t9b0KKm?!P?r~QT6!+TOZ^Gu=5cjz-7J_2 z%Ba?GG(F_3qwb;6_|)ALTOFHxpDRUF&h`bpZizKbz>;a3@MGB#%9Yk8L+P|AzT7Ej z0&t;d_(yc;%nX%9kh~YWWcG(&S9z2g^{Wy&Y*{0VXiD6*QAHg~Jk-&^%7C{d$8g!Z z5v&B>m8}3z%I^h_pn}QuRBDapditiX_P`q^-h%a;rF%YTn0obW;-_vpXP zWi^O!O}Ry;{dMA(lEb7+Ws{JH{MbgQ=ja?vmj@IuLdKN~WArV;Pn-&1PoO zq3QrHjU5mhbX>>@(;|~kyZ9LK6LSbP=&``gy%ih}0jjqsnVST+Dh%CI_M~mlE!*3u zcGU@<<2eZt{bCHEf`v;OhQL+sbxLMgmx~0Qu#3Y%T~8eOMC(oOpkQD1tV9*p5t0|r zOep(n+G~V^)biYEY6FeERAQT2wM9LQQ%k@~u|Ptc;&4*E0$wZ&TJRN!FUC z&QmesfeRw8`7uL|IcQr+n1r(vyWPp>ozx)ut^928E$ICQ-FLj^C_-+8$#2wdZv~L9 zTcE$46p22?{_=T&4YCd-j>3sS2{S4x*- zL+6I87T%taVH~y4$;$dd;M;!>^`^=2=e&>a5GGJrL5$etEGJprQgiUxN7rR1KTpsHt6UU+qD262tb0? zjnh3*Nr^SS)3|A|awD{l*rYrqwAmC4qOIzQ{|*SaH`X;El()xXVKH8Ofr2_(6HyWaV>@!1 z*2rkGIXYT4efi7N2@gjyV+a6LrvAF5*HK5?WsBi=17xuh$7k;1fvM}NnyIUv_UVe=0*Z*%QPYqFepV#0F|Bs&36c7?K`?DeIL z@(t@YEC~qqLc1Ff;fKpb_)8szKay4r=6W7(jyqIx5t4~ug{op9zlW%zCJ-F5ay;ZJ zo?@N4c7MR5_Nwcrd}}+6{ff%+mOQjIF#i5TvLiIhaU>w=2`Vaeh3C3Mq}kD8{|l(W z+Iyhb_dEW5>B0?M8S)Bn7U-G`iCFqCt_x9QXMqzNk8%%*>X!?>vOV)XHHh26mAPZ-++pth8)xZ? zyzUYU`b1MhLk5XZ zJu+Q_YV>d2ay@JSS3g5g&%hngXNSnoq4lkE`67dzP(yxhak#2bD_wT`&Nd)1{B@Ou zbUA4sy&TS-cHo_9gRxx>CC({d*Zxrmj;B74l6=>ab_+D?&o&s`&4y#?DN7$xf5`oK zZ{xu#(ib-XJVSTg!fsXAEwyA!`)$gZ!xUPU6AE9x#di?kGl_`+^`00L!TtZC>zsl# z0irWHwr$(CZEMH2c6O|PY}>YN+qP}no2^tHl2lUFuU-Az_kQ0w4|EFhkd(R5monPV zWc%>jcOLEasls72b;P?GQ*OhM0~43I8B#G3{6l~KfiV-4I$UFX@qTfiX)y3NBEmwuLXzj2{<)7lv? z5diPKKX0I?+N()@)3$uh>3_Zi`s#SfH^v@e;}OpD=-*ln?h`+|n`6z>0-4DWCGZ7( z+kUjiNq+6yrJRb<^`wIh>>#2%(;LD&Mul)cn_jS6mPvYA)>z;f=-%8V_e|nnbtIZL zk}Yn05m$~M&z$OCG8i2nsq3~-kK@FhYXi1Lp=cK_!?cO7oBd!-V;Ene;9i1|i@s-V zNcRn5kVOf4t#>ka?8SXKf3Z-z8#&!wqS2P<=WVDy|C1&6q{Ji8ofMvC9WU$YoK;h2 zuEgeetA^l$%N{5)VP6|kTiOZq#>J$3VoLJg6V0_(;RDvadB86hSLu=8EHEFkz5pOx zzT;Oo5xglO)%BjQr?x%iQijb&9N7rStyJ>j9&wL9d&-=%VaE`mMrZegk=^X)o)jD_ zJA_j5QrG?E8xHXmF)p|wual{E@hzJ`ZU;}nj&8L$6K5^rw`x4Ow&4^Ug6T`zUt<~c zYc0P@9xD?r)uV}^9}WB7uL=|_|J5B=*b_FkGS4TkGMox|syEBS^$_=SiqZT{NAPUQ!>hKsj23OM ztGbrbdKcRsrs`sQ#?|et`13>GJl#F-#JtgJV0re_S71^Wp7t^>lxQ2s{8Z%PfeBmNF0HQ((0P99C(IqFPaI5QWcz4}eI&FdVHlkmGKTC1?=(xhc%FBi8@DBE7P?D%_%6|2Rw0ZmxR_kmB%oYyy`nDG}YmVKakWf}X zGwXkXF#?BFk6zz!f9Q@mJc)&Pr5$^BymX)T*Js8b&T3~Jw;kg|-tb+pF|j-CJ%ryf zxvD_DV)4+%F_V@;0($oQENla}PE}NQ+Hf*Q%sS$8$=i}}akypSTfJN_g}767XNOUx z8Ay|N?n`l|Sg=)O-&ZPV92>ZY{UFQs7pF(UAG&AHV7>}%T=_ZDa&?IjN^)@N*Um|~ z%L$rmnvfRSe3wm3UJ~qo)w1fNp+5Q(>d(3q)h8V@7g&kG0Fa{daB;IM1k?>*8!nC1 z5Qb7`(p&=)Z#xfP&us_$KM~vsb<)?Dl*tzkQ7VTUP3@e_Q|cD zzZ`C9guzUDlXNkvdollc3Yl~P67Q+ohLxx>F|dHFSn```7JkAGdCxY&^pM5oD;pUr2GS{#T zxCr))Gp)(3xxTJP!L4pbS{?sMzkXr8XXrbt*fnz~18n)xeMUA&Y*lH85KJC9o7BJt zK;v)KIQy}Jzz6X?`rj^13r_?|*ipGLrYka0h7vjb%hlf@B&zu8gjuJGSKqEg_3TK} zIa!I0y+~w0w~IT>S2xTh`HCb@vy6X`W3$YMwCnb4P;eyXUR6z3LnB`j4tGZ^$A+29 z2BkG{0@Nb$vQ_k9M?v~u-rBs@)&#BbZaMe=Eu{O`cdQPNy6%78s@h+FuwWHA5o6p8 z2nnuQoj6+-l4vZen&%Vd*_~o<*sx}*^H0@%Som?c)dVcjgd6H|s683z3^~OEi6tZP zTkKP7GCOOc>}Ce9Sqw}{2(c5|FjHmSEXI!)02Up}ZTXZ&7QNI9`=n2l-zU?bi-p^N zfd8VN-XesRc?$9di{2zBYpTtjtzB(uE%lGDGI$P4tC{Agi0F?BU%qT0oU34_wR5hU zkv0+dSlJoBml1)9M`Cc7mLL|J?WvqZYE1N;at2p z1Mr0!F*Ev;tMW2iwp)2MOHReU)q~0W4uYOBj06W@7Wh1497bsnIZna%Vne1DMu#5Z z<|Yt7p|SSVkd$j*sLR}IYt;?uW)`W`iLG-YVoHL))A)5rRJr2^#17`=vIJ5r!bD{T zQZWsuh?;RDuG*bEH5pEZih%(=VxXq|0G3DuiqjAVB+?s`Pz z9p5n;oea*BiQ?`_z9ycstL-D%gcxJ2&m1B!f@5w>9pW|GI+~;JfBsy|Te>7Dq1L3> z2=sUf7%kZ$#v%(1l-*ErWEr^Dv69YTg;vcC0mt^Z`axG=t{NH?H83vwNZ!9T0nW0B z33rg@n7vMIddR%>kvxWyT_TM{O|^H{%gx!DMEb+`cXHHq+_0LJsvn!2mRnC;x+C~# z+C!Jiq~;GionqlnNu=7tJ7kx|2sWWH+skB;tflMZ8tfYpYI5CwsJ+2OvzB%8xxzV# zf_OPAQ5lY>5V`BuRXqM4d+%=V0vz&2cKFev4;vpB5xuHagtEhgY zT2ro5Mw4D1%nB08@~fkQhj!`|<#H(wi{ zTM_B_n`-SJ;-p^}SP_hQc5I-BeH~^(c$oBA8+uq~nnEmsh_Naro00vLR#ZiC=|H3* z{0U>0SD`Uw4GH%O28_10>#(vNPeWltXr_uM5!RR3H=%uc85v@uoKG-UK<@0% z;Lg1rQeveMPA+_|H=}6-Qt%C@ZJEp^{_LYk`B6TICcI;AB817DXCy)>1wqa-tPmrs z&z8z8bfuLkA`Wf0GNL?Kt0A$jW^^g{f{mAY_ z54t7q1i42h*(=Zt70fyV_v0sdan%^T*_%<(*(B~gswBiP@fgp2C%G>$ws2k=yZ4-= z^})ynMdCZw=E8L`228Nk>JL!5DZe-{Bp4$iUU0ve9s9wSRd!!6{7F+IpK77YYV!CKdc11^D;T`>{o|y}4ogh&?7sI}}mulfy$P zoEXt}N!hz+t@4=BW$y)iT5fw|Bj5;Cj5x3_s69VlbZ@(ffQ#F{g7rZXXhYCIW?sP} z$jN%D`^!o*#1w{D7n8!L{U_oBNP%fJ~*=83VSwd8PMnC#)^`4xS@qSiz! zFIL_84gU@kw=OR7fwU}&&m>XO~fCrqC8tRv=CPkhRq+194 zqmqQY#Zgz5>Cx4ndGCo2Z9LS`33GM{=ZtP7D0XRY07DbohZc;uBj0oGerTN!P8QaN$$IpdN3gheFVmqaItJPw-=lsS3>u;m+<%nhFrsy zO2^y{4D&L zh=;ydepI}2dOzwl!8Gf0FXHp;u|(`E;1wA=sCQR6S<}2b6eEqPq-8wmpS}n&Wjiz2 z^ZikB?=Z}ma@H2Ff5eTchl!P92Gs$p(4BAhrXj)!FJ-pm%ZIDh9hGakCLVMh>;)2Ek>Q;_~P}gY`1*#n1 z264wujGGiomdt_S8|C9S*Zl@DUrrreH!i+eevx{-H;xaL@yW8JdwWg3|6G^} z9NG0q*k3ZQ`0)?BiRfjaMQf2$&MyrNNFF*XRb=5M%sGCP<%VWySRW z@q)%c>#Isc&U%xW_jAhl%e+7>aZ!URDD^f+Tn##RNB8*>-a+fW4%3hK4rFGU>TOVQ zxwWyv*+$P^EX-0rRrPR5UPN<9;|iguanisw+`eJ4!#Y(p-V8&KmB}^mLqKW+K*VHE z&l*c%zfEKWFVG-EOMYd?uN)0~} z9-RM|QE3FtdX^SQ{M-9#^8*)^w)zo*M3!d~P&b#TM@s3s zrq~8nFZbb~q<8Nz7DI)Ds%XF}jUw4e_>yzj(ujJ|=+;T;P#*CYZQu+4JX&z1{4b;H zS-)KIM7Lu*>SBs!t%z)&xWfp^E|6S6Q=u34&VHt`88jt z?o|x7_U<#-wQmbVlv1ZNKp#lJ<3YPHZBic3!_IR>V7GO214#Asi?+o3voEwP@_fCd zcOj&_G5izc$=|&`TF_VDplh0Q8{ctb9s9y?)x|SNg0)u;P?XFQZJ(RBHcSADjvS}3 zPWtf*wQDSUBpz%h6HIrXM+GjcXLjbeMTb;K^%8vShP8={j6#+Mu#!p($KT5ov#O&L z3;unr6FQua=1wGPI<18HkURkmJFba!K-ACPs}l&ky75CFBYtDhpwWY~Mikt1RB%3V z%r-=sIfv(&6ZC^{jdlVBNk|OlId&)2pcHns@eq7EN+JU?0LjuAJ~^TpjF4LpGB8@I zAhxE^Vip!M(3*DxU{Q0D056>ZmJfFn z4|+|9xhEzN3;{KCnqnHt-v;#pQ+85mgL_+&Mh@v*Dzq)FXVm@{PCQ*VrJ_@*IQ{X0 zeMK3#Uw~LOmPh}TC+u;cAqzr4-Ncu4SN`gfCYtVrA!vo8pVhaAP9 z{j2zL^rnv}E}a6jeW8%QSMH}_Hwj6@$PH&y5k4QtIhyL!oL!xfXZ8a9Lr2_4Mz_+y zHgJ=qWqxT5Le<2hv@QDd_sHESgX^P#D_MI+zHJe}onXm74q zX?IvLvX7l;;oxHAm(UlojL(U(ru|FwALoR_^`DMaS@<;ud=ams;NJz z;6ofR4GTAV>5Bna;f4ov#u+4aooCpsy0! zCjob-@Rou1d`N~teNq)Q*XqrsDK@y1bJK15zLU+`gA%YI_YYwDX$%3W!!SlvB3OjR zGC$J6z970cI6?M{1Zpk}^tk4=n)LOc$!dV33xA^!uU!s)IR{E~b-b6=`ook$<~C-$ zWpVKiNsw<++n23a`PQ{6@B0D~Sesa{kk!T6W1YH%igSzh6v4>k{L)lO=Z6itYZGHH z(hO0vR0nF#$mWHZ+a|QyY%2e0%@W7soO2%jv-(q9hb(=Rjb-SECL3v=WsX71GIBrw zy?Bu*?o-bF2Ar%$6Xd30_6bb|S1*3rlvA;(%g#zN7FXJn*27SqexBYqtEKZgaYQlU z3auSyJ#8>Jo9hTyIT;XkYi12JhiQ5YNmrd`a7&f;)vBd^r;`p%=6YY3O97AhqQ-IO z(*$*cB;|zDcxT>*gKj6}jZ`*+`xZbHH^?NH;0c-x0AunG<^CxQP?}UFWH$i%=RKjo+<81#!o%l7#krL8J=yGvs%h^O9PM=LE*N# zW>CDn*;DM!&W%SV9f23}(N|y889zfKgS5b!jS5eX>tBQboFQYY;3kCCa5j7*Wgpfn zQ_4w<#>}hOz16tIaOh=`t*-M+DB?UlQ_K{p@Ls04z}^OX`!S%gI?GlR?>$O4qU+PR z6ZBz{cJF1Ih>orfTT&)&Y7ju5sAF#gQga-{kwbA371(?V!Mf^8Farqm-8wq+xY>9L;B zyhw70FJEvnJSG&5pCI;^IJY#V0wbH|5T)zMW!KUrWdg+H|I&Q+PZ61w zE@fg_l9}PuUgMa{3InO*x=0oMMx;L7YJEO9?7+$K2N95nlY>cIJDBMDhi2x^h{rQ= zmlmq0)b+vQ6`n5%y;BDRH6KBJv!X#x;ItE?(g4oV=$K__C6j1`(7=WmhE}A_mwe7l&k#_6g4)Mpv{G>EKyYILLov?oJYg%5BJHjES zfZc?&*|JbV7XtE%aIOnGjq7%G8g1B_qJpi_jM6VN%!G*-yev464XtL1cJ;NF%2Ed% zdc1$8Q5q9Pr5dmlltZ{_=gqjC%RDh?0p2&M4(E3e+7))CGxWwWWMj5l2fx{Z%#9UQ z>i;&n`YSSE8KdKp{LYot|;IoVWsWlmqb99UF3%d=2sUq1_XGRJf_J z1>!I&YU+z6yICKE);k)f>e#$KWGJY*AmmSVGM}ODu#a2&lF&NaP`_QC8ucwhlfgs8 z-Wj8Pc%NYk9S;+Buv~0i-y9hMJ`oxss068hNX`a}=|ir@ls&nWX@>S-{jAJ~NKFgGX8JTr6r!L|Q)h)H$ov&=4E z71T**MfMp+Z3F@9+UF(d$p?NMSXEuzZ&q!XzTnb-74$Q$#RG|$gwc~^ZNLxVtxaC+ z=;f~XSlvJGGJp(tc9dku)>W@Dx>WxGNh4?Bv+bYu@5g!JhWFkQC%-5I*q&&QJn?xy z&klmOSKZgQvbH~bhSOdYOjTYtuLu~8QAw@{LY946gb0)X6J_CmB!tjP%32PI?Q&jc$2F(h1BtTF>@vZAwQJq+T82=TQGOn zd*Z$Tsf-VLbdkyS6G!N)wGx}dbH1Od=Kk_G5`bG*4WJ!N=cG?OwZhJ|=Sj+EnJTD* z)YtnP%f3x;00P%?f!bnQN1^Uh$+CR4pm^r&~_Dd+z=@Ibt4QQGJgNi!TVc8S8ppiM6iDqb?<;5aM@ynA{r&NVTa>K_=fy^kPO$1oJ z$02h5am=e>8cl3rK3v9`Yfn-xwN&0%Ky_?5*gKzDl^#Ia=yOL>gkDO=3Cpo=cq~hK z&IO^&<8sEhNb|qrbk|f`Jl$jG={W=4T_cvvf08uBQ<&lI0fM(9?{f3&xU=4ZT(?+P zx$Q@YYl@E9HIO|`{MJZho?zD^y9Th^DtT3`C;4z@N-+6VL&SbgE8XCg2UBhDtB`M6 zH*c6bC;JF3C0}7SYm;cYZdlzcW6$B+9rRGYszgDYsUFA~8Z9)a{suYn$DwFDF__Ac zN7-&4P@R<;z6k(=Yx}&E9yRtrfH3u4U0J=pemQyhP|RfP*fe*0#9Xv?_yT-Qo~%R~ zaBy(8aNg$Y|1H-*C3Hi?{gjM8^r?USIjSvlFxr_iGlrj|8*IjUtcyoFJOm$IYF}pB#1+nFc$NMSghFFm_Sqs*aq&y`wXY?-Gkv)+=Sx+0V(`n_i}}2=0FAi8>^`mEnFlThWwYWj4s~0 z)k-!Q5YT(){|9>d&uUHTO`-@I;J=>!-))*}{&lBKiI&^<2!i$#YwQH=Yj-x~5;SR7 zHB#A*1U$u>tAY%%#4BVdCQCn*L>YAfWBn7<8;32$@zkwGXA@!|6@4^rSCmW&fBjis zh1{E;bJ+Sh^&SU|mf>SekuNa7yEJc36dE5u4{PFV1J)rR7@`}uLd%mSV-_UElw{#V8%|X5z1B|FnE3NWzQ9LutPfRRn-z)HiS5Bf zbA$&j=sYC;Vq)x2Lt`j929kt;>OT9a`Bg;pm~=bblnpsMTq+9TCFEdR^N|mi@mA1Y-||B4 zUFD$MZf11kctmSN8_tD>pA*8Av7?62tIqy+JFkOjR^1XbnnXZs4V5B9gd(J&$0r-> z7#g2T$;2*;1`euZpOQrbjMCQ!RpK5X1UZ3aYl}7lWD76vGC&36pk|8X>4RPd)#0x} z7AxUQ_5u=!J1G?!j=b$3Wy-063?HJMc+Ljo-~036V3cqpBy}jk;a6b9X4Pj1yj4Mr z)<>O%)bm?A+zJUfh-phoY(kQ4sGt-S+xd4K)MZLpfT60c(4}h?=y7b+9q=)LD>uIj zg08%V14zfulv2+4H>=VTG!S$MUa*NDz3V&Uay0k>Un~+{3NmLO`de+O;esd-u`qh? z;?X^%M$!5^v~niGUNei(j~|%mpT@kqXQfdJ!PJtqYD!>IH6mUiReKetn|wG^17?eI zeML9+zWLm~B{sC6U#Q4y%IIeR*pp3hkTYq}!$Ko?Wc%M>YJv&K2ZfM(Qt?(aa{v<)|U6YyDbYTqb>$IPAQM6hE{d(0s)uMsfs+sih(LCIt{nYOc=qPw`#( ztMGU^2yG~+0vuqD;yiXxWeQ-KQD@k2wn;+}_4)CA&#_j47olRH7)}9rq*p{c`Ho3o z{FBvCyzqsko^kB!K&w%V1%?Cj-W>ETiLj()v+!xStu+wPS{P zU!EgQ+1P`+&K-M3Oj#IUvtDKXqHyW7u$X}K&H?t}*kyTOJb>FVXT{`SQF$=As!ECa z-{PFI;`w_X5t?VTFWCiA$wHFyUT%bV6w7;^2j?$MaE&wSP z#G{HT*`rg~g-T$*Jxj4%lT?5;xklK|9Cv4 z6kXZkYI+IKa#2qX%r93O_39z5edoc2Qo5))r+|?eEjp=>d~aa7s-Eyyy~hN@D2;A_X=dGnF&PS`o5_oM5CNdOz&i8w{;tS9L!)=%^hfT8H zDQ6hJXo7;DxQZG|jMZ;>7r!1kG^iGK;;?f>@ka{>HDC}aFkVuv(@RQD*(tl?dgHeL z2spQ5q+KkQIDC+}LtXs&A4*bgW`W*9C?%5*{6kn2Ve=}a2eAg42j{u(e^Z+zXJ1Cs*a=*NjU9-$5l@XXn_JM}^d6JSSF>t2Hkvx@;(U&eabI zMDC}x=Z|T8njZkXJOvSndR_J=(GRA~fpz>^B^saTVWb9s&euOhoGxj+L6q$zrg91z zhbY9V=14Sp)N#SkX9&dvbjqG{H?@8@zKGB!4c)HEW#bJL#0h|M5MzYaDuuqwsnt2C zlZ*~5Bm?2_($jgv!lUrZgwSD=Sp$;}T@ju|-G=M~F@zX$Mv2GGK#1u|LW{GC8cY=y z1a{UCnMaWHSIy4jq(EzX*dz4}ClOR`WCh2P_6{b%mA!A+UnRQYrQJv_71oV@8td+! zKz1q0SZuTnw*cg$Kb?gIsup`*S@DwMOPY8A`m;@mv57nuaposAu%0zJHC?o!dps{% zf*EEXbUioHBo)bHi@}tOatxU<7Gjg0DE5p8qL7`%dk{T-Qqg@yjWmYfM4i!#N01U9 z2IC_%7DwfCEAPQ@#bu<&2@Y8;G1rI?o#{qZ{>ka3o&n`VhnS!w6*gv&ap@CKNoq`_ z;h-e;GUcXbzvFm@G}6qB;s_Ly@DW*#+P!1yvii38`At!PK-6FN$xrKhxDhWsdK)XI zCMF^q&h&n*V|4Z{Axmw7i9(`P1jSns`Vu9OC;~B<;N}V8_bb^V=^#iZc-&=p62=Vp zef_~CTm;-@lf?=Wa|fZ=nvyC4`3au|`>sLO774p{)oIxm2_u|kl7LWU<%<+`-pxt; zZNrA)DxRl9d14+(;9HV~FvVH(r@^6xu`cT)3wBp>14-@zK17n$YDNa}wfi?MFw0CS zhv7eQJz*NLTlamX0aFtDWuw)B93+9S;)|C&c$0U1(1m;;N4(NWD>nSUF95nw(G zSv=>G#FQMlOO@0$V`9LSuL{@UzJpO&h%iD~?PHb)6u}veYP3yyp*||2=>&c%`rE$bQ}4 zKmV^L*I8#9hpIPoXf~HOXQI?^%g4V$&=&dw1__J?B1WJJwuu6LI4=52Co}`Ck{xMV z7EUWXiV4j2hf|)tkp?Q8DjnbB$UOCuAOv82Dt54SlM~k7H zrL4_bEt$f1v1<41WYYZy|$trw++FwYuycFH_Y0 zC%!XdZxY9A%JQK6V|e0%NNHZrR`%00z2&$zwSNF71&3Qezo=};bBTnfo))YJH^C`F zi_`jh+FK!EZ9FYtZs=YTMva1tUSJbfhY4v)`(lqu;AZhsSG94fj~0vK!!JMF5*fJjopy}5Rh zC*>J-7XB!!kx1!j9A<3`CfzpL5z#0|ml@P1O%F z6*>g54Lrx@4)Lf(mTx{zLj=x(>cYAq%)~e@r z(5aEW3{urJXZ#T!l~HxU@3MXhA>%UUC5Z#iM6Ut~qy0-|@et4;%lA!~_Y0ttlO;cA6Y`hxHng`!{(!xmE@w zqe84JBZ7toPEB+rzXBicU~~STR6SSq(YY8IgMhY9sY7cHy%;>kR^YCWQGcy8LxTyuT>lq^H_=91Fm_@>GqndaGQ ze0PFLDvg~a*L7G)R8D1(vRcV(q_75|U7_VNE-8cNZ@t9JzekX6;%`9Ch%TWz!QcB>_Nug+qXl&b{k<4*1EIcwIZSpE`Ov z{I6d+Z;5Y8Wf;mK^^HZ>GDH$QiOSn5tkCe6Co%xzrgtcBy1VMEOmb#E%tMyaC1l2w z7@w&b>`E&8hmWT~gf-(niE>Bi!AcPv=DPM0Y#>A&_8}Kzop(5z&_N60k}`|>w2X=2 zDOmqxH6~_t*_0unqP$X& zbRNluzQ}exDRH#G{=_e+)5{FhpCmk7U={l*@g=tA*a7(tp^h%cNavqxe!}?s4EK^oV?PjPuyfWx5P~iRj((f*`er zrIKT)k1dd;_?51ZkJMT(LskBPPkBz@LYn#V>)2xXJ=X=+z-sLr=}`)8xuHSJf* zrgTh48!WZv1nO~*Su*?#xfpS&fwtK)D`3PnG^z5L8Dmq#br+!85%De^Gk*ZO8Pdx{ z`qV363`R2y-yC)DN&RVi>wyl?=CjXrMU$mNL6A0=G%S8%m^4p=VGoha(qS6Jr;?I4 ziE^xaEt6=;-MSB++yyjhAM|C`G2A32u8&J7swsC;>xukImlBQ?sInL{55q~_1Qr*> z&293uCh8PPB|sq3j||ifW9k4>fAIsB%`{7TND|X2R;`W#H=vQqkhTJK;kORm3xCc5 zbDdDqhN)oecdGvPRK!~!!K;TJ-3uSI3qMvzDmk-O;>|05=p@fbTNCMw1_B-RKhyFu(Yhs?25`Hg zaAG_1HIJ?n_kw60o~4HqfKD|0)Hdf;CGQ8qn(#J6dy$D>BJhFNf|0c@* zOLEca#TVi{tBkjgb@x+r^5ty!7k{~$X%3|NyM`6FbaLi=eSHA4`5o2oD{99lOOVIb zDQ>~{K#WXJRW!$~F@IASTX8SBk)@0;W1jP=GfM zF9Qp;uXyrRvwJLEZMZQ(ruccKl+I}_I${q#ZggvY3FMd=Op>g9It|Jx=U|7= z)KLzF>A<1wGHo|0s9wdnsHQ@&y!MymPa}4{%*8~p|C$= z?6H=4=#?|GfG+$p>V2=fN#4nw%71;x2T4XQuS-j|T#W_T%<5$zupcKBtmvR&jMb;{ zt`sBT<~bKTTTK92rQip$g?Gl)DtXxpcO1eOO0|QUG;HY5|29t@U`A(cJ;fS)eT^#R zHD#LNp3AQ=wy&fkm!~|pH}z(dle1;rb|$bbZo!92d}7L(to#xPK&(S0-n3IgcGT~{ zL99)=mRJIOkcKT#7{g$mkV;rVcu(sm&~lYT9YpW8 zvAKoeWLi@X+?wFAkT=YU7w))GDVc{CIP&q%>QaL2y`OwBYQwDz~jZDn#RTu?_f`HytP1_7PWB1+aiHB^R0Ylm=vh= zQHF6tb4jLeW~EId`eDHr%uN0gY#4-{)|bz{0H+ATxWmGMC0TZp3e*5ZYh8F$bS!m+!iz zQDIv+Fe`GLn9C)?@Ga5Zt|z%?I)4{Lv;7Iz{h*k_K&-qD4(Jco&TJJ6xT#rlhnT(t zj$OY|=dqnT0X|N^{_`2k9i32ch??|KKH~>~3EsK2`g1DEXxpmi-NaQ-LhbZ(40kU^ zyh!j53GKhjM1MX0)ztJrY~1WBo)YS>0T#}9ykCI$K`6>I!#c)+IcNSb_d(x&oMRjY zqHCo6Yd(!WvhzJlTn3*$18g)Nf}3Awkq*DwqytZ z&Z7>bcu2#5bZ_o}B))1(62M=!oTg$`z;yoDbVb8W9)4H#sEj>@yq`!$gQ)j*BW@0( zqhO6L>E}K2?1QhH38E|Yc}g6ogF_#HD13a*+gfK1dD?3Zk_3A&h+sEPa zu|F8{^syeDv}*5cq80_tv_m~mO-ZV6RO4ccMzhTYOg;-2dqlO z*jtz5ef16Ei|Nt+2J_w5ASoF@Dr9l1UW3|}_Lo(4v;HqGdNKEX@Xu+PrDK84-x ze!)2@RXwjGo1dPukBVvI1}KkSl!dB%t>kb{s-DD$QoBFM0^Syam2eB8r&9^jtt%T} zV{i65+Eb*;U3TZM5;RR28h3Md{ji&#I`Lt8brvA$-3fnxIcDog7iCgCj>H`KQ5Ld0O4sGtHv0&@ z3&tIylhW6#a(OjGHbP&8(%u90q{78tOcPj_lY4qo#SR6T?upU>T)gX_pdW}3->{d< zdP3hm!L40A-KSskn@%qR9Gl5ku8C5vlv|Ty&>%n9@mzbi-Oo;~E&UBTf5RDVS5f9* ziE@otwx<0)l%4K%QPTA(^=y3JJxrd~d9sy%AVmbxg8~Y+ecE5kiJ$LJLzc8gLTcwM z`33m|vg&EpZX`DVnYdRyQ+Dl^VYY)J(I5{QGTM=KyC9ilVdk$f%C1iXd3X8g&mgm4*qLf+T|O71E#@=)JHC?&UIK&QKan;d~xVZJx^uv&=^9n!CC z(Bj8BpXvgj67zReagHu78(`4cZJSVK+}O@kWtZh>#b}%xIuw0s4?77X+n44m$ozSS zRGim&N%1RY=n{S}gJ^Pe3m4V;h?h1k3?1UoN*dH~*pmSU)8SPNXW;bpVkR}!{R8>= zOAGNHN)vRzv-4Lt3S;P#d4lbElW3+$$8$Q$(62RVxv2KLTt;~;Bno$AA;kl57CTfg zJH_ZtcL9`WB+fRVXxqI}3SZm((L-xqKob4AFs&Xpb^JysfR5cOq6Q)?11);R$8yE5p!Atjfug(!KIrwgq2*7w%k59Y;KTYz>)?a&-zbNvoP=6$yOh#CK-wFy zq3$@-p-FXwg6zwMZ#zLO5n??w*yJqRo`?1T;}liScir~L@0vHXHsC&32V!DqtUF~~ z?_AZJ4ZKr~mLUG9!5w6g)iZK;bM$Tc@K;z*GM(>(X5Pi(Mbw%QVgk=di`;^sQq`z3 z>bm6&(6~$Q;Mh@-T|MQoQM+MxESpp)IPD`=@!~d`ZCZD;*^LfrI-Qvr_CAb7+MCkx{jO^qj zJTU?E54TIBikWDLLyB%%kIh1}Sg-jMl;Doq#6T9cHfSjg&$RXN&JMi310*#nsr>W3 zfHv823+eGHY@^{AnO5HSg#I9~%da-T%%5=VmTa(9;4!E(t-#%o$<~(9M3BoV;Mr}U zS!3Phh|nbFMw#?EXyi5Wnydg%1a_Apm~2ueny}vZouBq2;?b66yH!O#@VffKtfIN~ zVy}g5u?I9@S6d{A>8~!C(BP90&|^m~_9u0m1cVbf>fPI=x)X$R1gs;Q#f3${Qsbt+ zm9A8VID5ZlKx5FAg2CDWO>g4dxGIDQ(_dc1xC3(M_e{OU5o;1ax@Bz>=AEDyHP6$_ zLfL?F>>T^0f}0-=bUKbw+er%UGnR+aweg*Vr0$(9?+Ij;9)o};@K{2NdTnI%FzdQ{ zdoJVL19-v+>l8|2Pg$4Z0qB}RI3|OH6Ufe+(qAqAHr|ZLb*BrjHSe2?!(hADW}X#Q zGu#wl{a>s41EB0%E?Vlv6jtTZX5|AY_YO#m8K|@Vd08jp*kS<8>m&NWniq!JIcQ<- z#G8^QYzoN{>?Ij65V;t+=jn`W>Z;W9IlRAt1m#Q$w^`n~fLiL4T?~t97*>J9b(o|TAHUmZt;jB6u zabo!-SZ3_4`gj5E*X6&HW=Za>)p^ESD^iPr-vl?kT_QWVMCR$wfc-&BF6^^;DLchC zVOtRc1f&G(k^=Fq2Tpk>JfrTq>ak z#&T9|6K0e~pP35&jNOO)E*o=mwCtM8x~Wc^4E!@|u)7Ka@MrnzH#*b1sf6wC4EmhR2Zzk*c`AHTbRl%4Q(_ zg2Il%Bq9RD9hY}y-Zu2cW_a;PPIu$&mxN`#Rg;M*#;)6g6%u9}b(um4Wa0~g0Z;GV ziwv#8=CM!;d_dr@ci$wwS%a(w&9}HeO`CiHe6Ae^1kT@-&Ky9nV2<8fvnEEW~%IjA&D}hdB#IlWqG}b3KoH!LrGHNr$ z+Ufv8nwWD*LRszpeo1nauXU|(U)Z)vz$5qauC?#{NPmfKz&XFAfam7HLF^ztQXHgV zV8rh(3K_tcgu;>j1O~mJtl$$d1W>#Rzx*1x@RFRM#t2*3E{*~~L$Obo zVC82OX?mcQ44vz|oPfC9qHWU{hH`@_6Mb?ZKtNqQ8vWplzv5Z`2VZv`Rpr~Ld7PH+ z4p9&kq`N`7K{}-o3F!tmp>%_E3kuRLp@4LEhje$BFwgya&zw2u%sb;+-_P3qt?UKt z`+lzT3nw*i#i1`~DSrl@%a`<>m+I|$7`-pC6Ny#Hx_2%4!ZZ1Bv*!41)2`{0$6xg- zhF^2S0`o9Ok0#a&W#wj=sG!gNKC;-)INxRti59u_qR-ZxwnLG6cYhwjB5NWC=zWBG z$a;cb)zP`Zk7%GXqax1YWuIw2zY$>N^P=>q7ZR8k6M!#(Axk70l1^B$Y2p?8Gow;UgOUiUHZ^aDJhRG9=_eD{u zI)0#BEpG&Tsi~Hxuz*fp$p_{{<+9~=kX;Jd$ZSFfuh{$A(v*CRLyNxoVxD$v*^c!g z$Dl?Lb1VK)bH}vMg4w5WqU=3(YdH$~W!mq7No)K=Ep{YNh4I^dU_`WmT4We$sN zZD#Ex8L`oDat@-F*pb`Ihej+moQj%A)(`h95AfjZ6(JlP94Z_*1$8_p#gE`WhVWWz z3W4+A9ijTK&neDw8X86<{V&IWsurpa z>IYOkR0C8aR1?%ss9#XcP%TicP`{ztpxU82pgN(tpt_-Ypn9SDp!%T(pa!9apoXDF zphls_pvIvlpeCWFpr)Z_pk|@wpyr_#pcbM2KrKNnL#;rqLajlqLv289LTy29L+wEA zLhV8ALmfaJLLEUJL!CgKLY+aKLtQ{!LR~>!L)}2#Lft|Ag}R4&2ISfE55e7=yE(9l!zL1aJYk0XzU+03U!K zAOH{q2myovA^=f<7~mB^93TOZ1V{m-0WtttfE+*`pa4(=C;^lKDgaf$Yk(R+9iRcw z1ZV-Y0XhI(fF3{}U;r=#ya5;ii~%M9Q-B%39AE*k1iS@U0p0Ym0e%4L0S$mgKoj66;1{47&;n=$ z{06iE+5sJaPCyr+8_)yj1@r;>0Rwm8-Pu~7GN8&1K0)Z0rmk0fJ49$;23ZMI0c*m&H)#IOTZQ2 z8gK)+1>6Dt0`371@X*^I0`vkNfB--QAOVm8D1b)*Q~(+P9e@GA1YiNM0XP6$03HAz zKmZ^F5CMn*BmhzX8Gsx>0iXm>0UiUW0Z#x=0W<(wz%u|HfFAH1@B+X9U zKpvm~Py{FelmRLLRlsY28bBSO0nh|!0ki=+09}9{Kp$WLFa*2-7y*m{CIC}_8NeK1 z0k8zT1y}*z0jvQw09$|^;61<|-~ez0I02jiE&x}68^9gl0q_KP0lWb|0AIid01V&< z@CO6{0s$WZp8!FCU_b~U6c7do2SflO0a1WxKnx%j5C@0{Bmh1G5&=noWIzfa6_5r< z2V?*;0a<`-Kn@@m@CA?u$On7{d;=5!3IX2%MSx;J37`~E1}Fzq04f1hfNDSupcYUE z_yMQ~Gyoa_O@N<(Uw~#n3!oM78_))52Xp{B0bPJ@Ko6i7&C3l*asW{ z4gp7iW55aE6mSMO2V4Ly0at))zzyIQa0mDcxCcBSKyQDD&o)vME_h@(T8r!xO!l~!6Ti+!@>RYBj3Lnf3bbX^i8N%XgAEZ-qtfCVT~qD zGF}o1fs{)uRPgh6q4xWkslNC7l?@TqP$p{pMg;wGo`D!=y(@%)! z=f@PrcxJW}=yvKdo;`axE=e4#V;(Ed!mL`M4!bD~i$^`B!9y0qe((SPUw_G3|G&mx zVE^~}OX&Y&{RJNOKi6L*{#k#Kf#luuKSpj_x>E5;sacBd}%sO#^;?O6gs#E4H|7>^O| zHbw8Xx2!ox)CBuCNK=W59>*XzYO1_AV$iQ>3N4nu)IsCO)(;V6C9Z8A><9NZ{lBoMM?<_4Q8dt<})u zJYV37D}>wF;@;TjiJc)ZnU{{fh0$mJ8kYPDZM@9=YvU!kkZa6R_g@gN{IiDA}Z!KcX3#LFFN3G#6XQ#Lg5BKugqybVLW z&YIadUGC%K_9bI+Vrrd}`X5 z5*BAw9(eZU3JKMRPIw_d-!LPY@=QZlAz> zpZ2ZxWq8)hvbf*BQLu2=ZekUu zi$24d{MUt-dT8P0=*_qCC)3#_k6Q=!TK?9k1pYlXijLtva_AhcaK?zcui>htoF0=s*tvYuzh@n-FTM22>1#f?OdCC4V$MQ zQs48*oIA_1RKAR>XV-1Aa+VHOHDcf+d{Hi+(RsGSh_gyOxhX!_%h25~IIQvU#6;{=R2Q!Rvji<)Af5 zD#Kf5b}Jm*nwbx;86b`W+dU<>rFn@bJ3>4D)T@hUmv?5a!ugCq(ELou^buzvt}_$T z+;G|`^9QdRlqeaMFg_f~BlGFjiObI;5$}*Oz7+jflZ~MKm~~BU_uE@RA#Ss(jpq^j zm5Q#T#ft9YK+t|<$@*7!*?kMlL?%s>d5>K4);E}nrok)P_7mmN*RbN4SUqd@M|geuW{5QO+}hc=Q7Xt==L`Z5Q*{Ba}3(=Hl9Baf2d7&U^Yw zTar*#SHOuEz5C^!POo0-;-^HNQIAbq7g%B7l`X-V;1*7Y9FlV!9U2WRtc@Y`-Ha)s zW99%K!xkRpz@+eZT;fz8mdsLw%;~K(%S&7tmF^#Swrkv^=%2KLxId+8hRV^9PLP_& z8)%)u86b3xqQ3f-FOlz@sY|QyF-p_3MC{27N=aN{8z!Y^h^&&vw(*GB8tsxkri6h`(cWtcbyJxUe&}F~q@RI9;%4mXvlh|#4Om96zsF}?2^CKkCt_~y(Bg1^G>XO>iYi&dW53@2KA=Kcmt4U@of#(!Z}_eS5U%kSD&RB z%nUgS46wMD!>Y7Gf0tLJ8w{Yo@IlH#v!o1N$jTgkcs&2`diwb)Qs9TYUOb-jSdW-L zs2Tn7P5#t@SR-RtwUpp-b;r!p+;e^&f8q@p0HA?}a;?O?eW%FTH=? z!;TytWH1o$5*>Ko-zC*2c~Q_Y@z6U=;!9_*;d|1}^UAj^R%4r&tMs42{kMYGR(k zz4%XHKc%zz)m>kFu*wI%iuT)&2_@lO3)xJ^IE;={PkX=E5z{ciS4Z;mc~otG#lenh zcg8hp;e=^xbS!^oXA39r$-y#wvg{{4r-z3T?QRgQbVZZY`NrsYCtOIOUSmc6GrLGt zlDyVlRrlO_Eh;KXe--Lo=F{kHM-8Oi7teVDM#wg=6K62;FYyuEgIt`b0NxiBw#` zKl_#5N4)gEYI5DFTRP-T8S1XVBq?-cQ4xi?)8UJ!p(uF$r2)J1AJ#MZJy4^`$e;81 zr#aXEGQr}PZDUkLlz`?_`cK1R{DQ=K6hsuuVe+PHgt0~k_>oE0k7SIJ$;GxwDL%Ya z+eJ<2kb5&-&iCA5Qj~QDjJQFm z8>qc0y5v|>hbAan$!uMiyb*cmEFBzXpSWHQhCQBP@@64HsQ5I3?KSv)bS61yqV*HZHil;oIvXs(aOFa!Ezgv!YwUtTvk!g ze{7vl(EWRM%xPjL6?gi)uXtzz)p!$T$nkMDOLoD3IdtTPmZP6dD80!fl`%*o;4#OB zk&TasesbY7{^Q&VIo7*NZJi}VtXUihZZKJ&G`b^O~VKsYp%=J%#$-4Kqcaw=65d|tLhoq4uu=5Um z94q}fj7sUB{aW@>wWF`32owSvUK5noVDnpa%@N`svYF>b2D;#(eGDt1^h3TtAoDd#AvEZ za|Nl7Xe>92;SU`<5NCS(8v3I=xq-1$v6T?LoehVERgfp4VU-d1rV1o_hWXP7?`n=$ z?0@kSAJyNr`uBW}aWh)JZl2onHWNRXlzNQSjy6@`d-L9gwgEnJh%Uj#7ntv`nu=P=9`XNaaE!&o)kR6 zv7dGc-p^C48oR#3u(<}AS^A9Etsy&UebR1v-)e*l=FJXLbsL}oJO(G}WkA7#ma90EG z58*#5V{0i!PRvV&Cn~TirbH|CXhwxRPoYt|>z>wEL#?Dg*?wXkZbQc)ZO$lM?X14# zQ=1$N`WgdTaV1w>%A}9IS899&8r=$iFd-|VvzjY+1x^o)ixtR42xI7Z(6x2 zK_f`SH`^sfhuslmHe(@CL9;4C&{KrP3BAU(<#syHryEqOKj_if&03->XeviW(FZR* zo-clZsvg)d6vsPwB~X36XHzp{vp>XxeW%)+t1*)5?isuOQsxCGSLmrDO1|wDZ*?l= zBS96iS*)!`7IuW%&RJ0mOt3jc_>>hZIEIL&5&NOlT~5JO4)O@$yh)Rfkh)tbeK8oRVUvcOnxv5Oe1(Npx><)G34;*g0l= z8djC+%iUPJzv2=5R!Udqw?>;G%dGV5$;_MSre~)fd_%ZfRN*&1m^d&ntD^CLW>uQO ztjZ%v83ClXY^2pqpS&40X4qTG7n_TW_Wx8!EqFn*Duz{dxijB7)AZphd|$8e_H?J% zGS+tWYM_^#l>S1lW#qw0?2+%fxqVNjFg$?y6+acxx2-GHZ4p{kiC)Ut)nmTOn^N?` zajNQgxdWqS+D( zPK`e`pSKSVzL0U|S#eC`%=?5dLJJS6rz)eSLpVb4v~Ft$!uQ!Qai zS2V0*uA1Md+`@2i z?SzQyb!D4YnpEr!d1?33T>Y12C3VmB1gVs~U?WX|9VB(7pLcEiLzJHL_Cy4@&dsCO zq%u?%j`wtRI8F(O@znffG`F$8e!i~nDT4(+ZWU?8{U>2g%-#3^{-zTxyypGmw~s=o zxl@;oI^=q$b~hr!Grk6Y76^{DA9EFmfAw`lk4zncBc_Ya9*{v;*P)==gUUb2e zo$_tke3`9z>s-Z7BS*E^gcSewNOV#@??RfE0$rATF!QEyTL{5XP-_()XP`U zoXUAYG~5w?w>i$HfW_1xGc>2NK|EJvPU!-J_EdgYJr69jQEk0Dv?j5(pm?7ealRH842mx_+)b;2C9i4~c%1Z|Ug&a+E|hTBCO_m5Z@ zOIk{GQwfK9j*n7*`iX6{7}Cc?oyNd(k2q?W%ndHoTQXp9Y#(s*(j!F6S&(n_uN#QD zEr{dcGOsvVU@mL)S4IdVDm3gG8o_S$D7X?^NsZR4idQqs9n(`ku=PpUtfc8#9{N1O zif5TGZTx^i^gRuY8~3Y~^~dU;uN_-ApFVySe&+RkWHX*T+p5e8PB@N~ZQU&(zFb=N zbUVLGTk8Hj#RO`6-^3#9dYIR8jf9&qIG>y74X%krlT z`_GXi)E>{~*U6FTZ_M|A5alChwLE%qZ z`}V_@yVoiaKgf`%FQ<-H1o^OLtogE>+Vj$M3hxc018igiWX|?@uSC%{+1W)?TYHo< zI|{dZ)`U$YpW?1-n&*_8MXH-g#Wqx?(aYLsc|w)6(3b8?>Xb0IDgmMB&h z_p7+Yb!@E#xk}x*cXq$Jt)1#m>!%vzJ%~>jS>__lBfDMppGlw=jqZNbquZ~^#xMnY zD%ckvU;neGf@c&J`_G=r$HfC^PsNZ!%GKoi&Mqair;;-Z3qO2F)$%QRLZ!zf!^;1% zsPST0UxB9tI_qwT{2r63P;}Y6IB-4t{Q6E@;LU*%<#~Hx%SNT^UFCH(#^kp$Z<_8Q zgp;^CbfI77Zjl{SHf@szKX1QD{3?=%n`KK=4(`@Z_-f`>4)#>)`eoOGf=o#N*;Co~ zqUSL_#DV3}=U(7y?AKXoKAI02_2r)?wrj+4{HAOa>drf^DpLplQ+?m3k}hrf&NFqg zExLF{!*0$<0M=^%SA+J&Zy{0#;l=`t*@H*F$))TLYR?{RrdsDx@CP&pj#}02+#yrK z>eG71T2-Pas;F_=nVSinyPR$UQ1+NbvBr7WuQXw(zfNCYQ!i%AV0<4*m0sX8pF^?j z`rS*i!y>)RUGQg#M$_=ro_}thf|=xt?`;~!f~e)@s{|hoqQ|aR7(!Iq%hFN!&Tqds zwlX!Lu9q!YIh!RbD7KGCd%kNw407bkz16<(6R^VMP%_deS^JK~+y2Bi@!;*D;^|E1 z0*$gHY*#QTjkj4k(&ZD$rD)P7-xE`JiOyf>2R&mdS64I}^*`}CGoFusx7qzz%V;s_ zkc7YB+B-@6l!I{G@sKt0m|gv^GsZi&pNmtEiiVvda~XFQDsOWrNo_HZT(702F7Uop zUp?88M?fqLM{`+gBNFO;CoA-ozI9!4hsWkUf*s7gbz%Pb-J;By#6~{dl}>T|H0H+s z*s4`S;wj&`=&yOSrIK@7w2|S-RsG5fW?h6~>|HmbP~R_WQ%c<%=mG4k)yaKRSD5oyWvA<{*GZK^pU67wND7yaHpT%_ugNT#kBqv{AElq zOJwAGyitr{uD^2>C`+n@*WF%aYEENogKrAd9^qR^vm*%EOl?W)zW!Ja_TSk$2`3v|p$dKY${WF0)XH7?xL&QA zXt!kP#%TVi9qNmM2I&7>=7*@Ex8mp6u%hQfU9Q#xZhqd2*P?y(%dhZO=%gBF%{(exF-7{a~l{eUcq`ON2O$ zzx6aukor7h5IIlQN>+mZ=d;Wuih88jP>q8YZ$pyQhdnz_Fu2RxJ_;MB;SF{o zTA^Bot9IjxA?@jw9BW$z_xKyysCw99H-lKVK!8r&8%o2s)`)Mvo}iHF*_LCWB#32N ztjygJpL93{r+CibuiSghcnB!4P{6)XwrcXC&fVwB1=F9I{qpp7x-dDUQG1XiIm8Jq zIN+Q6tZ>lLK_&h|ASxKmm|TrI(keKxKvp>4ujvi?{sjxei$aCM`b(Y})LAna%Pc)y zPUk25>fvsFE?-6_MEw+klIjmbzSG=C0$X_8qWMaHnZw{;urOJpO}9 zU%Ks`eD;3jwH|g7s+TrpjhL#&CLUYP_5Mt=@Q7o1CXt_!UFfLHQa$tI1d=xWd_RKr zobW$nc$D@B#*A867&vy|2DBg17|z;1%K~F9SO9?N+Vopc7WfK$jV5t=y;M)htc91PAze{%@EDaR8xq%IP3Yof-;izuop^b5KdYZ+* zc%|a`g)V}=n1swtGq0ybSCqO)0f#TdpHUWY#4z2*Ns)Pne0e^U32 zgR!$w*WcjlU+xaWLG6ozVY6MEByU9zs_rhjL@Y=fWxEwCGLXy$I5Z_P#h7D|>)Weh zZ}}~;bJo!dHa-i$e@1!rG=KgK{XPDQ5%c+<9V@S>u+#Q*lTXCz2T3>2U^AsGRK?~8 zmcC2L-aKP3Y^{dR*?5;NwbG1d@pPrQE@6kQzJmraL=b^V z>PZ&?_wnoeYZM#`k_vO_5DJ6m;mHjl0$=h47A-MaXiR$@GqBejZqd0u*cFoWJxBOz zKc;5&a30UDTK<_sedrUX6vBMBN=ZL;oVlU~hqp3eHa^J^S&UR}(wB5td5$ONNHT5B zeEhEg1y?)DgDXTQ!Y1FDF-m-$>{nyioGQ%TGX_DrJg>Ts;YS0q^cHV5hP%v~_Fgl? z3CCs#D>m1!+2SHx#f=cSR)kx7FJZh6SfyL9GZ3p?vf?X$+iD?!4=-0SoLB1R!Z=4m zeoQ|8YsGIq{N$xrhIUdI?3Cnhb-76=Yhdjmr>3XYvXpvO6+;7eAEECvy%$y6^mTUIgml|{6xypvDCLh)hdAArS5`))=1aSRvmO0B{JLX8qkIa`O60Bt zFn=_@v5)wDYr$*Aol);7_PvCm>b0~y%9FxKTv5;K`tQyCy@>=gTd)w0(W7{7QvBii zM~*71H#DUk0_E1PSgm`^7Y)U2o3B19v&U`aR_=Z} zn}1F(Ew^bDznAb~>%;i;>5kY#>6*8$vpmJem4ZILP~``dw}L_$EHY;b29X~Gy+S0* z3YZk}Q?seeik~LUJU4+ke4;K6usRip%*J8-v+zDiVbz0P7(HlTy7^Y^pt2$`$@=Tx zsiYW``?~pn=bEQAW9CFkEEFy@#ma`K+T>;YkyF+b-U9XaN6y*D7J-F-xAfKGzx|@) zzUQ>xhH3b|J4_jRd??ODS5so+G0~CRNI+TZlrACG&09^@hSnbB4F)HsV zq(y;d&#_U~g8E%R8n4%Uj6Q2=<(-uMm6fGKF6q}NpX0-*3t;89s~yCLZznl3UcY&7 zL>siq)LkKo>^z?}LLzg%F#8J5>tbcZ>_V&6xlT~T5off^R zii&29m+T6QSKNV-92=kL2!9P3Mh$83gQI`cU(iGqH>@`6Aol~s-qGr+<{H&?iMr15 zD+bN-Y}&Zr@C#P^1frhxHUoeB!AUl0X-o7`y4K}J?p2_qEa$+CDpa=pKID`JYsV`Ry;_iHS8bYrP| zH*32K#78OWPq>uOvUyHu-scq9XX`y;s;i&ej?tG9_Jy&8C#Zh1b6s?*ob0vvXEf+JEXq-XK`&`v}NL{7rDD|c1tFRmdk%Rywbm5?ImeI zym$DS1Di_q%tf5B>HNe+`n82Sr$F;>{r+=TZ|%bl{Wrch~#sHTPL=+G*e_FPnv1)+wXU7e_qf*x#dsoB!GjP6Z`)?rkLk{U&_&I z4{x3~CPwr^hiUCS-RtZAaN_uxe8G!@<~}-oiqdmh$Ub6#&{p@OxMx^wp>5MF#;?!y z>0^I`@(l>zC2jZl^y+OfMW{m`T?y6rFb-pK+o3I_>84^$8ST%&bp`ReHPAgHGGj<2 zdzq&zFmP*LHUCrGnSdqg58=xr{+Af`W_r@`OG88k^!Tv(=}wB6xR-2^zpdNeb}$$Y zr_cG<8&ujAnlp^rvMe_ryD0R;SksX;iL3uU-tGqehpEAI6&mEibQO^=XKtCZGdqSZ<&V9v+o^mt3 z*UXmdr`-LR{^QTY0=8CCZ&Wi*s$~g7Tgc7DrhluBKFp};GNAC(i_31J&*0n=#{5`= z%aN~ug2<`FwVgpQlYZ$){!YziAGMXbk^O#3Izt&rG+e&_adLxAk=<_T(qqbBu6KV3 zGm?A^t5h_PlhKX&4exHHDr+Jpq8aa5tluX%2uMtP>;BF_OWsyU`u)i0N$92rpWULj zuYo^`J`yaICHXzk+lKIOf|v*jm8mUT#?2K-&(P*i6TB7GppUQ+ppUS8%HBHDTx?XRORf%m zgjLJ3sA^HnDEeR_JmNut+xt{&s>YT}{#o&3y&LFP2syPfub+CpqhJx&c-2r}`5CLT zZpz!&&Bcdt@E6Sa4!c{*GZp#SSVo=~4{mT2Wu07KUPvwq9(&8TNk5Sv;w9?WN{)Z_ z-KTOn8A)IJI*;@!?^f61j69p*OdNF5OpHuufqe}kNMlT^wKqs9il^kc(YHM!R7ZQ^ zhE_Kq0J|L^PQM`j^x#KDRD=+0U)6FO4!d0)irI!;l ze153IZQS-NeT0-&Uen*aorZmV?R*}RvY{3vAMsM^(<_CLa#sl{ zSF-V@ND9=3kV%_6%eGdoS6k{yq-fOIZ+9z4_8VVr$MNY~GICnaRgxjQ6Mpqnj?o(~ zO1~bA{DVFta;q~_YJNpQLiCH({7u1DVi~_;RlZCD8u_kmy7&wIQEam3yuV8|cuG*$ z0-g9^VU_jzLy0Mdi+)YVPH-#+znlMLO-%86O!`-&3g$`6B2AxPtVx&Ki_JAGWm4eh zpysSI!z!bSpHi&Jq^AFMUR4kt8f_|QlV=QJ*)Y6wuPG?8d>+$(W|c@|?e%o0j~)}d zAU;RQFan!IC0L3@f}@g}=cbM~_ZDv!`jE;6CiJO9=Dyi9fGBWB_pjCME38p+GYwuH z?GJxd*12~j1`-$#Yp6%l8aOVzoUy~-SlIQDF!0~9tVfws376jbL_Tlr!nRaR<(r<`um&mF{_@&TVgslR4x2Ds#s<#kXYQs&((nf|&~wnHOn< zV(V##Gjvfe>>k=;BR`@}Z&crmUEYt5zDVRkA(W5C)KLz*WvP*C%Fkz6f6dNop6oOT z8(_yUznh;w3!_wOYz;L+zphWHXx609vM?&z;%8gFWp>{5C+ksBH2C=`oW@T+ow zpQY=v;?$8tO0CU`{DQrDyh)ZlHx6X9;EKw ze({ej&}Zfv zd95Gw%YB6R4s9fGhk$G;_Gu-I_dqZyx$r6r;!NNlcIbwFn95QsQox-Hs$e2R2yoEwcf*~%7)hZH1MZ6`9z8|ebI6J%G2Ly@Bd&v6aFpn zqQ^s)g2ysa`e1LwDu%hP@N`xczQZf3;nm{Bxf}URUteDZSCv<`Zw_ICa#fLpz5=!1 z0U^N{x4^C1dwN^wxn3=SIE z`Sb7cLJu!Y*uEKaygbXGWZ!3e7rPBxczmV#L19m5W(MnMw@Nz3C+;JnqnJI*FjLr* zA48nvLiXI4%|+xlDk|8~f0{(^vCKvVlLW{bsB8(s8_~0loQV->sP#6Sy|S*Zher!6 zPiS9h<#k|V+UvZ^U3|n#!(%|Qmc{Nuj7Wx(`^tI$79&d)`U??KV4!zL);s7WEkey? zRuuGt7(w4C>nrqv>z||Xy_w%AY@lzL>Z{@4dw(&y6Y z@Urhjp;Xk;4<0R~mY_pZJE=Viss9Z6<9@mHO*EuLOe%IpfN%;r9RK{P4W<%w&7pbZ zV#89E##SF$*Kqc})QPOu;CE5heMhvSnz?@^?$V{5V;oY4K}UT|1?fEek3}7FBJ`B6 zvDoIN@jA&*dNiI?srJXKN*(Sam=@h^OXbie69&`Vx%jPUZF0sAC0wONxLnC2CUXi8 zn%C}ybJA7m^vCJPKBtj4{{cf!7ITqGE!41-XEF&#sI${}k4pQ-EY6%pn||uJ`eh}} zS0fLx1Vxkc_7`6C92Q}yZ(6-v(^+U!5R*NdLlL?vw&tt4qK3amb}O;+65cIK)P^WXjH9)hf&C{j6g8 zHMI*o$>|gX?B~@?8YD-MR0sTG@RqR4=mioT{ZLn0x#CpL+KI&KJmk%|GL`dIaNQAI zQu-+7L!*w7j0tkktKr3Ah?-q;vEg$mxl;bdejgH<{02jQi)zNu{%IF^M3X8#A8@qv z0Pj`chvG~{(DSw%btb*Qzj?`a)=A@I1Y#4tvcI|saphU)nZf2)qq#|A4tPSYr7AIH z(*rSqI?ixBMjm7)_`bfTJcM-An5T*AbW2U|Kc*_WWcy!?;MbPX!jSFZd3)w!Olg#q zr`J;=u;gW$l+kh1pNUC}!!zJ4p4B-ZQZKQpQ8kCw@(X=*|NA5~*3g@iDu}VuIo#Xh zD~G68Gk4$S$ZZKs6KQ^tnAJ0&^-mFru%oZyBctBU;Dvo5w8uEX-siCh*vmJcQlAx5 zwd*Lk?xECSifZTC%fqJs)E}#)c0Fh^Bik~mzI#7E99WaXb`;qwM-!e9T+<>L`6}H{ zkm7T0+DyRhyW-bmThtSO$I=q%;mZ8QP;?AP(N;>5pBL#W^^G=}EoEA~gG+978pnM? zKXEo`@lfnExd>nUT(NjeV|}8@)rn{S!U-)^2<2LLizyTj=TQ#K;@;vbF}Vr_{h#QY zv9G5>65$`6nM=~wm*tpk65}~McP)RQF$e1@zY^@i50@_&sr!`DTc7=?_`O)ZAGs~j zf^?>xMZwLU^o@seevL~HsVm2j=VU-dl)b~Z9Bf&`#@5~TS8mLvlXF(Xo@?3z#*Zo} z+sKfSbt>wgNVy~>!7|}V;R(zI&b98o@AD}WPIFoKqN)-uss12i4u;eDjllP^*lbqz z#90#4oUNwqejM}6Fm9pzQZk_Jti-V*ZIPw=X#-Wxo2_y<#GgCxHE=ex;V25Vsv&4E z=1>iJg~ydOIJ;|C7`p*D-GX zY$@is*UrGes;P+9mOhn|xp~3^dU}Fx9M!t+Yi}WV=5i6zrcazTsv{%uv+AB*Dp@xw zmuVyTKNss>UK#yeKa2F>=MyA2R9-G4K|JI320!-GG^fb@ zGxpQEX+nYL&3r0xA%=KPE-{|Tg(J(yzHCl@ElP?lCH=4RWKpA32T+PVWBkdTaO#E@hDWf87w#F%T6+4BR_6%YU zG#`U;BssYp=wT#;^ILFMY>&vrJmOhBH}~+anh>!aUOnbMAg6{~HSyzTFy-MzSSd>T z#g~EW5W#46@ra7aCt{Ir7&o0OjDERYVZ4(x+Rs(n0o{S)so3BPK8Dy^yd-8c-OlFN zh;PW0Z)u631192S2(-94Po@LQpD+=w@J!)yRWuRC<`%$$B7Qf>jOUNJALLW0$VRAT zG}O^X7jjPMFIX7)Wa&tWs=r2mRg>y)x89`P-dTDdMsi&)*mpWkd2oo>m-a!P^;V8& zL3Da*{n-_}hP!VB@&(VySiI6*^+s>mYr^{Hi~gz%y$P5}zn>)0ZgsxuV{})DDBKFyk~VZl8<^V-#s)Z%gqEy-M4ktp7>~AF(k(B{`cCuCgw^6Ox`X zTD#)ENhn^_vV&*LugU&0#Y}_^J0)Y5p@u2s0qL>>n~%TxDCD=*(G!M>mD;v%SWVOY zMMJGQxUY~n&|hUlXatRaYW7}zf%blHh_tWy4*~)!^#0e@$?+q_vSsdRvdr{u&h&?4 z%m?4QEeeyY2wKuJtPOs#4>%pGXUfS)Rt&Y&zP@Z8H{(Bdq^#pE-@7IcVRuxWVQ2k0 z?^g`S7bH|MCcpPYhgELt9Ej|o-x@viEb=?m-{4nFM#^gA~rs$@a>=Q1v0f^0lm$@E2Ey(3UmUT@qEU3RtXhJ+mQMsRwd z!X}W+Q65RcyHdPW#-RG{>n4e=b$pu=dFR!=N0fBI&tx=r*R#}9C5pAT(N_?Ql;585 zI`Qa{&PvYD*g&P;mgupxP8Wf)NoJ>S&M%@&PRU4NCVYx%n@IgGNX^+zmN{@K4k;u8 zLTQPTcET!jMH>`)zM1sw>zoEoN^L{tF}ZXp&$(<1mDJ(kE#*+pGOZJ&@;*VgK&^$C zhcEqV7y<)$DCo@v{DOJxx{Y73YkvLJeZQHki9TY;ZoDHlRA`$Ma>{qTW)T@cr-=TR z>5%XMR?Zvimz7XRjh+#UX!=S4*GH9mm#4N@J$KV|a-P>FQsb;|=)vxeQu9mLwu>@( zo2yXCp@ISn!*19stlQ*-!_Nw|s_<&_2T?@qwhmnlV#y5m(%ZVzjS3lh+zxer?+MbPS8_dD$s+Fe$~akg z47xI0ma2N^fEPlEgrz$_W)580O()gEtQe4_RWfl=ME8Qlf9tM_5c8B}vXQ27;f5}7 zOfw9&zg0|OPsDd0e$$`b-#VC_4P?*|`<_;cCp&frSN%h0XTQ&3jWy zD{jcG?izS<-q1)5h4kq0-_4wNZyy*XnIkkcbgh!JV|hH4R?=n9{mgXCyFL9Mk^?O% zal`bBPgQ*czj4e{a>cO3cWH9rYd8%nR%l7~rY^qyq9$%7F5UDx4E{}#pp)+*ZM8FO zv0voHf=T9Hyzyb+2eK2D!7{bB&%6>s4>>j9OuHaOn?KxJv0?R1uPI-xBQ;B>sP&g$ z4M?pi#`|pL)vUN9h`7(BC~G*<_|FZMBrDab_{bGuXk!w=3zUxsKej((z2PuJ#`~b! zj55=)&e(EH+j}$?nftwqB`nU5^v~;#n-i`?D?jE#A`|vEyMDYq_h}KG{r&&Hx%SN7A)Bns$c(Hql9iRL>^(BNA<8ag z)3rxPl$o7TiLNLkm96X@5&ySKaxZ)@K^Gzq$5q6TR_1WuhZ6Bf-OoiEDRhN&6eo2Usk&Dgj{-)a|s3!T^Ep%?x$-q*KH)`mC4$u>pco-=Yg zc8l%Mt?0_kdg5QvjJ`eDe{y<0Pm)4Y!dE$;)SE8IqHIa0 zNf(P|`n*u??=H3Au&=}!^cAXFxA_=)V;X+itmHIP-v+9b3a)w;CNFnu z;9!ne+_%1WSzaXWwhDn`*CUB4(twg4bm3buypjx2S}cSTIo%sS!~BZHdu)ux()h8U zSFkv{*+n@lZQ*$_TfC>`P6|p$81cz&3TC^?#;yY<^;*wan`$nZ&X~Y>ri!1_Y|o35 zMK%EGGp$ejPAQ=&H!ZPsI~Cj5X>`#k&ENateocqzi;>HP(6!G$YO2L)QbNPMCWCsP zRSBqlIJxA`lK*(MMNKVA=L0EM6P)GLIcS;=j{>7nnrP@&d1Y}~j?D*5_9=nlnmkL) zC+_l8`ecgl1yU^yJt6XfKZQgrG{c&D{9``R<1VGI?g$dLsoUMjuXxQaAwDo)Jy*BN zPT!j<+Gdm=gY`SlGvAcrY?Plx5_e+J$N;ZmbM|xug@W3{SPb)U>cGN!wsH(&=mK0Kj5Q(ZR2Gh_7lTq2ohE;iKzy;MpdSjOAQCt0F~`#b&x;w^ z8;-7Sjz&3$o_ zl={OOpNl_<*8BX{(2r8y)HVj#PIOXAI!RG+4KNjvYDLbDP##|4oJUo>cjq_(9H_>I zxqo0xcve&U7~QN)=A7R%k+JR@bM#z>0s1^cm(Wy4I{@RXQ>H8p%Ggdt0)^E2PS3S= zzQ~(+_5I{TPiAewyV{qRRwK;SJsF_5@PnOXaB%j^N$``9A+=tXRZ&Tug)$jUYqo}J zLQFq(RVb~iOJl6`38+#_V|ZV)*i5zV^s5y9X))ms6LM@eL0n`G6(4JV zW6yul>oKz@4ozFTzl`<~PzjzhaXS$#NdC$Eb)H|JgLh>`+i8);#}FZHqiOtMS@}hr zjjc4Eq_@)Pk2QtpL^dzwvjtrw{>H`n!{)`q=wAt1_baZjqfh)ApWMlsQi=JI->Rxd zQvBuP0>)24_4<{0H*e>t{BZVi3J+iCOF{Q4eLY?;D&?uCl6Nw^yEBQN)@+%D3Espl zZx-vi`P}3zlWd;Al=C|2ICny+Wx&UKy)i4#0;U=#&d7FqB_WHuxqG(DM^@7(QOG1OD(UJyn^fZ0go4>HTP0m6->o{ya(oj?4C!S zUG**26S*|j#f4rgb8kl_K- zIEEWyur^$cxp9mqu~5pvAgoH_x990H!q=Jx{bxpvz0!%|m)bFOuC`846hvL1KgWO3 z5#M^C-rS%_LZXD1bgh*w!PdI-snV_UeNOPpx@D1~;(afNzCg&970Fz=zbDB*gT=&3Pj_q#a%_EpXey*rg% zmPYUHE({9F%T;N<;+(e~SI0mV{p=7GeaKc>B!f&Xv_Iv$ zasT_@+l2P-D&D~=>yxyczD@nYBd#9(KqNN%DM@6x%h%oWnZK`R^L&ZWw79Ti#Jeeu znedJUsm~c`^^SVPwy7}x8jq*Y&@`K^{0`^_BW=@3a{J=xPd$%yg7Qf4dv`kZE)~pq zso&6CRISM*F1seGKYNWMToIFh9RpfE``l1yRQid9YSl1(ruryf+ArGyV^Q!OS2T~O zk^uja7WVDhscbL7%r1Csqg|Vtmz3NmTZgC1C-h#Jv^ag#+Q|<0h#%+r+`YIq8~n`n zRMT~7)46R1?F_LBCOoV8XMBw1P8;V!>6Y8Qx&`O^$@FrrGrS8^j_dw3MzT`^y(p2= ztaQiJzxjOo$VAq=w;a+}U+O>UOj-J%@R3^ci{tn$_IUp%n{1JCskvv9=`y|zgm3k- zgh;KVU3W|>p76JIdiK!NQS_?`cCB6&-0tMLTj+#8taz);HP%v6HYMZCx9|X63^nEP z+)91Lhwsd9G$y^tqV{$vA2ENun%xZ@&hDQROCgAds`W3Fl?iD+{_d?U7sfrst7bky zyn180?OAN??+hHNJY{HEW6h_!rf#Fx6YUZeH+~7cZaT%%g8NbAE_j`vpH3mYL=h{y zN$E26Gns9j{3!^ub^QHcEbcZvwtv}enVH!!p!|s| zfTL5&pxM_--RWwbrcaL73S$hi@^x!`Fb6kj8F(kY#FdCK4F)p>XA8Z1f*&Oli!Y-$ zzoRW4=5az#cFeGS4PRm0@-lebuw%GoiM=J6?5w|O66o+vk+1k${F!h5Gc8-ab96KX zm(aCVXXP~Gbp7=h35QN%dQ1zu{)Rewh&W;vpI;O-MZZz?uGKWfnX1pW-oFPM4eMNx z;0A*XWZ|RLpv^E^ny|91I%*4j0E^F(3nZjk) ze5+BqpcI8((Rb$&%TvqnY0H@a$aiU}$G%Y>!Q5DAmcyJpq^ARDUD@u53Q46x1Gabt z*7WlT5{s+zE|}gAK&x`uD2Llf_l1EUGG=Y?b!)KZ`;0A|+-efmU7vYDj`GXotKtZ& z{NMVQMPp00wXPN=1lHk9nKItEVkTPa)9zL8!u1kOR7mu)dA41DW4h)N<~<$-*%cKn zRTIS(k95b6qniGgnIV=i!=Tw1W9al&!|-jq&5@Q59v9t{D@j5PN>mB4QjGHmPwLz= z4ZW}H-)DmV*;Ao|N!hbP!`{NSGL-A-LIH8z#X{0_p7psi;@MD1X-TP@9zM6uV6232 z1c~vR!c0t3O;YM<>T1N>qVE%8Pqy)QVsT_%ou|1a9uNokM4c4-DJ#h#7Y63vIzp3M zqfxvAG+-g`=ao0Uc?QoSjLds^wk(aW^v(|1L^rSS3O@bj}TU4Mq} z_LpxXjoUW}!fD=XEM0>9z9TpI@pJ4*%*LB28Uh)PC&ChB4-L=0PN$l=bx#XgQ_>q< z+m(M~p>Qt!+u2~YhXgzrg{-~|7;xYfxs*nN&(xLMEmBKZ?)X`~k(SwAQH^QzL0t+s zRP27?=dY=0B~W!*$ru;3UjBV$^t(q1L-9PmyvzMOZT}oAtDk~%ERnyT!5*tVyD#>3 zjilv=oT7lNK{Sg3NP#FoBl9#QLb=I`7p-Lga6`l=-3j4A{AAyrrz zn-kLD-K^cwhi%6p7tQ@V2>t+v7*YbYnTTvI`t6P%>&wh&;tC0-?Rx$)i{s^hKS`U= z^9+-;cNzIqG@qg%_14am1)U9pk(3UCqjh14{NG;OtB~)}7H3{x4ruitlMaV}lWv9+ ztqJfLut@s)dOYfrbV~X4dGui`^d)Wkyx(V~na-S!Gb`s?e-pP~p}2_|VpH(hu}q#6E?i-ltn7L}cuvLzPZW6hn-e<4S5f=) z$DFR;50PBH*9u|74*K?NJ5zOMMNTmV6#s%77zsj}NrTLI2xvlxQim;w;n6o(0Tr~Q zZHH2qrIuc!+z`!l>3eCOGrljTGca)|EX3)AcM>_D6LwbP=Row!1mKmMjA&$`vTK*W zd+T%7HFm_fjVBiKvyr#vNqmDE%-y!1QKJ5OzxY~r%R0N@3z`QMwYkfZqFbv3IN_HA zRLL=Vy00lr7f(mSZ!$(Cf^Si7IKgynf!un#<)@OcYJUE>5Sdz0_vfm(Ba&^mlxn2 zRiW`BPcW{jTM)B%KXUtphIPBZ!ZRaKgg35@%yX%3U6}}ngVRt`WX8sL!KKlx8GLVO zk0uq5M2sbxuf>QQhB?;*U-g~~Sual&P(!<*e0&LAWYdZ>DLLno7p~`jTYS@YQAr8k zxvJ>9n)~U?$rj@+=|eQlGH;-!GkAc8LY-|*+zFh!W}jY{Ts`4+;)mVs8=<@soRdSp zbfxn{E>M2ZHNjVG7O)}dEzQOEV=}(PqacjWfZj(8JRy1tEiYVeDWo;Fe$I&fh0+o8 zdmJWRXN~P67wzVv^A4f!Z4AcI{nqpCJ>SfZpEJCFOS+TPBMhzi3UFOx1-eQ(*CthV z)@jj)b!y?_=TSG7Q3t*>`&65Frao3PdK*rnPNtjT9wXFnqhUyXZrPX1XlhzB5*lY- zr+q44GPR+3D*qa5h261}P>I%S?NUo>^%+~DJ02TKi93(6GtO5TV{@n!CYQ zICz_(HKOv@>yT}c1Pl@Wxe-46!bZtl|6gj{K!~mxyYA^#XXCg~Ll?bvXqa`2^0`xG zmm5b`@Ov3sN8fLswIY9cx9Vrz;HSZQmX1&NI_E$}4LkodfrB1C!>?=8UAfbVXWUOD2-nUz zjR)s3t$w2mQD~UJ(?54(4KA}~6Dd^&NIG@Utx( z)ziUWq(m>$xx0IPuHC`N>JG;CkH=(ucM?<8){(k>om)!HSz+U|s3zd1GZOufn$Wmfy|P@Ufnzmp{I z&@gbPe&>1LDm)B(6aMvM)yb$Axy|NN4Wat?zOz(6REEmh|FM3ePh6;yLoO) z*YjLqwf#!pb)NUUr4w+r?95N%g&$iN`!kb%KoTa0y>YVUYYHCRtCg9W{InjQ>etC~ z#l3L)0gZNCzg6^g^vLa=*7Jc-0;y#+AI~x(1C>H?FY_;tT*tJs7Bqz7b0w}rqQ%HdOJPn zdPS3c^BAvV2^niMPNs$w-o!`@dw-H~Hj&vUV`A0$5!U(8)5+6#W>2Un)1l(av=1L> zUmw93c{W**h5p>0%qzFq=^3OZI6sTXHe0%%Mp-`jZsllg@OE7rWLxVIyKg`*aVBeJ z&%H#bjA2uQ7zh3DE?+qb?0!lK?&W^o=Q&Rxg(p^hS|BeXKliOD%(}#6@O1oE?pzI=a?AQNa!rTOGsB2G@EOCHI|w=Uvanm0YI@2f|g6>#Y^cgK%n zh(1A&k!(Evv{saUvlNSPU0rAGiqaFcF>f`!^W%nr)Nh`SN$^*R$CxX^C7=+}zRb{q zPHA(MsrAg$YZ7)kBo<8_WIgvM{ccnR{(q3eDFQY_!U4Z7(8b zK12-MX_deJE&*eRmDGF1;{gLyxU3;|{PG0f>3|&-ft(6%=J1P=pT1Gg_ACmED`Wqf z4}G2QIXVQsVLSvY-Kj~DQy|JbN@KYm10(+0Sgrm=JpB(+!E=RjMBbL4ei}_aan|6F z?Yn-;#S*7yge?VEbKyqVqRE|)8K=bsHD_$X@G8w?jS|?E^QEt(2^4XeN4c};dk|Pq_3>wKvIC9Qk6Z4*r09SHg~$KmBcBLQ)g2MRQNU0%jBf#GSj+A z${#8$r39}BR+H#QS8cu;4J-%@z~V8; zzcaFrGXGFh0a&~5yvq$zAW3N1U?A(yrq_$6A%Fe4;>BfV*|T>;XtbZRQA3NR02bjG zKlg^Sr5~e4?l3=%aq&%1qp~1Ldi0jS#6pfLZ7Z{fJzJ%)K7u?a@j^Ql!TC&+dGVcQ-tvCP1MyR8F-kW@>U z=^KITCz{_>eOeypGIcRt&S3w|%i12s*S@^XK852>B)NX_ho3PGcD{#bDpz2yn4j!d zCVGO|L`BRgoc>|KFdl2OHHzY<4{=z9_O-(j#kd#ctWQ(Ph&-pb>%>Mw3qW&22+j&< z2cZ@2baN7IM+p!=dtEcu({^oBAgI{j>aPH8{j5H^xNEUo$;!A`CY%dbad<+LxPqN; zk`S0}v@j@;yM3`-mkAc3GB>&UxH4y0r;NbJfMThi$YT@D{>E(&L78b6jJYDXX zx^|MxDd2kPye5_}@I3QQzvNOcRMkP-QeWEonpdQwA(xRedpw-=}WRG%1{LI_s zMxBqkp)~zt(EDR=UpT!X6&O9M;StRXH=enXBWxl{M4TeF`lVHUv{$Vm!8S<<8nToE z6DSz7*jdl2?#3njB)SKoXnOP;IxnAkyUMaGWpI2wNe^!@7U$pK!`H3`!oD+phOW?yWyM}N!jFe+dxPkdrw>=7tMOrxV-Xx?qJKxiHhJ7?k` zj_ieYv~+y7WUQ-d4d1yzYdhro04Zmzdm>u0>LPM_J+0=#e3 zrFA8>ZK~S-cu4f)vb4*MPgTXfk1q7*bn(G)$LDWvV5Vz%m@+aB3%4K`Z6?LFU8(15ZW$Iy&=4eOOP>2xrUhO7g>m6awVx^L`2@0Q zJ+-x2>q}a*Tbrhx18)a>C~ip!MO~jVYxD3FFhpbNVy3n5!^&J~ijv+LzEoSjv*x|y zGbp(d=xku?t*YGIkVOjd`CHvJ=&J76a+Tl>1o1HoB#ABPbX1C{G1}; zqTnczy@<)_WoqyhV6AdoG08;UH4PDW;vX(G8+^yYwb$Wr_X^a0I!$%{s+DfQ2QwNt z>=yVf<-tEwKCXa6un;l>-#{4%xc~qgGO>k_ z`C%1puXe74UJ{Kvn=# z9QHoTU!Wc*ik?tdMB0oI8o-u@2B;u4Wm)k@Pj;vRf(P^cebl-XUItDZ&?O1_0HB7{ zY7aT?eyi_UiXmJ`HX_{0<&x8(^br867XUy4so^HxKdKHtoFUXlRnL%++Ku#T2JrviGle=_cZ7Bhm+!P#w8rX;kx-te`#=nmRjJ(B03EGtdJs3ZF*Wq|z z+6_I`AM$?glmOtZ4ggX}jeLS5~KGv_a8~ZnZrqCuOkso+comf7s&yc6Tqz8k8Due z1=K@LfaiCc`J=P@Hw}`Wa1J+kG~;f0-O0($jsJ#|m#MoGzk{W_tBtuEKT@8gm)J!z zs(qV8auNV^Jph0osoU=U2PD|ASq%8;hQn#=>@He0oZReO0RT*l1Ay28uicJBYxyTH z7w^7&bry1d))~ zVov`PDdHm>!uS{I%C2>*$?HyaV6qy&hz4*YA(1`e{TGrOzrB;S>0S?*S_$$wJ6a=m z1HxSEQsN8a;K-?31prdw0{#!TgCp!oq+1%BlhY@|e(2kOCO_|B2Lp z3=(y~^E0-c0MNmMilpBAZ=^vEX_SnOphYaIWillT_CPo)7-NpWKWU{795oOaeUt8p ziw;CSKnH}8T;)6|_)jF9&v3{o6jxEfc$KaqSuByjEUXVHQf zMJjeY;w@nEi?|OQaN0|Lgg7GqLUMzhHc5w^JUWWF?{l#-v7-YX`RFJSD|`B%XfV@3 zq=-d${kHa(lVRfMz|T^2KorTs_RC`b;spyXwM3q^4oy8pjbM-Pw}hWPs%>HBZf0uc zXkohhjhm%AYQ9m}9YYDM5^m_M004&pfc!qAJx`sA_+Kz#k@JU>#U3-GUF5?#@yvm( zySX&&+5dvfzfarL#?jg12zl+@CQj$P(-#7}IW!3Xw2;UXl>8r>fThhHwuK(i26cGc z#TM|bmQ1Pua2ctMw>|%tHU<~r5ca>}YO-sG^iz_xchP7-6EErzqMQ2vYs10J+VW^h zL-h1deMX8aZ_$8oYjogmf9%c$IzE#BL-fGRe!?LX2h0z}U?I3|po35-O2M><9z!Vc zPy+3cW?Iq8^5`So4b>Psk^1ykuF9o&9qL{iId6V0;*pLP?jRie^3aKP+-qK zz#-fRH1~E4g@X^B0&?(yr{=_f`avx2+O8jRVUP7OD0DE+p2HLV9#*mICwVt;!P-y- zF0POQ!i@Lw9u_PY+~^!yuOKWToBxVcw}b&O5MTjvNF3C!?(3m&hyq(@MX^MM5sfp0 z4+Hr85Cg@MyO;EiwWPQY4iQ8_nOwOxnAQye6z*dJ=aIOS_!%CHQqrn`vQk7CbyGH- zS_A_FQ1-$E6p?^BHBFBNdYPmQVMEHX2%w41%frIA(1DIm=m&XdH-lsa1nQU{3$$w& zJ;BjnN5B!hX6y2=L<6Spp##!Lyw3Dk?!lo3pUv*r;k9vuHGrm)A7z-LFF$+P6b2FN?w|wN=okmPv^}ilnH$GqIkOu<`qAzcxX5x_AQW$N&Ax2Zv2+s&h}2Omj51)7ieWP zNG1SagAf}K-*@cZd{OWRh6fb}36#&q`KIG_r5eb|?S?c$JO`NFsH;{b!ed;4nudqAZ^glr}vh-7R{+Zh8Vifoe4N z{0Ed1nBNaAM-VuYx)1Ckz*lUY-2@JzanET z46s-O0MJ2#_9pS!8-@QB`pfFkAe$=YAHbmlfK3YkutLIB;QC*2i_T|4gpX2w_U@9H z8CVVA_C~G!%qZ+bgQHZ1CLOt+w>MD#OJ$J;jx13>#I+ChC~H5um_;Z-*Dd>nbv1+ zrPc5Qz+x?MuxZvn`B&DJjR@XBFHzQcy?RsUfKF_p2vLx3bO zlqxv_kQc0d(?Oy!gZ3q_fj>+BZ4?nGhil;%k$Pa~`kj5f1yE1wuO0$hV|y2S&i9YZ z&^gq4IYJDM5PPB(ybrZ2MhbfhmbsBjst7KJ%iO9+TqAaW2<^kgZsfC; z^lOK;#Zh`X3?xnmVG2GhLkdHwY$IAYT-ipBrOKIPB!0$Vt(pg=ohRV$z``JL3XJhy_JWDHoEQLDklM(`$vJ8{jJXO9q59i_2#i9? z$Nsb+MgR`DG(p1XCpZSoUZ`D11~g`s-)?OV?e+z*mvBYwUg;z ztB>18A((#n>GcDj1ON!70DuRHsg&h0jetzaV0qPW2*H8Hs2fDeBF|$KEn(?0IS?|G zQ7^pf@Ncsh+?XGL2gdKvP_|73UjIgV2rWWBSwvth+_a<52ZMu&|LbHZHE;;maTbv@fccc7Q}r0#c&z|5WCjR_wurngQvugKrngb|5d3QYK%tCp4?sg+ z&LN=0(0GO4g1dF9YSa}dzU@DuKBu7xAkCBrsE%ZU@J($1cwC2iSZfh;EYuq`r1HU` zPDYfFDY_?)QG+$qA&mv-ZugST@K`Z|i<+|(z`=>#UY7_=9)kgfvAL6lrRi=Jz|6+c z(v{x<{L$Xj&B?>n9LzPl|Mn-q|7;G98PJF~PQygOdkIts2VB3S33%cHzzVpD{c|F} z$D)7k7#6V8^>B#LK}66aFb3EHi?x^lU?~pui2n8ZF&ObQoap~5r64Ms6iGCBM`IFJ z0>&KbKtZ%Ww6pEfDhe(U`Ov`Ui-DUz@N3V8^?w3Mqu5Z|7TBPU51yZcQ^B9I;vR;) z)<4OxnVUJ=xSQFVI+(fIfhPn1HHQW!C`@S!0KoDv0B9pI#W(&}D?%_Y@Nk0sfbU+& zAq+=Yy8j2$J{$n-mJg;QkAoXR0@~}m#88Mw6gYJ(D(pVK_OSC5qmIKcdIc6{V3tkr z!$;)^Hn{Y99~Qw8P7w?M=a7&tXB>kRKlKWo_vnaxKVmR~O$q=X<$#0w)m}$bH6DWk z+vOv6;G8|1YO|s~`GGdsp9c;u3G87=^&Eo{FSCYz_ArxwxR?vc#ggBTgL1)x&>n`` zBQyi!GD?xe01Q=e0SzSAJU$Ku_MjEqAIJYb77OTbESZGT1M_j04lds88TX2{=;*MC z2N_Yr7(hS*0zPv$&tqox@Q<2&#!P6yb}R(N1qRhf6Ne|m5nN;0CSgA zH0m{Z;{W9QjZ4Q9Gk3ML zuyHqawQ;jMju~Dmx03i70I&n=`hSkm_Qt8U`Z0ho7w~DJY%q%sID{01A7Y>$4vxQ< zPIhNSoPcBC;?3^KBi~j)pnN1S@NmNe55>VM>YuaDJyK4`p}^!i;Sl))I(uoKe#7Qd zD!8E7c!}zA+lXVB>=_n6_fVeX=oGT}S@A4-KsUOUZZG$6;B$habLI zhOnvpQYm#Hc!2T=yr74CnQiq<4dn0rs{PIDzN1qlC?TXr2_ZlC{%~^mssAsqzwxx^ zAmk^*9>zH5a2&?&b7Bvz+z`9(i&(o`eV&9Gm@W7n1J~|_Tn}%201aam>@N)P&H(0Em zXEO7dq|HZy`j{a?1SMH8a9F_qT1*IE0s=AAxvuA*{~}458J8M9T&D^F zRCLE=?TVVBUB{X03Yjn`cZR9+`xio_t!h@_MAmoPb>NdQVpWi+7aTG@EqXL@^0E>` z;#R3h)$%I3eM%c4E~sguLvCCvPimS|7m^WqyRYg}ko23_3_mMH!NSBAHUSSMqdtG! ztPv9qtwJkZ4qYQcX&^3T=2TQNA##WvsPXDm=s1-hmFa|GQi(jM#&l zaJ3ShkI|m(tzbpTqHhaXlKg8r2bwJa=de zECPZpVzhI_{g&k|YW5S_Q_o$Wfi94CTwscH27d*x zsE8;3I>!xh@%E;ykTUJq$@O%UW%V%iv09<)=^*fpwl$2;X9i_=hb7#bcMuuIc)k2( zxp)_yQIi>^eC6`arp2@1t!cfgnm-{lqGHB(G@i5o-dMT7aEErsknJ;VgZnVA6MQBM zqTP1A+8k807HHF60by0p=`s@dK!IW_-Pt{=!M(Em5K@)4Nk6tlw$CuxqR*3Dc?e+# zqh`>)p*x$1{k00u)hk?zk;B1y=l5zO8q0>`-O92UL_)Uw zDX(8o+TZ8KT-qkwKJ$e?ANHrW!{h5&TnB?ZUmmWMJ2^eRz{iK9D>Xg+$jd6U`|Zue zv{lOh?jF=M&DDd!uE$HZ`@v=R+o9+6-MZm-b6a@q9~8ZNE|s&hH|LkMyxzAHrN3tz z_;-I9s6Ltdmcb)nbCM9oV?cgKI{YR<;}K&q@tkVz!6W93fMfN1>jI>36+mjDG<^N# zry6ddZsA-d0j8rucI6mdLpfYQ)9?up8c%?tqxD-Z3hkrI+T3nqq|1gXFsm%E6pw0h zTl_xIQb5+H0XHF-2n}~FMGi#5CxyJi#voZEjlxz|T;<4R&Kt`^7*Y$xR^3~R4i#zE z-vTje4FBVEsYJ+3&BWa-5=)UVMcRg#GF4$ILC@$?2z*EhgeXj@r_EJSEgfnYG~3Ru zjheySU&pXrTaIdywLvfeTs1A3GlIA(7KbRA(qi=?WEk&B@_Tg6oI*d-(FDE~N|WAU zwEOu0NV&PHV2`6Pu%)0CO4-}NJ5fVD{pg8G0bC&7<}<*wVtHW{4SXF2Ol(z!DEPKk z(iNlyEoa%*U*Hwy1bvsW7_IfijJ!MecVX)q6YF z?Bfz40b$_NCUHqxz_>-2A>l5Fu~67g&*nIJTul^aq+djynE>S=7)NKt+%m7sld#Sb~@-s(KNaULDe@Xqxn% zufSLm3lorXgW1L>1<^q30tsRI`b%S6?9okzb=gm*q{AFfc_d|KAi&5UuK%a3U1?+s z82sAdD?ikYAwe!R04S{NVss^!$+GhIqaJ=4R`EQw7 zW6`I5L0M1;b#~F)$4{u)8D{5Rnm@!2_;!o)HEDs?@j5Kr-rK+UUAK)4Fmf^O_`IHp zvbXt#84&S|-udVL`v~G93f{Pho(FgDLT$WzEI#Qt_QUtdN|Nu5r+Iitvd4RR)rb&W zPV{biA6yLtrouEv1`=ssN(^{XFM=-h8bNnVjfm5Yeg%){>tqt%YNEw)`rF48cwc6z znD_HO#qbv%Gb;j~QhPuMIYNP;mhLlcGr5P*(7lC+yM4{&FBFv20mVsF-|m2&cs2!? zL$N7Ff}}+PU^oP-8h6C98`3NWs{&5!=J$()73zf88fkXX!S6RYLMnr#ACXy!-$YMv z%xhcXef1e5;MfB!`)at;E;{+Bz=5sgci~rf0#H0Sh*a<5~JQ z)^HfLRyA^7+6IWyXVwh=>ipE&0#$a)kYP+jZfvC!xo@?z==xARoxqbomUWlHivF*t zRCtvuNEPQD$>E>R-5)S})jnF1qP&|L-s>skJ-Z%8{T6X@85!QtBxxgh;A%2K3CyIe z1T0_tDgc{pa93@JrjFvxQZ+d=xh9)|QtXEqc{^BJF43gZ(RAZ4vh|)yv987^?FbxEAgn?&o3v{2_M~KEn4pppmr*+|yxKNT`ht% zT5WYCngAV-jO;X#^_d>@N+UYjDbT+@J^wKe!rJPX>@Qn0XLh5yXuxk(RV8moF+1+Q z`GM@`ITH$cF3a#~iv9bS4}FUr8~wTV3Xe9v-rVw$RLDLQ(kV5|b)&Pt2=d!6g>n6!_Q?4qkp z$PYR@r5B&GXTy#|lpvBd!>%h{DUI%-@n;j-wdaRMWg9zdsPTeX)ztPebjNcTH*l-| z02wl%mtM*_7BCD8bjbzVfZ#=JxWi#z-5temNs1DzeNH^s{c$syKc*RQ%6kVMw8eW zK4;BJXyuVTP=jR&I}IJZa+M1g*Em7fJ!B~+YLY}8#Ic(JMh~EY4AG-O+{bg`5S(;0 zReun>h&t!JR-aam^)OW6NRbMu?%|M-A18N1khBL0?32?ggz&8OS(;-x$x{}ir$|5U z+dziPT+!f|6p}%doyEw#m}e}QqwIR2LO5zoY{)OMZeiGURlP5iIh^;XPThQ${H;hrM=|xCmg; zPHOyMFFUA{ACpuaJ!W`yr-TYDR$oeIGV28oKV-p*KtW1|W{R{0}?} zRgBUK3k5)e8z@>th_9!YliP!$NyIq217cq^RV=1fF_Ys(7>?G4cb!v`6JEGM_<;+| zjZ}=Qo!z9Lsr4ua6jv=`l$J0fblME{so(JfriG{7`6xb3JfXS2ECJ1tU{0GZ(>*vK z>(x`HSD2=SG;YQMvZDZ_nW{%wZUO82YCeE556)|6@ z30}TEkA+`l&sreN9J>ixpmPFU9-qLI$N|e+X-OFX z_u;dLtC@VmRw~~PMox$X5|NoVAP!WS{DX=)w8-lg#WpEQ`By)KIvV}&z~~xSmI>9u zj{tq*yEPqq$0d$Ubt+>3OQED=QMG+atQC=!3>Lf7N2R}!v?Tt3(x2jS)?#3?43p_y zj95S6pT8e13wpn<+Ym$l&YGuIl^<^@$aTKJQG|~0=3r+?AiP6XNcKwn)m^!U9D{<4 zfxc{bV>@_ifm0(^B7mxgGXO^Q~&DR)$vUsugWyHtDH>#9BOq$Fl)3dLrO5EI)GP zLVTzLO9$L&&0fuUOueN}cLVqY1`kXKU7GVnMwdIFbx?4^AOgTSeRBbJy6O_`rdS2e zf?=6p({LFB&F$?hr%59K-#UHJ#Z?SxUWk4&I|C;HI{v$7vSZk-y9lXzmuq1LK>blN zf1xwUVbsZ``eDBtjVDkJZJ)h*q7mX8S7<&%PA>_Q6~!(t_;y%#nL=f`3<$M6lyb9u zL%%|1zOgnEEF zWm`EgI`>luJI>NU8DG}q`!x-j3V_$z-5ckbgeb;y*pNZ^3j5&rtfaP9=Vn=sB)jIx zoq5#Vvw(R6QwYnL_SLMR`Is<8$~dU%Gw&W5Lyo;cbiaX>cJy8>qNyj-`B8jzfl|^@ z#bna@7B{(uW=Xz(0Z&@I~xKw}QqY~N9VZZU0Qph* zhc7_^uc$>@3TifS_R^`4a6!Gutw20dH? zPf;aYm$Ir6q8po1PA>a*Xel=)8Z=SGkOET^enyr1Clumwj&uq-oiLAM5W(Ga(Ky(s zY@9ynz0y_(DrL zRQ?e#2jJ<`@GI60B{av|zd-AF{7Yg=Sqy#rsLNMS-D0D;1XU0t1_%^&EGQ+@bHr>W zvH(J@H!<-ySzq%EXaEYBSjsbqo^qqIw?^H+P42VXa`NBv{4DdG11ha}ax5AjEyOj! zMl`#tV*)2uvGYZvhklv*Ep(AQX#6*X1J=W9_Q1n>{t}C?z_7)I7x9Y=!Q%s1k4feU z-7QFDPMh|*PM zcnJ9OlESi5vejze|0721E|i5VVYIquO6OC>YI4lu1DDjPEIOK@9G-o?(JNOgIgh@l z)$~UghZMgDF(ll2&PS;$Ppo(0-glLmCN)Zl)GyBR0$k&jY$wii-xB&x&!Ub)3y8=q zeB5`|u4R;S($hMsXa!cw724tXJ2=i2)JSBmuAzEYT5ME0@0a9r#PbREs~}*%8dm6~ z?|9@A1wO5GPhav##Xp$oMK=cwISOSPt%ZR;1|pQqZDW+HjnAd!*v%49AYo(SZ^`gO z8g=J(YAS%Rl55|)|+XuMABCZSFJR5)ZCmmuo2YA1MPp0-3Rfm zT}iuskvX`x@$om3IhN_h;}v_CSZg$OswQsq>7mUFOx1?FI<4|!)LwgIRe=i55iP4= z^Pe}Aoi`LLt_-B?M&?$6MIj91B*&ote#^wLD0#b?rj*_bTtI@aCLMeOI56;avew>n z6t%pSR0rc}+9lQBu5FK+0{u^JQn-yy92i00-GsLyPdU@V?uM2py}-}v<*{ZB^l<%! zD!Uz-PqL^FD_@-SR}K%B#WEH^;;=>=Ce>=+@T`x9=u^^oun3*Y^&n^5t*#X)S+d6W zQLxV1G%vKZhF5dPrkUZI3RmU_=%}rY=V32swv`_Zt;pU@SPc%ef%iEEow#8wm+Fg+ zt?kVzc}k_P>6i^6JsvcOe?0pyhqUk?h|*$oO7GB6FZ4BL2!Hf83dxlH#CP>|1blXy zgL?TJ;lBIz3rs6!d~=I`Rm$CMP#mj$?Oh*5YWDd*LS%n0v^pXelM+27m9s|m3%|h; zV@EcyayKk9PL=Yj1MT0A7nub+IM_sIQjXq`s^+a|@Z+tC?i%KU;@{j;l|~_nU2YwH zM;>hrmG3GHvlo_sHay^j`Vj62o3(eyENhl{mZjBDcF?#9J0>>#D4uxXw(^*Ez+Bs-^;EQ?v`)s@`0MMi0{Go=o>YDGYSQBt~d5@gjiTV=6 zWp|w7FN<8O1gu>Gsad+85qx}Rd7^k*6_(}`@%CJdW`vxHO9*1WWTYmv>Lzsk1X|>q z_19*OS#b8xLulbL?np7c1|<#M{JcL8H6jwZM$FLrfO7+5y>%H92N)L8gu@e^Ow#31 zc}&}Q`3@c1jfBz5Bw1|q1eA7fg}Z&3 zv-UGKtrsBrFUYC`PU^XpZWU`gT@P8jh=~8H07d=FJgSS@Z(a_2>Yrtew%24*LGmRq zE}aULz|Wu6ofgJ6TY~)`Pgu3wmEy4A>L^p$2Hxdi8TP|9!bzE|1OZlQQB;#eVHg$E z(V`4~@t-G@4_x15ewo)z0WJVkH+O1@!sQSv1Sr&LqvfnY?{f9{^bJ|UWEKWgs87Wb z`x)S%8z9uPjg6_|zAjz^(NcQO*1w*8R zJi?a6&;YJ9RnWt8{)$B){G|{c9W@R9@0V_~`8+xq3C*}C{3^2=wq}2;Wk2=J2u$p6 zU`_e!g7Fjy&7Hj*Pg1ge!f(~%zzY~vd)Fxh%t%bwPHKY*Y!h@w0T}!)@>tw92wQ>% z8GOi+{s>5qpMO1jsOz#sla&q87@0QbbnhGUKgVHw7{s}A*Mc(l@h4&1 z*O>Eu!d#vgTbPVlUi+(6bs>aS{Xl2w=oa=#ahaft7V1X525SNr6)|aLG)q^KM#^VSBp5m}?P^ua(k%%N`4-?s>}*Lrzhs2=rxKmjCA z+St@Ssf5^}#rx*St~{(ww817`T93T#+D2wF*baWQW9p-YFX8gwKJ#V4`XaT5A$wyL3 zd5b9rbUdUl(Y8d8O_uf(Ae0f3=6Ul}bdUp3U|GMM+ypVpsx0I_&wZVvng1cdD9VUzVJeD#)ZuFdA!gGra>MfLN{qQepx4 z`viGt{gNk2(Z#Tzj6MFq_z$0}G zNikR~R8}@5*3Mg4#!S*XUoXd^nA0{r7tj>bnP958*dj=E=pw2eedZvSLd9C>DSgG} z%ObYeVj48Q-UQy|7l{!QFwCe2`SBHU??XP0!i0d0)V+U0t5X68XOkATd1mFUM#W04 zZ<#yi;+w~!GWv)KS>${OvJVRW3YAmy9HY^#`dG8eh#N<>pcw^@RC_{42)mZ+@4}u8 zEPEvPy!HgYRd9UV11Ir631_FpvLC(EPaF8xVWck*C@Z9nkH|lTz{y`OxY@&%jH6~l zqVqdhzL~!5cWW*B6Mqx;wB|W;o~ZAejm;r5C!Ijz*ZwX_=x3|G27*0>gO-PYz-#JD zTTSrCx!TESBklRjQCer*4bqRB$e@kVvQfqkFtFo^(l0FG#9I6ykFHPI9W5GbrldKR zGUc5NjogSg6W2FH0#V!k1!^!oQ|fFevWe5m+ltd*IR2U)lu4$D;boy6=RjBQ&C?85 zoX;7~;)RG<*dc(4Ep68!7id`f%d=pk4ti-{jJk|lAU&n-l9&ix?FcO8R&3j^~xXWTA_)00sr2W ztY3QGnpc{0-QoY64TnT;LUi+AVwo6K9&a&0y#I%oNpDqm^)Y6Ag2*DYtL7rJcp+rV z)|1?7tS9kM*TtgCjj=Q}D!C<-o}$iVwG(&dG*KL4w^Qt@3Q6SZZQgMeg5 zJzr2DKf*esA#g4r$dW~3)Lt|6%bEwV@ons~c{Xkg5~=HyCZ<>f$+?XjcIX_JV>Nqy zr%5FFDMxVV;qywNo40In(UUr9*HXv@eD7?}`=Tv+@9BENP|ykNmF4d9RvNo2S}Mk( z{9y|3HR!Wu&(EXR^O^%y?7UzSCz*f3u14__2{N zi6+aqTe#TL@a-}F{xPmb+n(~yTGyNN?$zDX=c50azTyO{xZUA1xG}}(a#q+#I$Efy z?|4vj#vCfy7))x?(L{|&K_O10W&5O6MeGy1qr@IX*DxC*nWxG<{<`*;>PE#1G=Bqj zoH6`U5eSG3%{a}TRN&>TC3ZCE5BA6A4#6%Gj`#etRm~SsidRax(n@74ho<(rcrWi` zXHZ&`>7jT$4T^@vqZ?uq?o+#^(CLc|`>MgyX=7)6bL4gb)_MCkP{3(Kl|p7Wk?9s= zD|R7KB|aM$po=hDnJU8f2r*Eb&rtCYtTg2t0Zg`ns`prtC9#=VX%?`n)h>tL;~E`KS<)Fs?B!G^y@OsZH?(l}?S zwG4|vqTWW49X{#C10&&Q_-yvU1PiiG&=0Y)z`0T@uV zWt6dx8Woi*@KxTbeTuetxgJ_h&O+x#rqT#wwxG+7D zI2rd>c;-RZ`;MfAy8Pcy#fh-vp`o;HGcW* z-(*n{6i%3u7^zC3P6dPtvyZtlCvPR%7PsVFYP_TwoNjB6J#d`ukk=+ic`J3g0e?G< zYwZkrADioIGuK8L;v6191brFg@0LeTE|RS4oV%C)h;yFbTiwKy5zE>P+GNlQ%DLAm z3!O;^7v(#)uuX{`U=B95HvnJiBghgMsV7CE0Sj7Vo}9^sc#MOwZmxAFLgqkTc3@#@1al|Apwpj;WJME|o=YW>Svy2F>vgl7J0@9C&wIcRUm%YVNE&^N$ zA)nDflQ^#!e?5unyJ7?~H}g!o%7-yG$byzXMX@U}x!~4%(%1Iy>7*>4^MkVNLg-S= z>-GgYY3gr0ui!^H7%QyEwZcye*S9IgR>+u5;CBo0r(j=ieDB(}G?A32#hsL(A_JF>nZrr#0f9cW3 z!NIg5JI$u54@D!#?L0zV^ue=uJ5BBOs?fupiadFEKlwnXek>kIq?$1JLkYW{^^Vq~ zI(mQe+=JIbRYkDVaM`hsw`|Dl@QLQc1{;-&{wh~7qa_GxLO6+{YHE#!zfJGZLUIvw zSA&-8mYo$KQkEw6Mi*;krbG>aHtQwJ!$Ym1Ez}<4hplGJ4X1O<;4S zRi7x~^*f~lWm|HJ5uSa96sEmFxsUidMcKaDvQFOZ(*6{J_hkzf5Zx|hLe3zE54?-2_$bK188t8Obl%!nZn&G}W4KV?R z-ud>bb!b=Jk@hD zmBO5^Vw|i(oVV45dP{iFX9uild+QUOqjlLS&funKV_4!>AZH%wx?;*Kr*$3 z*7NRyB*oZds)x|X(~m-CiI&iuRY6K-sf`ak!~}&xlwqyg2P#$$&r(fL_SjAnCZ^3yD zoiORocHvly+8HRXmoZ*S*}tYG<`Li*3P>AEx7+z74oCE=E020SZ%vWg%~=3%xfQ!w z49dB6g}-EFx2XRcv{%uYi(X@zx7|2M535ny0)(wXYsek0e|vRfUjQvIZ|*3% zgjr)t7CdV8F;$}v+L_<`4*V1!Tw*|DqQFcJ6N|Dy%tKG`96r&}cT`(rCLTyYRB=!O z6G+nTqJ)hvblU06U(XDWrN3yBqBz@$=?+nM-Ce`Nb(>1Q$|&aNWRLIwdJiXK_*~1~ z%^>V^=nt-OP0Gk^wgy!P0j(qYq~5!np`uE2o_^n!vG{L%U-6waMCBga6{!BbIWkuN zaXenwcJ0ie$XUzq2J7nlnFf^Ml$dv5Tl*;S-~oL;7XOuBd^i9)+oHflPASax7p7t5Zasdw9|PIKXz%zMYEq`>Q~5{)wF`BUwg;~xgS1rpeR{7^Xf zXftHM&mUHoxF|DFRt03A-3jqRkXIj!Z*LEeF0xh?-FNwMEEwjsLNL+3hmXNTe#H|mf8T@nta;FZAb$9wrlD01_ho-3r z6figs8}c?LKL7s8DsGB|%;{24Sn5(#7_iw0c3@)8^7*xnY7JywkPCH6a!@aYix4Gf<)b5UJPv39wGiL`S zbS)^v|F(ZOECQ_md03vUL+btfhcMmg{wX*O)l9ilN67E7*VSF;P4M+;*#9y39K-qj zwt=FE6t>O0CV>t4U%g9QJeO{IYG(Urn&f6rzkD40wq%OU^~HTgl#bC;XJGe?@M4#h z5n}4OIa}`E-Iz_XI>cY1{fSg<}tpIsjU(KB|5I|3jev#|uLP{>L7_A4MZ9Pyqm^nE-(7{~vq&AN>4( zNai$WCukA%r0@qe}_Shkt{>jyr*u0{3E) zoCT0B0WSd~LJNcxRH-C_^GxTc(fMzN0tgK#BNguoPk{KAF|uF|I`2#vFNz~i5SDSZ z)!T+}wgO7hbqT5}+USX-gy;xm226TfpI|QIf);~XA{S-k(2Z(Pj zjfFwnkqBd(rW4{K<0e5IIMsrbj@Q;`(GtHROkr?2OPodI88aP7f`0qUHj7cn6#_&1 zB(>OaXXMZ30&VwKmTW6oV{`+C!R{N!&hG~R3k4!7yNKeYuyj-nq!());O%dCAh^&u zaKbcT=vV-%K>dLYY1VNgn_LWI1GBL!bZp|mcSsO$=JpOA75H=HOmek9v0J?N8%`04 z!2K2oNAVr-gLv12%F|u_XXJErEMS$79V*Jmg3if;+;KS5CAyJo*vDR|DH+*iIcEYU zC4hT;GKUZeABSS$S2Pw~;%LI^bf3`NJa&erDz@(LqgYKW(VBrin%t3;LC-=loKd>` z7KY{ar61uh0xhD6;Bv$(J6elH4P?EwK{IvT>fJ>>kE?P$f zV-2E~4nC)YMjzfVQa5yEJuoK=PY31)3GdD#s9NP1oHru$APF}opBm?1N(0OgT~7Yf z+xHx8b0C`v@yJYZJc(my`F<)iL>xgQeh;jmm}Kp33VA{DEaAc{+81%#+s}c_>+8Ng z01^~mSup#O8gvNdlcfAm^aM(?=RYvJ&dZSYpPg+4S+3Fj>&38=53qYoINgt?86u}S z??bl(hA2jo&-(CRsm2a{hjCn2=y$p0qj7cW9<9G+1>Q%o_1XRqI>+#7g5s(qEQqOL zB@Hd+gAwRK{9Auc`CnBY8S{jPddqEQMYJqg%syAJGs{}0y^{;Q!E9i8S6SU6G!Wp= z+Sp(tG~V^!(_3nZCot>}>`MK9h{zWJJiWl`W5oQY(g|}Oc?FGg{!itDl>t?rf41;M zX!T>XkZa5f-XmjAfjs~1r(LE2=kttznxU2J;i=6YxOVW3dS|&m_6K#@KB3j;$KN2t zSie}zc9z(%In?|S9Yx9P?ga087Yt(cL$1b45b@jbr?h=N{rNsO;L`aYi9PW1FuLe|aCUSh z=ffRB#H2X1nH1GDSd?UXf1;=Nmk0HhDHe%I5!CSuD4E^cfljoOYmgh>+w1wGp_ltL zTf!|+*AI9@+KbMUEhq}IeaYBl%Do-O(w_qP*>q^qH$|+;Hyct~V`C|=;I%KoD)&RD z2q2BZf$CN_d$FfVK+mHrFY0*Gu5ISs%H}jd*0jh`CxlX%#tWbYv9XBU4xla>7o{km zL4Aq^(oVv^1(xU6yA?1nfP82TO{7*-OuDEwE@j9#xMN#s>}snxa@tFJ# zyJ8bS){+wzj!M_>;}d7Ja@bi*BG%BgDGd~GcYh^0_1Qa8DF-AzbR|tt##(299Cnc4 zkVjuqJ7j{#F=tXQ*5wBwmW{%9NIx+Vrveb0-6}ECFt!k(1NPC_%SuyZ?yF^fjIUseisy3CP;Y$7G(k=qti$GGC z`A#hd#Y6C?#ZuITezg)hwf)}C~ z_m@p$sg^m9JS)KQZSp^FFZe8*M)gF!TEW^(mkE!o8g4c4QB)Sv^)jR}30|_4NbDzwqgI1=BVhe#BPf40{ebQMIBdylGX; zm{afOPBIXkMFbg5iNxX3{y}2dqnpxwT;J5*7bK5mZn2Ap%6FbAx>mc@R|2Jvg~#_T zeVfy~t2dUaQ%({z1%g8{nzkxEaKi^PmSl!|M(oO5IS*I(c!sXZcoEJ!C1*`fOT)=r z62o=Xcx3)qTNHq)x6beoV^pJ5sA$wnG1Co?`|C`Sd%gv5J4%AEO3&5*K|#_B2ua6u zlwhhVk!o$^?=Ywz9WEcNj0ax!{&msDDP}z@qwhygwZ+7xCAWaz#Ji3AqhDPLkXUkw z+#^=ggAtI0=Rry)V9$(UPs|8rFLE*(_}pux(>4w6%Q94NN|!^rG|$^QG~YB&s5OoQ zbsRb#+Er7i0tkf0rx(@&n|w4xd{Dz*XUM&wrcdd>0y6TQ&zih1je!_LW%*XhOBSdF z!m7CA8XVU+9^_hXOCHlK^cDp-b7h|O{ixy1dSQAi3hmV2`K%95ggAee2RkKGnA&V> zBJJdCNZy7ImTqj2IFzBMy&mgO|liP}dm1;9j}>FW657zH)IqPJu^ z)5owWGC6!6VzG^6c~2q9M4@Gz&{ACICYz8g8)$ru@hTUz1=-dfb-!_`v>0B9DEA`%vU^W zS2ZgxGL2u?2EccZGX-zmG3EO33snV>=q+uNV!eb_5*f)cq{C!XRD)%S!Or{kvVnSv z4UPu|{{#ak8%Kxbqz2F~X`wK*F`wG)|9%c21dDST4#S4Zn|TO;(fCgn{mi@!uS{CL zz0SJ}LBE=z4(1bqLo=km``%OcV5|VfhgRHyeohZh6+j?_phG5-y-V`!>u}Q(X5KF1 zwO*j|HddT5?mutXpT>3ZmtH*RSHnAxQz$6Xp#Lf*|99o<0Z;IoOhonQ97EJC6%(^y z8cB{$fiDOg7le?mKRR4TGHsD39Oo|K)ZYN(Z71%C$<=37RaC!Xpda zvD*=7)VN4mc9N~Q&G*l=rjH`DH(W)JFun6t4VW3<=}wd>!0q>X50rHwoOF#5RDExK z^m~Tx60Et(cTB0zv^{r8k1NlSjFhpkqYot^VkwqvGP{rh>78)`K-ZrRZZ%hrC|C~N zt+_t!B@CncaP?(L_mL;a3JktMDy?q|ltizYRc-6f9GW2fHFbX#qI>9(1iiX(V+0)^ z0>WUpSE)8bb61#d`b$c2Fd4zX?$-PFNf_Pf$5W_;^h=dvo9R{Ym@^=|zX;_PjWb@h z{`d;bL&GCa*jBT)<4=;pOs2{8=G4~o_iC!EPWeA1{Nlbzf~~$Bj)!dvWUyk7*SLg8b5g+*%ht7*U6DST`fjDWgUC6=BmjrdnD< z`;)=S|zIK|*5B(hf1hc@kR|%jl^pkdsf^WRu zt(I>tvK;`OL9PAwDt7+UF}jTJav@Q(t+B#o^5}1M4d-_NRMs)Din6GZz)D6x6I0#nxl&Ot su0Zge`9H2CtBSiuvJ zd>|ud0ujeIH^<|b|D7x*leV{2q(+!r17XM@n!I2mr~+~A8@UaBHAFj}kKLFhgZsmw z#Bv}proJXe3Dfi8aoo_uaFdM50ExHuhsp)LhY-BAMr&Z~Dl&~s(a3FFuk8sA2MC5iU2exdPVINsfG5y{Po%iC5_o-y!w5K$Q0q)M@M(F&=9*YuXWH)H?KC4 z%-t`YKa3?Wb&N{BqcB#J-vk*h$6UX_V_bPz;&-@?@QQH-@I2kM+wh(nfwti<#3Hb1 z2tFqG0IU%;#ebdIDZV;E_1>6MwYV#NMV4UgV&IofI@KU_O2P4-|Ix*5D`sEHfiFK0A`Z)VE(q-!ReZZakK;=g$l@EZZG1>p%s(!`kF^`c z%`M&fZ{ip9h1TE4!iGZ?(re4Hg&b<6*d&&31#qfFdQq)0%>e8V);``btIww zt0%-d>b%E$Jg9?z_~}dSj(_`?-bzE`m~lE53&9+hoe@x9DqjJ-p0VdV=-D(pvCK4*f6r`@!Curd5I{KbX4P`y{x3ApM2T=8m+sosrU278lZ#; zbpCvGoy7<@dg{5~T%&x(#}Q5~t9UE#yk#0IlvI`1wM7J8%qV4q^g1;Pm6(W6HM1U; zLN%&QBzaE>08y&--#W!!?S2!y{-is_z%SA3RZMX0_e3oiD{B9 zKYYq-VfZ4b*sCWHZoRiL2~R(uCj*r7f24{cJIi*pItva4-xtB<3iK0B+2Z zqYf-Vm)CG{9!pa;!^X^j*904)Uuq|J3q26v5Zv>AmR>`SimQdFV2lEgOJ4E^%~6~! z#GEt%TfG9}+-Zqy7P7cM&xcyr_tseoj-!RMF|xqy zr5s2UpfF@`%F@QhR4q!Erk4zcju-9JktI4%^G4naU?0|~ZxBQQrrMH{I&6NU72g_N zm<8g8u3lq6y#=}9=ZHM`puyjj|T?QF}(fO-ffd=Ws3Z#^F9N3 zw6v_dqgl#rQ21s18z8K@`Uh%+chPKk^jWyHr=D{~CTG z4l>bP_Z6CywsG_K>Ew7B81rOzYxY2DzijM=US_PQ?msx5iD?WO6tslc0a6eA1xzP9mP6nZ1gLdM_7HH>5{Pau$AN8MP<;&5l#K@_|qB4FHcYakPM_ zJ%@Wr3MUzjZL?Z zv7KX_ShRL*_rQZbQeBSME*Y%6H5EDVGAuBepiC0d#Yt)10WECq9qa_prO@&DsNcuI z((sc=NEJoydbYwiH_!n>9zXyN%?vqpM7FS6Rtj^PS7#BZ;mu#3BZM4?6M7dr)-eir z{_Hcm%&*>OjVa)tjsb=Txirp0XR3@>aQQr?Z>Hc9!H=kbrr52(V&-44aJH^oc?q_D z7rGM@$5(jD+V^a&X7WjccUkA&<$RQZ6tG{bS9{vmP*>{@;>9?ceQe)K6tecmd8*77832Q;%kh^=6D=Z>q1 zU$6zG7knBWe^K`TMjYQCy3J&~*FR%VsJ(JxGtYR|QiX}(D*GcN1&H|%J5{)Bsk+@4 z^sFy9XSpUbLTc6aGy%}42w}}>G-TO0QXWNzD_kAn^u!F&gR}ViJ+-r1##P`{{H6yu zQRp-&Fp#)T)R;UCjf4YTeB*y?Nmu0viL1W9+aG!L?`g%ph-tzUU~XnZ=vWOz&MZ8$ zEQUG!fXM!~v#?fiW=sUceH;%aG%|_{Tg{lwcjJr`t}j?J3f0nxZ#zKOWVJ z>;?t|frv}QCq#F9_?|kMb61^r>hW3n!`azj94)VPt6?D?XGb4!UrWFhAYKS%;AbR} zL}s$*Yl^|$Wq=-bW_jZqgu%|h+-GB!!VzVCcy^$6SIVB`&}C`0i>9)?&*p;z*vKF-V5^55ivz5L-TJq+|XNMrHhq8VVVm+ZdrgNZq5o4rX zodNNhv4F8g8H=}`G`u}juoE|<+&wI1*imZ7L!@xOu4(3i znz~w48+smgXE5$|^TgryTw`ejCU^F~H1Am$+fSFYy7-~dRRv=F-Cw;Hb2Gd!-qmL!cyjxKfD5eyhm|5yPL z!{Ycw%8P9;U6)oBlw7PAL|p^t31V(MgmBiq0&kkk1!J-+GN(GJvG*riDZz!ZtqFa7 zi8c_z#K@})Ht=u4(=Op=_~tFGd&sDYe(TXA$HTnChjhd8zo&nTQf-FitlygFN($I-4^b5`+wP0)@)TXqIi|I3a`c=B>BKrl zmUd?16lc|74^4@@BnWbd%uNQ(Ov7kd^5`e4==hO1ZfpVrGxmtD!J2Q@zkN8B<&Ym2 z2j62t@SiFp)EL}@gO+a>)SAMI`4{D5uYiFi*SK>7m+H^zx)m0%=41UR4#ihDjQQ&rEI?CIi9j!{&JM-%OHq8X|v}+h8m4f6zgZ2o8F~E`V zGwUOS0yndCU=i35*I_q!``e#_FaLdP1>ZO43JAP|OC6O({n=8Dmw^73(4<;Nbf`sl zxwF5)Qu1SYWb~dQ+LBp$h(~?($)#`4x*?W)hdeO@WQ-KB=7>8ckEIRm0@cbfYgb;$ zIWt>YWYxzS&VFNS^PEI-uhmq1hycM4@g}i3}2tu5Za|e#%y+Rd)aCkhxH*}1hYRu;v zkDX1QzPtTbWyT*}#-DLB157g27u@}C|7_^{5OjrK-Z92bEy?3%WDD6iPf+-N%%i~D z1a=W>gD?2ec45wTjp`9Buhmi6km^q`+_*9q`k9dSaNBGswzg%ot^(PG$8R7`Vty6$ z=gmq95Smwz@5%P!&HV8{@sl$BFR8YFosf2nEX>@Zbw=`&tp#D#A^?ISLTbVwSWhzY zn&3O-&=#_{<0PTZBl{M3-O>iJBNhz7g=WFe=rXQf-eYtsLJb5&qO`n-^tCJs%_SA1 zoEJo$cyqf%i^I0V{(i{0uXuyba6WJD3ng!J%xWX2Bk<8Pi&$?!dd?M8M+zdu`;Rz2 z^JeOiZTAN44I?Z{aDYVx8D$?}MVR#g=~R@5Dm>aCk+(4(^$vq+cjST|eP1k%fx=ly zsxHy8dKg%~jv}%=1eNYfJ{0l_Z35z3ai@qHM$rkDg0uo%>)W@&bg!e}wLZ%wk3Re3 z=q(4{e^l>K*%B-c?QBi73?2_pBvYY-KtuRJIcsQbHho=bI{+=es=mb97%M^>Q`K8Cs($pJ`U`hP6 zy*b!(urnq3B*45fiYCZqVGYw>k7uN5nHp5khMQ@JWwX9o_bSu`<+%k@;VXS1WuBH5 zvHv0@PAF4J;2~dI9z`X%G#y_-_)DJoI3ikuT zBRxE+I29z1Iu?oL-_QgPDk{ShkYx31ZVeuE8High^-Z=0kNan|O4NNrhi5*zrFe|U zCqIto&{l+){+*g*w-v_fF69z9&3qquCnRjxH5-QP&@Hw*(!>}$d^MDe)M~QD;f9GYj?t1`%L%Td*GKPkvr50*-yp{ zC^0r023~O2Y$*Sx$8<2=Y?HuJI#|?wwNbC0k=}4@1Ed88&KiKH{|GF8E!Ibd>Res| zts*gPSP1JTCxd8zX3Do&pe%BC`$JF=49tgqB1x*fPr)m4Ofai09(i-h@mAN)=!eZ= zomtFT-!IBmsuZVsY#m%G@&jgEA3wT?p23P7U!W?VzLjXp$W$L8+NI9yo7W9FEX?z9 z?K7i$MGZS^lZ(8WrLj%HxvJCxCg~Bn#!*+9Iuy!{os-U!WJ?TU)Ss^ZUk(u1bMzYf zoGBP7P)49k99y1n7Sax2BZ`qBnh4ZBrb0(G;o6DYr0|PdSN4Wks3w5#Gz0H{$;Rq< zu~~jkaT7pAc=iMy)P6w4t=`08G-5k9CK+jLjW?N%}(b($#9~z2-c-Ztq~rR zE%p%c#`y3Lt~*r>iqsk0mjc8E(JMgPiTpwTC~g_kbn&oB;s~!?oJoHQgCT2GekV%^ zUV>|^jP`qP930e~>v8^Ho0o0^e?4<(>e+Y8qQeZQ8L#0g)!*-5wRUl|F^nxJoM-quX5uQ;|=4 zSF*v(6&Jlu)|(sZ)YX-2cIh#$HY}IX&O*B@Noy;Ye!%q{f(NX5FGcsXLBC$;Ry&+5 zYJ4ed6nrqla{H&&O2!(sWS$@@5Hr+SnLsuh2_bmrJn^C$;h>Hq(e#XCY8=47@26IA zw&S9F%D5kV**Eq_x~-VY{{es0u)i;7&k$8I?lhU=JgjmXD<0PBDj&(Tf<`5iUgjz1 zkfLsEB;%l8p9Bo+Jo7jIt2lW~vpi`=f<$eU5bty!xfv$Yi=$VqYV&}& z^{Y_!F$cVShr*XU;NW^4tQH-_9a#k}oMX*XS&xz6Qpxwnc)L}N!jg_I+Z<1iLCkxx ziDEH@k$9}(dmk=%6Q^@HhO!Q|Gi^ShNEGYQg+|h@Jf!Kh;o+*DN))Vfv>pb8-E-sFW+M*SHW+sv7A>gr3$h65Koer|3F7`(Mo3OU+S=Y^a~3C@y=H|L#Ff- zx&VMSzbDwS7p$7O5fpF>ow9f+f_cUL2*guQk^1&q8W_4O>?|V%db?|a&)9D)Qn?2R z$TlNv@JH&z5|y#aKv?n}D8a{Nm!eHGn*-oh4}(uZu(QhK7U(mjq8C+@X2pYE5!3u8 zS10NM>y3FR-gQ-tncfG2RuyW+5~|sWLu~=siWnKuGM;T-G<9X;{@4vDE{n0Ofm1j; z2)lIF(#;z?mX7UapYKy(DJc$3e4Mwx#x_&QyD>oGRH=k@D72ci_wRt;cg6&tOMn3q z7X`j2_(&2M#_9Tj>js-Lhb!L=PnU?NBQF;JjTfwrh!9K~j1e3ik@Pp_vuP zZlP=P;L=dwXoU!Q4_ynn+~SPBOjj0C!e*B%B9T8y)da454pkz3)hAwD=$ZNRMxQyo z&oLS>kIGo)r$p2C;=O#5a_kmSFJPkPML;|U`7-S`K7ZB&5<}V-lD0*LJ$uokSv#Mu zts1O15*ybldoe4!g9J?+(0Vi-CF_!`x_a8wMBB~B{d=qXsU7fxEy3U6XJ`?j zi1&3lkxW@QRvvYEF8CYw&+`@0=KiX(G(Rt%=j9xXzETTjB^9ns600;_L1Wu90|a~Q z;q>t9`Q_wu&2X$YejUX zxW&>5YE;ABfQ@mqF8wajJB5hYhL(N7nMLK{o=gQ5M$MD8s2=X)0JxOC*(|2lzPX

~|IJQH^*p2yh_PBT8*`$P8{)=|Uo*QT%Y|ij-d1m-rOvj+2hz$jJ>!JRj zED}^#X6Q>d4pE?gJxpryLSFg&_Vy+z=6GX*V`d0q!W1S`FlPh!r%#$um`TW1KyoXJtRP2s0J+J@MOOs<)c^kcl~cSm zVL3QM*Cv<~uWRBvf|SPj3@!hUU^*QT^YKEZKB3L`I4g~6E4~FW6wkYiEA@=#^jMOc zH9#HtK(^OAB4lc68b6}S8xyyGbo3bZ+3g=j~z+TLNxR;A1R;tOFVq5S# zqZs=eX)Gconckb@AcaQqp{AN^$@9n!1kgEMY5O{S_ zwTn9joip@};1zxH%>n_)2Xv%ykXoC?yn!>`6qPm^WU?`Jjlv0!*E`Ne#6tvyVx?(d z0omfpp#!-jA70`xLih(G>$*hNWJg0GrY2=HvO=P%>@Ig(Zv@XWyroFKEQv=%a4sY- zT;wr5Tp&C`obXM$NN!Rj!_V1Mp&5|c6Wa3FGy0=gc;tR7qIm!i@-pp!+(@x96q#C; zdkt9%dNeVEoHC-^`7KI_0BSM-97BZOA@-sc@w<*u5ah$svSrX-xwG^FG*Ajjzn(D$ zE|DvfXO&iOFq8|qKe8g*i13Zx1I`w{vmk5__T5(UdJR>PoB|M2RB_!9XiX`*Qt6IS z9#WLnbc9VJM-w1O9x*v7Of*84m5sh21^Y@M($BMJz9@fgu_I$v>&f3<-EWgnmk8`C{ljn;<=URy0>}48Wi8RhkQnp|Bgl3*LCsfom#>*PK`VmaT%87VCF#74!a2<_ zll|67Z8$)b7@}>UCkQ&}-5Qm#hXw2{rH0tb(NcDr1&}5Y(lRAFHZG`^dB&2B#!?es zeLh*kjr&EAtE>oCv5n7F)146k=CKc74P_IR6iTY*lg7kG%hr^*z)%xS7p&1H6e4V4 zB0k3^c9Q4{sqSha@*zmxK7-P$hZ!tnScy0L{5_!YPr!H~(W|9py$XnH{J2CCW2Ie6 z=GF+YB02@1qbPf^Lluc7cHHeV@ShuoOLYY@1gcBJA=fxjKCJQcVrExV znkXQkuXRxjn*Xf+>R1*pQG#>&YfeHGSP}6~vGKMkbZh91+(bX@Uh^0Hg2dCg>=7+c zW)x|X`%_-*)Y~6WrV?eXYeCR*Ig}VDK{8B4@0!m!b5pLGOI4HqTjXq?xdT&7WoDLk z+M`!`s{X70nml(PVm ztG_K82aQIUUYyMjHCVYVV#loEAC>(?3d7ywVg}nL@jDhcJy}R^g>V$!51TWBi(E3T ztcUs32_H6VZMzmRb`4U!8a?nU#665C(^F_vhPQcHs2Lz|n7vmaB_Gs-(;emgf(bZG zxg6sBUA>6{i0R8+ZF->w+Oa{G&2ID2oSKRU9#uoT(46!I0Zku1tLI^~j-hsptPDI6 z@rd^!_wiKlry_n_z`kBMu<8_T`S*tW-ftz_IR?;`THG`au;7-;+XqtIxYYy4|>P z^fOgPN|r0gDj~4CIiTMDmb&Q zQ5-U*9A&WR2Tu67zD$DY&8q=}uxNilu;r=`5?BD6nM5lOyK?U1i?1O`_d zEtc5mguv~cJ`yPz9CyhGr2-q4^=*VlS@1QBq;c?Qy!Q@5P(O-dupJcQpo+$E*_XA@ zGTDk*sM3x7+Jn7vhmq4Ki z&_^l6ONvpP3Cr90omq_cQP#zHrWzW~na|ETYYmadWS2*gK76D8^n*K%LLR4J=@5@e zTT72OjY1F#()Z6w(BQc3|4`3BvZa=((LTb`uqr>0uzsk7L=7RwEdVr+t`S`KDCj-- zNFS)7Q9l1AlP*_YGTb#$?j~iS)LYQVCEQg-7Lj#MM?&E?e`FdQAFa~OC0{#vVv^*) zp#0g0Ub?t_+9}8h5%hnVFNSmVt>we7NiI(Xa}F&c<6mGEt23@DQDkSl&XCev`_#Jo z6&20W*g~fr%3*QEz6N|8!ySs1cAp$MSY4L)3@GTU$TVV_R#A`{=WKq6tE-7=%}``! zPJ<7~t&w#Uq*loYiIOi#u}U}5x9QT*S93tNX&A?14+9GX{p8L|xaW;=ml51SrLkvx zPodI-pyhKk!$+pxi%{KGXTxJxrAWo??Korkwx8Y+Wes!U&I4-1E0|(z+EvXEiEwL_ za|ohlI2kYge)ragdb1!fpW6b9f|m-QeNF_An;_}h-G0Bk?~lNKR^e7j@Ae{o`FT(o zP#`{k6=u(nZSNIw^|O6=2j;u+r0M$N8~^O_uK#QXNR<+zW#;{9e)!0yGiEJFY{kkl zzxaN8(PjAYK?86lD*~%i%BXmxaO1W5wM>rS`GqyAt$394I*crT`s`iD6#|7s;L63Q z@(tH#txoq7)$O!I(x#tzxVx!Xb^p!IA`wle6Klt*qZOjf79_yy`LZ$MFvOs`+<}d4 z8vYMes}xN+yP^xxMX|wYZgy?EEc5h05GTx5JjjHtA_OGqy!F#TrOv_Ret`JIL+Ug~ zc-J`Na~wR0+4%ZTRnH}cJ6+Jx^4+EofvpE zHa55&=xmVNchhcjx5(5N)YJ7&L|#p;29jz+j74K4Zt|%8RpeT*fR%8ZlBK-WV^YzN z>ih$|dXbqs97}i;b@bD6J)bt0Z8GU<4CldZ8~|5l%^gtwTu3bES@+uy*&8Zyr*IGM z!s74rDaUwv(Vej|Qu0y--di()ti#I6sgmu7R8|QP#$c9$=t-ByNxmS{Ku2owD+a?B zQ~avH1?$nf>Gr@L=-3P8dYz(&lhxe0(^WfXUEIxb8==u~l?~Zd=VLovF&{kKwWiIe zGXTo`FBe=}nYuQm=kW_FB;$CSTzR|JD51>Ukn4Ukexi{0ikb( zgqw|7Eo&YQHa~lirKVccHm;UJ*d0;keY(Lvv4rvC2zSQD$mt6~svk7PS|6cX;wmqtx^y6$;V^BX57-K zZaeqznte$hPBlC?$}3dR%Hdb#8a2eOKMmt$pbMd~ooiDd-Ugm^2}xVvcboLFb^yu* zcY{qaTAh-^dx;NGq}^?${$;HHDV7S9a!rhEnp=H4r|(*0tHc_^ zO2gn^c9fD5YNn}T!)zIidCt$7BoNm^io$?LX#QP;aKBt6K3)x5 zjv+b7r|A$QYbcDPBm!?Yf(f13{ts)4N>QaW6fE%I*oafnkSiB{^p2({=dTJSyAPv) zP9llDYu`9EJxWBxFUcFyEpkLmL+qzlN(jIZ{jgXI?QkyGb|De|R9f0N=^cpMfC`ku zy>s3$JT3cDr4>#D&#jVjKB$;@A#%DpbxyB_@2_*~G_qMsMw{ce&gi9p#{{TzzCt%d72Y~{?ji@Hi~&#Q3BAqizs!*#S*o{ahP6u3A6xTvC@!I#+kl=FdfJ;-jdZX z2fV8Vk_;8*C*uK8q2%TXNg7<^U@0j%Bh~Nl@HL12KI?YB(lP#&hvn4TBzG*Ama|ge zoP5rnW<)4orme~DK9zCw#)2NayjP`-r$VuTs!KRqDWQTv^4peY1qXl#VZp7Y!Q){d zMoD92|8C1}`xkTwDo`*I2r7nNl1-a?^!UfqW0oNfvi$Hc{AmE4O<^h_p;LSN}3fJWc5m9JFBi3dquh2IrPq3uK$ z3gKqYuO}dxf)bzd#11gCIi0!4E09SMn+;!yg_N}{dq#*a2Gq`RRTtdv#;`C`=~HZm zQE?jV{25DQIcr7{nRLNYm#ZRf2TaTwWo6Z*_GqOndj8>yAv;VK+&s1Ps#qJP z_$SdI4=C1Yz4#a-^V#c6}kQpISW$KE)4f z9ptNTK!wyS<0_S<(JZOPeraCUvUp*gcpS}`daPoymuYp+lWtx2MLy?(MhV-kOkssZ zcBEDIN4|}C;TXWS?kuxr0mbt#m-G$1_u{^mfpIf=Bk^q+6jNXzTQF}%Js0a>FAw{$0s6!(~2Z}z9r*_=u7kPVZZxB|(8)A8_c zrLf(KqTK`;*34lHLd1WOs@bu#msh&2UdQxw9FM(ez|(+;FwZSTJIhrMq*VfYyR3>* z%}-8I&ur8FKmFMWA6It4E}AN3O>ioDls z5oSgM!bdJM+hNeEDusERs=M|mqQHUEd<@i-vu{hSJ*w7MW$>7?7sZ5+6Vrz-zyK0GCaG zD-RYtBZESHdYRg;jwF{w5`7oPLBbn96}gru$={c0w|*bjcOYV@XzvvMs|Z)BwSQ-` z2TNf2{1;OaODJmiTDeN)RnNM1GEB(euNeU9!-IhGtc-HlG4XjuiYmpMbzWIn7H!h+ z3V~Jc#*_V0f{I861+RYrag_A6i6uyPwb}5^aBMpojVXJmmg868?U{@WkZ0e?$k+Sl z-2?GKiXF%1k-M=)j(O{!^$*`KQ=qX90)3ncyLoVQ-P)zFXVnTi&)flRFRobwBU^wb z7l+p){H@359v#ic3j6QZl4asGuc-}d>=d} zHrM!=xYZ~1R@Q%oI7R*zoTqvTMKJn_334kc3%R-New8=t1~#l|nFIPmjSs{^AjOP- z2XvHc!9ahNFf0EPi#IRS@1cOlD2W4fWhERxF{L_8X<|Tjw#BB~P4#7YYv7Blh|kaG z=?l2fB4No}rW$xn0Nu*MfjI@J8-l121n8|$usBaGRc&n*J*RyNzBg%zxK5J>-8ni} zpSo1AF!S;>XNO}Z9#~`Gbz3jRdAuPOT}|9 znVSN!)2$_%Q^2F)z43h3)RiZhqe!vmCU<(hAD1V)y9Q^8&V4v%_5TpFAg$TnSf;D% z5wgPi2P8_DNFSEVmc1Jl@mmfmcR1KMlG}D{jdHdN=D#04PI{*K z51*hv;mNIX6&zEG2=DYA%v@m=)m;m=OaR=&xx=TBm;SaC&2NS1egTdnPAGcQA>lrc z@4}i=j^`=73p>pDZw=s9)mXlDcMTgVcjJG|Zom8Q(dn^A&+-JwaVPwcpj#OLZu081+g3uqbs@E?i zHc2b`s5OBf&1!tfKQ|(-^E7OJ^XjfTu*3IvNRfLb3Jl|uEw4EC-Mp*g^eZvnYP(Yw z5Ade=k*3GmL}wAdyLkW(glF`THJpv3fmce&;Dz1YAIpgb*%=IC2aEkX(!uzFpBIp? z66Tv|_KTw~*<*X`lis2&;L~geH$pDR*}$LM+)d60r}{z=!#r5qk0Q+3G`qaZKcbPi z?3QjTQGLv$RRI5KLfn_GRm%d#2K7BL)4m?c1x_n>#n{8&%Pj!kxtY)o@fA0zs}uZa zu-I8xJ-iBQ?Mbx9IfBW%K5vhbyKnIgugmt5>EL^QWSm8)q;=Tohnc-m66#b( zD&rY*X=nzbl+)m1lL;)rs!bH<_tTxi>>hMmUbrF1&}dG?hs2_U|Bg#^nM=PW4z)l0 z--f<@h86VXoIJq1eF{2LZrI(?MaEW12n^R+VKdo#icKL54Ent$6#Zrz&be#h_-^8S zyWQ!Y;q49#ZXv4zo|fG+cKGPYbI+M?23N2nGIWSPM6E8z5<1CB;{4EwU8$;~EKe&~ z$7>ElG5hm#Ku=t#q^uyCxuug}%oPRMZu(_2@&|7YbD9 zhLibZWu`0X+gx0_{*&8}zvzBV;(9uhC8k|U{ors^-8-|sr-RqIPfQv&R9ncV1!_d#Zb2`x>A^H)h`rM6++ zcyhw>-N5_9K>_2^l#f#9@W{%2Lo*75$0Sp|>WK5zn{DjBiT9WL%kPEzA{J?ED>%5s~X?m>v`h7Ul_IyUwO{r4mPcSZdU!t+5Js2Ya* z5U0N9$>AWaUh<#&m$%c2i+9&>%Lu{`z;~@{Tf!yx_mg_`88L_rD6;Fi4x1HPT$)y+ zgq^;w*Ex~S3<$1&0tCwpf(Wcz?Zi{cfTTWA@*%louJ+SS5fR8+6Q#HHqcCYauP;x) zzYegsQK>08c>poYXwF&-OM`!yL5D_z|L(r*{K(9)Nc&`oggzh4IPMpPfE6}35lAK1 z7_%y+os53Gj_nN-6e_cl8%kld#ta`Y{2}D-3Z}7juf#Xy7a5bZ1{*|xvX*%L)W+-T zp^K(|I&6$qpwXF#xu8H7pOM<;)o$YzZ2+KYaZ?B?!O#i@N}>+245%jzrWv&NBoyi1 zHDh5PtaN39oU8$xi8+g&U7BUfKVh$(w?dX+1C~npJ4}%sE3xmh0V=^mt)X1{=u=OE zhiwDVB|eUVuCX0x@W~p(ssdD`p|hfU(8qLj1N_&*@IjiSd;TUn@4TFyQD$4_pc0^S z%)xni0u}rS6UrATeYJ>#Yr2&NuM4$9b;vE~aR8@@GUHS#0_(7uSNuJC2hL!J_~ac= z%(?t)E*|3W1~^WyX~8r7khQxRvK!tsRn)s%t=$vtZA7n&toRqBHj6f-G8Z4TJJqfTkCR=d!ze7x#6P;+@bpHwz+~Pp1l({K3A#q5zz_(d!F_x;NF3MF~?XDZ()@;SI;V^tV$)#NOHu%sgTe*a=gsRxsvXbMT>7N zlC~%NzX6$5NVfv|p`UMG^#QM25W>mvpFZ&raWyHrJ#VMpW`>}UAgtumXqc@incR=8 ze_MwMvvypaRcb#f!%X}2!B3S(y01V8CF9)T8PTi|2Zs>a00Z6gZOCe^kTo3nvLL~e ztYG=K`gb=qvH9@ubN%;B^Bzuzt%(LfK-OK>#ef5IR&qzI$@zA( z;KK&>;Y$Wz&0l#ZTvGZUllBKLWnLSobF3ejTVMNBECu7*BKE*@i2VFg4g$FN_rNOal0Jm{=~^4m0F zFxCH6ik8=U1SNPe{ZPzvc*wUCTHLwjNrcL?Go|xaas$5c0_--ri1`4)ym|8z3%k_> z&?`w%21az@L*t+d*{*=Q+hmx#_TMKcW*Yh2I^PYfxf%Mh)pL}{1mc0t=uHv)w~G8b zXbj%9S``tetaE!4;5BK@)%q7AXyl2Jizvi`vI1?__R`7rl zLkNM;-i_wYsc-2p{mj=#45BxVnG63C*Q?!EE14F=JGFuyuo$01MXHZ=saG%YX?XYa zh=v!F>XU|tIEtzwU8FB3{ADobe7$1F-)jFp z$WLVW3gfovF(6~4l#B6jnYFUq-pzEvk0^OS$jR`P2QTEWC zK8i^%0JyedST9Ccp$-^&)2nK3Oo-oq=K!r%9;!-_L6NDry-@*C4Z=@U0%c4USxeEi zl@$Ro%DX)=I+%}mZy?jQ!jD1WCXr`P0}DuYJg)qnza=vC7My^PNK-ZUlpDVa`(vLu z;7ic08dT~Gu#&8?m7(lvf1};cnd@AK@i`oa;$ zBX25-Q1`8(w8WyS@aP1=BIA_raKit|0;q-wJdF7NL)wvVukEAt-JkKm`~=|*OJN_d z$pA^UY8}eR9ndrU&z;KF$icYr`bx+feG^S@<5t@uSA%c97{ltc7F}lw=4)ghQxI3M zfCd@xEzc^n%G$%E3$CCCLt3b@VV5(dkC;z1#IA0Xj3OKr# z%qh74)bv_Owghh? z(5pf8yqTa>Stf3eotX?2k*wZD0pP@@kbg+O&%IUve_7kd`HKD<<=Di0*bOp^)4$O* zk}$IlKNTIY82b!;8uZAi-oTw+Q&QpT;uC$PngF=Q0BFd@9Fss^YZSJ+TkrCl^P1Sy zIO~IW#C4x%DiTr>M?-&j)hn}hNfcPwpYgsr25-B6dM9UbNV6vAdsbSf14yPX&t_~k z=|+O?W+=Xq3eTtXRX^3?d$pGC<_iM`ql83EWTL=5HP&}H#Mh-%a0&(SQ^eKDhx|g& z*>c(|am;`WbnU8Zo=zCX%V)@cO!l0}^VB7ndZsd~2CmUjw9L_FI;hz4GFSO}zlGku zal%-2^#nNm;EgrUCp>d<0EHUJUP^R-BCSa}W=-hhz8Tu~PJO)cd2{6h`r2*mcaaY8 zPxAAJaXcfPbmI|%;)ROLoKjG$QNryPQB;WSt2>ELdITqO@HO|Y^`9C1PtbNgoX+84 z(!XQW85}cpjyCV(vn47An;a?n zp$)csj+DsLaE)XofLm`IZZqxlK1vSdKj?eH?22sT8r3z=0jINM=2V202ilMxhJW&F zHO0A0ezu>qaWdiH`fYc}LoZPSkT3}b{S$Q<>yGh-J_P$w{uu>n`Jnoy%0orTqi`48 zk88LPH=yeA zG9X7d;s*tX2P+u+4-jnFH_l+U0|$vMjX1bof$`qjQ0OE8$;CjIi+B-3Raj+fbX8^) zbYQ0kH<-)DE+BEdb$=QeZaJL2*$Rg)(AA5JQ9*muYs~H=hQWKloOk@YHcGD(C3}F0 zVFU>lNEHD#fVE7eBsktbd_Co0JR&ek71Ge_7nIduHPD)#A0ym6Wq?W7a}&3TAw3hv zWJe$ET#y$E7vyOP{sEG{&QUH%6I6k?h}_$Dbi31aEyYCkji(058}2QcWWu-A-KGNJ#U_N$`>yR zZ^0TPf$31J?mp|`oDj2tFcH@n+{;r;Cr3q()=YhrG(J?3fb zfLPp+wT8iAPqS!m%r>nae_gXlqb~=X05YuAjZ{8Lc!Yqk8N@l=x-1fOE%!h*1gXmr z(4karxRbkzxO#<+#c(5VNaUA$qW6h;IZP?(P<6@~Gsu|}ESH%`^wHaA&|FR!=WcVy zMPe$Mjbe^wP3HKUaZuYU(xLVq{s=?P)2fNzstdqRd=@KPMi)ON^y`x)70Bmm5 zQ|F1Fi2)tvr~6wVl`ZjE(VSiHVV6}=78ZzG8oVs=J+KqJ5pv`{w73+gX#wEG0$e>B zI)(S@Bhi8bhG^9oyeVs^L$nQ?J9w0sO8+){a16-#;rtIu@gYUqQDMqYezDkL9N1QX7 zfN#QuEa5TxqY?a)CpRHSMKRIl^&DDe!9J z3SEHO6#gYrUAWv(dRn}`e&e#gV&~gck!r(F+9B){!>d<>26T7kDiOL%dB($)L30bxz!u>ukmy1sLb$v$6_Ubt!V|+$IxumnN&c#tE~MJ6 z^Q}p@=CsAp-H=cE0!U9}CH`?9oU-BN=>*0+5dqH`^wWE649^U_929%v$@LLO?EGho zS6E*^8_8gNGn9u{X1JbEllvS$ypj~RXvO-&;6c;f1+e4fT4cj9Mg;gr-1s*y@HmR5 zS{H;LRKO05E3KA7G#q^d&c~m&5BvF{FX%eBO(G(Ap{rAy0N}_PxFuPkNRL;|ywy&d z^=o5eG2$zFne+5MficE9kQ*zv#vM>6#mIDvD7~qaaV?+4Dj0t?v*()y0@)wW588ru zQ>bpikB75lcpJPrkPxNZ>7Rgo!Law^ZTl`_wCI|llDrIP>N(JjPNFM^j(>xL*`*4t z@V_KQ$A4j&0J=(5y!(!UR06CtAhc^ank4@q)!d~a0;Z|IVEEDRd;x{#^S{qPdqu0^ zy&F`Xbk~rnS}{&By=3%MK@#^B?e zb;Nb^(O|z;@3(**@2CT*FJVOQ`>Q0v4RBGL9zzq1O;(SqHtK9z~>>ZnP;i7iS*tTuk z$&PK?ww>I`j&0kvZQHh;?AZO(TaD_jI;VfYnjhD!wdxwzVDa|)d0G24=l9M%HQPoK zq?YwA0ydAmcmCWUYWe79wL~ilF){{^=iTN9n`FB}?Ix2AOhhOXACS@3^4Mm7f@Vr- z6gj@>M<<=rP65ZtntVuX@<(!lgjbwsrwa9yO_zh2Y?w-$#I!SWFL-gJGB@9VI_v|; zIV8DlS3_yZZs3ihil`f-aiRE(91w3*(tPXC6@xJ$WgKwX7`nD&bcgTrj+m6zNcJnT zG4vB)pFC;~T-TI@ug&LRFX0t9oAwh-gAJ5u2VeubF`{}M;2#N&hmP+DeLsAR`CXwa z9NMWZ9b_o&WHv?3+&}-06Yzz1N5TUfdAh+dw}_TRmib9sz0o&m1LsWQhRmd}jh)Sm zZ3ZT7%S@Y=bVT=$u|MfkQ#>((bKSQ{lh5!_ce1o_N9Gn z97?-L8$t|AIxyW4p9y-9G1VaC!wQ4)_il-|ni)?8ayG)ur&`aSREViIaSZ`htkvgs zAFyNXY;^bFD%*xVUeQmr<1u)}HP%?RtZb9`V4}plF)Uri(S8LIvJE}9p(!_`W;{Cb z$`*U0%AsuLfBy4R{2To5e{!S4s2<9w4J1;d8~H!Hr_nvk1K}_+dkT1mX+nb11tH}= zchRtEfWCX)c?~;zXN>T-9sLRrqv+RMqU#rI1Ws?QJNBK6;##w}JT(q~@yF~|ft#Sd zDBjOUy8nqRH#1h#*6gP_rDi7$h6tW{yBgyyG6m^6BJCS(XyCH4Chn@_usNvrumT0G z28|0(0f$WrJ_xB)H~u3WqQP#*GaviFm*i)z)So2s<_wXR)8lgCzu+XmOMHko$TGY^ z64Xm$hhCG(@t{77Z|>H4G;M61_pfZ>l*j_q^|WxD{vVy{M%&VySap`(IxWOMqcIIS zza4kc?Dfk&F#c`^bzA?LR^!?P%zJgQzj$bHLphhkYQ65cd6?)$&&}6ZL4lBx`_3hc zQH^Mk?13VLbqH97;qfW}yq(fD)o)4N1p~ysx4a)Fdd;3pGg8CTA4$Ch3a#7o;@~vh zp%s=v&yE|<}Uk5$mHx9xv8vINN5Hl z3J)COvqxplq!_U+E01F02pM;Xq=V8E)&Yi_4@(;};J9~h>c45sS9H;ZA z>wSJVcpTrcsN19m^jpFWyqs|}zJ9NYVq$GVO%L$MJs_N=JWEvfM|6^S6nbL$sRCW8 z+OfV0uKlSedLj6DHZAEyI5E8!rDb259B^}yJcQV@=5!*WEjQ)WsNo`p#2nQAEN^uUmz5DEU``0F$Zn3k=&Z93cud6#g(SPO-U zM4aeSepg@fVUDzgiqkxX9R%sy^&!Y6drB;~5%j+b^9 ze^%ptCxP;+fYkG~wK9fHg_Q8|k9rtNU*oOv3N|rVw5VxwED6~0kWVv-XPqhe+U#^% zKLP#o#}Tm%a7R}9DthEb#8XhEYr4DtUljmoC|K|v={b?WWRWPK-L?^JRF_I&fFp16 z&f0lSnNOqKJGuFGdg^M_7ug0e4xN2idL#KCf)!ly0m9F|6#@giW9(4iy>*u_SEJPV zG7@$Tvd^MvSTwWjYe3Qlzp+^-avKL zFe8Ry)r(?DaxlaRF7&@2WT(orbh0uej>(q2XY4iU&moeubJPoNTp$*N|aCO)`wXjQ*Bu=NfWK!q1 zu**iIM3FpZ){0p@DNz}y6Z14p%gaLQBP1Sp zZG!wEq|(&~+#cV4jHqciMmeLmWglk+csYCxpiNd6J=}_yE<-f*!%VB@j$LD`X}V%d z_(v(clz+LNm$)%tg~pQ>?{?mbqf#MnOe& zFS${C;jBgc`Y=9?6c9CCvEmUZlNv=s>)bfW4w0lDp~YdoNDw}+rMRq_7l!w@2<`g> zFy;Ad>=!SP$L~@$D?o>vbTtHY6}S6=-opJ>2NsUfh$5TD>Z%KW*QF|RnKf8Wx8$Vl z#&7;uG|}1PWQC=#y6Xn+rz&}Onkca5BYgHlxy5!;+M7NGUGCk|vzG5Ivi?vOdc;;Y z?oP&=kSf}#Ii|g;0}ceG{-0tXGy!fOAVsFj!Ap_8?hIwS8o4!!Ly)p(()vza&y^5! z%>gosR~M71eo_xJe3nPrj^^+KyCj=CbT+64)AT-#fUr%RiYRbQ*c zU?ZoPxp6My%W{vd4JIL@vIIV0=`Qr3mD9 zOE=(iZ(8_PFdkG;?1X!~Vt5F4Of~H2B&jXF1Eo=kDCCX)n}%cqT*h(yeU1bq z@X8P2?@YAF6nbZOiNiXJ$DVN*;1Px-N!q7~yqyH(=S-(32f7o+rMbpb4yrUhez#n3 z#|KtnNQVmSgzI#S<42Tmh$71ANabyS;|QN%FlTq*9{cOKZvV)xE}tp34SQYNq~;3) zo;WaY1T4i0SEEzQkHn`z6sW0hTrZ_hrvrxVqtn0S`Lk2~j*VUr%5t>@FpH%sR1Kk0 z?lmBYe4&@A5dZ3gP?n{quY}Jx8(WbEocfwZ8~@RYSGP&HZ;h#1yeR9YXc(UC#ci&b z$gyMH2fe3j(a5w_|g96M1+trLdTlG7bulPV-Xgavdp^NO5$+;WB~Mzdw} z)N9DSs(y^$*=7VPo-^DA$POVUF9{ab%6DV1!)hiV&MIIwxI+#8v}Da{sx}%OPBl?d z4x-R=&=)yoUOvozwzIfcW8ObEyN6)c52}Gc;yL~O_$B4v22gx?tmf>E!~3N>Z)X zq!f2VSky^NCWu&D)Rn94u>`@msYXT)xbYu95zgkASfb@mk!@Euneei9_@K^w`)U7@ zJExL=KE=pJd@zQLulckh+fwmT-Iu3dV6DiPosy^>WdyJTShA2kYUO^_wB<>2X=8}s zKPWTo9pp@cd8Nzt0|NKSnu|2o3$*`C5M!lh zF?6T1`VnUXn6>1Xwz{G$eKVpBNil>ljRknd5%mLbv2O4&SD2%fA%qhM#?|D&$wnQpRYU2}@u?8 zpDXT=D?Rm3cJ)(D6Lg8sg##ADJLz80_?w!BV-|^4>-Y6XV0o&JKd9z|dRdS4xM5}Z z+3q10W`=zcLT0bGzA$88vGg9yt3&5C7a?Ck4!mii+lKS3D4+r0?CjYnX5*797j^8+ zodd%F*yA+(ZMU72{6!G$>6-Qcc}jZ^&X>eMd(4H%40ZX28lOBME`rgbviaIA#CJ>5 zIGxz58>eiWe7T#KibHt6VEWV~4Q}_$)UCwgk3luKkPQM0 z)<%%)hYC=Mbj`&Oj84Y``C#5w=ORvg%MF4 zfng>lxLEyHMS-d{(Gc9YDrKEyROFRWG>hv|QpH-i$F-k&p!`8PMD=6J;bM~EbXGnF zAaUR++fJsikM>5HSa#DD`Rq~%`;W@MO0)jmnzRFzTG*2#seOSV?85f|dXMAD1)c)E zi?LH)Xx35!I0JhJ8UdP?C1(I(5lBR&&`ce(S_(bXSaZPr&n=Aw(AZ}WxEf4;_X4>= zfPXAePhgy0p@m9q?8d8)|KD(Wb8!blfVeyA>;QQH3`xE}Js*l|GPCxqUJUl!7F zSv^OdXAiw3Ax+`ToE$XP-f)s};tCasI$bsX*eYLVJU1ieNcKiyv(}2az**2tpIp@m z+Sn300`nR^2Yt5Ge@nfs!;0hsz^`{ADaLTvR3D|%btjdFGAJm+5x30vN40z+j0|CR z?Ogc6)PTV6nat;7WT@=>Cuu=L;WIAG;HA&|>pqH2u4y67z%PLAX)5+$x~y(4uaEop z)XVv$E9y@V|L@02n^YNU4_;qQsd-ybAJ4Kw&%uappJ3b#!&#* zHA&^3An_*fP_fDAwe4FO{s{vQ)KOidZf;KB*X8}N)U8W%y6OlLaWr&*{!`Bd0@#NM-~H;RKHCV*1oI=+$T?m0PNrvaij3EB)#s*! zb>r*TFTu;KJie%i$(C7%*%lHX>1&nNM`>!_cMD?yyXOig z8FShhg#aINCG021#-Y}-=J zo>9k~P<$yD?XFr`A#e&|UwJCU$8-5(9w!faY70SxTK|57`u$6={2a7P@O}8r_(S;r z)}nkS`Nf5T0Rf%=2R*a@|3b_ELnr51F8{|s1O6Wd+RDun76=&Z85juYe}pSD8afI` z;%NSJbzCgfV9F@u6vWrkeul(4iwv;vX0*-PTBtrK3Jn(98BRocH-I~C6JD+@q9SRf zh>ze@|in|b9SJ_b!W=vvDfIXe)G|LeE zbPB`_Jt}lQ9F$t7Gd@gee|38pO0{x?7i7U{LVX~{seoZ3c~xJBy+}H4)xCWGpyJ6Q z1obVwrA4h40XT9SU$gK;e|7YURaG)>JhqV8sDh>ED;~gip6&$FKszAM;Z}-JVVtgQ zm-k*CcBF(^aFmM>>l)DqaHC%biM=L%KrM-u)Za*lOiT*8FSL&JA|c`tEtPRZOuS&W z8~W|PHMXKEg-b~r{x;pJCo1k=r)(}<(y`{>82QIomOc&ylbG!;s*PT9cWpCA7Ms)@ z8d24Lq3m#Vtm=;;PsjU;O;Om|+EUe=$;X<(ei_FQ{Q8T!noui~{9GXFM-(qYF!3>5LV*QN!Uvfj`M}NO8Tg@6CT?Kh zz(@`oB7-qmKOt2cjnCuI=Tl2`OeXc%(*w4+X&Wly1D0xG!>PYm8Jsm#`@&PeEr8@VxEtdlwuvy6bfVZq5E0qo zOG&yh3KuGNkHDcm-X5!lzxJU=EN~mU%g39G9^~ONxT@Nn2z|v7wY{#VwuCfhH7yLs zf0&L@C#x!)5R*ydbe<|kp)M7y;rn}TS>KeXLxbLvcWMcO{!F6BylIg#kn}hRwC-Fl zL%DV(ciyde{Hvs(V}Pjh|Mftr0yODhk%52^`GJ5K{udWsnypzWA>{ule`-rKE4_yX zWT45ujt_JZIkpv9+6zCV{&S>_Ov{W6x*~!?gbDTHEP7Zm^N@7yTg2_|!O@3_`c1IX z?;S&;5J|JOA$;^ZtwrV6sJNIE8Ch;VU%}OTCpcYkxg=EqTppM?LN>?0SeFoFyUdZUJqaOx*s2p z_v#gQEmwe^$_81L%#Q}+=0Ga|o-ALY6 ztS~tLR~!c+r)B9il=S{)C%ML8 zdYLW-BR+U}ZObP@6)lSbuqLrnIW=|xN}A{z*;of-Y}YV}bBRvTXDU&r@W7^^X&CT0 zw(zSNSeJoFQ$sRg2n3?V!gyyNL_l|7*f%;ySaW27P9L4ODaYksMoe9nX}aLWm=@{O z7VM+Dvs_nUbW`boD*i_n@>R%~+ z=Z;z%%IpTFJk>tDn3nSnM&qA0wJsXdz#3V}C<*mJnY!8G9G3DiX}$lIuR|YiT{d zle*Tydc$N7@fHbKc3w~f=2R7@GB0@gq&Ekk4i6y#AiYBq#FwGIoM(KsFk?;PH$-%! z)B{i+SrjUOXhWs|<9fR`Mc|Phoaoyu-wZL%4Z@4lZNX6bVHNBLnY@@QjSMS7WVw#{ z&qRa?jtCxd4ouCpX;4Z_D>g8E!$9f6V7SKK0rqy?)KyfgiLI%OL@A+}b(A)_$^^w$ z$AytACUw4>t^wE;(qNv1nJmNMRffEHDag`th%7lE4@XJ1z*g`=Jv(b4Y+$7&fj1@Eqho;)8>2nqvg=4ROG|5BG3z_{$9IyUiG_u0B?p3!r@ zAQsB>{VoME>56k;6|EZ1l`Ysq*y4P%OL5Z3wZ5_`0kLm`ZpKDrT=uNbf5Uvdu#L!x zHsPGBzZ(6ICGIkfOf^;W(0i89G@Q8CzE8ooQ8kqNJlQ(D7$R6c75p*%OUaiMF+QIJJzGkN%3g#l z$GXg$*y-I2A|=wCY)ZoIId-i;gD7RQpOGZ54VKQL7FBkcq)pCv4*iA7j0EJTS#L$Q zqyxqMCvysCe0ZtidA_~<>*UGp@ugzHvcLBi{D0Nf1O!C&zgoWwP-SRo_wJ?O|FeTH zbi9(3n9$OY8n_|;r>JS~M%;xs0SG8>8W>36e>+3LqZIf5-s4UlrMwCMGoUpmx?)TS z2q>Kq7)bViGl2j9I-7CZgJc0f zI@EX@Qec==E7-qS72d2`rhBXb?$Ig#4~9&>zg+A4M`}TnMfC^dtCzkvEhAK1dR%wB zaNLzd#?JWw#YBz7js}B5ZKH!w)t3F2M}6_s*gmz8P57*W2JLHn%9gy?WwL{J6b(?( zsh&ZtN<&ljm_c(;lJ``Rx>Zju&SJL zz6{bXv4ZR6^^j>)_Dkc%E8Xjz!Jsrtn6ha=pV5;lFd_yIB1>Rj2a(Y_F9uG|kPQ+Y z@U~hk#`SyBMF_)Ac}w+!D-4-s{@4_UG~~Yy1I=v3jw(Z#bp2FzOhbtP_>OF9;7>R5 zW!Fm7naK9+D>zpn5WB=Q;t^Gl7)d=Vvs)M|r>s>i;s6b37?k^#4bQa>sb-O8?+IGK zq|iT#QQZTgH6AhZ0K9TzE-m;62j;4MY6QP|_FY9^#6(U#@Y*8R@H|>Wm5xFuT}9q? z$M|#c9TQ$fnvU=>K5H$isxRelJfepvkTp3_e#iNKHLZ)b24-TG5cm*xqGn6caU4$@ z%#6Rlk>8@E19b1JvNRfn=A&eOs^sthZFDxQ`}*?1WL$GdPLO8U7a5eKpq)s_FU6DQ zi3CaH6j1wb*xpB2lTH@atc6C~in4ki`Rf4nLar`7U2VJAHpfW914yq#K;$DgSAYlU zK7D4zfo@qndwJb^{JLf@-(F@~)V1$b-3&@w+X%6T-E0et3=_%YJrH3nQ~JqZ@gG(T9JU&HEz>_ z%E@#0;OAf|(f5y2{Ly1QX%Z0L1J&tifj_NXxrzPGpt-Lnnl_~Gf2|5Yl;7!(g~uHgakayObFv-~@sq#PclQN~GRFW|QZ1xYU=M~^ z-CHa`2z1Du#?3?vB94E+jXL4`{Rf!!QZE4iT>>pb+@}_M zaRO_CwXY!t)fZoMO1W24XAy16BAvehdk?p2EPRcw^?b6G?&R@5Y!36&J-%3(R4DMa z@Ej&Ust0>Ok+0x$VnzTGl}_x?g!B`-Tq{cM4*AK94O}CE%xDnRbwER$yx%}k$T9Yv z-RhxtRG%__+!nDxEirw4&i5`1>OF#G4hgcnr@N$nj5;sxR-2~%Lu=yDATG6>q6~Ue zpP;3FRsG!(ITZi!8?$2(de!2t{a74J4KAH$$BjTgmV11X_hTSz zT;&>pujZJ9519d+fqUjJvor6t=~@|;@Niw!S2KM;Vw`>5Z)pdLU&)9COH6o!p+rt9 z{81mNUM=ZJ@}At_MCFGX^;#|_2C1#q{J9|8TefhNM<%XcjD&5uleqpLZ2p~la;aDd z1~m(Sk2?v%k%PB4O<$hBga&aVVGo~2Yhg#6hsQZw{M-Pvy&k~hhbte7gZyw;lFGxy z(87i0m*c-3NGG13sR}M{%oyOKy`hbh5HudY_W(pJME#!iw7O0X$XHsNvcy{T7#d}! zakWm%k%A6===HgCy!1C~5Kqy0~-Kg?MqJ|te@*n%qvs_4yP~zPf7?)Ngp&&%n zEU~Bymuz_Uu3GPwK=<4VC34@3Y67pP+*8NHnxz1T!pRbl+i`qtGAiGk$XE+PT;?cPudWC;S zv)y#C9`${BnZq$K-rOwit-q!QYdJ5>jQ{}apyc$epUmk|7nutwg3HuuKHe!Jkbi75j;a0;RN>H!UdHdVh0dSfGT5+W(Pk>r|?qqb+)tktI@-~XhP~gpT^gi z24SQVvMbH36DU6clG~h@-p*+?;U@KZb}uiz29O9oK9GmcZHlO6_`rY_=mP4!311?` z1K-WsQy!G*ch~)Cn9_hpU~u_}SK{IsYRNY{Bk9sGXU%MUSg&!>5J(~qC*kENleyWP zLv1Y^P{U%NC@}X@TYuhIR?z5<0Cgc$re3I$0DnU=h$tm~!!_yI1|{18QZT~*qNb4k z0V2sjqLPt!lqs`N1Brk=Qt(O(0!M;BvzueWD4*9zt~Xi_ZPn%7=3IQpVoUK+!_AJ? z^LtEiVc$yj_Kj%ZF=@z$3TUV{?-UN8o-&qv!#JpT;G7xoxe8B%6?A#ju{K_~TO#+w zkS-YTJ2#ZWeGoTGC`f0;Fierw4be!=h>TnPR4A&LhzAu22n>MiFdv0hR5cAP*E|&; zb$8JXZQTWE$gXe0z}@DSv9@tidL#7n)OQ*%x-v&kHktT}SvGEn?3b7J=-?4x?a*Z+ zR%e7ycA@R6D=hwmg?Pbg?yheA==urp=beq7Zuqa&zUwxnA$QinTgx2TJ{?oVjOl!Z zz8J;5&pxqALK8rern-c=b*bLQ;z@qBNIZ^X$wuTe)DI=~AQ%VqJ&KI3Aj$e}1x4}A zXBepwN|gxVo_7TF9e|hn8EOYCBIoF%*{T0Pj{~8Z7`JG<6?b5{-H+?QAlaNy_6{80 zJy6*3KyObB>#49c#IXDNKxA_TS)EBrei+t^B5GOn;=&~K6bWI{GSfoJz18i7Tj^otC}Q)eM%2ca_|uN zn(G0={Q!V_`~XEq0!IJ=&arRmcnkk%%z~qUb(w3T=`jpl*qP}=aWt!_t#BZ2u?}Jc zby458pOIT>nqmqSy*8Mi@9P+OEQzZ#<8G&N^Zr-fg)zrK1Xgs%wX$H|X5w!w1&EPt za1W59gx{J>6%tGUam5wAn%7c)|tY8y<=f!VC^0?h#dqm78N@A*pS>pb;ueFgl0|(2}#?%G>O*p zr!EG^Z_2XFC6SVpxm3|s7ZFMV7_oZDG8x7fTqXX$=kXY^&TfZeg5R zVYelOy21(M=UGsfz?pjll5hR@+u@fOVJvJcO)r=aKhUY9VVo`CAedj5D)wOM%~6!w zi&X%FGH*BL4^WzlO(e0lCJq-nclL@5+~pCsZN)xkw4}58=9zW<@v0ND1A+2Mj2_@} z`_qEd>HO70PufeD-0Tom7Q2GIZ&Pd_ZB^nza>_hYGy^rg8$QCAB3!5vf zVa)U>qg{dvNYAOys|kqs;D67^Z5jXz>*bwvon1lF#G?8;vW8+|JL^)6Dg?)8H{|*Y zd>@B9(snpTebunB{S1WBSlD~R;ud6WiiRZIXN+&qZco-9e_!8!KLGYW{obJN1}5V4 z6Td*Ly#PPor_<*PsXwg7u;MnauxbenAb%9UQPsWgLQiUwv9VHlBR#u#at#5p>$KmU zxk+?z{RwG^x{h@JQXh;AvbadpKa>p19Eo!gIjp>!gXZW2yXdQ-|O@`fapk~;d=|EplhGhBVNE`l*f)jd(Yz3 z-XJNg#>s~TSCH=TD;mT2)7zV@U4YFO3}hN!jF?4*v8tZ%8hKH^wE|i7 z2V1Y3c#_kMSvxWKSV84@v@CaWzZQf|f&gI+l+bugV~Wq-a95zm8GFL7%Gvq()7U5S z7`HD~ft%s=BdY_v!o9}o*qT7dtf}9Nq%4T5`p4DCBWo*snWfR@xoA8=gIJ+Vx#qp5F6wI8^r__KO%#M-Rvc z(>8I-hyX*YD`zSIe>7HCwE;6iRdGldepcY0IUu81=6=Bx+-|EZ74d{q9fzw>BHA*| z_TE!xRb4Z>3`j$rS$lI-d1n1!?)rf^-kmv**B5^cu_rpbhZdgKBT|z;KI_ zz->vE4~1v=>*#eoy_b@hSd(i1)-4Eg#2K`jJ(qdS*)ALs)(v)}$MIQ5zSpf*N8V1! zHh-IJS0-UnqH;B*@@!#dKpBR`SR`#NVw3e^`}~PBcV}V~_S+w>fa@?cIv(YWt`7e% zQvhKfI@2)b&=?}2Kky#{{u1{tH~GBQ!#Or-pahwV%j!xEUxW_D#1cCu9-~1;-$GgV zJxZWvWlW@WMd#r?2%-fi7Zo$BWRk~cElo}Jm60|AEQjUWniQxp1GIs^$nC*he z#Vi4Pw(OA=zuqmWrH{`ZlCY=*y_HJKuYf=icC9gVgG#1RU#0`8F3Vom1tH`$f>EC1 zAUZ)%<-w-+uHKzJ&-_IUnm1sCU|xKG*iE;ZH}-V#%|V8T6+tAunCHl3-q-h&CZ2(9 zN#Z^nbXpQ9n?>Z!o3)hj&;4AHB8opZGIb?XY3Avd;Hi%P%5aX0C&|{}*jKUzS^?B` z$*=}KLY8+`$BotrU3-{g2{`PItqfuNTXmSKVHa9C4J-y^`O|uzY4QWTBQfhj`|-Sd z^g!6GImY2NCzy*DQ2bxBenFWQc}}WVy;OSY(h%igfZnqp1?5WkTbB($n)X%Rn9Y=Vw5u- zhdCXvI0xSir;xmFEvzrBWnm`@NIo&4Ja*>;3~j1xw}-XBdU>GwGERh!U*3L%jW9gA z`Vt$PXh@x_exFp&4Vcnd^ji>PHBG1wr=RjAm}oO8H3r@CW5(lZ^!s4OBf}7}Qd0uI z>NG~eGCXj?vU9qnG)2NPd;qPU!CaKEcFb6{+8SDkz9Kt*wkC@pWa^zktMov2?u<+v zZ;kE?uR^Q199=Inj4FnKOmUPzusgPcJy1UKA9r^~+$2yL%1Du5MvR?%rw|)88D?V( zQ`xh&Qt#@e-Cau7(ocIyjFr+1ytltA_7G^*N*d020nFy3RSxY~A^_`eDJugRzEL$ zI*T2cnRn4g&H)g_9zd0qObp}f9Z=~^3R#sg_x&2lo3`LZ+)m~#6C0R!;TrIJJV>;s zMG3@MyNMt*I$i7-d^ql1_*=IoRnD-Kbul(R(AT39h-(!8XYe<{Ee(Zmdw_(57{&q7a92oql7h)h zn?ifzcuI@lB-sGhgT?XnEqX{Lg2$3sg7y#f-*~qEU${tL|Fw&EmtNOpT#}-#A}1vx zBL{=?IdriaQ2iOiy<)&*?Y_2sYQ9ZvKPcYJbygxa{e_LxK_vvG4}R#;TGK&|jBf2` zDg<+gw%Z(PETF)MZXQ7w4Nw;02UoZiYXzwb1dXqG{ayC&K9o~W%ucsX;Sp7E7}Z?< z)h!{FAnZwsUBmjfs>epp{z!WE_m!H$xkQkKNL0EJze<#nja*CC6;^CZiVB?x2x4D! zAC3<+T)0&=Wo^9gYsThS=^xp}R>v$hoFFop2j>fe8~}v%%>Tpn_5H)9FtcH+jssV< zdV=0Yfqb-`*53;IbE!O)D{ipLrQbL=mwnp*r{$q_E`S5KgIi_ma&(U;=XLJUW3scyYB{EBF?Zo znaDXTCIBav9tHQGEcDeO5aY6l&4Ejr$BC4n3%dR zcZ`>Of2e5iU**@q8U{sc)DJ&|!Y1`>Gs7Q`0l@mL{;AYd@0FEOf9?5%-f2^f02puA z)T-?>Zy;m-h|V{P-0o$WrIbetx!$*a^yeieQFs#r6+&R7v!&=8Z9IdXDqH1V@}qwm zAMi?Db)1f6_tbpzU)rY5&zqdm(n0T)aEKPnOTtd*okCzll$z0i_5%uz(IU1)y4CbL zaKOJbW=^);9cJ%JmM2mfrmRRB5s~DE_*6|R2{Q)a2=Iw(7DS}&QCIacTNvtO=x7T? zGq0dhw$m`q)UcQMfun1!~v*0*L2Kk00oB$SY^9Edd@|<%q|A^V(R=GdJ{jetbiGf+E z+*!5G5_Nxh6N)j9CgWEj+^@L}g1HCY{+$)>@e|a{oEIp?Z-7y1Z~>uh zo_}v+H#&ey&(1kH1`>qLfRYmHy=hbE)-8hb3G|w28ucQuM&Kc$I`YgvQbV9-VxXxp zibOwf7ccfPtO@FrtxobK=jArJE$5|-tS=~oG;jYh`FlyP#$2~?RA1U+-k}~2It?dx zdqeNfV6pfU9;72E1Ut!UivP2tZ2^p(TFXG(b0h9AEP_jPwSP$_Z7sOar*Y~crJ3A5 z55eprOSWXpG)qs0P*8um&*|8m?id#{gB`%|Bgw6bz558sB0}!ndfj90vbmm8LdFrS z-={eay`L}K)V5Z!f|MArO}mqQyp*z#CpiTy1IETFWmKa;H5jF(G9*PRu- zSG--3r*ikjBv8(P&sX{AougiIBBb@Lu- z%pFPHw7M2Ec+}W*-MV6zhS%T2tgtf}&Jm@|gpivWg}E8&xxKp}wEzIFBd4D}hIxS> z-JWqh-T;Tj-yeNkfgC}H<<9PkN@k>?|M})^d%{kLuF#~ zT#m=p8_Mm~+c=pj9w)rJSH*d&9BJ|;g zV&9brH2WRPH}ra95%(Q#DhlO!JJ1q#eWaSsAvF}kh9!`Kzw7*v>{R?^jK(cbt#brJ z6wf&rRXvCb>iQnZLadDIKW=nHw-D$0i!e1IH471HyZTMAw19K%ZRc$@<&Mojn`YmX zHn}OzUbn26t&d}*0u8R>zZvupzH$ai`PWfn`>U!<{I!jCuGL(N-NxlvPcJ-(KJ001 zV@1+E`P(rQp#_Rl(FP7a9tGvL3oTq!NeioNj4H0I?VR)u^Pf%`sZzMiAs^uL!flPF zc#p7cS4O#Woq%`CW%eOAptIMZ%@{{BsbQ7nma-?J&3|MoaY`lpTRH8+f$M)Ob`d1Y z{|h_z8=gdc8fb()^pkHZ&Th)LFVP2kQ)$nF5R0CdNb*fxfXVvjnFadR()Pr%h@8Ju_}RhdI@q^EQ6Sri&T{++cfszwFCC;w-yc};cd)k9Q``a2hXgL{~o_= zFzeL?_xCAH{=x>fUvh!XAEM9)<)f}95 z_Q9aQLl`Aj?|6mIFS~Chd%EPToL4Ghf+H!`ZVSPaSKbuyIUViD%7O)MiT$OkUJwc8 zq61tI@jJ$RZv_mxO41o;hLW7>K$PsvL!am7^v;5~=g?jw1z@7V{R`3B{1>RLohVsi zQCdhDYTtTqX~^kY^=*k>oBgpIS>Qp&>CH#&f*OUXo_O6Tsb&O;pueNK;f%a0qzO9I zIP;a1{dc_(u5Ek~z8F7NXRkvsD>$;&00Y2aCZ*juvx8*yZ1+jhv$0B)Nlrb0zhEp8 z(PeT31OBbsqsR<=lqo0@7Tcu@Nh>8X1=xmbb$rfRkk(U1xWcW0?G*BlkeFGPSZS`2aX=i-vBa`Sz7TN@8Z4F*Q~n6n9jCILPIO6N8V68Y zdZ#-lvv28>SyrayQ=satxX~P^f+eq*Vn1pL;AkPp%g=3aK7BvuYI2kx2|QI&{QXXj zR=lmcaUC#ZNH2c$r3NSphVE!MXK^p~y+5bmW#?FvrzTty)@7^F%n0BJ7Pe{3>)~u6cx|C@AL;eYq5ZK&t(v<-x8S%=G;pv^+4CoL=@JjStOH=JX6`6@ za8&=Y{j*uIC_wCaz!OoYge`eO8gwgcc$^iQ3Z=3*7aYgNr1=jD8ka?uLTSU$(u$r& zBDviXAT%ouuN%iDcA1pq?4ha$?xWrFZp$dxnp}x&nEPwj$4(@(Y8fq=+fROYDF>Zg z(Dr~Mr?f@pI??8B0#dSWtO9uQr=BZuqOvhPi4O;j(!`S!7N@W|(C7W9G2pH#%#Lxy z@b@M~r-*WjO&rkdtg|hg&0g;*h?o#>o{amn~1u3-yPcX7O3eS>HARWIFcI(Y#BV=~K;b`#`Q)`XtM0kP-K3R6( zaZ+N0>5;&Hz3Abe%Q_%QX>VxFN0dSpfGw@IZ&o0#q(b-y1)C30fUv(2fF?kW*WO1Enup%5K${D4tgE- z+f_@g>(>^Y<4CbSx_36~s+M+JRaFSYXrByD8l=VD-3BZy@JTai{OImQj2wDvg31{q z3vIsx$p6_Rn|jQD?Yg_v_GC^}DWB7xNy3Y@h?rV*+AmDR0@Q>Cg;X^d z{*fGHiaSKwe~`=W48aa7So6NcU?0Y9YXC~=@A3b^*Ex2F5%}viZEU--ZQHi(#x^H* zV;hZan~iPTR>RZ(UFXfYcb)qM=IzXy{d@MFkhnxL=Gb3Q`KKWb$ZB(s?_dqPqBo=H zz594yEbYi!7)Boq)H#Id!qBRMS$9h!<0o`sdhk&Ic3Upg1x$Rh`dXxv~8IzDK#%*sG!8ors7O@>|GmX(xt6igy$UcT0{O+cT-j=im`t zUW!^uDbCi%%!8awB)=x<0y{F43HV`Bb9v(P_Vn>Ue19 z6bN}eWkWrce^z0m*r3V&5KncmyHaPg()&%9({B1Ym%{G{BYss5`9akf3~x~|O&C)x zvy{STqokl7Nqd7L4VLBjPin&nEUcQ!aWf=Rs5=Ste#%i9XRr1`jg_T& z(k*rR>qdg@7$0BKqJkp}Jun|KG`50ORgy1EipH?;J2N!u&$JitJ>7K zR#mYNmrY|p{`<;IUwPl`u{V^%Wgjdzl2IuK6>&7%4U!$2L@nvEg-ACbruNpuQ)91V znQoCyPM5Kx)odKtCg?b}Pn^p`3v+t}_Pxk2uNJIpJ8i~WtkA>A&Ye>;G^bxDP4RM%(=}@WlPc7T z>uR-y#(f6<-zG|$Tm4dM2W&BrJ~m|;oXdkBsgbttE3N3sCef=&JrZhaNO&N#~6Y@3zl2_dzQTu^- zvxrLZu0^JG4c=R>v(>uZLM11v0QP0$5RC)B>A-J{1Eqou#eYDe%8lU~%jC)*hFsJ~ z9+VokFC+{Vj#e#VZpJijD?go9N4*!M2bt$k9oX3YE;nU*nfKT zPV1%OTeVlO#=47@yU=~QipG??iiH=);Lyya*J-Kd-6sim%%NPwzgdlY6Jc1$x4m)q zH0qg3Z@WG$a+8RHbsCLU&&v+r-d=fN8-8u3#M;Ou2Q47f%w^63d^q=gOH~7EWt#$I zNtVM05_I!XXqMtbg?8XhYAozw8k#zkqw%PGUA@DKFd-qgW9!c|G70_n5Pi7#ol{UO z$S{vSBT|So_-51FvOmGMzr)Yx|8w-lYG0tfScz(hn|D`{qRc)u%1-F%04=?l#F zF6oadmn1@d6>bjNVzjGc1d84%T3uSfVwLfRr3{Oj!K-IM_vh1n-5t`L+l{BV+TKH> zAgAw`77XK;j)@=mF_#*l`bHxndEq^NVBc{B?G!$^J3&qXU5BJtWFf0_GsMnc^yKwf zit3Pq=gHcP&2r2#MAvhIaMD?Ui(kgk7V8)z%x0 z?rg!RoKv>|g{l`e=0NjrT}Di9)QL!b&695j(1*N%8D{16{z**ReZO}Y*O_R4xmi5! zpJX;2CIvvx%%dD*WVpr;rVpoSawd(VmF?Z7t&yyHAJj>&k&TbQ{ah+ zTrUr#jwl%5_4uuY9J%K5d%L?hy+1v?OrE=M{%e#23b`Nmmd~r_1G)PA-fKa^0!3m| zZ6Yo|MX~O-{IJmT(%Sbdc>lc32$wGAnV0e~RbB)LRx~N(sN;oZZ~yA$AnKl?%cf!A zVaEFRdamLRzP|SL7kT~Y`gne3C;TgUwRfZ59HQ8fuR7SFvRhg~k7xGTL6h9=m{PIh z<#neGTu(dr37>2ILg5$wItn$|vCgR1H+EX6WTh&G<|D@i zVvR<{Zg`RZ$SQm2H5;*hts{F9)l4+(N(+LBl0YGGaQ^Mu7tH6u5O02EY5m-Qs%oSv zu*f+qQb~bN;nP1yN21fYkuiMbtZgk#t|85RScVYwR065+1f9c|5rFJGp2A5>ZVq+= zP}UAfILj}W#U49>DECMjT`s#_q@p}Rjd!+1t2M4uf28>{aiP)@ z_7aa$%7hw3N=*o1u4^|`rC{<0E~+LiDfa3Y{O!N}J=}yB5;DoxNO(SCWqwSM2>uY@ zS~kMN^Q%KRJr>;J=Pl`p_6Px%$&;um5N(k%MNSEdFQK4lM1`VgAY=NcjwhVcAMTgZ zpa->_sMXso4ooadcgYS+Ckjutw&zVEF|zx_G5Zp^3IlVc!S6q#+FF%sj3N`9yF4fW zTGO%+My^g%IL8oIkQG}=3$~K{>38Tsq3@Ij*7~iMi&jsgyYnR@ly!&(;v+TNKz7c% z)8L)=>VkKSqhnp#fqM!*Om1GpHz@=+)Xb3A(qHTwfdH(LQ24hf6ic z;=4LH>;=^+A8p$0zCrM>VXfemsIsRIQ=u`cH~29{Ghge7($FTY=nQ?!GK#X~qW!NN zI1$k6=8LwyqWyk6qZ-WLC`(VNz!|R|bFd`8SryBd|D6GEPCgwtdN^GfLWv@HmMyd+ zoTpZCsU*vyxa$fN2pCISaRXOQYg+UtKY3;gGU1jC0~UQHc*P<3y|tr(?{IWCEuX2M zp*vG7Or`SRO<%32Nb`2YmZVdk69{ZN1ndQNw%?ZTyHHpfDLM3=9lDe(U_~E6L`k>{ z7ghh!B!~znmxn_o}7Qmmi3C8wNU^*@VJe2o^J{Gbj9ALt) z0}^7#6yLzRk}t8%Ix*2ga!9>@=fSWMdJ;VJD10U~`4s{@t&Sx+eL4qh@HFhDgLbc9 zAVqQ`@HAi2h4*E9L+#qSOlVWJ0plKbN@U=Q$lMxm(Bdd)mYnaa*|PNQZXy~mDEvPV zY*WQWwyQq3ba|P8_PNYNwb!)$e}6}NsJhzrAc#qMyj5m*rJd9_x1Ju@lw;@FEV>Q& ztsB|piTOphN2SO}omo!=-OqkzbhuS8dSKTeUeP_Ki{zff)oWdNw?xK$MJ3fWTb|np z*$(KE2hT1RSGz59BobH9?m9k1WS@W@Xb7qk`9jukFnC7;|7HQIXT;beAsd8L6OyM} z)-LMkha($p(PBZ;7&3cCk~mHa3QEf4Dh1vWV!|P%I z>chC3xSE{O?RRce0+OLT0qVW4d);^+Gn&WZ)XYp!TGO4{xC_2?M)5 zlksN59gT1R-$z0m+JMW&1j6RAN3R$(Mv>Sl@!r5KZBo(lUhkX*a-)sdz^_8s>R{$Z zS+ZO^-hAaGwkFVI$D=d6j1W9sn8ZKwkZ9`3bM2BR+NkBC1~=ed_KH!+Unyg##}3r` zgC`2Qnv@vd7lsWL;?8(_cisXvHXM`T7jG(x*}HcDQPGGp5$Pev#Fd0RnP_5Fn`$U{ zY)gMPXj9D&@;OfO)rXSqv;_#oZ1)8F|^ zXn9#~<2G$a-{kyRQ*v(~kG)T{xNbh~`RfsVuH(MyjS?3AHDBot3T|*n%LBcKXcsLk zt#RA{=m!Of6S;jIkM`R9#78&G9GlD_@xvQx8YI#8GMwOb1G_#BGH~`7Y_!CmBMkI(Ct?M=NDIv0?8bEkXs2-` z6^fh+NN`=-M{{wqG^P7V?0&jam2yp9(RmaADzwB*`DGL@ey52jx?O?;ulGe6R7oZI z10+W9A+F&Zjx89^1uAB4HKy%ir%BMUD;11Cn+dE*vNB51#WUqknqGLG@o%$;90^QUz3mwVXpDrSfx8`^Q1Hrn-ynuTS#gJ%@@|N zAPL>9@4O@_iwE_1lMTv$@2{a`j$8*X|2Dk8j4GSxADJYcf0k3gN_{l2M6I;iX0ym3 zW)#K18Pv2z+y}OBUaIx=2Z0xaw5O2*+LhNU$j^;Og5%9D)~jt4?f&QModL{dU!l!4 zo@@11#&Xm}3S}Fb)-IQNQ=OBGPg6~ybE+7c6^vS+F2BYcpKtR<$LHFuN&AHUT|=p~ z>7L8O$*$&J{p>WlybMEcuZk-n>hYPj-1?;_{zGQPT?gt*Zyy7@fxbQ;DmtqbGGQxWJk2j@&a6U1`?p&sT7HxE37 z9!}A_B>|VBVd4D?_FQLR-g&T$+$j3FoeyN4;))n96&53yP|*nP(WLr78j}~83h%58 zSh0w|vydS2@9_*UP7!%co9ub=o>pNA{Dsp=}Qj#v}oPXrb_ISp% zP!Ho=ZL4yX_lt74UZySW5OGc5*CmF{idbsu@0*}R)Js-?=q$ z%kLbY`pOh7Cp9UrjfhoYK|8ab`aE6_P=HU*v(Y-LGyeer@WYtxD<>(XyHAm!3PBMZ zS*`ckV5Z#ymdPg;c|fAtO=oe0KwzNdeMhLViQuw4uYjI@>GA~VzHeEUJZ2yz3(Y3f z4($DVLSLA5gxzK|{TI0$zL8|ha-~yyf+47O=sMq0`8NXIPH`{cQ*Dcys!mq&tD6*z zaL{<^Pnh)L{?k?mBeLgzkEM~&E(`u=07n+)9HN2SCgSH$f-%F5A5&}rpUr{*{C0iL+ScZFs8!11U7^Yu z{lAlZsC9T}WCcc3lb?dC9(qwjZ3@&$L=$(VA{I-Z#^X6fm-4*$_TUTnP1IR%H-sHJ ztoO($qKnIC4B9cr`!n4wkN)_i99sM&(%wJY;J7z*#kheHh*e`-mpvGtd8$8`u7 znjg&N{}2=a*<ynz!`vsLrqS%9s-82Ct#rG<@U~a79)C7tKg>)C&CMOeaG_^be1f^#F z3rV@43bY8Fb+Eia+KdGW$R^G+XF)_;ckDBB_$DR5k&5WB3kB1Oq)<49u{omlwIu|69hh&K-T1Q{cP_qVy<4fXahYLUf|dD8vrVoIhwtolGXNWAjP2 zWXWQSgl$-=_&ywUFo99ADvc z>ZU9ZVDy~lnpEd}yUsV`+9MY7tfT1hk!m-}_JSEfUn%CG)A1accEP%iN3KGQmJ`3M zzxiubtR`x5_}4mJ1PtnD?bDR+rLaTG0yU-W{y=S}TM*X82LyfMiN-B?qKpZCOuC1N z&kL5b! zoWRPJqO#76|0^2dZMBa-cWuTo(}GE6ezJ%RhCanE52xi^Yy=V=mGL>WJ}dMGsWOYB z-GOm_0dcyUL!7yT_!IZ+6oRYr)tV8dH?5mDD>dIf!iCf;YN+7d@k`mpQCZCK7&rqk zMHUGojGMKuU?$7j0B$ACsfog$2LA&I773OH13;xhOlAwW8v6SB86C-O4d$t?w7Y~ zGkJVW8x1NrUodM4@0gZcQipZ4JxLp3XQp!Z_AW_tH|x|306gC@TpPWCe_9^k0edVSOH%{J0Ix*yFYIfLy$aWt#~+heR<{x>5e*P%`38n8PcJ4Jd> zB%+LkA~fIHzT+9X5bdN`CWOsqVpy`TXLz>UG#KBua-(mHN*w>$A=$f;dgIeb z|FY0i%?&%gaNTD+umPTk6d{-Ic{M<(Bly*bY>SPLX#LIx!eO&8ztx6sVrWUxw19q< z>HnHuV=rH$JDIq;4rAX|N=**?O+;`I!>Q}dOqB@FR@uBo=IV#qMK#B?xu&EEt_Pt1 z!CGW(Tkk20uR$nHi)vxpdxVEs3aiVU*LsNl6v0YhhEbdFb_O0kw;Bq1`_%t#IbMG_ zK7o@8anUpCtK3q)^p2YSx1dNTZAJm7GW-2p72Jvwzpa>sk}qf-C{W) z*?GmOxc8v)X#kA}JUk<-6r~Id7?SZhiwg@e!);cOEy?vIM_MUJlUyikgW%jiQ0gio zcl-?@5fZ6z=#HQ6=;Vh>WO?%+DMDJ9ASoEgk*TSgVg-r?3JZ!MTQPl_c&I}aGv;x#Z%F%GE+Hg_?p07LWz3iAmscG&qM(&TOg#SxM4dqN0YQjjA#P^RL@9>0(pey7wQImCog z)x#q1R|0{wX|A8lmzYrTFI?GOY0s9O<7gaWa%WQwByj2Rk&}(lLg5o{c$#SW{z;qA z%T9ngDWh86_ho-zVw-F zATrN~<`lZ|BWBeubBw@_eDqXe?3HRyf^!uylsTW;tcHGEo_CRkI+DDOgRk_Hn$K&m z0TS^_$bPt`A}mPbDX2^lJ!GkB>lc^wK{q+}-JUbTrO7+XEcGFjo+g#%PS1Mo#BO zz>Yxf>2{mv$2jT-rCCY{VfL6%6^+|d046f?JW`@j)}%@X%6X!iiX*pyaa7^_T}%u)Th4r*Dl``WQq&JdBZ|z z4Z$klp5NW+uR!p=J5KP4_`eGVsQf%?nv@yZplepGFt?=;Q5|u6nk=k*qD(R3VM|O2wA!Ruq^0_ z10zdL4pDfkZgSX)z?*%p83CC>{qfx9Jm>^c!Vk?UbsHIS^H;A|-EQa8zq=a749Zf* zPccK82GF;l@4FO*fKxJ!M|(@$kX#nsd>8+JTm!@ZaSenK{Eh!pPOR((_Tp>sUN$%T zi^@zX(o#K~HrtFlJ3=hwjG5+U#0I8kh?>c`Y`pD7A^qefM1J3zCw8R*cvUA;YF6yQ zz$~LYUJJiI-!)H;Q8TElKT9MzzIc;w~_jg?{^L=X!PKqcW{K ziy}>A>Q3n&70aOvXk;r#L(!hSdwnobDxzwE4cDYrfl)bC-t~*1u3b%+RwLTg$`V@? zE-L!nH6{h9C(8z?(plzsB`oRb6IIPuswUEQAgh%b(TvT=Z~0Koob+$fXx%}WDOvKh{rQV+C>W5Ky(a- z<$X9v!&nu5kB8{*N(ot@LXktAEfB>iEa6UgtV02ghhRp{^4R#MD~P9~nFshbc645% zI9~|*A4W!Q5T4Rf1aJFvN9v@9oT}%(ImOc|R6idb61Q!=)sf$Pc6*HB!F}rYV1;v( zxaj7XY~f#|YCRaTa@#bceP8n{Gei)p*G@wpE6nMJUdked~{3=!d&W=&m z$yyDZ7XBOd-!xIn9E?*R|HRwZ>y40%%TE|{j{_z1U(S_OnMny_5*O+vbg&F;;V8nv)rLon4-Ju69U(SsK}V$G&==G57> zA|WyCSIJ#WEh&5Oi5qPp3lr{ab=$lqZOBvKGV(JRM(cMOIxW5hCW=#>2M&F)1*$(imk`3DwA|s;lXU6TVV6dyJ`|#Au!P@ML zvm9SR2-@3GN+AW=gzi@eq(3xp zDgvMEKd@bRiq))s4Kvf&h3*+k2gzt^jFpc63W+7bf+&SnEUgtEV6goTpwx-3!ybP# z2NTbuzOgtTir3hcl0F|qMoFKd=3FtVpl&?>W}Tx(RY0hQCg_8#0ak&zhD9TS-YQ>f z1VNYCTc{AdqnKr_P=R&K?QTEv4ZyUj?Vil{NYP?p)R5XTuMD7ZBt+IQF$ek*7rMb6 zda#ckoYK%63MC}g+ye;!6ITw=QQy|&CVV@SD)>PkMBDiIIfY&)ijag`m<`1h9+bKW zJeK0YOboE~6#^+J&3F&WN2O?S63{oF0|G>pt6`l>j`ZR2z_iNX_Hon-PeWr*8n#i3 z0;`Uw5xLJMrFb>UUdQqyHMg46)|E^bojb2()-1J>U4N$Dl$|f2q*W=j7p$kvx7S!c zJEyz7G{F4r+wf}&o`6G@X#or=*E?Stsg{cguyo4Wu?0$`Lk5!}XX?TNYJfKiK}}5) zGxV97yW4}v|Dn4=kqCz4OE2gh;6o4v`8S&p8%gH^cqNFPA`oz^fr!XR2`Y*i|6BJ$ z+cR+N4$6{hNFECeiJXK2_yf7d?D3GfPaz-nFE=;NJYa;_de~QfUC#%wiHYJnL+$8m zmbV`V&|^{2l>Pn~kfm0$`OxwU*@dVq1O!s(Y2A6=hSob&<-xm{5~xVvx{^@bV79=B#LqIhMfXbsCPVgd_y&FeY;L|!0jS7*_}8^SQYYQceJT7vf|n~p@X zmY;SC6WIX%UTPTJfnhKw22wn-1(@wFJ+);FLJ_@pG|}ndu)R;^t!jhZ)NCiZNg=WU z8j8{yGgLV`3dq`;J8=Pwm}hU&n89opGbp5(Nq>rwCt2Xa{P)EPhinAdt!-8Bo_jtx zI>AdeAbho5MQhxb&T989`R*?ldC_&(Yzb1l9 zRw%h1i>2dwd=fQrHT$K9@RCf65jVf&J|%|f(6GDH4e-*v(6q)*t}UUb&17Nj-QBjwJq+6O)3#N)6?dr7sR8a6b<%(ynX(3_;Zm=0 zBH`XV3=}?GU7&a*ba-PQZC@s7Ci2;6zsuU&kh5WUz;NfHHf;&t*{j2-7;oBB=6A-h zTXGcjgg~b(scI9U0~gpc3L5n;d=h1cJNuo#JkGO2pDADS0Lw%X4Sd#aS=l@cjIPjps^|%!ShBlY+ z!JlOF*R^=_N`o8P>)pQ^IFD9_(NCJlsHrn|;&CJlEn|mXz@hww^*kT63%anv=TSDS zxGm}LJa^~C_nX1_#T5mek+eay0vq%w5gukbcR&nsvIQ_xx4Qw}FPu^WOf7!Fd~3e( z>CnA0zM98|l7`6rouu7wliDgy$0$8=6qiwNFcNFrdd9ln_M2MZDT0$tqK zkDbxcJquK?{*!CECDf7(g7Xy=6txchpL(Azx;vT8iSO8N(?zp}cmRU78bDM92U)d+Y7= zVx5M=s7DC<*Y)(%3Bqk^f9As?xY6LrM6%@Bf(QV%Q5_d4ySX>&_%}0{H?&Jum*R_( zY}@$o=jdUo$jdq1o*Ly6OlT{|I?k9{*V3)q&FY!gd~^fqaN9Ej3cwH{TH*bC+}#?l zZ3n*%Ox+Iu`wsW#4|ccgzMZ0^6_CEIlA0Vo=T>E!(0D`-D*c^g`C`XG_xM-%c*e`HxQE3g6zd7w z1D(bv2!FL{%avt`Y`54)qkB8^?WRA?My*<(qZ;-WH| z>^9eguBn*28hAa7Oz{a@)IQRr%Oq+743q*dW8svTOJeg!b-b zCWXd*+w(z$@Vj9A&N`v$(+)~_74lO6R!4St&(~#7sVL-v6iX`Jk&Z9U3gZ6&at=WaS@+E2ECerNz~^n$t`JgFdhTG%0f# z2G!*$@O8Yfct4bbBdlP{In)gkDSMj(hZ91*zYyGdjK7A83Z9vNupq3AYhuffXMNR-tVq_6?O}wyjUJ3nfSS=cKo_EE%X{>Jm zg~72Vy&Od~f!W!D6_^LWvAE@$X!DY^321h!ua+j`E^dttu4n&|_$N1xUa1Z7q3d+r z!@A|0oInVzf$vqo_8EEqT4n#-cH{ZNKQ@op70_xe&E4OoRdx(%Ip9p#MqyQTZzIOS zN&1b7X;M6aF<<(!Ur$bND=HT7`*x;sj)eM-pYu0lUB(!-karYdjgW1?N&(&H#xdcT zNaU2~;2C$Z_tK^~cKCA@w&*0|k=TJv zrDv|uUtm_L%0HbhblJ5&i0aK0aZEyk-p>Lo zm5%&ItjBzCWA_BkOR>I;syy!o9&?_bV#GkT-dE1T)E_;hC9U6wY^<8MB-$baF#P(MN_5a$v*8MwBU#DsS>0&u6%~c@>oDh&&%x|bDl>+Lo#bc3Ju_S5<(AioExd$A}8X&4B;lkS9+09OQy zbYq*eFcfYME*|0R{x!tqZo*LB6dQ{EEKsk z_yfNWI8XJ9!H>B)_JZ=`UHD6#{ljl#0T~_SPQ%_qE$@EY1=mI{vOpn!@V?LQbLz_EorasLt1#YpDyH|M4$_;+IJpS zL6Gmr?O;A)Ycb=y{w_E>>i)(n zQrGdQo60F6hQ6!8MtLJ+e#l(Q&rMWV8VJtT<36Ldzj3a|lpfAH(yO~o4S2J5w#x%r z5+&{qQ$g+0VFI@}KFfthD7MW=3@eOWT-$90g!$47k2nDnNadgQ_MHt@9;CF_Gv1uvDx4VmL)T(92oz0& zouxIx#Cpt5y;d$Tk(_T4=8JVSc>=#p5v@?T@^>|t33D%BIC{>Sfj|J{C1SIDwvun)h4P2yC!>9*1bqdH0 zLxrr-H~8@+$p>EUej>mc)3WEWZ~eUQixkOf-xA8%+q%z$4h4bwz=00$)GB^oYG+iw z`;_>p6&Ox{63oo;kltz}=ceFF{0edLXcPed`$(z4QIKK6u@h7k&d^w9qXM3=IdEVT zd=!gAvn#BHd-K@-yUUgDpH~^cUR4wC=G>Z3rmP#KH3_||tV$NGpS}^&5XYRQoe_s7 zq^0Kf@(jW9r58bW19X-Wos&p2HQO04tRjeC*-mcCeWm>2Z>JKXRdDa0GBqDF4K*Y` ztno=#kY>7aRInq#+SZXKK2`2DC-nk2?iq%?02#SM`U@nD-L=(({U|uUiBU|;FodU^D*$_OJi&CpxIV*w@ZO(hS(ogC4p`c zA<~p+v`l8f6>fwM$COBgnXI9XRJb6^0-11m6xr$5k0OfGH{2;WHdgEs;fw`X9gIKe zyYHQz_kP{eps5CaGsjXuDNKSs73xl*G$h?G0e?GQlH_M6|BRxv6 zHM1!>k%{{Fr9r(ZHYLvfvQZO}@fxj!X7NemM5NhqkTL$=4Y!o5S3VfEh55J#zqVtR zSy{3peGJBUDQFN{s^EdamIocNaAYkL7@=bKHxpMA!K>zDIPS^86-P-%&1gkBDI z{~k?U0S;C&$f;UDj^1+tMGO5`u6&lyEZ7s+Wt=R`tfbiV#&W2_i(bFD-abQy#&t^f zOt4^cmzqrv4EfF|^&?Ny8RX9mfuV9@7MMi2IfI4yl7GuV+OZU2CMT4fkl-=M%>JWL zrI9!Fjq5XYjXjl$r!_8}T~K#n6?r~H{Lhz857rE`9Nu@}*6TixVvphE@*fKMDtfj8 zU$&XVwn~bPq;xm)DD~m!Y`kj)!xb8!@hS+U`KU6(ug`Ub>I{`{QM#0tSyxGH8{>fS ziH-|{%Stc}{GcIp8C3$k9#w@4iNYmx@jL1EV$^fpOMn04!EKCLrBGn~OXDP1Q*^(Q zoVAs>T=EQ{4329h^9nn(lCJLOQLID>^zFTM`YT_IH9pCt9X$z8LI32eY7{?TSs*V_ z3ah!w?u&}I4R=4AGc27;u2r>6L|_+LRv4eBLUwS|%Q^&8ooVN~<%!Od>7J&@BXpEQ z<;8`1xzxj=skLCuDTm#)4o<`n!DdL!r(BA!)%ysDLF9SQmJOb7B=q#@8ZUZy@iaXi zho%(kqdPg@EK}=|q?mlx56w-M97pjLV2p-893`{VKC89Ml(C3?X6yRdU@BD;R|wlW z+<8*$K94NVQGXRUP&c{V5ze5rGXivAPN^3p|q)T zCE@^>7|P*X@`Uo|s7q`q1&5@Zn38eG+d~O`hQ@zgp1xBq4eF$~zxv%YuXY&%S zGyErDH|I0T+ykqnWnLCty&Z}{gn=dJN2pCCY?LefFD)bh|) z7P}ol%rA(Sv~Am6x#RF=tXPx=Jkz&sG<^{W)=bv5UQKmm10esi`-N_w$n{F`Pl4YK z0;2gp5jV8GshNSZk*$MG>VG(Ss-jg1%l`(tCv-oYHd^Dq1QY@ism-J~SniI=N}93~ z$UF-p`E{G!vXc%n`r^{CMSlds#*s|g7EvQQer*8Hw9k~D3Fmeoq@zS^^$pr3DAr^s z2Ae0b1{^r=e>rSOP^666pH>#_vEHyhS1AU3fOKe&cBbQ3a`l!|@z||Nu_==rXOGAG z?F*7Z7(8f>@9$lI=;z83PLs0BRF99JM=+9W)wg9sWm=|KqGKXYGHcXH!qLFd z14OydPb3PaKIHRhT?!ruQ5lG97&lToHZ8nueMI>s>O(H#{|9c5Ed)!YsrLJSHF3X| z{jjc1=V|=Mmi4+$8Y)1E^A2{8t0r~Drb7oVy<~>-E_J>+ zL#<`irHw6cTv_WeeJ9Z*xP?m)R_uulGbm$6f>x22oJ;(Pdk=GPOt;#Sl6U% zT#yk~=qTF_CYKcYV}l=?>&3<2_(6hX=^Il8QNqhL)*BZP7hPA8>UDdO2))(*1C}x6 zaY)i)(^Y?_k5>3qT-X#oeK66v?{?Q$Fefc18~}}^(K%M z2*&jZN=8m>=B6rFP6B^VPqCvyA{r+g>JBw))<(E1t~FK`pWkDja)4A_g>1_ls2of~ zBV#%WG9F-A$hAKpAN6|80}VYxLT3B?`T4=PZg6PQY;551!`vIwB|}i13SFTC%Rhlo zW#vGNPy|odh4e408hNS`;i*{(0`zO%aIG2@cYlh!UON@5Cxj}k!}Uzr!EA$CTd5zY zgv!V5>-MYf({dx?Ch&?+@E~MJ?u7F&Id9C5(OUWetq=`m(D%+#^z5v+fUnTYU&$Vt z&4msde<4uAcj*3fI(~C%_g1gj^Hp5;IWCanV;VM6QgSg5{qEsC!T+O2fRB@x-x%}w z;%+bQF#|bn3+cFMtI)41hgurOgvsTnNehgA7~?f|)Fj zkvhR}*u;(I9-D`gNRRC#OVQtR56OV0^bOWOq~FCusgbP{WWV8irI*DIvkO1_JQkB# zT)6b+=u0MY1OUH`^jo-i?a;du{J4dH`}!#hEqD`3&;w1rHqN-ZVW>FRH=EBCaZ->{5A|BQdb-D zpWlJWIc(A`4ZKPDY|4#)GSaH6HqEdw@_0es{-NsI5WvvHt~Xllmz8Chh0T%3O%~xm?|X$_n0dje&lL1mU{=u03VwLheU3}Yb7_QO>>fo-x%EcGJJnW zmfP3oF39rQ%)N3mdbaiA{aD9Oj)YE6ucGv{+yf0voH+Y|jJ<0t`3)$wXkkBOn2~SEDCvsuU0QukCqj_NhzydX_SJ zhKJ;+9$E)W!O>)+vdo+~UQnOl_t!dORqI z7nF}{Fz6Ql+(76}0VHMz1;m{^Ri;i~P?HWrIoHIy@6wRbq`?a3(Ed}2Gqgn6$Xx5p* z1Lxb!&nPNRoxhp@POf&9V!lo1GPNi(}j6*Q<+*22jl&2movsG9XEJ z%sh8H`6z_A~_MSG~B$L3&oEzE{Sg9JO8+G z0nQM{@r?!|RG#?YY(0OZf-Tc3ynL@u3n;fBC>5Cx4Xe$GQx3#P?CQI|RsfO|WD3&M zxIgX45bI$O!i|I-Yv*rf84+!H2|)G)ZBBO!mLAU$C?28jOI1o zu1uPZ0z~gaK89mVy9AV=X+?eX4I7-9h=kNtfmN?dw0|<&PqngLpJJkdYKmh-98*j; zns@-m=&mSdr<_&^Jr2#J44@?4)Y43vjmwXQrleZhDG!_NaN(Ua2PXN%VZO%swjuMQ z-$q3F@>YL1KN%p8wHevIPeoBMT#LZg%lObVT!hQaw*$kQ-VPL&0{l2r_=-)8RxcL{ z6rX4%`NNrhh1`Pm5@NCt=MYbehQS<7^RsQTW&Hu}I?S~q*1uop22A0I{Z3UAg7;L# z#`I@RZHGc4o=@01t4|^6GGhjrWvm~^%Wzph(q06;G0yu8CmVG8*rbGteUvxcH$Sb52kzud8xQNk~(7K{b@DMZh@ zTKGE|rKrExYKppMtb>`znx#46*`lZLOvedItO}%P0{HZ0e%M^JWCMuZ{t zp!yNKk;hjit#K053|PBHweAChaMBZD?PH>0C}Do~a{v=jovfpwHPj#;KB5Z=-|LQl z{*XZ0C>95roj&NRe+EHdQE1DQ!+ZESD}$CIW7@|jvyFe8B}eH4&UDpnGxS9_w%`r# zn$s^LG}}MaJ1|G!>Pjd`CrJ$a2VJy|HgfjE@bguWZc}aIQn8)>QbRmT%UZD`DBp`9 zZi7U4YO@jiJ^$y>Hl0Ap^jlCk3tiQ@4E456)AxWIHVA&cKs&Fn4}5>{;DOKlevoXs z1QS8Va9nXeBz`{YKri{PP_?q^xbEl{?n!S%XC2g3KxL|>%S3KZi>qPm$O6%onAqzE zvG~gk2Yq63l7N{RJoLww!mUFP7~%Q%unaxG^r}|=PYv&x-@{|g=C}OH@ARFOH`Tw2 zOa>4OP7!LAG)U(|3nT{>0YenRqh;twd_hoI~VW-1NW?f=u2luw&Owd904;AfXk9VL9^j8m#)7Di^up z6t8o^YKtG?I@&o{&5G&#G9lzhcB=2vT}~1k;Bsu|%`kTmVv?h3xReTnQWGR;jTcb1 zZBoRl1QgKf{7fndSg?TP4qk5l6UO?$YH`%CP;M8%h>SLm7mOj4X6SG{Nb%cJAIhIG zHdb+cr0o6>vHw%Esya?)wKJ^_;$r?%!{yql?tp8}%v0!xMe5@kV_=cQvF>ZeWS7GL zV5;+f-4Lgjswm7NI6>HeXiMv+T>}x*GzYF%4I(Ke63KfBY-^35QFA7e8)y!w!IouG zi*IkAnp8yG`iep9I2V19Y2z++1mjdbS)un^u|}m6I@F?Qx36u{IIEv4#I~TuQ^fJ= zL*NA-J%L1#g*uCy+sPx)CAhvw=3)5_pjC+mF4A@zEuT8@r2Yg>K5dly8y}qrw3Zya znJI}mh$j|6rTmJTK4rA#Ew%ZV$&|Xi#5yoWKB2)h&GPJyt16Y< z#B-iuwf(NY^GCY|4o{=97pOVD1E2ZBd8O1d!-$5Rs-^83*CZ}uqg$c(8=I^Kkge}7 zbw83O<3ioC%bWkke9VLCH1vsIX?qGy9!kN2OIv%G=@X<{WI=j8E7}Eo7m)7<@>>dA zonsE6Y5_@6h=>`dpCKB^d5dsV%_4rKY#FO7N{zda5H2nke)RINo*^FejCfSa)ATvj zuc?8^kA=vGF@XE-4{xmqdSr%){!Zgw^T(Kb zE_s%=Jhq~Hq`RhMD&ICFKghzFOT}O8S<7t)5)m@}r8WwZZ(EOVbA~-`o>AA~ArsWd zk8lhd4ualcfL~Z_bD_Vn_gpUnYsVt3<)HdfvbD`74sBJvq}8oi+t#-Suqn_e0MbK~ zDTE6r;}IU)(MWP$xk{9EXx-`6$hPRC6XFcEEk6(+s2jjGSVw!1iQ&TJb;B~V!6h~c zp#o|Q9)9}v(g%z;q!^!;?eSDYw#oy_8LTMh*2=)D(yHVp1+G@>LUO|7f<->k?&2S) zT<|npc>zW6NASHcwpGXtK(}?zcotP$dm5?@T?-L1UZ*Or!2^y#Cj~ZvwtqlcwL9f@ z)>c~HRkN99bUbU)Zh$O7YTY@a%@SqK1ElD+Ctl_roF^l43(2*!q;B>jfx~$->qsp- z&!~tDvJuzTNYN|GG7;x#Jt6uw)m;8^bCnoW5A$T05Vw)27OyJ+IP}tS?rq}Ksb>676UaBP&di8`SZfnj2Sb*E>oW|F=RaFlUgZu-%b#&G_k;6^~?}R(5(s8k) zHTZh8>M1muMj$D)&`G~w^as}W64oBX21s>qWvzTh#?%PpHf)x2^_;O0wEZ`iyYogjfj;dS-lv32Uo#uq~*Tla9Vb@>EG0 zxy_d>zKp~Q{?IbAkOre>TC+o2hU<8SG~E)?xrRxL-IW#F#yBZ8Ed*ue36}-!MGP|{ zciV*YE)EC)NxUzekX=$k7jXk+l zf*y3DO2wz;Z=K>xq3sUPyPhs_m62tFfbUe}f>31tcFU7P{TE-LnU4_W0-IkvCZ|Z~ zb_$p@PX_##?DlHfBUlH%m<``|UiU3~JX(V~DQ4pUy2>PoQ2}_6(^NBfp_ou?g7&OY z>-zbT5$Jb~Rb1C)pg{NV9b$i}WacMCeX&*CJnx#OV6Qk21=!9P)^xT;jA-rzdz7=_ z=G8`BBJ*$SsY$~)E#Zx?Ez1t1-*EHh-+G^|8e^TGC8HZ7wR=nwVr+ebK!tUut zOmNNH)<_?rHPP+-dl1VGu+xw3um$*sD=+o-Ym!edh~;3KE*@<)CXOQ!GJ;IRcGsGF zMP9%wL%CQ}H|E!qYgefbH+n%;zcg z)Jc_yzW9pg!X9}seB>8|fNzJ&n+uRZbNKE&rP?NtgY=NY$AT8z+ErIUk<^hHYuTD}u~ND@%|nyarC)1aXr0&FJ9uD z-qOjAA`9X}@`Kzd=Go;yhOsnpT~~IIqpvAhGm>HfkK%vHB=MdnaqLwD4auB-H2GRP z7WyOO(Te48uv7eewIBt+P?I#(7jxE=)&Lo@KQkG+rlq-H4H}sKq?{N$Knn530Wg?8 zN-(HZv%Yb3%fw8hx?MelYrj(fS7S{;G+Ms6OB((gmhY4-JK;l@hLwZ; zmm19mHgoV@f|9NVu^d?eVk()54Cspk>FioY{W3x+n+M+AO?G2$KrIJ=xx|iWgzo!C z4OfJqv%k~3qJj%g={Ty=rKH6PQPFg?+nW*y`ShcJHKu~UuEhq~sLqFOZyD&QNGiUu z{}7(+v>4A@0Yzh(Xb!t&o;HwGNuO(p-_|w9Uy0XNEf6>98k7U}=lhb)PK9BG9Bh!B zPlp4i6ASATuM0>#2yw9hK>|N6FMyY`)0dN{r&BB;6AnQhVK+a&0M4?t?hQ(1j!MHq zQ$myr^u2&3^HX!s#=8x`$l36yRRL495-6+53Zycd$gTq%oX>;Gho#_do6M;q6cF+o z58bwu1$6AE3o9nfVd8PrqH_MXd(Q#|R<7Gx@Rvrfm&fh?#a<%d2Md2h7#C5o&*vRO zb~QU|(+xK@`>Jb*E~tNrP`lRKI;I`$%H=yJOroR>C0|11pKf1u2M$Iu35>}XI?I-> z3+C|z=S}`Awi2&^n4zE$zbG+)v*r}#!|7DV1fG1|t^Qw~1H0bD=y{7zr5A^bfhP4> zZj|Myiq${XLVPMpZJP%|I#qgig*Kw2VFvXAtrIbnqDtRl7rc@s~kGTO3#ruq|;ejR; zA+|OPin1m!Me3=C&1{tLevIxOI?{YK&{;_EJhZ1?F)wt$rWgq0P{4%;%2Nhhv`8O9(F%qU?`dThkp`~9)y zU5t3jEH@tjzh+W*ZjcWJ4JINKPYg3Tf6*B}G4u2f6Dzxn1Jan%GCKUI5 z&t9BAd&aJ($;gC$ge?9i^J!K@`@Y>~ZNZN@B{TLE&cf6R!%1>&_1QBsTar7utlza83Ofm1=BQ|h|v&=zp zr$y86c-MEDy%fV71spDbq(P27!?_&8GPL^6y16ZzGmMGeRnHsJUTB_(YJ4uveg2^I zGfNuqiCJnKaG1=^Y|%CtPW@WaA?@z7Lad2^1IG^;IM{3X1n`t9X}$reCLd+39dTZV z?XSn88l3`u#9+te82hl;O=@CTN%yrB9d>f)L46{NjQB3nh>o)pmzj@xE5?q4b(r{m z5VDWKi0{_`)uhW!CnmuKUgt|R8W{$+e?kJJJctdciVjmXcL8*Xugbl`$QSwmPRic`DQ_E&1<5a z>&3?eam9Q*+pRWKi{iQkGX-r_mI)ma#u*vkYKk^los7i&5;?9bLkTtoF*AQ+$OZyH zGbyO=P3NV^gv&flXCy1Y{WOZ#l9ops2iolqiCyCXchLs?C8ftFcwsk>_%~~|dJ)es zdA3R?CKGV^xZjw?o|#>hjNtD`i9Ce4))jJ*yzA*J5d$*r1v?8+2N?w=8X1h=AUx3|j4~YNv)qa$cV-Od4evT? zihRNStc!)Ds7TfxSXDjJ3R%s3L*x)AoMRQQ^EY9L!=rsg1*47cGO6q>ywA2eCf1M^ z)AU}XhfB^=iOQXiBh~_={-j|oXZ>K2qhgg7h-RW03*#nYz&mZ=N#!e%UGoPxIiF=> z$b^@s6YtKK;RQ-y9npZO%X0-EAjFFoU@R^PwcxDb-OK5(U?H-m48qGJaH|+g4DVFD zUr0%aNMzfk6>vqdEE@Y8C%^0a`_U3JRDJSn<5_&-GC1+oH1a#3{u5qw%^~0*Urv$X zkSzbJZ^=AG30F-i-kL>R5o-xpumd-Dw1oX0M%Oy-iit_>E`UxAFKy@`(*kXV8_Gz2 zuwP?LeG4237YX(7&u2_i=G2W9XgQh(bmI6is=e<7}8jCIi{q< z=s)LHq)w*Yxailiooxb25MvLEV|Yy>M6Q#yPNpV+wKY(St5gWS_q{0b3S&EdYN zhhq7N`an(EYZQV}pFFxkcD5VR9<$qypaW07D~Cy5D~Hvod}ebl1N=18L0C?oEKh-Y zhG2gA{5#w#Fr)!qLvgy!B$JFgvPA_YiHXcu4NgQ%J<<`9`LY1G2RKzm2QlO50@zZe z;aK6CBV5>V9QIN#C4^>A45f*UObWyVQwuxzsR1~!;^Dj!C92+KG-oy49BP3knt=`% zjBJDu5*dsn743Gzy@#AB2};6EqS|6tf?ZR)%aE<9#E8u^)~3Mqd)^pFboV3%wKI{v zLmxB!6m$9Or6j=YFk=fi^pkCxb%C8OB-(YvfdJG_>an~Mq^TRWIY$R5apMs6GbMBO zEvN!I>vu`vlEP_2JOR+*g)2WKm9yb1e#I%BL$2LFDe@2=dYC5CBh;l zm$!W|IeJWl41(Q#mR@r9S zn3EW}_kMtUDHJ6wDCo&g^l_c*ikoz%5F>8*e1`|D#oF)|rE89i?ws{~F2rnRj+Q)F!36RGL- zNDh?TF=O_BN!YS&sRu8Xxd;q7y|^gm38l1?ipU+PWveXm>O0c6m7bJir0bfhL2OIN zyjC)qXV?u`4Wgz9xhfPPEtJ9o;nO5*g~PylmUZ?nCv=9Wl<=WJ!j)1lDb2Ez&(uw? zY`FkZ(cnCn01cJ*>NKFf1|4t~l=m;>;AGvFLnv+iu%b3Au^AEJH@brU>AaO6GHn>K zpoLWs;$pNdHu3O4w4l4=tNXn4^9osyCy~rFRMfdER`Ekf)7JghACI~-zIO_>F&h)X z^IkW4*yf#I-DflK5n7U}hC^Ou8HLmG86^PG)jHAF&Wc%50|t!%9u1!g<*V)}p12OS zN8Mt<7MpOQre?Wwk?NsIT3O1z=0JwVHbTF0m=OgUerMAR8S`awh9@4%zN3)IQ(>W` ztP>$_$-ef3sNeBxEZLNDW`PyP+#C7MW6N5(Y54qOdXtn<(P0x;y#Gd|YsSlSTS)+0 zmK8g_>l=v`^iL4v_QBGivnGT`XB#~=F4Q_CJ@_k-XYn(CR>vE7fKzz8XZQ4{nv)6s zA)8Mmb(^26ymLp~F#Ub`URHw>v0kX#3llOw#^6g8f8?mJ3YY-EzCyMY?9uJywqD5d znRsA&s$6Ri*mC@{N-xvXUa5{obiD$|m@$;mI>DaLvj?syA>qgi@y6Fy@Ms(4+@~#g zf`iwe&>SA~=yF;QPK%GFO%t`f1!k~v`@HR&av_v|z3{qqRr4J<;LtT?4vmv)<@&Iy z@?;qZd*YI0>8*VDCFyGvSN_?JHn+f6U*rz0^-1?XwYwl^Ip{2R09 z`VOAutugIdD2Le8c^ONRc&!GI#rYhAb7Mw6v2!*rF4`_=xx4SA$jnjFNw`7X#|z1q z4|gO-ZKh)}ab~=AVZR3b{R;qiebl>>6bS*60QJwcZgwnl`M+L~-*f4;MArF|78DIG z@%zX$3$6`sybJigoKmX4z$&q;wLrJZn`iW_YMp)KR7y6SHY5lQ(^J1)i+&*uc`$?= z8+Tku?i->N;c93i$xanrtG!Um9ypbzq4Q@y6pNE{2UOf0>zL8Ekp5axr*a#$uX`9>7?LV6+oc2)LJUlB#b z^48@&@g#iV6Az ztmNBc*Zt&NhbJ-x4%1Bx$DP-TO7D zU>f5b`4uUn-&!qrK@iR@Qbw8m`*&~T z=UHT}0Q&%5pMZLU1JG%jQ({I()!8ivzcqq@a1H&`mxdHC!2qz5sBkl!4f5my$D^Xb z{9w9qsH55bSXKHL)66KAnfue?mqjHzT1vCXM0a#8+{GYMQtZ-NYGACY)Jv^jYD{<>W-E?WuGu!F`1g!S7%Ur z7xclcqhBwm?YF7-6W$?oLJYs_ztPh@AE=(m397_wq1K)4Ts5h18XA!I;wgWA)t^JP zeD8jD-KJGNCn`x(_khqRM;-wws+ieGJ#G;1#Ko5R-Dy$X?HQtSRUFEE+a6uml0gIn zk$eHp&$Ya!GW}EC(GZ>Ur{Q*t>~u+_zBolbNBu(8a+x$hBs(@}X#bGZHwkU-Qp4m6 zLVB+aNp~k$%Y>NeeZUKL(e8GI)+m5;8pZ`KA26<~?!rq*=NSAV+*b`tkzkM4>7y8B zzz*H*fOX%p_#hv@k!h+ddqhQSmQnuezo`efzc)lu&-HXUNf77nl}ogWgzRfw-oAR|QS9Vz+1@%6*z~I$lrTBB0_fQBulpm`?Q6ZO)L;NK zp?32hr*7*g!x#;E;H_Jfzc=pTJgqd}mbJO@08`UEFQ0;*Ew(Frabty?d;8@&h)w)oxL{EqnV5RUmBS&CX!y*SG^#eW`WnXamf zOpNv7>SZ8vrqEfh-O9$^zpL@Z_1FQEs+aQ|X4}wDKV+U|j`J8p#FHcS1rghKwA zm^q3+w_|IL(b)ciTe{5M=75D*iO;oO%*W(=C|>I%0y6=RLosL9bj&SfM{z(uXVy!A zd95Ft>RLLKQQYZN%5N6-ZYbj&N6rJGIzX)+v(dd#8E)L3o(bIWDD_rMh-4ylNtf9r zF_mUT`dO)8zq}hbw1xuafKD^Jx{yF{Yt{u_u=pj44x3?3(|F>XUqj2#?oTo_IM*PfWSv(^nt3-dn|pC1%Y+jwFbx~fEU2(K>C|2ZiB zzrXQ+uQ&Uj#mqb%ZH=t#&72wkGZM&H=Be;EG!W1;4iJ#Ye^@V1Z=)NO>HpkMn#gq- zy?CpBu@+aTknB==3z>q{uLNO<;-U)zbwr{?&-#WxKgrK7z58_^h+j3M zKK0mQ_WjEF{_iGB1M=or=ZQ3}`MOLJ*fs6rx{O-iA&1UB`e-BZuS80rc?nG?gqRbp z8m{;^l3M}JEu=fmMX!;K(Ym@@bi3HZRdEkRQeArj_@<>Al*#;)eyr`PLZj1g5VsbZ zFNm}U%7F@Ot&V%mRk|7ynAL#9YgJm%O&J})5E)X7sKcyN^x#(j`Y6jEXtH!QyBNunnMTf!BTvi+~3M#Pm_+m|&BorYL#hV@C2DlF?xbB;P zhpFdt+LC?`&EU9C>zWlCuHLU-@XD>-6{L+w)9C6P@g4SM0QW9j`?`T(zk5)=PWY~E zL;aV>Rir?0Uh6h{mwM*=0jBk7t2vDq!++oqtb;&!RjrBV3Q+@|--1P8Zky7>du14H zh5C=1l-IjoZo=nEFz{_vpFP1@AimN7cwR$q>8Y@|VH`fr_nJl)-5R@OloDZuM~ot}ek_APtzb7_lTX32>z@p4rGNN4xSD~WuP zp!a+^!y}aN4g)8>Jd94=M>EYsm! zdI+JUq8Rg_RyWO2a8xJMV+XU_IG#KkD7BoC3Ry_O#JV(<|XUye}n272E{@V%>V*o;+h}H5H7Hg+b$OlXjLN zT48;1HJlkJ0s_f7_&)T4!%txVoyATW>hoNH%|C&Mgq&Hy^QmnGgYkey8;*z~{rdi} z(~3^SBppOdOrUn&wKItPn_WT``_p-Tx$r5_%1iKVI{a@m5!)QYz% zGCbTwe6Y3AgrEKX>eq7{pdn6|^dGhAHv47brJ!6BT~l*j*XNk;%00e{LCq^V5Fyt; zY)rT9txJJXER>n+yT(E13}xk2(Zj@@mzU$GnkA-c+VDJek5OX42r@p(t%SzT=#t~6p;)St%d=GXAQjQP?~X@qxqHAZ1{}6E+*d`bf23v?gLCHjakYu zAOW56LWNE2uOPxw-iRncuQY!?aoeISha%EKKk#sPXp#|t_!G`(WvhQTSV+3oUVZKA z4>kwU!F(zCL4f!*lU5rjUb_oHv>THloN_o462wqS-Mara|L~l)TUs&pc=7+}bupk$T&2c4MP=trGkXFlxTnNUi?B=jW=TF5h%#iAbyeOKALMz*4 z+=mb27+?ZCfHZW>L=#jzA&!hKi#-nD)$yf#>$-E$)#>$KdO$cVVG!WQ=uLHZVR8KO zz&0@8j{Rn`suG*t$-BbX_EH;(9ZqK1+`KI~GZsuqjUM(jMPU~yf!5iZwXSW<-L8yG zlwPX5oDkQX$zblM`dFaSB%jLU0pD#iJ1{r^1CBxY3Gggj;qxB&3)+})0YE#$bAP`; zoA$Z4Dij})NJHiH<;hMCiF-^QugxHJ&hkn#!dzVpNp;bm4KFqd=qRD`}K&{cpuR4jZnS#w6Y zpWScl?cdHeE3eNOW`XEn&EMw47hRw(isa9hY4je>ja~hBmlAf*e2x&&M3z_M*KP8| zrvs>THP_?!*}kH*o9CYoB8mIQwfKdYwG_gt`Lp1A>si4>vZs%4tj9(F{#*j20E+01 zoY4F;9iVsNBHQV`r_pfjms!YY9b&o7`mNh{U@tzk=Tzt_u6BAYgzAZd=^@rD(66Di z86tF~L+u1APGDU$g$wd4p7}OGZ?d%PYAC?Ge ze9!J}`Qj5_i7)(e!-Vl|Dx6-sZTa!=nBWtOUyDL8zY+fbX^P87f6svs2*@HF2uS8X z3@>TVFJ-*{Ddushcj}BQnfP^GR!(dpLWT(>{_Ad_flW9mIR;DSZ+tU9Y9gquE2cjV zN2iW&d?0@8+UlC*f1m4=>wJ}!Rg6H1VU6gt>xJPGIbJ>~87XR?Qx?WUX-4UGc;Om( z(6_K3N0dcA2>_{|l*jX87Ex~$$M1)2-GBOfM+82PQ*|FyQBg~2o!~Fs)L;5k-xO0l zY?w)0==WUl1vmR;P{w`fQ&i8+=#q7AMaD%Rmy+0Tx4K`_x-+U2gfVs7j8x z#m*E(a@#8q;-vch$>l8#X%oQF;wzg^B#gQra6tuyav%(c144;xy=uTdxIi7MRG_uk zzYsHSKG;#HAGn^0pIQ9NFDNp~-<)i?!sJSlvShwPkA}~R zU#O8e4DX#b)hDxOm*)v6jN5uNORcb!LD%Z8k^}sSp40AvU{*(7x`OsP4*^#@`K3aW zIgU0;CSHhw4T4ESuUnD}0R0n5VXa-7m_Z^2zp>@*G#rd;y&01 zZCP;_H7^^GmajZ+trP&-w%cTh=WXSW8cTMP`|xjmzz6QbF{;Pk>WGj17uhG@6=b5*=j>Hi3?!2@V`bg>$56xZC>9<%N zDe7cZ&u!w5)D${TNX8_J2(3Gi!ATcF00FSeXjnqB;8kwZ68o20KQ&Z(i(dOCz-96o zllWN0pEpN+sPD7M1ESbre|d}gyY(mFB<58*ot40N zR&Vo&Jze07P9ZOFU&ZUE`P~5{x9UW7XY%H>Xj0{xUeNbKVB3x`_2fBkc@*%teWQ$* zlbst$7F-B9Q4q9XP)OcLFGw)LaL%N~f#diJPN1ShDBJ3+cFZSWP@S0`S+4Gkbzfu|inf&Z8bUL~Br@8`gsCNb z)x=-{^q&aTGD?ckND~b@RRR(G1d0?DCW)3jF$!iRD#jaNrHBgvllUW_DO*wIk)cze zNye^pIuS!d1{`H}>u9O5|2C$!X;TAZMM>P_$kNDOJ)uqo5xfI-2bKaONo(vzt+9G>?NhQLz+c0Prs zjukTU_2v9%K9q;09H^kbBye*!iOtl&XY{sq;v!cA_>FZLI1|8qKGOd;Uwz_;cyMT{ zoC;Ad6ih@=7Oi`cKesqk@ULk$e#=$z^8_L+(S_u)U%EkcroBDxe|b|r-`F4zxJO9+ z`A+Pg@~4AA!AB8FN^-mb134ef>{It-yldOwr*ja{zT7}zE*=1lb(tF*qFsb}=rw(6 zZCXbw6}IR*@@9aza3n$h9e(K|EXE$l;8sGo8dG=%N}YD}-^mXtQNvp(h};c$$C_}| zjH`{Y&-kiaRd2O|hEAeUSQxT;et&%LHbVoiL!=*Q2Wpp#D$17c<$5@jeQoggE;bw+jUT{A*vOQ>usqu@v<)K&h7Vye z2Wu8d7MAzvgcYPQ+ZO~%*tY{jt`m$B88cvzS4#8-OjjWk4chQJgZklcy>Afg$AVBW zZvd8$dGZ{vCc6_!2VopF6J~Ig$LD9;dlu%L_;z{FtC6o20>hjR`Q72jms@eZhKD~0M4$BIZ@}hCNfeVbu9h-Bp##wprq>fHF)ZyNqbh(8!%F5( zaezr_%$Sw;&hG%Cm=R}-U=}Lk<4|r4Eh*K&#s0bmd|~>)`Uz^<*IwX$rnDS zHK$`zf~*5UZI-B4yrizo1blBI-i)?{ ztx1Sk7f^2z-wt>%c0yuY{1|BuPpsT-=ofa@!|y2$@upZ#&Q6UkmQJZ;XpoXm*B`KY zs}0T!k;u&Fr}U?P;v6tAzbkuMEValSHd@hSxHIvHmIYG2(FTKOK`Sl20K+kh6(zzydr!tgfnVyn}m8^%wS*Q7k%U+!L^T=D3Nq3 zDiD4M?g&#NRO&HxEddi+^R@%9SHH#Xx=|j~);X@{7a{qpe!bE5#>AIUwvy(0=PAW) z>Eed0=IsdU$xrtJqplH455xmv8Wl;6*xE7vqwjoige$OycY7Ex58~Z{zHc@h} zCbqw^H>$K{pXr-jn>RXI)yk8;q9j+eV|5}-G}0c5ftx`&Q{h8Ey8uX>+|dC@oB7+p z$={QQPCKl)g_q*gahegk1t>Wbgkq)hpI7T_tfIomIv|cwk->jd7>CkVwMIRbA3L0M zCWR*~f>e0#28F;;B)ep_=lqf8?>gF)v)~3zj*$<{VL=KxnBt}*u|&4W*>!>HXFMfY!h3k?;J>C zMx=_e zSZ^cinnq5pbEG!5d7_9s>g&2Qpw5T-iH!WF?qL zstw*Qso$?4ub|a!6-jzjpwrgfptwf7YH>}os_xgf9C}J_EbPa4@tw<}|2<6<6 z$+Ks{hSrf3Q?@V6j4EYY9?43lce~f3AE4Xu-pW(exM|*(LQT=JX2^ZyYRQx=>`~oC zdLGJ($;Yt)^s7H!#)($uV$GMVAUcOA2xrZ;6k*$F5E9e~Qhj3^JLYrr%GkI6W7gk8 zhRIF` zHU#!qn>jA*@Cid{jYXG%tCE&w4!@wYkC5wj68UfwB;ISv>3ANx)!$$k$cY1C_GNIHenO2*tjU*$b;Y zoZ(Z;5dI>y=1Nh`D(;jNL_3|nLvbBvE49nNe3*TO_k)UbxOmEE5*H2aSG_aD^f%Cr ziJQhO6o?B9*efLzd2%xi1?qfg@lDX#p)E3K)089z%|riL?|m>TE(Aq6DS(eGLMf((EiSp{EJ!tZmf0h`>ULRA}%3n2M8(&;B#fvBDkQHB%rQe+VJtbn?0VzhXZ$a;> z2uVQetw7Fgh?9S8y6G!NI4!FXljS6_u0_iwYUp&Zc@(XV7Rqj-IsSI0?W#co=>=CE z=)4WjN`XVR11kZ|E9b{VkvPW=i*L{fdF4mPc%GSaz*W-wG%>*%%q5QU zYnOFs-tO20?SZNtbW0_1nY0h6U|>bEcaXUnSL&bRD#p*5LecfOV_OcL`jMVt;KN%yP<1~`;lTNG@QKC zAC)KXRXv_cT4V>J=Ub}dOQY_8sqUimlAicaH`rS9^Cj%&=9F=U(Nh({um^(OU7)pj zL=9(aIC3T|W}{D?&<@asYyxEK!gOf$YE#5z2jNol&f48vn;xXOjiG5(&aY)uzKrOx zJ+ucPed^)8>OuV1L%LHF<`x^!fN`M5`S+<4{(dXRe|PoPQ*l}ok&&n20N<$)lr$6S zL}-xzw_<|;PDkvMU+>~fWlq}#fd#wT5 zCWN7uNwY-pY+5xDemD-rrMHJEN?KE1+lt{+D|mg7YR=OiA2&V!D9S?Jl?*ee#YEa$ zwWHC>LbJymcaTu@gI2;^tsJNEx!LY^MDV!afkn%l8X72_n+NW(S(G`{XljdzD0Xlp ze_fZbp8sBj-5k(3hQms$)sy`1oNKWH?5A$IWo2bAUbG-8(^AgUTh)NOe~en^98gx; zg!J`K4O&E{svB%T|9AeJ>pI?#o15`;DIHydXa`NqDpr+6GNG|ExFBe45$nDLf@t=v z4tNU1vH11y&Y9i*gZb#T*Wz=>^7I0;1~lEE`oH)(r{GM0E)B=FZQHh;iEZ1)7u%TF zww;M>b7I@p&VRLATeW-FSAEmfr>f6+`+0z-JOo_<`lItQ-e^YHIjgR-Sf@2I65-(#8<$%Iev43xDdh( z)73oyvcXHo-5cE8$7vaFF@e>OzYV+Q4@0@j)%WnnL)M(%1jn4Z=C*?LrUnVFeRjss zCOJ0=8r^Ei3-;MDT@6pE;y~&s^T)eyMJ%(F?WV_LntPBewFd&2KD#!KO9#JlN?WNA zn@_`n(-8zAfGICkMk?u-j=>m31u|Wy9Y2Wx<>4hOfnAImULLfy!>3@JF}^9Av8?40 z10jRX1y$ZlliWHA{9x_OJBt+*3T1{tV+#VV@}fNo(mTb#9QlGO0rlP<;xcJiHICp>~}s6f=$|{m%GFK9m}!54WWhU-X7j~8KkoBREl08=va2j z8s`w(H&MJ71@H=vVwDiZxO)-@;X$|?Yks$K)ezZr>OeD%ldiwDl?8H`tp2docz@0f z6N+2$tEEu=Q8gvLruN=rQ&1DgYSWSe#6fFZ;G@O~ZN`*@T!Y_EkdA_Np9Fz){+$e| z3s#zOV=@a%>|n2l@pva9o65ZTV50PYiy=OjJY9dtK5b2hoe?lngH>1K(?cOYNg7`` z*eYy-Ic&`IG+f;0R-2FAB686xMvO4JyowcQndyRuIE813 z=L%kMpZJc^k%Sgiu>+C(kurk<hQYuWb-S4eNjhR zyjhtO4a|H2)AL>B@0W20j3* zqfv2|wO-yh(6lBY`?`|Eq7wlZENAjNT0(8C=u&!PktjT+^eSe;RDs01NW^NW)YNKb zZmHGSt7DF`S;u&Gc}yo2fTLgETuw9`+s@YAi;{M#%drEbl!h#GavmAoRFvEPd41b0 zX{lyPMo;i~-)cp7{|p+Zyx2MO26odp^=`(c zZ^}BCz@6su(2Cgx))>TjwYRj95G-8O5r6jH$q_cZlE(pcB~8Z%xG0f323l86{IefH zV%G&N<^bIK3QYNF>MC>zsy({}^{hBEfhWoJxF#L4hiX>(e3FlsxMj zsev<;2V#AdA4b=@uWH0dk|mHCVp<{|`C>uW#4A6rcdbRvA_-TriNRKs5Xz9p(~nGE zfl^7XXp}5sHY(4i5HhbaaoR66%MHB2*s&hrM1U{J5 zvYQdtK#K0Qu@zIsLdfi!B&Lq5V+2&#^3Ri-x$?pkVLdcG(p4KBKjh3DKSs`|=otM@ z)xWO!Be$V1q<*cg4&mTu{}WbAU7RCdlhd(eCP2K&;!;BcfIC9(HI{`PnLk513KRzx z&nE}cFpo@^dL~#7M8< zVQIcxcw<r$ZAG?m+^t25#JFe?ChVe~-76Y07Lf~3brX4cX4URMwa?dyWAA_XqKao{%CdQE07B{CL$-i^>86bcEnx0TE}S+7 zf}-C1{)uI*F9!1NxlWwV;mEnWM@6XSDJtuu5(L=uoN;oFU*kmKZ}}@*m)4^qXeeHj z=Iwt1ux~KMdd$_01&m+kT?tZMts6pjSABx>7Z!A3Lsz7)G3^|NwJ^#X9X1vj+o@s= zQr)`T0D6HmdRy;?E60wY&pQJoHXPWjWe&c{f59fYf_F=jT6^OGTq19tPEk+VMJj0j z{I>aKfB(V;sBqz26X0*=EGu#?IivImlN_oh20}d!aSg9R&O;bnb63!vc2sryHps}j zP-rhrA1%x-FuU!#DY|rC->BuVhX6kaNVjfA0tRJ`DrAs-NP{h(n^5{P)mHcy-oGjB zZa`t!86`r_p+X1+KnJ_Mo$0A`Q(Jov^$hf$vGDDjs?J^%)*O@kNsqfU&7q{Qdm2*g zW7@{yVmhd)ZSAkU_dXY`%g~;07sZuJMR;y_Sx9lXI4Q$neM?~Cgm+{`Rgx4CdWvk# z0jz!cRE9~=N(8gvqT8vjA&E6uR?6SM;r7qxObk1ns^O3%Kg1m#-9yS<~1r zF3^*Ld;48`|MavK6U0uE>0zq;N9 z?EJ0>@G;cP=EjMNn3v_kDl?4jV1;n;0Za6VgWEC=2I$V=oQc1ovr%7cJ<3RBWnznl zp;Ow9H?yiQ<4nEo__jA?e3~OzGty-m-ET1j?jQ_Q32L2cPJd)0_*Sp5^qYXRelr4# z#k?m&sUji`TZ!Um;}rP*g$O@r9@$m=^=khsf-`LQAHn$cMzq9bPDgb$s81~%9daMkn3_Z5VLx&j%LD&$l3^0H%xX9~$pJk)8@Se?c_iTHMUNR{3dATI)bH04 z|DRiEU>)QH5Gy-nQOiS?u9Ew8WdOv2TE(d(XVZ5TfisP$muy|EQO|RwF8K#b>>tBx z@^-teHJq<<5pTXSXGrmmDLHaF$JmQC_DNVcz@SMYwQ-|E@f?djps zk+fInvFcHiF&_|W?bWs1!^G^C6IQ&(o$gYteV_qGvC6vjmR>I5PZ0p7)Ffs6dv3eJ zhX|pUU-0S%kswMmUxORA8z8J)>iKzop;xLjemDz8A{mN*>vVHNaZT|^D0C3WYDW?4Y3SnYu^~CQZO&MR@aq(bQqR3@Ord&vJiqy(4oR_ET zWL*!?bjV0zyN&ffT|5+xL%?&EZ0~m$5Mw*vL#=$sVB;&$U0|v#U6RZdS^#k_@dy-1|fagI_RI#9OWEC{52l*9Z-*|(a={mllre#M(m z8-LK{+@^R10Vqj~DvbA)D5#T3jmcpBkRe^Ne^Wx&^CnU!MHufzs}8APgX~a&&ku|f zh@482-n34!<#f@zvj#?9H3z0g<_{;}5LRXw60H_MErG^9G;)Q7viuc3HIh0>?(d`o zc^_U{uih1g0G5-AHmlc8ccL=v0u!|#HX$?FK!Mn#2k`X822dafo&_71vy4N-%@%SX zezg)H0{Tf+i9e85oq#7{wdS3adTLlS*ch2?#VU>@FN_+*j4qZhz9{CK4V_qkYK?Hz zbGyJ{wZt8U`f-|(crV3MDOH<(4b7P#WU_&bJ3)Nya@y{znyaV=Ah|I;dR?9rpaq?IsAemGsHzTTY*}KV!(PjW=UQLpiOfJZjApkM}vFs=KU;^q!`{iXCNhv}go65B`A;&ubl8M#o6>||R z2;fgGwgc`wE^!V!hs(++`~aZ`t%zz6TM;5%292A$qM>+6c6@{?P*a$*PG|9mG30uo zS`i{An@nm6%pC%8_0@vzx8O#PTx@tOCqm2+m_II^rKVUsiIxmaJ(^zJvHW7 z%LLXF*}Dy@l_Yupva=(5f`Z5pc?X#l$lK6UOLUZ$IL#J9q19GeQ~;sm`w!o4SOMnN zAaVj3HBlz0P^_5kwX$oMdveQA0gYD~F=`RMnlS7K2BXFlp%N%nT00(BCnaG{8$Hoy z**-sbQ^Ebd$g6=O9(okAzIwBwhScD{!qv^D+u6=lJP5~yc4aSLai&Qr08?BL44cd3 zJm+30wvR_dh?a4I58s^_Coy-IbeRVQ2K!00AN7a@h<4O zK;y#Ipled;YFgOpS$6s>*Jqel)G2(F+d6N=PUYu@y_*ebUJhzNEl z#Yb^s0SrMYoBH9`0)n?^Y&R~4*P!|B`7W2&4^3Dndfs{-d7P-J9u;X32#&A=ryY1) z9XFC<2qY{E@|1TQ$4BTmyfydi8XhnjX>^*Z-L?0ViE9&&QunlNJ^y zv+K*W?k;X4{@!1o>(gH0{Jq9?wf1Y9oS2*^mp2-Me~dU!ew+pw^iwT?zNUd>9D09- zZeHFA5r6d@$-S2uSfSl5FFJUW0HoAYD7|to1iP3#5tmTq`a!X3n?AI=Gs)SxfPfKZfh>9Eh#xbJ z0JzY=5x;+(vy89i>qG4)l$Hen?UDy7U|rVJNR=qssTl}MxJ#Ei{%lcCa6 z1{Ucgw0Wni5Zg^~k(s5FWopJeuz_o(1{9cP?tvyGO{-7y0L02&7K5~OPk$97f?z~k z6Cw9UbXqIKa1ot2%5+xq*rZrw(FZAM2+{;->WH!f-I24I2EE1AzQnwu6$78C^Q^MW z-Ii60Yx?yi!PVq7)*v6z7tuX7R}v z9BP3)gFZvu1Cr@Von>eK)^XV}Eby_ydUDMen(!*LS-uADMndzO&p5?E=jADN@~280 z%TW%-sGrNb?Dk?%2Z1lP3Od~hBqiGU87$mrHe;r(3)>cMeTso8!M+W#)6y@-s_xIZ@uAT<4#0F#cz*R6eH$AMhK z%HhlzYKEk*R7RB zhvW#pAoCSx%hKWJJ(Xt6Yb9C+>uK1NHOUT}!o`Qe~4IKR__R^8n zH+K)H(twy0;4!AK@LR=Ue~NlxRlRbRRVNC+hAL$F(%^o3+l5_@_bB&yy;Z8;7(bug z0Dz;XJw{&H+;V1@dBz}OdB|ee^7L@Y>R_yxnh4L&_{qAN`x8Ew>Teb z8LCl|(VoCD@hOb<86LzNs$KosAP^+@2wpT|MhvtK)&d?KX7;!@+hxhSIg66ku1O>+ zUn9on5Z+C|;26kKU3WAis-cm-i`1O0aZSC!q=o&NH^X`Piv6Dgb;~0z zvuI9jwo;rJJ`il~WdBk`8{?$}sCWV+f-IL)zkBBT6+8K!Ds~ipTJ&P2`lFZTB@p=q zcVn#9b0uv(D{sT)>PTY`WeTG-3ty^wf#hz1@CLWLzvsUflQPZp5=Q&M0ebL1#7tjv zhjfGfW`gRc2sK9b*e6E*D{RWgWBWtHz$-|Z@+c8^6LX6gx-Ddsc#3CYnmndN955J` z>uom{iwN%6b890rA8NaFfF_+->rL=b7PMvC%qtOzI`%B}7fzFn z>JUx$vfhT6tzYL(ipxEdeLedK>Ynx z03_J{Z>k6&%3!`JzzpoUM1)3r6%NnubFRWwi0>&9a95}1k>U1?%zcHDzK~ot!D!b| zA|S^r)s~8&0LpBR$AQw;Gs}(fy^fW0)SSYeGCEkr?t^|R*dSo1?+OHS%(RkliY{>R zE03A`mz_bd)itX!xN#w%$izu133j3$!!}^!k(Gbc=pt}!z{=~dJ~Lq7akVrJDPfdQ z9lWO_sd81uN#G+X1MEISy9#SOTON#$F0Bj?SJ$Nr!e<0d>W5nDvikjIch>E&@X@b% zRR3;l2b+rsTE)m5ad9>3u#BLo@r_|4{Gn=LKA3pxO3;ac0(LMxD#ne?hY!V}WOPi> z6!+il=N|UVfX@=u43jHFy43}EH`7@{fxEcytgFF|!p%<9m+5qjKSh*~gQNhJ($$Jt z4i|#QF7J!Ua8hV!M}$e{UMhG}-qB!S&_P*oe9Gr5w^MsOT#-kaTcTZ#UnU@HmA)kh zi!vNSh4gpn)&;r;4Asw(;4EOwAU6qP_06AdVbh4?w;R?@f+w()%`&S>& zZ#Uu|+1TBjM}t+^*sPUcp|cmxUQ%3_JMZsT5w(NrBrbO1K&gxQ zhIBj`xCNS)Fml>;E}Izn6Fxfg6-o7NiDjDD05raWsy{8YhVuyQ-zQ3Bp50?Mv~9RL z*)lP1I2m*)i`KTNczue+RRJ!uj0U~F?k?N5t#q+37JNz61oaF2bBBI8emW&z(A_)o z`fa4CB{i!p3{BKoR%i3KuZBMf4tTIDsYz`a56{czE_SddN`j5Q{DI=xsp1LdYBOTU z0jmdpefy_FsZRDmB>XK^1$r--_=fd8MjO%GC+OC2n>wL%Rm(7D+uO&9r}5RKV%zIb zaXO7^_p!K_t52xk7O;(Coelgdolj&wgZuMV-#dRqz<5;V|1zY2;fz_9dtPcR;iupQ z!-8w-yOZV~Y64|AMF^V%zmSsX0ji&TiEg;A0q$a=r%gTQ0mt^*wez@a8(F_3i15(#}7@Ic)>_ z`bL|=v?{~}4RatmM>*O!(T(e7}TeY7|8=bf390MPnJ zSFSuBI`kUhKFkK|2*5NLPh?WfaVt6SM$b+^mabF?94<9Bv!^TDif-^r5LHFp>uAnw zAP3amyg41xE;O;?Yjl-eK*HCOYA*rvDeS{B$U@-&vRq zcz6I5wn$7_d36S1Ad6K)?c1a1o{garjQMxs@|*C@TT9Kua~NIms_bPdd^Omf zuWv2FcW>;2FI5;_uFQBY(A0lvQ}qq?Wb@u~4_XFCwpPp_7)+#$0R0?$k={x0(yk;O zo6Z>0bfS+|gMr_#tAw*7m@O*)mxOU~&jPePm{m)e5_ytrgpy<_Q^W?mTb1jWa7$n8={uMLPykEuU!6)-Bs+X_NTP{$hMNWOJ-BJiSqx6^(IfQz0cm-fq80mNwVmGp3v`BSY<1oOM6=wlF`59^JFwlYX!f?|kR z4Fr2Uyppl`ebyE8KOWVQOT#8hmRP~|P4*{11eiK=w$C2Bz!BFM_ak#-hoadWpJs*_p~qFBFh!0fW6>9gkqp;2FR15qIdV64mzkIdHQ>pcc)QzAsd zlap6lLZyE&EW;siq!Etqftkt(EI(Qwbf3yL1LvR-&wH+kL0wqY$I$t&rr0rf5X+UM-`N;#Bp035@WAB*8Ldv7bfZ)P}#d zT)7Um4kOqAz;ID>Z31FF7qG(@_tveH_>Dj%If4P`9!@l=sTF*o#vOy zU6hnketn`UFe>a(MN21k%vc7-%*5fDyU7$7!h9?M?6CV0kh^-OxG`B2{+dF466A)N zR=t%Rsmb|zoh;r(E;eZ17fjY9eb{N4*`@MVQk+p}BGk`4J^wHuy=jOf9Z?>OIWW!K#=I`|B`R)6 zkFNYX_Cpz=(&4GB;QOrH19@D3+}kGV>dI=`QMGps{*_m@ua>^ambB>CNq^XQ37re2 z-`v5~0rU^cTrJ`RdaV3<-9oBo4~*1cY8#gT%=c{V@6|4+k7;GXli447N0{DPDGFCk zhR5c?t2O?r;0<0mFjZw5!YZzG-S`p1`ygSz(de<2_D-DJQkxeBsFxcMx9AJ)1`1z0 zM&E*!5}i96u}@~HvRtg5Y;jb2HIK96rB+>`3gD90N@UE+MVULa$}t!{UCL7UMD`wl zdZEqvTpi60O?J+UW^n6W3-tcVV^zD3;D_a7O&34_Vsw?THYYEBK~PJ0OR+8bruad% zTyWzqv19zIF(o%weIG7rdT;()FWqmEQ6Ag-+i_V-`HOfh)5N0n^-BYH`^ewfao1;* z*8aSD-|;HvedA{9eG0Yf>mKL%WBV4+?z8rljf(ycV+gTL)S!z9CnxC)?~`NeXRQqU z5?5U&wmjcy#8B*~34gyB%RW~(P%H2A7@eIcgB%>EpcrfX@m5rPqT|k$A z@&6cv{O zq#Dt!>HCkJ5w!=kMa<7jrC5rH0PThzf45l+(Mq#*_37cW+WD6wDd%Tag~++_cQ+Zl zzIf;F)R?Qgp%#q^%5D(?!=GdJXYhLjucU03PFzc% zSXhU~OR3sMDiA;3Ex&hPKYMzBpI{4%bR|mXll+p9!Q{oZ`k)+tVp{h)i))JwM`lGG zcTm;fc>3ZbAH?vb}rQi9#>68#enVx2{oo0GXZRh&|e+4brU^(`c zL}|*#phF3+0pB8tD#Mc}Hn{fiowW}?f+PxyIeI_zA@=O0bfc%U#9yRsDXLv9kA#fN$ks3=K!6 z`oKWF$v{D@y3#p5Fctw2q0+X6{pi8kDRbSLctB$4h?BT^=X4Kl#$0D!G^H*N2`05Q zed`dO!J*>bkH&8fc5!i~h)nU19<4vH%eDu*j6aV$SMHHC78$H40OLf=ls z9-VuQwDg;`9kSbyxA{x!m_=(rI`7xnoT_k@+9t^iS)zv$UToVpMmA2U11!PqSuBHs z-3CmcYyFyaJaPw+x|B}2X8ojAo4%>&#oP`G3Q$Y;Pl|QJWTlq}6P+XS1q5 z)9PKNuczs4kS?;sY%}@S#T~WFX0VuZ$tu-DTHhUcNmtn2GTKj)+{}H^wlZ)qMN#bX z>oW0h7$m#r_oynZH@v~OeSfqpUAM}jcaddVtlfps$2$TD(Bs{<#2P0E29KL2Vi6HX zIp3Gjww9&|1ays8V@zV5nxei=PDUbctbpWbdaf1t`2@v6PAeqB4hLd(k%5cEz#UUmvdhnM}eUJ4X91j-g^-zhn7+^bLQ4m}1-H^{OxV_!m~`g!iSKFIuYJ zqDy0j{G;CMU|)gD;3hM3co#`1$dqe{bbOYHVh{mfLY~~6K+NbCZ9H&hxrOssKzCxa zQ4TolljJQ2sl*~brgRRd`LmPp>=7xGmH5~~XGKoEB?^mtc z+@D1CD`&|VJiXc9FXfo+k?{_n{5it@rz6k1{Cptzx_OwvRTyQDf|@w1^% zXB@7o;Cz&zj>H#_epjM#yU+&U7>pWG_JE|+%StA#j)52jnW52zakM#R zOI@P^AmA1+c`Jq0!J#}=7tXB-dfviY3^fImYz4N4MU}#h_qH>0T$?@e1sQB~&ZEPa zw8TniJ%Obx=9*a~8ikM*T-RV`SK2hDj_F_)4p_c?64lTy;L7M+R}dMg(nzcLwM78v zPK4seB7!4|Z_*B?^KJf!F@iL$J{6RV;b&!;Eo&L?PCS3A(1 z1p9(IPeUh`oznV)&i}K}^AUH`1NfxUif9Z4WzJC+)mNgmjazD~ysLdGrvKu;ZViLE zYOrz7Uim%IYws2;yx>#1D1Cfj1>`lUPU0@k_Z_3^~D^{vUzzEDITfDn#!5 zwCyi=@}4c)o6{_aVcFK%#=&*q-IFs=_fw4v zZjmzYGQ|2l84>Fx>VZe0?1p2yhQwNhWQm50aVEdhOia%FWmM0m%95Yt#dz1KE}-`mF&rBjgyV?3Xz0{M zti_tM2YEc!)R7w5orz2z7*Hi`s*?8FiouMRCtec!GbKw5z935!a0HAc!)b}py& zY}K`OeuXAb3iKhqSPbrC@|N;|K73s_M^y_FcR^LnLOM2F5Q1O(w~`pLo^cH)(e zqIE3SR&u%%sglPfMVHeq=LJ5+lSEY355ZhCiVbySn5Ap`o|&y}SYpH6B6oy_q%bK`z+y_Nt;n0eL`l%y|@wE&f2oWb{+OZQ61DBpN97> zto7OE9=c3Q-gZ~m8kpq>Ej%I6nzQ-}q!H54EzIHix-yn#__HO=EwIw`_O7^f+o$)M zr)aPGn@SvbWcd>DEec&-9kgl@K-C)bwrEcpD`L`3IB9`y0Cwy*5*wl=pR)^1slc`O zY(?0qlJu3BQyL2UklY3ll@z#v#G(|T)nqPn(P>X^G|3^@&tF>Nd*U4u;wv2`zq+k3 zQcekST7Nr5N`3k2u}}D<7sj7T2J!LPiqa5_BDvsbAo3^Us-pl@ z)4c8K$A+Q@0FFN8E<{t2&%uKGK*}c7SWh8>dy#j$fdx;V(HCn1_R?B>T=RlcaZzqNND?F2cA!BFql`Tz|UqC8=nb{@k{*%nU_MM2i-uw;S*DK7+ z-tm-JN02_n)$iL0dLuAEEEv4lQ)E9lxILMYQAFKi)QBS+SkfLuPPM9ODAthpY@pBw z;kush3HHgV$`yN_jC!mX~fV8(g!zI3G2RY3w)&a0Gz z(_Zc+pN_2kh6y(Zw$zu5KDG4kIBUZxX2K&M=9i4(%e7+y@l0Ve#6#CI$5EywFem5g z;$LYkMlE|*T5kJYFp~Fx;x;dvpdT10qxUMF0mQajvJNvo?D)_5oc;B%q}Yom;Hnwt z!r114;_VhH@ws*x=98UTz6FWN9DP{h7cSGO@3C7qE;;uZ_UECs9CN{h$el3ymaWQr z;%Mnz8Ue;`5DGii7+6neJ2h#zgLw^J5owDkgFsZ#aM9K#xkJkl?%E=mxBb;~6)SErwLx)V=WRlq z^D4>?H?`Me1G?kA?!K}T12q0=Fy<_Ky=$mwX-0jx6kXH8zzSpAPz-2Iewyq%O(KIbV-$YUNg^5u z+2UD9lxrgA;T~yg&WX`d_2xbRucv{*8+3K`R^a-Z39*8hIy{>t!Y$AXPKc27g9FotxS%T-PdR6W#x2m zNThHIv2rBx*hY*s)4~f8jC*$v$^U2yaW=0UCT(qgev=dqp#v&T{8jGrXc}`= zmpej`(kq?IkYNM|S?>e@u+rZ`3cNpWT)QU{miP$vs(pcc114L=rMDjF+?8Qt@Cu~BdXBsHekMh@USLz^BQ zrF>eo$8UbzJ4d4!DcLhOBCJg!hTt4~Xqx$(gCgt`n z3Hk~n8k`bJG=MXz3Zp8Je_g;l_ew!!bdbN&i&9i>W@tHNMEqk#^lE!c5ZTSlKB+y} zgT05-+A}+FX#dOT4DjLTk9!GY@nJ6ivyD7Prsb6WqA{Hx@{U^d0S%1J~0gXM8*v+H|=YqSyg^s?O!e(UQWLh zx72%gF6<2NC(VLJNf3qJ7o0c?h-eYUd_h&@9>yn6;|w(c#fN*5Jz0(*I!fG5u%@;| z7$K~#HM3`sVM1W17SQVf8u5Xn#R)ev&el}711}>}nL24HCP{O>u8E=e z5Nw@d{9n1j;jUZ$P0NCgw+Y4+ubM2On^WMN7z8Y~tkXAz(GxT=4{`9v_5HdGo<#m@ z_8Ib%8!Lo>UJabFyWBgwzBL~+q7zgMmb;q3YeX6kke=PKu~-;T=u;k;QgeD%eb77@ zCDbT5Su%*Sm&nOi*#7WW^0z4iWLOw#G7?DI2G)iUwmz^g#IfvpLRL87i|Kjf;60yK zOp%KBDUroV*BkSJ&l(ol$;jS?X59~0S&Nj#g@8W&{}R7OJBMg?iU7ljAXr z`MV**nr$oYayKqqdQAn~j?fZ~ z222(}lLt2)X2qdMuy;f_4)IwTtS8(>*b$o-Jz`A9Le!%5p0eCNDb^OpCjAToA}$cxqCt>#dsK?mQv!RHdb;^Q>9T=E%h+brCsd1?AB2U z!oeCd5XLsRuDh7kaT>2~OA-W|HdOaY9O5UcpSHj!Mx$GpsGtR>XdRX-=AGlDT9O~$ zs_ZMVYzL(ngBnXMUeltJV$YNTF0LKtICfkWr+3JUY}-{nS@xXq?eLOq(C6r^>RsYWX8}U$vgDF z>3FZnU36-No84g)McvVE(qhic7g8qLh8@QjZ~~4o#q&IiT9VKBsMDPnfF`FxS!U6^ zgL?LEM(OB*_4OoLN-$4sy>jQxxh~LM<|aINoE?Qvd;Me`hKCG+7yCVJ2&x3&fydQ; z9fkf(?L$lkC>~*TYrP^>z6L6CaSZT!%HD#tHShGtvDTi$22vo4BNzo`7Od_@6)~z zHl72>6}3e|{7E(8j@_k-GvNaCR^Z;aj_Lw$Px`$;yXku6#fvfi!3Y}Bwem>%?MfRc zLk&~;D{i(gR+wA(ftShy@(3nh7dfHm>7F})dz3xHEC)j_lmo&NDIC14SR-SZaTda?S+zLL0{B;cF506oQ_e*)HbGM5QJ@Bt^|NfJeZ z195|Wl3_dNfN>oB@6YFfN^<;zC8M7|*Q^#?7UJd~21#!O>m zpt#?ju%@WU;yR3g%u0?@SvODA7eW4Fd0?5Py(Oe)U}coQBZvZqkJEg*&a<9pjBBI^ zl7~{J9W45!UHOjeM3|l$Dl)+;4g+yQ=EJmzSnrXE^Dy>5++BwNkBo7fWm6SZ=OKLf zFwH&C$S^SGb7`6jtS$B16TYO=6b3X<+-*?w^;?Ii{ApI4P~|X0S%aayRNm;V+i6a| z2S_=NDE4Tqt6M8&@FF>}YPRQf&{O1;ZKy{vvW`Ik;|~A5HAyZ{)<*Ox>TE+XqAgA%yY0GhU}XrE>a z&n4Fp_qtVqYMdAfYC`;)-8@GOX(W(lJvuxXNRP4ULyVTqr*y=;%l==711n!S_1Wm*WRa0inaD%@u77Q z)Y7`@+zL8K&#lF&%zPuhsZ=elk?B61)uT(jwi2sy{g#4hR~_n9_4$FS>Rp}yjPuE+ zOO;pcj9N?bwHb4FKOJFGf06#kFuQAcM%O`^TwNrwkZvouA?Z$du+Bjtr87o+(BV>F z^JJzuY-limG`IB>RdmOc#XWd_Aw^S{JO@KWon}PT1j+_h9fgckK7SwyqT0r3MU;mQ zyfRog0N?s^dLxd`7<7z(>TkFGf;-_XErp_~Z8v z%A612uZw(@SEukP7J&?`(n`Y>AumpOUe3EMdSglLVLxgxr&GxQnQQXmCY+&luGRrZ?t?YLBEIPGzUf+< zk!-z(_E{`{%dO(4Hv5vySE@lC(F-NR6Lp!!KP!JyaEY}2D(J-6;w-sezB>h@RpBY+vK0`heV7tUp zS9w?fCRLmRP-_y#{@=38?A34YTy^|~1K^L#K9m}3OKptxy0MIV9`K00)C!gA{FJK} zso)o!g?%@?Wjzzo7;%DXBw+{fng1|0CjCAh+_m;}qsw{;Pg}v&FN!)}z?)<>ll~wj zG$6%En1@&;?mVQCMFC7&Cj(?Ko}M>#-7<~e(Br|sD8~{SGNO&kPWC?tKX?5^C5Ob) zm(9lhnzig+z`_KTi@9s;*G(4ulTB1Mptu36bAJ|?5~;e*-rKHnWNg`?W9iuglBg`w zCOe??1wnR!BK9vjM4|S>BZ>_)KbILPdt8rk=BqPe1@3#YKL(>XT{FBdrvFAwGW?wY z);KHK*u0<9(GJ@A5qz3$2~Zwy*021P%s|nTt~NZbUchnK6mWMMaR<+i*q+<)y@mwR)5| zlLPrlaLG_q1XAmR$-+EmCl{NeQrAX>s0~&zIsoUJd4m(;#qeF#dO@)P;C!v(TrpdJ zbu%Fg&1**io7AJmdn|Z%W5_6097-z*u(QvB+pou_h@rX?-S}|am7&Q2a7|u|dL^+a zCI)y-6bu^ijZ_pr|Gc1igRU4~E$S(R;VS}ffX*Y=KS1BUUS5O!>Qt~t(1zHuT#m8R zx+;AWYW>}|RTihd%#~SktI)O8iO`Wh!}CD9}iR zcAhQVHk@CNi%u3~d*`4wx+S4W~kS1b_gRI+4*EBf98S}{F&UDjrPmo~8 zNp1)w_`7P1pfsfP{>Ybia$87;;?C)2qw6LrRrx6VLAV!LG|o#JCaX)n)|{djkL4j~ zPP-`9nKss(be&pM2s9EVjs`|4ew)N~$eRL(w`|TPKLf5a?ne0rH1g*UZGV0_h-0yU zqZiO_G?%gXt@pV6Uf)t!#Y2P2hdgnc_*%m+>MDt`f4FD}TNl^oI{m$qxMgjAl6jR2 zjprb1j#(~XbLWwriYEpyJrdl)GQEGdeu3&libjjw*2{4qf9_ekG0LptYZ&vAV;n$r zErZ4CRKoR&Y4Fe1F8&-#r{lir!X<8wfcBVITg#Bp}0_?>i}Sl08iUC*VGunpzfLEoX?E~YpMMDIx!pZx*L zG2;B!z!_ex9|o;@+{iUJ&cOdRO-1tcfg@&A)*AeQX}={6aGUpPSQGp675_7!<6XJS zu0Pa!8{qQ=6a)LfPLeF4J^}Anur0)-HQFZ4+6PbS8mt5Q@AEK%6m{|MD2TC(vi$59cq2xyZd76)h&b9ppi%%k^GMr!!wft&vu$ zGUChd=9nFTexSN?JQT_^RyO&8fd&(W&&c6L6;v>c-2)OGDfX#>4G-M|`z>4Q=OuO` z6a+C_JqPNv9R}*mEox&ds+$?b-8ZaOr0!@oTm3Qxc%{{jW%Mi;Ki&J6K9erQdy&6LhF`ANI$B^ z9U1mu>UR;Smnc$lQ$UsfFT&m_IM62S+Kp{YY$p@jn0R8_ww*lj#I|kQ=ESxpwyppD z{@PW$_Q6-XkNTv$`k<@2dfjVX_sJ1^JDr8yM2mPO+e#VMy1;D4vs(m#L^|XPq$3r5 z&?_qcyY5fpDPU!@8Z#I<3wm6@g26#}A!*z%lI}{PgXer?xbiUTMakkrcu(FC^%cKI zT9a6^ZFgG&dH9+q2EklV_NT=qEC!RvonF;yut)pX`JENwuNsMej~^qc#awL3(M@C6rJuCEcHkdL4$ZQ*5sO99tw*kK$@m$AqD>mW1Ld}j#V z0oL1w8?ambx-euzye)lO#Do4(w|=`wB$%bS{gGd`7mOVMu5wpmSkc}{l+#lVK;OOb znLHlrtRV1x$KO7-3i_zqbIR;n=c)5WP<`2qfCZxfUSpOR@^|V`aI@j3{nf;vbeJ2! zzD(aGYnsc3V3ZjoE(;u(?gRUf#oN#ipB?#gw0IvPo!vd=zy6}Af_pub*g!NOitu`Q zEVgM)G@ta}44>)s@gmorLlUIu$Jk!2fHfDuBfzuN>Qpn1Lm9F`PNGxmCa-1h*ASBI z$>UBKp^@PBb7k=@B6})TcK3DGaz6t(+*}SEY_McFUARx?>ARuc0e|bB$ z{|(UF#AmI(A(^sstiC8T)6dib0`57RJ((XMj@)mcAT{P9$GbjsgwwP<-=s^U_M;a zjoTar@N~T>!SOe-eEnVy)|^?nFu9REruXKK8NDPa#E#$_>3z0MzNw3VQm355pE4{@ znF^k0jpS1lABkr3skdpei9(`P3NP=1aD>OHPHar(jnQYV&!l&W3-bFbz24hJum?IqRf}Dmuj)$U???fNt44JJ#D7?h>KP3&|Hh7i zj%LCtl#{E&MI{K#(`AsU4v>=86RY@Bp@SatO7taQ6F&v~ryz=9J$18!IRCiQOaQ(0 zfWvkK0#STKPUE?&a$&(^^3eaQDf-O=l6-X_eK~V%OJnalAPL`}*WaK@VCzD|=@`#p zfZlkpDSToZMFbWIIFknKRf5ZABtm`E0rmb`mVqVMT^nqeK38IYBz;`f=iZk?+v8oT z?pK55zoR2cZ`s|fyd(=y@L6y-o;{rBcq!_P3ay6g`t0m;X^U@7C13O7wEPob}4H<%;K1XD2ofPHxdC74Dy zgD16G^uK{}cP5L7$2?^q#Nx@46cS*f16NmNqQ&m;8O*B+$~dqn9;l%MGtq$-%Z22U zds$T0|6KjYx)RT7p2aadp}d>`j@Va>7bDQx@E-~U7tB^97VsaUs3N!+%#w@>+=JOh zjp%VYpi#OCq_7=*%_Ktqsehpkff=H8STqgwV*VC|H=Rx^`V{U^CiH&x!s~sEAj;=v zO`;Z2e=)2Rp`9Rfz=}K2o!Jo0iVNZRljiuzr^OJ4AlMY|uGq}a7tERgOV%8zdEdf@ z7@@_~El^Fanl3p=A;<+qH=OYr77k?2I=p4|!Ne>zeahQ5=sx+!I~aQ2^obEu>6p@{ z$@skv+snRdGmFk1j!g2l`3i_YF0|Kfkezr=sU8YZE7zi_1xRrWJjoHLAf(RF+>NHv z-^gBUg%VmnQxVCjr!@)yd%P*U>P>k;)n4e8!K`iroIwCOhymW|N&Q94_wU_Iqv2Pv z*UeY&?#hv_)>F>hb`a=omMavv5z#rbN1ZM&;;S<6QWlrCZ^4tHWO6RbP_A^vT7EiC zpyqXPn3!6F5xuJb2hb<-R3?;#Nda>fY$^cm3~XOPG6YK;)n{~WXtD&V1ZE8#nbd( z&)nS{+`;gJ2hS?{9#}BA0FAt{|F|>|RUm7%2y2A~! z1p$+x>G56Y7b{`FyL-1akRVFG2Nfda?;KyiHfiEdZlcK&2of0rGwYkMwM2?j!~`vQ z^gSY<;&Dk%hq*~dh5_gB_e3lahV&;F9@x(tsv+asIFSc`^oOGWh~Ct_(F7&zybvcH z3tH^XfhKwoIdWnkI=ki9lf(s2KMQlSMx!1_3$BkgyTB z5|cP4RIAPrcE*2rAB=W^?G2#=fj;EF2|XMeUUxb#pP$Dc=Z)3-#vVtYj6w|e+oDjQ zMhF(qAjLGmj6MPf~DZL2<}vfqi#I*eo}kBZcUUE%!fl)B@%K1Sj$lmX<{A8xJx`?1}Dgv*T8qZz7aM5JsQ)qrY;t z!LM2eP`;bGY*0$Hhj%#Qv6eWtxFpI31UbS_x3gSlVy${%Ega>6Kxj5Ga68-s( z$itsgLIywlhe1leoPZP&Y0_D~f3*Cbo)!JKoTvc-riQ5RF+M~{+L&kThs#khpuY(< z@myJeeU^Xg+#9l5CXK3uvjchO&F9+G2RXK-EuzAQ@>TMp;X1$!osZP4_oEuI(IvcN-Vfh+a-|muTj)W6UHn2Vz-@^k7|K zl-fFbE<;PiDy#uVnbyUhA>^OrqviR=qPAqedzD4}GSf}?uD%(h+RZgSOf7Zg#q-l~ z`G{vSs8BDMPiepxVr`OqT<<_r5f4e&EtN{%)F&M5sX!m^5ZUWtmSD>kD)P~b5o;Id zeZBmbk)5Nj*YJ#Wv%li}boFt0IDYzy)9E#6S0lG}Z9gMtH_h+nHoUE~Rrh?&U^l%+ zq~6y>QH9cLu;in3v!jF<)U9b7 zVfVp@5#tRANVgmIJYu03^6VYorPvF1G4T}I(1!QEfMBos2L zyoD)bm3E+zbDk*1yb5ACi9`as@;YXAGW+y!Ts3=gLx-B$l;e2tjEy~Uundf3Ts{p2 z?g?>!GU*shS%L72as;N5h!T%gC{6k(gUCAm{j0{GFgHg0ks+`$SVwKO6F6Debr_62 z!bHPe=o=WcSkhn1YJG}k>Ip!TjlC!zAy%J8?c7gXkG@`Vu_%(vF&5$;S+`OpnnE5( z$}=U_Y7*(Q1me}Wg=s2kffANxc|o4LXt8ZbHvWgV^+|c+`O|?clnGHOoH0+9lT> zNCT}xYF2$Rz5!28xJg`$jGUe8g#dV8N)S;@$%Mn5L|yaN1)8b2QbgrjJ|TiTy5c0X=r1~lYC5#kZrG>v)qrK+kXQ#En^Y#NiW19?4TC~i6= z0tOb8kuv$e7Lbr-j|?YUuj`Ws%ZwovH$})N>$;4edjocPM1sL0=xVGujJ^s%cV<<0HV0je0>tli!yrd=sn6i$*ZU zQFfYqt6}KL`gB}Jon{d-j6>-lDU{B7TONlqjB93%lmS0tvKC$fo03`gJ4iS}D`N5Q za^_X9nr)>xUVWqUa#{gdt6p+VMn{T8r3_a%h`3J;b_N$;oe5pDe@gle(cg zVq6h5VmVOmfmN78;7{C*%5*aXW>cgg&_bUY*r>es%f=kBQ@Kqgu_z3zN;gn!Ik>dq zG&sr+HQy&$8LS*6xZR@Xs;)7JyI|hPt`J`Z9m8U4$A-R_mOIAv6Yo=y6XaK(>bWaA zc$7L@GE@q768~QX%wa{BKZS9qQV45aXVS@3?h^nWrg#VYFZwwhK~s3#47QTVNTvaf zI4Nkzyt!M1ODR46X*cm%C0_O2KPknK+{4SI+g8)z!8(fdU2UbTJvrY z5y4OYMecO1ZPC1}yQB}aToscwLo7-Dd<0>y?1fip9A&A~a%EGjmr!5et<>O;!>i`P z28|gojkxpF|!Hw+@%<5Gat}ZK(;1nrgipCc_ zIVf!_-J6(>>K5ymJMqv-htk`$$r1rYgE8bO5PW zl#6j*W7;?Ev`*H)tw!cbwXt_`NvM`Sr50qSn7>qA)RJZ~S^1d`of}Ass~{34Nnb2I z)=@ov{Y0nn5KEjg03Q7^Z3Kx4q4WM|toVgIJobSN9JwhIgAzh{rtxD9MU!r)_|-fW zug6+!_T{Weg399WwViM}D{q!#rUA1FDY7G!SP-B(jPhP8!hOrduHAi$IWwy8!N*z^ zv91ws@eMyUC>snlMCam&ztzASBP~vgtyj-NF6IfHE>|L9~)TmM9pGMqNdlctbiK(l zp`dSxeJaY4XyE^pGLHpDKJ0DZ2Nix%+eq-t{U?Hc&B}!KawT&wzYj^%=-#6Z#Ctrs z*e`y23H(&}tH=HdkgkCWasdcKu@vc$p0d@r1Q(9=t!noN2CnH&s#E}f-X|4Foo9DU z?CG~4+!LjBm(VPmw*DouDU{g)B}-a7_rw^B5#2C3gB(s3v-aw7JY>z0l1r`-@dAm+4Rqa37&tkzsGPCO1poq9qRQ$3C&jU| zHg{@8{>h8|Z7tTb!gFlNchF~ita4k+z<7R_RBx4)k;bh0Rf!8j zJ)E6Jj+)}Yx7_X*?&HsXN)n< zjy8+8l$W=*c6yp|4On{TJiZ2LC;mN82eEmHC8XFMJk4}Xz#yJj+tj|wPvx1 z<@=4dom+0bM!R1(PS|O;&~yR^@2SC+vs3}4+MW-87x7lccI&DA2(HBT&RYZ`lFFpQ zpJG{^pXb2SO6XdLXK2a55`vCTNxBeb1r2G*!MDu&-C;kt`xhd-ufo8Z=ZR6Q2NXID+okr_<=>TyQm|F zTT_nG9kzJ|0A%6DMb56K#Ws!Y1oE+lJHV2w^5|t7@hT^y%UV~(F})jP`*?@zXRW_U z0yGcg(89@GtG#rZDyngLQ0a(TmUf695ejlU*z{maml3I41X=0=@&LF@P#L}9aN4|T%ITmJpS|NtAnKPZ_m-+%vxFa=D${g z_NyW#;&~Ec-lfW%mOE!4@Crq?sp1cK<{bt7h3yGv(AM%VlgRJ#$hki1Bgc#PCQ-s8 zWTaNoL90Y?DTE`#7zkv1kqgDMIA|7Y04PDvt7{|0J$f0t=q^>_=H$!SpI*8|J zY97ttpROSUaT#5oLg)3Ck?8x5yTc>LdZFEz-EcZt6rDU3NuC_y_Ys0c0NoS}XG*sM z1PgGG%%uQB^MVxc( zc4B+1HWI}doWFQ$8$atW#A}u^wKg5_2I3m9bsUF!WTGGX!^V0VVrp~KqxPY-vrGT2 z?d|D(uRUvIiw9PhI(vWRpv;I{x5r}wxT8Px^)*`ky|EuYgCr&&lq7?$>v(H9bk5J| zV|N~zYo)3rxqj%^Yc!#k>W=5I^Dv54mc>MmJ{yT)+CgC$908!e8tpe@CNT2D1N+9W; zs=urb1roE(0=29z#KDthFGJ8Y7I@GD{HD5kw0b?iz_Mye)$yD-mMb3|(@u;O+dEoX z{kxs?eX46Y#!fXEA5ORDCn+ERDlq?~{B{tCX_0&?Ev^)XW$m;x?Y?=a(>(jVI7h@X z$bkND76#01BVsh=2_VCaCLB- zZSA)4=<6tffq}&o3q=&4LOxP^VsXH=+QpK>r(FPBop!$FVGAcO9(^g}xgPJ~-eo8+ zUyMrbq}^S8d1=e{jVstFxZAsZzS7;;_=cl4XUTu@O>>75H$A`B`iZ<{8fN#5C-E#` z(^=M9xMsyW#&{GkbgeBCt0U4f zYF@n{!lC_r-`{&+`A-PhBz+>WPGiOQ!)fAiVJWme(W#H?Y`(6X_GEsqwxuJ#FWs-; zo%*E0kl&N)9xgb#?nRtXMPY$TRuPgQ=YVU;V$tq4b z5UDvk`Km|Zcp;KXh2-SqYa6JQr9!6LzIEG=Z38)Sbv*yJ30<~QgA#5+pWnA?=-r4nPTO(`O?|o{ybzZ z7bsJ&<3@OZ9r7jqY-q-AVXolE;<>;oeyQOUZp;%c42|{>JEG6l#%N{l3|6ySZFK3v zsY}N}PTtw+UHTw7V)z(bF$c?On%Twj1`TzK;|u#00aqYBNCuomk>dj*a^q^(FK9!J zW;FwJ+2sikFVwQY=)Zp?C;4r`g$s=6Yc`}(1P@|?luG^(PxUCUGJQmKV_PUqE&^n-DT|i*?nZn0b$S_##%10E{<^L6V z9eRU+{b9>PCLH0`T|Xt#-dv597L}i3-EzR{b@HSMPKK>j?wGf8C*fkC-yDqx&3?t+ zJOnFwerKFn8{&OkNL?*g<^*R4;|EgLDJIsDg4Xi(ru_DzC;nv&rx1#Z_-EN9U`_CN z&?hRwyWMAmd2$v$9iv?V0fUB9gH}Pu-Xg z3t2l-5MF`p#jJnYRifLCQ>fpMlTI}JHrqQ6TI{C&R$R)sW=|MTax!l%Te^DWh2F9Y zcdWy0cA=LZ!I>wWf;0c!%5EB3?e}ca7Lcz%uUR{g)NV#uM{s?S$H`~a8uSN{5gd8H z$^Z8oL$u#3PwW8-{9pl!T@roaM!G0CbV%N@Isf-eN}ct1y__lQ`CMuNx-X5zAqz_u z7qliwA0dzTdXLIYQMrn82e`YnHmq3%#OBPwB8e%j{|8*)gbQvtJyhX-rDh)Miuz@b z>zxyaEEB~d>A?>3IP-VRYt$?7{CbJd+pdPb-!hgiVixUyJ+yoMTt8#n#X@9kUSc!k zs?YzNW{sjtrye}X7RoyIgxo{Tl|VJ|dV%C*wGcBU=zvvYR*1UxXL{6-$Cp5E;g*YU z^qk!-Ip<+!)sgY2G3U>q+Y=wyJB0|^mcX96^~gsj;W6?JJ2|xvHJd+x8{I_WP1}Sz z&-a1(&`3y$ncBl`!qlSTgKa$%zi60b8lBtFQp;c=1M`F#}2kJ2%g-8JN4E1gxbWN6T0%GHQdMi>?i8HX)>Q478?^ zQ?0QJC_NMt9QgtC?!hASv@3hnGa@KpwK5?LbTFH=Y)@pFn233y{cJtU-Q8UkFipJf z6JclF$@O6W)XiWMo!FY#*m+*}WC}S6uba40TK@M>Y7n4(WUxCLcamG;ukft?A!D8F z{`nVNd(`KGaLwz_`cwiOh9x`x1QL#)3ian6#o#Ujxp?u6KftE_Wq94;10yKHup;wj zYWY_sxljocqk}5M4QV77 zJirGez85$E&0^1*d}E(N-Jm*5e2Ii%oRH5CV^o)kZ%!{GWu(G{a6Zk|3u%eR{$lp# ztztx1Z=&C%;%6x)4udWU;br*Yoh4~C4qa+qUH;v1#?>l29S6hDWtv`+& z`2jn`d7MMovy{F;I8psJW%Av`>ny0?v17qa0fFSy=BzSxgF}lhd%|&o#ol9eAUIt^ zHNIy5f`|!|aY|R}77k?5NTu_A`y_d?z^mQV$!Kfx;O{Vk){3g;NQwJ7!O=Kk+L`{* zV`RzyP|xp}We$3rTeu4CsCuSAu$a=; z<5pEjj!v?nNf@cU${7Uum+ZlIp1;vQ4QtU|*8gZb zfh_fms$EvmFZ`f@-WnvPaV(=DGHH&~sWD0IK#pWiY86YCRBdI|U%1x|ONVNchzSa) z|LZjHTE=pEe

Ybj;DMObcbnCtN#AgpnR_B@t2-VQ|9NsmT)1!37U}i3U2?>NVgtdPj&&sl@ zJ(weV^8$YlW!iR)%4n3z{B)y^hR7f|*GZ&8u+m~{ukx>~=R!hFCaY8j7r%32%L$1* zf2)7ag5Jrj1W8v!P?rH6J^-k#v+_+G+ESCrs70W*lmZ<`$C75`5Jy%2K>`ELpgpLt=`r%@uLrGR$$#w%dGB_|)GIiq83ArEI7n3y>SGb8 zq||Lf&sz@+D3!6VK=@V~ZT?9*mY2?Ry}3ENe{^Zc1J={eLIF$`>lbYUwj%Qo{ebB! z0`Fs;f4MXTD(oGMH1Pi6 z8qUJhTUqZjn5E9)JwVxHL(7rE<`rHiP9jA_0)OA>sO)O&ZDN+OV>8PQA+A1dP}kE_ zwG>@{^=BYIy{=RLen z+d6tF{k*swCpQ-CKM2vxn7}iTxha;S*ncm(b*c$c09F7LCeYQ`Oihl_IS4;w7qHfr z&840wPe!WTI?N&7x@GO35bpc1)~eTOTB-U%?YbB;I)!+AF%ib5rw=PCztg&^th4IA zMWsEzU0(v@z)lAL-((Th2$d2zkB}P>4~jqXjFYf0JSWP_@88ex8dsRPC$b7 zuyHA;&|s4(y{Lf~BLB3x@`IjA!TMS6PznKdCB;h(Sn1WA`pw~H*a31ZPQRWmt#2;O zxa#fGdx~d+==l{+VtG!TovOgj99>BK7NT)f$3EnZGniCe%tZXdhw`ywx@UJECnTmK zL*fMCLm$?T!FH^miBvsUmWF8AB1tr~H`p3BTYGwQoX?VyQR{G11rP2JqP|4xK)P#@ zeG6R~&;U50q2?+cMb6R|b84GM zj(S%GoyRpB7El#`oyCEV9^*ovqcnAlcqVE4 z^|;(ayRMsZ0sZLDiB&tUa!45gE{E~!S!^B?0$r!&FRSa%rR_)8!3D^Wt3}|Zpa6RyfNt4Ak55VS6dJjeKqLX>On;Lr`pf_ zf!G|48#`T4-RQ6cbSer zzKSCYnF+KC2l(pWyfNs=$P&VAJ__>{%^0j9q^q};=?Y*x_ktLfNbH}JeNI$B05b0_ zxZEKKb-4C^p)(i=rf&xEz8;+GG0uJ8kES*2{Lo|X7ow&!J5@uuUSk@%mDwAe3|hkb zC)fBCXm*+#HeXlH#7XFuq~6VF3eLdj!`eV~lndBX^DvyYfh_frD{ZE3U_P2NutO^;VRJu3}i&asO)+%?Pq%A zB=QqCt^Va4k)`Ir)d_#_x?lUy$($!<8j_8aallaA#ck#pwV`@BY5YB4Kl2-NjEJjul~wSBu7`n_&C3 zq5*wnI+5E#(rI-}t!L~P0hnf6!3=R8&!6Ua?HD2R($HB2?FHiGjSdm3#?(EyHdd4U z9m>isHRjUIMl(1##-mmMY9)wqRACFbF{;GPBb+)*l7aUDOvzJ%m({b^Kj}s92Zh3= zubJ@4%l7bH3@t%=gaF4j_0%2X#-o%)+*gAFwCk9&X<}i5QYmqe7B=onNCiT>af~0D zeIDJ1(x#V8ck1PAVRgL2$UNtP*4sdr_RNR^LB+2>&NCwh1X`j%YvjyGoHN`oYbkC$ z?yCAy{_}7-5^sOlc-AK(OL)TAH{FZB<(o-Doqjwo{>ePe;LxXy&WfW*w|@>R4^5+0 zNHAU%4)f0x{@*{OrbtW>2it><&C~sMD)+3G6`7>++1fVbm099=kCu=E^8_!3k^j!N zw#$c_?8U4#)Iah8*&}U!$F#w>0bc$Q2yvOTU<~Pfy(MI$sdOjDO*q&uKQn!H?m7`` z)$?Xw@U{FmL0+rPChR9blCEzhW|niVC-qIzT;AsD8XZek;a%Ka(eQcYFWid63!xHx z9Cmp1jk|x-PxnRtPOX)pu$~QY{wzXB2|i0^DMrJiRsDnjUTDmC2Sg4mei*V@bBi;& zUDdQ38!bf??v@GqC?pMohhx?>kSpt#FA65|%*%e8po?oDMV#gqC~fon{rg8DZ?_Kp zY5fz?=`T=8S5!rg+({yUL6A2`aJr4glZ`ZM_+RSBK)#a_!7=RtIp~T?&QMnx+btl;isVrwO4s^(n>xg7X%>jUlt;<;Zw059*dKZPo6z z=3vOe7wo8ml@$X;bPJ`Q6+1n2k1I@zq1N>Nr~|GW{HRVN7Q-8~3+q`)_) zcmH()Yfkw0lS)0$`)JKKUyvuefsh^H zMk9?DG|ui^^sR3usz;LV)eXoii2>LAnBjd-&os(?ji>zHCLH>g*O4?6y=sG<>4tT^ zlba#;X`4)4GQ`a^ z;1z2v&ibJo-j3!=Z}WXNFFK_R*ydi3xzUOfc<-esX#{Tdf34b~#y5eQf#7kWk5M7@D z#lmv5T;2+y+=CS>`wm>c(uw8*B@(8Ny8d~z;LFX6-S#o-5EyN(2GyyqR)Tag%MW-Q zdZ*5;fyti^`|Rku%SJQkoLgPN>1yv+~;oKa1=WcZ7cYBw1y}I4sRC>J#{UePcWVAbGnGUuebyW*I zlRFxQsK3<2jVOxO8sLoBI<-lf?MI{-krH;x_-zdPYYp-L43_yWT1T%t>ioC>vUr=m ziq{_Was_~=m*K|L*zQLC>F>Am#@w%mzt$YtT+hEaU`LIQt3N+|R}NX3t*(i_!Ph`| zWuF}L;fD5OlCAKsgr%j@a!DZ)DpKCL>PNHm65 zDp;aBH!~~o;?IP}QvRnF4UsGY@hC$!Df5nIuI`RsFN{;iQN0;vQi+{cOc}sM_Ls|7 z*V?W2Hccku;|~;VFe#UTTjW@am_<@xAC%MPCv<1mGe^4onxBz5Q}4gaTZ@4Hr24kH zgENXiK}Lzf-Y$==E^XHjp|`#So;vyJpVY*2p&w=O zisY64ZscD_VB*>AT3c90BX71nAJX~;Uh%$3p9fyt!p%_*(2CZU*06HC+HGJ<*lnMJ zprL~=o`3?O)zI|9YkpCG`A-t)(N2Pca|8HL%Q)fmy8wY(uI}?2y8(T(I6<84P z6Kzk+Q*F;gMJ9|?lzMI`>i^d1kPsMA4x)pA1c-rvNd7N||J>`XE^rRu|JmYywsG}< z2LXk61O)+s!1(U~huY_M|M~sDa!bpSnZt^csYg>?mI=)FTDHPLcjCAbEr$YhNPjC0 zV~m@{C$L9@+@rjk6B+6o6Pf0i^eST|Bv8{?Ht~XAJZQv^b!73lPP;R>p0!d{?L{)F zMEgG7!W#GP%>aC-rj4Szmd29{oy0lf(Jg+QKX8yeMsCkq8IR&4?iaD#ApR>dT;DNw zjQ8HljG#tB+Yplxz9#&MdTY=bw;0<#7$o_8XMKk3kA{{M=Y=64MufBS>0mfYwbc-$jZ~ASmK=o6VWG8%?pS|-DRSUCOIOVLm9nr@)+bF z@(%B&O#s2e)2e7@lhBYk+H$dz9CQN1NknG1cw3IJL8cs|OJaek!Y~hULUF#%QaTb_5`Gxo}o**uQ`-|NY=D#C*vcUfDc62TrMhH`KWXZQoRmYv@?J|*< zyi^!;os0_-MA*CZ%X!3EGPjOsTCddJOR|;k1eCDkhI$|@0VKp&rq9U>J)vbmAvE{Q zM1XR<3|f5hr<%hVEhaimiTDU%mX{}p3qUpiK}T>9FQ1)c*`>YmU?ZF_v(Vj90^y*w z3rzKY+L*EZb@9;ajcN645<%5%)tc)>dBzzuqi9(}c8%Vq8cc#VQcThR*-YOnoc@=H z@w`e@HO~T^yXs-8z-JSS1}J(r==()cQ47*n$i+W4?{jqv^4_@V@w1SX z8FjD!X;azHKIcj;*X?}VD0{jR@~w6!tH)S^%G~^7kVe3d$IxEBW*RA0E@tfzva^TJ z=(;urxmKRf&1al}p?2+0Yv~HrMe4+}{o3RBPR-h>m?=WkHw?;C0xJ$aA3N9m1F$r~ z0I0Td3splpR9LHa`BWGnQqY+>K?}i8&y3#jOV$TTi9)&6BwqXvS5I(h$-5Lj6yXwq6NY+;>IoQhMA6JBDX6!w0KY}aos+wo&Ko@%zGy+PPix=Jn56GQK@Gdc`%e1xWUgcif0$gi(^07^_%Mlvp z1U7zOoxb`?M02Gz%(To!f4SYJBE9=KS594gcV8aGy1{5jr_(KbmfW5dOM9rGoKrU~ z%)ftC=EB1IM*L3q)&MO}@fZ7*M{UxYu)g zU&SFUvRZtjA^cZ(2ZS}E28;1_(6Q4%kGMxCPa^ucD4*ZB!o=~II`1uD9&8Eh+l*Q%m4?`3W(|{Nhr>66mju3r@dWuty07129U8~#ZVS{Z$NOUkyavz1 zrI{mSjoyeZ=s+hC5-d@6k1JSxzg5t5$Fe!8w^jCkO?KhS{!1xO{e1Png*Y;3T}Jji z3iyRxm^-2@W_SD_dBtidrjmJ~HGnqMzxMe}CEy1-31`)i%d3B=Kx@O+;@ZU=0+p_Qo<2?^41zjYU674IY;rtB#?9KT$Ms~L z{E^z#t&}jdnGw)l%L$ILHM{4;FsSF?Yxx~=sTjv`v23WrY9cKU0}m`43YdXlovuE^ zt1$GD*Eers#850~i&G{(O&Ups%Mb}1sM4SnQqDjq24;Vgbv4sJF!h|B_TaM*6b|#L zk#<2NFp>yK>(&=ddWc=n(SHe|H!ktZZut!i<~2W=G;)#{*=XntKv(beB26n@eVzPV z-8_8ftuSdFy&c__n7Qw*!9vNLt`9-_c@aH3;ViP6dj25kYCmafE(Hpm|7=YwlsEhkv-^EE3TedCB&C&__XPNEkqX9M z)e87iUt5A@cgjuuIxh)F7H9Y+1Vq#MSWf_Va-6#;6Cs1+tSf?uVGYvPL|{)u6d#J+ zl5d3XVtJ5Zb6^2xc*PVXiYC+8Mjg~otdxZhhb?L&-_SOcxCDhLVmGsUX(vjmdJIZ4 z?rhpmf5km0`_s;BS!!J0!rlvl5j2=4LuES+dNR{nOHt2Fv!T{xg@-<6pInYr?E)ar z!sEbMY0iq`SL&I1y^lbIslA9c5*wAyIo@JhcSXD!y6Asq5-Hc<4>q%%+BhNuKjLXK zIhWT}>%2mbS1MT)M#Z}gcQBw1Tq@tjgw;l0bl&d-RooRXPO$Nvh@UTbDCx(D5=7=` zP80LzO@=u`K&k1ImUcWMAzm=*y$3p?1Tab`aTDR8g9Y*?V@z3X+gqY6H-eRyGXN!qw;Ipn42L;`4#xc!kS znT*HSPpG7|_a9+X7FIuC*6=2qb0Cm$j#buEsnv>Jzm#(82vW(g)ok*qQE=m8>hX0= zBsJ>UxfAlk~<_@oI!mC6&&%6y}o`dz0jzgkM=3mIsU4>nUq?RVuF*b^cd(nM^)RCATxHxuwQ0iGc7tAKY&kpq}8Llg@l&a zl5;9X;w1xU=$}^rR1W>K|#+9$Jp82BGM0zcz^oblfv8xYK2WK%MeqqJ}pCRSUFE}Dnzf* zuAe1nl~W=mO8iN{piwyNa=f$R*GvO+*VRq2p_7m$3&bJ}VgcDp(d|bkz z2C)nOnYz`(0ae0?f^5`;|J)w}YDw52q_p{G@MimXnv?@`Mx*#uP{1BOz0bmT^dpfo%Y3MPiKm7_s7SukB#4GntR}CNHlqNy2mUopVR|( zngf|R0zFUnY{ip|V~XTCwDMzpw9B!3`XC-U^klMQ?-eiA6_5EUT#6X*L*+vcd_PCB zjM&5ev^g(~ZURK{`GaJ>8MQ&*9tc!+aSd`E8E0BWkL|^R{R>iMVZZFv zl(Z`fo9STrv@B-S(XpNFhut1>Y;7OMO3xJ;lySdje&urKy;%vZYu8H0o+msj30y)3 zzjDl8t+_H{o4!AQEHV=71#^2TY-NTL{yjS|Vv=pCHwSFsmMF15HSLBM^2?g=#`ma) z^h>^mUfy4EKgw$vs;q6SIZ9d_aIH)if{RM{vS_51STOWoU|7pVan(i-;9SW#LGiI8 zY_^67{|{g97+hKOb$iFQZQD+FY}>Z& z{bBDp=UQWocL7LJ#W>)Oq7t%LXwDM-W`=|LE_rr~L9iwYNi?jt1r=dL<#g$iYG>%& zlJK%?pMi=ZF#uz<$3{O232B*zr#%CGE^aA!UZadZ5j_jko(Y*c{)gF?RoeQe~L^Sfgtt4TAZdE9iR?5myNqc@9K=quz=6uJ~cpZ z7%sV&w(un6-nSBu6{4PMW%8pmVtZZkin}sJ@;(ThR0|<30V5e%sXj_GsZ4xd5NjA zYNh(zQA&*?tkwc1q?C`iBpx;C`4XJW`2#~7@nB(eP`;%sV*g@ED@9vO47`H%(5ksY zM)`ipm?BL=d05dAO8WlnqAr(e4cn}fjjb5y4?u|_Ornt!`!!$7kT=8^PNlF7toe66 zw!1rpU&ZkG>O4!nZBfjPuk6=GpcljNcYEa5Tzm>sp#yMo-@Rx_Pq}P+RZ&?I%jMa{ zM9S|;*H6Q&b1fZpP1h?SJJUy^0b2n+2TJN5bmuH$@*4Kw#w1#o`M4U>F$c|yTUVgC zL%@;VCfig$&@By}{!wkt#3h?cZcQf4sVg;YPC&|)4{X&v*C<9(2T^1oPbw6NgDHAg zk>l&bS%6c+8^Sm_VN+ed8U}hW5yh)kBABILq9{~g3Grgy$5(jetFawTen@c!$X+_c zQK7f!?dT}~;IIZtmvZtSIn1n_$lyVhR6yN+niHj)j#UCRWi4e~8%skSVKQC35O7wLQyO%H!ehPwl6cUWPgJlZvcTR$SwwV;5ef@E<(Z)1 zEK?-1t}fToahRjXzzpCG4U!`_OuJkCnTqB_i0{4=oN{QZoW z!d(P~qlQI7D$z6pG>w|#t&!t52*CBw1+{}t-V^MQi#_%z*?vl*2eF5E6J?oPYgPw! zMo<;7Krz>HGpmC8?;IH7*&?o{zQSm>-v7;t@X5c&GA2k_iVz6pIH zIMSlleunQWZU$1CyAR?!5(bY0V;TjP1u2ypYEP~SA(owx_Ef?%?e=*bRJqxMn9aWp zFY*lzz<6+XhYVO*TYocr|M83TJ6*ffN1T?6nU-gaUY$K$WVDyp?{3ZR&eZNz&q^%w zU9@UEBOv51U;X|ndtujR450F|t72Zotg7{F znRjqfeKsD5(FdYTvQy*>^GXPf#^b1nRv~_E(C`0ddYBTBU&;je zpQ+&wDJ}V>^uW9+NBO0+07)ZJ3V(Fg>dMm4~On^=;@|V8549F8v~u z>?aICH`G(pIb4NRp(8=G17I}YmywuAhB^f;n49-xE;^@9C?{%^R<=GgSfYGCXs04n zMXaQWfz#8p);c!P>NmH-ek0-p9WV`3F6hWe5r*qKh@{|X6S{{)fVV;Vt4??`qh1%9 zctE(i`OU4JclL4F>h@+9W@2 zp8UoGqy;YcbzUZ94y$%yHEPu!cTFLa zj7FU)N+0if>F8kj*P`Gb^dO3u+zM^??S;SoB%R>8J9R#b%WTRsZj58;!pP*fMDP|V%+1WFBNE*l+%ef zm%1Vuss?EqTnSGQy@SJI#&N_;`Ezah_ENG(FFf4WmHfwLj956*%bT4y?ecE_t%rH# z81ZD43EZ;Jg4(?|bWkjr4JtItgeqaPsxSRL9=Beh&grNYIdSfqq|qJFK$EJD&iJF^ zdXw7LK}kk6KpyasO~&OQO^$FgORH*{oC998kobFq%ofdRQ*(x}zCMjb^m@|Tk~?j) z=h{;KLdVLO@g?WwK4<{7Qn@#d3G0W9nJAF{l!d-Mpg(!b5sU{cP*Q^iGuLWk}cH z<$S3Eu--}vQzEXM7S|`)Bav6krn)}m8&F9}!&FG*>{}v~RGnU_jXo}|blfUqJNO8k zPa|y(&_o(7E`ICR+C0Oxm_JJ~|7O7p&q*L|z~&j}29xDgWv#Z&MvDj&QTAYhHzNHc z*%{I?hxAliAdokF{->^2dVqGh*junRzZ0GF>o(`A2~ip@WEywB`(u98`gZT-L8XH` zAm%z|uwp-Az8`Z~eD9;Eg%3dN*TKcp!AB4VC~&^L`%voebanEi^W*uxO={u$dX{#) zn=1f*xbFSS5Bnz^fuN?-^X6mK?xw@mvB!q{d_m0zx}A!?PDL2Mk9RE5Uz}>q0-^_o zgYi}l`l-iEdjzY|P^=!rP+Ozs$1i3-a)A8F+hZ?%pvD7R1%w!rnLR?c{^w*46fgMzVinaKE#&cWX<}803%5J^4RqDiL;7 zL!fbo%n2jYBjJRx9SCR>14ltg1PReEz)LzhxZmfv1RmtBj!8`{F;y$vi-84pYY;G; zSC!d4mY=~Nz9-;DM3K1C=U_BS@fNMR@(`QHBxa=hR9JFBhNIp}Fs2|PT4innDo9>n zc+fs*qg&P+>@UbdgVbMMV22$kC>F`Grd=XRAxvKyd~56P$zmsdzAuT;Gx<ax;C&N1po87s1_=$V*_MSOfQlI?o>rlyi6EP}m=-BqzV|p1fhGuTDINCkjupui zv4dNN8VzZO^ur-O(;U^z2Cd>j`5!nxQUg>H;ldIp0;@r74P3{?j8m?*vw*T7gdn)e z!dQh8ff&%y(Ni>2T(PKf&;^3Mfwrr$<1_}ddyJ*VaMA8>2%jd)e<_>k03%Cl4{iH! z{QhU~Ha#Yt84u8p-=r$?b`?pw8fz<;!{m(t>J7rzvL23%fO&{qWNdC|nKz%mXgssEloDDM~eg>)Fmk6 zq2~tjXZ5w|IwM94E@kXb0)7{LN^SxlYU6JUv7+6}q^s;x)Pn3XFW@q$xG`9r5*u5a zoC!e`3#bxI2>g@XDjNQ4fO}kc&#$_gIKbn*g*n5bA9aOJ&$s4}ffH?|U0HV%P}{{X z1X*bedWR;XK;_B-i^9v8&T)b>gFL^OJAn8P68J;S4Bu@qiWpb&3P3ZF+(e1qH)ZfM zEZe^^dA?}yi4K4&Hk(-ofk(@CE1Zkt5<%~Mu(bef#BN&-K6S8i>_J##`{245oag-0 z4mmYIK~KgA)J!sR%#SwQE24Z-09U+l67+JGCbB8JJ|IE+O~fk}APAGA23t{|mECY3JDRDb7WA0J{kD)BQLO4SR4Z&= ztjd|Kp#l9pLYz|$=N_e%%S_6zhsvdgBf!DbiYp1%7{H~bAL*dcWyn!0XZ9C94N{jk z&>Rh&=#!>^W(*3}K#UFgh9X}v*Ub?8Z&UiW?;pWD{M+a5PXHt}wt_Itzp%q79ptXW zGE|s>iGK%|H2LVI@MYxrbNn?*{_f+$!nkeLa17u;0=#85Af#Fb24SEiXa+Sw1Hj3P z{;kRiAxvT3s=_gxrFmv(|0>s5d$dR*L5&fryaef3nY>@PGY5D&YPVHc7QrCYYX4Xz z5Ot&l*g_5k_yf*u^lf~<(Y}9eyvwanCScL{QVKkgLU;d?x&AZTV^`bw95OsCpWTvg zNLTXeP?qo0#h6{}{_5%ng<4ZZ%w@oOAF;o( zr160985N&)qpV@!LfU4`Hn~Ci)git^?_Z-()8I@={Np<+-fo~;m$4eUgXB72Q$9t! z+&D^z|J&>BvNp>HTb8U-EjlX%uLm(yq(UM9Lr4>(b%+}TIL$P-Rzw9UR+max0h>*$ zw~BF~`r0%7ubiIXJtr zHQkc%*GQtw0W8_=`L>D%w|7}$tCpihpX}D8AEPRN{+TZl9ypf%b`;2iR&TjcwTZ?J zu+OK^yv9Akiya?~pD-Y3sWjkt{Lupe0#}UM=?=JL{pzIDHG0uY#68Y#Z3kwu3zHQG zd;ZlWd)`g1g-E<+l{1DXKrf6|Qc^z8I>c+Cik!S+JSbbUPs}YGkZR&(iL~4q^%#lP z@Ko@xTQ&p)y2hqohx6fBjuy;Y<;~cl7HqFb5{OhXP@5h0#inZeH}P*h^9sM5eal)9 z+BKrLB$peBZzBOmLQ88{!gZOQIr5TJNflz171Mh02W5TW!o8gBYJIlU0ErL^w) zkQTCk1HZOAM;at&FXhdaZTG!cK**&FM>Tpe0?EO3;7^9Cgu4N5el9*VQ`;SUe&8}O zELjMS)r}0h^Kn0C>{7~F4;n-i59k^Aa`k8-w%*0I3&VT7KL^STwF{gdT##@!$DAYu zhb)ZPCV_3RRiNFVgAmqs%#ycH3#H^#0|j9%<^L)0(UjB674gT3n+XeAmizZN2&Rt_Sx37n0PJ+#S1kSo*6@|VSWw}g z0lUn^kR^Y99p0`Ep6(txcI(?kozvJKpr8*}|7LR9EV9q6*@GS~I_yGMjR%bmUqx?o zKi$}w^LJ??kOHvuDGlYxpa&AZR}MF5#rM4>eI)auvDpNEkXJU8^Li1K;)mq`;Z?x@ z-I%xiLegoG15o_!K#C{7vX&jA))R7UAc=BCaI%(@XN_dt(<)p#4X98^ z>3bK4Ro1Uh*kSbhJ#n!kwEZY}UWmM2bdi8F&wqKS(+12|df7lZP&J;l{lljoT6f1!Oh z1-`~8V)P|oT`xR%Zs9OqU_gdab~;bG0c9@df8?mE51Yfvq*mIjz zyLV&mn{3woeYp*?tGeIHks8Z}6IJCE)C9xn^f>BNBbatUA^6Xhdb8C$xK&b@<1^e< zy`z%Y9w6Mbn*3e(i%d&=!SG_7+VUUtICDx-|MD>qj(Gw2?HJX4k^7(PQN4EKhI1C* z7`MO__R}zres34sPMioL$Sk;6V-R#?kBxdDn@~cVu*p@GO@@ui2`05G8)QO3ZELE6 zs~Uj#O|~&NFf3{+n*VH?(7$DO;ZtM(jHqz_V?Z^Udd2M9sA)nS`2|xl);WVwP1@26 z@4_U^?9b+}2|T30Eog&hpJFjsfN^(B7uu1sfmuivRPF*$Xi!cPI`Lnm3T`ihr9GuN zZ+pob0o|ew4r6tD@8dQ;fv-ezTEJ>0Z|`7Dib0! z?SZCIc6ir-L%qY4#1D-i5+0)TH{scDq^F{?Fg!x+6edF%2<_0OnXhF9yyBQ9X%ZXK zFkHH8pRx*MQI~LI-ZAO#8{oFr zpt6r{ayaHw>QLmPOFKwrky9!aTYMf^6X=Bm;VgoMge-z@%d8COrMN`7VHIc>?|v8} z7d4S@W4+z#+DRIZ8k8oSlU1C*#K1vv=CX_yC1VeQkb@w*X1ZT!{fS>G}H=tcp#gr;0 z70)0O)G+m$PQ%$fd+gu_jL5>85Osk1N+`V_An7~8Ap zWHkFv=qE`kz0KS5lOxJ8r)_0Ro2u1=@47aG4ZBcY&)p>?#qz0=d7S4=-dqFrs;Z~= zB5yaBpQi^41VPNV)5pB(H2}YKt|Ge?JHC0HCzx6Z2=!0D24zhU(@Bxsf!}6gk6mEn z@2>t0#GNNCX{pRiqpjY>>YIPp&cM*X7D#gf+!50x>qGEcACOH$PZ0g!d(w4dUY;4w zA8zTHfwHOO`2mDCh|=y*E!HB~GVwbfR1;)@Cy#HYxJ*t49w9N}@_-Nl@BHO_JX2dZ zkcmQWk9EChz%QLg7W^P$@Qt}3_^dp|UKg-2#(`nJ>KJ_A4r|Y@`Y))XvfjS5X76h) zm`j08{6?>+_a;kRc^~a7pvEH|jqZjCGW~eiDI{$V4PK|y+@fV@gpzb}o5gaG!+umg zPFKlsRDgiA(9JiW0FenDxWYtBM>6!c4~ z@SkT8YQbx=5-usRQyn)I7a>%P;<&^&Akah2v{~VBGVduog)ujlkt%F#y^AKf^9lBYx>Nj$Qp3 z4Q{+}$UKk3JU8+_X9?4PDFnV5zr`FR#K077fK2kFt)q?=6F*N-R!0+ZXM_T{W67ba zih<9;q7$MYZ3e4q%PT$JgN=AKPUf<#N0(Fr&aJOgpM%I>YYRHw+>oU^a{5~*<0f&h z_l(KY{hI_Y*a0t2UwwG<)Y0yt`)d_qdyVL45W1MDy;?;o-tUk_o2Y&{`Va14ywXLN ze}}@4=$T^cBN4$;J=U35p1jo}y4ejn{bp+`^HNSid(wAUDbe=}L<1PGxy{%Wy2RFj zU9Ocr*_*J|@6M-T#$M6fXe8E9Qr5e=bak#*d|IL_SOFg3c_NKy8qi?+U?RjI4HG5f z_@nI%8@{sKRq)@%Fb@o%TN_0}j+}5zEfJBGvxWJ!;Z=h*JO+=^X)jR7(_8RfGQ?S3 zjJL#%4dQ7nw8=}a#4mEkwLKUeuqv@jOrDk7T8Tks58L0^JqUjKRBhS@p2Uf+Svqkk zZDE*X%z$`h31!{e+`PXp=wVl~{1?bxf(@pjo2N2JQ~!V3{b>*IkeJu-(55yX^UnnMnr< zuZSn~v2%R;q~~^puI*T??uwKgt16H&Pp1L8Q8W-5uJ6D|?CLf7b6^X48karFy^~s9 ziN@Iuo~Xrsb1Tf`$3894wd#u3DLBog%zyx{r5AvQ+56-XY5S!6qjqC7ooht9K>w+~ zp2f6HkZatlj(E~5clr9S6Kb_l_#K3{U@mI}Qi+R0Y@l1^lS$g<@WcgA7~k>G>ataW z(@X^4xxQj|?((N(Bt5S1ZU?`b>5n}8ig7S2O+&}~A-Vw;CUqbpE=Y3aRahKrL@w@! zxV!6G6~rbv$}S&JByeupRZWt*@+j@D1n*lDEB-)bR*N(tLp>TWa83{_1zgASIUU*B z2~r0Ko`NTMF1~fP*Km`{OX{V=r5swMV;fIGT11QuqM8aBQ zC@gb#2}n)5E;kYnNe#b)?P|*NiEQqiS!98)UXpN`719*d@(AFjn!7l5t(te@p?vD1 zSe!l(Qho!QY>Dh~`lVxlN^f#m2{xi4(#2AI9u`$KmdoWt-%xR{uBk+$|E!=7)b@M) zi+Y?)EP?5m*|MtBmk~yRAMkX&9%#ohfYeq4<655mi(Cp5$2^UdRS(m|7HAFileG(R z0!|>Cp08vD*PV35-`+E~v`fy!7#k1SzLFC;#;Whwg0UgG%Z4c|5)PQ_7ZAvIXoDZ9 zhVJ7T<=nm$b_|Va4FuyVIPXhMbbgf=xx*%AQjulD_e!+j=^|WhHsCY7AqwEMMefKW z#C~GgTP^c}M^RTrx%rY5kkc=){de;x$&VBqtx-U}C5eRjuLB+7Z7#Jn#b2KfKnArayWyC_4({=qqIe2l1B$%Qm>Q&dJI%r9z>i}oBPW;*klBW7+d%8G%caPs% zt{yp%x%!@N;ys2eJlHS6Nb;&xo`Q0__DaKU#8toW%mPoM_^m~D9f*!+7zh-1?wO|M zOnaWUW0QA+J39i9BB%rVi+6^=<%+R%q$$dP-xSdBm7N|?T*?mpzdW~fNMITT;~yd5 zpzN@<0D#Umjag<9mHB~?Y@OorjVQtfJ5WWvd>^)KA!90#h9k0Q7N-fYgcVSu)WG9b zby@37K`>bM3R>Wvq-2nikUR5A`bYz{E=r2gHdej_{Tf8EexOU1u>=i+!6BN`=B{vX zz18L&KKLeyP;h~zj|GcOW|>Iv8(G_B<<8GpNr0W8GfEJdtM;;2$y}NEJ5%^52mFZ~ zmeYJ~PAkKXpFAWDEB|9;RP-FR#O6F1UQtk{7Mo=fMf!qqWO$R0!|#kpU=$1tP+`Q3 zQ1#}WBWhl>!DA@j1FilB(eY0m`SQvQJxr*mbDT6gcn>91xABOOx$E8-|LlSD4;qb}gTkGlT3PI8w$+DzsY3i^iV~cmT z2)s=gzg6$(g(6L+fKsy>7zN%P5$egB?KDLok)%4DLJJv!t8i9qUcBLE?7 zBI-L)qy?2=g>GZ2#3iL0HOtnN$_^m!aYrhNnaqd?LIM($fBv0e|_+F zXeb5kBy&tL8{*D!$X$ogb{DY-`dJg#ySbKwMf#MQtnd&IHJZjje84wAB>=xl8=$jN z23tZe$C-Eaxie&IQ`(0s?#!l2gnqv5!Q2*d-pY}q?3NNA8!~|JuMjgaRI|!>kBy=i z&ast0m??ca6PQT&WhsGl1T9ck4qI{}mkvrR0>4Ofa4))rXGzCV+Z%S&Bv?^+pEPoT zk=>Czgax_y$U4I_Rc#=eApooVM2l8=p8V-#jV3Q(gP=qhZyo$!Nqd{(ddOjNLbMcf)l?w&SmAPfC+cncBtZGX{1-&*) zh^2SpM+vN2U`LCg6+b@(#Q14Xd6;I1b;`r$@@AU--JJ+^u$Ms8ssPO7a}JnNKC!rn z@7u{WP$9{d7EnT=?9crIVtTIi5{#YJN!e$(y*K$g9OqQ+=i}osR~DjJdEZqLEf+2C zMR*pnxn}ra_q0yvbNqcnkwnS{;lh3Sz>0hr(K1=%4GR+CC4ar)fpRp^@?{@#T z2Ngn6aj$t^-(vm$C2DL{QloIA00Ef_0Rf5qH(TWYC9uKE!P@Qs#e(_YGMf~NzEV2S zI+dmr)xOer(8)vil(W8{3b)$iq+Xg^#-q-_Co%lbAPKhUXkwdVF~(x+E$AO}Z}GnC;r^B^d1=MVh~gF7n=~lKND1QzoetaXFr4kwSl|FZ_Jy zutajP{{-8hOlsmVe@EO6nqpEbr)t%TCLUT-{kWLg5zGQNvVa2)ssTdJP(g5L!LX8+`mNzC|HD@%uL}qt+qxyCNirO)lVmDCA z3M`qDKT!G`(Al}zkbHF5Aoe)>W~@!in2dS|mh)Rw)Wst3kt8J;!_SyXo5@^$~Pn;YTX`@|qJVLuNv7kkvv)CR|F zK2O4dhmw?lO40X^a=OTrh2^@w$X_tUpe zZMXIJb^3M>U0Fi(oV1`+0O6~d35EpKeiqCwwdEFdw0^lFaphZ88!q_ z9?TVr-Zqrp(g$9rViD_(DcC#z9c@Wdn?4B&m@il)bfnH)*2*y5kV$Z;A&eUOoG8+L z01t&CN45n7E+AS9)$osQ8h1>FTJsW=w~P{SYquhWF#&UlE%|kHqS86_N*?8vg1#8kysu+_eq!dnZSVPWq%}y#X5X8pQ3J`Kz_R zu2CGBhYb@@lA^sKLlU8^zZ=<`L{I`m0h~2T?F~TDwj?uO5MWwvZ|QBF4~LPS!0Iql zl!#zHHlL?(K_l(W+ooLMS6W<;qC_*cBc1QeWKfYoD&nKe>z=+M-0_jl>m#fKmlm_X^Dc>ZpdBr-9otjqb51{3ypKDXt`$tVmKT9ttC6= z|Eg=u&Kg9xIl$9{035IE#<>gVA3K`%eivu-a=Q<{|JVC3cQbb47BbEPjamS46y+fO zDHK{H9;i4q2YRnxC0zlXDLL*m9%5@{k|Yc0PJ}B#5)H>CDd~jxx_T6QfMqHNf#!(_ z8KJLTtK+!t+SLLcIVY6p3agS->Tc&xX>EmmnrN~=@e-mg*6Nj@Iz|xSW)Tw+0=@|u zGMUDwMlg8Xz-HXU=ElrzrQlZ2*o&+@t1ftDKkA*xrqLS-vF|5|rJ5W;zMl?i=?d{L zl?v3vNgj?Km4KWkys;IIgzj& zZeA@(hp+OUJ<-;M_en?20m>OJ!J=v;?Hxr7VFe=%gk=@7cg8o~^KSs65O=+!yg`Sl9{RVbK?*;{sJ(mJ}X{!z>>tWkv- zrs)9_R{| zTpXCgft2G7ZZJuc)xQ(LA;uH#0w_%3>43bqtZQwW?Y-51-jmIpZtV(yu-=SuT*v@4 z4}f%!gV)U0*t74292-=Q&-hv!C&c zyb*h%U5b-40z>_wlzy^P#G58F4T{IAA0y7XGvY_c41Wd;6%s_0+strVxPpQCq69v= z{I^r)A^iSBt z$9K=HK0xdFKq+%o@G-+rR5q$5rs$?z(5^O0MjWz5$2UC8>~7SoHk2LPoW?dnb!Ot&`8hjgm`PWU z?Q-q7e7ygi?iG-fUl0gFS>hj#m>CEXLZzX99ehmw2dJ;3VqjgAW zn=ECZu%;@rWumDDrcuClO@xv>i3|Noh}B4`3oUqpbgGb zfKr&K)jXOPBm6m*y8@TL0V8ev%ThNxa8oHh(;$cnJ^`#Yyl9m^vXSzH@S;pg*K{c! zK+}9uGs&1ELFrH<;iLVisTxOyZ`|@ig@O(o^~oNJtuk`O?Z7a?h;x&!n!eKZpAfM; z*lb?BN}=#_D7G$qg^k3DU~hn2f_AI*l<#Q@W(&3Yu8jfz3~;A$PpzP`D2v8%*A+=R zJAxuX^-2mn6>y>QElspgR;KtAcsP zm2xZGsidr=mu%i%v*=`hJ_n*|+a^3G0_NJaRSKlV>W)cX=ExJT^882!g*&^jby%xo zjU@+4do<-{M(_0{Ul=^)a9NXt*q*45B7#-8k*D(kTY~z z*N6V+_Rcq>k+V8^Xfg7dlpJpoRIOU_(zdiSijhISQIV>8mi_lw1aEJG|CQahN0RWT zRp^;a08;T#RbEglzLvA$x_53Sa?_vdU}L6y8Q7C(B(v^ntP>;3G{!HTjzUa zW36mQhrbRyz2aHO{~*j4QZGS(0#1Z@XV@EQGS2#ju=9hYNQyMkpl~eby3IAVikX5D z!{#xx+AWZw(~?miOLHxB1|mIA$(}eHR2KuK0&x|$7I3CKOYIRrGi@!NjP+J%&;uz+ zFb9hV$JGaExIY%(RjnDi9H?*It|#?vdWw=#q|)C$f71c|TD<<7e3-dt1@MrV`2Bsm z9Xp^?=itS1XZvL!#gbp2)M}#HN>l>u6>E)vE}8nKPryKBu3tu2Y~)ah)pVy1n$q4? z{zMKz6iUAI@yl4JDn^X)CVPs25a+`#_INy`1zv4t*H0sik>6#Sa}kcTr(Rifg;$t>wQ{XDxoVF zzU7mJr0(V5o&GOOJ_Dq83VV|K~aK=IVt~5GX)60Bo7Va@nk11A16k zY4)!qd{R!e$N9o)N*WJJ5T^o=sK%-<<7m3e0CW)>Qm1JUg%<+9Kdt!#@$VFF=}>ohi+onX^hK67nfli%uBI#DS;4Z%%0`aXefFJU0j zz}Yj|z)w}9BmxlM+L1~6scQD3ZaMx-#_H|j)V~2$Ucu%VNWCB-9{%^o6V>HQ+W08! z0Yq?C1~L;sP$5wSC140^Fsex#*+OIy*Ua$P!3X}g9+#K?BTNBmNM+4qz&i@bz=zu7 zr*EogXp&BH#RN(FPi<2kAh`+z+mn5*N$2if?QIXJVvwuyZf`G=f@5IVazPRfd|n2F zD?uAkK*%n^vjM@OtlSn3siJ?c{8vXb^J|>c%jMRm2RYD#gF@$8mP*rbM3X68;u|jU zVp;t8wI%U^(iNCN6>SCwIS4x%W4zaPGUIU|<|h0R6r%&e$`xhJ=+`Wno8nL$CTxPH zwd@3do1cs)=z-+E^k4dRAP=+Uzk!FVhg zYkd-^yIeUV8xH5_a;AX;6GfQ?^heBE4>+_RDqJ!NqLcVE>Hb&JPR`)R$(m7kfat>E z8<+yrhc;#$=1W@Tc zB76R;(ipZ|bi3(KV+VD?GoS+vHC?51X zcE$E(!eDKBt(*hSL!@#xse?u4Mrdz85S~~4r7x5b)gv93iudKJ^^(Y5@ucBw_D{dh z=sREyB6evYQ+E5&yA8CHmgLFp3IJMnC+^N|u&R~i0LeAs)#*dR0cQRsLIGO{_V*xS zZuzb5KNZi9nzbMDj`3Nqh2v-h&W4`V?}q*`Op}Y74C@9NuBt`E$be-!7DcK|>s7xh zhH6|e)4^}m*l4y~WM&|s59Db~ZN2Z}4q5KjC4a}XMoit)!bm46^s~+fP=^PeV!{oVS!FKZoA^Jt4#R$#iGBn3a* z?Jn*vvvR*CAp{jlf^dJt`fs%$B4ad=mzb4uV*ozra=1yHb=ZnejdWM~A}EBPon4x@ zB2&r#S>oRd=0gs%!Y_wy!2Y@7Al6wLn2+$c6CB@P$9Q*yj$`}@{>Wv%h&s*zt!P!Z zkG$5buXmSuDPs@YofX-N5tZ;&FkNqlA|!r}_czym975de>wmxdei_5Fzu#OK3m1?t zptI3Gd%(c~s^P}gcvz=YgEKi-B6IXH5a5c6OoL*#8eXR3wS{-_0UM4xn7r_32m#V^ znT1ohi zSHO%yN38TR7Pp$5#6)LR-Ly5fEphJuXoRua=n&P2(0P*wAK(a-1LcY6OY%(+*vR4y z?Sqqlf;20OD;A`@0Jh>HMCB)*tjhVZ7VZLbbhh<5T}r!jgJH5sUy~)_;%zm$`xRXXl)3+|xS58m+tyyfAU3z&IjkoW*m>isM;AEEHgtr;a7+i^4QNhq z*;!Nc8prlXvyN}VXQeS)ACYWg5`+;PQ(B#d%~AGlr_H1THU#mqA=k4K| zuV=fOx@Z@w0NU2+sZY3IN{t$om1Z=7(~~Nko4rv`%gQpXr;20-*+A0a$I@-iN!x|- zQk8bB-n$AxNSGb0SS{w=m(qYo=Xy?;c6JXAzXMTI75G(#iiV<89eSaJGQbF+MYpF% z67FfPQr4&53U7SJ;;py6a91y!`5L~%{&af?^b!7p1K8p`)9WHNT{C8AOP2GjmYogk z?aGIj=aGm414ZpA74d+JIau|kn(6`2R1G*S+h1hMt%2+s?#FNqO zbUJ<~DHhH#;`2R+t@#Pum^&N!j@P2=FZ^U~h|yN%MAt|VmiW!HIHyHhUr}qubSbXw z^UPGa0Se0FlcMQ{##`%-3iY;I@9Y?d#P#)hJ?3ed&4Z40vtAt!uQ^{jhvM-vQeA@} zukl*;(L?NlymhsT_ElLI58SKyv_ExuiNKIPgO1H85$~O)Gj9%SlCE(4TT0Vd&8~XR zj}3~(SewUPFFeoW%5*bocJ8c>sf8j&LZ6nZnCoD1J9z z^*kzA>T~|V`i0bKq|Z~5&Q*&W;RA*v28U8c7rt`IgI#6@hURh!`a|fhJ|sFQoxPdm z;Z(|Gsv?@M$$WM26~8$UbvazUlr!i*>xL3Ur_q905p1|JOW>vA z0r&td_38|@FbnZ~luHQF4ol6!JwI6o#RxQlA0@v}&zst{nvGOaLbb}YF0Z76o&h-n z$>P4`RenX?gkyAU@4`&+oaZaBiFA7N@R1`j1YY@DD`gk82Rhu0%Po%z_p;^~xW;J} znax61ege7S0mW#dacC6gck<99Y=}1vAT%BF-}c=4xQ8~J($;258MrXweL)+i8dPtY z%g!_AD&4k>QZOf;o$}_|%x<<(hOhj?)yUyM_RVMHR&2Wzy2qIthB9++fVq^)Kw2wF zQJE~Xe<#0DVL84GDZ=N@RhYyCi>MJRr}#8pZeV!q_$u;-qwFo*?r(9Q0&kN5fDmT- zLkY3gEO12dgO0}~52v z1IvrMclUWN*$#b*2rYgQK|5S0Er*m97;VNHfo%2{La3fAlbFBMCYwypkFFH*M^_5Y zE0fjPDNK#H@D<=yhJXk_eCopK^0*yN@n+QL^X~dKdNF#y-AU2|`5Ye|d|l1BZ#Lh1 zxH}tmKi>X4^?FsE#ql#!&XhEszq+!U-&YoId>P&z{Hp&9LinY(@R5H%@;o4kg=5Y{ zYF6O_!O*GYWqYjq+BP`&YI%M6zIC0jRkonGJXK@7&9)zZoOB4-($D&Abk6fdH!XM8 z!(hb^OSrCL;DOv?bp-KKZRaX#{!|*1=sLq8P!0JIK?pXpg2>u zWx{XeRGyv4E`A5_%sAgugxNY|P4)*RCsa87r$Ny9=?}_Jb#V~bHtwwmbaU{TTNUja z@Mb`1B~)5P!2w!3F_t!h?^zp`|7i~n?b_V4yrJ9>LvB`U)S@-1L(1kF8Agq4>f(uS z_$uSP40u%vn-DrW`LAQ>`~C8 z>WAtdYJRBwq3(zJ9~ypW{GsWG<{w&qX#Jt>hxQ*je(3z6>xb?idVc8rq3?(O9|nFH z{9)*a;U7ktDac2G{r~H#BNi1Hbcuk0zLG%y`|Zg8dZ7Px)o>{eMWd+yv+D{8FB)wQ zFB)?v`mePNs8kc{RnUP(_Zj}*&sE_6ymyD8u@Hx$@gF9BnEYYthv^??ewh7X?uYpw z7JgX#Vd;nEA6EW9zP>Ryu%O#E*2K0mvCWBX+qOM9v27<4PB5`;+cqb*lQ;K!_3GU} z_x@PFPSviiK3%(a_u9SIKdk=4+CQxS!^S^s{=-(YVL+R@Yk-sFAp5#Gw9p@ zc%1yd3ns0reIG7w>EGJH`tQRl#@h)i#@qdez2;)P{f)f;gk8nu#UTv<2gwnF1L65! zfA)Wc{r?lT(!L+I(thv{hs`VPM|=VQi8dT=H)nza0>Z5EKR=t!>)!7Gr~h9>7dQA; zoj1f2ukJ7!O($+w1VsPVq+zlPB_&T{OPfmGdlD2CwB!E1WI#jcq4J+33?X|*d1J|| z>8{P&V_fAchZE%NN##GXHnEfW9>7y~J5?#``<}D{ql%sp1Mc+Cy-}c#$d3gW!72+Z z#beFb3?1d$I+u24MD#MYFZk7P-m?Xv{*jVn!X_PJ0hT2CXsr&pCQX%4rJnZ}>-QtXjy6q2`M zm}gGJXp)Ec6N%>bFtk!BCn%&m;|afu_iIf(P$jJvfZLp83l*Rw_{2gpL&zKOj zm>gTDinD0=>O#C<@#M~&-wKXu%7V}d6h?wW9%~kh$8VHP%%~zkcLxA3i8VCwBc2ct zFwjsLipM zAHDWgQqJ`psu>=r2;clz1Dy7svWCCZ=s|XSGOg@4W=d&AjDQ`Vl;aU)v9W&bCEh4G z#EH2HFMIQE=N4|%wmDl3?DW@&$fkmLF#Q|i6KYe|YpF*B#ng4pi?;F^wf<@II?W$Y z5NIb!prj(m8JeYKW8+>GCq-Im8Ex~D^jv@YbvWn1IUy;K6Gu$*vz-V^g}ltq_`sHu zQLWy!^|M~YashUF=Q5T9-!&I8l5DzaOr^V@iJOtFQ>F0KHVwG3 z8~2#$!wJc5Vu;`*MA3$pzH{63>5oEz5py5Hovr)~T z^7MYM^SEdR?q0jc-rxHlRRgQ*W4JY+WWNv7Q0WGCG0&#{XfVFrs-Zav zqA^l$lmJS_O`v(g+IP@LObUtN{E0X)LlCRsFU<2tO(yjC%vkS;N2BCB4F`tg5kTD3 zh$-HIc2XfhW3GBNY$%c7Xkq*8?+?W(vjj@kd!D~=;LVzk{XA0|;drSX5c0;L;x2e% z&dH`Nbh7ry95v^oqP=ih(96u^cSb1wW~{i$Fo5<6DrGcfNPX{oJm{L6F0Z5`RZWxe zf~mE#XhRo?BD*GH4%Si>Xmo%~Qph7dP43(TswUrw{Bv?#riDqG1f_y{qHM0S(5 zF&>+A3l+Mo9d(TcSHEb`do5_yKog!sDBt1QY)}r<&{)LX9p?mvq?R!;)DeuZ9i+XN z8vwM1xUFWN6cJm)MKPA&iKAqYq6edX)Wc?XA0!pjHjuzQwKs?ZOemeUM_>lr5@Jd> zfy7E?XC&c)ih^sDx~a>zp4NUBloHEViLIAr9OQ+R4jqhhen@kcnlwb7x56C?w`B8_ zE+WIibuh$nA2!mcT;<7KbN5QtEiG;p4X8$|_$Z~3=4CR67X&4r5)raIzjW|SLRz2a zQ$UTzkhOHs$2TxS<-e|ChVbSQSN|ba2#k_BR(TT zJsTr7lGWvGQE0qN2=>2$@lk#fX~2*wuj>Lf8kLZV8dWTC8NXo*|9)eWX2Eh80eH)V z#L3h$b;vzp!HiXRL_%a!I;1f(M=7M#+R41!#(BU-WJ5QsxD4T$NS?o#36nnE6<)tT?I0fOi%jkRY*{xdbxQ!>3n`FUE$-BE1?I*7+3wujf0rEk;S zT#aE*0pO=NNW4LGyGS*nh4307AE!J<)kE5HChi{FmAKoRo8V3mGZuyp0sl;+je5A} zP~HPkGeW>1un21`i65m*0iL25i!#$iADQ;)Ir5V91@vBYNQwSwQ>OyRRiXx6YOTyz z5F*>mVE>v3^=n$lEA=8Jxl%|tqp@0$42$&mQxHO$9)RS6B_a28g}cC{u5?`&P@t@lx*?nKSS6 zqI#6BPg{UY2V!jsi1Qt;mK5Db!Muzk!8U>-W76369$-|B+g^6yqobKxP9Zm$*81{{ zLG`4NZz_a9Xdfrz8-r<9oD}@i`NE+?$$|Tz^8G>MDvB@CwngmsK1E=vu{P=4Tuv!U z>po_0;)P0&UkE1;o+%mzc{s1Qj|(cB>#`M-5qv7z+R8Bpc$POA-nTFzQ!@1;*Dh|K z{x0T}IuK<&Unz8M8+!yP=SN$$SJCSunEtB>@*zL|hz%8rO4zyZCzOJIn1?JT2)YgX z<2Td$&LMt}Q4Ar{HDZp@E=bx2`zMO=#(WZzClaQSPbqbVi>}4y~a;rh&t6$E?n6T zgL*K;c2>S@l#o@Udo6LeJ(3*L58*!J4qb6@7ANh(+&+o7 z5dzJ6E+P`6I3!aRnagpL?^GEIH2#@S@gKvU9};Ag?MsSiQ| zwmjKO;F7c>$4j&jmvX`D@gCtn``iU<**jta`u-ftXs+*q{$R^{o=#cFJozN&%i?Eo zU;aL_v!nA&7qR*F=3zQa-BlkzN zw=HUWCVa(>vsXolRIpcI?n;+m6g6EQ^XIsC(JYHUl}`-|=-}t~f|+j9cXrGrpoeIp zQwyjzVa1Q-UUGsohnX2z?#IH7)_xU)-}VQgT1feLLgi1b;_Cahr3-Do0x`{cFKi{{ ztRUp-pUgUlgRy4*k6#-cJ#B4y{=7Kl6?9Lhu);E{OmLm{*;`4p$XF1~WU55QAe^b9 z*dx^ISuV?$yHWHKoGEn+*e^@r0u#wi$baeP~_M~U6MfDxQXuwJBQoLbdXzM&F~i!|P<+_VYa zW;m)+Mgyk;n@&Cz)t25G)OFfpAa&;@x2u)!M;p)P<^w4+-w$A(Kq{IAs8VFcUw+Rp z$|`W$cqJH_Ntr%FsYWGpWSB@inO$|!Q}i+9lYT4F^wi&`Vo+duWEr5uFK(h z1Ai`tu4)qv!yv=<{-N|hQ%!j-tW*6@7`9j)@)!7Qp?FiFuVcn6J^yA&qXG_pnr(BQ!;@I1d2@%QqWe z#ElrJnCQ03zepO`(6`|!{v?!5b`anw=QYr(*ny=tj-UA*_a-XpBDO>Yd+@p+&R%Sm zHTGC(jYd$}gn{dlP_jC&5pvccoHK!{)Nfw=n{rN{?Fp*`6oB{|0I%2??Ah2)wW)RRcQxy_qUggS+i;7?7^OcM-JS6 zaj*BuFqmbr3*uMP?!^=t?L=dm7;I{$^;(0``v0Z} zC%S9(vtQj8>A%ZalyIC94NBMh@iKEg2lls3;x8GYNLaUqS0h87vl_}F!rH^}u^Kdd zS+i|V5hvcg!{nddYs3ww8G{41-#&ES!b*1MBxj0xA&A$1A$cQ=@et>ViNAxUy^dS^ zfMyo80%9NjLO<9&$`+TZq&p+duAE~LmO4d8JL<~+kd#D6_QCxtGP~yKc6l?3l5_GD73$IDKt{LA+vi&ToJU?oT3m_ldkm}Q7}Hw#(cDz3BJsm z74^7?ZnMbkN&l2Uuu@y;Q4@8DlGYPM+P5h_1-?6vKX}@?4Wypkt`$^l(XSsjF4}Bz zM-+<9KhiLa@L|DK6iLRdL$OzZ?#d{Nwzg*-WBluT*aka?-dk9zc)1ULlG^Exq_sO; z^@luj+hbbMF&o#|QT={Uaa*20)6MVL;I>vd^T*U^5!*n6DDn(X;WH}+>AwiFHwdCz zU|u^xyJ~h$8gGuWF(I;b4a`^$Rfk%N88wsq=ezJ4*Sj z*7oMnqnWRcsV2+z6S6A-#0ufc1F~xxy4)dE3W44!-bpSuy9GKcT^4?qM}-d2DvHy{ z&$PoEJJ_L_h{>9uA4wDN32CA7O`&H=K-$+g`TRctJ;E(RUD=NCvRR@0rjbI1Gf8JD zD^J5v$9)?CB1;_*YJ)vIyU(wNSEFHssU`>Drwp`m5RKytb}Sjxz!1=&_3sx zM;${^gH!zoqB5f^);hKH_w^}mwS7IH7QfKxo52@@xH>XtgaojM&01nH4psv7fc7;U zC*~6R1={gPHVlWgSfX;#tWhrY)nmlA2v{hR^2tM%mNp`bJVNa*ty{b@Wp5DE?$sSg zulq=cs1l1FR6uiaZoI#y(a(x_RFHpmDH{d9g)3j*UvoP^@21$lvDc0f7SM$Hh9J*c ze?*gHK5}AO~#Z}f2#(4*56WKMFV=o*`>R(SR&&Z z<#JWQ@>8~P)UK$w7fr2XC>?3>z`1Hd~rkHzr ziLES$@+A9mb7LX84yY=WpuGR`!KI^?je@r&)9>AH$OwF9Uot=9owSXY%R!uqa6thf z!ezr}IAX`E7`ye3#Elvyf%pkcTF8ah8fLleGiOw)cB{we^O&p- zAezkJs4&UheE1nGFkv=tX18#3r&CBZusXzTDVNdU&Q*UjL&ns(mBHjPm>Klr2_N!> zmFxaC&>u2PdF7T+z}l|T$2gYF`}nO6yd9R*LWWlBs#D;pagj9+{a>XtpX<&cm-Dr~ zZyu+{JBld#N-=zHFh-=5BR+dJN5*s#lr-hpxVHzw@2}%P)u~Zl^@rg=rurt<*%{w5 zkX%q4zSG$5==||>QM)At6T_ntQM7C-Vy6=(&Ws)E+Bp~!;08obarcnYlXNK4^aT{# znywD=+T5Xor#**C0w>m|6x}vVCnFBZF+>r}k zij|&d!AGJkB*nJd%I+`O4uPK9d3vFPqT&PKdnnlqXNe3eWqkaDh-FY(35pd`&HopS zGaJ-9+TmK{DIQ37zsxepol6^Wb<3OW)o89Sq3=z|`_u8XeupPVqCu#^7?#0hT=H6y zPp^Yw^|M)3H$hHbP6xM-4{0nT1K{cLIN7Z6aOI(PUz!0bhihq;I0_;^~ng- zewlt*VSRclIVnu<a_)uP~^g^(y1!WM)+$ zvIaBlrg2?{&UEVqWk6{Y$QbR3)he%|^yM#22`u#SkE^Q94G1J->Ok&tgd#9efyhj* zoec{{Ax&&W7c7R6b_EGZA%=xz8v_S0loSIAZIO}I2tgyA`3<95;c>HS7#Jk+ofDB& z!Ah`6Eh`~pG#m+T@|`j5p{b6K%>$>D!Qyy(22hSnNwcc*Attv3aZKVt*hm5}s{@yQ zgY`p`mO)H>)VQX>-qWyA1c~QA(7N7mVSi7Dur&j}paIQ z*388s;a|qBuTH2)JF~R@i+0t$tkI?{+^`YABH`+(((&0J=sh3r9>2pG@ z0)ocGV%pJ%$y(nLiW+iGFi`AolEyA~+gB6kXyn9W?rOXOcGePL<{lOn=Et2Q4uG*u zf+T8V)8y7iMD6xYoF5C-ajSo+4XeKHz^Yo?(k~BIrQ7s0eEpBdcE6&!^6=mcP(K3; zN?xFP#RxMh_`*zy9nES~CZpI^VWQ536;0>v$3=kB}7Wl@XNr(V2LdOmOX5$9u?Pro=H`kIqpD)1?gNl#7xs3;PB16x{p@ZiaSN zy*7>&rK0hKJ?XE3ohX#XT{2-b8%r{kSG8Fa)3=gLHzC~#1ABzjBplZzXTb=Dm<0F_ zDXPE)W%*ERxYa12MoSvKR&_+6y2838F+VBx>N^pxyp>W@1%F+t4%JGoE~*yk?XcE7 zufD7`WK$g5bo+p68(6e^`}b{sL^1d@vyK^l#pYjeE{)T9b75~(i?Be|FhW$4tngK= zNXnX82k7(|M9~)g>xrnw{8VCn-v$Qp)APhh8B-vCc0j=SmllvKRXHOpv21r?O^g4k z4a%A+g_So;@1#>3vfYR3#sa3VG|o zv6rCcmRfZpF~(V>x)ZtcM*}Ld@HGh8z`N$WwLImjMm%?RxoH*I^WefZ`!TmJOQq4x zxFF|x*D*2*q(L%!27DMdY{$X;i|BgLlPi&KH6ZtzWj!|zrz&z(zGDjNSn=-*UWDh2 zWh|W6UDL0Vet03Y^f2vQM1=R0tT5Bh*$jhJkd}dEYm0e;=5|6Rg#mwGM5Dz>1Xl|K%N)F&swGY z#Rpkss1X*&Y!b^iut2>Iy(=*Ftn0EtwPjqAoJPX1WcjuoII86is_m*!&KDnbwSKU7 za57op_ZzFrAt@9K`UBu}>%nC9oc+OF0p8^JRs$^k;3*4YaKW=k)+G|D7-4h$=9Sg3 ziS*|U3r&B(ZDy$c{d*p|p-n1Zj9KbSUeEGP_xruUCc_*1MPGi^LvZ^JL-ZjzwoVC# zCE>kAUW`|P=rdffZzjg{SV1{IVUU(SRPN}#pl;;BYT=rS-ya}P05;E_y>PH?^Z46y z-bV*8=Kcl^eSRVE{!sAGFKZZ@xBY|P{XR2?;G&Lm*oW8RgK(YKxH@iXWj*PmUf}KU zWD=J5?CT0=9Uzzhz1t25dqJ~c@pD&b4|vM`Y8cGpWQNWjgWRr#9plwQ?h$lUma1Vy zc2CwywLAmMOa~Nn7kc2zh;^omLt_nP%nbet91S%&5I$y)U%4--qBgzRT4rIPz5C-8^#db>2f_&DOb z_00Bqp@*0`VFV<}GiT#ZgJ)XuD8(-unGi}TnsRZ9%K#A&vzVD|A?d0SkLfUK@ZzTb z>^$;O73pIB@uNS(U8TwR>sn?LsH!1sk$);uBr*qB%ffTl6=cl)I;lj?b!)gYwXAV;73$e-Do5Aw(x^M8OZHjfM>%>M6uFXP`=T9r-{@b0eqV za|#{cD(n_+WSY_r|lxlX(2Md43^@|!}XB~;L4TZ0_O*Vcb zicmu=T?d3~(MOV477hf{EdI&rn-@0?Id08iyGNMN-8{;!=pXZS*6WsNWp(HC zA)(NHmNnV_4tt6d)-ImWPlOyX#EMV&I9!3~lg?oscv?3(nI=B1BCcv@Zx5Y3F+|+t z7+%gdEl4#>86k)+>N5CY?D{|!3xM|j~JJ?c9K$*n%Ybh)?~piv9-gLIF+ zA_T&lvUWu1uGb;`ms=Bb2~ICutvr^YGH=cb1ill@kpK85hPk@k{=OLW*&j#!wd=1? zcy{K)93%OSV2YT{GcY9ucYP8H<|KK)phpb_U3s6<%yc^llJBP~La1WtL4BYp2;#XL z`eUW6xo4so?tF?CL1US)fOr>~;7dcRs1m6?6Jg=^C1$R7>msN-G8)!aD^FKP$KL?6 zlzY5b67Ba#9v^xcXs+A6fd{YqLvy2&q-K0&s@N9o94ZX64!%cP_VfM65o zR*70-gR*jcR;12Vmet#F54-6m(ge!iioSAj=cAKT;BQ~ALMz10KoUr(FkJc-1KBYK zAq28Q&pOtZ5GMq_mejK*-`<3I&tGbc@etyI zCBLqiGRALEmsdq{k=9jetqWX9V=%D1K(MVQS^$Q=o z5M>lY!NFk(!kuZP=8m?(u4Y3>QZ9R4dQ`iR^$_0A_ifIRl=J{uNeqPyC5)plRBmN zKf4*$rMy@GMzf+&xo5u9-kVZtYW1)}@Jfg?1z0$3bC?D_yJT)uJ;-h~* zD5w7OsDu}_U!!d+x@tS)*M5y$rzXx%aLQpYJXngy{Y!k4OI<1rv6QpQS~$QZa{X-s zU%noA2WfLcjwQ52ij$aEwB7852+1VXl7gL`V*t=gZoT2r)|^8&5mmLq!CG!rBKlHk zq=n;QCe!BiICNDGo;17yW58+=_(=7Z>pyr^a_IEOBAZ5UXNSk?l2$c_9C+xxKm(Qo z|Ey2mwr*0PuqC`u?ThIq{X0jonV61w{bmOQ1(-NFLRG?@P(J$5(5{2>ew>41^~j_} zwCK$#;ML$7HILj~LCqK91bMY{{b@3WcQ=6Hf=M`{|0qJRBF3}i6rS>1l{CuHmJ*0b zms7sAIG6!Nu2iKNvk8ZTyQaf*e(bJ1iyJqdS#J%xJ;S=ZgnC zbmh1q>vN}@!|~P-3-cMAv4QG_zr)XWeVEikrcD< z`YIzpn}UXACI}KGaTQEZy<4Jk)ftLR{B+f`ka}{`n5;2Jd3GGJ+V-9^>X*`YCwj_t zA1}u?i;C|;Ng#C!{#gg{FSV>3mcIebG`TQ;Q1vpXrfM8#I=& zGS9eQ-NRJNIIH|PKlN|?LR2en^9yV3! z-p2FGxbE3|i}5k!xTQU}YhOl8(l{04rg}NiH^FZ}TzQHny1Ded zIBEIvwg5o(c%FT}((W3c&T#-<0~IX2Ir^A&d=&v(FDmKWJfHTe2DsQwo}E0^g1Eu) zn86!a2^}itV~@&_+)72$ho+J5`5K32xSFjLv?Yp&wFiK+#xQvLE# zSw|mSIq7Wj`RegonQL|Hf#pTPHpAaZZE)o2r^4g5l5g!cI4G-*sNaEaZ#$usjB#O@ za|@Gp^&!JroZwKswPheHPU$E;mlmP)g(2vUEdk#jx|~Jj((X02RGB*M1vO%`TTnzD zlXqqGAZPK}%l=(lO7$g)h!SjefRv^#L|&e_(7Hdo(;cH?!b)dxA(Xq!sq);F`#{@C ziK8zl#nhJU(hYGBAld?({`>L+#DA0Y$%R0ww0=sTH$)i0FxoX7Tl1Ap>q7oBmsko%n;Vx=|yn~LDn9_U6ER2QbCok{bbj~gU( zmgW073buzyj6BJwmlltvN}br&6|uJLJ`jrDZsMh*)p7$g(`ORv8>0-A2t9Po40-)+ zN*G!kKjZP)IjlO?Ahhlm%}}gT<=Rb?lc3Wim^8AnE>CWdDC}gu%2Xw&Ir6>WP}Nd) zXAyJF`*80v4|hRKhr--_Qrr&6$l7S-&{-8au0CP=%QBWn4~I2G6&;v7jm)5%zVu;j zcE)N-F6Rt%_qsEe@eh7A!cO%;eED%}uV!Ox?X-20ANKy9jxTm}qL&9ON6Th!@sM*d zX1VmWA3R>|8Tp1dpLNBGu!bAEL1%BWUcQrCgz$DvrB-`#@JOv1uhLdW@~H@OTX2@# zeE-E#=60U7#FymOm{4PL-YwKX=Z@IwAlR4scuxxo@C#=YH* zS#4}%IWtI-P>JJ_wC2#fi3TJG0XL`p?_K;^V*4Uq8*fyP(q8ng^QS0~v;;>^_Nir< zlsopmt+cM+1tBn!ui~K!!yIbmx?+dAX5zvlGw z@#$|4;g;Qv`UkJKIDE^!KbYLSlKON^m6H^Cwc#fQ{VinrIc$05>ff_Dv4{fUy0C2$ z9gv-_92jY^-j79-)r8oU#?*Zw^rtvNTfRAABfl)>8M+M?b@k6QGjO8kE$$x~l9$#Mu?AOoMM4au*35q0)IRjk zv>iZLDC2bP)3@PaKvJj9ZAOy*vn_$(hmUMhZQsG-35i7(ts!BRfQ@=RYBy{1X;ZpT{`s`WeIlrSRI%^siGA8Fv!bd<)b#noWD!OkyZpW~y$I3$a*gV9FYC}~ zh2MANIXgqc--I=N6UpRvGNC)RbWvFnhDk2J?C*=0xJ7$ihqtv^^&qglFmdV!-2BL> z!0miZJItoPrKC0tn?H)#&b!{!y^J&$*Dp8EZ%wnRNAQE2k1@A5qB-YFhgcRrBOo9D z=ybhNNq1QE^7@0G8dY#%4r8lUeTc(58Ct=OSCr)qCVvxu1B+!3DKn~KMy9Blk^5|_VI zmXCy;xxAu4N;;De5Ma?ZXWtl}1*%mI_Fd=xkDFuptclFi&q)LevWN0ePWNl9wb>R_ zggkNE+tV&jmryH8P4i1}!I+-%BghtGJii0IVXx<3MOnq$KaVfOv3|4|0MZ7rJ>S~Z zn;C!h-VM12$2~lC2D5oy>$lI%8R26_uj4}vdZ$5u^PV)f%qeL5{Sy+*J?iAH#4az@ zH+cIkV}6i@Vi9(RCu`PYMu3KiT^~QBSiOX|Ioz-@d>grE&w!ipIl6qqD_%oy1C%pbb^sxc7MK{!Im94?6s`Ql<%-6y1S8v3*Wx>m7a_87TkR;i za@~xsuaKbZA-hBXcFU-CTOCQnobD=%>S3~6YeB5<*ByKZ%k9XM@sEnL}A#!I3UmR0a7(9@e5ga_Tf!`~4~y zr?zJhb{Xn+pH%FnCs1kw7-01!u$G_y2m)w*KUKh+X?*bzhwZF(*XvvU)Y)OIRJJC+Vd!REi)ow}{ZVG} z;g2CTI=NJ`EIx{OiNIK2Vz*;(5+r~L(ZRaX?u786UvoIk9N%zDKc@1Nym4<}-c5Kr^zIyg!Mt;w33~=kFo-glV0r(lbvs!!((pQxwAn z%Ngqqu7a?*cLxT1+Wk!s%=>pMj*AklHIj<5xRC%YwPp-+iGRfp2EMF_h(D$vpGO2+ z^SAjT2z6j&(*@)brUbl zL}bg$BQ?ypF7UeBXUXmS*<>PK-~J- z7nK#sH^J}cf44b*J!>&UcQ&yP7CTdcS{$^r1cy%HW8v*BR}*gI=c7|2S3MyNQ*Zw> z<>9RB2A5*QCbJ?nf>DnU$8v)qvE7*yxUL#SanjE}Bh_pcJASOkQ#Hj;ru|oWTW+8EOjiMV2 z85<}6v#F&mp5!qJ(DJ0@Z*5@k?BfIt_;i4WOII&C;!t2MO}-DFc~UNWppNJ;Kjyj9 z{ll21(rr`{bf6I;rHfq`SoH6XD|LW{QK6K(rYbN+VztQrj%>%fxCZ$)?u%Ct#;A6E zbpEi8<6-^M0-Qcew2r<@hg@T6{oEb(r!kYTiu`4I(%mUyaO-)KSd!He%BaWCfjQu% zSBQK5=y9>V`vagvQIU6vSb|*?9LhHXlhJ)AB`qm&sdFLTlTx$6UUNWgjc*I;P8r-d zMW#o2I$x^*ibR-CanH5(E!KnEa&U5#w4!fc+wrEM+8UQGTq`=BR>S4Iws$QsPp^u@ z(Y_FtYnCC;=NfZ+$~G}eT5XU^nt?*rPWRr&Fk%B2bu9m2bwg=#Ge8rgFv(!KIG@xk zpi9LTkB;XpIHZ(whTkE3jTDcIwlm21Ys+cc3e#1Es1ni;r)jjK=K%=iye7R^#-oB4`6r#~tyUHbmg`MDKSqv8q^Bgugilj}~}SI9cTKZHt&hMR*iOLoAB z+Gv=RQdwPKK*S1%&?WQZ97xD1S#9LO3Xx74GVKo(v&IZwXBH8otNtSUeAb(bvk2+s zkl4nYeH8woysoh_TNLylt6V*!KX9!4T*JYZ7mGWkwP$Rk2D zI7UbX7sEyVP2h>iQb+v{xfo`4NR#xFYUXx>eC78;+rF0c=!?>;r5Gb3GEk3+;!1n> z*Fr~GaD0!X_VWb&3=izTkdPhD#LG8@6?q9WTR5G&-461{N1nGIKZDvhs19zsuN0RN zfBy01GW_>Nby*y;4>(b_ovA48>k+d2bE*G*YOmMJDlGC?4C)YC{ZGj&-)8MAJCPM^ zfERsNpLZ=OeV@jNlP&{%oxRuKhDU`@o5_?c%ZSd1h|;)#44f8^rq*IfmJ_=DLg@v# zQX8F^FQI@ETZ{VL5BiPYdO3B+(M%_ur4YqrnZDC=R$XtgN1zahS3@@g~oxV!`ARV?5}VsW>?b|6b2{}oj5}km^in!XYzOR zyaGbj>jo$`OMvp6UP2__u`tDAvBRoiyx36Ge@Zz_9P3Tv$CeAE!#Rint>;}Rrn6~H zJCVL3+ots@!@_JG3ozkTa8MLzD9S6tOPv#u1OIr2~m36h|<(z8>G6s?Pw!+}QH>u-~_9 za9JN-6Es|2(;nI-Ps*kfb4&3|29tN^NNK$?-kIuma-Nuv?Y%964v24Om-9J+Cfrz? zyt{=L@xVabofeM-FQ|d=+`|c)fSLVDG@3v>>xCyy*~p&DELkeA#g1?wN09^QwHHS= zi|!YbL2f~m&+%rh&r%1&$<{d)0TyMX3%fR^sgUX)%0xNA@}%1P`r|w;5zMnsAMmpk zk>%EX<3993ZoCZ)b6v-Erc9nEifnD(!88MzWgztj@ubb6_h`G^9S+qc?@}!2TrRK4 zU0;Ko7eeLDngs(CM%T@|E?a41ttf}$R?U)<+LBvd!wd_nSTrnqPOQ4hvGHX(t;2fp zQ%YkwsbpX2L(ip`AldOy6z1t0OG=14a%L-)6Mv~SMCM4Sw}?G9f(Qw&A({4%PtSR5 zOMu8uqQ0vz@tDw#Z!=GVdj`KvOXZgl8+i{kA=JD)LrklLwT!!Tll}y2hZ+=mEd>l; zPq78tZ$~nlvoF$9Q*z0{Nn0&Qi`b)9e>_|50r05w1_Fa-!75uI3V3Tg<9d z(n?q_fs0NU(}iznLjBPMV&J759+w2BuFB-T*uG zEt^>u+B(1HW32eeY3TkU97&v!{BlG(Gb!rF%2<5%4;saM602~sk{x8pYfO)7-=Dxz z9*R<;54HSPGz&SWAMT;grB~rOx^`sd<&WFh(F-K08dsfQ`}by^4=fsmH=-U9J&;Bs zOh9EARK^HB^nsLYZtNUCc}p+;!Ufdp4ZslaWnpzza`l~99#vDkxuy6g8ktJMiN`~>#$UQHgfUb?_mGSi zqp&22780q8QG5jAzKb5?&NZnQEhSX@v!mEWZMb4GT?t;vrX2RrYkT{0ssXXtoi}*D zRQVl%4F}M}uJT{EN3fnYZvl+TJU`~bxH~!{l-0g1S3bKA;BYr&NXAWpLu;@9MM_K! zW}tuJ;6^|Fw+)(Kqz2vZKn2v6PUH&G2;3H~ls5SkHhairV+9@1CL})>Ne;_%_9s=H zV%C^_w;H7CbBrI^A|G6_f&&Vzv?l@={%U2mx;ODe1`=bs6oijqF4cl77(;{Eh`_+i znE%yy&mv$YD0E1@(#x^h)-#s}Ax&W|+e~KGc{Gs(N0r%){XVGFnf+)R>Ip^PuIbK5 z=_j*TPCGVBirNyqapty3sJ3q&p{&bd>b{EaJ zr#ekevw`c1?wi>r_mEq)L~*iA#ZHPk=*go@RH8A!A~?eAuw_WNYH=wksSgH^SLmvS z>@szn!6?M7Ztl_OdINtOk)uZPM(ji4S&6yW);XrD20#C>F5Nk17TRO{4D>M%oBNd{ z&nl!=Epvu}#rUr?7g$yMmj;T)hT>(SaFDDbU}QW^V_mY3!v9QyQtV@TB$j2>OzXe=Y9EDVye#5=e_$8X9EYIj* zS?A07Td{!*RyU&d*Ba zU|Y%f*W7xw<^o+(DQ@+f3_O!vk8|zl{by z7SxOJPS~TH!WniYqRyt;q}D283v$RC{&r+mt+I{Xs@63rP1HyM_%Z5g%T_v(eiK7A z3J=WSTk%7E3Hgo{@5wpFhvh>(af2glZ*@O^RsHre*#OqYN48V4Wr_}L?Yc|_QYUMP zeT0Jq5?aN7Xf*htXsX5cI>l$_Jj&0kv&5rGk-7)Xjwr$(C zZ5tii$;$Jr?|oN&JL?CWn^#@4YR+Q}!7x#M2xvXm!9oOMHe*lTb`nC99+;0jy4P}g zv#;#=6iEoq`MDq}Pvta>CUJ~}gr$@!(!QCQqWlQ;*|ovUT5I7I?g zMucTY?#Z@G*U&9)aF}f-8xJlLEHr!cisckoTmj3#3Ai?@Gea3@*4ER@meg9S7Gp~x z*7-%0SlY`i(~~RiVQbiuL_B+gTc?nX!qT6moXz8Fu~7hapSUU~X7tLw>@zw2|Q+oD_PBYH>8V`@#~gwpa%zT;Pv3An6z&K7mww zwFUzR}WT zviLWpTAgdt{y62xpoaSVSd3ES<=5ZKiu1;`?SpV_{JbKj3Q?SMSbs{n(d7fy`M<5@ z1WIS*IN7lo?xHWKa7BzbQD!YsAjhmP)_%D5~PC?9vrkr&300*dM&s1 zbGI^?7QRgbFRm3g#%RZ(R#rj_^R~V=B#&z*be(<^l8-7M?_qF7dCH5?=OQx<1;m4h zyR7()YLA9Nf-`q^X?bH9{BkD69kel?dEz%l^qY%5LC{zez!y1u~yN98U@3Rqf$5!uak$S~&8lkgf1 zRyNxaw3DWB(wqUX0p$N&c`jVLEQ|aH7I%)~YAxvPxs%63yIs z<>iYA<@=9i;$tjqav5{FT||!!Nh}66*aCU2VHf&nC%4~nilIu&{!IhSu-F>*hj`S< z(eQ8U9ljLu{;W#3^7S$mcURM{($S1JoW;304OdQKsuO2x&~oImE`pWzwFV%yAU8_( z=Q->{wEW-Fda=W&19LoTTps2_m)do9E%K7H6>DRBF0)9`60eSIU9~g3~dpXeE4x{0JBp<0?C>!QH0$9SR1}iRvP^!U# zMTmnx?aGA*uUaJ4%Tg+2)5ig2y#x5u^(SlOJhQrZ(c$!PE6UGNgJi*HDzk9e7_Szv z+Vyk>vg=Dx;8cLwN_yfKd(;mJJw!TrSaN6m4Ye#uwOVL~sm&4(+|+dcA_Hi>o;mE_ zpb{);G$`}-9;B-KYcEZ!o$SHjq(X%D5MpVnf7eZdBKdChUAETVCC{8X3gb8p6D0{` zOZ=R%>>}624cE~%+>R<{g{`I6z-sNbKt2;44(7SWe@g+BjQb@2%!@MGJwUKg(e)8( zSAs_9{4L9kP-^3}>sfHmEVT{!VG#~b%^wcII$v2RR zEH#6nv?Z3@qPO#va!9VvH9N>yGyX! zbIla;?z9q*u^Q$2`w5rVi3XvgnJH#)QxPvW${zz923bq6c@g=Gp?ARj<^M6surQ+b zh}3t8;emMHC_*0BnjC=SrpgmeMayj=T-rL_Y8e_MUNezTxs4)IO;W%>Y1FA zb9fFgoFm&TE)G^4PYKjcH}|bPDE4z-rE1<@K6?pN`=!W@+8LFYg;a}ZV9ic*~pkOgHc!LkD-{V z+yrRxH&-oaGyVF&g_P0Rp~h+|3&bg#=xi@i$FXo}WrUojw5G-pP-yqIi~LR962p&5 zDq>Jb$m}G}VTiy|;)_cMMhJH%dTMq!1RL>@BU?(TP}zwLV9_6$aea)4Bbh@0h~J*c z?r_F%seU-5o-d@E(Si%)tdcQp)WC_VQ%B?^XCO6HdPWM9NMWZ^K*Y4zPRoehNJH$p zh$iMZ%rokLt{F?n)3NJm%+dNQ_wzLCXvz1Zl>GOA12wj&=}H&$3M#r5v|Z}*8U?Z_ z64A5w*No`ydfR_32JB)+;03V&4%rXa7UMJf3;xD=z(^Gc9?4|HXErDPPSNUnKx5L& zoW!8QZ6%(4cD~w-7+riwgIs$`^u<~2I*|DR}BT6_7u}&+7 zjTUB0PLhyX#NZGdP&SkYMW;aaf~}AwgI5O=0$MH#NX4bIAie1+69RGoe4-pvnw%V0 z?k@I&x4OF-G^?w}G;I`lzo))G0jc-;>x)a7D8x-XY8~l3(L_YsaF~49^5#Qew z|8}5jAIQDn`vlTa~1+6f#-Rw0vbWhbY<$sj&DR0ta@9dm|Rozk3 zY)fH9LKr{QDVL&W`P&9&dyfmSID~&#L^L&-pJF|h@EWNxm5YegOORr|{Gf1C@>Wwqh%zn^8iO0@^Vgl0z zC-0}nAdbPQ53!!Rcl}b(nsBte-RD*lh?O;LFPc~l!k4Y&jG@1)dyGsEM1>sX^aWujYf{cn9;Q2T#q*e8&fkIx9eAOiK&^?!moI_|EnL z>fpXc8Pj+T--0UyplF-yy0H-2|F9DWDTq;}S6#xz$1W>D0mH&YZLwoZcJ1>;^wr^dkRS?R4$L_~VGi5vXid7H z(<613x|iGrblu*5b!Bi|s()JW>pdw%Si5`CyxQM_2mt14VwBgDwZ|YIu)o@*zEu(o z+dT_XPQM`tblycnfHW&D-9OK!&E|J;*g)4Idu;?2fc<=%%clB;PECqi#Tp*K*f{4C zGip-w$WT}2O+YXch`6J~|JJSCzP_vBy?$V-ryN}Xb{EbYXR;t4B^zRsh+Yp}O8hI! zr3>5Wp}E+(k%;wMTJ>8-ZhDXugjgRIT^oCRH)>STo4wAovskpz?U~`{17lKehrxk% zhjx&Z#~U{&xf7m`aG=>w)~;^_nDNgYVScn^D^#x>8bnwOtI*2sBa?@8!#;A8p3*cIi5>GREFKop+fBxoR zoWq34e|`#hv&ugU$bP>#v;$ahE&;q>jPEZ1(MBoc^^kd-F zJQf!yOis(2&{{Q)j{U_+!yBC$UX1Mh>T55{Q(vI_*|G%#3uDDO(8@~`Vhas$bvpp) z_+IMKk3=lLCP*I%(s=JnLAZW?hUAC`U1#VC=Q*+#S0i1c$rD5L z0iqGvaKyv>ZZU}E@<;Y-c_s2y4zX6&5$>8VzAE4Ia zQhFkW_TMZ+w)uNhH7+0^cnn}5p8tLPf4|N74{@XV|H(4E;I&fU6i?ncQD5v&GoG9$ z_L0Li;)N8+s15i9dE<71nR8`X*52W-6|aiiN|Ev1{k-i_uAQI7DQh%CS_i76dbaG{ zQ||MIq8%ZYUfzWFbfpma2?lsb5MpBhD|g-Nc##tkX!2%Dm~5SqRgX@*w4=0r2A$rTxC>V+w0>)X`f^tn;3 zl8#uYP9IiE4$1yO`!99l7-F*o4YA95K9{orb>>%npupZyd7GbP=3loFOPRG z)~Ev9tmTb9MH4@%)#9WT%R=`Rzy9(OIL1myt0Yx85&+JO3|o^Vg`Q5|b8P`iR#GWK zUEy7dF;D?fpddyes{x|@d=OVtcA0D#`%0XM9D1!BWm%2PzimQ>x{nMVi`@O@@nUr+ z+qN z3Af*bL*px?mA{A*P9uU1+7ZFVjmy^JD+h%hX$mj?WYZIho^?_Des2NjryJ@r7kyC+-3Rr?;Wz~0mc?)KmnFe3@LBI%H|P;eGi1Hg=-6_FgC_`mB5;?oSrt0#p8Klz+*h2qU6Giv;34hh(=&Mqv0xX+F+sXJN z@<<}fKV@TeBZGAF0vh!HrZQG9z+m-EW@{Xw?E^X~v80zHDGc#g;?8>|0>yXZ|SZix7`J3k;4<&Z`~^_zQyojL(LM<6d~no`3oe3e)K#DP>VK z($zz=)XaQ8wG1}vR~^%z5zq_?$FpjhKFYGvwQMqa|bZ0e(}vnB5u-Z90SR+qr((85Fh6Tvw*)iKR^PHnjj!nPek=pQxCoS;u87oNVP%_LP?S%tSTZ#vvDbxsqmU+;P;gc zLz0Cyn__@eJwbM$2dZ@DqZR+CRuB^YV}rpWmv$(~5=X`i%~3H#;$Gk&^f8Ilr@Bgf9fYXJ!DS;`WTvbfPsbzY4RAr%@{RF+-fEu)faUE#NlftZ%jX+e=h z$bgJ-!j)9F6=X5?CkbSbVUn=cP5D{1!yw9*tpmOClCvsXwzT0f5S*#B7mY%93l`&# z^nem>X(}@z?2J*lSB|0r%-@u~4*mC*)NW}Oe;t&TTx9s)jadk=yhqIcCxOydB_t0O zDbjtuZzS4C2T0r&$sf0_4jcC4B1^&f7&o=u@|4E_MB9v&u|PG))~ zCUI-gh3_@J1U#be`i;je^7DWzqrq$bAv2v(Qm5RN3JOjBG6kX4W8|y-Z~o+iwda5G zCj+~OO9Ml0)J1%X1qr3;oPY8slfzbR54W3T5fBzdsOwxLCeF8N>prm9br6ez-im^&QT+7t0j_U;;7yK{5quw~| zssVlu(?$%MF}IA{R?aPS?iBd5+b4HVYI!w7&b{$urVMDL`yqHJm;Z<0ky5&6zKBx4 zF^9s>5)DJwu9-BLN#9Z5(~rgth-%2q$Tj}Q-^JVf*WZOv3jW96ZTo-u2cMlO|CO{17yex4mYgU?y{cDcxVBvmYl>cBlbM>`X6!Xhr2bOJ|6FMY|YmB=K zQ2_*&WrWSloTtXFG7G6PKSwamIz#Kf_Hu<}7MNwR~}>uSWfMm**Nfvv)`k}yTit!v5n z!&7-bf!841e&(z7Fey1(YNBqtk;@#=6+rAGzb~?5sYgn2!*hxifbs^r)eOGws|4TN zH&SL-85Oz~Q>oX(6-&6rUNwy&V3P@xW}r@+hVI+p6rU$&nDab0DGCJ$Ibu!gv8*~Z z?TON3Jo~af=4GYev@gKc^_1A|6th72OQXDxDwyZ^AgOMrynH8XphbFqS<_0l4zOV# zWl~0Mis#VS+Q-U_eR`N0DQwW}x}NMMc%0Rs2rZ9OoV{K28b}<=J2~}LPYC?8=9CDj z#3F1AxkHEw0nUU&hWipB7RMzVduhtdEr8Cow#7F=ObJNKw6l%R1FM49GI6}`;d>j+ z+S;`K94&ydV#|NfW0moZpAYPG0W__g%JYRN{2qrVG0xK5Q!ybyfW&3yBXIkamc2*4 zmqgldd&VwcNr8JkxBPDyVFMF^wuJzy;;EkH$Clj&_D5f>k{2oaYCby33~oiM5p1P? zcoVwlBxegAHYextu6^wu4I+@An#FsjG2?=)fR)8tA>4 z32H!J7(gi3sTBO@urHL3gM;C%Mn2}5{w*Uzo-{b5-4Db#4K`rJFvX*e_YHz{1fCsN z`fy~6tACSAxmr4*vSGP7e#45LKlG4r>JIdDScuazR|Ne6AOyR*3{6hC2sbA!9m;=+ zoNs8$ZKRPGM|0tlJ1&e!$etzrr&(4}ElR_@JF}uLV+eIK5ZB#Cal<)b%wDGr@iMk~(6kO1Kds!)JMqC7291^nSR zw#_6-1o8F&fK`x#fW_T11XJ;?RW9CNs9#<8Wkc+c%d=+Cz@Ngb+!!!vLNl&?SMnts zOmBWjU@ru=%iV&adVhawIbXziJda%){!b_&-0X&=Z(9BNZj?`N{_ujJ+gg}Jj>HNz zRWi#sAGfrT_^rn>`<(}R-i=KlJ_N&|L3+SO+k_Wj7!zMyqrkczl6R}yr&c!!u^E^~ zzuJb5-ihqEQ65!{&aRI$DgPg4uHyCftNQb`NTAuyEe~@25IL^gK=L%nOX+wqO@7uYq)Sh(L*AYz`O+;+8m^~$zXs(9{^TbP z^b;1mcA^oT?Yeo-PT@1QjNlCjy!>4u!kEqItC=XqJA}Vr%q4t+y4PW3Xti?E-W3BU z_wsr28#pmaG|G1AHi{lXRZ3h&6UBftoL9iqG}M2PpP7{PkmMB03xP`*{ES{U{H5fA zs*!da?_|3wduXGa_vG}z14uTF!Rl{aY2f}}oEG@DLbFajiK-^QCeD=^$MaA){DGqX zdX-rPrso&O26{4+QQuXcR)R6~iI6wtzw2&cczSrN*+UaR!+f{G{RjEA1L*U3tZe$2 zIlBxgBZN@mKyfkzH=(ftz2_qap#<3~6v8sui=gYuf8ptZFXArj%A%E3UfiM&7FWZl zEyi1adWegD@RGeAHn1MXshNu+W#u*`wiFMT+IIEM;#|vdnaVC`s@y@dNY~?H*zBv_ zWe73<2-uI{Yib{{XKUXjw|9solP+nlWNeDm_T(Jiz6@J4CtrCaqJL&o(iG^{ULt% z4rVn5YoS>(mbHtj_PRUAqLXC1mT0SWp5alJX+~IZrpcLV&I_9*tfiPc3gCmE#i#4Bg|Ul966wN57?-;fDEHGs)R1`kqb32lM% z*EbKGp*e+)0Lo<*-=*F*J#L|OJjX3bOR@m;GMhn}%F@-$@@|r&BN)YwQ=kMf!^iv2}-P%Q60%|fJa{VmVZZ`wD&rI+OR zOIIWR=Ahc(R>CD-%^_~>F} zaft2!v_1pbGb1yAWAOl&n<&4q0N4Y`6@4^u@Z!-D=t!~^MLxv}Y~l#}asp!rP`Vrh zB-yS#OkufUsZqWBw=9Cd?Zr3TrFg;W3vuI5>y;H~&e5567fPO;;MaWw}l5PDl zgdJ+?*%ciP@ynCEyuGlU%A}DGPR$c<`MR ziFkFv@dr%nX#rD6WOfSb;ftPjXJa;(Y}Hq63p8GKQJBWI#8%xY8g;P+z^vm3nItf< zAA4E8S*KF}Q;$3lBBT$LH1d*EJ--Vjd>M?6NLzyv30HKVZM0x}~K+=qet zy^GJnVr-sfbUKaEB2-ptBv}ihbX$$ILuE7_M8oZYw_*@sYBQ|8b1BFXAb%V^^0{Yl z#nX71RMZ^;%0Qae-PkD)K*7W8OV^fl;5D1_DWz~bo0hqtVlxAlwZq0gMx?9yH1@oQ z#6KI*rk<+8Z2w$;@B$V1$6RpjwBQM@_0&JX@`C|e*a$lGKNOD56&E|v7s=?lCczKw zZH3Ok1-lJrv*Ko`5a+>IArEYXCm0N|@u!&Ff~M#ab~k+*q%7J11X=Riu#R)3Co9FS ziQ^aJ9OVO$e>0u>(#BJWB#VQm%T`(*sOl{$Udy~>BXEzVY%yW(aWK_OE8q6w4#%?xW-iOEB z^fzHb;E=E0Jc-pyS82)jiGZYJwnHIl*o_NlOmB80r3j_BQA&B5>~ zk?G~B=N3zgnIv~uvdfpKBFKi${*~C_?EG6!ChvwaHv8eyE*9b~{$=q#7|L=sgb%fQ;mt`&7G19v(5s~=5x!JcZR+Km^AyXjg$A2Cf&Co?dRmei$$qr zjfRN*BHp>L9lH}$OLB$yrY4S!<8oW)+WB#t=>&JwbuPHp(EJy^JI|zcVm)fw;S#;# zVw$E}-A#hk^_lLM z_sU-ctmF#ZkV*#GdW~@;KvCIpDx8=!DP^{4b>9o_o(ePl5Q79Yhka7>*K@)X(4I{Pq0wnZ)Q~b(M;8T` zEu7o>sLLpD5Hm}cI`XAK6=>r5Dee?h8}jAdFg1KRB> zr0MTb+LEH+Ju__>2~O*64@w5CDfDE=81PL#I5Tmm{QM~_jYbXV(EGw>XR+=(i2+-% zOft_vP9sKhHeKAl0f{FLAGE`wQbBaU>GMUI(3Wk$yjCsU@#|_MvTnp!*B}AJw@Wb* z6IqmGR$FDp8?Cf;W~0$PE-?lfIX!c&Z?pUdM?tF5CPd<~o2}#(Q#<1sxk?wNtddfw zA*hk7^b{f)qHA(=V2y$vk1lk9O}|BbP<*g~Nv2c%03*%3fZ{G#ElFN}6(<|u5_>_L zwj3E$(BEMSCML1p5mcCG3R=>|4f!~Pj@O?12nX!QyAY-q>J&?jh^lT+BGF-^%@3-yyeSc8$ zW`n8g$_NSjnVG9DLYkV0P zc8TivLo{%6dJI+bvYBx97?s^l^snc~q{YGo7t2(eMx0}mLjiYd6b32^iZh>L!(gQ4 zxf-7Sq-O7m^>kAJci$=(QIR=ME*cZzAbb=6+fg2;1S>0fx*&!?piJ8XLrsp2$&gLX z8;bLFGgd{~Qoc7ghZ_%ofRL`r=HdUQSo@wY9aCpQB3%4Rje6wj`aoG=PJSy4Vw$f* z^nSFR)DtP*)Z?ib!Io%-*WORJ3%;h}9@UXmZPr>Tg)emw^RCD<^*A%%33nOW0^9h2 zqz|JV4aCwUu-Mo} zW;h)DHX2||r>^*C&j5vx^_HA&N_qLpt<0IqjL1nc@H9!T$KkIj>igDq0J}X3jn&}d z@ZrlcxmOvR>>A3K3P1HvA^eV20ZC*|vrRq1fPKdbgVbwmsHo^Wnvt^!>K*Q-z8R&R z0UKj|rz01?lxG8w0JN>76bQT-)43Fhe6cD|hosytNn&cfq1z?du*{c@3&%IJ&_3*c z{LawPGw}TZ)h?G-g4=BhA~po^E)m-v(sOg#jE4XN4<=mj4-QlU_6@(>tDq-Q({OP$ zw*Gryx&uIwiys1E1=;ShoP$M`?BQguVv=0=II%?_UOxaJ$?7A|S_mc6p~v0zSwV&w z0+IUFWVF+1oj9kKI&E3(!NH@j>#%J$+*PHq^V~JC&6Sd4$rrZOt%BS^RIZm_4>1u^ zRQ@caSi1}xo7nAbV{ZBV7GA|e0BZ3`GhV@S1b(=bb=N+-4Bmxst!w0s`tVRgQNEp& z{%}bnDt`iS&2k7~yt^V4zeB#<=Y(uvl9A`Prqu7mhPCEYAQ~Jxoc#A8$=>4}8!;y3key%+-c^hc!mAh20!gxodJ;BOn%7KUH|2A8RqdpHbB&S}Lp378W~oWx6N z`sVg2M^U3d3BnDpT-_)`dU44dCRaa*@*DF%O@injE7;QK}0B72I(2I<&4<1wh^ zE>5dQc-j`rIv2i$?8&s*c^34-esFby>RBvUg2it7gZ8;I+R{1(T_R9gj?Md&I)#H| zym2s4$lAac3;ssQ3Jrm?HVpO!g5ID1>M0Ol3pJc7EqCx%xU}{rq)CdkNru?P4IxW4 z21C8Zf6ute+^qH2orW1>m3mEiJ^U(cNilA^)CQ)>C)V|Rxpt<`%|%+$%+%$M#8uYD z){NYB31!25+t!pvskf~*;pZ}`{`)-VeP@ff?H@-+qfZL0(9Xr;de&d?-YK=Jt#Ug+ zkUEd`m+8pzRVqna^|E}ICp6-*3U5+o_&sU=dbC2V>RN`iKQ+o5W?}o;Qt~nCs)|O- zMgn@MT?DapiN2P?R#?_rZO{6=V8(kVqFMLAt)r>s*Ws|0uAp4-lT7>%#^o#NIIds)#voE?SuM((oT{8RldtENFW6N_-56%*0 zkkYQh>D91&pPlGoh_m*U3J zpPC>MOOq6+Z3o0_etmOdKcQ-XPw~l?40VNKVu6e^sI4!CMAPwZn}sFj69_;2!F+-^ zar>{LNWb94p%o({4j31>Z;=O^wZiAZ=~Wl98k}iak(31Xt>bIUlWH`fwfydd-z^4r zGqH^>*a>pm2G9}BJQO+-;HmeY@y*7mxo}1b@)3|n&4kAMB3Q0t(n6B}|JS-LS&aof z_4dkwq;e5+^C8kMB+InLBg{pf7BX@!I;`^eT_~;&mA7}(#yBRm!hvPq`BvKbv6^aO z+(XvtaQ<1{&n@$GLu13w^QVfcNMs0xj@xrPj%Q`h?CAp#&pa)#$k_sh(?nS6zriT{ zyDYytBZt&3kP=Cbc}hkBr*KL1c7`qOubd@%1qY2>atp*>t`FHwHL%NRSkfaA2V~04 zZ&mV{i&Ma`*7Bv%#kl|!R-xd;K2$^u79)s`*61!pCqz6mmV8QTw7leHWzq`FlIVI& z)H6+?bKY+)VeKz4)(?}S)g{vf>(_|a=%;uwvf@Endo9dcBu*%Rb)v>bx&h?-R8CIV zo(ntrC`1&Pj+Ii9Gar~$NoHxkLLm0Z7&8ZOY_E>RMs+7e3#cMUJ%6=?#|$1emtzA@ za<@=vEE^%Y2d0ud?Oba$_K`Uy?)fz}=QfF9YXWuAykFrXR}aq_gP))ylWB9E?YEy~ zMTotI{YiFFC>=Jy)GQ5brOrcbRx@ln^pwVbOow?pe7$AIGH}cw_Q00&et_xHNkM~g#ZCU zah>$4b+qNwkRkG4;$Y#JgU)X3;Y(~9Pf4$(m|6x*Qtz~Tg}_DVxM}WL_MK(|9b#Y5N^p9tJr^Ub2jt^>wf1gKFLY3129>>%}3rQ=2hIjwPL^ z*Eg@Nv6U3y$U(u+yrF%h|BQDKuFpt`rf9BtvoSEa8WODrTI@iQZS(Evg=I%Np@H?|)h9l8Zv!fqkxMQM0IQ(FwMqG7;jRRNY8 z*p&u3uYmrZ$R#}FZ*a?$>>V5^6nnO=bYAY&UA0F*XVBFdTZxCKAAQ;(6Vh$yrw4vl zJKMT?;C4+Cw^Oiqz772i1tG{JnO8B5P7SVeTi{XoX z9o7O_3CEZ77+R@A2Jo1Ut|=jcYo-D8)d zfk%RD^y>K4bDW3I7nJQD)Xi1YS4B8T1aX?9-;ZHDOiAEj#PXC#s80kD=gK22+s-Fk zZS}O9jZQq=f3j3{jg53YhmPo|#OUq4LG~B`jC3iPM;jdhU3gE@(fM{70z(S*^4Jdq z-B5BIA|V5`)uk;&9{lch`PP;X_oOimZOW+vF%)GMzWoJ~JDHV0O65+5et2$so)BXY zeL&>a#a=_*SgB{M6~C`@auuUP>l^vNW!x^Ths=~Rs1B^hgvM;<(13a~P>X^H>sN6B z2Q{@9mCy_$0GY@%t8o}UrW3}fFk+(*V^_l{??4jtxxe>7eD!&cf8XDaOnl~ zV|BkI!P!PBo3UTNf4j3KM;hXprB_U-Bn=UKekU&*5o$ldASa9?=Px77?gTWY(-h)I z3kx7zkW`4n82j}k#f;@9wT>`{?2#hCt{b8jaG_o_&Z_JSOi3fk1e!WaJvvzS3Z_&7 z7M8NwT4ZgkY-SQrX_YZ%PIf%Y5>N%)AjKFwoG0-H+VLjgX_HC_p-%oumtI45iFa8wPX0s#e;~-|;%n*j?p7r`OjK zfZA$qRUv*Ih({ntBbvb>BZpKbyV&gMOF5W4!K$yjy$@uNh)e=&RstEF4@-Dq$1n{U zGGYEv{`hXiOTt)T{gcCqaNjt9((PA)VFI75)y+Vo>AIzcauIws%9$wUI|^CYe(Xf> zI6q?=$b>VO#%cQ!-WE|zLISta{Bfe+_;Q@a8w3@eS)#zw9Pw#2le>{BN-;fhgrWSt z(Cey=!}_7pAYy}R$glsVScmOX*YW-hk<+3QC_Fv<;`eABn zA(J|m;jM1Yei~`$_!6HD9o79B=9RLyEaH9czu(rYThu%z1#wul#D^mn`-<1{&%(OO zLy~PxNpT$Ob0fKUHlxT5)pK8nb^Jti1~Y3sP7C%*0 zf}YWZbo zd@BvJRJv@{;FXhNs2zX#JhlIoCJK2*A?*&s)P>jn@G{QXiS}3rn4P@589np0BHNAQ zI$D|D{w*#>qoK3fv@jb-er#9}l$u(VV#gOIB>^>{NN;AY>V@iQMJ&5 z3#g|33(4zAF>~emVQcp(S2YA(U-0Qw|4Xf>(z)p5v%X^$@T9(B)m^-waJG6>p{^FV zO`7d;czn%1F1PPL2Wm9bB}9IA#EGv^JzSb{;I6ik7BU#AZRQm`X^5p=7#)3JnLQ8@ z)coXPBI(4K_NVpWR1D{Lz8D?l!6^RygAIA?MxoFF#tnxiTHkT1f?|4+q=tdl6`YWN z%91Fq;YV>F;O2KAri(cvR~$!}YY5nq{#d(jl4-7a4McUG!@u5$RRxzxN`QVOHbE%_ zV#1K))z`Njfg5OunNf?i1czZ9*ixv69d3Wo#gW;cFK_>(ykHWf07MD$ zY5>k)Z;J_p<*egHFp$S@JE|lUpT9?bIiGsbDNX8%C+vN*Fy2!9l&2k9?Tb^?oYti^ z=5nugnAj_V*+ST!O+!o%>e^jzyS%o7Ky(<%2A~Wsekq)T178+9aT(SQg5Mc(heD2<+&MVwYt z>Qz#KB9QM|T~mea%yJHHTg`x;YM*fA(^&FK zd~|UZMqx zFy&Ee18dKp0xV;iDk$51Ogwj*#sEPnEqNp#S~hni5;()ZF4V=ogte8=?v-u;Ybu?l z6Acd70G=>Ct<(W)(|-e#WF13mD;e#wLY4G_l6ALz>ehx4ocGW@9S*1njj=RbxJc{mV$YJW?W4K`o4cc zN_aK3CDa-2!tsi`QUm~Y7dq~mZ9&KrNzQ0$2AU{^j!MRjIPZ!)o_EXz!zp?x>o_dx zV!h!u`KU4r*hA&Yua6AJFkE=bYjeRaxgufS;WB`613laBPXecV9TG|Wy9;<|?5_t8PfPECSP z+5I2`ZB#0WO{l!*+Od_`lpu}{wgjKag(Jlbrv>hC;M%%Ti7znVlu<>Z@&DoL9G^4! z;x`@Jwr!g;NhY>!8&5LPlT2*e&cwED+qUi5e1Eksc3}o8OjDDzbfyf3u9PouZILqhq6_{-V`49JHejj`s)Q6eiOaXzp z=a(v*W^tsNYWMgPIeY~~xwcntzmfiSL;kMBonC1GktfL}D1Y5qyCtcxklDbA)S|5~ zj)Y&ka_!;?u7ArQo3D9bcvPJfC35yh`@rNX^~ru(y4xIMY)*PFDO!%Nm$RbnNC8KZ z3A0xsy|Fd(Qa#~0a;+j(`C?N@S&pObCMIe_)!G>knby~pjL`s6tu`w0s>%>Meh4_e zE*ja}zw}u4jo=eq(o$y7n9ty<4WdOvCdt6DlhMi8Ai=Q{?zM=ikD=70gP1@{cmXPO zloWfQq7@+Ydv0K*ZTORVTG{4V3-sIRklZHrRM~&X-hvlKUjbIk1Tv2m@Gamfl;aUp zmB>frB6Cgs4n=^Rh^?^U+^2LqUnCZOTGVx#3<@krEDuD*ZX&g|%cg^>kbjmrzVFm1;ESfD=vcb}??Q?Q`% zE@zHC!1W3a9F|D2bgIMaG)MWxzJ_eT{uw$mM#%c*4*b{{yg_xc8c0YVe<%2q zNgUA^S}O6hV)S<#+UGWPU7=4UNC{_llEn+>W88(BNC|>X6lhv3<77qWY`V0nHCc8c z!|^ge#@hx`_)?byB8=>78<@F)AuuwH0{-&P>Ws&gbo-9C{NwaroJS zX|ZGbJIQ9@C;Av;WC9AB)DXd0&hXU*l~()|r3Wy|gX!x9ucN z3#aVF7?{VJH!cq%QRTUFB?%; zx`Wd5`zz73Ect9ZBH=nH))g!t!Po_yy}7mal5tGEQ)JOd4ct^3i*t~RLm@|636m>2 zRDt5)EJ_(_){bTE5-J+B+_%m@|8(=qNX`fjKJHE5qhuf^2IOkS5e2c5cqvC&Pfh13 z!TJG7B*G{z74Wjqa=224QrD~nwe#EX3weS>aWJ2w4z@pgHYYnpU zw6XRCjN?*yRpGb~`gT_?v`r)u&6$Mxrv zc!;MSKWJi5P8EF$Ug3@YuSXh{$Og{!%q<_FBUw4t^AJNoQ_@(;z1vv96HvW~b-8lF zeg^0RQLnJAN~BClV%MwdiTXM$Y_hFOQi|i~ZN@?g0L8o6$>Jl3;vSyVLuTg)$_8q3 z{dGo6=&<}gBa~XCKK-k@dwTOR)T36!KLe&t_PfIx>{xkHIXiANizUP2MUh^^m;{Q4 zwnQE2$%!EYl2l)(lOPec>QX$Tj-?bWT2=yIlnU^PgQc>L*namuYIr{Gy&B?JAx-mV zK63od)KI%<5l7hwe8JVk3!(3#mlfEImCeJrDI38-QhCcZP7PiTy_bMqW0jWWRKRn@ zrNHxwXYZo2?#o0DEoUAa^tq;|sRxn>78%OxpzISdG!zl>7c}gI65v@0aqHPC(q*3B zqQve|xF0MkFxCeswUQ&2;VsbO#hV@f5?Z|;g?7}4AdO8jr-I_OhYpz<~hut>(}uuT(~nM zXb>~PiSw_qYl5Y$9aK(QnMD#Ey@uYaoAv}MXHvj`NW;I=GDo^iw!n28+O!U71-POU z&_N;OrB|B0W=jmV2LD}MehJ0rM9y~Kd9l>aSU5AU>2yS$p&g<6v;p5=zxw&zA2j*} zvbg%^ZRZHKBd(aX_NKRy&i8Se4)tB}_4|QWJ;IB&1P<3oWUt#NPgB=sFccAHlhxg8 zZ~eUuq%8YwaZGIH;80*hK`kS!{XryoDd@c|>BH2T$dJeI)3cB9<$;H@j%g?@GPB+B zfy_^1!oc+tTE^(xGhn;`k&hGn6R(HjG=;bhPJAUcOW9amndaGeJWGir71)6dmiu*p z^ePZM`;XG*{7C4cOv|ToOa6GZ@``Jr3<(cZT>(k_dnjIjs)~Qd3nfcYn5!~0sS?5~ zKC1l`^7^n`ed+aP8|QfX{-Fni>`k&(%x5zxd(0vuN&GD%0dVH`*@eeMJSGt;qMGjoV*KY1&E!u^Er`;5+$o1gk>H=5Iaq1C9MT|*K~whP?%o-=62D$ zm;3d`s@&f6HE^Tr{kMSp^BxZ*EY9`td%~v6&9VVv$SjaZKLkB+zdMqqyaT43QjQc^n9Gkm0 zPHGq-(6|1d6vfY9I`a$)F))eJzO9fQA`YZ0mY&tK4DF8qoXqbGk5$~N3`#=$k*1O_ zhCv-8rZh&xZr@;B_4yE#f3fm&cI(RW`U)!tN*=GPrw-cJXZZ6S^+G$WfAi~rIOjw7 zzhWsI$bVPlxBl>d7XSM}4w76(-UXC&Tr%`&U^E*L_1tSHemKW-ON~O>i z>Owz>KtPIOKtQDaU!_vf;|_4Dob{s)a8x*05bzV7-&WsuiookJ0Y(!8Ejzi64K?lM z7;QDBgRX@RlT?cB>oGOZwwiiOPg#o|4VDA%b2ADG_>!~|sv4gf<>T~4PwMLj{4mvI z_&!arpD^{swBSCKh}oJ{B{k8K^my!#`4}-ZBcg=|{o29^(NdjK)rO&k6IAId)uksN z(yAh}g3FR0V2Dt2Hp&Z+dTYLpL>p8MyT_*f6>{BFEY8aP-rf$}*Bt~`q0ss}iQ_wY zH+t2?;>@LxC=wV+Cl(h*mcke07FEpsb=OcBYsFw3V!)ad=bov*^(V)BA7p0C>b)rsGYP4w^Dn$*xF zE**Oa@zdFg20=g#^3;^~*og2(zqmM-WJUDVnVsLU8bWnbP_Tgmlm`?n?Z=_KsT9$f zR-T#FR{blG;Glw5KL`wk#O?86%B&l*66EVZWCRC7Xz{+4?(FyY0m9|eWqxX39bMIO$NvS0SmWO-}B zu4!n%M+d{vbK$P9W?zF3%FF)DuKP(JmS7XX+aoM1UVPezPH2-foWt&TD~`##05WF$ znBiQX@c;`xpQ0)?=bB{?I)~R&lYgi7IE=H`!gjmBr;wDAz1P`tjlK7joBa%OC+CN! z#ror4A$u=TJp8^)U0z-<7Z;W!NHK#M`|P*C&(ZOE^4sJ6Yx#lI*_^3?ac&|6ziz01 zuZqZz^Si~0^=4Y8^&%5s1#6zb zm4XrwT0}o@`1n9psJ_TL3jf@YA9)OjwCf_7X6TS8cSL=lmM9g2l&RE6qYoatvw^3D zjsS&7vqE$|k_(#tJd(9=DRA{BSqgkcF|~k*W$`d%KWUL#j^y9A&v|pSlNE7;QPHYS z1%?5Zl#$XyZVg8xafe3I@A>sVJh;Sx{OWK~{26~%cp87Og52Szvn4G{M;6YaDw;qo z#I`bF$$~#kbzw*uYT%V;KE)u5K(KNVr~Ku-!HGZ~7-t1d@zWTqO$!K$AqN3p2=t?c z{&`6aYYY+Y2p+8hD(v@*Rb(oz@q5QKqCrvVoFG=5fQG{lqb(87loP4=Cgz@?b&bw zCl*XrmnLjwWzpcC%&iZ4&H>8BQ)+Z8rM)d3qI>8J;OGDPBt5&P^X&k1dL;bM0Z1%j zoN*vM1TX1tcr7_l%sP^acI*%2X|5HXLvbX!6*=3)@e$|Jc(s3QabRcJvaY7(NMbT< zQ@4m~sp&|@bLCKKY!?S$my$?`FXj3T{MAew7|yC zHH{GKt%-}QqT&tpMPR}Vt!3uw{{82l*TW6HQBEWp&I4v(8t=(-mj_2TCKb4G#AHbK z``yW{$#YpwsncxttP|@dw!%^K%SV$#F7NI{K37KyN>)0;ohvmtC_P~({xmhE2JA!!K>>I2 z%;PWtrLi7}nt}~}!Ut6NRq_W#LNb$XkYsw15pu^Jbt4VE9$w!>)3n_XFT_+Xc;q{afYt#6L_e$ShTKBO}X*_YXXa$}ly0gk%E2yXb<# zHRK)3P&v9(B0er8^0Nnq+M2$I-6cu?%|)%PpYUeWo!%@YqQd92Ftz93dk`@og;|DZuH%*?n0O;Oi)p23F|m3hNsgDaje;Cwc`}RAs{j+E=ND zAu|+AW`=(nPQn-7f-sPRxQqBBwBS(0+3TV3!7Zz$VDprdQEm~gaAF^+Zk1m?!0V)0 zs}7e|Tt`3u_(BKfPy5C_{`~-5QF8)kX(E=I_uoa8c@jEaPDv!|<48gtHj77qy^rDK zXm4T6wg%XQ+5)pRfZcwO@=h~N#0p@bkwxP?(4vMNe(Na2$MRObs1l5t@}O}&LZCv& zRiA|M^^2L>wo#VAL!lZDljmG}H8Ha-QcS2FJ2--?4TGU4B<>fcj(z5iw%n6$?4;Dnv^M&3 zuX6e9<>+m?FNg4*ReV=bUs`ghF2rlqiw=bx32esDs=khQKLqA1;i5u`Sa$lap=rR7ZnQ)4!bjIGKE1Pke zXIGOFQ*%oMdt27Tl12N}$LSRhLnN1H2kN+iw$L((QB=OYH6T$_&>YfQlL)zpK3PI z4B%cBRpLycNkyHJ+H+eR;j)z*vMQvUzggA=oty~So^J^hr1YZPyoNg#k0P_JX(T5H z(h*uk1|`?7h102M=I=u|)!nT|Tr9pM4jcnloxCR4oeGhuW9cJ1U9Y6EHtmes;NH1H zHNp!Pctfu)^4h?W=aa{Ejp?aq0Dvy31t`$Z)vRLg;-%!ytPe=Tp)Oe&@K8@8my(u( zXbH4Ps|WPzsv&G+^gb20D;(*`#*ir$zX`4k=LvgBM%G!$P_`vG|51{QYjCrWa=SRp z5kZ2;?mx%uT4%Qqf)l)`5c4l!1L&1m=gciWvdh-XjKr8uXdy0d-x097b*6Li0}c2N zFJh1PW0H=g?YkS}POV7qe+)7Lp9M0QHzdMx!?I)8(uNUXp<5+1B(8MdlEDwa?t5)s;lY|$Kk`8C?ShKFlWsya)iNF9 zQvL5_Y&Y9zwY_sK%RTY}bOeD~0Gnrcv{abTW6KrOUHkY1N|$=~D)ldj+(! z58H&A4n@431V?9A{8D#%29C=7+Q;MmCy8b0XoE6%?2%3nC@av+(~B+JOCCPI=qgx8 zzlh=|0J!y7fumqQ)GHB5F(6tr$>oOwZ=pRB6W8OFBX!&XEJ6Ki^qi5aZOM2tAXkJl z>Ru1VR`^)=TF_nn_tCuf<|jjTb_VLhdO5ByvrsLFy&`I66})>_Kr0d-7N)#qmLg5n zysyu-FcQ0^GAa7V){~Ps&MrW56RAFa-Iwhc$Y6HkO-s6@By#-^U7R?CX3H zR^_chZkLtQHcfW`DLc|)#x5e#LS6^w|EJi zgBW`P5Ig82mL3*+xq=om9>icEPb_xOO8XH}8kv>@*2!uc5;AqnOiape)i19l|FA3| zw$K0u0xx;BTQSH?9Z7$Z@^rTl_MJeS-~^lwD=(i;dR0d~fRkgbeWQ!d(B*CNmbvbb z(++ACHlmvvZ|QB^!}pClKFQfRf6{G>c`SFRjLsv@gwP!-%G+--ZlA2Q>}}#(q!0a$7-RUz&CSKffCQmunJ?d zw5`>F`tY>A2LAO3vJtQp9yFY6PV`XxU3vL z#phU)q7$oovQY#|WQ<-s38J=@9)l-?kfvynFGh5vBQVF|0KXy7Y@Y=4R-i3tP=w}u zOIbI`4)LjJww0q1NHbFn=R|Zl(Xmp8on5zK%g<{6^RG#gXTQ_3x<kwbn*0fg}_6 zao{#FF~FuKXc}Rwe*7DWGuV#mt6QNq`96VGSZOUrV$3hZ@;hn?Kgi-A=}u%AO?X78 z6Q`A2&@loPSneV`4zj+_t4+p#yE_oQcWGW}Gc>ce*ZR^42gmA{3>>@J2|SjUF;m@& zVLQN^?AHHCb*=VBJ++ji>BR$0(65nNWSj4Zsethw*%IzKwk5gm;eOkWdhsP*l|2TT z&U!xxjP!okb_xj^uY1fjW5U|y#R9n_z+yNRcQYo zMqb+RF@mWFbYb+tY-2S@OGMYa_P>?H8N1AE&E#~!CTA0%$Ln-XQB>k_H z)5Ez)H<@C$ZHB~v7i7l}!nxDfb&eOGwY5|pzCiPh=HG{0RDk9+zY}xk~AsMG!A)ZhG-vC4tju0p5e(fjva9dlPwVknqU&zLb z>9cw(6><_i{Zl4I3M_UBv%=w4;IcxC4UolE>kV=eK!y1yqw>#gPCLMR&SC>&|3?Xj zT{nZ=*NV`eL;4?iVO!8FC(%+`J|+;vlhP|p*dcgoCzLByWRp_#2I@D}nm)lY+K-69 zxgMp%rNhSy=XGBdvhRabEg)omZar1Er>-8}nY78K{J^yE=diM1HQzg+xuUFHIM_iY zq4H-Rm*HG_UNp0K_m$Uu8?dDwJ^>JLMyw=!fGBi8zvi~K)v_dJ=ltlSs@;+n|L!Tx z|F#a4azEs&vLtP2&oD9gy$yPEuj&LecgR$NhvPbId&>hZ98?K!DINaHhC6^=UK#dC zv}_4A04k78+PC4ql)W{54kafIKW^Y%pe3HK##+~D9@eB}m6Jykv8<}A+g_Fm$=7Kl zsculL=YA7wxR@%Hc*CeJB=R;ww$RU(d~%DcAdWwNOL>VYrhNE=Q`i-VBh~_aj74dd z1_#U#^xVF_YD~b$%-vD0ry4~WsH{A|T@{I_V5&y8!IkZ_6&bsmY_siJx9JoqCw3mz zr57snd2|4dy}!E5|3Xhs@z+peO{O3wxXx7QRrAEypAQV@uBEPoITiH$`;Gd{aaeHk zbmUfWx%jXwdZm!}AWc>s(3JxeSyNN}6ysPXps2dIZ$FPn?6nU)yKB3in59@Tt8VrcV!q0ga`*j$%H(;x4YcA>MsgoUGqnkTHEO z=1YO2j~499%V!HdBcCZ_&(djM)Wy%NzK5cR&(g*p{)8v&==x{X3uT-YU0JW)msv9Q z>pVEa&%BV-5uiXku{t6vHWk->J|7rIoqk$A71!O7y7)4a)jy`(xA`K~v*ZtWg3=0e z*P>@e^GcaKK_$0T{$^nquhg{RN6-I^vY}l@H}QM*MN)A3b-t;ggH`Nv+Y)t3!gyw> z33gh~GPJFHB1%x6VacjmU8+v=L~iThoty1Lvu<0Q4>-V#e;EfK+PZU@y5ag`8yZ(B z&Z2TtAPw-od*MOfZiWE4YCFuU?MV|stLl+-A|tniwMz5z*^2Xx zw5%E_-zUkX4>^erGckO)=HTG|$sU0WmG>iXr{lQN$ALyvNv&>Yj4gLT>o2 zmuMm0-FcftWtJ3SuWmU{@B+yZ*U+a=py56E z94RzY9@a{3TDCgnG9#=S=@S2S$!C3}xz=DlF7r=x;OllzeBRrA5@c)R5M=+Or`AY9 zRLs0-FhoE{&3KVVmht&`F0GM_r*KHJLG~&5niel|`Q)$kb~l;wRbEQ0md7=Tmv>;V z-FN(~N)Iz>W=zg;Pi)ZDmzxr@(VMaF(bKN_yMt__)YxjAaNqQdP057tGW9T(>q2v6 zy=sN^*4{OYro@fNg_pG{BDrsu+V>&Z-)IbxDYclkP;`CaujB|-BDZpIn;nP?_0N-I zy*YqDG>Yn=#Wi|^|EM9BTW@Vl!5EMhX9njbzCbcdr|#_UTfC#GBN2xkUE3RPC#Ty$JOwH%srFR%iA^`!ajND!5qGyfTq+5gdiUetBcS#p$Wl_OuWNbeR~r z37r=p+v;$RztlamjBHpg$VoN)!2kviAxL49iP{+`5)Qg;4))U@IAxzO0RmK;k?A`N zE{{o=^;>sgE6}Pn3aT>WSN}JH2n7v|{0`)|uvp6T!f1gG`Q>j`&C>B@gtKQ?e@WQ= zpidaFbUxKy_53l2U+N}!!~4zoxYY3S@@ng6bX(qg=N5&pxjC-k5PX#I|9;g;r(c|! zesCQ4zqa%ZUL8TaH!XX`;u@ z{qGXHmSW0uMFVkKZOLGTfDvLLdq+Y{vTGOuaXROoT|IkAoLRb8{WT zW89FLvf`Pco6Nw9vqS8&zOhoVwd3Un$Aw9k0Vk5^n_+FwL)GDkL}0%2TP%w^;ALnK zyiSYM9MlT#!YDGOS2I`J2m&(Y60bEnO6I56_+o%hs>@G(bV*Eu^?ujgMAIQ*x`$wv zwvr+pw?o*$%B!|wyR@eEaqAtqpGY4Uy-_lU(}JaqtW?Jr-dnG*|2Km+F!oQq&UfN> z^cw^hMEd_1`SJXJII@R=>&_cPN!MTD`gL_=X{|R@!$x97mhsM-d19F2!%lH!Y1yLF z2(kz)65tdTdS!RNE@ZZvp6m>s(eG%#;ssN#B*4KPDYVdw>cS`zL=5C}gr0?bZsxI# z?Mm|{s|iETnFC*duwQox`dm4*9&KDnr*^R$!#3YsS+z6-Q>t&S?Db2W{GDs*1kBBQ z8;?}*tLk7{EE#->v}IDTbc9lTS>A#cPRt+&tg5St1g6cP!nQSKT^QOjGn0bCmc4A9 zZ837&rc&9T=&5`Psi7E8RUYcZJcrno8NMglACxU`WYZv{(*=5 z)eGEy6Nm!^awV#L^s2fW<$|%M&eNbi`w(}$?He39P3&VEno3*zI5sMFY#_dO^uV7S z)EuNs(sCPhjy=FF03#`2ERb4q3~s(bR@$N_T?zpASjMGtf~s$(HbN=I;gT4%%U$|A zf#k46waU?^<)u;K=%YvA0v*-;F4HYUHnZ8bZC_yl9Ca*NUT0{=09*5|j3o~>hO#CY zYI6E{uMsXe8uj^W-M6{xGiOn^r!Uw#*=pU@A&-{ByMS$P)O%~0JL0aR7P)gkiuKw9!Y+>;CJI*4e7S z2$_iiRvWlBxXb9xjwQV!l=OT^`;<|-6CNlD_mL?pg>Sh-a&YAYhX#8LjXU;6s%GL* znUL5!3pkKiv(04&Pd)$V%Rh9-i~ErqEWQi{G$pVij8>8@Cn4CQvWibZZ)AC%e*+cK zRlZk3niwQQYlI^ke)_1hpV$8jjq*1wJBN%By#Lv(b8N-hl-7@`T;@K`f|A1-MGFN(t_x-rasyvS{Bt;yUZN{ zygwDmLlYzzm8Aa$kc1>q2u{p(Ja|CZHmrEy4IQ7!i-)uqmE8H~T)iu_KN^Ms-hV@W zl?!Keq7IiFAMdrZzZ{>dIQ@Qm`#5=d5Yz2I)4(3yTVJOBrBOoYosryF+7G+iY> z_>8@+A6`D+tshpev4z9(1&6WAPTRKwyn(eTr=HGnJG^h31l`n7a&oJL`-11hLyRs> zpdjN{lr>`*ofR)8R=NsIMW5) zn0^lB6Tm=s&yco+Lp*baw5}5d@4b>=!lJCpeqfVJ!GYyCVT2}b*cmGao@J8?tR%0B zul<8MhZPN8;J=7@Kw`1d1oz7YnL3g>s8h$7BE$&8oac$9w^(E?TnT)SErvpd7=jZ* zkIdZ8W!3!6`Iwet^Te&P%r9;SJU_^OG6@B} z_Hx+RC^LEraFLv1CiLGW10rXRK$w+OFfLO(bC)5|a;=$ z@GpkU)n=cVE|tY9rh0OH3W+xbRSp}`3{()vSj`NSJl6W|11Y#^#8Lxy^N0A3rcV14 zzTOyr&y(Ds&ep^RKuX-wAO8;+`?%OW>YY4^zwx~U-ng3lWVH+kXhQ>TQN7yQ#7;0Y zF}dXZ_I_<-X@48eiJGJ6frB!3Nnt~`U~hUs{GsUM{J3B-5@m|bEt|M(q3S=$75(;9 zMawTd<^Z+iW+?{)jN_pTEmrD%phkXhz5k*e^K5V9AxAF-(dv~SP9Aa5!DGKMi9IA8 z=>0GrI0Ce|e=(Gh{xv}>v2}0?huqS%YJpQ_{DMnIJh%a5f}KL=0+m5=p^ZP!OJ4qE zT?)h8o1%yo2>W%Wf&qy`vsMFUV82Z{4!}nDwr&T5Msm)vwP&MPIi!L4Rd6Ao<#`yy zr3VQh5C3a3r2Eu^gu6+n@8k5s{;Lf-ojG?&ELNfCEe=wbBFgXf}k!4 zAsniPPnm68WuR~!sMH=)w9<=iiBCV{q3^D+-1Ci&%(B=7tpNkuTgMB)AnFsQ{7UPA zio;$PtJ){<{$j#s=A{p?x8A7a8Vd)^X8P;~4d!HC34Lz<)#`9rl$_#O!8L^i`cqi{ zUa@%UG@H2cYId`0YSL$-+peqxcY>HH$#pQ|28)LBcfybW-oQJX;=vf|dh}HIuTUpj z#c^a$QABL%r2Fv67>9$Mc=L!3!R}hj@yupEAX``ATBRq0;M>MS_|-U-?r4PD;v`G3 z3XaFY8rg^Q=XwSsS@8shz}@7?u*+h|_fLHFl?j29g6#yD<%_16;iR;xi1E=a)yN`j z4T#W>6uF!LF=SJXYJ*q_&&?0^P`o$!|^=lR~ksrsW zfdS1s0RfE?ePzDo^43|p8-&^EkFH=)QR(ll0v)yoCtWXnEAlE zBneEUz>ha)xj%~FFfN!JhIeqoW^B#sNOiDi9E|N-1W?YZCxbuG3&S$Wd>~kGw1tn* zp755Ph8=Y(E&GF)q@d*nh0Uf!tcD`~Wn{Ud9GW_JM zVKQwyE0MTUU<8SwC#Zi)I42l)*s|9}zF@rb4~RmWxaOj7V=Ebdgyun8p66im zfJBYX1SP^mf_*cr-2U0G^->9*wb!l!C-PmpUB}BjI&pfNb8uzr^tS&_$n&#-o9a%?`;nbDwATErkh{2TmQK4g zh(5lSeLCbUPoKj9`x2$lk{wFnAsy<-rhdY~j`Z@b*bNfQGHoPyUKDbkEz^PpOI`p# zJ9N4aYV=TUy`TCwsf?ribeSy)gnR`E7MMW-Bj+btY_i8gE{C9#S8IA3$CU|-=?S6} zV7E8)(8CJAF)A^O4a(EzXoiA2uum;yOjJAq87Cf++z!UnJ4j2S!6Xf{k;FW4b1Wt2 z3irdZH!@L#vPox@28q}UCzd_TGz}}jK}SkF*v zN1$qb5wvd`^guY&%oaN+fPn!UddF!Xe;BqHHunn2GtPg|h0>X5L5V@U!@MMxKWRvg zp~dk(BlnbBR0MHGX7Iv%&wl|_x%TAxje#0?(fVjlsDr16;k}x8Yu*oj4O~-ciF+( z;rx<(`LC^1u&~O;r>c$`#~i$5O>0t|52_x!j+lrGhkn(Cz*=se@Q%9>$4`0Iqs5Vs z(xeWE{_p0pzfXikjDO%8m+mIP)BjcmZ6Kdg9wlG~OHJWrTKgwfZlzMpB9W41!BBIgXdNk$P&S0iAiP z{E>r*kP*Go;v00!jeYgC?_4;`;zv_oL*_XIpw6G^H2tt!^1{Zj4tz(q@izjpwMQpb4v8D{mJhmAS!cQ1mw3nTLx{3DW;tiRj0o} zySP+!@(w+MbbsADc+l1Lem;9Zt068Nt%#D=JH`fdspJOAxsDfFB*F0&Hmp@?hlz8wd*bt>35LKhnqZ-;D4xwhC+mb9_B1p%; z;imC5fKa-tzR-X?WUAr)-oU++csFg>T0O^Dm73RxU8VZo7CVBa7yKH}8)tIfgU+HH z&1TOA|MdRt)oGi%Diw)~Vbc0&Y2N^^`JdnvJudP5*NDO>S#v5ezX!$ryTJ6uTX-_3 z;yLJL-4O7$<<7O!HMb;qXPpQ`xi93XXDi$Ya!MlRr_cZMN*I zG!2;qrFGM*w3g#mW0M+tg~O18tWnKyyj=2l1%_*EglxvNr z0Qo19tioKIOGFpSi&0~7ycjZ+ELIo?5EhA`Rn3(&+wZ$?^PP+$vF5{JJ%G5N%+uKR zfFIYY;j|W47@_jDkz5+0_oVVakqC}arOcre!wmnGc(5=uZTzi*sr&{mgg4pB9u(kX zR*d2&cJkA64n??OYvv+TNTi`BIPEf606(A`lqPWHS+zTKbWXl|?Mg9LuPYGSyR)N_&&3w&cD7HOqpQEN3RIytAb$2iGDHy8x^trlTh7neoWxIN zCEWZrlV!A5ACr>7@V6YvsGe4=Xxei@n5YnWtJ*5mTEUhDFWLvgh!!b7DGtG<20AHy zAq(6ugl7fD6QNA7>=Qt_yE1*}93^tV;~e21qF-khLqU ziNaJAztRj2RLh-7r(yOesk(jz8DS!b);eW~y*{|0$lF4KlRxJ0g@o^mw|G%Y4KZ7Q!366TyjargFCsP&*xmE%xw-2B9v%wAfmyXu6V3L{Oi|lu6WeDwz=vJj3%` zzWb;cCJSHio~)kmVizbY@7$le+tkRC#l%O z8BiFs>%ys9BkBLTtB)R{7jYxh=QkxwmW{@T!h2bRs)F;8KeeAG33JhBGJ*HkcygH0yr-nh{+%7@iW8}jz(Ei~)6ZZNccZ((!Dzgh<30w=TBbg6eOO%pB=99{9~3ykE|!7rJ!z#+s{5+aPRbk)Pc z_xt7{>Z5Xf3<(kbDoKZ|b_^Z_)M`lE8I8j%ze;!gkV;LNGDV%5hfYNZt&>z4)Y4iq zOeY|8QlSb1`*QYkv7tX23~q%;D)gL>cXCVYQ>Vou;9zv){ieb*0TpADe(lE0MVHWw zB*929dXBR-pXuAycyr2eNQU)aPT7ygCh&RtKPOlZzVKQ8S$`(UfrP;Z3G@|mB)Pnu z9WsYLrDn1M*O#32UtZkUf?ifm{l(p=sy-7E&`qKCH3q`c%ae;s1!h!Gh$@;X6!_*i z9AN!50<&B+M*iR z0ENo(t_^^=8SjPKz*ZIb&u)rf5G7N<~l!W8tk0ZEt^QGzLuMN)42*Mexbhh1T7T|kR*H#Guodx$4XR7{HG zbUg4f=8~Nn)V?t)GRgB(ziwHE~Oc6qOYdXf!7_gOc? z5wy>YN}#21n8g|qMGE?ww;u-8YkQAs-lp?Fq~}vrBLc_m2LxTEid}E-9gk0+0OJTN z{q!x8>=6NO5{8Q}d~KB8KI?`A9b!5|7IZHChH0?-$abTPqQ*n?qAJy*DS#j@IrUVt zI>^?U3L$SCoCjA^QI9~Y2(@IbdXSD#;gElhfjrx8`=Cvj#ybzWxHoB7f}xp z7ZJTrpd~)?mzEg5$X#uIW$}5L_`1hU42In`H*meX!jgUbUtzwT)5O1&{LAImQR4J~ z8jtnTssas}fU5F?DqUEi^rISGB&4ML8F-WRP!9e|@9W;~%178MFU%Arn`k=jH~0Ck zVp;_yF9=!vZhW$WdMaVhiz{Nz)>;7N-uH*hg*5wks_Qa{(LGA?e$Y1FEp%lr^`Q(K z3g|P-l9ho-aj)?fq?z+#8c{4U2qkYesUQQP<>LXNy`T5yh*x5M-f6rRo5f9dtLsvC%$SG~m>#C10{1B6VAvg9b|6^OpPhbz{9VP~i6Z$q{4#UZL zJM-gsZ@u^`Id6DTSHzYqvE3@hxDMx#1^8UNiTVqkYgJ5L_M((Wo~=1sRZ&8xgs+9z zPB*omHtcStNTgNRQ>GMJ=e#%+qD0qujsO*d6u1JK-_+ZIQ|oCgN`lopAP!^a`gJcl z3?nu|B)V1L0va&w)l3(0f zmz>WXMk|34IjsS(7iR3*oekdAzCaiQnA|aU7XeMTadP!<*9o69X4jeI)Xp9xwv5XDW|I{&3q*46F(JFZfEYCD;t|CouA4X zK}NY@y|zT~Gm1x4jNmd81sPIl;r4Y{y*Xr?y^$n68p?(EXgw}8xO(9hN2RH^${w)uUCC*Z(`DM+jO(1C%2!$IrLPZsGRT| z6p!~0k7hikOeE#4gZ@%)GdY{=^2OC{x>Z=0t|_!aS#ZFs zEP36rKlu6{LmmQig@jce^Kj^nVGcCm$kq#-FI&mWPgjs@{2@&ZB?j^g z17O)kMTP?F;i?WIYiGIs`FrcGAmV2Fi{LPLHQ!)Op9$_U3apcovN31*63rN3=#KBH zhH&zN4zTS5Y2=PM;@fBnx@)^>qEon94cnktH_uc~IRp$-Ll(f^t)-DLs@zgg>Vv+8 zrh;4UuIlhE44kg2#Qd?^gv{>szxXSx?Fqtsc5CZB+his%C497dXaOlp}6WRquzq5l$NbFxt1SQSp%-YquR4g;t>mN=;0zry*dr z$ah8ozFV0m>5ypFl6IaNOgPw|mS`1mYjEQXBT0IX5rr7x!QLM*33z+6CwkO1gIEof z?(LIH)ll8tme88)13CIOS6@~(Oi>|>Z@7hJEYachu`m*d6xAZjIkrdBH9mrNEsMSa zXtd98*s)IbrA4svOy!dsvD<)#8%Vt7<&TqkDbv>6cnn97xxHnR)SVV7?Y=lf%IR7K zdNPVy(zv5KXVsA+znjtg69t=apjQh(Smq+~447zR7BmtXoPi|4b#mpeRIac(zj)=W z`Jf8giE=7w%Aykz8wYh$D>27={o`!5pSDk~=SX>X_vS9Oet?DueTr=Z;4q?^>>tsk`J@_ zx3#;oD$X)O$!-V2#p@ni&1Z2k+D*hu+rk$f}&wJju{{%?`~{BTc_<;fLluXKq}9ep~HoLqow?$=w^ zv(4X?wKWc28^NFuTkd6Z{Z-)Cz3t2r_LZWr#+9k4c|#my45}Q_)srEv0zsbMAv#3+ z=i2~chU$|L(?9LI;BKH1hL+61N4V$1Ui({N&!j*bVW9P0MzwGfL%rbu8O+Sd<40+f zttg@4GJuyAK?F9izWFlPu-@@ zj83NJPNvSzmiBgxYO4Rg;QFsAYnk+m3H5)gvN1x#l^mErKtbw2KvMtZu(Sq_Q||w9 zx2D~3oRVSF9IV~1D(utDteWzUNw2A(h>4&WN18y*r!jDzvH;GTvdp8uV1E7?R5=vE z_-v;{=cUodJr=SOj3i5mU!R3%pQ~Ba+QYPAho%p!-|`x?_tsnd{c45V>d+eL?q{J~RY){@PS--0Gfn`a0O_jY4~ zjm({r?_I8!l#i4eg3GX@M%bt$S(zgb{vc$HaQDK9mI9bFhY``M+3wB8DuQ>0exnRi z|4JGLVTvLsW0jd-4a2Vw-q^bwBp>06^Qcm2Yn-&W)4)>2a z1|wf|7WA)Xc%9&jWP3%4k}Hp{&@}@I_)3)hVaR5k02_)$CLkR0EIk9VnDd-Np%4B+ z@I!j|p$6Q~-wg21#1+9UUn`g*g+nQbi|OrlVb37g>5MwJ)w~>ib8xIC}uoK`=uMD3Hc88U9&c zXk$%tWOj!C2rnyRJO~p6>LtGvHYKUKb?2_K(*k6=r<3(UsQJx+`2&?wm2onB(BIO< zGnoohkEoI*iJ2UANywnsLo8m}@hlR{C5iV|Zqq!D#hQI4?c@{p-ZC2i% zaksF>%DSo%U-^;xMW|zQ2^=9Vkv|J&!d9ELK3q%}@#NvcW5)Sm_%+yp@*@hz$tlff z(E?lQydKLM)QWxgj0`G zMt<)$AN8?f=R|lT6I`n1-b;SD%=oTXNdlr=|I}gJ_IzJIM&jpW=QI#1?u}fax_LS| zx_P)bO-}DLUObRKe+}L3Jw=Sf&a7=bSU(>eojpN+oPB=Cy7}I(Si8M12LRu$t1)^iq%t!C7mj8$%3S&@PE{xGW9DFk`f4yO|lUcz3rzXPt48jlS50DR?6i9 z0VdmCaM{STj=%)Y+2N|*Bpp^tXW>jUO&-UY1~o$tDc*SYDLA7E`nW{m>_m+IsB?xS za)Jn=^}8lZuz6J=JHJ=%#DKtrkkS1>*c^BRHzB=crookF3sMSlOOMmjYx0+ zv^GzrOb|gh!ZE}n9CSMgQt=!st#bNQTtG^UZwWJs8sz%6#tz}AQl{_NXyMA@hEIeD z{3yfohhy+SMeaZqfNmmRaotdd=fu+Xcu#3z^gsnjYl{txCV0~@_)cj7Q?oVm*FYH> zL1ZMFaG?9K;xvY+)@q`>w2QNah?<<2rbIUxgia;Bm`4Tv;X!_Gl17(189SzIvGzI+ z=MPH%bdoPO6qebbd;Wa~%`+VOHwQnhd4Q?OBxR_ezh=E30sew*Ns^_;M4THQ-%HnH z+qu5rz?)nb0;5D(&ue~w`|ugUKUU)bRKBgC3%5H2pR$ysf8V+WCIQ>i%PSCI>53V& z{9mn4=Or@!5A=-~qD^#v3A^7RmWt~1($mq7G;iq?m2s{+;FN(u71hB5i_c-PaG}Eu zv(I8n1IvWcCcW2i`^Jb!YD;iSZjt=B;<&Ovnw0Qnsie6_%@nf$)4ElO|CD!o4CkeG z+`5vevRQ%xYij^UV-ADY`QjCrP4;oTZ(!NQMl=iJL-JkgO;R8vOBkFeaBgrt)(R>! zPGqk6xa6hApAF(G%OGgp+*xc2mInKq^-n^rS6^>#Mg%dy)!yLne#Zqp*; z589W2&ZSFXKA1eZOMfneZHe6g&TB*)!$kAsTQ7~CgC4HUq0XLq5r<{DE%Z#Upcr88rW7*DkRFFA>&iV7%oGUJ8$kio~uQ>Jc*Yy4}nuWVB)kC zfm2eDLAJEce_sImYuLFq9}2Dul3p0Mx}$hlzyMsrcOYt1kZ<_zhs`Slm{5N0rJiFm zg`QiwX+?3;zwm-O-M@d1aGALMB%rl~ebg#|u5T?c2;MH1JKpVM>g{d}oA>ALYK&Y- zQ0kyKN<_mu^@D(y<)p>q*+pG=H=BhoUrkn~R&QV9j0^zvJ5MTiNuFmU9@4kiTREGa z3D-8dbq#n;c*}4)k+V0{qv4`sDl>`0epjDNfyzjaH+40K(`mD$pC z1QQ3f+Q>ouVG>(X4oW0?nD7a{c+{K2pZ^OAP8r(eQ{zZP`nCY}8@luoHkZmNQ*>dT=RW8MU$1g&@yHZ>sJI zgq^^TrX553v(7=e{uvXX?-PG$Y|}0o3TNB|?SZZQmeYYGin(l$<;M9f zgf;--#YC`f;*aoByCSI>e%dGajF~@z6woBv7O(vy=&@QZPyfWU?0w3iHu;cK>Y$~1 z=lSkY{fE}p=to9af14Klfep~Zcpsa(rtJgYP3uwxt^?(<;$O&vq)0O)vGvS09c%UU zs>LH?lLox#75PQu@|cza!3x9xGi2BF*keGk(y*Ipene)WZ;r@9i8{e_ifWz)N%)+t z3Q#-d@OsfRi+4B6+9oaGg8{h~4Xdzn6!uZXW3g;H>P+tl_>oUV)e7?<YbkQ{;JxTdq5 z!u8<6s1&AC^!X^l+sJAE{2+<;IRf zqP{)JyHYukZM8gfx_sYXX$camwQ&BUg1(uZ)qULSQglc zW%PbiH>jSwA@j+jfMf(*7j(G=S}8ST{l>bNhv{<|Q+rQIn{?zMS4x=i5(n%RL2@i0 z(3Dy;R)WSig|*{KA+eWSm}3(j5p$Xu3!~MCme?S}q6)3IL=X5pcVNR$T1$_1{$-|o z&~jg9D@E%s6rDQrW1(!>tO@l~A~L(%h1LfF$2QG6*jUbpMRD*6AfibU_|qM8+H)DV z^W9hL$rHet+3Nc|RCf)cF9EQJe82kGP`AycJVFdIoJ62$_dI0VC(P1`678`mrPlYA)?-r2L&f)4a?-Yu_u1toS=oa_!c%JnAHEYSThoTK-eG zn2$dg0W?To&i2w=%{jVbpl5Ai!&I5&pnSJW)G2BX4C=de7xR=x}@T4^la!Gy*+mOp9I>M|Q@{`7iD79^*aJ|^;XXot_tX@O_#vH9 zNgPNm!|(?S(KLVcS#j%^**5cS8=)iIFb!-(Q52eN%iCgEO+^)8%C+9|%MI9b;L^mG zOI9GX;yXN$6LC&G>~us8s2i%Aism)@!Dp8#X$FaWBW#U98Xe){HJ0RBhnMWL;4o*V zyl2XBCYb$hW(AOLF%79iUd(gl$ysML2{OvkM4tp)@2FSO*d=UQEc>JV6vcA>`WA%MFOrag+;S z%>CK`(u`9T5b|@noSxT|q^5t*Ypa>q`N8!(rcoTt_yPcI_juJ=Z^{S%vkiqNv64BwHDs74do<0-5Yreyemo~pU!dUb!{ z4{x=sCMRI-Y+Dpw_Pg^sc}TJe@43dCjz+l`#pVJtgf5P#9A| z;*{BF7#Uh8K)Mcs1*_3ci`9jOvXds%)g;7tPX$2!2240Fr;;-WP8Iq@mWFxRTB&17lJ{xN%f{-ssb9}TJ4-rfSUML!8Y`BsTI{0EK@&tz%;B-u0J4BZor=ayP^$%&?+>6(=Ix7c4U^hp7QGC4BQu-7SnH(6d}GEb97Ot#|&DaMFmvNPXtRHW$octp)YP z>8yU8)R1!YrSn3hc|m5Co`)>m=cN2rtm~asxyFDaWU4Uk+KURXx`$`gbnmApl^;$Q zO{g_S7P@O1{PW<{`mhr4xA*Lj(mio+0)K~1OMSRhschCbLoJ>heAwoVcAx!fyT`3@IZOi$eo zYZ@INj<+bWg3yXlOcXc<&ro~$Y>FU5v-X#a4oFRa2cWvDj~bK_tVCD8ujZAV!n{&1 zF2oi0%wBjvLB0Wm9%q}Ta@Ufgs!*D-MA6!_N@Zcif1Ge8had(?9^%C8NC5F_gmbIU z_pA+w6>n>E=%SEsLw&=uu=Adw@zw028|2gnE#Gaa4#{mdQx!DoaNc1%&4MKn7yJq2 zCdNtTKU$Xad6&<&T4C7^fv_#U30Q`krRQ|b0&Lx_j{P=v`RiO(+HP6^+o>>sqxV)r zjyEu4(!2qL!mmP;{IED{Zn;=ncGFk?dn+!cy>kHXW%c+vo-k#WBF7`x&J5}5)=NT4 zV^~dqv<2!o)=HCNL5(=fd%-87AttpL9WeY^N$pvHUm$z0t$2ZihONXu7EpZJr@o3i z09cQmbq5FA4wgJH^!ygSMG}s~-fkKGbOi>ceKgJc`6VmpsZhe_R&G|NIEDBdJe?cl`NsPyXu+o{g#?|&n$oti zChH0jvq<)(q z6O)$hbDq=meE3VS>m9%~O(m|o`CaRWa`125Kn*UT0t;6i4u~~;3S+rFmhaOFWt?4t zByNxaFS5+B9LeVaxllbTpX==N`3{;4SnhIkM9es@7GzJCm*r6{b{J1M+!0gIEO^n$kqVQx3X zM&|_ZW`jvw&lMC<9D)V%B`twXu`T=*jt}=-J?pz!*Pq>odVJ9@udZRgQ*Lg0{ObW- zoZBaie?Os~{!u;O_iisAU#HGCe!IE&-7OrQwe9F>2x=UiDzXQ$R`@Q}0$$8x>6R~F zevEl%+}-}%9>|XT6M*`};fEGH>o*KTKztIEKQ`Vk#ZT;uVw#spQrg!k2*h}sBY`}@ z1=Bwi;C9;lt>0W4x-9T1keikFp{Vi6uZ$mH&aOa1wX@Z_xd?T#M@X(i0zaDp-80! z`v~+_QYzd69d_@9B5^T8{vAAd$0a%r!LKH~ZjCItiUvMEGlpNpV~w!&M|)U1A;&ut zF{dvg*qve!V2TdcxVn|dxele&P3kz(fi7lAjPFa|{`~?Sd^LdFE2mE0Q!%F zJ8k#4443?WNss@+?B7nNCYCM+PL|Hr|B+_@C-?C=sl`~01qf*V0|-dszlGTu(Wf;3 zHleyeVg8RM*r4v0)5d7RkDwwmrP_F6Z1JaO)`GTzrK#r(`kK?)J@-FBC_)5r1Qb|3 zmJo_Pt(s@}$B90pqEpZHTdzOZV2gW8?VvVv`q|mp_1R6|3;;<7HHFOULP+;hI6Uue zh7UV+LKm=iwk(37q3o7q_w>3YlERW$GFX-OV?=OPO+zs*(z~;!R7c%)k zmOIKa{i}ddtAL_0e2^qc)yXm-vgW;wu#zi+i4jE^$m9lr%6p%J;iNE|_HKtIqKK23 zB5(LW1VfftKT>AoOzu;B+AP8=e`u&ehSbX`TNQ9H!~4)U`PBFQJvBA88#w|;v1Xjt zTCP^{Lh4Gd!v9Z)xTeUG@q8x!b2kvSvo}&y#%HHt(J_ijeo3ZmW^B_+>@sPuJxADl z^>eBa8KDVE)gWP_QXmUAPd-fX>{_Bmwy!D&gFZw)Q(6N;Y(rXMQjxB3{3NfQ{J^EA zjS8seqZPiy${*HJHsP$C73URJUZ#m=IBrQDb6(t1;ul zdaVO|*Uc`OFciC`PcgI6rt%ZPI#uN=<*T%s_${>3Z2cx4^|vF2NkPJuVUAJBP=Jyy zd=hF_u4kME-rDEUkggR6NnP;6xQEgGl?2cXYRy07s*-VJ)#!o>2TP8+#Az-e|EZ-Y zA<7MdBW`YP?*lIQk$}n4_Yam+d(Yui1k8$U>-}nI9-`K#)q_D&vE0rg7e`^-{}twa z7FA@2JS$S-1ef8I1+opBF`j{v|C6G$V96IDJYo`2GlUhc*My{5jdLIJr62`V`U?Om zC}Az^LJKS9%K*DSf_8X2xmfdS0_}ue1i@H|DTFPi=`s?dJ(hHx`L*)?+heWlCsHe< z3=W4&T%wDlWptDH+^Dp1mHQIbbD+*USU)>cTH+QO=~TvP9W{L#MCa%eF8oNXA_A8g z+T&A@ApgB0vM)r)(d763zxiD)vlHM5HZ)55Efa8OqaHM?_KO9RFTM{$`H0o8KJ-&Q z`MeK!TbRAXMdssyE7BiIjPvUE1*{fxKqbdTqfxTP;Q>>|)7e1zzj0YDK{=*LX zIS+;1@B6KCLED29Y6XEA+jW2b1B~kgjodl$S-89xRoPdmin_>U*ki3ofgtr=PI{$}= zvfJO7tnN1|E_&K;o7hkPk5GuRH*h-=A! zJ{eRw2^$?N4!o&%CeYQ@`w9?aT1Ll{8x>D-@C@bSJ4bIdN7!@RZ3 zIj#o7(chpj@(?mSa-LD|(JARsd8oWI6YOUcqFsw&IE&Q8yU6<}e#Swrky;hC98C); zTCtBEIMLd(wgXmVZ*gLsjIxKs#sYoG@x?57`nrD4;??%EfmxIlw(SnKPQc>7>lrH6jF6-f$tS&9m`@Lat$~?f>Y7+PE1rgQH9VIY4WtIn5{A7B z1+|i;k~w>?z4{cm94xW~Sw#8|V#>6qSqz=6OZ7Nz2z;WQKYaRnc|23Y;b^3rOc+R$ zuu1EX6+pECMgAQMWfF3Z4KaJddRlN0b=D9VXA}fy>vGehuHcerxrR%kRBr(kYMbv5 z@&XO}b!vM6310Y$3Ph}hMlpq(WBn;}q5dqrgit_yJomLSuFAW=v5*6Bb_7tzu5O@Y ztf;)AR9?A^2~Sn;v^tHpiz?3;4NOG{4UCem7$9ZeP>=(end=vMWWQa}&`S3Uf%8s? zh#em(^Z3DLH$J!se4D9j8t5HOQ)F%q5V8D9p?V1O9C zJ2E3pr7p(bZf>sjl`8|3H;aU zP&G4G25T8y_Bm3?-)25UnhHh)-QKohKp^W3y)#9x+H#|Pyj`1#5j5e4uWN*qYlq|C z1Z4}CPOy~L8B}n%6Jtx&{=Jg*8AQVYtKWOrUpI3DHzB?(3QVWoL+YH+oaUrspc^^H0#o zpB){fTcv%Otg|#7#@vhSC4b>%Ba5_>mjg@}ncZP03z#gBN`;uDTF`&A(9VG&JRkpt zaM>aUFGKa)WVF8Bw{I*+eGAL5`}#%?&1p>l`Kg zTwPdJ*ZlgiK`x))^H#)Gm!fyQ+g&?uGWoYK+4Y9FXQQ-ea71Yz@b^urZ zF|6isq*I7)7Lb`^px~(xCX_Yt90%#xkNG`7?$n7(?qT8dkc0XnvlZn9R^nM$%i zVB!!vD8*R$^!VB^A|^-}*(H64^$lLic8SRx^Fe78sOAca7M^MT@ytP191#}aGGXoD zz3eYl>Dx@ev!m^;o5WJeD9ayHES7EbH~q)l3$tb;4s>1lJK7gHPg*)N_VeI?3}_{K=hH%aVB~o}G>EN?stc_UVu&qJ zE*c%t_BRvGMU5vDj84FRI8oR*l?&So((JWQF#sg{+W>wV0ufx6{h(hEyCI#XFE0!t zyeINEqGt#bHiVg`qha1spZMBUTrSqU>(;2y>CKQG=ZBywceNYi-AN(3N-1kf6p_Zb zV-v>1Ws_D&-pQjY4vP-8ey1{-BBd%tY&fM`18}pbtF2mMvAJZR@q+0F>G_Fi77X)L_&;XW78_osV@`!kCei={P8$g#`W8f+us1>|S ze9XUmg(5K{O^uswDzOx3xpt%9(rD6r*HbepU#!C=PVhc@42%L0*(^9+hJ=7@&sgZC z;IJ4SaBEHX5GL#TEx=QGeZ+R-fqIT=2k;vm$@Z7k^lpH`vDt5=P*`n+9p3th&OZ(Z5$OQF+S;2(|9(UkmTZ*1dBD94h*iN#N6AZyACJc zEJarqL3tbQpCd$wl<$JjjAk!#TQO_R@-av+u{#vs_EF2~QLS@jl|DkywSnt(n`CS0 z!y|Dl@5{6Zzu0POPf5k5(C(s*Q~_+Jw~R=|O+8s5t~HFZy+C4ur6BV%_a*$?@9D~h z(&{Qy_`SHYhKc=rGwlnEsE17kq7OvsI~%MGNd4!ud_OOlKK9YFTb-jx0bF}{&4@F$ z*P@*Hj$3!&sfIIAaJ6G2p=jS1fU2N9TwpH#*8DkY= z6(wyg_{dL^h#NE_Q9%(HlBM&R9=p3a%^4LHhStj3C~t%h%17@7@N3j#`EkR4Po$_J zuu2wE?UEDRR>fP_V2*q?1>_x5QruqB$XrBFEcK;=ECan^t^&{*<Q*%n#roZ)$*gwwM=ivMuM@xN!2&CD(g9ySutw#zWV24g>cU*6qmaM@TDvlwL|E?UbuwDsf*;L9! z2RE`(dg*)Brfv|;oB{k*&Fo(Mw%D3OETNM@ufgg#pfl)t1h1l(fiXGZI3+!aGpwkGd|5$`XocRl(_zdgt0k$lG zz8M!A#Khcnv=a>uZxTc&Il5+lv1%8!P)A{wmHe6D!AizSS_ly7cfB3vLkb#R*u9Nx z%JgCH;KhYbfoZ*PSBdjLkL}dTkvdv3~<~x+)H`=KV z%RrSSebcaIa~GLuG{N&UFChY<0-HZ~h7mdO+b#CF18pIQ_i&nwmz(*eGBl|er!9Av zs0O3#5N)doN??mDFl%GwPtlmQno>|IQ&1{B z{gHGXRQ;<9KW&NuhY!IfNGHttC)iRq^VeTSP~BG^A$rZ@B_e}+eo-r3Qss@8e;W`j zSNWaMhPd6uRQ8R2Xa~)fDrgNI78`wYeDE&R2~_~iWRe1xL`S2hVr?I5t*BB5e35Pe zL&oUc%k+k@G@UgUnek7jD)tE!c}QpSB5%yOn6v%mDAAg$Gzu(BeZfQCD9&;A4gaCi z?raZ<+U~3N5X~vHGgT|b!|1;k=0rK`ez5H=dtXWSUqeL@pjOQi&6zi7cKN(a8iu}{ zVM74zIp2r;T*FCP1&o-@aZ#(iiioM+7y$wcJvuHC5-DtgdLr! zDw3Jz`8}MQC6c08QqYkVHUjv?~{r;Cx^R1B%S|ns4kpWSEbfMoDV$3}46U9Qft=n7+%Z6<(Y??z9~CFbkFcb^VSLW%-(y}1lO7@cJ-6mScQ=Lg(M7|i(23S$CGo_B#~uJA zH@@P!cC+RFvqZZe^E^84yYP|fFl@*R%MhPOC7L&uaf|PC8F9)kMsW@Lg=3zibTg-) zu=Lm;4n;o?bHPWgQp;>6&N}?a`AK@Hi5n9_Vf0i5!yPi-{mQp5Am4PdCE>O(EO9gc zRKT^~iGMpD$ZV-OU;ZtVxNT@n2N_U|un7-&MukYAWKQUf-io=6{##ZnUdFwyLkGn} zj%H!W{JI3R(b_1|-vT*~wg{@sIMsAfLT}z2;T$)PHfWdy>%ciM)n&7sBUWXh*v!&G z%|~VY@&qA}LE*%1rDik)W@xG%$3cb7)qFj)rhS(rR&ti48^`LgqYDh}svCfuFNp_Q zx(VutVuofZ(Ea#YIcflVB!LAQN?JxzW#+t6a3%<8$ik)2PplMwtX+)9^e_bKdQt1| zP(F8<|M0$2zM}AJ#i+Tk$U0N_N2elyJ{1+i!Yt1q{?X5)VD(5BzU{RLZTZ0A#b2IZ zE=`|k>TA>>5voVATTqSu{Vzbiht0*;5EQHTrr?8f@Z0z((!FfHu1>c;H=87Mx&Ltl zhjV+AwNT==;_Qc_q`Pt^i}1zY{>0wTfwMBp?60kn2`KEx?jIH9PL(ZIn3ivvQ26UL z=aSsE9ONF*J4t@#eFTip&Od>e25?rl>I1LIq}!c$Dd&poXxWY^e1!n>pTMsqCIZxl zZY%`1dpZUABC0&-3mNSvvOV~%d(Vy*cZG18dg2}R$+jWL^^_Gs_5?h1@BVp*CnhFt zH_gc_EchUR6sPRE$)(#U67Y~Pfs(6x8SuhsT|opCvflRa`>isNT?fQrq4)O*0vp;% zBF%@0ea`bt9 z*v{3-y`k8IzFelj<_}L(vd;--&NIC>s${vC~t6 zj?)_}l;EdeRV$$u(-{8wTF7dB5h1&B3!~!vpSz&b0HIZU{aR!*VgYl13!XiBPG0Cn zjVI4GoMO3Wa#8zpG&5(nRjc&jAf-wzh6wt>qWA5xPyt}^$Nqplaeb5IeBv_GsnD7^ zum@|wV*32^HL1r@;p}X;%KPcE4D@dM`Ag};`*sSEhJA-M&rf*eHMGvSMIKH-tq)U; z=7$~0t2rO9=8Pd$w6UhojjYM-qVl@*rVHd0Z%cu&m?3JF$a1e~iOgmL6Uh^Ee;A1Aky4mey|*z+*wmLT{X0 zb65@EHbdl-dL0#4s%Cp7f(BXmBtH9eVNoQ}-XHbBD9!^DzN&Z8P!~Ay@u)|fH^|qv zve>S8^t18jy{0vR{O5;HoYDDu-J#)iQZ6*YJPiOO?ZbeCA=T(kJd#Skb}EtD{0`LR z>@@8g{$);RWbm8tNMLMPvTZTZ6xRVSI&2J1)TTnPva+hGvg#r%qMJp5s_55dSSNMG zwMQxebU|USHOQGf2anmfm5|4`JToj(bAP4OMSM-H))x`pZ9N>4$sX=TB4L6oY)?2{ zq7Wd)qz{bmz5t5goo$pLR$7S**?HkUiYT$lkRuI`@FV=MbaOIlZPY&6!*JJUvz;rf zKpe`?`%&VuKJkETcy{yhh{qLb7Z_G6LEE2`Q)l9yIaKUxqtu79^_N`85j%-kL2Jj< zgH&s}&WG}a-o?>?3Ch0%tJ1`qzfY@$hNA)51h1i>c@6MRd4JSA53Fo((#|COQg?sZ zSsc=_Z(gs}fF9{O$zmyzb!e+;%lOOiPnby2AL-C;_l_N-ffD40Dstjdj=u;LWj)8$ zu05?9`Zm8d8 zQwy;L+S#~NzN?;YP1sc^FCiQ7{9cDv_4WtbIqizxa#vr>f4P&XaPneR|hm!ZUfLjX#g7(($#)#Ds>zjjL{tJ|4 zBH9VeW4Uu7#u8Z9sl1tIW~6BHZ(0>#b2JXf6{zQiXFH8b5>w``gVsC2c@{s<8g8vJ z&Xy?5?3%~`ePBuLhMXT+SfU=Fy7cSaJ? zN*=-gr0`+E$u2k+f2e&PYUS&dV!)8qXk*Kql||Q`u{34Iv~^=<@DET?R8$(EyTA#w z@se>I$FZpp--|A3A*O>5n!?55&phxn6dTh0B;;QePDpB0oR)&kZWDpNY`nnH2(;;5 zUOAaJTjEMRjF!JTtnXfg4WlLS#2dJl6o(ky;%(z1D(}#091~<0X(pp$xW`U!R#~fZ zevM5WZ4_eF0JN9N2tkre`Tz^S__h+glZ%WeJsvFB#=*h0(eIyf$F7}V%&ZScbMLa4 zq=(^;gQnx{kKfhyTYOyGzP=vyJL6lcGAfn%4a*X;5-X$poePFsZK1gurn8S0hf3&0 z%3x#tUZtlU_5NFEE}?kzZmXSoop#rqEUny#Z4*8*8Ga^G`}(>JTRG%_G}*i}iM`O= z;GC~gsj#@&yvHog8^t)c8SqcV>>ggC!t<|7G=vTfb$w;6lHX1Czipu@4Mfh+DU}mE zWuQIrA!|0Qd5dNB43D=>(qd-49!pvx)T#rK(CD=fvoi&fQv+l(7`3TCayLdzcH8EO z=)bu%6rI1T<+-%$>_jR7LJ>z0Q~DMV8jVW!>ux*Ox*55sPa5ckz>%n_uEAeWuxbyd z+<)<__*Szq{RMGYWC>AcSU2eeFDVWERkp;t%ggg|`u?h)HgG8xsfLt$((q(rJpX>S zf%ZQVgYk-xWOaPk%q2#khBnamXAWAlqiQ@EKhx_-M@@tHd_#c;#DiCH^Q33vc0zrb zxubF#P|}{SYLda301gKQ#5_KTY4a_&(`nM486IZGYZ~2?>U?-2jV`?dkvx)5bZtUr zTY4+wUtWaNb-S9BNPxC}3DBsE;ZPgWBlpDTY909!rixJ&?2}|&o5hqajl-xf*?J&1 zlCEoMWwg^QhKSw*Ol=qF`K9^^)Ac^b_uRxU2N}-&@$X=!0%!mBSE;SI*#&x_j|{Mu z^d8gc*5IAn@8ml!95`URD_J1dX~d)_ z(*9z8nl8CThjUc757{T%_V<6>tj6m|W)(afG?P90N_+4HxC43uaJ*lX*kWg<6$LLF zzNcS&`&(Xu$ykrve%abCS5ykYWP<-9((vTC_leCACHFBcnc=wfHxyQqp%z>pW`e%6 zCwtl{xAB^eoT%tXMwgL4GpNilDWl?{8r#GQ^=5;?RCB%?f=Iy%Mqp~u-J%3hQ(FdZ zN9E7Dpa@b1M5V&>8V#8XD(kMKQ6Zjkzy56reSvQ7oE=Kre&fUwOQjLlR7hjs7#8h5 zG~l1rB1I<~Rbw3VHeYFtLaz<50Z%jviL*9ojVkpdns~O-Ai9sU$2p7vt_nCRuua64 zgGHL+Q^QYu&8uu=*|)(q~l`5tk*(LS}I)=%GKI( z?3B0xkQPO2s_rH%6$PBwix}}jPyly?#kf-b%bz>t3L#XI;l@Tu>walbK@|#b=%fJ$ zS=y6g4`219u8Lg+*p*-Nee2@RB0$u_A&Q%Lle+VMYC@IIMNG_ zsw(TdV>pU3U}F^`>U@ER5l)K6ca;-# zksE`ie%g7+b3#0uJB}SSM?hPDt(2UEtKYeC2s7K3Dz$L6&8y(`or|t`K8$c1sN8}E85}EV=Fw4x;NYkAjQSxs{d<2x2pL1{?N1w z04P@WfJT`iMv?i$!wp}4#uLzu6b{zarSfZb@-4^2%!e(6g%iG-toI(x7v4QzM)`t5 z{P5LKKDhO*EI`WW2|U>!N1T&#CWse(cia$WZ0`zQ&gSvr!iWKaE46pD*#y|WuO9i; zMOz2Lh4_xc;JN2}(lf6Y-X;Wh3KG=-z&f85(jpzW={>GYIl6FV8I0LF9@Fec&9|VU z2b1Zsu$W1se^z!hpz^w7jRFMJ=l0z5F+T}@9cmR5ij8$wKQ3Fdl7&~9%WS75dW0RK zLA{-#Wfv2fH0WjchvILZaZHM0v*#o{I3*@D&4@oxUsyZJ;n8L-Va-`Six4dW$nH93 z=dm=zQ=>Givm2Rk2kJ{6s1AMJWonRC{A_3xe%S!>9ojIcygGHY*h-@tWC^pRcg&l< zk7`X_uycVdo)7b?V~EUA6G0JrN#ydK*6@Vo_igws?q ztI?UIPi@4AY#&y4&1g+QAYcF~H#w~qYv05W$HDaQMxQyAq*&i$QZ;lxsb(r|2gnqL zmg7t2&t0i!w;OQsDjlY=M%gowKNnr$Vk)Z+VqK`JFdY5uleBN8s@SLi+OLLj4;F?o zCf?tyu~_r#9vfVCK1to*3^0eD$RR8e5~dJ}&E9M?i$4O3f@!$?+do^9CP?3Yl4|-H zPXyYA{psrzDm=`J0-A3280SYbgaL7+kvV)#j$pVueXsqK>0F9kM$!9sK7S@?#vL^U zvhH8@vI}pd_>%jwrvYGD5`t%-74CwUIYu?vo~6$KGH_LtqCA0amut;3 zZW>C(b&YqpMc2`Lr@7_R&tP@;qml zn*N!gJRI!yp+39;(bfFAizZGaI?G*(f6>CVLuS*o#dRfocE?*au|f@c*NGTcewsaS ztM-|F7LOzOW}Z)~Ew8+{$Yh$}a;5p5RR9{5yPg#6-|b~NKM(ztq+@NIxHH#lD#o~6 zVEB;i8`QR$4-H9jbz7xuSd9yCZXdGsm+#$rXKb zj4J-rq6$V&z96Iua|A zo1$vXO!eJz$q8qFx#fLT^z=mgHgi7YG__3soO&L$*M|crD`3@xl27LUMhG3eYp}nt zm#c1`#to+foY4)4D7wXDuNn#Np$xK|^$5EToe=XDTFCMFPLWCDUHPqg$8Jn3l+SEm z<#v_Oe5*h`xEFJNW0igPtQJ7BrqGOp!uJju*BpR!I9hDTuhJN903{~(baJa&Uwl|^ zL=1_|yw2+Hc6i^5#bvcT;5jw$npya+pf-Q3;V_8+uI4>~Uk}HuV~iO%9S>X1v_5=x z2&_Tnavti^@2@}!3mHnyv%Y0;OQ$KWVP_%thd91pJ$>&uc3j+hEzQl#gY;I;ka4Cy zS-uj~&)GVC1C24g;Z((hcueC?cSwDr{vT&|8BAvTUb1zh{nNCaQ z8-le*-A6kjNnPtON|-|#&4i3ZIXuSjiu4`6pl=ij3T3V!wwcgUDm4kpIwYudRBeP9 zKT9S$xscdYIpx|Fs=vN%-<$bk3kf<(bx!?8L)wN z=1mx|A3eC2hQB*8e&teToWMP0eAAvG`ZXaj))uQ0(s4q1`xdo%LBfNeLO6a*b0Z%K z>#ci{H*IfQGx9i9_rV=C(=_DKeoDQ}1)Itqo)bC*{^kGq_}FVUEdAkLk3<*&BV2u0 z;hvaB*l$PVH@P=78KsF%G~#Qs2Rm)A5tVanlJdXHmF7I0vacv24JncuMe}GM@%i0$ zmVPEOpkYX+M1u|x5A6tnhSpG<(Y|IlY$n*QZmjn!FM|EpAWAV3tzQYj`-Weah%Cw@ zr)e^pmqhNYz*!xerg>zI^J7%;z6T0gqn#>=;LaEdS8^{|4sQM|t%|Y@Hdx#`1MRm$ z{LLfx-Kf8&q+aN%9hn$U2+R~uL00CRR>z)jx$d4b{x!2}<;&9pr|3l)O$$<`*q!Vx z-kc%gKFYSmamZ4_Z~HJvZNiqf8e)(j)y?+aTl`H;Pb|^Us~@!q&e!a|qnPv3AEbUJ zzj{w&GE=KJ`mcj?cxC;JgqKY@Qi~};Yj55S@0MaS)A*eBCn?)eV{AD|xQ$Xy$6P41 z0t~6n#}zqK2P^c->9326()8Oh@CmdqqfmiW7>2}W6%!NOvk00|kOM*<4oVeiX9rww#(mnuvYson6&CWKQ;uRr-YMkQhL$;4Z6H@V1Mq(<;z1Kyc7?> zhM_HOZPcC1F{paaCdIeybH~15k50-?G&qXuW!8#hLcki5k7VW>!FvWzeB!r!#o3XqrRBQ#ZYa($Zg6AJMT-!Bp91+R{`-=< z81Jfi-NisO3VVgf39*T)SekQ>LV0oSj=>X~PFRi!#q@|nU44AyajY#yig9Pr3fsJb zQ0J^hyJrC$@4GL(DW8fDukwa+oZDS}yJd;J*&L{Yf%p&*Og|3<$}yk>TFrJ$Pb8vSZ%^FqeZwp zpXqs8h_d1rmU%N}v|dm1GSzEn#*(pOYynB}bbd(P zn)~hYnU2|w=m2&o@kBC}2CT$D#6HE45h+6-V z@U`2dTU*R&_6ME!R?y7FHcmJR5;23UDN(JJRvwe66OC@Q=&rwCyD?v!L?2@eb5sW` ze+>Tqz1y0MSpve1*`(yu5{lfMiu=neQJBeVwi_`_6h2yHCYWaCiu)rORS)Q41 z=*~Y^t^_+y_$btUI`rwU9>>U>6l5cNRi`u-0#lM~8wk_c9)55&(=q(_Wz6qNH9Mt? z6-EP7AaLvlRB;t(=Xh`|;cz5HUla1NKEGK;tGV@`H+6-?cJqf>8_yvcpolPL_YSME z3as0jEsML;pmcA2dxvo?R+#0!J|7)_Xzzu9a!Vw^&8%b|&3Z{fQ2*Oj#w})(9(o|h z<=X5BPCApKR~H*Comh3Fr1Y)aT0M5Y9=waezG&M@BGEXvUc#XC#z!31%qO3{5+`Hi^e7zkJD+@q|fh)A8-cxa8v#*KN61X{V|r4RZ>ha!a+Ye8pB@XZj^r5U`=vruaUYmIxl55RB**yTvw4(U zN>@}I9gbI2MD+J5EBcWX{fH2v53+7-*D4A7W&$BwKKm<5QlQ&N`Z)UMK${G4dg2#J z%xiA$%TbScALNQN9{ca9nFAy^xQ$UH6Ad~S0j~IeS9dgEt!A+hn*ue8+S{i5M_7Zt zb~V$(1nbxCbz9geMY8TLW*#JUnuQxJLDO+Pl@aFYh$Pi z`tW;wjyYB}VjcR(Ap(er!Q$}OeEqxg0&G8wL((_h;oNlWq+|J6ko%VmV3rVy`XbucWQ<@JC zZX`Z>)VaH>>_lJ`TyI9lR^o9DOhiy?7w1YKVVkqeb@YS^L^tJ36yz1^M<^q=m%(C!>%4yG3SIY{}(q-8@3cR`lcQdF#^-j^h_tLgTW=rMns0YwfRC z2XVLO*Qk$mHBaTAbL)-0_H?09QLgUd79mJZX6U{A{qct!O&E<$bGkeWCxdw|mACe? z?MP1Nm&*noRxsM0{S574&+q#nX4UU?dN2~6Nis*felVD(s^FX-rYUwEd%8ZrSZJ!0^h!HhiC%o=gOMJEZ6c&f zsbEJ+bY-k3u(g*=_19pGu7mg=4Vml$X*)&L?vbeA5}|jD$Wpej+Y-X`YBu&%K4l{l zUk&CVC|LnpQ~KX55PH`)j3tzFgBSL03ov~x!pTBsC<9&j1^(Q^ys1w)TZ2C6EqA4~ z%(YVNKok<6;T80UuxG2@7%C4kEtoKDFC9G6;a-mE&3^guO|2%qptwYV1&6nKOy zrZ|?**=J5`-$1MD*1r2`A8E8?SJI+uBmS{I@}YP|k&7fxHS%>}sQjqP0P<%<`}Ba`B!3~?7Jf3ctpt)_jh3JA*m~ecdb(hu`KP?x~@8!O`P8BC)Smq(%SnB|Qfpp$2AV9f z-8}lQV+h6VcN$obSSdY820a?-2DS0$@D~S5}-8Y2WzDjy}hka7zpB;8x-_Y+>i0-KcHX#R>738wc z6rD@Gn1wE8Xr=NtWEZ3KQ%DK~Ann zk^V2+YMrSMlzl$4qKXny`u_FNpR4_o+)gZ1F{QQfA zlegR%5fg6`ozFfd*MRyoB0Z**UAvxn;MD zv>El%C^Rl#y+0i36OBrqT9&x}db^6n4l=o(n#KDp7Br;d%@S^PI7WXzTp7rSI3i12 zO^tdw)OVgYbI4zr+SS!#rVh)+rKiyAX}sSQ(AE$Vt?0=JDp^dE56gMt;!m4?H6mIP;dHt=N! zom|KV#b&BSJi@7$ISTQ+uX?wT0Jp43H51rWE!`m~!vIT0C0y6nYCn;oki*DAEdtm* z?>Od^;kMc%u*h`SoB~+S-l_H!hC@jPvG|vaY-3i))m>^9@RD~s;KSa*;+SV-LWfp| z3r0q}+eNMV;6#L?SR&f~P(c|KNe(<>-DR6Q+^<65zsF6_2n*tLpn-V}DZ}yW@okfX z5+UxpwbhdfF~otMi? z(LhEf4nrlluikMyW?Ae^)7A{(n?D-UeIGS0_~$t6oJv7&_OWBtwf&8jGhG@5Rnq2LUl3b(9#Mi@)q0JA&9`Otp znkukzOFUkNW#mx1XFub3pN?3(iG}Ep3C)G9F*M`8B!R@CpI0kK?@1~ex!J|XOX@a^xS${s)W3Ma~&w5P`3bl<3VvY~$( z6>wpB3hFj*P&di36h-0*OBN0E@_S9#*^iNfa zy#)?rXEbMh4(*B@pV0Aap0>-ycrv%%N#Gv5sf?U~Tb-T#C%H-|7fv3Uda0s%Ujt$3 zgt>Ibk`aF=f2+1{>j6coEpJy2$@4%{_#gArjXk+` z!H{@ojQ*vb{ppjno=$!N0z$@5TRZztCj%0SN4XK(NvP)sM`v5&zL1LxB_p!OOGz>c z!e0S5r#(h@3Mj|NY_-ZYx7WJm7Y-0Yw{FQV+;1;Fwsg~!FsgizG*QI~Gll!wBWulp zgQ-cK(IuUfIc7#YM9q(eOyvExp)I^04}W3g{enLx1Lmz3&jh^>njH?tbR|o<*b$A5 zTdEU0Z0)b#%u|>oZj07@SAH(aQ|jUzqdn&AuGlvt4tAzOT+D17@NFsNDyx|gJX7-( z*osvsyD4epp!dPs@X~>K_`wVs|9(b@l9V9`DKIjNLNP)2!(g^!y~e~x`(`bwmt`wU z&xgVgQA@4kKgo{=|AkP-NsbWN8e>ElXBJ^UnKn^?GQbzgKnWhvO{uE<1O_b(3f9FW z33)OLJ(uP{QL?urgaOAq47OtFwJ(~ zrN1UCe|EGbzY?DIgJOfS7FR04ZpmNT7*-sC`ZB=pOMIm%UVOtQ~UGW&pfLWem5H8NC*-QgIj4*B4&ZIiN5UgBi9cfmGUR1)ibrFE{yf+c-` zOn^)MPfp$$<`u&jk%aEt4YDT8bl#}U{o6+<;187Q75I|Sm@Wp(kr%OZk2ndZI%&EsxCc%FOq%7 z-}Q8n+e}pJ=JxR8N9Qc$-&;qK(YK9-8_Y5xi|O6a_ii|-g+3Qj>;|qO7#-31MWRmy zOsr$8GUDjYT$@97C3ckcA)ic%gpQWF%}SRA>qzb?N?+ODn5&@}%OcMbhm@WMs);jL zb$_j887#S=$%!*V!rh4ztFACIZByFSS*FgUl5lrfvP0vU;(+afU`A0X2he=B?#^@R zd4@6C79JFKVr6>YFZ@OmVe@UBq+ir2b4(oW$*tMnjG%J;N%S~RLW|`htKgu$S_+O} zgiwlL6zlE%nwxHsD@J~#JS?h<5Sqm>IU-exAHM{DTIy8>?wODTrwk)WbrV2lwcN{t zrXmxU5@pRiLr=`UL9A8iCbrFSJ$)Fk;H4-X^qqSo)q}oTU=NTK@yrU+MF!&vZswKR zO<**)$Gi^qw8g*`av@f7#OgpCwDN+zUH6R5W7g5}f{(D`POgLI3)_a8T9S~?hWGcz zx`VQvNz>gnLyfL4J(;OU)YoTOJ7QUO~DF)k5 z{$fK5A18V>5zzac>K6++d~pp%4|b@2`aq63ZN0f`=FhP|bjlTI%*fJYkV0|#AkD2h z5oj$JbUBG_-r~+jEZy-}w>@eqa-Np;=H~Hvxtqs{iN=sCXW2aTSTj~>yfwC@vGB-X z8&Xs>d{L>}M)3=Y?M8JdXio%&mSe)<%94k?@2j0`hidbb!hD}C<|W0aW!$so(;{8> z70uw2^Q@6|1hg@UoW|)K6A_WWj%vN4YFJ8f9Y?}UFKbdi3o_lmw@tpOkNVYthvq_Z zw~UgeokalAjn%y*(BD0>>dKWs{G(^vVjBewWe%HW5@IR74dWPWN30*_U5%~WVD$wP z+6b;YJnMsNSRom9^Bs+nF zSXz2cPGyr)e`$Kjo#wqxNIpmVtDKWxhv*HpX2EnUZlRLk0;Ue1;Y51o_#oJZdcZatb7J z+iF@f2p#mdKce(cAu#W6xa26#=5%!y31OPz1G40EWAg-;*>|pvJYoSxU72^3Q~4kyDS;1uHYP9E;4Z=&==sVc|?9h zZrl=96FSJ)@K%S}1ruz^a9h*P9UaEcuo zu8eSXfTEXp$L?9{$Icu^Z;D9$FW6NIj6`Wcc5HEqDlRXR5CKwZFhnsUf;JSwm=D=H zb&wI-_)(TtUb=uIwP8PV(upCtB4G}z4DciaSS4JTeT6Q@hG3o*W-6bNW;@| zBep&ju`!zOJy>`zy1Ok|a&&0X#ecWc)xrh;-57ZJlVqUhhr%E-eUw;}yrb?1w*ifP zMbj_QY`!+N7sFke0`@1_Y=7Y6DUodkW6L2PRjifNFMUl2_lQ_g$L7bp`6IV1#GBR? z$cb;nyLZb^mwrgN?ibU$%xO%jFTxY+kV(S2BP6Ua9c2!EEhZE!Jf{%wPl@Aj*!1OFnaGRc42F@#B3@K*p7q5qF?dFKUZc1F0?DYy#i2X7T z@CJ6HM-Saq;Fp$Y@Q+QbBS$rIZ@F*9ZUWFN^)5Kyocn%Lb$ac$Pgh0b==g{}h3nO!hr8TdmArkZ?mpGosxcSAj6t>?t zRA-Q}sn{j3y|+~zX6h}nt4XY6MSXu>(9Y~$$eVD4fE&1|RaR46J7-`)ZUBklv2HvK zt7shK>A_U{3(dF;x2I82k*66u?eZgtJ3(eHY0As2K_JtjXPk=ywVHoZ2x9bsgv;qy ziD~A9U?R`Q1>F3apR4yCwp5amDyN0@o8z zwncax@McbkGfmV82KSy@@UvKUbMW-SQ!Fw<9+F?Aqs0#bU#|0ct?$K>y#s0IkQp^B zqdC)-Sd>)s9?N(#riwX~;2}~PhU46kIz8OAp%>C*NrLp!Zn(T6=qm)5iOKrNoOiFS zoWzNh93qJFdQB;Wu;`L0)8~ud`ZdP&ED!%m-nSD!z9bm(Tu>D{jEK& zWVS{=!P4Tb#RxIirB7vG;#vpp%i+yZn&BD=~xYnil5YqAnGEC+zi0Y*iBN8WqTnCME=~A-rk`78G`voki(|3w?>a z?{P0#Vk*mrp>T}9O<~e1SPUA`8K7lTlDDXGj|VPSX=pvUjU3`flxgiZ<&_ESMx_6G zqot_iZ9-3PcOzao6od;Y)~j{l=CVe(;itwb9XN?4E@#Pq1CJ$_x+cFb-$;VYhLd*p zBx@mnMQtJ2H-p~+%W1;XkI$~syRf5x*}l%Zo==uw^=d029w!TOceeJo@0Ig&(VA>N z2LE(;ua=v0#3P;lff*A&c33Vil({4=v(?Q1wWzF}+t>WS`3MUV7CYQWMv{(yf1AO7>6xjW$73D6!1wBDxb*! z3lE7QTWiTgb%7s=U(3k5{K;LoKjuj69Ha*Z{#1$`P&{dnge-G7Gjg4e2^*GilOH_u z*RdqZ^|TMDt{5gv`TP4#=C@Yf%_qBK9tFGcjJC~q1rQSX$fR5tjzUJ$mSeAjkmzbY z4Be|Gm#@F3*ey2tdCKaqW-0D){C&ZJ5f&@G`2Z@<$U!7$K12Ax)9k03(@5w@wRF+Q zPHK$nE5&Y+QHZ8(g50^qi=Mupj?p~<*6^rs&ds<=$uX3Yu(%?h}H#f_`S~Hac z``3aZ9xc-xrOLr(!R(pD=KIBBV{0Ko^|InM7+(aQYDhK-Egw}S{!J$4p~?Qg5i+0Z z9OI1`xzL^8a&`VjpW0_VzLkE?Y*;J!E*>~=w9i$v2xX;grKrjE*?Nv^*lj@Z4y9p; z){^v)GfIm;jiI_@4#&otne4SBs$nLVFJ#iIRf=XO%bnH;A63YW`J>P{ugLaSGZIFI zO%=i(9XrT!5a;;BV@o@WlkP1PDaD1da$LaW%$%23gX2xxe6B|_Pl8(WcOtGIIOh(H z6-G;6TuwgN(aO2@Nv0}LUQUP_REbBVdMX%Aw>92Wx#WiE^DtFZxX;7w5)a5N82@u&hz=#D6Tcu`p;;q^M32v#GXD|km)Z36c5TT#hm z3q;r=fxZHzrW)#%ho!R1DPl6jCOOHLb^G?wvPpXrtzL2~V=IN@v6tKSE~O#N5aEav@P&4#^@c2XRB-?k|&e+TzGUG|Eel5_1^6oOSOGadzTt-fVP@px+ZQ zUXx_A@Cc4m5kwu!>2<^U$HU5zn0%g(>AjZ|KTm~}l4I(c(iutJUl+^6Q`o2ebyR2{ zuL`V8UrC{ssou6}1L! zG?RVetU>*07cSF?|Kh2$Nl1QwBD0_8aSFDy~+YM3lEO$+@q%A6~E>3iBTe2iAFx8G3-rP)+`$#-@i0`q3DbaKWN28x9>YKsFE+uEykgGMBkZO`Q3Dc)c=Wc_|eZ}Hk?h=aW8j)TmrG?(!rgi#CHE$ zQh=*Co3c{|)w+A{;VyG9EsC9evnhUMbDR z$`egIPQ<)^Hkk3G-=D~S3S$KCt35w_z4Fr@d92JVR2bWA>-vcD_2=h4Guiv}c$a5$ zX9{ohXd8J4tofo1e`_EReO+i2vWZHp^du|2DN9=ypl<(j^4G3S7$m3f zSyZTpRznMKcEleuKAXF#o&wT|#FcX`8;|qq=j^|m*qcOX0w05#`!!PM(?gL-F|%SX z!tvAW7?tytR8A(P{ZJ(uzX#ddv+RErEOL}NuUUE54ID0Mx7)lOxZ>a5X_Ty!mQL9l zj7aq%awwfTOVe2Q!PWC5s=wctgzi6vn*8BSHE%_mlqJ7bj-={3%c*9( z#~e<-3DMQe?W$7x-+b6A-q&Bxl5l&j<%T^Ea&`&V9AP16kqlit8rIfz{wEMUWIM_= zShHPBPilICFG=zfTDtd7CM2=X-TlF)TSu0sn0sPGR~#|AAEB@R;+rr$+~i!!I*)#F zSXa`mT~?-D%68$PWktHRdAk+R^>@B}H3OmC=^ta~cVk6yjsV}VeHT8&1KE)QcX>;n zY>&$8ojHs+xZkG%@bCkc<{ldjIizsrfy{y96s zsL)O}lPO=DGK-;%vOjO5fMsLG7>hd=`s`~*gJ2d!2z^CoLNyzQTYX&WN;)Z9fW3u4 zn3sV5xY!@h{^)hocqx0ANKrqNl7v#~x3|&%KyvqPg6E%m74YvsVWG@7(b5Ul*VQ(VZGp8wN=NLUsuRkq`U+kqR zgIP#1Oweh~mWZ|~$yg$UkYfuS?)J1VC*cCRrB{4IgqLCBN+;GR=PUo(%ldv|D>Dk# zmWJd|DIsJ({F>A7(ks#vqx*{>bB>6ypv_@l`LoeXOt#0l)_?s zZpMMSue0P`fk!^?s*i=Ud8#JgozYVCzp7ijX6SBIXKHF>qBC_;-(1bjUu1^_jUfUt zeJj|K3wuRg*7&RJk$+Kr=nVf|GziTW)!|l!QB(Lt+vFC6lxW6$WBxvu0sHyTH#B}o za@|CIc<4R77m-661PRUH_iO2*=CT!5FRO`PDx68whej9YddXqeQ+(A&xMf(HHyAR) z-jiFsy`m+wmk7bbo{3cV>?S3PDHyMoTPkL5W)9(H5TiuT)FsyC zr#q(_rFS~aG`{V!9zzma zD-TIH&rsHqJQHzc@*2rMNN9-Xe6`DBX~Mr38(WX>OG}`5t&@|`1lv4RTeIa7Xu9Lj zd-L?Oba%c_^+H_o$S_L6I$5M4<);mdPdj})G zO>AwK)#0+ncn*Ekk4j1_0PXDi4dOv{61QvKnRQRWo7-{kWwz}9VaUg*(F}iMqIDOC z;+t7@pIb_XWf8-KMiKqiI+e*AMjg5{ZZ~Ux-N4gv9eH?nRfw%6 zeH8sse9N9`ycK%{IDVMhP+mG^w+uxlw`Draf-F60CKY>xklyRwBG$<4KZk^YJby_d zb=@Ca=8T4-dF*8{NJbV`9cpT8-#J=aNf!>EFOk#Uec)MPhSs|N5R>rbi0(P{Wf(<= z4jDQOb|?Nx^l*6Bzf7b-0i9Q|PEXo2$iQyPC^HVm7rRo^`+a?X&++h}xf{ojiPM-_ z#+LR3c7)2LH>3gXkq7$C6rGc1;i#&mC6pF(!xHyC`h$JNqf1CIJHcpIM=hadzrvCt ziQKqP!V$gP@`pWq4@^o4-4VU7KM2DfP30YEO2f4$dPLH**eA%H?!vEmu{5Wng3a>T z49Q|CR0NV0uH}9_V&&Oed0QPs6=PGAHB?_p)90y&9@b_5^nPm$i+0SkXzgZ7gtPqq z<6$qJ(-SMXvgwa;$J*J-KkF(qg-BBocody%)Sfyeqexq$J`*fK_wiW=LOBN`O7a!NO@#eHb@^O_ z(j;7${)ZGri?iF7t+Zxdw;albv0#Fes(j4qeP|}*>+uC;S|h!@a#ULds)-|UaRuw7 zPm6{KPiVZ~ptf%rx)vTOBA!1t1ZDddLhbFrKmvJT{YIPU;EQOyuGQ0E?DO76v`9#P zVm`#cB9usapn71lxfZnzsGmIulUhK>RL}RF@ zD!`4J;FPEWTk^T1we?`w>$4cu9?4a-^=qE2><nayF|TL_Pt4+sZE}LI!2v(#$)__M@DfP-H|kVI*E}B`2?-@vG!E=9Lr_P zw_%I7@8;;+!1KTAMZM)k$KIi&pafqc4?~DKA(nE{R8WM35KMCC!P|F)@Uro+9BH!Q zFNd@KAfC!}NS!U{D5H*s&|u~j{_`qpyPW+8;f?noB$J&J9y_65CTN>fjnBxeu}v2lI@EzB>k3eOgOH0XvhB-Pvz_R+~Dx ziiT}x7*l3%BnD+e^PMta?$ii*4OI=54$-1HY{#acS*g6*v>#Uy(!&>j4HAk$&Y!`_ zEO3|!2@0HvI$xF+fC-$1QP)|3;r*Y^REZt7!_xhiGgaZ+HG41Ff62j14qtNglH-@0 zyyWyHXD>N_$;C@9Uvl-5>zCZT5V0e}cV0w4oW0H^>o06G8zfC<0?U;}UfxBxr=K7ar~2p|Fw14sa*05SkMfC4}X zpaM_>XaKYTIsiR@0l)}g0x$zu0IUEu06X9nfCIn@-~w<1UIX3$cmTWrJ^(*J03Zku z0tf>{0HOdffH*(`APJBHyah-DWB{@NIe00yqO)0ImQx zfIGkg@B!cn_z3URWEuaoi4`={10-6BLfEGY2pbgLt=m2yAx&YmP9zZXk56}-7 z01N_#0K$lE0*(O3fD^ze;0$mMxBy%Nt^n778^B+{E#MAt4|o7P z0{#J>0MF1bmp_=77tjD004x9w01rR_AOes8$N&@oDgX_D4!{6l01aJyC1Dpdc z0GEI(z%}3o@E33kxC7h+9srMke}E^zGtA575BB8+Gynzw3xET_0}ud+03-l100n>w zKm(uyFaVeUEC4nD2Y?H}1K{3!nqg0~i2| z044x4fCaz`U<0rNUI91&oB%EWH{dnk4S)y03*ZCr0|Wqq03m=dKm;HP5Cez3}bQ3_vCz3y=-S0ptSm0QrCdz*j&apa@V5C;@x}lmf~C<$wx6C7=pW4fqbI z0n`HO0QGfCm0{_oo4f4LAU;#Z8lt?@jl>Gm26PflhSVNz? z`f?|U^Z(jpV^2>me|LGK;2^M*vLm8vIchcfr;v_NfG&)V2CLf~N};_{Ylr$I_fv=P z$iLP?7 zxM}X{6DhaOU;3vriIOI_T&bf|h^pTfZF?|0vDwtU^(z_J`2w135ug^+tbH zDqkUCsOI*vX}tLhs|nm&m><%OaPRF#x=Q7lBmPiJA+fVNxrpy|CG-{(Qc-KBh(a!R z_feI9uhQcy8fI-Z%oIU!rR;X-YtWt3y!z0b*6B6P4rB05D~?*zNwfPCwWfe3%vBeG zyS@PFB#RlU5Rs=|N;UUs|>8zihQj z4@;n-6n`#uO*Ni{8Rah^53Xi8jn{?v-B+h;d;Q1@)U@np)rcM=FQBkyiO%lp{LJ^* zD!VsO`o$71_hq30U5QBGKHGQgiOB#xla!2R?4*F9=WXb++;So#Bs^JOhE59?^`cTz zS+yA9Ekl$Tn>vz^s>H;Wmn|Ef5iGgq;XJ=j3$)_n&M?ZDi!^`b4ynVvrE0ZCz>jffVykU0I$%N1uKW zMd)L&P*|5XAG+J)NG#;1=+mS<{r8&4ABx9eqH9D^ft14U#=m2DRcqQ}kxl7DZ&-V7 zJpGD{X#M1~JcG6&g48cN@(Ye&65^w?Tq_)}^7vy}*qvWdBT9|%`f9=o60`&jLBYZq zQYdlFDX-0GlAmZ??KDhrXDgqvGO}de+@$gtgXn&wz*SyG4=Y)mEME>0!L?vTpp%*CkGzh(2BpYB5oj(kb^K0TOr&b_De1 z6!lcWwNT9O@(jN1RiBIsNvGK2&fmvsq0H2J@K zWIwi0S@(wWoBBjD8PsBBz&_}U5?{hS&EgAx>Xd^&WTnnSp_;GAi`WOb6x38qArrF~ zPHEcRPdeK-6EY}RhjfNL4?sGes8m!YC;CekZZEe>y&(4|M*}1Kimp#OMp?NLFC{UZ z&HUsMX@9nk??P{{o~{Qh4j)sL03A?`_eMXJD5OyPj^OC&Blcbr| zhxcsMm@fE=Ri}e@gcK@AOs%KIjL~O~TOL5GUBE=eZQwgweZciMJLzEXHO^PziPdFw zs-K|;)u(P$CRRGrnpx55(#pzIhrXu0gH=z1Y#KJg5ULBG&zGYTI%61=l!N{_qA~}b z;@AUEI^I6@$~n9+LFiG5SB2{-k_@4V(W&PP$0eL*x_#F_}Q z^!im%QqRq~>2}t^9mR!SH6>T!t>d_vQa!zH%-bQs&d#dvpQ4hT)o{y&sI1Ef)OdTs zf$pd#qqPLM7LBV>-*w&$HI_ecO6&+{qTI_)V28BS>^;>+&A6D0cD!F z*n0tAKJDy(Em_a~HU1Eg+z>sqn#XpIN#nZy>c{qb5*j7w z`d9&euR6P8OpLgGRL9N$t171GK79*Q0>)cO#8&Zc=*Y$v8f%K&W{UDN=~uCxox5{{ z-l=Y7{Y5MmzSgWM6^qsh4WsXIu;9{du=gM$Qe?B+NHTh{QI{QwRRUbM(7c| zJeCZ9_3KUhhuyT5|Dfxhq63M-ZBZv3+jcs(*-^)~ZQJUKZ5tiiwv&!+qhqt9lYREV z-sg_{P!Bc6s@JNqYR!LsI9(G~Vg30d?rc zvAdQpGbX|uCBPVZ3_J_c5+%g>(V4u@dTj)l+qsA;r`0oY8)q`u-{Ww>k)qEPFFCI@=gy}h9;C0MgF%F&7^QL6xGyfxRNoXcw<|xCeOxKZj;wx6N+c)( z&h19m`>bwG_NUVdTX+Y1(2O@Z8GQxxBod^ge&Z_*+r{~IP-$>oOgT4ONSW!73eOHW>?-FY220h7gFu+Ilfmo%A-`+%U(G(6Yvlgi1dh1_SIcIyg zHTILC^uwGAO|&lhz?6Ak*^E7mPwD`$=Ca~a7jPk$4Ie90#6_6DG=x0H<{XCS4EhaE z|3YIN%tSL4L3l!@?GER^GH5>XYpP=IBrZgW>f{y1=s;5Fe1&EVK!$S9rpVOf|b5x8d6YQ@gyu<_4y>8}k~ z?HvV{%!Pd6W$*QZ!~+Vo9u+3y6q785uoN>A&K;vFLueyCt18hFa`Y7U&Gvs9G6Mf! zLk5_z@3%ns--yiL|Bc8zaQt6HrfgWKY0-Jx#qh1{U~VrA*k zzcL=FVKu#=>NT%0Y%{Q36>X{rk(+E}9UcF2L*J~C!1mhtHRcohU{~bRe0S(Pr>uiL zv$;yd_v+bp{}ZUTpbuz?=2t#+Puig=WgeDIh_>{3qbjiP+9MX7KaJoQS&%`Q#ugJ* zhWtbo;(6`R)L8yk4-PMQZ%jj6MXtY=SP+X1kz&FgJO|oY2TnmfaTn!X%Ob@8ikbmc z_pi@CJQ!I`fB&Ut`M}#)=X6Pw5NVM5Z=SD*oikB8Xf$vZGhGMnt!LsOTdn1koZ^~` zR7>(T9+X3#M1&pIHQti+a(i+pU`vS`2%3LBeA=0O!x0|zQIx;^YklcDR>^H?`)qY0 z1TDr^NhlqZ}iSE~YrfBy^; z;#OAA4?Y0xZ|+u3H&k>1>G-QlqZaq03$YxX`V#j9e)w;{RFB?@rTg|L=Pud1^N5ZN zl{JcNWxa^G(s40l9r0{8vOqxf-UwX^j5= zA~L!EiO2x;8Y4~rSCLuzuZqk(=hc*+P8K2lbK*?gGg5V~!$OgIR?@G*__pg4-gV-! zo9uE;=PRQ&de2T;t<2mtvE}AD?prcsRoqYbA})Ayq;y#hwVQymSPC{vsh}!ecAsQw zx;q4L)z*D_NZuJk5?DUKYMab)ztKKBR*Nx05|1zAO*L6rpa{mab&-=6wtpe-GVmr^ z$taX97bs)%E_)wqFCdcAOrp_Zb)wYlcEU3apm5}$B2Geh2~kOWH^O5O?gD+u+lk-s zwu54MvBs+EFQ*$VM6vE2rqYb|+37n)b7_n@>+osqty3m^t zBK^gAn>1ET-Mw}Y75@q92lsNL#u~V=k0P8RN*G&3hc`}q@PgE2mZu>hTMPX zzFU3fu%9j&Z7fyKG>0E$;ac0_4E__1-y|b94`^RetD)Z4I@G;DsZ_P$Vn|Gmd1&Ed zaT9J~EtQf$JU}2i>$($~UDi1Xt+a0138+uMCtRPztSObPt5jyRB{N}a0{M3ycH)lj zR1*)8k;Y3mAgL6sJ#7+41w<-4IbUj)=GdT>UsUyP-s~!2EI1v)v20kI$ks-*pHUO= z859?Z;esC05_t+8$~o>>CzHDImqjee=$L&wRx11CHI5Ekm(ueb#2zGEi;BtFfUr32%=L0&Xja<#1rluF?^=6ldbha+w^ah?%<0d^nTyf>^z+45E@qQph+Z zY|`aJEh$YFu;#=IXJC&N&{H`$-abuMZTbG-9&B$I8^Em#uBcb};A7nJ06N$_K4~Wnyoov(BG1a#Gzxoc0?;>{ay(c~Ksjb{FXqaG)BJ zDY2PnHu5IB3?kPAeVsJ7GPDHMH+u9erz9*6KBi=_ESd*8ai7!gpSW@-6%_>hc%L-22GsP(iF?rpinYPc-#j9tmo>j28IAA#?6TBJ=_l$wfu z^GuHm9Cold(Y`aAkXG@-b4~i;(PZ(Ov1=G0x!E$!4%H;?RX*);%uh2bDOARJkv2Mg zBD`OA4AK9wW0b$_n5X}=V?NlW46QeT--nn>1T=>2+*$|DwILm@vV?!sHQLCh8q!pS zQ&MmGgPBN_H@gapzU-LO-v4RGD4qU4b`1G{*fAfG@~hPDv|o0N-+$OKP5>lK&2*$? z3C$v1!{3pn*4ws!g?p=6%I_s6_tcCXP@}Jpc1waY0Lu}ngomrJq#d`*@}K2uQase& z%8cTAYWiml_E0NSU^ZUA767IUlc+ z8`b!pfRQSCNO*BMQMJjV>c4aPT^!svl)K1O63+BWh-q!Q)pnajh=}YN_%3>%kPmB; zr(PjTE-tdH#{eyR?T}Odx5@K6xP&bvM*DU;pnMne6U6UL+1oFCO=3UzesB_D1RzCUaG1rinLlL$%G} zaY>xhTmCyqr1XWyTsb)aWtnEb#Ct~}F(3QfOMbhQw40+_y~BZXw3<&Rx34^ky&(<) zL=S^u-b&Yw00Lw8>HJPr$PKl4r|UlLq5zSS1ES=0Sk61-E2ZHNilY{8u9dL{F7~Qi zABsv^%f-YXPM7h&hf7B!CzW#;>4ArV-h&lm<`!c%_z&hy z49mp2brNB4bSO{3hqW$9nS!1W3Q`fk#S`yrKNPfe5i)N4@+FsdMj^|$x#J%X87}eV z&s{v$C(})>&(z#s2j?dlsIA%2p$u3t(_ruw^M%n9UC1YKh-(2g+=K$YGup2N;Lc1T z2Ysdcl(9?tw3=kJos+j&VMFQ}f-6u$r(F&UeE!(^g3UZ2Is(oXs!Oqdc2~ zddmF*CG0VCiyq>W9$3=odc~6JE|UORW6G2Z4bx64X5k%uF}QBcG#_+o2sJ#uA&Tr` zKk;JVFt|c0ESA0KD_gUVFGp-e$bWq>&@Q}S7|iDKCE40k5XRYryw?MY?oLuHjPBGO zv?7T#8@^<|9+wY=9WT}I68%5$9r!y+(bg{AM{y->?f%^dJIiv?I;4$w z{~yXz2Jv0nitXKQwQS(dzRdPAIwf+p``t41a(m*ASgKkkxCjx5E!`j99K9ISA+X#q zhP%>!$rN1EkvT#~Q*4aa!@cP+#E(ACNTpYRiWVw9iL6h~YT6CK4w{~OZ_lyhUvJ$T z^tbSBa=r^#Ek&um-uqC0!G{nzXEbpubWIlBBTLkA=qpK=j#mpn^_8Mdom(Fw2)W>oa4nZlJ zqhd>B%j6{>nO{>ClRTznftR3_tyg+NH}XL|IosZ7ryhsONO$ZS*L4pSv)bnpiqAKt zjf8*REXsk{R{`eIVIhr#+bB);;i!IYX)Uim&+({xE~9O}hSOft%Ne&q$2MK|K0h~g zt?lr}P(H4dWw42@;Yc4Emyk`s>dDpz|I`y(x;JP*9E9OZp&@Zkz>{l2!yn*s|E!KL zI3PQzDgVcQPC|Xf!to81ZSv$q+|}Jx)pb$5dD!$(`w`G+bEt_nEN|Xd+5=|Xji6(X zd069c_)5O@7>}gV1^%GW`h!(%*6ui@nXV0$TAQ~kX~lJ^S*1L|jrt7WkwcAI9v)eMqF;cY=)coNoRx|Zp)oHu$VOpG7 zQ$^$?moxyGua??LedP80i(ACJs@CKkn*SZ= z1<1VYj|+r-AF2o+lQwmjHcKaSMwPN}H!ROYqLo8x4l{}sb%F8Bah5j?Vw+jNB2%Xju>pWTK_sLkmp1h>3@Bs*aso!&eo zu51lddoL6QkJNAMo7>R|^2IqHBV6DaboGo{lQl5mSJNkMM&meuBNFJ;Q1zw=BE@gK zl$!Hk#{J0eLpZ!g1p@33P(>h>+Q6=usShoO$_eVV7}cUXwyinZMv;~DtI~FC|M3rn zW7HkP$ZOIimf44m4W1k|T-#-8M+>Z&Mrs4jbW1=KVX%On@nlLNQ9h?jv`zM$7To%` zP569PQ-`9iM0=;v5!fc|fCV9p+BDpDEC?YvOjZJQzRlwKBW#z%BHcC3Vys~2! z;|#Y<9L_bw*SfS5lEb+wUb$7o6}--@i>rNV0dh(v07Hg$uR z)*C#@1r-S`PqdJqYL1XsG2TN8(X$JE?m|~dK_L~@`O3ByrEo;2Pry)Bvc0>BaB} zOVt;E!M8|?9iQvmg!#v$un;zUTiKdic;cr4D15+O>DL!uBL5L}66@Hq?%uhH99D8% zl>$9eRlXt_W9Oh*oV>VTLl@9Y@1S+*&7U(( zzzb5_B;Nu-iD2!|p5roP)PdT$U%XQA{9)vb!>Le)0y@wB-P1OZN=W`>mPRYF42#fB zwQ#IgQ}djAlBNz4s8E`<_*?%>6TsEb0=qGK?4OWgifN#;q^VbH>J4`=Ok>q_B;;`LE2n(O{(wIZEzVjAC>c-E1W7r1kQfXzm z!K|X+Hb&*ECWsPDWK>wZ<-qe$v8Hk2M1PkQWD%EovMOeNZ`w%gj+Sxl*CQMdANJbzf;C@+o5-DsE+wYmiT0tJyou3=F|3z&O65Ov4 zlUv|sFF*8W6akG%Mc;M;-3ln!oymbnMSuY(>-(A5t~Fa`6KOGMt*{;e=EL}>#;;UQ zmu8CdpS9sYR4%0Jws^!XU4LX@w~4BYc@7QFfHqaBN5qjUFfFuSX>f8(!%yLE8-h9X zRq3TZPzqT~2(8BEu6VXL2k}@v5}N1c=K(Su7yjr|>R2WD%maVjqh|{b`DV&#%%0r# zOR&6h>_|~80*bga+E3i`rR{f5)E_f?(&I3eIjS^vY}B|mC`b7LxPh7>+*&IkMyoNC ze+<>^MObv!z@hEUEebYBZ&O;I<%To$p1#k|V(f5t#nlHLuS>N3g2MfpaCa#_7Xhn# zPuDy^{kE%;y9hzSX~DErtmGHiUjjn8RK9;hHDXUOdZbIVq;QErkQVFM)(7UzlGq;m zP_)Ert#Sj$f&K0+>xb|^a-?EAO&AM$UUe?(>q6z&f%%ULX$P=R>eL;f#UE}FHhH6_ zfpS67W!4{3yRs$MX10d%6w3m?5YfABn3~KV6C&u=q-ox?y-q$=E*hVv%x#wCs!Wg|AH;IHrUj|Iv|ssc1;QBiPTQc;pt^I5}&taB`l$dHt(4abuzQ7+Rb zAiM;l>!={TFWjg!AgPJwSKixAgEP<8v}`61W#s0RkNx9InWnJ4mdMIUj=XUX>|g~* z1Ar~E5-}(P-c(wsh5F$%Xh!seI;f%(kr)jDJ0WPdWg{w{JG_QA5+ATpC`0IhBWk#W zLy?bBP6PT+gc;kB1U!SLY0-$vw1^kAkdg1OW)*d;sz7f@8{Mi(JP$5gXa+@FaTfw>x(LS<9oT z(y$OBy9z>4|Mh|vA6I=Ih!qTqC$}OL%fA^=dDnho;V0y2weRg+`N<#}WloFIF;do> zM;N+`I87lZg*jqmh9JuFU8DIR)Zp%OaDOlR+P1Rp>hYi`@l8bZCP<>{OhO}9*MSL> z$UYT=0f%-<0c%y+dkD&aV$YYvOzk~phX&NC28vIPCX6J`wB@(UN=f@#5?vLC4-+w@qZ|6aLAHdN%!c1Yho$0OV1 zF++@_H!ftrel)~u>-gK6G0IbC>^8|(6*~6xh6k%(=F7*x3cg<1EYiA zIx$}$VWdt_MNx6)nV0rV67ylW0;2`mq(w#>X*&2{yjmb^jt-z@g};?Xy9`c>KOm4H z_#ZVdxR5xVm4;5QrlU$7R846kF5!BLBKt2!lHcIj+KpMr9^g?HP_zi$rURR&%ctN5 zS9x)&GUl_o%9#gN2pBv?JOq{<^On&>IiSj=;HX@=_#DA(){w5l@fEF|bpp@Sni(zi z4YCDW71DXUCPrkx=l;}l7RH8W1j_hC%d+S{gGzIbL|_aq8Ru%Dgrdp1R9GhE)+3FMD$i89S6brv^>;!`gN4=r4lgV2^F-NC4 z@rE57KM&m;gtkoH45(Qf>zZlkx8~8UV;%Ze`vDtP$4zEi$9_*4ZhZ(L)DE(v)~cU& zXOf$1sy!a#6_nb~LI5U~-~Mjx2y~T)tRr|ArZT&a=4LCA=IcV%tl|5d(j$EH#xBCBm z;OLVp(BuFO1mvgC|5L&Y$7pPCVyf?KXzO6}pJ6d+%}y|MsQ=Z)+++52=LHLBq@d6~ zg+22+={X)uKQ?l5pw*fp!7frm7S~@}F+QmsImta?rLZU#>QXS?QCp`s5X&V~EnP2M z=l^0&Y9x2E7qjHh@oq8)!^az5IZ07au_@9pt&9^pVY^C{b$WuFlx=L8+d&;UjWrh}Bug}K2LZ$L^yI?j!6w%s z3@{0#@-;21M_@N(0f>bD^fh7KU>f{3mz)qV$4rIxA39-pP=CrhE)9K4ABZqHSD1-N z;w+8v?O(_HlL0Mqs`|W34tdvhLU22wct{;v5SQd}2<1S93=CvK?J{7C^6Tr_KZcW1 z=EFvf;>UVGWUUhtl%i`D@({Chgs--(zw7tJX4!!g;SVxAYnqUV&7`x#@pF8Gc7jsK z#hS6kM`VWr^dnt^liB32#9-AGTsd^9qGH82NuwhT%7eo9>dVKPb~d*+-x)S5*>ur` z7G5Oy;kAFeU04#?%ejDM1>sRS>#Oxm-tVvhq0Q1b>$2~1V>fJ|q3V07TGrywA1l!} zJshVFq|cfvMvuJ9D-hJKKk4S|Ocr@pq|sp3?>JHw}?B5*ES5pkAZN zi6wV(N;ZTOU?WzsYJZF_dh=z%kyLf}KmW%Wm#y&Wl&~ArUYhlcs@J6zQE?OAHgESJ z3chFjKvR=dy|Hq&=rRn4$-C6#T-qrZSe@%l2#vm~=L_Rm`0%=)0%BdeooRtAso}%S zoP7rz`$|4?%#q8T44KS^u2#vx-@?94@r`9XX~KSBt%hw; zy(FpuDu$xv)RMLmRA$PZE~RTrbn)?eymw-^>t}kpzIT@tqt@-`mbZDF;9&PS+v*n!m&<8R?)89ZyI6nKAQfbNHTsGLMq_VOY z1g-sKtKu#Tn@dot1bvVUs zOGCgx5CP5h)X1gPgNM!uCn5g5cDAPrMR#C^|LY)YpW4Q^IsyZ&5?f{ZRD2!DDsBn8 zpgp$Z_6(q@(Im1e{%RAmEhKPcjZ6$(;qZ*@Z^d7x+FS$aRfL(X_S4q0aGhLip3Tmw zIaS__e*Nt_>gVF6!ndh4-Re!<^aNvKA@X-xOOE^5Ny><~|0D+yAkd-VNi z>Ac@I(}NARn>F!QCBpn{XiG{7UNAi5;?zIVu>3}m=XS3}V z@(I_v46(~-l!MAS(UXjNlgnuBia=|)ursYeUK}}-!lQofO&IsCvi?fq;5zrCi;7Q- z)M2bEspT%(k1EZ)&orvdt|k^20IWZX%Lt>=bQcNL zl_Z52QPUF_)jH)H#i2h29>gYu;;PRTI0-q^^-0{h5i!y=R!^Sa{J~Iy8glkSZaP~3)DB3JOnww1 z;5CSScYm~C8%@VkXqpo zBD}qZ&~Hq5PErn?hQr+KuT?liFh89ytWhuhw|fyU?L0Iq)S^Zo)&3^1r&)U`25Hab zE`F{0j_djiZy(?d@)&3gzQtA1TdmTo_>xV%n!sz5i)z~m#`Zts_-rUL1;I2RAYXsW zKxF@a;`q7K=bd1@|7E5-xAyCKbO4Vt;p&yvLCIEfFFwh3bv4O|n_T+7+R9YYbo^{H zL6?jREI|XoJgpC&nxaY7v+Hk1s&|imioVl3=!C&2&aB(-*3;93m3WD~S&Msj&g*%i zLAAW!qc#K)nF7!^Xm7jn|5lxfcjl|gWb6k_Q0$vdI?Dc-P_e@3zuz-{i~?kyxKwbU zzqT<;B;({&vtV!{w+Q^s;ENRV;~F#z!O~C2FexOIFh<$X=gCxNp*s9lXaKMDGiZvT zd}slpJdI-;()w^zATb_M-x|cZI3hgc5sV8LITk z?_OKHWSsu6Wo+`Y(WEh#AAnsXTkHuIAu_qK1|M?-KUx`!C%Ar}mugluX*y_{WlPMr zkvfabhLMz++O*yR)G>)Pcrj2G|0ud&+?S#wq!Oy7tQm-QgXHk|xHutEbXFYi0BZM~)W=v`99VvN+(fnkCXl5Xx5mX2$JSi~IL?gfTfxNS& zgJq)x1LefR-Qm87A^=;ohTrwhC#YAcAEYt?+8E_1^hr{wf6tE1np09}6qwIf^JYmu zoyiH>N!gT4OMTUr1vr&OuX#+?;D*18of+nL6&ojwZz-voLJ(Xi#KLqVQoWv=ul?+| z(KT}S8Kh*?`C9rhlL~W=_C_CT^6G48l_WP!jm4v7$CG_RlaKM&Q%H zg9Y0zk0N8q1#W5~JYOGlw4Tx90VfSX~5nK+5xZ4Iq)3>LIw_1`D&cTF-M;hu3)h9)iLjY>Mlhws^SD+`OB>WeEe{CR{TH z!$JdRm_BJ#ouf+-60gMHEjK-L3Pk(cQq-nyuM0!ifk??RomL#Xp9A@r$}>}SgTGNL zr)&+1jTnwWEASMk^B}cFTT;yNaLRc)Quw!YEa`PwM6xHGbJz_k_q8`HOTmHvG*&PnL{rwr86x&0Pff*)#umyheFaTsmSwiAH~A|p>Kyi ztD-F;CP#%o+5PQA`E&l=t5Kxn&Isc3tS74rgR<++`Q`)u!)q1I-b?-O2xXVDqbEr< zqx@h;#r*!6o_dyEMC7>j&Bx@btNh#k?Ixb%=Id({SI4yh$QxAc{&dmI?kpnSdN$TH z@N^j2J^(u3Wt2aDB}MSlu5Qd#?3ZH-+1GE8gJa@5Cgp;G{a1qn5dX|HKkamXiFjoE z+;}r>=W5xi@fW0P2&Z`iva#3+A=$ir-6pu!pud0OnHtvN;`VUk80-Yogiu9yYJdL? z`{fZm=r-iaLZROr?Eqi?Mm7N!CtlDu3>?qS6@8W) za>cGZk*lzCVe^dNfaReo&>{A{H@+&|1ev9b?umucXj82am*W=vj>po`bRx6lEAkCl z?PXavT2!X(DFW#WM<(P)o}B3o&r)_M0L$w`w*c9dZ9d3u5Z}2&S;^Rd5;vjtzE4j-$a4`Xmc1P~|!WZNqM$HQXUeKq%3^9V7)L zerqFgTn&g5eZw2`$Qp25`j2ot@%Zz!+ZEm|XEbzm~mXCvEWbprqL z3T!;{GOGKW;>6?AKp!h{XGCGsl^JtoeNzbG@DTAml_(`MiP|uo``EU9e>z^0cZoZnoHt2 zIb9!`ze?f<6X?wtNIwdYwII}DLB`7xG{^Ao{2fQn7v%E4hKj^n8Qprv@`Gckq9BnH zP-*RWeb#y-A3Fcxy+1XUa((gZEQ~H59Q5I9UI0&GWrn@w!wFqDb?$D0HSRBpKy>=CANhuvt85s^1_?oeS9M~V zB3jV2*QLYe-2Vt`IZPF1%m#u~psw>}@mR2>WH8DY>pyYKl7G6W!>94gIMZa)?){MC z_tY#ffJy9!8Rre=J(nfhp)5KL)hRjf)0ja8CsLVgkEXv%ItJR(*qxPTkT83w>vg3#Q5OkF=M>HG|cdZbP3Agh3=m_>$^x92cp^aip zK>q?7p+e$~S~~KE+3@>10SPBu{`>g9KZ_#=(MS|Y1v9&GCQ5Yi=Stc-yaQjeTK7`l z%O!yg29oC(8dA$x*jb-(37+mxmanM1e(+aRK7s}g0gY{rqCz4ItqjotJ`$dr@tVka zCJ5|jT`v#PP>V9nfUFSM38|-!*mogns0Vawk;2jwUuGJJARd!J?IP@*==Sf+Py9c5 z|4yC<&Rv*~@mr+u!?+Vox7b%1ZC71b*slPyf8n9L_H?G(r~l}VsmRs+7)pW>wheSk zkCFUS30e0$s5_)U%!;+3n>Jh`#R^?K(2~)>>$nzgU?5!~ZBSI-w3OIrPEsl#G-a(f z{{AZ`Z#D3X;cnt2E<#Df&2KtGekNQz<=`hZ$e#SnZpD}yU(v1F?^du#_TrK=145xsZyA}_|sWof>Wbp%|P^9U^vw{@2MO_ zQuU=W4S097P(AJojUjDd)I(j){Ks(-kLX`0#jqzZ=ZE2-!{UGMU(prG%VtsAP{GY( z=j01Nqj_poj0-Z&bTKQWND2-B_eROMiKKU*K>^wTzAp`rE@8vqGbM&ra_g1NOLq@W z&~KwawTfrz;pH!7&^JexQC1O>d|atsGRPJSnQX1_UJXk`Tmi|{D$J3*cu0bIt9Nh7 z?q&Qx3k*&(7<^K9Se7W6&MX)fD%8I~rQ=2-V-QopW`0FcZ|YNU_$}K5i)7(UjPn>B zoLCY6{@Or(zD|mKXA-39sAJI|)#Fd7eNH%OQC~kNFLg6`KX0C-;yzj##7K)5S}w`J^7BEXfOJ)x99NAMf$g^|OctcLcgc=&hQ=+3~ft$)7- z5e$cySl{bL%xRZQO+y;If9Oo` zp=!XijYD1lc^ouRMui~de`0;`RiGGD(JdHOfsY785N28*>2?mn=^i~&PLp1%4tMBU z9%1j3{fg}r3b|KnzEn!_Lm4=6D6{AeHR(!}kHg|tAkoRH-KHS`Yal99yhhT{Y+f}r zQ?WJPRca(K7%{+E1`L#}*O220FLL@BsjZbZC3Ush;p`7CZ9Azo`^dGa+V~$&57#YZxmcd7Rib@}VzOjdhl2tCLp` z2Te;W9TRv*tTq9_5pn`x5HD2q>t#e!3&YHkF(+>SIn|joSB#U-? zG^)^dK0J!vB#aNpYOz;Q4xN8qV9#KIZtxYQEh%S^917rA?lZy4*nY;2Kt^$8cl}{G zvKx_iI>~REzJm8}tpwoK1>7{h0*JwVn(xuH6=FGOoSrFxS66~{TgQ)~!OHyvfxXrH z*-7X+Kh~%2T_ll~+cL_|y?Qcz$!b}p%8~|f(uO@UW>JCMrRc`h*-b0;bq=PQOL^2l zf}u^jrvbGNyWT3TtyezmTov{};<+WCj;CH&XX9vRV{seXAjpUT9BXMni3oO*^pq)P zweiAzM8pl?x3*#;DKc!pXVJp7#Z7o&@Uw4XFsJF((#sCSU4`(EHXrAW<5!D=v#F1_ z=<6FZHdl1YH;gwTwVdht8kuITEi*};_MiltdNdS{99Rm;BQI8lfw$ByY>f&pHCK@u zA|>^~w$y~Mf0cwXZH`On&6Q+t7o}Q{pq*qlt8N2cL#uy+K+C@v;{-Wn74}Es6JH1v zNrM4-Ang*O-Bo^|Sb8yr8qBp@W(6M|qFa}>SBl^q>XE_i zwQ541t6Ig=V39lXvZSqTWu4p_ITh&IZxQ9h56zl@Nbsa+S4fVq-EIQilnbj_iqWKQ zV+0V;HZvvEVxyH2GdwK0C;2FfyOI1G-gc7_>$+Bw#*Q>zTjlxV8Ea)eEpy!HepQ=3 zN{e|?UKij9E*C@f^ESOViq*q66#m9gwCk=RmwB^#zuCIkC8Q|%{(EhWrk8KVQ1oZV z50n!4IDV>@dv=+z6w$Oh*Y^;uwI8KqwoHKO?-M7{hHH&ze;>8%eb`wtG@tQk1{9?V z%CebFJjCBjA^r8pILWac>_nFr%+k@dO?@)xsp!I7aSA`aHoq2;?WfA6;tQR6IqRu=~a3})yD0p+T_Iz4Z1xhjK_N_d>1NnKx?PZSCqHCxEI%ASbABMY(3_lrTpBk<>PsK7V6gV`77zX0@q7HRAA^w0aAFbKl!@oyN%JxEA4@YDRe9D17z>ZQO*LD2}j z8K%D+ecrj+qNBFPFAtU#S~Z7#DO+uLUO&Uu$7NHfwaHY+{zPwXf2l`(@TWY;%fQ>2 zkmKKN{2oNF9xYa-v&9}D`7+}xokpu%;ejP>V>a({uj<_6s!%1($^&}m;K~;htCPJ~ zysrp6%YQBxaG|`hI=u~yx~GAkauAy#qYF_C>7sBpuk>iRus-M01` zlD=Ie6M>6Z-zh8;V!Mu>I4sON;e+7F92*Y^fp1i7ZiC;zWvFX-ZF6?^A5=CZ6&bYS zz=W!_gVoQ<4MdeSj9(9W2}VeWL+!|LLgv!*G@QEl>;(t@@f3QFFq8v#N|7$H9LarE z9Z}Y@Q#8a5KEuR1La2xHX?b0{c5St-5ocO!Hc+|lXlW7X$TyC@P(=Am+ieV%QyRON zmFs_`lWeO!>d@vRUP>Bge%M5~iH2G04!<-a$HgZ_I|$ zBW-C(H8?Sd^FAwxw6*58+$M?;BU!y~94^Z|ScU*bhfl)iu|7ybR|Kn|4*2 ztht@y=1`d}=HhRJybUqJ7sI2KzlOV4;I|v-aj(zZ9`cr6R!g-1o(4j>?oZo%j04l7 z56q3UR@3D|3t?^64S(^fOcVJi*-Aagx8}W>(h$^mS9lB+`@8`W{|kOw^@-M4nslX0 z>ygs=R`2TW3^(@-5X_!O~Tw+JjGu1b> z=vVy?4R_V7dOiaCZ9V7;G17^W`NPX?W^%6%5Je{MP}P^(i+L*k>Ttbnb(SpY*N#JE0aPC#9*H9;)TJ5jC%KLc^lN1o%ZxnRA!otq-|!px?F7`|0! zr^_n^#G5+MN@&cu2&^6r(ZmU3i@uDp38w2ko1v1pGi#t^0Ip%PiW`UF6SIET+S)wt z$LvjU0DAI*Hs%gx@nOq=8Di-^UPEh(5*CUSjebUs&Cp1NGnYMo^yLdvCW6|MKCFo)`pKCRiRu zP_Fs7W1s-fO^H!FAT6vimrV9LWv=u|wM}@~6(+crOJI~5(*Uk^=ps?a~hQ z_g=E~3GCq@($sp;?UwE3gSFaaa{?Y<-MQ@`YJ_j7kq|34)AuRq(YG^S z3ppXZegvw&6X;_;n>7^N;MhklT7mXB+6Q-DxFQ$gokuhK{H3DFd39O#nx5A8PUjMk zU1SjGGm*Uc1J{#@-Z~^!2<=F(Ey1`4k1wcxe-p)8Ns@E}Nd+mc8E?=wAGa>pn0|Z% zZUWfGI1c|usq5h;msb)bq{;B%($A7Z16mG_@;H#1Q&9P~3Bnm0eD4#PPc3{16_;RS zA_iIW$M+%^I^*|(MqCHTc9K$l3x$em8yeTX(#z|g=>f~3K8CW zDQfa^Qm@)l-AC4>8n(i`6;|skT_fvkB4Fh~-B7(#r~NX_jewJsSz4{E*RKJtsjmR` zW*5d%HTa^UmR@Y(+26$db+EW7V?1{cmnh-d)F>I+wr@GV5A=Qg=4JKdtRZrqKW|y! z;_I~4+0E70&GV)iaPsuO%6Ra`8eyFFE(w&%#pxarlm_90iamSH&K4#f;`$~430$o^ zx?^5Md5<55tKFj0EYXm2t~@{hy6y|E93E@mnRTV3GLf`%<@%AdiAX@+Y9F2l)Cjup z8+MTYwM>ZUcYFH!(C+|y{}}n^m_B$yX=CAq%Pwas0T z-lwSP(oHl-Dk13>?H z;gjIT0meZ5(Vpa`ndg#}->kW3sj`q=Ck%LuV#A_J`ASyum{SKo{t3z6o z3)oUNgUbJoLilF*IX7_^ru}+?N%QoRzw`hSmi;%DyQI8ZQw3>^d9i4hV7jO>BTFR# z)UW9sx-A~C`CGbEqv1_-R{d^E6#HQ45{I8*r(g&els=gG_P$|;_~)`}^9KQEGvc@khABjX4zo$Hw{W$+l@l}|X`r=M3)UYa z5Me;?FzW_Rn{-s;a<5;`L;%_RC%nyuCMk`jM}>FHE#5UQv5n<_u=P&Cfj>{!?hovE73p1Y};?tc37VA?weNBwPkHOjsv z$^%hsPtj;*L-Z!{eigcs^rwc&%|V;!KJk_lItX675?6F@y8>KgSDN>wYq)TP$(7({ z`i*Gb)L*9lQ+7l9EQL+)M`983V!>$}jxx%e!clnp^ZzF4(}7WVvQLUlV(32RW)mTJ z5&~;7-XW=oLMK#-7xJF{H2*uv&Ury!`YUg#U$CG^0F)d#;Ie4;u2#V(D3eTloJC(> zkaAVYHML&4NYp9ylA!0Hr9e5v&wSEjoFfx6jdanRb@Z21C^qNu$ac6eMPtIly#Iz=?^TRR8>Grjr ztpX4l)fZqAvO|7>Ttfa(gYo{iwm-Yl?6d7ujaEzc2W-RAJ{_9U8o-i zf4EgjLH+qr=6{uQT0z@5!e7@YU+n2w4*o__uz{OuDS1VL3^Y6MqL-}rUAxOBWFP1Bn7Qyf%7@q_(m@Wub)&*q) zpU&Peii(wztlNh3fo2XtyjwpB$Rd=>3)Y>G6r9?DKSDYLYqLENDS5|3x{dXMIGxVx zB|U|pc@}79Mng&nE2inKXzopw9%|8pBuW4c8m(N9!+rkw&-Ju2%>q1&bR>CPVL~`yDsJR#)hP(eb|2;d=EZybyWym4UtmXHBQBA@0qQbhetgZ?uTIuq-MojAeMvyBWjjId zB>~>w){N8%UNB9%j-I$^}K?Pjq5K8!fblp(+=8MgJ(*G9GBmk z!|G?6fh6NQ%>nPhssS;P4Q<`vJpNezM$|6$8m` zktAHvdifCv65+EI?;G!Ibin`?9(DF9&CD?CUhbIj_R<+yY-7ow@p*5AMffRc+lleD zaVlF+=X%Uv((Ex!GOZ8ZPc=8*%66N_)xR_|boIxp(o*78jv5s)2MywE3y_R619ylL zJD%z!NZ!EWOxJP{971P7KOUXFry1qReN2a$}=A>{A6biVKjqP6Z?=Po! zGcolI_OnJ#(Q$e3^Vy(&bx5Uo+nzp|!eifNmtU5}s-%tHrB;WIw}9a#>W;aVJ3Eq& zMCQ3dkk*Q$sAmJd!!e6@;+mJP%jLwQ!KoA)L^E%*PLPWKmK}U=FH40~44UNeSWs#M zC*?kCdGI{k(Bc!zojI3#_0pAnB8u$9SPKOcR~;(|Wl*38;As zpvT{y0NqY8k)3g-5tN{2VwkjhHGAfE(6md(Na3$koI>E6(NoC>3lMWeES9gQ&XG%# zIEXmn`+Sol!)Kr6=Kn}~puD!SQTV8crb>O$cZU&BSaMq~K)5YCw9<|FncV9q@YdS9 zF+36y`QRudt#H8NV}4fl4((M=ztrAUc_gR*ay2jlzREGp<#k-dAyLU_`L?%7p*JrP zj=HZX_zBeYc(a@2ADHd;?nc7(#|-tEx)3VdQIf&Y#hQW(?Wf5IG|1f?^!wo&pHu&& z8+#+M{s#5f>Dj&O{`&e)fXx3QiHf9POHP=jvo8y3HyWiIOP=&-`iD<*UX&&;-d@^` z5RYJi8Ci=(7=;=(g2@=!KS^SDZbj5>b`exPRT@a|m<9W$2uRy5Kcr!1^ZGYzQ4(9y zC&WDq$cxA|;Xs=AO0lv+`J&JZ2RPz7`(*qhBWwjD85uHC>W#UkA~lD1>ZNJuKH*jm zapMN*M@NaE`JIQ(uvZBLiYrII5LcoX^+g04$xEvAVpbgXvkFN)<%5Rk!=Hm&92&z! zJp%2Cv9pOCh`RZty9<9UH-9=UEW`0~N##ai?x_tut3#)-C?k|1jqRC8huBmMju zO{7?g?J|pAl!>l*VcDqdYCGpiD1=S=6aD{%7_a)&N##Y5;YkDpxy5@jdD`YIpf$p-eNCAPs>)^5H2d z81QY#j6d&3HRflM5&z~APYK$<*<(E4o_ui5qzy{-ma0Y#hJP~4XpG$qcF+Ww2Ly{H zdjH}Sm${;p!`!FFY6)vw6hzOLw4B%#MRx%i7mzYm%Ha5zkTA@!H1X-9Is6RB%8ofS z#uuf7LLpd4cdEvV5>cLa!YOu~3K>4^?cGzE4>*KeEy+US4=h+?QoV<@cN>4owrio> zkkz(3rat`C$a`=iwz|{+`C_Vw1={(WE~>tt*|jzt@nhOXwI3I9qowV1xy=4?vP|>(i{*rO&2!^-@b<;jN$|b$OtNVQf7 z`ypz%Q~ZuW+j6N=;ZTh{0-Tqn->~d%r3`ghx-Gv{NOp~y>(^Q0 zNyi^maBQx%sIER7;w*#=N$(xe1*siEC|q=BNk-`DBQeW%tNQ|9-K+6dSkN zI8|VswjMs@tVbr_Cg_@{nkYh{I8*sI`q@)G?Y)cr=MwoGIAoBE;5wVt-f#eDb&dc8P=XT*Mw zwJ`7H-VhpW4?3RuT!W4r?7@Bzq42B?_wkM~!oILT?}cs?bOWJ|6E3Q-D;f>EC_~n=W6ec8sRS)~F4@r~F$gTJL9HU*wn@u& z7WpYiXZza9m6nFbn)T%zFPbN381b|LSNN#?aV?BXkWydAZLN)CKMOXU1QGm?D+U*K zg4Rdygv4uJ!5rc3Nm`On+M6e{r$;3(ai+`nu!=!QXUs*~#sttn03xg@?(o6@|TTJdDT4rbqjLc_#TiEJT%<`OBlT zH##^i>3&f26axBBIZ8WEAJxiJ6fqr3h2U8&++}WITIWw%24x|(F9xc>i@7vdLReG` zh*`%yYmo%)Z>6ThEMOI40&ZO@UoE)c`JjxrKr)v-_uqd}*o2)gy(>7!C*;vu!ArJ2 z&UaG)$p1S@6cfUephy9ErtN9@->w648JPa(X7rz}(f?gY{~zV6rLCd4sWXGB3iOYk z|C=cKFDglq5j735jtc_#UuZx|RJ)Zg=nv7TpFbG>|3aNjZA^__EbYwcjO}e5?42!L zOzCV*U7Re9=^gCM|L?*7JKWU+_6G>q6UdJr|ButsqcI6@d=U~6htLfV~@cxrZw?eWKN_<>&F)o;uE#&>${<+_reoBzb3`>?j~_ZR7j{VEiFk&RH*MrT&V3JA|~QNPIz^^?ep(|$fGPjj7t;^AZ@P^jD)H+8 z>Dk$Rg=5ERptdB8E&Q5sf!{|#^Xsxw_T%Ydtoj8)&(T5_a#C7Y+UI2DM6enMBrV8O zE|~#qBDWB%8|p8s-;1jkn(V^Cv&5;8bj$~^xA3Klf98M@za4Bw!h&ZLsH zfc}fwP4VAdU;}#Go<~f)ObBFnH>%;ppqQV38 zsIFoYG-`5ow$>f_;SWfi`|s0Gg+Y0)m6n^y1K+V6vdYOAv(B(oG8AUI!4t+G^7Z#fy-gx-F81?W%^Z(4{sq_6d{vI z#I?ijW6e)%@~K{kl*2F)&J=lI)&Qy`{L5oF&n-9cO9#EDFG@6%2DU4a-YVwUj_f?H z_vS2DZxe*P-Y`>Y-_j5kTO^1$s8LeAqgo5JQ6O+KZl6H1VUd>|lhMy_PgrwIX__yB z{)WC1NcAPbgq+`GruzP~kLo2KlbZI0-<9lfwk0L`+kJy!hwgEQ&y@W-!4x`X&4tjR zpT(gq50H#jtY)}6(raJtN)$JYzQ}g} zQ`|mM_+N?>8xq&uKshl4yYRZ*CuXimzKJZpKq4?v5V@?B7d6%?*(*LyTF4D zFjNO|fEe`9)q~o4a3a2HHnQ-Q`B33~8|}`J+xLDA|HO5MR<7zDp786N||w%R@*8K!7F2#ivx> zw`sHZyNiwtkS0xwW(%*KpDENH^9LvCZA#nA-hegn zeOA1<3xa@Rm7j;>OaMN(_&jGuPEMiv=Kw`?avbeG3=_7Kd-*l0WES7Ah=_~3aT2>| z#MJxP^m9WW!m_&z*H68wWE*-t9}9=rg;Z1yui=w0u2vGWRM2)LNM=-Ez)^diBCvsA zW3WySj4>UG{NRtcW2eU}V{)Gf&zqu_LccbPE&dc@j$A9l*!9aqjpCn6plQL72#sl-Tw;CS6fM-Dy0 zTcmQ=vGZHzn}!BZ9b3M57Di@^ z281SiJ_l`uJ2zWV%O+dd+t^!v0oVJ5-x|b~^i*=HZZjmn@K_7DQAoMWN{un^tzN+e zI3J8u-{L^-(8yp^FKmC(uk9gqA|8?rKQIyw5}E_VMuHa4A=lxLTSf3NK1e~&a(`>> z-NCiYzpiKEkYA0-j;iWCex6Xxk z*LgRHW*xUlw@a3_>!WLnsn=_)w$-+Wdd_-mMOihp;XLf;8nmWwZB6SJcKsRzdpz!y z-sWZTKc)dH;e848v34bmP2jOANofKEA{vUn&Almr-|kB-x>p*THfbgQ@m~CzxmMb2 zV7OYFZ9851?q)R!wzzWsU@opMfX`fK5RhJ5*tL$By8AMlF-f}{+W2z>ZVtLti*r+) zrpl~$#4yh=n=@yI`1+lTREkdB;j56NCAe(_Taw+;LCLTRi5!0ChNaA^d!KCBoYzQfjNk6bY{NWFgSMZI8|FUQ11ok1;sdrzy zYPvFvxVJXGsb4AQ_$5ziJv<1OubhGgAr%2A0s5g=FTGj&Q~~e-HYJDdRq;i&HM!Sg z4ilYe&iXrN0X=g*1*texqFI@l16i=2Soj%$@P2>htAa$Yu60d%PnY4qm~oX&%w8$n z2zD5B7$l@{G~u$5LtGvbNS1E!nErUr3=N~ItDmP&Gk$oG6mMB)(Ix3j;pQllQI{7o zjM=8ze`Nfosth(fEpdi-#&6@zywPw0@+~ikES=IzPpn#?$*0N|%Z=Cb6rQDXzO^1$ zW#6R7dfKzuTWsWH>S2{-6exe)<#`B)#W9ZJo!t zS?z(>_!r2i4w4$mf^XNr1zW{X<}#RtR?@()KFW4s@NDPkPX*yUB|l1eXI_$llV1ON+mce@A6$P&syn*d4Jv5a?#~xkH?Q z`DYpNuOOaDoF&dg**Luq&<(`W&=C}TI?_K}G+H-3`RW?=a!DDA>)v#fq5}12jWrj0 zDwRF7%ysi?a^tx(6yOR%wgzZIUj>qOwD=IVa_(9D+-{m$<+cuH5d1 z%^IalGbWVh)c|Los#M2}+nIajJNL&t`2uw`bq!k4REHee`Uib3BpQo_;Oj5XC-deE znyUjf7_k)p&el%ph8OGm2J?yxoAhj(Zdvr{t(plz5!^HQc9Q3ZlXZk^Je#RRRrRzl zS{g?AceQRsYt2;klf60j*&D#wb()Wf#Z;YDb+FstyV!xeJ$VuA&66*ha=35#A zCN(8lAwPC6pF!IZqd|vxl|dJY%-N;_k1{+CIZQrEq)ZVl3jJ+E2-g#y@8$4(M@3lK zr{o){u|L7`?r~>FUURXL0g74Ag#3{&SMGeDD(4_{xpWweS5~9T05>2>kdWBGfwivX zr4f?ve)4I92P^Z7?i4E_5qiJ2mu>A-^9jjYqF%?kZoa{NGC5d=py`~%hU zQ2iuVfL8%{F$Cph-q1-9!oV9CspMhEiM7Zm;Z&JWn-IrS(Vwb+KVO#v#`>_!#vO}q zn)gIMO{et66MYXfU!g&}98yL*$Rp;IV{>zXyKpSRp5`c}Ci?z!HLJN6VW zwtNv9G6hbf8s#MM{TRDuYGd(hdvhrF0q-fh1# z!4fV7;r6dLYSuP3DCwd~DS0_54dngVL@fBO^O z?c?$KpTb`cxad=*yz?N9hVM6|H+QxrB9A*CdfTAFsW>I`jJlVOrcqK4{qtZ5dWQr5 zs(G=Xq$ErpLx2KOrij(z-5ql_dRW{XoY6e*4_Dm3 zzu&xxSpfM8Aa`v0Xr(zG$E1gEV7l))@C%6c<@vnQ2iMp3%vd{@&|D+349)tT9P}{+ zYxw-{s_frY(fU8XoTS{i2LZkEFMa*;J)>jVGlMYRE9&c{$y6A-c-$jkfj!EqmA^>Z367l;cajwNOw z%fF6jZ`n<)=67811;f_W(7>F+)D^@L^$z^HT%aJ)F!*d}f06$a4-6BC;$v8C?8R(f zE-mRyEq&Q|gP|$R)X-Z$&?DANCX?*8-HbhJ@YM$PJPX-p8Fhm~2W{EJ)DJbX!cY(M z0e3rmor^kRztoRlo$Xw7Jb}Kr6g^x1SFqeb9_-B1u`s@$%%1{SPdOx8duWqnqP;~1 zRC!tp+kTb)=0jOCiAu)^=vbUGytumXojil(>KIp6Z0Ma5qGHeD75(Tpgu5-6(J6Y5 z6+n;?;Vlw4d?x#@I|9M@&r;lpCm50yAPg}>I|_+Jq^Ae@uv&rf8C0iNQTU!NNBmUV zR|cpja@1PkS$X}ul@5XVjTD+jP}|`9YGU$F0&(^l&n^y4O~phQefNt}R?hujbiZGz z#qYIWKUIMG^k0%H)kjL9F0U8%EQwjNaX=euvI}0OXHp>FGc7aw=EIOo2lz$pt zjbT|9m}g?litK=aQ61UbAYzu1zjTXyjemc8uA~ORA=7krmbI{XOwr9aAYZ50<=&L! zd?HF9bxtokNxulEX%F>}Bu*heo2;%loy5?LB7mfJ;y{<3DVfXGaK)_>oc2UBuz7ia zNuf_`obXd%E#wz9XGP>4lZR$*tCZc`LJ^YATWOm8TJV8ec!v%t@M;Q7<(O=%?Z>O!#FxFPWk6A|7122 z?*(;++?S^u@-HVd+%89YY_O4+c(aeq6kbQToeQP*xn>Rj0x@pLrBmkaXtA(3xW7_E z;cCASsQw5F!Rr=T@`C$*AJtpPwA1`|vbgw2;(oCng!t!UUA`zCP=hsSFO9nIGh!i8 zru_|y9Q}00#y#QNOYRlnPgaM;Q4LK$MFl@XEKR5 zPK{}gxHgIeMFUNQS9DZ>{JU+sg1y|ZYw4PDey@JH!9{nYXuM5^WRfkvM6KY_+<&EW zhlei^L=qBqN;*9QNGhUwL{`Tz|BKoqnt@i}|0B)3h^7z8(hLtzux$E~-T9p>eEvL6 zLG3fUOhlnUd!Hm0w-sv5I~U6wIW`YkXkqI?CFEJO0AWJggJ3A^Mk7wUjz`Hio33NVU3g9lQ?Mwa7d#45f5J%ZPsgi z8$Sn|A$9c!NE_LRR>>Z*hFRB(;%SoRB`cfE5Pj*8i4#Edb>7C?agf}*AAN4%DRsY| zxKr8?ztybL^@UTmz3n_$ILX`D`|r0<_o+Gp_v6cY_vUf{4DC`QCS7a7m}@630#!2r zQR9co#A&FS5_>0d6Yah@rK5!d3{gz)gLC+*dSD|AP^>Y`*}w?<_6AWxBh^0}J@65` zBF$0a$ms}9WtG5clG z+#OJ_>V0};vM2*gv{R5*n2W;0=qE3{v?g?AhXOWSuswX%K{k0x>BA#G{2;j@;|gc5 ztRA-zSuxb3j6Ow-U7AbabIjt)2lf#9rmxaot8d-B3K|mrC76*D@Vz5ygUE!wLnltw z0#G|_^c6d`l7|7Bn36&TC)lM_r3O@LA%1jVolUDF8yKbY!9Th|VPvjAG+nM|7B$ns zWkBqfxBKNTcG`ZH%{LwzpZ$LMvROt)a*+FH|Dqg>Ycvo3<+28snWKXJFFFl|Ol=j8 zV&VZCR;?=0OL-n{QAvoPeGSFCdAJl;93VSHLZ1{QxqPEd{zHV1)jN(+=62Nfgt2+j zP=d;W6ll{1SJWi0pGDC(pD-L((m|osIN8XcXH+L4J=ex(Sb2rkx$U;M(TQomwtXL4 zr8|j|{`%scN9CLr;%14UiW39wlR8XH`und}Gxfn_@h`1xLtpHe>yNU2e)D%OB_LnE zyeZ7y!1~$(*ino|mUjtBd;iJw3 zwEWf$Bb;H?P(P1WOuVJQ9R`FHNKOU}|B5Z3-&7mDR!ND7n4a z9K?JDk?kyex+)~<^;N)y9yfGxa<56eJ()AXxs@H14;9-lf5uLnsUnXc5W>@J;0s7D zw|ZRqDV;ZRi4)>&&W6+oAp>19-$xz^T9L!o!L=VbZiOZfoyS?9@~vxjy2Hgo=W|-_ zHa8*p#PM`ST^t2F?(XvLd;B+Vc5J(|1?f$4_tOLCulSMK)dRT1PaRqi_lVziqpK-C=M-ZRVluF*86) zOH>qmd#MS%@?ipF5I~zO;xD^ENdO@i_E^0fIt^?TsOU46_`2Q}ipjkZNm0LqYtGC; ze%D&7^uDATa>WxWxDkOooao4_Q0vD&CBM+gQHRcVi2xd=B&uz)lKwp%RL>xAWim~7 zp%Yt##%0VP9RB_*a)2yFp7{*h9fMx}!#ge<3sN*pp>NGa0Pv2NJFq?0NR8v$a&l~z zckko*!4I(56+ETZ4z$eSD$rN_HtRZTdEAy5i%P%1g=MPtge-r<{!k$H1Bmtip(;Qv zwrCVwVzsq=B(k<8vO7kf_fOF_|e^-Kv0H3)}&DUCSpN62tqCGM=E=7b{N`shl` z_n3nrk$ZJ+1JZEUJ1+w#B&0l+Tjh4Ob9~Gudi~e}3BIGMn2uUsj|NV25y)^bIQQ*O zY}+U>39t)2Xvdzv{-VVMR`}a9tcY(iiHo*O@^CFn;~Z(vYoSJ`f6@!Q^TriDzM z$hMgKfL=OD_J7vwfa~b!$hh~^j8AWGCP-q^B`_(g1CqM534)WL8H@y0dU!6<4t7g5 zKgP4qoz;g0+EM*}H5bl%!v}|Q_I_mExuenE6nrLE3GP41>0h#QHnw-#g;S0Y1ySa*NdKyO*_AOT7+ww zOI-dC19+F@x8Sf7X8mSid|6KGa8GjYfqnEh^|!BSs(h>S56$3;P(fOK2D+n<>Nn%O zMU87iUfUT)FF#pBW5(*+rYTeEYN9n1gHIasp`0k*BRt9jAHMCbc5@-nR(i9j=9y4SdS z3RBzWp@Q5YvPatbf+I$7U>pcQ19yZ?=t2Vof2<|3>V*a0!x70#eY8)Wb%1s>K}=Qa zL$Ak`=qYf?&EwA`XuQA|FKv*ACB+n*WbQWbro?5V^PQ%Q(bckCeR7>v7bPb@eoj?T z1F*9wTUdCA6?#XN79nDGS4CG#LvuOM=i-U$H)>E=qtOI+ag=WOI8TFKCovv==?f9{ zTyfnMao=6KhLRtD+u_>D=7l>A7^NFEl8Z=Y+H!h_U*aNFW5vsjpbbkvl}$tc>#~h8 z=LcNwQcB?<26>mc$u;Xlq~j{EJeMWj0sP0JC}82$84N+&xM-U*&VQvZ>{qR06bllG ztP0+k8@8BS@)e4k;~oyP%m|QrC%%^5tzTuEh+8*HAw#9PEyDgS;N_%Us24`)zxcJ` z#l5$!E=gt(WUkMkED;+A=;62yvkN%1T?uC&9>*tGLWy5#Z6oRCSTmf7VB!Aj2OK&_ zyb2hYn4I+GnG&6b8wC=^eFl(nrE6^Hz~<%KXR~lyQ^>UXNFLT_jR;fb1YQ1RD^3SL zT#Fzn`?dP3V77&R^wHv7|6T?oCC<2WzKVu*5HkAYoeDoGFLH_l&vLUmO*zodbtx$D zTb$9z(6(@jGc73Nb9dC3Epwr-0O~@j_pb+jB6-TSZ66yM#J(u}A&Dq(`i$x&2}nGF z%a!yfp*4%8nD&XrmCAQcF|(WV7pJkAcn3e1ld;$0Pk$~sr!0YOC`SuKM$!PInq^|b z{jb&cA4vKIsm4XKfpBk9>jHMXq|2Dxwk$KNjpuK7e|ggHc5& znM1wgcOmb4pTGX=ERn=xyvmnuxuD2T@6H68Bp^TD{4KfSC^5R_#gac6rp3zv;k#5qqfyn$g9a&7I4%jzkTzdVXp3m zHGPah}J_aE*05EOP0Y_fRyNaq|xYX&0)A}*cw z@DKt`QIZ$=q5$UnIgpQ|uo<24brrch@!i8&7K+as_Qt4XpwtgtyQz(tT0ax@*hBMy zR}fO7w2_SYqWOjQlE~Y2@ohVN-lmJdVy=KZFuXc~OnGv*Fq2(rO#k&M_#|yhS~XMU!mbio)XyUhL-yxq5o!1#!aGv%;g?z%D_${S#qy8LT!F2I#!27R;oFO z&=j=Fvw~4KVBetQxabq9pkCHh8K#5U#l@}dwJqsav~rqZdB&Nq_*WCTfaUM6^;2o4 z*8ezj4z*)y=bnzQp^_C$PJMfb!8!K1LVnB5&kgbesbz~^hrU-n+qXs7Q8rK-{jSkfT|24TT4Lqa{z3gB{(`~H z0|OmR0M$=8E!iZK)~h^I>=s9%VixUknM==fm~4f6C@9y`vy zyMt~PIqetTBO1|mB+GnfnXa970k-m)nOv(LQ1C#q$X3yC8pK{LgG9iNX{SWy)j}WP ze)EFb{9KrEL*JE!J%75=kQ{GQx4-9q9)SX6RpQ{FR(-vXxD-#ETs4GHI9ML1#iyGU z>_b~iV`B&KRVl!T1)@23A-wjG;cLmftj>mBbKA#dBsgl6^hhCP(*W)6FKu7E|L$yo z`od_P18Uz<${LlPsTS$h-8;~}ZhD;`^R>l0;F;gexh#g78{retjdc*4=S5ONUHtfN zb50^IWw%7Y^_j^pz8@F>m4=JCmNF_cS3{b{U53o*2!;xc4|rvNTrnOCEL7MVL@_BZ zn*cTDs&a{y{n-0OC6aT=)5saE9qA3g0>+mJmWQCsjgtOk)CA?nq#|H_iU?9;M6dCbtn{zt$g2T!97 z@zy%!v>Tc{cm0uISyJDshq)6loQ9a}b?O_TNu>2GWW5f>JgcZ@9*a)HO=BGY<-pqA z;{r(zrKySut~|||Tby&KTb~5w>!Qgsv5mVu7+rfGYyaFAryt(kayqwwt-%r(@acAjv|}1Rux7|K+U0d@id65l8t-KVN5OLt4Cu`cia-@|& zS$|TTy>D;%K3#9qASraw@*Ct!=5VM z5Z*gUT!vH@=PU6xdlRX!+8L~)plLSrUihr_hS)^mCqAVu2Xp~7ag%KJy0 z9!k0KhAk|Sw3`Lg^4KEFS;pQ)Vuj}N{-?ftz2(E3Y}W1;A(6xiRY5s(vmhN`PHDgp|fO+S~GBD(*sg6eQmeK4-=E+ zCZf#OrF0%w@H+gZM@fEgkCx|o>S47gP@1&qY_ulUX(Itx%(`4p_VQZ(@z)PIdBga( zwoD54uLe~ONY~^0twL4ERU#d{l1yx7H&!^UYKMS)3!0)LGu5*4!g+GQ~I-#l8o| zp>1Ute#57MHfD)=SzLQfYMEYVXL-V-5LFc9wNCtT3w?-^6=c}9hEtU;`01p$3lvt{ zhwd8-jVpZC9lr$-CmMGGz_5|CM^i-b0R1GydrDJ4EpJdNf9(1i{an2qF+ei}j-MT* z=Y9q-2cHMkdMb1hN$MXJU5b4`iulSRnt$F3hLtEKOK1B(iq0v5FAuhtQy=wz4 z?E%fzR1xHNg+iZ#5_?FBYT~>d|1Aba3zeQo1q4@17Zzg@&Mi@nthvGV&oSB?&-W|n zR(Ree13M+6lb_Q$1e|}-{E9gf#=koynx}WUSemP7{oQ`2u&HSO* zr#zLRn8M73l39Y+gbr zD<6&Bu=UAj_v9MZn0|y?6wN2D5C}|;e?C1xtolv)Hv+_hm0@Epkru{U6QMN$DV#Ro zuVQTR)S{=)8w>H&rnAq9#li3mmF&d)DeZS@%x3nZSuZJbqa3^DJPntuiD!YH>H0_I z#lyfL0_Te9r|^wfo8f=e(j|wEp_&>M*uzUAepMQYiZ|x`F}{DhVLs7VJSG`P*hAD3Vh zA@S0*O)wN5d31G)wK5eGxBgx+sEFEOhBO1JPYS%wIqI$4Xpw}mU#m04n)1s4cb5Xv z8C!ZQd3cOUT$0UG;o>>tL+|ecmUQeseBqko`3Kkky4C(-J3+X0HL1?FOYbq;L5s@^ zbG;kMlQ-M4{ojbVg2$`Na8n(~&n-=GC2I$ze9bu0n>!n}&pnlo**jt{ni;Q}-Mc7; zTUd29=!f!OleX@kLtNKC)rLs{_Mh8cnPu}QSHWpfkcD@z_@%qH^bQ0E{cPduUbhd~ zS(jb3+jb6T7x%Xgg_h~wlXy4KnvX%v@q^3Ip{4GuTF=y+xl`V476%pZMR%hD3D7{t zEs8&972qBn4zn%}hrw{eXZ6R0mv41v>!GhP-sG4;MGjMUe1&@D#E>5-MtA&2FZ2FN z7V6(&+1pxE`8NPy$QC4@X7Iox!*Lq_u$oRQ zRN`;TWYgQ=EhZ*5dx#u@Map=`DyZjMo_KG>+JKF1Q%_c?{V?=KXK^-JV%Znrvnv!s>5Eb}Nu1kq4p=^q6P@E6-v}j+v~YZ^ z6~LWxr%XUK)5KJU+^KVfM!%PmNq1}W1hQh2yAL}zeU82UH%kGmXZ@&RLYyvh=y{el zO?PmD^~TAW3jUrduE!`Mt@ioFVpDrCCBll?D1N%=6?!<;FOD1*Le4Wa+n$#0*3LWl zx-Gjbc`3%^iy|%ZTG*nfdq#1=^M~Jjv5{OL#ba#*d*TWENbKQ|0~bFfyQpR2>=V3g z9R7Ivq4DqK*3dWrBgKYn)j?eLe)njO71TZKz{MHEs##FmX|T zUl7Q}VdF);f(c0=uwrC52~S=LQ_U`AG-q6o()CBL5Q%e+TFKVNGK5uy8^6phiq4qH zazTw%tXpIlnr$V;nGppVD|He8e_Pz zi;7O{Z4%bF;Wn&dt+BL+i4pcvS=a97?evRfcafZ`)?E*T6278ee|Ef)J9Ol5UR_Fi zKw(=vpEYc*WGr86eBuK`~h-a_scFhef1M={%wbNajY~iWu^j7Dr zZaWuF9UB0VfhxU1JIdcb1nlVoNr~V`aj8`xfD#6)v*enMC$B5v!zkG*aSk&|CSyDs zGtfuZeYfEDgr<^vv?wvJke~575IPTsWXCMf+dHhQ$fbg47wt*QUagkQ^#ACKrV6)E zh~mD+g}ZR&|DGG4XonfjyEJ~vBD;aZ&&XCpIwo{pU5xQn}2*~KRDxiM@Wov!+wA6zG=&rjGdN4U$>;GorE6aWOOVDWN{bqOt9|tC_I?7 zUF`!aARU!$L9Pcnh2TA1%AmezZ6yW4+G2JC%FXJZueL2SZY0S2O0v*vQ?V1QuO`}d zf7~#5)rR64zW?Tgrmw~zz6PYxxR4(X?|5sbWW7aR~SVtSK{G)=zV@aX5 zG-+N_Juk)rua(truUeKJqamO}lwBM5H^&TEnEF#M2GNJ&qfdRiU^Sh$SW6Za9V6%7 zK@(k}oDfnY_aE&yUIu9=SrKAk>CD~SHYC7ao5w)sMc-SGZNcElj1GJzTmAko8mtmY zKIVK;?xe#Q1(RsDRh->nPE6_=<9oPoOl#ND!=LMs??>q55zh>_!#DY+j}VryEP;1|vyeIwtd+$PzC8)6)eL zwosnjM#8gO-iPHFA+T^c>jwt?I9QOzb$Xr;>e*``O#fvx_!6$ zZso*dmjb*h<6cU!cb&I6I;20>Hn>FyIoa><4+tm^^SAs&tcD~eBf5|(H%X`9*QXS4Tn_N)K+kYWbIJG|+)W6-p!pDetWU|yyn$uLyf?CKN zLx7ZBX9X;B1lZjiEt~>xdijZ2-7#jqD#Z$W!1j#cGbt}f%Fr0G`<@R^or`A{k&`MR5J#q(v~oqusM zrXzQaZoRZQN&lM>qpy=Y5AFk_rpO#8y|&=(h707I;M0!4&MNPHOfehv8aas+4zq9` zGbm3M1kaD%`ZeG@)9;LjBOKI-`H71h0uGYt0PHtKX^2IyFGe z(5V7osI-juOz}l|H{cigw#Ov5pMRn;A1|GozOD>194nI48(9|ly<+CT=!b_^Nd@6= z2(`bPcIYq)`3#|LUs(}~JqCB&+nFeO90Z`=sz}2$)2q22Rx#7}&!8l*yik!0h-A(B znsYXGcgQZ{b9rQ-RQE1beJ%N6PNN!wV&bdFkd2{*=Fy%n%gq)LfLOP|zJF!yq`b`O zW-P4D7Z2PiN zlO4#O!dXyHBT*tlcWS7L7Wq(8OcM-U>7Y9^}#%q#KeO}WHBcltxs`km1`uw0*a2;}L zCPq`iyj)7sLDl)~`l&hPFb&Z_Cdc9uoc=2$e5??pfRJI8E|I~S9eeMZgg2f z1|zY`8O=T?20Epdk5dGl;t{!@I-c|IwsL2clPsJHeY2*!zK|}~EPr~tfBgCPBI!@q zA7BC1L(WqNt)r*ana&qIl_-oXrS7dTpq)C{yoqS{i9 z;`$~V-IZFKq)64~f`1!@Chx&SJE1(Y5(4wLvrgf^Jg0~7gXOaco2h&xVm9p_B7LXX zjm!PTAYauhh07tvj~~gfR&8e6bHt92zs*|llX}c_EQIK*nx@ibOz^x8MISr<9|8us|!P!WbyGAZ~ z<8{wBCrXtmz^^pR+4SX%k+6T0C5adLZ^BH5+f=8Zj#2K=NnlWj+1QfaZRNmi9_lv8 zqgP6m8bt=b6XDNuYQwWCG-FQl3|c&Aum#u?O}bV3x_=atq|m$GgdYMirFdSHt~FVc z7uP;--0X)Z#``tLm)IZg&A8}PySHAodZm$l9*_QtSv*Dt(yf-!V6yr^QhVwA)(+Ok z!)gCro$d!*+}GEaKIKS1h0{FR*-TNsJbebeS^u_-FRpMPKEGc}zQ%Hd-lvWw7G6fF znyb4ZUVq`XmlZf-HD=NOy{q5!CQ$PDa77Y&jfA_r=o?cNOYs{rr`9F4c4IG%yCco^ zF}F|m8ljxVe|E zIWL{THC+U#cJIYP zF5XdulJgY1&4NlWkC7Pn9nBah&Sb(C{==U_w^wVb*P6#h{fK{yd>#~12ps{n2icsSjOKq% z1FC4Pd0%E^PwhluC$br_T7WX+c4baae}odtr;Zoh7Mfi@Z*6W$q8@{EE!{j~YChZJ zYX0$w$Y8WUo)vc7!ax!Zdb6i}EtVvq<9}0D?7zBh1wY%4IM+7_IZyX{|67@MRE+mb zx0%;(=`=}y)SAG5TTdPNFiiFjdJ{V4?qKu#!0&R2{qB{>!}k~zRIp=>w(#>msDWK% z6YrRr(@KW@!Jh9$Vqo)!{T`Bzcfxue z`~e;Rd*c6|_`fIq?}`6=;{TrbzketG|LchZ9Z4Jkhg+e-o)=QoVqVWxfFkFKJa4K0 zf=Uiq;(4El=oQxG?u+P}hEnq!Q#|X4{~e?=g|_l!`RbyVtI-V{ctP&+iA?0PRK$IG z?ng3zcfZoM+}$E{qZR=?Kc^OcNMmx}$wZb94Mfzr>*{I$e;ym?Dj)uD&wm2=Pp4>q zV0`!O&?(_HYW7lD?GAkxoF-|ziTB~}kl_*0a5OYx?wsLWOR~BP0iDfvlYb5$J9de& z$8M|Wd|{VV@(;ukrMQZD$9@r0GcZJk2VBB;pb(2uS+u@a7BW6Lc`pR!ed*JYW>}phSb&|b8ao*8Oo5k?3zh_J)S#A$}=WgZs;-Bk{ z=bhUgT2#*ZUjp|G`7tO*+UPelkM~yP>v9IL{-zo3Jx79b#h%~L-(?15AlQyUo%DB? z#yxW68YCqXIR3eM(toyKBWGo2?sEBSh!zgy2usQN<~jeIVaqQA0;+Kz5d6)dLqTLj z-LC9+5b@7Nv_dZ+0J>|4dhoYc2gISES}@*o|L40{j<`>|d?mxm7xz~@icN^zMQ7*p zqwwF5W!+oaDJDAoJu7wrGPZyF-HGfsV#<#JDhnQqr-L!9)qlPd?MBI|_g2bZ)KUuL z{beyb(Eo-Q%-?NsLbbr>0_oMwfZkhu(MdBI);Y)B z95TJ?FHPhA`+s|eDM+Gd$qYvPa|xxzyZb9fUN5;ud>>!?DAS-=!s{^SGM-)-oTNtY zW{^Kq2@D3>Xs{gX+S)I+F&BaiW2;Ho`uFF~=^Kw)id5-i#+l;e;(KC@`ng&^r`gZe zsZfgno^;*axLfH(E$pnoX3)qs!&#V+7kYzV)Bd^9k$>=ZW`4Bnqos~zVXUd&Nexs; zE=S<->t&nav)E{umsq=qZ$(bE>_kI%B81G2j+Kt*v4KMM7)Pse{W2O1(p6GJg)PSI zON6-_L#^=9l3kzsNNaj8Wh{QC-wd89Ir>G}iz%5=?^Ad$E%qNhM zgP$=Z3K+u$W|Lo+i&<~|IYF-&JpQdkak@1;z-8KiA`=hE)`(lSt_>Lc23u|Kwh*C$ zvxy(3U4Fx7(wnj&u~2y~);^$W)a;~)NsvlgX@4w}L{PRPzxv_m3cud<^r23jd5P&@ zyv(A5?${FS5D*aH;%Ow2NrRkU6RH0Fcb!H4xvqV%K&>F;*PBQSqOhOrnz%cp+pE(# zNbXE9Sm!>*meTGT6#pz;Z%3VV<5*J#C>jR=0H1+?B1!gCX>#1TP1Pl57rn%d?ua^J z|9@-!So%l@O+WVu5Nx^?PPoxTR1=&DCgtZ>Xn|@fPYpQfLq@&W4 zJed_UoS(k7&j&fL9P@;dGA90rBNgea0l}1;iUv+VI(?#{e@))JLGx;vgk9nhWAA{W3@oR^S zswCVW1uhnXGE+<}pYiK$)|gugO0@8^buaiVH?Z(Mmjt%EPKZ{lEY??9L<>$Rz;P>- zlPlHTZ(M$qKJi9^1p${VD1rvW~RGqDW5nXKM3*M3Ol^)1K0tYJac|y;3cG z&6Xl)@iAS{WkjDbLq%cjI$~GDRi{y}b0n?qAZ~)fJ#}cZ`bRjC(&vXg3<>=tMBuIZ zHF6Qp3%Q3C7kcFsfu?r4R4sKxIiOUI2rt{&`Gm9TEm-)rszT@N#*YQK*~BGP)lcko zn~pi_-Z*;h?r|F{q+9YuVt<5h^=k_M92H>Spc&FQ5){3y(sZ0u7-FZ0C)i_!o}c-f zZoZ$8(O8)tt#lI!wNW3%R~WM%%zerl-Pt{qoDjm5r}&s7qPR&`C7nyMlAPt>1mKrT z?)F5i)B2O=V(jTAN7;NDgNb#cgSxY*?^UTV*UM2WHHHLt)*ZR^o`0CV-O@wb-Y1P_ zYl*G*7aCN^bAg(!fO`8o)j*gqaRgWZ@UxUWeeug)sRn?MO2kaldvaehqC~N2ym!y3 zEWjl~fx#4MUt^!>JmI=1d2F4Lr1@e9C{*G_2o|pp`7{f1Q*={@CqE&P3_>}Gg zMc0>P--~H0s1A4fm4DNm$CWzo^+A28`8{1Ahisy&a_(ocHqH7y8P68CIP&WuA)2pf zL(<3K=s20LI+ESG`?{TjvUoPkwBfortWY$Z{!35>`Yq8!4MXoHj?UVv4aB z!;+qkFoB>Azkjaes18?tOd@{CB@R6ZVUtn1)FNA>a3pI-F-;O(vCzOOb|uX8pbGcH|cY%W~WQ+{(fB5+ma<T@Rqus&28S?S{>83V6`>v+_AccD3ya1;F@T}h9PWq`-xh>hi?h~m`Z_k={n`a3Qb z1QYdvnnxKN4hegF#e2G}Ua=p{H8%8Txa4%{vUZD|k7_CEO_#H9awh!5c+mzl1E$D3 zr=}krgI`rUe|3>w-Q{yg%eL-Dsdihq%BL{hFECKec9kg7Ka<{?ybIdlAqTkIlgA?5 z4}UR2$>^GVN6WL@qtDhDRZQ1Ez?^Ts(Q!FaLfWk-G&^)4`RA_o#1?D`2JpyFn2|hJ z`|{OjLn9G8qXu()$68c6l0!kV122kQjd)&mYlMGwmT{R*lbP76@F%mt=%P~uf^^|H z;WjUPm$Qz{vo8AW%LYEr1d4=_I8N4X&wsT>P(7*9p`^-+}OzSyAyC@yM02 zD)r@UmaGAdf>q*NNXwHS9}jG+%@mgMc=vlQ)x1E1MD7D68BGV`r8ggg51XiKzHhsg zBqXkWT&!Y8?0_qs&lk~Ml;WN$zon_?y|v1k^gMwQmR5H<)vgBy-I_Y)LD0yKt5c&|KnRO&rF`E55F^j? zhvy!P6R1et(@3nHttzQ=+BHmRwJeKg))6mGWViMp6LgQX+}JXQsK*Km7=4&4+ZmF^ zU5~%yS*(L&)x5it&PDck-Cai72JAlfC4S+ikmJ6PIo`-{4Wm-fE*;}Q9DmYl$Ym7g z=Si*1pP5aPPn4@FC_MCX(igIfHY}^Ph6Yg6$o;r^F0V(b2O;S`AIoU|$s-oYTk_V! z7Ll`4=N8dME#_u;{fJAi=6Q`~U%}%5@nK}_h_w*T{2Lh)AB+~!WlGg=>(O@=u)t-x zsFSy*dlAtsCE2N~!T&xgihtWdTiV&I|1F`FhHfMhZ<&J9Cq)k$oEE+a`mX=bL89|C z?qapYqxda9Wnio2{O)J{CT9ct`4+mQip6_n2c?6Js?{W!Zd&qQNH2H`G*;PA>HZ4ToAK!8JP?ksv7FIOQc9-13u7=YN0>XF+1!p!JMj z-M-*6HE{WhGCf{zqWGKbD5hJSwmn|ZmWM5CQE8~RlsD=UH<0gspHx?)867i znYV2L>P+Vj%X%fBqkl46hAEA}%f%ubnNHoAl1SufyFHTsv4o9M5+zU^14(qWAHVaI z6YE>_jf~->>h=YmBP|o>b?cWlig=BQzCrSM z&Utd9=TP3wl0}PKgR(Vk6=k2oVW42OR28}40SR2=T*9VbdVleKOf&4Q`iWBdi_9Nk z6p92?B2?=E)u+OCsgu#RDf(FDWj+FJ!R8%}@AjQlpERK$TLU36`(d-`-L@Sw2-qIq|>6wYdd43`HbzY zx+TFE>Bt8o<&nPY7i47bKIHoO`7x*ry^0l*eJY#Ca>V|p zfi5b>;-g!++EviiNDopAoM}4K_>cw^@p;NT1$hzllOb$95gE2L`q4$EHi#2Yi!dM& zc>I-v83){M$wjWlLKQvE`WPXFkQgQ$3F%+nW}2{L-~C3ot$!PCTNtD)tG}!JKIrWSH!qj0 z^k$=<%ZlMzA81rm%{6vnQWFODXVVD#(vzc2in~_Wm(jZqJThD~;q4uy>EPq7#0!LBEkRNb)f2R7zG8%no5{e9F20b%n z?z1bibidnZJfftF=o|5XEF?`TSxq8n5yc5PeUtKBNzmmn&C zX*f1nW*lVCb8#k0Jm90R{(qw4VJdY0iPw3q&gd6=d@C2Lu|sdG2upJtyAXmAJ9*B) zRY=Mux0+k;YS%jE*hBmek<s_v_PD`MOG@^3`RE_Ah`j{;C47Be>3fg z`O__gD)H4pFgh0D?3C{csC`C9_DRQ#on-cVSUzaNZv8{fM*|nr-G8%;r} zC4Qv2I7;Bf9c;(QrEs3#(5YQ)N-or)0|uzM*`ybFL+OpH<2TU38}L+_h5+RwKl=CS zt>5Zq3!Ze7#>OeLSfeN07}m@4+AmsH<0P?p<>=dS;>cp5*l2>ax_zJZ`x~(~I&a21 zn<>YxGE%Zr@Y*G8Fn?=?VoJzjVLHq-8D#pHy~m#w8Nw*Db_eeXRQ30X-SF-r=q8He zlYmL(1Fr+B>%bdtXnH=2UB^<+RQEGI-F13Xt;DQVmf%Lgt9b3}weKF!x9{4yGO+AB z^D)Bga=E^M|`F?$1KcB&z;pYcz>wl^DkG|~tId5qTsZ>)h zIaFD6-~AA&&1C)k4i?`RHKkbtcja-_N}W_5Jr`F^vr~VQD|G=&wA3)*>mPM~@W_QT zp4n;#ApZTSsVg}Qew$n#e!ri4uvk_-UIADhvF+DZ^Hsykx#u=s&Y{HXKW*mZhbRS9 zZRxm>wDm(pOMf5Z%fa~P-qO?{(1fi+$4CL+d6L$gw1*OL+>~X3Z|~lT1stf|h?0t~ z76rFJ{1YoS*{UQuQuDR2@SloQP9HsOsI94aq2t>+lVTU>f3r8!tnyR^uFz&wqkhK7 z@R5|t0nH-~IQ4pGy!3u1Sf=i(+CWAr{b|WmX|=?#pnsS{=KSb)3=Gm9;1uSRIek#| zWVOvXHSFzd($ne_=aB#lD(B6CcR|iPJyU6@+DTMHDBuGs_LwzX(s&Ny!ntKTKT#^~ z5|>@Yp+rtbQkBe@z9kDofZpU)tHjV30}a7CL9qnYBG8a!K=ow90LGUn{~NBECOg>+ zoD+njbbr6gMt4{LMt>u2ly*QhuXwe?u8aLrCG$MBABel$NO8zPH)(nE>s>nRAMD}* zE?4HIzY&}0$Sx=2IYt6t+{>`_`(1ms_-~c%_(vs%W(n}I4eQ^9?aj3tDWsM*9GT)R zwu{O0+Rd0(quD&P7)(rG%=$LsU$$Twl*;4D+kf2Kl&%wGF29&Z_GQqYk%ZlxOVC1d zfvm`)xtG6mU*Njo{nIT0_sPvI?~n-JDf8U6avMboOX}7uxmOMgZEC!>^Tz0%h&s4q zts<9#1MjyZ^7Yc0pA)gDs2BtW0VrTezC6r-4nO7ekO;}rS|D4=rpVGnx(b9^Fh>O@?+ zfHOwxObsC_#(jMxM=<Y&s=-)u5+h;Gw&PyZfMRt33_TniA_KY!A7#|SBsTwCp!HNQs0 zMy`U6KlvX!6_h}eSIAjS_U>4?{I!qww`zN6M5p`dpX+VN?ZP`P6bFA1Qx5qNud?#~ zZy3{C>Q(pP_L~3wk?Qv2kWFvmKYwONfcVudF&%!A|6;HvatK0Z_Jn^T_VT(q3Kil~#NoB#6 z29~ybTk3yKAliF+v|mw&(&`Ulg&SQI0(~1z>eCy45l};g3))HIe<6-A?9!zc$-5*ybg<*~z9)J=%>Ow*P|%KW zOq(HRJ}=Fq162qj9R}%Mdk5hh_c?I#$92K5-$Eb!E%gTC)qD`n8|C%Y%-n0;VU&Mv zSNXls0tp-!>u&;(%QR4arhoeScyq|>B;M2f?=KKkcVzm>S2^_)xPafkj1|&~?7Xs7 z|E*FKE!tt!p}smF2w>&}q|D^358xEQI`@<6|9b)naV|cGnL5SKR7T>*8@+DWJe9vw zTOWOa7D(u*X*{^z;s&?JNjXA;IV)1f4GLI9?uAx>9{w%K)oAiq1b=;oakV-@Y1L#O zzkfWAX@n*F@qJzS=QAyn$bCYwn0h<7+2R8N5Fe(~lRx_TOYC#*ZhaJNblHga^>T@+ z6?kJLHTxU25FygE*+@+bLrM6pCvX0n&b>wYB7y-$3CVvBg#Y~zfMT(T{<~BDrA?2J zx{VH19lrOwQ{EnEJAYEne}AjD;v|E#j2$fxxG{cvtH=Yk)_?-72h``O3uIc3k9iz4Wpzj0~!}sp`K#YMixQ_{T zycz53NX#^tg`~TFcM1%^Mt=^w<8~vreZLzEEmY=D_?2+y z{ex!302fU_VWr(4aQp&kGX=gy`*J5e`Gc$JOOfbY)Ub*6G0c)p`cJf4JCL^P!2I9Q z2*Cr(=xDS%k|uvJSdZ!LVE32Pd)e70{-7By+7by=;htUZKdk3AKKKKZO#ALHZeW5x z_)&~F{(pgO;*>>NSDHwPm&YUepZ`Q0zM@HdUSJ4v`Q8>`$^QoqDTVGH(#}qn34SQn zDy2mSQl<;&CCS%f;4sDUhFbptPq6dEqm4bPHOe*l)LXQPV9t{#;_vR@Fui`kjrj)} z^y(qD>K+?1BD`8vrTT;Yg+9IA-%|^tdv`y6z<(PGJ!8vh{S)hZ8>EWT2L0a&N?HWo z5Dr{4=Qq&ezpvo)?Ft_KU#{T46jeEuKZy zdw+S;2Aupf)ee%pl>f7Mlrm)n4nj%U>M06Q2ib(GaqZpp$|Q!^nyYa=pAOG#AO4?X zXU9^Ud&_%kB!u^4qc&AB0`YLK8m%Ky=`eORH z(C04z-8cGHnESW$ zD))WwnL4|)bk9|sv)0A05+cx)q$&_=t#@)%fY#fI9UjhitMEmgSVk3lCV#bp8tyMz zvI_Cfs4H2#pOm%DS!B|%VGSM$v}PN&j%}3wjQHMV4mis9Jx9D`(kc(HvGddM1f0mJ zs>Tj*fbIMdwVJ&4^Uu0zjICMM5 z^OF{K{XnuKDC}ZIcFc5pbSck#(AxQ%*WljH2yiA=DQ~=&AwLF|&Z1XWbI!Eq(Y6Kg zyMy~MT1}aWU1`YnnHO|3^cjsv%&e33K|SuOVC4FX_7oUG+3QR9?thlu+rM#2Anjex zcC~t&In};Iri7dnF}5ptb6;M?&kvYbv^<*o*!+D~ed=hY^|ae*(gC_k(27KlI?5dj zYioptKcZFhXM{iFYNKkJ^&lj3pYr=9tZlk+Exw4_hAel24 zO92G6M($;iYU^y#N`D)Xh?X;0v9)~P{8Hs^-WSz&+z8J#R-v1{>Y?~1i}b`CJg&dj zhr#5Ygf}GN8!}vAq986bQQlA$val)S&@6AiGsoO@)_!^UjO@iBN0Cl$>Z^<^_s<(N zz894zjYsZ6HRF+LKr~*V!;;hC_?7le}Hj;bS#bNvG=E%!tLm`Zba>UfkS48mVf0oFJ$@RN|JCsBS)pk^&|PB zG0~b5$etxi_{9HAR=%DVscR1HDZ9DOr&40p9qLmDQ(>4EvV@oD>psM?F7E)`cw=oC zCs?~@8Rd~Wn&_TF`Xe$2@b)>A>*O-#{0R@e?7nTLvs!JH@xpT+XNW=l2T~p@_t4U- zH5H1t7k_K-XH;qZuMau-a>VlLEf<*PLA#r~CNWRC`}3UW_PcNe_v<0&WtGGmBaE37 z8lzMIy{Zn3_h6#4LD!;ZjvsE->9U%=wNS-|=BylkT;kBv83c?C!csG=8LoA~BI7<3 z)f3JI`z?Zxp)gOsr8(2%cq?a(!n5P)EUdj9D}O(8quHyMUOSL)+dgp^xfoFp_v(xsMmLp=HdaHv z_+0g_Q)a}4dOu`qCN~VP0mA0+MhGy~O8D8()bPt|i4Lz7Tfeq z>;xBUiApc}IrwIMdH>WD<*yelx zwRYREd4p4izxQhX{wJPr&y71i#t&`Cp#RY$FZW;z3vbE8zL{=2b~-R?&JIl)OMi#S znT;LoVD5jIpqcSGJ7AdfArC_xR3d?hRU8WSKn4oZgmz5m=FVR3uy4;>qENnq5pkyCv-r zCSP+CPXjDSC6IfD1w6)Q+x5_cA(md49GbCc#8Gv{E=jL1bs zw@d=IhBrazAp23d=NhGNt3`Jg(BcRvqjh;o`ekY8Z`juBAzF|f4)b|`#2CfkkwkU% zcB8TD)2hl`!%BGb^)jd}+6 zF+BbVyQC?B*18tx8HGDA^0ZY25}ie=)gTd+6ViU6{<-;6o?E*2c`Mw!0R^v0U z+MYA%aw%$4z9P^TybZhROYonCV0ojC#S{LNPZ`aBf|kC9Ac)uWPXLIupI?V`?l06k zw8J8gb2G{fN@RXW_aY#TO3Eb$T$Iaub{-9AT&k1vIu<>|y?<^pz}>tlv6Zs!2p>2L zH`~usi`{c-e9K7*Vm!+nR4PLn0ht4T$FC7ZDIEKL@Ya5|7yc*dU!+Tv&wNIRqeSU5 z1{S8tZhe*%5w`lC{jMg}ON2x79A$10PpNJ*gEoE+4>sjfRN~i^&o$b-FO0^Zx*yK= zW{8h$D_YJ941WMlVP`~8H)BL_M0OR4P4LuI(<{yJ2O9%XT0~|5gC=35s*k{f_2~Li zn8$4!yMew#LkXmZT0}Tu3eO;m!R~5V`@t8)3fU zwhFTy1`%krmnY4>I9*e*YdlU+!A%kG`(#`Z@yIS|Thw`3e8)|y^alnSHL9OO^Le&~ zdk@BBh`H~X^cUTI8F#Pl5^rzPIj45NdiyS}G=D?eT~1USd(F??H_bOIjy^R8Pj#d} zb3XNPxwxo+*Le*4jrs9j9j-lTewORk?p->}J6=5Y1O8ye>1^=D9Xix|Jw5Pwh^X&U zfP}lzFh~SZ>J|i`PQ_K&uPPTQ_o$|$Mx>uhm%e>^uyo=oPD*O47c;KnAYN=L-Nr)S=Mhq3jGD)s8u$4dwXx(A>GV zQ#z)cC=EI*vT<()GaqdC#^5TqAw{&af8o=)tfC*a3rULJ0{}Of*Y^zVa#p18l7EyD zy!lST^`EkuvPvBw!c9pd?*%}`i+XuNtBz^{Htj~kT?rR2L^`6r zZ!`=!zYI$3)g`VRP)cUWstXrolz%kF^+ZoUbps0Sk3FEsgX#mLM7^Un$mxrq_Du}# zoOlQp{xU-~z0K2@T)PGDQjgPZ3-OaIi9|PwOq>2>q4=4uBVS%m_@Ha+I3_;kggbXT zir58wP2gYJ9$zi;30Xr*+|q`^t1F zJkjFP@~-K+-+>3sWSG1g(C2WaTg^5HlW_O??2^Jnm1T?!Fa*!hI)TC-Q7xx0TGR$8bwdTO1bamp*E7 zrKEO}#vnuX2SWbHh_o*KDf2i~;>8t;cTDa$lDkhe^>vb~&ermC^<+LAB7c}X|c(Di7A)4C~Y-yEn zDi7GbV$-Tj9E{yA?0;M+^LZA<)*;ZkH+Uo*g2@Z-#imlph?nlD5ch`f|5Qrr^l-Vn z4*B?)w8z&h$%ZcTyBWwW@Hr9gO~!@Gg!LR&rQ^t|c34W1I2v0?j1a(RrdI3ow%I@4%U<3J%dk% z;KmahU?@i}>VJ@m7~g=X84{=01Y`{xk_pQ337u+@*%+QSod}zJJlFQhN9JSN0gpw4JHyZ4m(W zg)%;fn{@*{bfF+3yS$2X_6iH@G*hyOPgdq_%!A!IKIL2b-1t%{)5|)##>S&(`K0lrI~6A(ztixueuZC1siYkDuVhI6h*qVvHHmHhOd|IKjx1x zq?RL_AxV{MOEC|V^yxc|e4kHfrqJZ)yTRGqy~_0)e_#yubdhE-RuK5}8fQEP;Uzo< z5hMiT@AUm}qN7VZZH2LUK@d!;xtBNVYlMuS0DquoE(%I|GVMO`tZITeEy}KtN%_wY znPQCr!d#*krsY?)_SO0T33P!*sS!_2rP|WT!c%KMU3PRdSykFr#H2ngE1~eomw09^ znYL5`jfZTPI&6(scjv?oJFj$9OD&H@-K--pjX5X#lKG4h3oJ6c==lP{f_KsEDz=`w zw0|4WYI!s($h2_$Y@=y|IXCV1k8D~3qs|apm5s9JCF`d#CV3LAT&AOQ4xf6T(vxyK znCP`S&k)9PSqvrX9bT^&Id>{~$jwycn+=iD^arA2d}6>tL&R!s zlaon1D}ebY9hwvFg%{1D7YN16W?rhX%0$S^_Yeh6d33VSB*SArsegT8k|2Ei+JAd8 zKR1mEu;*nU2J^hc+aAx35I)+HG_LCJ>yS`MbtzTlw3w`M6rf6ZvF+EmKLr1QVSlE0 z8!wX;P9@&raFTsJ`E6B3+4w=rC5RBj@7dGcX2d+q>D%5X!37L>LEQ3@C9d$%IP`cA zmt7vwLi(Cps7VOJi{(9TwUy3?k)@2|0*Zf?=rMVblR1Z#PN{VK_K zzzwEV3jOg#V|0;zbVK5)-1~hoE7|<41pIs%cR%u6zlnw4<&Gh3{We zm#0mXn|R#jN*im9Xc|v9MORoCsx2xa1aywz3XNx}<;UVTLJYY`C;=~-kAJ&qbKGmWuq%q_t zn9@W4akD4ODVL~ys zsx{sEI(aOQ+P0yv(~5Ka`ruvT|5T0ep6;HC%nRax`=U!?`PLcH=6@rfs87!NE;F(u z)^PXA65VufPWKkrTMmABh<+{c;_~@&b9yLEGsDrgT38xPNOd)?m+KZTIQQscZYN ztx4*~1)sF_@H0d~lGj9Sh~Jp}p(pp9)xl*-h-ixih#D^R8KB|}l(_t&bH&Z#)troh`Fm$ucTd(`fZ8ZDY#cMG_4*cm#e`E9 zjtWKoa9NBVS$}i0p6mAg#i3`>BX~ndS*!-5U2a{vocaXpniRj@IElt_u1iZF0jXnVCOp3Y2nIeFgc!R0E?ZK@S%^n6WF;-G~g@CIf5016l zJLoTQ{?3o$5D}};RaXTF2)v33-HapS+StkT+c$4NJAZbmvYtqNK$d=bw}y+Ho0#|> zt3@jF1bHg+r0a`ssK)kzPgC7saEEqm3p-B67~Y_2b2q@@b@N7eFnD&4imRZNS))Y# zp@)Van7JZ4Q>5t=ta5A;T(y=Qr_af+M|AxkZ!>D69hqXx)p&)Ci$Q%k>Uomp9QK*{ zmZpWeet$MWFqs}+33ZxX4$F_}xkz#H6VixGXccV0aFcqkKK9!wB20bUQ?h+L?5`n{ zM*$r%0BAs$zeI92?FQF#kQK zhQoKQAD8X0J-Y7qZSbTVym^$Yjm8!PT@Hcpd0;TVt$u%MGpV^xGrDr(t+Y?oYUq1_ zGcxe<{KWrRdY3Afa;Cr2F;z3b=(5%F>5XwS{yOF9g zoJ{DOsiuFu4JWgh<#~kXrM&^&Xy39epdkqv-@TiI%vC_#comz=&wqZY%mF*5w)!mx z_~>g8Ma_BaMMo75MiYu&Y2V(ASRho4f5uiV3V(ff9RyryKF?3`;>jU0fA7bo`(c}o zX_=9((nh7Fy}jXdK6lYzZEYJkD}=iRHw=N_KnQ=fC#ut!?*m@#wy^6zv;F~cVnT}c zd9#kXoBQ#4EMxDNn9ehvuUcr0LI!&kfdt{TE1xVJBX@v~`-h^b zC>DQAdK@rijYXFz|2lgT7u2G`)pRad0@4{RC((0+g|6_eg#|ZJK zb1WD90InI7jg&&nQ1?|CLbeW_pyxadqxdsb7S;8flX~Zk-;j|VGZKgckdc{yel!>A zP{@;^rVu^aGQ_0mOk8%SkzzH@Si!3Y3#6dNWoMl_-;eW7a}hb;9PNk2En5?s2%dj7 zhvTr3D`^u5oMLfB9H@8I3y9P2urjf zyo1B3_?Vt_=&)yrl^x8umn@|m7U26vO22R%jyoP@rprGdE9~_lmX*a|ZbwY+o;|tR zIz6dtxLhm7oU!@phs5qhr)kn+vbuj`G)f4m^5zxIIn7oc+eT-z(dR>)7^>LGd-~2@ z2_kz}8q5zrqKCEKSvWddY_^XPZLJ7xhD=T*Xa+#d$|@ej#VnC19a*wlCi9Ig5(sp? z1mU%IZ1mZ;+I~GEjuJI}Up4mu9=Wy;S7LuL0G*QA|iBtjo}b)e^bX;n*4*IjfHNq3aa zXS(i*SbE42{1=bKjj=u1ugQO)oBg9FrCe&>}BA1W-Ia$dPVq#r0ode%AI3O{wNw~Dz|90 zgjD5OMtNdIJ500Oa!Mgs_{i}GXU1?FY3o;Y_#n{R_wDD6zmoFzrEh;LIx5to|1PmH z)uyeD>s)9)@9`q7j}^Porz&qYkFe%+JDzjD3{E?^H<+q+oOhoIkgU@V$ZfT5lR3TV z?WR%wx?up$z;$#NySKr|Seja?gZ{-pHsO3nFzMS&8I1#@N5fJ0f4Wcph6pL(&=a)u zHJJT9P}E4sdt^{0c+r1b z!Cg&qPimoUpA>SQYF5A$$$1-xhbET7PwI2GvgNB42P}WqfGYzdN50qA2HSV|2_Ib~ zw_IL68)+6CNI9*9DeXixM5>jy-bS}p9x{*XdFA@Qic3 z==z^Z4cfd#q2fT*~Gi|0^`3!4!(;8f{Go3+2Qo znO&~FANw>*$qaAv#_o;VFK{zdYa=rp?CZ%YBbBeN`YTr4kT`RY&EL5P*M0HS&jB~d zVZO$zA-(R~gjto)M{qN*$o0C5yWy_8!S~(&oNRylM*1s<+$ddH@iP@Kv#zCghrHTX zcnLo?mf%CX-oW+oZt|(w&w=woe`0dpo!sHn+O*L!<$f#J31(Tp!_3K!u>ji9-MqS` zq4d*)g~FPO4%0!!H~ThpGIJZXJKQ+i*XQ8kiR{mXPtR%eT^5uc6DN~b4l!4>k}b;C zvIl=YuDmm=KQh>s=@3cnz_(t}J|%O)oKT|8d__IXIZK-ND_aZ+DFjkHoUOKq(t9x4 zRK7Rc&sL$?{H6zEbk1znVg0hUCaPcP>ZrSTxRJE3svjf>9vxn6_1RxV%JE5c=S0PB zKJN)KqShwqQrF(_1^lTwdUw>>gLnP z>$;|JBn);>pE)tsOX~GLFrSf)L!BJrOUa?zY{n7!x!Y#USvRZmH2ItIF1RZk)aKZ= zmx~jQr%|ps#pZwBO1R(RIRt4*Q1bSuDw7rSFvj0Yt5i?+sKK~u194Rm0!I!ho7sO| zcjmcj?BHjgl(=Vz_c@6jDLIyu^4u4}iG^ohfMrcfeC?%vmSdaiATzDA7}=0%v!;23 z%mWa(VsZ_jcAUw)6g{L|s_JMsA5lh-I+LZ%oXZ&+v#&(iB_(% zd4{`V29NWw_gmoIM}+DRJhMzKUuhD{u@etgv$vRX9c;Pp?wg(RR(scG7`lIH(fR{) zuGZ3%>#2m#zb6_Cmj5t&>pX*_k7-c`T1b`ZUSL&l+3a~-$;sbba*9jx+;T2^SBg-Q z+F!dbk(Ej$?LR&Ffd4bvkvT^H0N&8-((BFmRQKHyL`5+&1kHYMAAit}h$n3;8xKm# z7KcQ}{Trm*(Q%3olxmCHC+&X{2}M(E`YmT_FtscebQALI%lfQH%*fDE*rI0vHfUYz zt>=Y3GfL#Wdbx)3M=hzXuyD*l+u(UVLyG6%44=AWlm}j>hiY}Di|+akH041sBKcP+ z#Z>K0a~hnE9U|`OVxi2-RUD0YLrn|^XD5OHD=6+pezv45w>hRo#u9(D32=OLLMCVp zA=hm4)e*X0ILEcaG|ljI?|ExAcgZPabd%iiPUt{nh>?LI)I5Fyger1ov$?LzQ!?j# zn}n{G;jw)^RZfEdPpp9YAj~(>iFQI^FbTE0q@v1#1|u@Zq&@8I*G>8bDf{w*j^cub zBf6Qs;d7lbaE^?&fJJ|Rwu5A8(79iOS-&RCZJSf*X1q9IBcpz)26B#6hFOIHpf*{$ z8?eY@)9bbxsB!lU;Jk)E)wH@IJ2*LC0%ENLdak^miKVk37}+bM;I)mnovBKhqpOdN zQ(?0zT=Kc0n`tu%Q7_h(DDda*H9rSgE;)FoXtm$GdkiZ-t zoH{qg$6KE$_U5+ChP$(xV6imf$n@o;TZk~1da$sNdX7KwC{%W^?u(?2%Tx$>(13>1 zIo){)f9$Wo@*aQWXx1LX|5r7hIBEpAVf~odbgFzIR>a4>$fzQ6YoWD0MaT%Cdlk0u zM5G>Ox+Ao5tvzM2%oyZ?^af-PUTwv-vrH)+%}Z?LFt}1nJ(D zC!A4L-Q3(0B*HD)ol$M6BSJWO&CLVkzQ!-oIzM@!5&eHn;114h7p}ywGUNAcp_cBG zM2FX}iFeOTFFRiexs386(d-$oiFY%H$Ni~nKO3)|qiL}LJgVzdJY*0pkgfej=Y>as zFm(O}$Bc(<_v|n@PH6UI((0*D+TKj{)S{-k1AB20U&l;suiJG#!_9U#egqwehVN_$*NED>JkL7v3kQIpYDw7*vs;eRGie0Gr71jmrXYS0Dx2vpx0nn2S z1j^LWv&Zf_Pl{WzssvP5Bn7-oVYz41zR+lsG%mWv0;jr0B7ga2-lT5d$)N?G{sH#3 z@QZ7))d-Hds1DQAP;FUsP4Cmar+f>J8S6bUBF2AY#G#DKmtmMKYhg418Pf8e&(1`W~hY}z#z-bmi}W0?|P_JI?|SFSP6m^5ff_;XVp|LDx=wu3AK z#7~A?)<`q`Bk_C>4M%7+7nTu~!-HV}S0Mf(pn8kb{u1uRwf_-LzC7dKK|3-xBtt0D zWW;}Pe%`Kj^8FU>kQ`n&(6ri@>o?}R z$kd&u)-)%CPz&h_sPKPO@PZlt3?Z8hskQ(;)x9~?l9_hlfoyIJ zW#~~;oCMT3%#8{O2~#8E&XyM$F?=!+aac3}{j>eKq?|m%Y3}vj5W5^ogDu3>#9R?@ z<_qAnTML(dK($9E#ncCZ`m+^Gz0dWQtjDb)2sTj{{Qbl$Ir@LV z#wE7sHogV2w!@w|_f`}gcE|G9ZTg;d@J#1^rKx_q`uTgAG&b~^pmE9V3M63nzwX6S zZ5(A{N?n`HF#gug6)D#C_38wn<^Fu}I#k=O!0jO0?{2O2M0A=%i}H{Ncu%_|4A*h< z$!@anfD}p~#(mn3neH$WdU~3{d~AO-vWK?`LzZi=P=h5z&jW**fX@EOFGh z{esjjoP2`W70s55(2=hFntH>*p_+GVGCHl7>`dl@X(j6Aj|^pjdi^^jewUpE1eZbJ zzcPR2N+p^VmAi8(MOQ_3%A3bvP*R_&`G1aB?_Jn+DNyW@APi`0>jsPtgDiieUonhj z@4Yu{FO=0Ht;mkO;#PO&xy;Bumip@q?-Jn^u5|(I|(MyLt2>mk9+WwcIyAFSPmi9htfmfOv z8`N4y4A$(4l*CSS%CGN6AEm<(7R57q|z6(vOQ7(tegzO5G>gT(aqjR%(j%45}g4)P-# z3r9?X6z^OGGL#1utH1A7_V^HZJ?mx?>n55y(B2>S0t`_YQ&p&yM8SS_ReTOGT+btY zuygLim#*0yEU@GdyZX}o+)tvKHi^lgk0c8A2;`B4l<(2A-35Q3>oD|%2u&EZeseCk zc{^ReC76q;(0wRc^X5ZIH~divb*o{zBK;|5Ch-huc7F!SF?|-#^+} zW=bfyXvX#b+)Tjo^tMc$l>5IbQ~wmer>#>hz<9%KPSdJ19zH18P&z0`t@64i(c)i~ z*Oy@bTIz;Dm;3)hC-Dyfe8V%W!*4C9^c|}`07#Ihh9Q5^ydnHAIyIvw?%dw7zKiwu z9qa$)8>kP4X)UO@vjj9>yI9X;pf@maxuCHg%130h0O&6jbvH)ZAU;f*$-j!p-D4UR*9E=c}h9tf3O-f zCZs?t*kbTw9gDoo{R4*b8{f83WfA<{MDYg<{XlZtU7z#=J*FC~-1VQ(=@P@jo%~T3uEdTs@8~=e7k|f$Rg3mDc3v<7arQGq&|{E$kozN(LVPO@fpW+_(EpJ|z(EVV>sYQ|X1{f1j&HUV2D12GU}i{y z!Ecet!1E`~wu$1)Xz9`~t6Yz+YG!?ZBoQt9{^<33`TLYLQKx;4I;50p_&~S*9UbaW z_f}fWZ#>xO-${fV1sxL)yG*bBO>o^_EDa*C!>fQV%u%?*__3OJ~+9p}3FJJK6vPe%xBnn zPO^2%!nB}>n6Z_PR;@H4Y-dQK-LkyW`G+^y`{wK7`MV1&8hLq|48_l5mXm+#2h&B# z1~(f?mG%&meb=>DXBYQy8wb4gmZYdF_MUDj&zz&Gs=vme^FPM1*s5j0#>?13-%UtB zhLbi$$46k9qxW=Yd`6kEC^9-`WfYKQ7|%QHE=Z?KXYHYFS&V0~=xon(BzbhRcyUO+ zMt>`Dkl~Z^13&#k(!)HG&`y7@Lbj4t@0hKVD1rN9bV#mQ^wt!eYXMU8ic|GSIxk~Z zN_}waM`X}+Xj1<(UMKh$omKu?#ijczddHB%X1xNWPM0&<3Oq>{E!7!=R14Qyf!OGR1^$0}xqHjQR^Zs= zU3SFlB3T$iDG_R5fgS0sp1F$)Kw+z4XC0nl9q&8gk*QF5DzeeV znnV@f2P|D|+79;xaQiUGuz%Q}2xsV^{i$x@J7~6yo}b6%=TN>fLODp3ZQynkG1lmI zgwbd}49h{X(ZRuHi#1nag*9jDV(?{qw>0)nxHN|`ry1H=AZma9z_xkDK+Mjt_0mv0 zHGP_9Z5#?cqigS>VRL%xcC(0}#KV+atbKQEKy@n~Ao~@-IoFuhCo!By$9V3k-G_6C zUCpIx>lQe-<`WyeMLAF}c>mGzQJ0onmhZELS}}#|`)WnXw08`gm#tSj0p?TWzHMHU z2bOP->qD^q{Wa+iZ3WqE}iN}At!7_@s36M za@(f=NIoy4)2tGHTxc47t(CFVQ0Zhory7QR!^m{k-zzwr&Ddf#*I4CDdvgs*X1Tfp z8$%L&jV3M22g-+itB(d#Iis9dhWp#@(&^M2#ASI-g$sXetHjo3n-dE_j^(X3dd>00 zs#4GZlw&xY;GnWJo7D+k-GG$u4(v&Q2W)B8hHfKLa}CLnEPrVRqHO;O#}%( zkwgh(ESIILImbkqcu|D=XpmUWmc=j56c56gX3($j)}SW3k}f8k5=kdeXN**o+po5` zM?jvoKR|z0%{=$XV@SnN)TRY&7OQ$&q9}%ddfxQ3kJmzPum^0Hrptlsir|_>bmQcI z4O88ecE5XY4>?@n*XkUjtXA5z>Qa3@uMQJ`x)1TWJ4$qZx!G>$g3IZ!cPDo})($Br zp8oL-)z4}-B6+k&#CXmZUMObk(^*55%AaB|neKm&j+6iQI3#RpHyc5t*ibQ&daEDy zVv@1#QM9b>X7>A0B#K=+ziaw=(}V0nv#p*mciKQKZ+**okD-H0_v;N~Ue}`-2JgE; z)Nq^-U}N;w8>~3%CD(ZgP&kOvv%|g3gylYqyZikC)X>S`I+FknI*v)=h zN)&&Ksswr8oxY=1%@G$SSnaSywwe9Yxhc51Z3ahs-(8Bu%^{W=9-iRjD&CyABy)aW z9J{`XN>8%<25~>)9RmA4?(4iG%($B{;;vL@cDL^lcel76-ndysN6woC(&|*@q!EBF z?048ZZelg6jjr*t$;2)X9ex9Tns+q07U+N0hFPsu3O#B=^j^zsn*nGp9klS+ZnwzONMLSmi4vh zVCRE@@1`O1S8-#vh8Gr?Z`agj%k+qJrrV-2JgvH^!azI->2;5T zk?nl0lY!E~d-g6GuVUTRt8XlRoi|(wR4l|hcXl)%|14gtsWUc!~q5k@2?M6l)+}UKz#3!NRE}uO2m|?dh|}vO{Xx>{sxrPO~Ij zY_4YY`*8rIHmw}*CuQqxRe^u2(>KSo6!z`&6n1jkbUu!N^PDJws+VI;27i{F((H>Z zS@5IMxv1}(Kvg`o7DMc5JVTRfssjZQu~x6m=-W@K{PqI3vuA`~rl!*u!qi1XpU;%tyHlCfmQUMY2TdSHmo7s2XaNs^B8V^e+99)U8 z@)kdZWQ$79gkf1f-2Y4ElHUmCyeSJuxq7vPDxIFC4Z6(}XMBj5qAv+P5J1npyfCfM zrf5&Cl-UpwAnr|`@X>#68I2uf(+CnQO-G~koor@Mc;e7xZzXFbVX0d9 z3t%?+IE2vrrR_IJns-?u`qo+kaw^BaWuTDY)NIK6S~KeR(u=t|TKH&pQ!(V|FUJBh zJU{E=#d=kCb@19I7BAKzmOsrlgC(BCi0Y^Q3pWQKp%k=L&+nQC9FG!s)( zOE5GR71QZ8PL4Ue9Mh_|P4vLgvj_-eomp_t<-GW#i*!Ww-7-jC?eP2p3oa4|mb9Rowjl%};@g!DmC-3oUX}|O8 z&)c?j8CGAITIYY)aeDlw8^XfpsH-jqO==t;EKzFb7+(=`n5n&rTk|n)m@d#{Z;{TB z%1{c~h!Uj1;o#>b>1Z@89yb2X?;0`;8rTvF33>&{V8@GKWcBOdF{|5786As#wOc#$ z%@~<&7C5cH8KSg3<`*@8YHrvszN17l&w=c|Z+v6Y>B@g{QL3QWCXLsu{p6NmxUW`S z63^Lj<6k`F0CFg3 z7pK)ZMf&ps4qD4-H)`rYxUoq(xIyjU`L1to++G3L8&p||Efcf}1TxO+;%IdiBLyL6 z#b?Ww#{%X(&EI9cUf9a(yiJdda=) zVIPj`>cA_0RI-$|l<8%qp7&iTa@(x=s72}L8V{J25^K*J$BXy)xFs*UCc2X;UT9bI zHGhB7gu=dxyV@FiW6xO;$?nneh{t@Ah5MCrd-|L>wD<^BR|$v#={kjuCY0cnWHhw< z3V(y7jti2?9LzUt6*$ia4sbwCbEM|!0T8wOOHro zz#QE|3jNaE`}W6kC64c_@pJu9rSu@^Y5I~-q$+5F2xEDLxO3&N9}OdW*FQg6%LLf2 zuJvSr-dC!MvFU@-Ea)((Hr~zV#bpNqQ>;G#k5cw18|w(NuDkuIL|5t)*#WsO^g@3i z?`w|@4#WQMnRTEG7A%-ApAZmU5`TmZ{Dy}1FI?QqtqVF(AGf(h;m@VJg^eyJReJn0*^(`+Hv6ZufjG&i$+N#Tuoh+Vznb_TkyfcL^mLb(b=Bbu+4Ebji#%>o0nK5JuQA+@pU=IPRz@ zX^8Q+6E~B$BUiJ zWs_HZ&vD;7-%D8oA8)vR^ipxH@wm)O!_N9UCTEE-nGbxNvgzypx_Vtn^^-P#v5$pK zsuwWkb#sdx|I=!5&~bMl@8W+u0c&3JdW+G45tEvsM0EX&<>lqy6{76t55<@iqsgXC z438`Ab7UF#SIlE`<}27@)G~|pU2_D2D2o50_x{I)-#7HHjW=hliEd}trKVq5O|@f| zt-`E!FZr>*gdm1WYhaQ}PTOtM{N5Wa7u*=x=e@ok>{W8wZK~;X+(du3!j_(iK<6ci z$qNHj$~?aD)BEX~qZnwn5lvUg`etV^s05ECNfX$!J$-|eoZy`@_>+Qt zb~KIPW9I@Dt);?Jm7J`INhg<1QQSO}ugy=yv2ne&FUd_8rw4zd_`^ahS^_Qh1e-l! zJEol);qzWPc4xJ=xei+L|q>v^5c|@b>tq_IVA*+Asp;;gz=AsqR)|a6i33P4) z;bO$GZVK7H5^}=9Q_6bV&oxvEKbkeu6kfc?OZ>$H)KpMg0`L@4Fm zV5JL)b-Ou%Wo)JEU{R3FzoRx??5j) z03BMKt^oe4@2VD+$#BLfc9%&-aJRAarUt@M=${!i)7rcth+?QKY)iY*>m$W2IY74L zEn;oArO59Zuc|)6`hWYjRS!lG8Hb~WJ{gwbkAr{KZ}Jhx=JKujruc#DkG(S}JZ)4m5gGIu|4gUBuC3f-UH zjP-#G+{y_HblU(LM3^Syou^Z#=-i6J(&VS!XUjLZyoMAX{Nei%V`j32!;Dljl*)gg z2{oG7*+Mu1W5Ftw?m3z8uZv4eJ7ny){j!1`zHwhch3v6PU3Zp)3x)!+%L4;H7rK{z zm+H0q5Y!LuQ)yJ?Xx4gO+2lwkW7z6GhKQlMzcZ(2gP8DxleeT}M;g-xTTj$gkq3x~ zPht@R9h^h5@2)bu-c7cmUAH29U$=ki-@2hhY!4tL`1f}KjL~rlE_!vL(D;;Nlv}xQ zIhVZPj7fb}9Edi^T_IoOqOGkHO{4djeJ$Hti>+8mYg8_olgKbHb-375tz4eEQTsE} zSkT$_Yf5Uzg(g$R#mmFe-QooLR|!f1u}(u6wo#egl@41JwbothepG5D1)6_)6%Oot zZuhwQUqbp88>@ekNMKZaZFPKl>2$ax7dp{0RfFCwpY`=+t=wr!tv%?NpUhq|@UyRL zknB$D)k)^T)0*Yn#`^jmo7eC@5f?vYdGw(9kmv#7M+70#GiXUum+rOohEwbEP2|?xP zby87yd^Wt=FdA;|Nz?hffSIgQ;Je)zRTAGq8PBTavG7&DW}O&Jz#zvFSjU~q{+N+>ajPaue}loNtKhU+tYKRUpE zS?47A=wD5Mx|>19MO47Y^*bh&e!?Nfan|}3-`a?}&BYo^yEEcl@2D^Du}-3qG3tYvszdsl@*RM@e98VX0`^(=p+x`Ku7YnsMQ21wfJ)?4%PToD!AMiH=50QS0rL7bcy z4R(rE3}PIm@U}o$W1k<5bjdt;ja5doD${Htt1#<52f^$`nP5G_3t0ev$YSYIN|~Sy zrOOXo$*kQpb%37G{si7+j;i&DRwYJ2U!)r3o>Kqb?K^)i10RD#E6kKv_X}(JZwJ=9FUwxHhlnk;!xu1$$1BZI{ZSbojf>vVh%OUpI}XS2@eG&>>l&u#7IgB)bCZu@7nYq7svC{xeM=yuR0h@pO#&dcLs}9ywcJ@fXiHhTR-Z)07 zQ2?oMCk&_iyDon`>7naVG6?xvLK9O;#sz8^f&QpTv(AmzZmzQ2up zAHjc~TK_FJH6m!-{=VapN~(Xw~tv_vs{Cg*bW1sso3uO;+C^I&|-2QU_WSN45AIpJCxj_4j{Hwiy?y)rKwhJG9av{m&q{$Ayv8Yv~x} zbsc%{ig5gGZz79C_i%pws@hM-ttM(>%4GfQ0v~~h;e+ZJAo$vT6V`=ub#W^d;wIh$ z6}N6n6tw>gu9O$PE_Tk2B4<4kXsk`(!hj3`-@LWg423I>CS&(gIhfw}KFMdc_#c0~ zFvI<&xX^ah7&-SKAlBOt!|nDbrmB`fLaq$Dt-(@iz?Ijph_rM_9=vBL7aT8S#9 zj@orUTjoA-eVag<{QhcV?Sx_Kd-;Edly`zQAB=L)h*6}^z~@)34U3yo=cLz8%{_V`#@cA?fO{LDGFx zdCd`Li?2L~M%kZOH%t6GI;Fo_wjUFx#km`ub_1*uMU>Fvz5A=o-e1;PcF})|Zl{~% zG|E)rodLHui#Pso;mjUC`l)t8P>w(6h(F%BPR>OZbTzwzB?x_g%-b7y#`b$&9)!rG z)@3>E$KWzATh`cnE)Xz6e{lf+rJ-FVKkDyx30- zX4kF0L>Rl@HQp(zb$hsFXPJM5zg%u~^kDyLEDLK&D}Ub@*8Ve@#nx|;*Y?tC;LE+U zKJ(YP$5#iVhD)Nz`w{s8tAWNK&UNlqZU;#;A}q-SrqVf908@)KSoZ`T0-fnB<>)Wp zu~sg14Cqk6^X)`SSC0W)l?Itqt@_f@T+G;In%OCIN+k#}Jly-cGjV@l7(VX}7?(+A z0k?B?Z|-P*C5jxmw6L~$1e>B}sFZH;fV_r^?N^lIj@(Y0u(BCf4fv%RD~mplFu|C4 zu7|weyu~7^%t=^RB>ME2M4`%R8oYr?X@77JSG(M1ZaAeAl=mi`71Lx?!SUW$h0x<| zNj#|D#&9^Dzd^2eseN75O1- zba&TprSQ47Nv`fPCcfEb@jZs%$cxFQcOAciKTSyj4kpOtI);CsU`Y}MFT}Oa8%%ft z+a4sTkw^OonXX*sc=xPNEJRMQ zNr#_+&xd4}L8}bAo>8a7M{RP+=RB_aTNCp8A6l;9!nFa0qb0-I;c&IGEc^VigZj@p zez}bdrJBD~0&YFx%cp*QwL0R3(?gY}`Ntqh{J`fP9%qxRx8513T_3qfSVS>)eusDazCxt_oq6X5o;PjR;$`z1XWgo@#oqK=Vm^7Rw$Am?X+ z9AX4gQG{|)b%|s)^KYo>VmKQLqHZ~;gsr4h;k87h8GN3>d^@z6UZI;7af!e92=}G_ zV;g@*FZLTYfV*RBFCpUlyH1ybk5ZI>3gs1e!pXs>*!2+%ySfw-2Ibgl%(dTH9EBHw)v;43WRroSx#)pb`;_hD?*w>f76QfsVJEA|BbcuVNf37Q!+iJ_&YcWr5>i4EA!peqqzp}*hN+9QEjEd(ez=D z)f#ka)*j&j`A7lw{e0{8@bwYOm0C;6>2kHD-5Vl)j}N0-HSE?;$^O;*AAJInf7%=) zoQ`^tU9YXzY`+WakR$#hoag^)t(bpeC8R=YkuK^n%2c__O3A@!gZ_3vwAHt%D z1d}h(I~M}K{3jytDG|12eYseOm#y%K|KXsca?5P7kv2;>EKr3tpA@PiU-FE%!=v^V z$)r{q-o$()YVBT(iYDQL-`j{w`@e&%!&c%G(wLV6;C`t=RVEcUm@PkRJ2!t(^YA2L zr(11I+l2Drhh(m#!AjkcGctI1a?U``jQzZCMgx1Jqq!J=h0J9Km2qFeh^ zj6})L37SNPa|CvkFlUd5vRi*&xT3!Ib86>2T@-xSdvqJ}cghWxv3oitC3h!+7GO}x zRDZO9uXS~(Li!Wsa5jjr`S$vHzgnfUQMH(3T|q%@1Yp9!awzp)q(MIZfJ_v=zmh(6 zdaDoEoBcL}%T+dl2}?kmOE*_ip1JNa&0pYk@|(Y+i3kj1n^$+KmhFEiY6O{SDYs`X zk7NAPG{`1$In)a)T1RKaeshoC%MlVI6hCyByln-TOQV#0kB6QeI1&CbT@w>vyW9}k z9TG)s+qNt&$%%$Hyi^aa@Ohws7VjUp9ZfiG6rb7-7n)I9p6;%Q&=PIbX*9BM4f@y; zTb=KMkZ?Ygs59H&o@;+x_!_wll%(G9iq)twQ>e7>rb*5I;d)z) ze|fl&j}dFT*7nNMA0p=ATVn{cS#H02u|`$J>wcC)rCP((6^v}@QU;Gom}j|(Vlh`1 znJ<^2zxEi=Ae-HLwYRFm|2Ujj-j_G3HPa*LMeZi<9jZf#)rqjsQLBc1JM*t7M+kIk19sKFqP`Z_LwYZ<43>C>xF7u zj)^T%t;f&v)cP=FKk#wL-(XG8OrusdUlMiHrq-VvcY9uBJMCwQpz#LZ;0%EUy?VkT z@T2+c4V2Dm3ei(EYCMp;0#T3&6*X~&iAXhS+8nlC@0Ne7=N6rtfujQz>Lrb9%U9*{ z{$t^FuK}xWw|m&h$e@SLZ=rmsSvivDG+E>+cSt6|E|Dcn<2;+PJ`x(5C_$)?ciRqr zBDRgR5>Mmf_J>Ecd9CU@CXvcQ?qD(o`}o26a0JiG+5YqF`F?c3EH}?<-CO-Mjw8FX zW%m#HaR+}xGAXp~Nx;&J&iw(-<7EYhas`GR%ymazuQRf+?a}ZavToe))Kq-g32+!4 zp&0o~D%(1%_32HSW|yterTPNB(xS?y`Wwfm$M6NI1pTjb3>Ll*c(aOjkC#XIEZFpZ zaf2}i7bSSBi~AgA2&%TL{@wUMb-C|U#)f|cxGFgg%3(=?Z?5iDW$a@}p zjNr??NZ)e3{up-4N<+LABf$KE`ccS;>#-mjRKU-D^Vw%i71!#lodLW>W>k zJoSH8$4BZe_tmI?Wu0mRcP`5FhCGUq^Z>-_-H(PTZ0fwO3FaBU46N>DHy*S5xn##j zp{ioSksHs!bht>DRvMn1^&jc)TSuCj-pT#GDOk*af;)CmdblYN83zWBrC(^^CyGb> z!gNI7Z-{Be`ji;X)a}=F1J-i+h0!594r_l^)(OKPkwaMYQMK1!2UFHtT#jC!g9LGf zg3rZn9CX(Z9XUDogHt5qEq;k!6z#1nfhclmp>jrQF1^8b&5#Vm$F`4i`ow9vAd^bB z4`g)MA=y{i~!@r#oqc7n9dDZUx;)5D}`6boQj@d=Qd}pNO%J$N6FSXllKF z-?@a@S|+Quwl-8a-nsl1_|*A@+(~6>dSO8ValYjXD-+8s=zhp}Xc{kx+p%nO)Y@@F zfoNU`G(ACEnL>nmIpalh0h;?f$Dn`9`QcVs%@#=(JlNgfH5LA6CGg`5SYc>BOGg-M z6RL_!wH;lKOl7U}T$-TlY*gM4!)2>rtTGw>kwCw+#pE4L>|f0pSoP(f+88IkBA=eP ztY|wmGTx$gd!C=;*$xih8?vLmYzH%28Tbm@>>X~rxnonC*m8{x~< zYjNDbY%!B-IrqEFlxW%WTAm)n7)P*&v=b`yx)7GCmGKEpyF!wkoGI#XN~)YgraF!) z!YrE95Ez!5=iEm`>TX>5zIHk)`cH8|S!bOAtKS7;1BuJ&#M-(JegKE=>HMAGnJPV5tA(lWGmAU z3Z(|Tv7*a;gQJ)cam}ju2??q3hDKF&u*uOx#y`U&o zk|pIxFMoV%NzDZrqvM7r<>hI?e=V-FdVKJYaV6zoR6Vq@O>WCbf#|P=1!2y z2x~iUfX!XEx^Y{;0EmCdrTP^@aHtp%w|Zwnz*|m>{z16x&OFzzzeOPJrX0M+GJHtu z`qkc?g_s)zX*|Fn)nwb`?LKs>~wrZb|&Em^w4KsGEarZIT z-J`;(UeO-$Of-_fkiEm*U0KCyT84DbgNfI*xnZA8RI7uH4TFE|(7$w0yDKZXzDz@W z+nFp@Uas1TLO3BqTL5K1n!l>aAyjD)gia~@;V}OSiBv37#eB^3(n{prZDevdw!2}^ zcUkjJjfoNsQfJ6ao{Sc==@?9l_Z7-;0>g`Z#5>iRbv~Cvgt&5fnb2v&zPDF(OEu=z zOwNnaCv_{$mmZh^jo)Q|O7+a->>;0l?K~glR<(;FO0`_Rj?5#{sCevziR8IJL5&6= zoxEF4VK2&ZJI(t3h4Ytv7~60!gOd_L#`rfsp^3D>qaqbrYJDD~fuBXlqa7*hhd2zN zS!>}npMal&zPi)pdieX3d7V4MYRu+mP5^F4>{syT%(KqQT5?@L9(8%0dnu@QGo!O68h?^3(Vukzcsg`|cknxQDMh2yeP`Ej`o?4} zr~=*f{m>P7ip%l^)-)q|SoNQJjU7)&FCVrlZ2eg&8OJrN-UXO_U*L(eE0Ick8BpN@ z;$SOP3J+#y2y&}`G>k>?PVreLjHm~%754yhWM%F0Gf$WI`gi|`)4a6i3g^KuT;$@v zeB4mbNnpu&ycGps3%cFI9;I@K7JdmrW0bWz-2P~$dlK>bg-xe+Rs`JCr}<(&=5!h4m1kxyR>Qx2Wg z9mP!!7xsvDCJXi}TQ0(&5(~HW`zi+T*-+lpWcd zK#K@8m!+5eyx(@;X!brAUa#31X|_k#7T%=)4Sl#QF58QpFdU`q(-YI|n4mWX2Pc`U znkU8U_vbf%j>=Zk!ns$MW^YM#$*Q)N#VW0@`DEv{eZpE(6ML|VR;=mazkvu+LR-Y!S%`BL?!`&smrqG30gQc zajICGWHcFK9sNbj!1T)@qQ4*^M`pLb3h#B3gK5!fWs*KcSaqX09&n*uc;K#Tgr_-r zKp+Eu+^PIhu6mbLclx70;b&$5sd>tti55&18m>QUcqOmPp(hV}tr>)sW0uqa&ur`5s8d&gb&55hI3w)p+O|od(Ok&P+ zXdTs}VOV21H&oK5`?a&twA^QJt{jK1(d;#}s`P9;K_G-G-!F?bh0LaxY?4SejS!U5`xP?sqmBP`bk42J;Qvlv6 zwa;9h#iTR2*Lyh5v8~S`vG^NKx$e|O%PVy8ZwVxSO(qt?y=MlTscX1@(5d3j(Mdof zwr$qtI@;TPmR`|4CgPFFbRC^#`^R&`+sD~zk{R~g&ybdwbg~tm2sZb9HHd;uCjSTXrozYvQnB- zuB5p!BbEN5AoMF`4YOw(KJdBWmM%H<6hsRtR2eZFIMr^eP zVhtyh_Wah*md|w%IR~!|TM&4vhK-w`k{6P~dQpD#uLl2s{>NH>TtX{OIA&K@?|sGc zNKADpiGMXcOO@_F>qAeB3NFMVQ#IF|QD)x?cU=TV`KH`j-~TuW@`x8QyFx)oef3C4d|I{Wwu_J==f{z%=? zg4W93Ag9+smE+TY%bLSkf}>P6(f(bTcy$YEu4<pa4NrUM{;_!d{=OGyJJW>Af|S{E zDXge?Q#mhxSuYA<|0DsuR`&oTPgmkNV(XZw^0fkH^%{=|ovbD~AnDRyljkR~zPL4h zeZA`WrBvbbuDi~c;1je|p&*y(;u2Y%HlOZZ$MN-p+m)IX()@+)=vhc!1RExYKPd*q zMU(w-!>Ay{FM7S?&RZ5dgYtpv2mZVH({C|1;5FZWZ=P4j#iIa+5%YfSn)MMF*6@+O zF0l0(kTEz*1WA8cy}Sa|kYKAt!R}Eso6teAIy;-x#cIt=GI)1=(6C>&zzC>bY0yR4 zd2@S#&*gGYX0y^1A%;q{b(zNNcDk8+;IQGpupVg2plP>Tv&Y5ekyX37_9x~`r*a@E z<&O$~-=6QI3F>OwhS(*}>*ZY)VNynZ&h)=&wHEY!Bzeza5t$=_X?U9Dj({~p`*Kms zm`D{iJ6!Nk4LpuYnD)x)@M zTaRE!WWjWWCo0kAYPXn0fsswSliFtHDN}=?mYaDdUxvvAamDjeBAbT0pZXRx zWU~Am6rb-@>~`zigefe8g3jG4ONZrW4;#n(J3Ld#lGUmBqIAGSO&#*2$OQgV#q8XGuk&MMsBMfXwyD z1~oI+OF&v08u1s{+SQbGJ|l78I7YR;!w)BVGU4bVml{f)>?wSODazAhjOz^EI&E>`a*KT-vaLoC|_86wnuuZ&b>@&KTz| zu10z5&l0LcBK(%XJh z3P=-iYZHiuta|H9+xgsopW~gW76cCEO6fbb3#g*j0xLGu+k8kb%XBkie7&2aS(#u> zvFl=ir{dj?A2niwWDM=?omIl0N;E=+Pg`@)+Y!d(s1cc67v50mWpVYP9@<*1vW?&C zwtGXmVSzKL^}m*W5sRmiz$J_)6~Q_v0sWXD8(U)2RtxXdY*({?-W|5(jSnWsn9^v% zoNo1X<^@6MISY-wk?m-h^4+ztJ#V>vcIhnYZHg6thFl!l%m(_xnpVGtY4|E-XS3Ky zdr8Ri!`ac4HgmNq6^>R-zaz=sWbL#1?8NZr&ZIkoPHxzhy$atIi(Z%1rxte_KEsiw3}80&I6s`gqE^adifFJ;vZ*qEU#Q~um$dVd1mjD^(M2sc zIutm*XLW(u9oPGj%t}FuQ@UjUWjtgDQl{Mqqgii(BI_}0d>=@i#A^D;NXf;h+47cD zf-ZyKlEz_sAP;qJj{3>ucZoYbfXiMo*T_-R3>kWr3lsA zphzrw?rNxioM6Ri9YcvD;aYknaq!RkW_(CtE)m86e5ysV{PB5S21_Qwq3o+v4iL1!D%9rX0((X`&P2Q)J`6jmiq;Pkb!Wn{(BoOOHNgKfg5b=w4 z4yh;ggQAl9EX+dB7ciM4STj?M|VB^IjW?{>Sw=LH}>L z>mwQXXKE$>E!L#)fFBxC2?iTErvs_{%?1Y4T#%uWUP&5z_oHPE)yB$?ci>u( z#dPX=+*^Wo{*ba)@SPbu4RLhJ@-h*=e{GYgvz$aoOW|a--W<2QZ?s-poXNvizYvxE zxJ$NwDIxSaQ+k5L!ue9s_)p8)N~cO*ja8<-so|q_Ckfek@B^_wBt0PH_ZgUw%&fnA z2yY6VUD8Nt&5yYogUeUcxUvB%P0Zw5Q$|f_IcD+FT1+%&(^@%LxGGCtyk^L|UhWr^`8 zAg`;DoFrRqS#k=#4>UwwNQzRh97ZP%60V9<*juA=q+eKttdjv#=T?W!ZopKvRt1E@C7k`@c(t~d?4s&=F^4+$ zF@eSpo%D6;Oa`|l1BFbA2?ZA_F4wzawes$&T%?~I))v*bbU4*UAC&!BU02(La1Qf( zcAPB}vp;v6e8*)b;jq;-yp3E=Ss$T)9xvBz?dyEiJZ(nW7i%_EhY6KF+qRvJwi$#4 zs-;zx5+yp#FCT7EgPBVg7I|7@YmGpTK!#ts4-{r=ZR{!R);Yx~c?{m^1!0XGE_U2(9D8O+4^n1SPTVBdx1Cm3Tdl%k+lgc`q z3@)UTAdX$i$MF2~KL53lZ?;fH?BsMfB}X2uLap>Z`}UF}?SA?OiOFN}ZAm(N_hG+? zj-PK0iuI{+4pdIgRiLjMY6#k*OyfN}g6fB+GrMsSG=9l<9~K+N4nsDl zfQ-L1AqnCK+eY>4+GXkr(ztwozf1Hq{^NGG^>MPV)Lg7^3I12|n}?Ui;R@OAw8A-= z!`$u~`FNU`01bbBkqkK$8+FSj1!YX{rw!6Gm68&hV{IPmDorRQ9E*?nD+9<(63STf zB(hQa*&eR~X|@vwjj9YvS+t zX9M`t9&?a^Qu^xntS2}hA&gmHMmz4x^VGHyi1xQVRND8;9_luJ2*o54S zEEaX$gfQh%!NVB}yg&!E@=O0|sQc*wI$R|U+T0+o7DhT}q908sQ}~q<20zOtGN)wp z82YRkN;~fcw>NtC_g0rngIW~k!el-Q+2@6v<;bE)%g2s%8@x+ z1RHUxtF2>29VpUXy+j8fGg*y3?)$SkR+4kCr8E+7np)L=2aPRxKfT1?$a_!eNk9$S}Tl>-}^P7|hOx@$G?@tD5gD4K~8e=~aQ; z4t*#Yw`3yeJ}1RxsfKa2#Zr1}rSrHhU4{VG0;{oq@`ckGtJDOv5)cYJ*_gSl)vEUs zgz6`rK3G#(1)GUd-dw8w{vQh6GeB1CtTEk|z&>B+J|LqWI*0V$CoI69txW=g!FF%` zRI2le&#h7^_Zv65`=jh~6~alPD~|W}_OzViX-4+DA}nWw!I4MarM^LvlsDCu(*x0! zjyD*8!kN@Kk93;N&RW_HUU2_dFWDW81Lz(m`dAT&XGCXL8jWfms@L_Tn%P)pbZFa3 z+3kXIg-dzdZtvj|WYer%vRT7NeH9GcTIMT9g?CrrQL~1>eWp@rU9+j1yIOXEazNd@VXp|CNb&T-Cd@A!~!IL zn!xQc4laKV3^J3kZT*_Hp_v0jN|ZBxm4rS|7550y)cZXoh(#&$8hm~7LvM4 z#hffiC3b`wXjd7DM+IT_lfKQ>Q%{g-b}0@7UfXZ}5^v)Zk0wJro=mI=--lOz>~tBM zpP)+OO*H3(auWHrtHJ6a7(q;rFp*aG?LJjF-Iv+nIbNcrLDM>j7aUN*2<8 zT0P;1@|0DIGN?!uUTProOK{HwJ^rZ8*>u-gvd+@z>=gm~`*5{#ks>1rg_ch*VY7+X zT@`K_DUVed^-#J!FPKixqSi+^^Ycd3&?dwIJBDY46?;Qnmdz9qzYE=eSyNSNN38e+ zu>Edc&n;aY$0mIU7I@m}&CZvz4~LCEBM&p2IAAyNXXR}BM_*R*0+Z8)E}rE2U=*Jm z-f7d;~7u*LyZ_!XKx_QWebYIuFkmYiKe`P87-SxEzo#qep z*nERO9fP^V(|O~+`hzE|2n`=`%T{g=OD)Cvzlz5Tu#oS(uD)%5I?ZSKjemK(6j_1aZsDn^!mYUdWIw0>c6TC#DP!a!>G zd|HtEawwJ~ovbY}iYIYWs$3YHygqUV;P#AE4D;ODpE4~%eB=yX8cpQnTy)*T$nj?- z>bSoNXEr&6Ri@H^!w!JIQtHonQpW_iFI5@SGV5=?^Sa$W<#McY&X+x!<{nLET^mOn zTiz9UgStP_ipoXL%eOM^vVEfKoTeU|Tw1!t(J+qzaL2#u>~}re#*5z_0cDAVht*kY zMK|bnI_c>CO5Hk`Ht$KbP8dmM)$hDd=BU@QLD)G!5px}Xww$R-__HVcE90MSI3LOV zt8vKax!eC*ckp~`x>GxI#q9|P1jYUAbsg`?V0H3Z)A&!e;M~C0tXuJ6 zJXUgGBa0nMcW<1c&KQSbf8LYVH7D$A;%oO=LMd6SLCj~=4JTVV0X z5!&;%LL6&R&p`;9HQY!U*pc?sEdQq)ID}9*fFITc3(nZbf&}4oayv3+BWJA#vL(M?9K(o zsNm^;6h(GVN1QF5NAaps3@m`yK%t%i;!?tYyjpyTb7Ts>S`~p7*X|>HB8kFF8uTnK z647NQU0N8XtL;H0sK|Sx;DuM|-@NXtBp~k=!u)a}zf8lo21Z}Z-2JS_B;RC6Z!M7#GriL%G9wa1!2r5lic~Acf&rl!)j~msHCX* zH`t>IXmj~yAUu6~+MKZ@bIxOo+ZWUG90GCLqO*tf%eP-n*Q)}jfaBNaH%JDrJ8=qU zZxfyWDUGbcrpIzY9@@d*2}nhMrR|-6vYILstKb)N2ENv$#76ecB_7rgt-OMy^d4I< z+XhJ@5Ylm>l;{~+JN_(jbahqRpX07J>dTVA^l{4=Ib`-;-SKgMM~25efW#g3;sP*P zkJ%e#D`|DUXz>JKs>_IK5!;PdEXglFV)*RMcN>L$N~2EFYF97P=mqpE<(Urh;R3DnPSHs(X>~-SpK3C*c&12mOs`IdRi( zsW<%K!`SF^fyZQ);fH+2&Ccq?@9(K$Xx}n%?cQ-sk(YW0P!A~j)^&PoAqumZJTpQ5F3ZvW87%4rEOV)1$f>~mw zlmw2zPruOY?k@>4mse!JUSxesool2U#m>pksrm2M#2b8A)n2~x41AbeLQ&j*xk$wZ7=N$m5= zR7yrY!}*qYGvzj1mR#{oLmf{JpT`4#!r0vjv`O@t)A{NserJ?_^wvO?W>kIg)Ysq9 zV5@TZ`_GH#G@7rNE~A+=%cb>=q>h;BuJRf6P9LXCdm^!S>d|^1JOuo_y$Sj57H#`%3v;cUMFeS10;(TJ?liMEre+ z%-Ql>w&@vtZd;LQd-L^zmnE8Ik&t|X?@o08zIKOzz~=#fj&v<=@8)(_4@ALSHi_z6 z=c4}vDqIasq(>@ubi6^WFY^|?F6cz3S2Y`^siK4RdH)h#5*PJolX&f^Cg5jeP=~GL zAHTYNeS-S+Hg43d<|dFp`CgY?cj3_PK>ejnWZl}pb?pqA)0LEX<6Uq0ynjhs?)r#C zr5J;@Xi86iC>p6G8BrAJM~9r!V7VAO0_rbZWOBIUs#t>M(9AkMaB{l>51?G&SG$cv zAYjR{_zE6`sJDZuM4ioGWC6Yp2eyn(B!JGQsB}RME6ZL))Xf zU>75?|0ibG&@D z1+HH=MljazCW*_X7<^^Vnm z{Qi`G^kLC8P*DouMW=?1i?bM0+o#+=blm%ERXoP#D@)oF+Qp@5Hns`Z^O8lZg5H&EXiKP4?38t;-StX{P> zG-Sz5bv*b#JWk=@wwfWQsLS;OGNyEAeux1 zBS#_`CYn?e_3=cFvMTm}-1QabWdcA)2wn`~%-@Sf)1&TPlJ{areWUz~O;=dmf-KvA zhd~$Ob^MpIHYnr{)wW4S&1eW{f8<-oRX*aF%g1P%c51&JUR3#}W6zqhUE*wfImg?B z((kwebQXzp1**Q+=Ww2~>8~h@e6H`|6i9=`#+(X&avEAp*(CTw?&LC0K0NdAr24J;SVKI{ESErOxq3fhQuyWeD|!5Kov zW(5oQ1UE2eWvP#i*hS(a(K)BLx*##Gwgfn~s23x=H z=?`~eHr$*bb_Ko>>EF#+WLs7)IDj;!w>iAxB;r#z!XAks1Pgk16|ICY<;c8$5=H<) zbo^}vym~5;)Mng$$V~nOms5boH20q2HuW5@$HtqoRqukF(RZPn#99a2;_oQ)3R6rp zD!I)R#D3Ap9U+04C?p3+{Ww#!$ymk^u@oc>3t_UsLz`r7zV%L2~orckIcDo%J@xY&HDTty2ZrpQE zAa0z4^dG+9$AQ8=v~R>kzS!e@ndJ#>|M-!>n<_0SgBQnk^&(;YFtLAm6$}2rN`Ox<)ZaaK`%(S zmO9n`2q%g7HUExiNMdJgAMB7D-xdzII%CGSf-|i0%B!`PTV{DqpWWVFHzAuZR z8#>b`kBLGumMM3CQ-3oyneI*TMCV zBYWQo5|N(dV$RdU3oQ%)D|M%eU?HSfq!K$fbwv3#Eg*t{EqkL=q zwBB6`*0j5$$Y>0?UINmT;z}wiA+1DWkUVy(p>SwP;ISi_H-wwLTziNNuGq2t>SxYj zg(Qt>XM$3H^;4@is>EhL@l6Apzh@+Vw>bzR!O-diVo;2JF*`<<>^Z z@Y2>~Gd|@uzi)86x#SN8CdyOTa$qpB+GDBf;?QA%xjaO5cAbJcX`BoHS9RA8+?0Pr)KvEoQ5kQ zRmc#3vGHlSUqlsry5^hw`C~i^N$QbqU$4nsPKT7P31XS%*fM|q9s}RUJ6+RK@Ymc{ zfGOiXyD>eN?d2vqehd>4fkA|i91(x!&on4x#MMF7SIYYLe*O4S@Tk{#@yJ&{9o9reJSCxRw z4JzK`6Z~$MU&v1q$nfHW1a(9iH3PussP6TzRc2`DPcE!gwa7O$W3ee2)(T>LnmJ2< zg1>Kc6+Nh@8)l|}@G+r5oY&=)=6wQf@CY@(Vw&yGLmu;93i2N)<|{9VsY{cF-(b|D z##WTx?K&sG6xx)Vgu>ADCW7qr-BWwFI!i&t#=DcYw+7@$^~DNfPW>Ol^R^PLe>pxL z7q9a<9&yCyECmHx%EIqo1Xq|0fDJm=|uj9i>LnZHu_P#HcKy?u}V zkUFOVQ)V()WuMOzKh3l13qyc9Jen&(wc8(pG(7LUe_fp?X+q)3Duo@5pU7Bta7rW_ zzhk%)xM7g~`D&PY zYbJb0ZRVPJkKa1egN~~e)Y`OJ+n6ut1#0G@aTLx0oau$OEdX)?}EPLY>eU~QLAFg`b z^YC#SA_aUr2_4({LXM0J%56PiP1`#4-?YB2nCJavtm7Q!a}=^;ziXJRJ!Zao-(=Fl z{f+ko8rn7QUbYg-=`~7>`&mlrnl>gQwYywq&DqIX&6RK5ra+FIBd_XzubeGrn~rE? z(lYLDotL0BV(ViGe7o)`DDj_KH~%YT%GtkuO-RD!4|gSQY7)Ib#flGPr*~mex4^E{ zz>MB6|J32el3hQB`rnJjNcqO7 zYUEgJRt81^8@>8{KD0Z3Asaqfgkm-|5Ok3Hz~0CMjTlGY(?}vA3`5OBRwu z9#l-I8GM}E;?bT{SJyNljyErr8azA7jtqc*S$?Zk$!U^CLBlXIGOlrU zOu&)%$iwZ2y2Q-3Ht5vLlFo6;+et!{B-eJQH^Vu>Sth6V18KyyJc8|)N(@AZ59xCi z79Ef%_D&7KjE6J6+F74Z`bdG9Q-=Njklc=E@G>*uH+Su$x`WmV{s zQe(=d3Ixu7iTK}i$b0a=wcWaqpNlX2*Wa?@i{b~~&S>C$-Rf94?eUgCx5Ya@N9#3zbD5(f`A18?@q~4+r+r6BF`aN# zmEn)^jxemjntQeH02j(BB$fv`dCI!01y`ojoFZA*MQypChBh;5FJ*Na z{en`jesM~4F-u^Wn7fhYsz)8q2gca(V`N>tv=Vf9uZ5a%u3!hPAIL=&Y1s1L=gR`* z!3WlVx8KN?6p#vaRf>RCx>2#g8Z;%^T&QnMBup^?5inGo1B107s*A*_64F2jMw!NM z6FimtYl{gN<^H@?0m70MPgqJuz5a?i`V6jFSlB zXw3n3K!DbArKkc%*+w8f4@PfjA*zUspwG=4*}AzWpCs!zxMzgJ|LbBM{;zd2sehBO z!5ywFVs|lblLeAvF@om~x!+=MaE+>e=TlU#b2`tM*u_}IT-&3i!U`q4)$lT`oYd{S z=?myjs`1K}ypuUdt}6Mw>Us9T5WAdRYOj|3;K8iZeQ85!T^Tr$mXdViO1ODGoha5|IWoMo?nYw1$O zcUOIPNw{QgcvnJajk~Pk29kp&e~eD$y9>(3?}EPM_nz@!(_Lk>eIIr`4k(Xj%KSpZ z&7;;7*RoJFZ!UviBGJT}>*#1;b$nhp-In01$ki)GA^EW`A6A&^0#vEaOC%twHOFMj z`etmIh%(l($Znr+!L2$YOu=q{7xu8$6Lo98*OH8FPHF!Ds%V5GYpCUO#RpRZl<22a-sbi?9bRdHQcsT@=nv zg`8~F38F~@SxZfsUFJ@!U695A+e3)0D%2_oL0pFV?or0S^@ux@l|s39>P@%aOYU+I z0yj%N;|um;*Y9U$uzZz&hMdU9&yBYp``l&Ck45>4utufKYj*k4Tf8FlA*v~pWu=I} zmJ{?~O-)8NJ$mf)BSQR=E$cTh$tzuolq5=tTFsENM8XOPk*dRLejA>7ZIj$Ics+hd zNS!`wj>@R`j?3?Mt9XIcHPH$6tQ5koFbJ2-hJ#LR+TzYf15ZAGbNVlJbBG1FX=!Mu zLMQ!{Z+N$(O%De&jg#kxM)PT^bUnneVTm6{B6F;3Ngd-t+3O<@^LI&i7u;5iLY8x! zNMmynWjjXpYm%RCnd_`&>`;CHsjtv~gQCfuWs_`0Lcx7THL_dOE&x+Trz@WG? zwX`yF*_(fy#&{-w38BIn_K$G>Me}dMS!De*Tn$&i7we7b@C$7A=ILVpz)HiUF?*I= zS0A#Mb)MZ8dCq_;UW*CM_)RNYe^xM`+a(JZ7gvFRy3bCRbws|=I&`WYF#K7ingAAQ zu^8Vi~>X@Gz)Aith7>!lGq;<;!zhW0^ z*o>wL)Jk6MaHxT~b0!ULk_@8d3&RtBr>f9|I!UC(y&IL7GdCAVCXUu3ip91l4NU{! zj+$@0(wruLFXU`FhADO->Vg)l6SY2A!sH8TIBlYGm{BccePW^ODtfl{9JRBIbB68@ z+=v~mDbs9ArQLgYR0*V*L%YLevv)h{R%uaCP>rUQCv#0?r$k09nfNt2u1MycZ(d~x zq=p_OKqkg-Mgf5q`K6~1jVHVbOrQSA*8w^f=1Xi@ZE!A1`K?l5=M9ql(^G~bT zzk$^C=s!Wxm@CaPNrsnvI~+H)VYZSI;)6pzSO9llPrdvGLdt4Khi1}NACSZD0lfz_ zzxiuSPJuk^S~dUaVE?~H*YUsFn92V}R{}qDBvo1sZVvJm2Zzea!@RiNh`3p8K2wt% zx!pj2Xuw-XLrioP7k}K`G;sB2t#{w*G?dn9#2duuQ!@V*8QE!n-lfvwbw@~7`S=EL z%nLwwY$M?IgzIz7RK5b=y@Jm=weueqzQqaJ}ix)X-|lTH9!1$*Fx(jS?B zp)VA|g4i?LBfHCLvGLwWGFo_W67YU(Gv_!pst^J33Q@eq7b<@EvHSPE+Ojz%1~b0| zuXmre`6$x}tnP;mIP0Bx2F>+myw)SwsMWo`Vs%Sn3a-Ay`cgmRX1^?IezbQq3^xs)l zH!$1^Y@VXk5JPTz_S_>v;{60*qmf+d7xMjME{{byC|iaRTY00>RI->D^PR$?-b^z} z&5EK=C45-4?$S1rNqYw^F@Y;8jTa}O7-37wjN$J>H;bKLnJ^F&GoloA66#yNK|eTuiqaVnU_9BtoTWcl=Q{!m6p5?zI$orYpB-6pZgc?p zVdQGSn%dBPOdN}q30-MIkt>hkYkAnCmaOn8JFxe#O@CXc;@gzY{xYt9zkigl^X6a$ ze*Jt0TDERpG^5hkvm>4sDw0KM7%8iT$mdXhqXzWT6<%P~ zRtIBbX!PtDS;*QOZ&cb8c2R6BCd@a6m-!L&%XUMHO`e%TSksu6#*;>eET;OwNn56{ zkGp|;?|UJnqH*Fyk^H_^7nbGghjWINE&_FB+%`+lI@*~oy!-)(P`2iq(@;HhQ zg9%R*k7CVgdD?(Z5uw)stT}E7K^VjZZ-N*IEDWh?oP2;kCp+h*`V^6|hGKfzgru2A zN%$IOzdB1~lGo-o*q5q1m*q@3GEgCQ#pp?KZV1Q;fF0=fmy5xF5cjg}or&hh`?|F9 zahp3+H*c3z0YT8+UHDr7c;`w(@M;(2@(Oq{2i_UqzrDG-eYifQRO`2>LJ6OnyNv{c z*5;lr!GiCbhXqX3i4J^Fy=N&${oRH0$CkG4!$czQYnP0+D+tIj;Ke)7esR{Zv zYuP*)xDGjjs$FkS4;mx;Gp;AIw|=vnv|TvogX^_29mU~(pJ>7ErT0D)p0?hP za}Rg-JWqgs)s&rTpO#0Rs-|+oCugG;y>jcP^u<<}BOG7K51Nk~AvpiSj5Jlw6^au$UuD?HI^u69@dDC{ou-bmF@g(xUeBl4| z0bZq}!#&*OEuj9>8Ny2#1sIuEKq|L>jQ__Q|IM*~z&E&Dz6=w%ugA1Lx(B}Sqw@20 zzgs!gOY~upZNL69y%I_OuRZC7pU=xzKzbeyyH%J$(tpMD|Leo$XXqf1grT?p9q_l_mTl(Pjd;EKs6D{;-%o@Ud9iq0zYUj;-_f?eP6%jYu z1nOaj`%L5cM*Qgjn>!Bh)oJM_!n0sTcsjj>N|a;<@3DRlSpjVYpb8So z5pc8QSfHR)ouYg$l2l>rgl!6ZnZfASw z;B7DYHp}JD&v8E*eMOriWG4e0N!fP(o6F@X`c?c+2P#0V^db!)czb=&U`x2wDKy1@ z!_`mt*)m(?6K`-JwL_4`W}&8i#v&zu&49izFy;wP`cQVYX{?KfJXWQ+u*? z#pBb%pL+|it-2O}7JAnm;ZPV~0kg$Wq6|?#aNB_w?~D7zkO96Ic26?TdeXN~86D?p zz7O?~B;e_l?^S7=6UMW!^e5G<69BS*4;<*8#~mAJBtQ1aC4J_1eO>>cl4ywdEN?hL zg@x-CEy42)q=Fh)(DXs-S@ZcztD~c_S#SS&a1lntIHOoIf{HFi9zPwWI^gIKzpjql~YW?@9CGC3?A(%{b zze3qx-wv^u-gu_xKK+vWdb>L{J_?BeHG|t%5@kuAM+2xA^8`JQQLkQ)J)K+%iV++? zljyLwu>USA4y7p3pEcKiv+!>KRY0o0e-{3~frWo3xc@G-lG<+09U(1%`<1J;NS@_q zk__P->Vhk0n&(9iKXt!-+sDm#clQWeU*OC<8VcnzW2ey9MJo=^SHK-ZF%s(MS&a4_ z3mQ&J$2OiDEZ+j|c;df5&nYEPr&u{Zb9lcgq2Xk%Q22QU8bITnVX-6E^E@E{e`RZ% zEIG;Z!U-)joM@$WKQEm87oh?*aQ@(f?DOy;2Ms6XN}Ezp3?qKDF{<^JT{&e0`18f2OyS63?4TXr)iIz{KTvxAJOd}722NaXr$3KBi5;sQH)*bG z_`tFBeMySvdF>^PJq)yb^u5mqM}*^y^%H=E-_s}(+zmrw)6W_Qh= zH<4*j*r&Bh{>%F%{^31EC@-rx#Gc8^cTip)s?a>oz-Xv}(Y4$uf6wD(J~RcJmMcCp z9YGC@EE%169uj$=8HLv~KJ0l0=0FWJvW}#97IJ?>k-NN5to&S>zXgz5Iz{W4{7tj}ooME_zg;*5;fu0AK1?19 zeCAYsL?dM)SmUl7e@O+}pLX3OWfMp~3vEyjcAVcMF_zh`^R0S7BK}JTSaYPhyGQI8 zj!d9DcVYkkP&WR}!2g{X_VkjoQtE&-6qA z29sS`krM86f4%>DY~i8tnH{-!C{$I_x}FEs0c<9_vE%WJ=Mk|8_AQ|J;9c5{UB3v{LFvaEQe&z#OSAwx?gSA`TF=6bh(I#M4wqBK*5>^)N`}=?vu0+=syUb^U{0<}+#M0rkFTQ{3~8bAM`q z!qY1V=6UekLuax(-dDtUZa{xMc8c}pZ)g4Of2_Zq^|!PBcGlm{`rBE5JL~_K&hmT( zgjP%;Tcw<9-~TJCO3?dVSG{ckyq*-iq0cP-(Z1u%<8zHe_&8E!yl`JW&-qZrpCp^k z!!Otj`Qhj267}4E3h`enhfn*3wzyns)6}rf1JPfFvYJa|l%wgm5#^1q%kezL*vB_8 ze#qwhuqQ#$SYaxZQQJafNzrk-zDgwOJq(o5B%ap6MyHz2Xa# zwwAV=PVQ{EXCCj3!`?Ie;eK^uwXWSHfBMYr>jP-eqHs^)>j#9mczMbtG|)Q;9ZRc)_y)tUMyOvug7XVou1k8L+~?Rzkj$c;OAczgT;8> z7r*dpf_XT+x6suE)G>>$ykmH7)9A~u^4Zjzqmzyf?krj2=b>X7drwc0bG+Pie_gv- z?3q1{?`Z0;$G}eli{%-DBru52dNmYuzIC8;*75+UhtAoY%!b4BFa}Ekd%ylcAsg*^ zcd@`G_(1^G5zH(Y63?|)1Df}$o_zo-9ZGoJ=N`bKqW!wA{p@pP1DH^@KKopmG2C0g z@UxH0xkFWo`*V-W{rYo`a^bnBe*izqL4p13Q-E<;Om+;&?3K*&%pIbYl@-m<9}qiJ#Vg(lEUn7se~VgXrq3@$ z-U-8DX`S2yIj^2R?2e=@B&|3ew_c}?16Nn7d2iA@Om>qOT=+dNOU$y*RVIGEK>Tn0 zev-m46N3XL^o&4&&p?|ITmf3oL;3 z`xeUo)c2Ux3g{#+Sk{9KauDCK>cHIw}!EI_Ggf-;{lg6E2-PU~W(s;H?e;_j@)ni$*{QK&L z<%)#`(QA>4O0wL!Ddv56{x(NATm4aL* zdu%4m;&Z5LQL0lie}X#kgFiF%io!kgGI!Hl37tox;o)0vAmi;$NrDBweNWjY8QvW1 zH`}qEN4IBPmdqz;?%!%2Zh#pUkqf-a-HrD{0tZum8HKN|)}-j02EORdxu1a(^?X#G z_E=aRoTlEl%j>q`F)KPIE6A|V_@>-Efy&#H&s~$ORWCB0f7pD%JU<(L%xsMO_VHlP zW(YW1taU6+Uh};_4Y>t@?JZIY^NV%`yYL}fQ5Gn1csdD_Z--XuvqcN-7c-Jg(J zwD!)U{tO>Qf5X1}oV-bGlT_;ji}yV!Wr9%;A`wJ_vT0ri8oQ?&rs)P{F8s`>CBBMd zou0EACJM3An?O@;=KAdZ`vRXy`!Bk6bBE!$+ua4=Hl-<SBR*o%%~#y_?TEgv@7J>_==< z2UXmL6@(4?5%`?4-lscEItlrzX1qF>S^_;QY)W$3CKE@d1L9tJsV*sB0o8HrgLj3G ztj5*1f28`rLEF=D20iyBB*Zya5ueQa``qD+5!TV1yt#*<@$0kqs17Iy1AIYLnEPYU1+S=doZZx6oNoAlwGi+q-3#9i zs&0NpTr>U1Vx8KnyC?jO)6`>IlR59}I3?K)&C|y*MMla2wWAuZoy5;Ul{&VR4XGa! zGsoWESu~hyc|8cLUfLK)H{P96PMOLSFQLqdYoBNj`B}oZ=cX@tSB~p0vYt17pW4q) zfB$;RH2G`-t`!@*~w(!x(2tu!mlP zuH16Pgmv&XCrsvr`0KvCyu*t$qSi<Qr(Fg28e>mB zL<9yD=P;HshT^qLaa|NF8PC?Ye;lc6`jYOult+xH1=4S2;F(yJn*fNHJ`~`anrGzr zh~rOSK#W|L`OZ!RZLu%l&Gn7p@4Lqq7>n~qPkhQudv z7l5*hdpQCt&_DZh&=ItPHD>KCqB-7f+$POh+6<5M8z z;{YYe8A+*$IxgV`E?@&0(QTP0%T1D>a5x>}`J_|Yu}w&(h+F6kG4%+)>qVrzignry zW5K1RTzDMt*3`iAykwWtqA1A`bm}j_G+I7{NMrcuHj>bnY77U;Jx15*v$X2&mys_V z28t;zR36Ir=OP=eTB)_uFVA51h=-~R)gvo=P$)Y~&ElW{+` z2n>6mo1^1Za74%!JBk^F0mbf94npDr7Zwk7G0+ z6UO(~t8bNxW}`@!K-;#9$VKD)qvTq{gN76Zyg(PRW(2R-QbZM#Dy@Teu7I#19+Mj04G zGq#zNGE2VT7}jZYe}jC=1WBZ}2@*38^9S|YJQ9N1<0w7T#P4Zn^u#?NL$Vr1R*BY! z;JSkT;F_VPjpz`foZSptUE4D~w^J-<@B?ZgR-&HjQOi`F3U7uPe>QDd)7Xe7xcbVb z2(%fly4J{v%6}H1m@N#)QuhZNV-$z+dpgl_~_9iQ?Bfznf++zE??MxZ3ZFz#8jd@0` z>P?A4Q}%`QfBj}Sqplp7^=_pMk~ORvS5)wss5Natfs{_MlN*00DRN^QGfKgA**W9= zNvGJGG@`e0YO^|~GBZN7M8P9U<9y}mo{!|&bW|=(Xsag0qdyGt`J|eMj9$JA&P?*& z8ugx`)MR#Tv5Ct|GI%rN9NO!&)mW-f_a%w3{VYWlf8}Aqjd$mrQjrX&j$Xf`u`LtY zr(^I{#qIXg__?P}O>MGKjKs{Ol&gxCZKb*VN#NwA&ef$7kvCh#0Se{_uXTH^0->~` zUGTbV1`1@fevRGyQ<8d#j@3la%xnrWHrR@EzdnBS=t|;>vF*OdW3$k7WV3!9;(?EYgB)5mXe()J_fYq&ck~HQvdZ{DMIL zr3jwLN81}RDo#MjWSg(5V=PkJYl08DKc{skWf&KHs_N*Qq$38GuGL=ynJi{Vd8f%G(*Qz<85Ex?e!>8@sw7b zy|LIro-5Xv7QC*=i-0$fD--CwMcncFsjd`zFh+Rc+KoJWN^@?=blxEpmE4&>?a7ty zf12JnbGc+)xO>c{Hd9W=S>AqH=W!TndLA4yp#s9awrN`kwn)yU+}<#h<0~bc+UYm7 zGWOz)QXnGozNhX+_DQ#23{?RtuC039@w3}3ugw6FEXeohIoE_oSzRGc^TszpiI4fZ zJ~!;_9wWG0*Z~tNs>fxJ?($zR^${=be`@E}%7Q~&%p%xTXLg(koHqweliae>jWlxV zhsRSiKZ@L4H8ObOR&;*4@nJ2kXgXKa8MVF3$l1z*c zmBBk5Mc&MiS-0viWpW|4xkVGQOK%x-Rdz`Xg|FfT(IhW6<`Jjd`aX}LO6+!!t?@-HGcJE!Z1F5TC|C9en?bJ*E`_$I#$lP zbyd8y)daOt*RfTJ_ElR}Jj}$Se^vFxE<4!ZF0F>;F`dwe|8aY&i72V#we{C}a-2Dq z!QB}b9THOGE_%aE7P@1vxfbi@qmbcOJ6|IgOx3ikD=Re%q)eAAE(GLSM@aOpC+UuV zuYG7(BP<$6t}p5o3N8v$<(M}Qu~xpqN-XZ9Wc0M(qjgO>*Y>_pnNBG#f2Wys;^mbk zj+7tbz1$OWPF=0M9c$~%Dvrwa3{$aOO$ZyLwTLf3#%bJg*sfPOqOI^Z9q!>@CJM;Hal-W{6?1| zo>`x|N{WB+gWf|uDL-+Bz*JOhe{sgX%YMvoIdfGPH6?+er{OWPe?}Oh7*!6T{odo~ zs-c17wCh@7H_9sGoLg(M*ld709Pn!O=@b7|*RW_P_Nx9_Yife4>%2{|2|n41SCzk- z@fTUgwqwKSl$zmve>l$}Yul;JDi6GG`-jW;CP>-c%!zFqa}`{={;KH(VR3;Br!JA8 zlFY2zB8%SL<WYgQ6gijLh)d4_`8y%wjtmWQY+)e?JK%(0yUvd1RG~#@@cX ztxfMRjyyHYH{iARDx1%&B%T4!7<>MrVSU>4uW|TP8LP|X^!d)2equJ!x)m9-?5R7a zB^_yB{pPC`AVEI1c{f_yHvJ!6Befe)HCB(5nAG@ALeA{NIxT72snsmAUID{`wX{}x zdtA%6<)5aHe?-S;W7Kb0qxY*KWxU*f?{PEb)W79MeEdGBsoE=0#%!HzhZ>qRmb6n%UcP zjcPDkb?6Xd`ho8FkcV`ryp7FZ<e=)S0;zyGTKwGq#+w~+W&#@5A> z{x7Jn9CHnR89L4JKPXeKlo2{#`xQE>DMm5R)vh*KhH)mB&0D?l#+k0K_DsuqjY?Ql zf7Z17dMb~|x##m|!74k9equONs}^2J3DS?4t7LvTt2%7j;~g4j`;8!;;sDfTF~a&= zS}efuaF%NEy0$KcR$uf$kk@T6oi7k6e8d^naU%V+;OxmC?qq#A@3Xz)ohZE zAbT??fy!Eg8xwu^eseAuz_Tl6M1~T~@T+QRJg{76QU zC|o+M>UPE~)_;wUe8I=&vNKqdVr(sP4SC3n{2iVu&butQ40b^r)7}8vcz*a*@#~vN z(=jcbr7X$gTCr^b`23h9@Z$GF{T!fxZa?JL(-|ac_B!$WRoc8lf4Wu4SG$VMH^YTj z@Loon5nS0*elZKyoH{yFFS|ZVe?K~Fm4!%jhxQX>GBj;VI;Nz}BB?SS5O9cJZHCil z$<+u5A9=2$YYuuScn=-J97J%g&8D5;P()g?Dudc8&(0q#5HyQ1>OGZM^qK6=ih7YF2}_hAhkR_#aFyG^?X z#dy8?_YJk+718-st61de;#=WOy_y(~p+2_TcPrHw!Z@ZIiL)IMgT+C+2d?u?`%xb4c%W2^L0l=`}` zA3R5iFh^sj^2*s}OrX4e!8YG_mFLZVB~a))F^yg-cbfB(bVe@DalwS6CO zDv6pzLJ&lV5-r;31VKa@y_e{9^v;k3k?6fMdhccQ5hZ%>jLr~cw9#e^UVhhg-A`TX zeb@c0^{#cU^B-f%x%d7apYL((Qr{6v%-`{WplNuoxyf*nOqR0mC)0^6wSWI$-C`#2 z#Lm7xYtPVWbhe9)f5Fmi_VP|kA*>%rfb@c=J&_aez@%ht&+f6(wUt>s6Pom*Wb}_>YqogieR}n*E5r(zfSs zYoEG*wonwue;}xyCTFl;z}NRBe+@Xd80@7fy zSdr|_0Jq062Ft&EuMdOv?MMkQ-*Xdm7i?d8>}EE}e^Wh`lftCsaQUWbz>uC@4Gb@3 z?<)#?fQQLu=QgofnZCJL*JmcNBS*y<{qp;pTn8fa3&8a?Q6&P1(~2pbjijf!^*!DQ zu&Ru(S@cShf*s^)&(kV0+5ZKS>YEzTWg|2A?isaFF#Hgi7YEFeBNXdam{5_LwBD=5 zB=7kHf3aNp*HR5v=hx|W0AggZJ0Ij|X#)?;RK5eMsV&6choL1DTJ41g4(-u#QvmQ) zr3_WkP}6}*%++LV$@}f5{YhO64_@ryi({f)A^ROwl5qXjVWB3A>L-DPY&&L1;uKGF z5eF1U@ouyXb=7y&;^(>X_1Da(o@rqdT*si_f7KH&YTB5H4GFL3Cf?TEO`QK~`_+PW zcFgS80X@jF4vfYVDmz5q7xFZHI}JRm3y7Kk%ps*T(C=dxo8_zHB@I1(o;fvAzkw`z z??|bZFC6dJ?q}+JyA*I8`z2F+c{baE(dqk~&c(nAn$6t>bZcI}9rNriw)BtJGgrSv zf5bs1i#bHJtD}ew`E+ax;|7?C_$QY;&aO1+`m-0yCrrTOWi3nBW^{(1g@g4XL^AUm zFNeG*SAurS!xhgPpSC|)WZY?%O#2|_&$&MBg_@G{!iK88g-+xHKTIWv{&ut2BuSsn zzD+kRiUYcTJslQCeb-U6r|;CvdEIX$e@?OnNhxcZAK-nhXl%Us54t@OLwcpDiw?sz z4PB98USnRLTuXjt3#!jDns-hJOf1|d5}oU@)Zz~$_^R3r0q?z~6HE7; z^GoR!i}t-p=1qTt5-+@F5XlP&h0Rwj4Dh4>B%CBF+xtq>IPRzMP1f*v!bLjSe|F9n z6oclcv+WYJm#X6XOD@B5;4M=^hELPkf{d@Ko6=<@^?m(*;2nse z9g^C3Y%=qS%DG#e))xk&aqEv&SfS~{)wQh-DcoU7h}<;8hIX*HEjm)V&b|N-H2)Iv z9TJk3_IU4kam=8pkdN&i5)Yw)f7vYlD@P@_c*An|NYh2MTPKSfyVTekCGL#tTaa*y z>i}7i+*nu530iJlY5?mukeczEL}9bqUMU}4hIvU7wWzma z8%egFd=Yw4jJJuX&e^zr?~>X|ZJ9>ph`UqTHm>5DEy389spR6ql=9Xpe~j>cSov*M zDzH78C$52LCxo#*Vhe84 zG&G;d2R-%MT-ZK;<``Z3;nmN#f&^nfKEVmS-~YX?bi0ctFi}o@5_&%}ecI-?7GoP` zDGbbBF3|JZ{J(&XC%5z3e-M-2yrZ$oPVXPo7f|0Zx2Zn8m@)l~j7yMqSo9)G8~8`Z z5mM|`Yy#Z1g8c|<_)04YQ?uVf6Aw&(C7$x)v|!w*paZ_gm7sF@y=V@`vzgOk>5#Vl zOY|CYkUJ5rPrR7;e<5ythnX4H>D|y5 z?T^xP|IK*1oAm7}8uMHlF8VzCSi$EN*WP_e<}e%+K_9|1oPoe-ZzhCx6c{?3HSsyLhXH zHn^itFO9VL?>GGsB7?t=eN2emn#!RDQV?TR6GMFYjZ2mOlLF6h8l8sW=J7B!DZjGj zjmwS*QRO7>tGDm)*`PjB9BNa|_?H!HRc#D#4PR;@sP##%ra@^Ji6a;~Ahl!m+-;%x{i^P} z8#lqC(DH5XwooaLuE6=h+JI1}6bqroQg4{SGVi|o9H=!|gFUW$!^>&7dZ39AFq`+i zJ|yaxZoi76)Kn|Uz3nXvinZdS6l7UxO&&in?jfl`ex_74i^1ZxsfKiHQub71%DU!?Y>yCXOKt7 zAQqoZe-u(*=;{8uO-8+=2{T7&v$)l}^Ep8(fqP%s*Eajb_@MK8B}hrsZQj)owRgyv zUTLUae^b@y2}6c^PYEZ`G{*4lN4b?cD639&hKowg&uDx6dht%@mmGDVa5sf zV;@eJQvCqiH;v=5#Bl^ESgNj+cUnZ1v1lAGce&{(;Py4?P8o4ReJle*tsBOKj^8l_ ze;NNKNAipn@TK;cIKL>PlkwFBQ>#wV?~iKCG5}ith;<48y^VrEOFO_O&EqfQ@A)zH zs_&TWjU6J~?ig9+>U!Fixf$%&E@i$rr<3-J%jm<-7@OVeG3#DngFc zMoD%ys1|JkFDdF|#g4Pew~CeBN8PiVe}9b%{a|6*JU}Ij^7WdimW_t5pqFVSi-!?7 zNWfxgT>D zILTbU98S8jw^>@qyWeZO{o%Da(qhe+?;xBj&XVpE!3woZqj^sVPt@jGv%^IS8JG%} zx?lNO4?3f1<-t`3I(|&P=VdUge+$p76b2)F6v^Qu98=@3{fQ_Tbl-RzZ_cdqVHNJ& zC%B&HAmIj;DdY6H*5^41^fu<(f0C%5vDsV5wcE=~C@1uqIF~QpN=JYoux^e0BGG3o z9|4fOqfL$!?fT$#uMQ&$4MiUn?*bUru1&g&xu{#vQ$odC-C1MUV;-#6Kr*MK988hk(w;y7qa$@r(79-4>pd8bXowsX2TX+T}iz z`Slv_%l0Nb3xkD#KiqIpe=1BY7#6F+)#qNXQ*}FpwWB=SwE!yJ^N#92H4FARQ~^@D zY-=9|R%TG;k%XKYQle|aRASwsQBhMfqAj2No*IX4zL882!A*Lex6W1{FH#riV6p)Z z8OZ(QHShRgz=QMYOkIe;?1U4f;u~FNIB+ znhQD2hQGxzGyaL4oPiQ31$bMAzN>mL_j5#8wM6e}j7FzGsr;u@{RZqpF>M?8oM z-#(eVg#1lfq79p|1o8=1J|<3L<#Y0r)~YEZ$-@EoZGFiYkaT2>M}ut@#F5e;g}?9o$?(N4Q;*{fxsn@|})jLogu{b$Bf@I8n<6UCh1FXHFA= zUqai`={$7JHP8oc64DszfE0^`&+~z%2TDm#+_kD}HxZR~t(S`4l<7b5r}~h*9N+?j zhdVDlHp?Wby*dxo-4iL><^x}r&S;J@9wwjJl($L^3eEgVfBSc6#*Xm1rB(KLh(SxA zxO+}TMef!3O1x}zRZNuCuI$$ubye0Zi?8s}{6*6@uf1t7Zz~_<>DMeN^d;p@mAruf ztJ?4r2>sRAVm%09rfC|(ag5V!9c^Nk|B}3k!m^DrhthL8{&*(RRGY7}QM~61h{j52 zYvM76SMrCie^L5MLzBjfBN_W>Uc2{hRfA(JJf-fE`$db=$055X4_jt`cd}R!zQ$+# zb#cPCXpZ%V1%O_gP=@CpSJ41n&cep1(*FMTe3s*Nx~o0ww6hVMLfzyO>;EAe&q}^~ zS+}PB!F$rgk+1dlh|*2Of2)6b@yaWx&=L2~Dyy`he{;Y?gxd7ad-*_eqt2i93R-q_ z<-jPeVa|ifxVBDvCgI?(Ibzey5iCk&;=%_+B$;L39AHingrQWLZRik=A4!`^0OvUA z72@X!9}C1c>7~c^@wkr`-K9(-JS1iZEh;_)ot-XeZ>kzqQ@+renkS{%NQg8uqP3Wr_}X5`V~&e z>4UG2w98hjAFHzHZ;#K|;1Iz0`?+TWzTOQSfBvvVE&;U}W8kHasC?XJ@xDOXz&L9* z-zNx@=Uz=Nw0EyD)=d5rhG35H9JEP>?J#DFmG%KKUak2d$UD0xks;9ghqZjbFK_OhbWC6WrMlQigP)Qj-6hOk9d_p3@s13%Fn%5Mx~z-~ z7W=pOvNrl>rs~*+%bJMX9Jbi4X-%fE-_4H>PoP57OZ(X>f0chreopgx+^`A8dOcx4H=s*|)0HSBC6ethD9p0# z+xL3_;3ns-&~wyiYi}sbyNN;<LjreV85g)=5@Mty649Ep+YZWcm3L|ld2r=Y{UZq`Y(^xKZCwb z6p%79?|12e)Ktnc5r3R2J$t1!>!c#D9R}+Etyv$`Rq#eGBka*2_n&YRld^U4tf+*<{f7;i;kVGY+2V&!$EOCnXGo^9eLKXFd>y~&~u%e|STlZKy=~V$oAN zA04q>DfZ=#EOl+>G*^fVT*Xv%59iaUvswJk>KyIDiV_g`a7bkArt z;(tYTNSqKj%CFuye<-*{7FCCjcCwFAxm9PMA}1HLg2PLyadc>1+2M3h0iY5=Xcg}& zX-MMrXq-=_^gXqbGDuS>4 z?NduN+Cezy2j)Jz)ZO*IU(jJ;p-ft8-ff}rf)53c$gN6NL_?y6UJ>TnA8ET8V;G{B zTU}m4-vp=bM%Jk62s-G#PLM+7jNAN4D_+L_mW&_=c+hZVRoFm2v23{qCPmVCurm^( zNpTLT+Y?<}e@;7{qOc_2o{=1sc^<5EE6jm|A62lVJ;nKJoS_s-)F-l~y10fXPPp~^ z^%UTZUJz+FtxJdfO!c665Vc9Vr1JZnRjXTZo&gi{w-dcXiS#w6s%^3d=MpA{3mx*r zP^T1AR#fa?4|z{&)-CwfIZlJnlQxQLWjZ~Mbl-_QeAor;(+}^YX6eP?sBiY4q9{*aNBZBt+7OZ@xjZT_Hdm(PPQSM2tgptt0uf z(no?he_!s}IpBsvpXM5!xp>0ENNZ#U5r5jXNGy1$Im%7qb`BR^wC;EV2a(0Q48mI>9zh%yZ$I5|(aLVQ-_ zs_P|}Qfd|4$DbKdIle&* z+X02ZDY+eD$;02yhF2?|{|IZFAtb7%30yi8c7xYD!2vr>#`oXMnXf~d5;sklt?&`W z90U-t%BtvLKjE175+D=XYnqrv`aCPd`{&e(;ce4j-*-t0R1lJaBmG{w|N6yd;=sWQ zf0}(b+;8hFabh@k(!8dgDt+%Q*uCR9#*B3&@k-0^@C6CyOG@MKulAN<@^ntl)DJD( ztAjgnO-$GK$y_)~Tsi#I5a3YNT{)b9aXr(`MyM}3>m7?*e_ zG&}6ZjOC*QgAN&wUQ$LclnbF(6Hi&0f72aUxC|}XALEDqBv!2Gw~n4`sX6pkd1*Ec zD&#Yp_!mV-^DpbS9J$u=@^K-*6zpcEn}k2lAFzn~7U*n~$QY8w_yF}UdVYaDPEuMj z%(HzLYzOlvq*t6_{oc4A4lyfspp7ptTZ6y4V#og0oC_bt11;HTUphlor+A44diXc ztMW*0u|CivPXxH_ENN%Le}0tFf1?v08`He@#c{Kk+v+l-hSm+vwj-`o!qy+q<@D^V z*-E;343m&k|L>Ayps9ml(g)#As@Og~{}ZR4ZvJ`2=i9bIT)>TOv7g^jzPQv@y5$Ys z6dc&D6%XU!0hUj+FEZGeS5yr25@kSD%FLU!9F9Y7!f9#}o;%YLp zWOA9qXBF6X^w!-Akc zDLIlV*?mrDg1?rfnL~*b*Lb)&mC^l!r@xVQW4cn5*3YK2t|7@nA~_>+ce_UT3@gXb zLJ5L`&0P{xo-ObxFLw)*f42XIG_|o;cayWN{(c{@7Gu#=4o&wYOSXBgC@ST^cOLnj zTZ`7GA|+8F+eL=a{6{q49Y6lB=<=|4Ii5l|hPdx!$sQ6l(Z$6uk5+tmd3f17q@7xj z@Q2=Y!+`^DNK@ZCds@h99h9jn3jC(}RYIEK)1kmKPR6aZy(oaoe;)1qZ$xf-4=n_k zpFtiSIJR*MC{Gb_vt8N=I~c!i52}bW&0(Fv8Wz4^2M&~J5=MZMB#|oaI=G~f2BX^-vvS#Fa zh>uQ9gHfmVkh(p~6YlSS8>-BK&QyFe)~UIk=#37Edrk~Ne`;JhZz&d~_~d+*M0Lu? z(~!Oud^KAli3|iWxpHB0NCkTdaS?0)%bn!-E1FfU9a zY+7(xzd`NWUfT3Q*Q6KoEp6(h?aFwitTaxK+r`}rrDO>Y{y|*=0RM85_v(ih4Ao;Y z%?bQn>Xwxaf7nK~h0s2%byh0)e3>|g3i>$iu^~BhA#e{>gB&{xj1`wO>YV6jJqqL) z)Sj0WMpYy@ZTh-OVm~(;YjX9m&d*WK9W*embBeC0|A@MKeq;7t94m^NSp_9&w8kXn zodmW0{7VMI`d{+QzLloN`$NZutOR|q)sOO!;Cz^Ge`G9y<2071yQd{mN^FLc5@fq( zk`xbsx6EH)+nQc#%VJ&1Gv9X_*j_{87n|kB`;1gPWT@cBC&4>*_3F`PurmL!h`)v) zM9mBrPJ|@u|D>s%y3bq8xL)S3XXwc-y;<`}eT_qxay2hH*UWkjG>sI-8-g9uXVHcj z?3u?ae~C^u83Y~;Ni!Ixc=`Pfn(UU^crIb|*|V1gLZ{-1v6lV#kB&~Cg?r0uOL_j` ze_6%QrG2XTl%7btO!hF~B65Jy*Pn6q*&I_`-7T=n)o0oyxeBS-)9(tOu&J?|Tu}a! z{Hqh;^C#4x3kPUdCzc|k zfM)PD3-aR5(5T$7A$UI77KemYHDr@Kzg_=-*HTNe&QIjO$cOX7&=Ri9g<;~&bM4sN z&>iEAYm<%Vk^^I144J}rHl+u@#l2X{ge$;MoG3Bd5`I+4#T%a$jYntd6#HuQK#<^V9>ia_CEv%YgPKDQ+-W-ZqfjggX%4f8LjjfUr|SPb*4g2Ux|~i_^7aZy)C8b|8Vu zp62i6%@Yq25B@`=+F7yrrBBE|Z(uqjSf^`z--H>dXp5hlm4#r&=nmz!fe4!;P2EOG zDbu88C#`~XS`k@n%v_51z8{k^%pp$N#FVj_iHjCJ0!q+a7?xIRyfpEQF|5maFaFm%`Q(-9dt@6qn$JJ?Xd}YvFqH0P->XwSK-`ZO z;EYuHxl3U`eeQjUVRLgaoWzovS?jhne7gGeVM{kMT)RH1eg+?p_wo5Pjgs1Z8Si|> zWCtpW^ENqF<1FXS$L4NV%M%O`3Dvnfys|%Q$DP}x5>iwrci+>Fd|1ITe_Hc2cxb`A z358^OpG~|=y$Wl5i|?^Ylz;VF!IL1Milqz&t6F`xGxa|39v)8UlX(3YKw(sTx60D) z5%R~krK}OVZrVcDjxhUg%fZEE+*UiTr-2o#yyqVuiK-ID%Yq`1T7AGzFP}Hbu}ycJ z+dxbcxyxc~jNQy&+hT0e?tZ8KV&nI5EeGqL$KB7UpKnzhDqzN? zS@^qqK8r4TkoLJ(e{vuLb40eR`-3u@L(P&kh$L5Nsg9&xH$k4637;KPGqcEU#k#wZ zOY+OI_FLZy-x0H{`Ei)$Q+wOL;>iwp?4k1Q+2HR{Zl6hQ7580XmBj95lr`Qz<>Z+7 zr~eT=^S`?5tX%m|@KpQc4A8>TiV~P;Qj_2bf%7maL4h#Ie{F`jU6C_^oTJHua-+A^ zvcZhPvXr7KCvW3N62b6wThq1=24^HIa#W_ zLb_*;Th4nhI@+!OlW~1xJGPzJ&r-P0AqN;7GYJ0vgUtdDSNGImEn*4ASrw2}2jmzt zJYO{z>lacCe}vyUbCAde)Ks*mEIZTN8`jdJ7o5{gy(Ig;2wi$@AX+Ypau@TXzNw4v znpntl^FMm`;S~ih?vIMT$Qa}P8+!cV5mdiMCI2bj(hGiU*iJlYf6>tMqS8w0r>ulLi zo;dsMYq{`lFs)qvcPqWS)QeQdM@l_kP$JtFXX?|XI#1EB3m5yEgq=Mn@7&{4DKiDr zei>{kf7AP@A+a}8G+rbn~86Aun$%jx~w?u}XCZ~w!LLVM4A2igbMfWjuL zxMsiU=@_5*7bZqJu63swGE#_ZANkmG3)VGnlS_G<&25m3`M=v8nsoCL7x& z*|lwQ;y~c$rv?$=@P2t5sGVY>BWT-p`Y+y~k*HzC5AipW z7=N*F?Q|pNy=kV9b5D}+)CYYU)-y;kDmoNDf91&9L>b!vBi@3=0Y5IKn3zS zaYnK+dM0yj0$X1Dm&7FYT>81M)(aD~*MISu``wn(|L7BVM;Xo!$C};CU|aCr1n157 zX}&k`Mew31A$#xod+xmQ4I&Wz6NLF8E6q6;5653VBGRmR>adpb^Yacd$Z!5aDM=wl z-`20HPZCObA^LAfoV9LZkn-qp<8QkgVgMzeHwn1WNBBTh-{;#G#qJw3La&JtW`C^X zHBK+_h_^m0y)hx5b*T6=)s*P9o=*Klx9f@Gv;L~NRHc-6W}N#TFAn2LQlh5cU%Yat zH=+Xxlvz4wyvOqgeqZC8j|1tbw9P8YiEOJg?e98(LsI`?z^12DGrmPrWTkH;1U6J{ph;Pij>$dW=dzdy6m-ui096k4PU4L6Qu{G8+ zDG}|Lqbxrq@aIYOKWWM*zS$eU=}YTLqyDXf=Rf|@!7k8^4vI;a{ZwF)kY^#Wa*QvZ z>C@a_|JWqQ-qfY2aokc)^=x>|^ljy)!T8woYNC|qz{g^F_CU%d*~RW2j%OTtCadO6 z5lCl{SoGdF|L;-`;@}dq+Dc_Q_B6Kce+%F!&W}(Zb%MS>xBmIOFXTMMjjd`(>!U`v6^rp^ zSKI4S$R`sfROoqDL!wt{NbvQk`s(J;UAS8Y{9D724Q5^qtoJd#(0?n-zN*Iv2GzuV zT7^t4enH}IIU(>NL(-wU^p3GM8sxgWWGzrB5kQXi6d^dci3|7X+%Ah@Qu`}QQo5r= z@b72yy<`cJEA_sT+T@1sfI2%a3q6{=$~fa!?>7+?qg&tE^^>D~uPW$(1(oj%Gk2o>wXgs6W*qXe4Owie?XOtfu z%ETvL|J*;OW#mBO6A^hT*F&LGS&bvlLK2`T-TDAiWLsMCGAktfcc%??OuR3|&`Ia* zjT3IOs?E}~X5lC66?@8LwlYt|XTDdze(z|5D$h)KA+VGZtq`KE2wDy5LvqS&FAO^J|RkCtDBvi`R0^ovu-MvO8TSf`@zCp zviXVPrEg1ri=VP92A`47Zs~A5+)RY(dzezC$~GEr=YQ@J5itvXS;qA}uf8~Cb&_Sh z*efozXqTGR~+c#iXQ51-?*+&cF* zLu!$!|HmoW+D6Txj6q_FzE=_|a$$hsbRW~({gecZKI<$4zi{{5##=XDSI8fdQKxe zswJ?Bu#p(()v~Gs``F$K?St6kIQ6=5Ofc8;)Lgv{q&DS$pcP)aD9~U#uR_Xa*{xx(Z9AeVlQwjEpx_T1} z5Py#Nk5wggePe382{qpQcg>tLyoS!RySmajLl#89Ul)k(e}lTttP0r7{I_c9cu)Nl zs8I4US^p|Vu^5j9AB^>V)Jgp#qPP~+Hvh1zL@D@PN5CBI zc^U&tU^`GwbwJQoMHqbea|XzKcL)cf2KbG@KgE{$>DPZR2Q9$C*U8vlJ~i zPx}U8qAk-;8m>r5Al1V=`n9U225p)KMb!E8&$TUB->`Joyl7_{p0B4DD}_YHe1DOo zylWQLso_vI-B{2$3;Rr+4aQH>Kjr@&%W^T&_@+vpDK+xdhK={12>L54Bq33a$Af6tVTqFPHjCDxX&}kVFPD> z$U5H1x#<3h{YGZsliS7e*kbf;lYb@-F}V(E=uHbe8h;VtFMP(j zd0!s0w@Ac%zmN07IFf;nYM3)nc^jtMF+p!380Eraylj# zcrmB(z+Xh*E(xQ~g}e?mF*zpi7vlo~8qM`X#U`Z1?lMW%f0*yLbAJ*~Q-ynJMWFkW zBU3W!bA-CXkCe#BU#f}+)Q43Hx4efdY7u?_Sn}FTcA_(q{N2?qgrnuAfAohjuLOMr z`{a9pjj37Qw&B=4+Y{MlKL1j&%blpzesRNmnU>7xm7U>=Pd_nuLVNd0Ied?HGdjXS zPR^%+M+F;5ekrSc|+`#(p~k%XG$Y4A5^DT=2ew;eRDWXax~-GKP`lBRbtz zqLPi)H%8D3>D@>vh5Oh%lmw2mUdp1RBaJ?Cl=g}A^N47be#hV4_ZcAh5A(?!(U+Rw(}IedQ%2FpeL-CSo~nHa@@jF-q1FGVfub1jLpP~wcpY?2kg*qw(Id?D==!X zF{^nk9RH9KOMfJI{L(q_((7d`SwPI6Yo0~V=j@cq%sNwT%iu1CpVYT{O{bKoA$%ing4--o%y8$hF$+mK>CrG8>S%Pd1vVTN&0;jUBzJL;z|(t zdB6?+g?~<%X#N9$hTHlE4R07qA90q-=lmfLGWc?zcSDGlJNpuD9wf3$|DutM+D_;J z@U~X{+sZ0fJLu#Ocb_>%U(@T5bysyX5xjok`7%zq^x9-n(Q^O&5ugH~Ix#_p{r(OKCy z0`Pc+ntFlKUyQ( zFt8>^ekXqe|@yIq! z-}MZ|CCv~sY-sIFHOOGI6S^p;Wa$Tuy@Y%T!3HHzHN3IeCeSmb6!!CdWvDueqxLtnBu4bFJ zj~n5s<5-yP2Cdfeb14El{XjzjHnVnPWy#9K`(0BJioq>+P{%xwYwbGx!l{$rHz0;bnQVAe838ud0T?baVHh*%V0L~1l#6*G`)cyVSSmtJTGE4Abf<NHKSSNLv$`GoL5XNP8!II($n6vf1sBB1Z|0LP+ zJkjg25oL(!Uj)TB6$j0`1 zU(Wt%o-E?Y;c;NlCN%1?QPN1aRi`MOFXC3^Q8fZEj}`5IAu|6y@%c!NKhyBJGava% zU315aQ_%gls-fH?>13P*GzYDtctc?Ynwibjo@i2mkOAg^_J7c*`^Fd6ImF=CM=zW? zMM+dQ-{amB;}msiCHfA~AzkL|>a!kOO#%=dF!PgHojuvlIrNTqc~J^~CarT3Zy_mO zpx1TE_pE!QmKq&g@tgFBgA`Byq5OdCPp2Isp?#`go$Ag1q4f$nRL#$*_x8u78k`nAoOe>;DN*|}De?!p5)C}B*ql#t=+Jn06x z@xN!C?pEQ6R<6y~+^zKP?Ey=i|tLFxDY%x%*1Uh6+ePfom{ zM>2|Uy0F}HMx2qE6@WmD1&!7Vp?JCArhqLu?#0~Nx2ucp3ja{`kvc(%Zg#{Ae*uSz zXirPcaDQ36@`CT&N9}`*f7N<^H7X@K^WBx3W%FhSdIXeJ?u1I+cJSb4=e6jK8ea*b z=-)W^bUmN!q?}U}#M0{GW*h8eJ+ruY=k%~|(#j%TGrQYgka1_4zhM$kStTF(iUBcQ zYVi~J?3xuB#^fICb2nER@+&UQ*mSpQBkFb?$A3(JqtT}}@}mN#-PIBC>7PgM`*<(_ z_rYgegt5to)QzT0jH36co(65f_Fq;2Yranw2hu*x6%x z!GFODNV9isi`_W8kEEL@*e2ZPIKFF49 zBlS#Nyw^_*suOF5gw4N%IgASQJie>M@9)G#VIglW*Aa2I)Mj9HMKFazr~*H56HLTE zpF_A6Hy(2%u;>0bniQ@~j4RZTAAuFGsee-7TDMgt2$r;s)T(V$&vCw0{v(~Ra+lkS z)D=;N>yH^VaT)aNmH2#adXYrCP$|KWl|%zR!7Vq)eNIkszLH}UaNqAKX@LBC`J?+v z&tn&0(Ezw46WPt#s24Hxoxy4O>7KATZOH~4vwLQE^0Wn?k{lo2t?!m~eSbt5 z;_$Ot*LT+`JWZ=0==@_8LH)`Qb--^RD@A`e!y_}LiFV`kNBy4|1HxHx{2yz-`u>a^ zK>RxV=oe+hkE%+Wiei=LLr>Z3CGI?riSMqknC5#Vx%|`RgVmFP*TUjPL%gzP^r(5b1`8pp1EMnV|wEz95KjV zn73|P7~t-qfQ`!-Z@-;;tK1mG#XU#v{k>1qzPjga*8zsvh)1FEg@5oWknA3BAX)nc(QW z55`Tuc%C|b!|tyx>Nu{~!O^?VScrcuovLFvmrT<^9C?IaFPdGYNME*5_f?+ryjVAm z^|tkWo|GGfH&j^_Jva;l_PJM*jb8^Ham6#QMynA?uE36>**Hct+^3@W`ACVjA<+A2#K;6;G|CA}DaKCg5FM^-N2%Z}Sx{`|kDp|*D# zC)h?y(3o8t4TRhmI2hpbg$k^b)4^S7fe>oAVZAMKNvf-`@_)A4)v^KIMQ440AN@sdTzQwS^UBqgu&)IcSv=b1&+TjerN`<-|Dv;;4e@g%R8!~2tMd9EQ7 zM7h162f;R8(SPv_0}5~0_)mr{-W&cpgJykZO82}tojFWfu(yaq9(4_NOQgTpa{fbn zLD4}Qi+jLlew#!2ZMekG->YBpog3r#iZ#F~$P+ZMFK(m}OIdlf`7Y&8C?a|_{J8ym zZ2p*`f$gaf%tmL4M5j(d=*{QGFa7{67s)5X#n;#Cihuu+{_PhV2Aoqt)Huf^riAx) z_!1!pqF?xjlN6 zV(fCLY=6v6{LoU-)4SIBe8snOs}gu4SV8I{KnI+hL})I#-p;oH@&<6Y*`EP6+Tn_> zy_mC#r;08UO5vPK!3u)?W%cH~t;(lMVEeB~P_=i7dtdUM_KxOZwyT;cn`{>uT^me)DA|kD{v0dxWP!w7LHJ!+|FNt`!j>NnaNiQSP-w@c ztN{N>z5Hl>&rkk{ar86RA@Fx_42?syWRH)4t%q5+{#mIbE`m9A_3hyaGk=GeyrQIb zw||}9Q}$R7ABf3|YoE^?ndiOEK6b=Hn)KjLl%oBC`YUg~8b|JPd!m33KI?iaW7(=4 z?u%7$ljCJY6X|i%IzmLX2Q278WvgW~goM}wYJAbG`8O-S?!{a(lN{n}K8oPC(tbcp zFw{x^xY8I`)2FD1ROJaE_Mn3M34YT#Qh(J!I^SpIkEUAOx2UDEHdBheVpfR`4uVIc zqb7Xi!ymrZKD=N5E%Gknsm~R2#5WcG?dN?o*r8XDhqr|A(*AU^wRNNfV9{L_RSr^M&!IrlF`*Hzca z`)reYb-Su62F5n%|FD!^b$^FzK4nmRcbC%E;h0>w+xzSz$&AR2&s1pje~f*D<4R-1 zqtiz2SmxgHxnHQ0)+&nIT#=S-pxNm|6pUhx-y6H_>w0f%Sa)-T(WFfqK%2R{OuX`? z!+;uL9zY!Pe7MH(D^qU9zMFz=Q-T?G?+4{e zO6hM#5P0{c02COp#X+BzX_M(_-S;YoLte;!hqo@y9j3oZ&YfVKFN3Wt=Uki3`Jci4 zxtr(a^$Pobs`0bxoPXPj9sOSYls8cc^*<{Mtoozt{q>Vs?+O_kBZ3oa7~+1FICjmm zu0I|~2nrnlN^r6i%i07*o8~IL{3k!75Nxps6LIrN(Dc0vX?Dct{PV9>VFpC9o?qV| zmEBh8&i|>X|i2Oa@*eX3Lm0ABMJLCBEFn_mUOW|lfVdhYO9Fv~o z%EIJkJkKOZQE!~~+>rfo+^?L2;CK9VWr;TWpusTt6s9%S5_rIrRiNymXJCz}3|67B ztAv|W*+E**7#^J1;24lF1&`gXE{_r zYpZy^eEKtWmMz9bCnGhS7pxS2|iXbk5bOHj`Bc1-f@j z>CJ0LNBWqoXDZ4k)?Q~D9Z$05+`#wz6@Sy;d@!nXc`t) zn)s>Xvwy#bBpdxWHSm_A*;bSwyfjY4AnTSU+VNeLuhGM?NiDVt=pdKBvfy0z^#TDw z7bPm8p+bfv)93-KUK3^FN3rQlN^~;eD%HcHDalQ1dP%7};c8ZyOZ7=keY>fk;intR zw36!iil~Byjnod`IFAT7bn(hk0r?!TJP+b4Sbwh1Hew@WSnP(JD;&aSk1S-(2=%)> z;j=C(yviok`r^wwn-k-8Idr51#49R?VRXKYOXv$yig7nkUW8x{^C1KgGs4r1^h}qt zPAAvFWj=e`a=h zyU12>Iv4dlslS;_=|@l=)rjjZTzP3rP}I-r5)V7s#5MUdYH1sg*RsdBsi7**DfQze zqE4mU*P_Hu=6FlqIxopw4K=(}sH7V&rhndd%*Y99ItF7<+SZwYq-x`e+zNTiCfkT| z9mR7CtXPs+hWyjfc;i{Q2Fcc+Tzoqv)uwnyS&N$QN+0ABiJAJj9!KGNFZ=~6v9$Ec zxat-ad(|%?4TQ95d=C-t15Zhg=;qp3H-bh@CXRKc0_KytznYC+jSzCky8Vg^nty+P zgbfGXKbm?G%Tf9)vDSzTCTFgC_LonX&+^Kle3-gum8R*mD;J1|9};NIVk30hp_ z+MwdZ$_!^!&8X+avDY!#NveNM8?Tcci#cqeruXe|?kDnIq4d)3#NnY3)_*fL0Y7qh zS)4%r%v?8vS-)(h884KrUj?Un=2sMp@wiToyyaWD6AYz&aTCU%T*JWXq5W9do^y;K zssw-H+r@saqVV3M^0b+tlh`=pHPfb$Ex1XyyC)(cv3}OddEewhZ<+>8)h|AuuWy3+ z&Lg&iHkk5P8YVL+s7|)6>VJLa@82Rlj8AGM_lzXw=o#`(*NzsM9U%@Fs!^uuJS5<~ z!F!PC8OSyE4p9`f(^m-!>{0TZjvB)}u8@HU?XYrNNC#IL47-ln^`9s)$y^%xDei0z zFlI1$|W?IsB3iOTe8=7 zVGXrXw}wDa0;3U0@t!3$j@fMPcV}&SDnG4HRIjnJ_vk;|1BKpsFeF3mtpaL1#4PsS zYgtnU)gtYRx~*A?VSjvvG0d(osSJ4{t6#U1INl7KcxPi|ef#mGK*n$+@+bF`XRp8ciQLs}LmLtY+`0ME z41)J2sMt8<3YYb$`zat=pKKfsaBoj*sS&t$LG2OXwR@O9siA9|H6(~K- zwl(RE+ei5RYC|K8YkJwh9?$h}LK7-->`!aD$!>r9GF@cd;32xrg8n~`BaHDYM}uwU zU>OGg;dntmyMShCqtlr+y+hU{8+c!3)6gTyy1X?UkbiN_0=k-jSg-1w|8Ul2OTDXj zklO0<^?>nl?58{VSF1V0qb6Mi_lZPxS}qpvxK5EX7VEn&vcK&6kz!Z;gNQ0m)Z;8} zLe;`Q{ovu_b$X=;*lUaDmT9*wn)OJjO#qLQ>ZZABX#aUm)dl~T=cL61cMB4EA*#!B zc|2bDhksOGYmf^izg;`Mq;ft%IG|%wmdD<7u7?tpyVvk2Ht-qp=HIX>kFmGg$RFCV zq&%c&5*QX958Ynzb*K03vxHNf@jyhq*8G6p!1-%FlolG0 zM_-y9dY*O{<5K%fgx)9B`yz@WN7+kD*DkPjTY*HRg{@bxK zud&l-JY)zh!+x=4V83VIjE2rxH!Ud8_>1PlRrJ+cz3tUw6a)+TGS|AoRP{-94M5E+ zjoV`(!-sP#lFtfozj+-$Jo(9ZNq>6)K*F6DsrqlQ3is&WS3b?yv-A}B{BZL~YwEmH zLVw5KU+kX~3lyH6?=|gDJ|r{DO!wOG4D?Q9&P$YVHE{5)MVnI=UY=PcFApTwd+&KG zrMV7CS<8Xn9rrR)r%jS0ZwJZm)*rt!#399CB7Qwc8K6+7YG-wOCYxB@d0>&2^V0sz z52#QK1nGKY4*fbkkG^d*y?dOX@Eu&gn1BDAJ5at%>O8(rK#X7&e|OOxAz@Oc|o}F z?wxADlh0Iu5#w;{E3{DVkXayh+428mV5PmSFV~@}4b!>Suv43_} ztZk7b5TQEv_V!Nz@vn{92(uVv2kL{B53WlhbqSj+U=`LJ)(5wJ{=MVV-?BquK4Lnq zyZmHgHiC|3Y^sv?{%b`1k87ms@p7km>SCyBibY~3m=Z{s4ZP?0fZvb2JFM*XxqmT! z^BeZXxKAN3@V^d^U?aJ0qlW3d+kY)@gfFqh)*~|Zw_hOZJ9q#8!?{k_kegxB&^%x_ z?Kx`3@>%_Mh^VfS;9K>6*rNLHzio~2KO@fh{}^%qy-R-LL4lBt!6{TXx7#c3E$L24 zS@UYukxwo#I50Nw|KU`QE0KTR*r4u)crBzTbITFv&ya%U0j~eE+5UgNu7BU$Q*&VF zsQy1x`Ty;=tF$!;$McF-Mouj+6aV+m{YWf$tmIzPyn7QaJy~shjA={T+GzgUzm@*I z@;@DOR_=cnkN@8FC5z$1!st;|0^Sg!vgGhrU=0qg-+%wmvfm#S{u#6!FH-wWv~RmL zA<@Czdrya;61TW>TDch@*?vjPX^v=pR{fA=a&rs_h%1~`+uh?ss5Ly{5J4F z^vX2DYb$Pg;tjsLvWSL0bG49;+V;X3P66Md*NrQ2a;$AX>L-<6?GJHW>nhx_YgQQK zStGE9m-2DEl1`!~cy*lQb@k`EFtB!PdRs)J7jZ&vp`UTP2fOPWb9ZqVKx?}G@~|5+ zk#=^hip+-?AS)K))qk&VUf}5JRgRjR(zPp6lUqKn}!+$6B2x!xUed~3Z+k%x4 z@5zL!g79OTKg^zfzk{q>??}I!$StYg{gQ7$|V{(r=vAAFfRn=BJg~HC@{A z`eMa4w9ChVDow~LaIsumxfsZWUbl)Mg6bk<+k3MmDY|)YuDFgO@Cjrff%$2T>7&?L z3lb~eds1C@(0`H_XZ4mj^XWm>kR5yFwQIBtvU;=3dQ3J2(<{73X;#O392EAMrK6DJ zE<|s9?9wf&(oLi*?_}^>*ZHE(+&oXx#s!UZ@g=dKO3PIjT>bs9$VtZbnCD%Qe1PL{ z9D3XM#Xb8g*9Uz?)mE;1hPcq7$L6p}^7k$O+qp-U(tkJIVleWfSjKl}o<`Fk)|_6S zbLh#FuW!0Ol)CWWKwZ+~EwAp~9ee45r*trN=Rw|}8$Urb_`Ruie_=ZyMp0A9uC-Qh ze>Z(_5hb4Z+f4IVdDFn*m!QLS%fzp4s;AJz!!QmG-wUMsq*)~*RAS8;C>lAM-=)=r z+VkE1yMLw#2%aWO^E2Q0a6O8@9>P`|Eula2<+4^CQtU-pr}Ri4j8vHLN!w885WBXV zY+X1L{A8qTCtbQOl5yQ%q~GK=>8@1qQs`8(_Mkm&vSwUf6>LzR@;NsP(~5%q>fn;& z&@TVtbaCN!eKxV7yI<1CHmNQ$zkgbEVS|~^sehOuMh%c$-lk2;VEx`PQ*Lh&s*4pm z?N?ukMKvhx8cQEzLem)wMBL!af)u#aPBmVCqZlU;EyL<5_E6;-F=$6exUl%b^uCen z;lq{*fG)_qwFnNjVcU0=bHMkU2Q^S{H|_f=-5NaCzwdSN|Ew;N?Ed96nQ&T|*6_PR z=YOv(fd3#SvbX5DJy0elL1vkv3PiJM2GGA=Jts{^Wy8>l@Cv`Aw8^p5PLFS`mm z$ThuO9AXv-^qZspR<&)blV;;>s^cD-WM3-e_fqJ8jioRDiKTk~ODru-yrU?$=zo|1 zp6FvK&Dh)a1}~W4J=9}br3SCJZ68RU+LIc{aZK=pC{cQSc(}>`(8BbP1SndZ?tA*M zc7L&{D6o--dEv+NR7agLSH5}c*vx5do0H^c_i_!G&KBLz2CV}T@r9n65j0x!-EYJc z(XRj=A4I!3ny5xj z&&7+t7hBocx)DI9xG+}vsga4tyvL8vQA0xRS$}C&26+fw z-PqQfhRj7sX@lK#_x-G&zb+rqHaZ+zOz7>bLfgw|{_pG6VRIjF(h* zC4$wN5Y{+@7t_wf`}0GRD}O+3lOi_OPyPKsou9fJcjf;|;O`#Yv|Fd`-y7&dOsE1U zDJQM^=np1)IL*9dT0dpjJuvV=PyVxoq;1n$x+{Li>3J&3JLP~KcKq(b@x;LIXt-Hy z<6O0P*dr$FcW#2A&CH45W|l3HiY@wR^~R-e8#GZUlErxxhjM4i*aGb2w28rh(jF$cHJ@#IX8q4zg`?Q-<>~?GEF|0hQpf>)j#-pzG>~{CZUU(E;E67IK zKIZU6L&5Ycnb#Tz0)MvV6QVhSTUQfx{$ss9KAfiAzLyJ&NH9#oaPNPYgW^1&ObN}8+H%-}sfUr(NzxBt)Z zg3q{3M8BZE-FyKV@@qLEv>|$9%E#65%_qT_|10|_$=sp}F@HPIfn8a9{B)6Vz%-j>Ip;m$-}yEN1_EkZsQyZ&Q)FB4g%z)u=8*( z1H{Lvd{XS{9Cba@L~`z(hlyC*OsG}rNtQSjF%A2v@%ZU%;I5nslMwTpoRYV?vuAJR zK0cd$T*xFBAAid``x(3Mx7e}iW)9mNaN3(++TYpXjp*KY-#=SS;F``gW$#XH3SMQ48^SdmE<>7y<78%feWOQ-&RS18Z3Wl3U- z>`%?pgIxU=N6Kð7YqAcc9URoL2?Qfe~M!;d5lR)2X(jn^&I3g`Q~hbyWU*_O%Q zs8_E3W=AxfS?nYm{n)xd#V-k?cAXU?nA3b4&o&Irf-1bv%OmQ^3_N|pdZTo>$p zcX*y{3V)*pCTZHx`ukoMSBY&-tIUX@F%L9Cb6$?4b(iRB@u_0AEvT^nyA)vY8|-VGDm=M7iIPDfQ@zUs{1L-Pq~3%`6syA@1>xRKyZEy~VPj4a z0|T_0cDgTSOWKD9^E>=o<63cZ2{rFkgrD%%#((+pF1oUCKP7zG7Gc>}f*lrxm$a>Y z+zLDm)Z5f{=xuJ+Lxte;>N-?md#$?2EglVxA4^h2P&{>9{lz#WBS=71_nC#qE2VE9 za*h3(V!mCtaIqvN=b{TTY!wnBWlY%(YkUXz?p#2m+=8*zP^?bz^fWk`LFX>MYMTxB z7Jp*6JTd@O)G+o!)y=Z3RW7qlObR2KIg>Zj>aGMKE!Nsdu}NC(493*3Dww(?(#XpvxRFiKK&`T3a`$ilp|}`)@{jJT4U9P9<{#drE1jh(tp*d zdp?&?_p9p76K7VD3%&03OTQ~QH_>!qF-Xu{AfUVjU^PA4d}_Tp2>|ssdUcQ4VXUXr zH??b%%$`}3Z3*W?)3>LUL}*wx`waU|;ZE90-frn>ydDIFtbA78f!o~0_bGHvu=o;HJ!b&Gvz1Hf3Tp70XL=DqH zQ|}YZ)keN^o%j=*cgrB)*4XC=g;PQ0O9f%lCZhzJE96O2K zMr&Qb4L`XCgn!dC@caXQZQ~wP7Q{;IKY@ECc;Xvrb4RQca1z>@dM#RG*w>va*1DKK z0bk2zvxcjbrOZ*c7M2yYPFQ4}4D(LEx&*kSJWZ|LW0)qa1dp9kECBPZsK);|T6V`# z!4)o2RL&6G!@`^9$yPs_2Y$AtfPb3%{TJmv9NEIHLoXtDcGp<*MgF3vcl(ezPSnW6(ruc$W^Jk* z9#*j5l2$tW_^GQal@Ve_td6gNgLE$3s3y#8D#xx8OJLK=OmoHaiCkS!VcXUZ+$?kF z?t>>E{&o8{%J_bFW^}afkL19pZ|L&8etmtySVh(W#AkbJF@Mi+5ym%-r5Tbb9qr+z z(EqZ7giZAey0^S~zGNQJxXo5|espPOp60x=$S7KEzRx$kq?I~)u|uY*KR(V>suOgn zyDZs%N+fU`)N&K~7gA7Rb%5_br6XD{!ElwyoyUOmcr% zaP#{jSHE9TtA8lMz1K`^vFTv7UwXg}9UAuJ6=m(DG&y>*wsmX4aZ|q~qU1cO9W$lp zy|&omcRv2}Wa~a9LOh!w&8OBS%x$s$|bqEU5($K^CW$FUo_Yq5oRs9MVpEGP}D zLA$lEzIN=ys?(fb=$To~MqHJLmR1}Uwb=N9!La%yC?;t*Lo_W4-GP1Kr!Crr z;q}W|a{fDYU}S^?Wih@4``k4&{iryM9Xgr^Uw=ovx>$?(?l}^)H1B-vliqTQR&=4h z$ZDp(5aO|}q9Su1xag`hMfn;naGdNlUKq$#P1?ArX4my@z)81rr4sZi-6tofPmSsL zyX9+F^G@W)8(C@~HPrCwBKKpU*}%3&nr_{Ud4Br#;Ec;EbN|>X9RT=g0*LuQZT-yKa1#psph09+(8ggH_yjYac z)TwnKD1lAk)S4B~1EMy&#V*l{_V&$YA2mt_;tfliZ6LD^Z`Mm(t+c7`YrjYRJa$c= zI12(_|7~}7+pQ7FuhAb0Cj>O#Wxsx9a$BVoZ7-+as6eIX`6pjHHw)Ig?{-MaX};Q;?$Wm+dVaCLSvc3Wb?gya-LVs7_wt-E|9`p0 zXbsioG++GhjLlYnYv7bNd&ev>_NXe~J~dXfoAN7e!8_~D0!C!If! z09I+x)a@_}2)HcYxsl4kn#9VwaS7{e{`sOwVOA4n3sxwjwQ{z)icEqi7L;K6 z3oqS{9T|^Zgl8)A1rgpsR9lSWet$GNkk$y5?Nin-*>#8JbZ{L_Jx^Do>oaa;RM$}f zXuRMMpB)qB16Q--Sv4p98kO&G!rMAj&Q}7?nO-=^>4q*1k`yJFV0!R(|^2YiE9Nx zA#>2!ojgFdeQ12~6g%qMFb-3meWY(l?37vulDybJoPp|q-H`83DhrXfBXYaYw za=$PWIO)#xF{n{r( ze1>fXU1oih=u+S0WO1LCMdmX`4 zRHbmC0iQ6{ug-vmUDuyR6ZWk+K~DqAM8PqM`LpAJFnekwDK9aO_Ab(qj_t2J&S56q;7==1V zib6>`j?(kdh)38Ym$JUmusoiNTN8=w-eR`V5 z=Nb0|p)o7Z<=in$iNMA?*VhYPMQOU$YJncn5v z`)lGC9YImi8Gl6P8UnU7I#U9S4XbNj_BGh@BtZ;4e&|}Zp)X`7!T>Ws%)eZo{S%mG zZ6fsNGlOW7!g}VNX|&y#&xYI;_k6(%gy=_B6wJsm{NY{u;?fCe&-0siWo zz>b5${VaaA56@=RMUz^AZ0_jxML*4}Bh9_O;<8?unW_j7xNof)HIU|Kd$EZs_5)Cc zU6jv0iUQW>#$$gRM0H%#I|E6tnRxpGSoZi!E&NTXM~cUe77Q8%+FsvQqqDo+xFG8v{0twnR`MQ0FEQ z4o%2PL+BENn;b`Nz4t*|Yoo4xD>qH0cT^zRKtDvf@i2csc1^B-=Oq?P2I!Mfns^kz zQd;jL8oU9jTS!VMoD{7K7TSs(Ke}w{=5ssMOrqJ>&bQRqSbP}<4n!mjLRa3pquA;Y zW@rA@!yCSiTMdCx3m`+=ip9CTNdMFR@qR(8>QM&2$Z33C`vxmNDvG;C$P?Il)$b{v zW7W9L*{Xl1b^F)_k-ZIF`(YWs)}td7iwxC1V|)v`_m`Nnc7@AT)TH?EB?{u{sZLF5 zxGRbNxh8mJTQv&TxSqGDQkJ0UkrCDR$V8)_?)b$_3TbSnxnPy;5OEq6e~vV7BBHk% zu2t~lL)rvZO97fv1J4n9U>vT((&Z2g{Wp(H->HA0|9ka9J3iYod+YixR_IzRZs|_~ zR!h`)(~Jxf7q|cLvRtNl_#V1CO2p84Q{O@MDvb(Q?Yn$`k44CK5@JXPvu)mj?(y-> zCr+c~*eg9m8wYAzVYD3G6o0l>QNVuGykWscm~^AQNog3@n%HJ{@?_w&{+aICXMo`G zkPv?;1>&wIrtj3Ybva?22ihi2+fSa*&%DeDoq;W$>DZ%uGs*tkS2lm~SBgTmNX4sJa=SUJJjXtJw{fOoj%ugK zu&$h_%V81J`dm~8D8=@{CbDkfO}V=osX9?!;;R1MwD#=SBav|nI^t_=7WW06H@wQL z7WzvXJkDA98;dgC>{vo{r@_R|e}fl(?38EMp(k>h?V*2&*zqgegQ%Ro>bj@#4OxFc zecU=07FT74ak2L=`<6>!m2GeS-~DXB5n+gx2OMwubdpySx&8 zM_o5XWg*#Y#x!h9w)FbKQuvYgq_@V|&U}N!RQk;*8@3>=+W--#5d$x_BJk-H1N+j( z7L0e&^hnhb%8mDI>HnBjIqxoj331Q)H$~5QfJ0cahh~aY?e6sC|~)nckQ3>fuVm8LB$m4 zS_$lMr#i&XSIXclFZFtE>P4S>0fEK0ho11WoZXk0^=`4eS-hB9AyL>7-d-$8IJ3|6 z260^RF79KrXY1CZAvbI>@*u@fpF&UHGR0OvA6}?m09d$(8cv7^hF+0P_rV1f&Qj92 zRp*vK{V>4Dl&1T_%l2C<>qLLcE!Lu>eUbooFJI6jOG_W4{VH8mjrv(e2s@ob zV%NX#RD)wx>UQQ!82t+rigzelOHZ5t5=RD2ehX9%%bDmC=+e|X)1u>`q`!hI2Q;r@ zavN&c6HqAVkba~2;)#t=V<{9a&^k1x>Y!FOaXOY`186vp8Mw9M%27@o6 zi%wih&8lF}Z9@7CfL8U#M=U~C|KggwZu_Dg%~n{D&>1!aa1Lt{;y^0-91AV5jPSzIDD2H?UDdsDck|%BmfGFUxiaP8fH732x)3zVp7osF0{+AT0Nhn8HkP#!R}!jy#elMVF~BA_MYio#c)&I*AgrrP^;C_cF9 zefLXy`@Rd0U7s>xbFsXEeK3}1hi{y>S}f_7n>G}XYc+p0O@i0S`VDEZV-^;ksQ&m| zF`Y8Hl07?b!vUy?7HAFnFIXxK;hM|4I*OoKr2Bg3-lNz5y8Zt9{H65n-@o%?hmkDO z4%CX+^Y>W?yhRz(BqL0US2>31_6HjUC5X|>q1NSSAM#6`YuLrryrDeZQmXT!6xtml z1@l$6jgfz*O?Pk0JyK3M4)OB)C>r)TFDfdIIBazjc#01OpCL+NW2&$|RX2YC&B6yd zX-xL1HLS!xQ|zc%I@g+O_GcIc?dR-jwCYq_RK$~vr)=1#`D~ENr*Dm0*P(_*HWxfx zzoY>bve|&$qcE@&Sy-qx_0v(`U*%@R!siz)(*}PROOIwSCyhr1`#fI7Vpgq|6EDH} z%5GrQu=A3r7sm@5x_eqP(z!-E6F>2$^{;isX4@}VU-ARjLMUww&gZW_F}p);|DH50 z3B55ioYJPmG>T=VK;_Kzp#ZC6qOh06`&Hq*7xmZHbPjgsxur6@{w-%h?+xwZCa0Ls z&&Yp>nk&I&Q4B7uvpGJNE?*Fm*DhEz-96Fc1H%hC`*@^kL^0 zC5z-7t^&g(H^8_j$pT_n;IFnP);#bT-N=6<;V*{H?z41JLE@&UMVNj2jdm(p!eW$A zGq6*ck+8{!>g*JkibQ-1PAR30*F{d;OA>OMOJWDhl9$8CuMAf%q!2-)@=jQ#F27cU$;5!!2P1D0~yFAc8(H}jWdTW zs^cBYp32XcvivCSH|Dl26n2%ZY=(c8MH0nXP~`)nbHIk0i`93Ig%?23>twd*)}m1B z8ZpXQHM+rz80ovJzVTY-X4h-jEnG}9Qs6B@&$Ro;RDP6c)z>y(g9-CrZp_Q7^-z@X)RWGW$Ay|js4r% zzWL7mXv#7B43Jy1qXLg-J_$!uW0Sqo)dgLVS&dGR#>;nSbC7q=zLYYam=2o#{s?if zrth4bzHjMG7~`%r1nGl@{0d#A3JBn;>^ykdmSgD8a7s!A)*AW0)sH#9S09xsAD@^8 z&ZyNb?WPvi%+<82_QP6XGp>Iowz|8hB0$2%s)5Ftrf7 z>JBJQ#%5U0dx0=e-DdrhcXZ%DljUu2k*T+F!sazvyL>_zf*!Ogj@Qojiiyx&D=zIA zqF=IpjlC*pvisB`%~6_nV4+VnKIvExH&@A)aATbnPID){y}Ul6DRdC>Cz;YC$B?Qx zuSKAhB-x0ZAI~goqRM|7v<3i|LgnG7HAus2tnsaJW>sDX)4ikl+`qpz>9b9&B4lvw zf?NhBPfPEeUT79g8M@!Z4N0_>GbnlP);wd81mrsxjg1|^Y%B~}E6oaNWUmo@f07-4 z5i{vWFL%ju_HfS4bu`av^(hM(^E8z|H3z7!VJ}$@Zl8WtX$gPi822ppv9%&ck{~Jd zBlyy6;qVEL!drvfmyN;2UPx|}R9EF{hS}Q1^krfqPdpnU+WWr2R4-S=6qKdi3`YM> zH}>!0ZO&Q2HHmkz6;b)z~zls`=x z?Jr5F#jo6_la5;zzWEUF>C9J18a*iqXIJXHsb#ewImzmyJ_!R$uV?*590Qs3PN zo_Yejs;>`!p1D$;G9&CdKQ|L6<7oAC5ZgiN)rLx)hegMCI*7%h{mMQViYWj^~}PcejFw1j{auG~G+it*Ctd5+At+K;=%v0bDi}yBAjH z`;&L6ohF6xP7}NZn1+e%ej?ZGcPss1wt(oy*2aG?k#Z&M1x}S*q<9Wfc-2uxM<*;5 z^}=R!bcm%bP8Wsz#+3JskC|B(KA!kLUF$I#nZa|^5J3v+RZaIlr!iF1c(3pjmP8Lmg!gA-?&2Bl1kumgJ?sr+5E|kEqEl^7 z9RGjLXV$O+8{77PnFy6|UOBN-Nz#V^mZeHm7KG8;zD>Q6`DA9zuR7wQyr@;Ip-;!N zQLJ^B>~i=-n(z8pGT$jOU@$gpIfWzNWC60>=M8wK5f z_TK606#voBDQe}{YL;VwLZ8^y^8WHkJ_JVoB&6>X0$r}QpEva)ldHt*Gt!ny5Dv(myPl2qQ>AanZk1xuCDS4-I;$>aL`3c z@~#|yNi&wLwMK!Zd5+k;UrY_;y$A$Jl5weGvU^W+<0oTb8fN7nZ=L>`tCL|4CD0GS}mF;SfUg~>k+@e3=bIokFZuewsJy=p{NQ&ED-6_9gzpliz#c2pgs zY03T$<`tHw@0Pebj|xTgk@tTIOO_3*XE81YQxOC_zy^*HXmJQ2Y)W(UMmjaR{IH07!+4WBY}(>o0QAAo%4UbgC&mzCMs&T>HzmV2o|roy2T zjhC8kgF@lYLp6vVJqG>wo<8|FD;x@l@l_~A6iK6M!ga23@?F1xYLP>ulUvdycF7nz zF*LVL4&y^B>*GwN?oxlT^2HnaqHHVcEB{`T!fJvJU3m1<)kRAtTo7oidOFiRzcoGU0`u70ep|2r9xuePwG7X@ zOS)6is_o#3G>Ic#@Hmv=zAaceiFtN1G@+~vufqc*J2NW2B{)audXK5S;(rRvURX&p;= z{4Pg1Dtb))ca;L3#FRx*xp52?W8<$KpesTjRW!g1zQ%uv3}YxAB}6jcxhiM_ylHJ+ z`j4`uj9)!UygJ>%356=407)f;nNKt3u>i$+(LNWZmhf6lcX4tPWVUY?!AiyH8>qwe;Fms)RsS zcQZ?l{mg&e@T%hlLgB)3r1xHGoWW%4eHg0cs%ZC|W62ab9NF?16vLM923+7((if<9 zHiw_OThG>=@PfS0Nl#&AgD<$M%}yv-ilKVD-PVmiCSa3*^BA^G$PS36wiM3xlZ7-V z>%@1zf)3>q3)k^Y@7rgKz*OpZjyYx`gX8WB`4)ebG4@%_IhIEqVS9&A3Oo1xW=Ae2 zY`%(bCn<>@Cm~TqYh1O9Q!yMh5fWk)V~r>G3B7d!m-0s4q5GFdg~K2`Dq<)bqOoxA zT)}mcSg#iBg^{tX+94gBHv3wgMgl+m_UGnhl6jchz`%AJL_7TOu=K#bAd-{y!? zYG{Ad_WGum-LQNjL#ptR_=pjkg&U1hcea90Q}ktnH73)cmpe4AP*A0>g50!UZItQd z@50;kkMBHH%H^`^wTtamuO>iyNvis{sF3oxNZ>vQo5s$9Y7! zvhW|gBmGTAvD-OSN>gLK6n7Q_(j)^>loo#v(P^-kgoC^njNqESAgAV4%yDb@_ za9jCAJ9&+z{^QxZUjI|)*fhMYbEsf;S?zsitBz&*T>~Z;K>H!HZUosNTLetuUr>Ly zllA`%>3-hXq!l2}jx4(0MJ-g6@p<8iMt;xuxM8gCrOEqvJQ^PrGYCt^j4CF zqGW<(_SK-a54Jf#Fn!vDDdk*cTa$m7Fs;HUdeB(U#2kB=)60}y=}2ln1Q4p&>O640 zxH#>qRLABXZ?{`(^;N3m6|;UA5iOb6$ddHQDDl@vt_)=rfTCf4p*eI5Cxg_e& z4K*^x$=n1M|5t`Ae%mNdFTAT@g*m&@Y9G}O>8T0A_{1_>mWka%vU z-*%41>xO%bl4Am@N*YJ=;K%WofAcXrBN6woz27O13z}zgQb;jT zfznLJG9HA;9W8w&^`MD`fpNer!17ov<@i2f1M{bfK{a<0>2I8$&tf5Ik6qUC(xMeU(93+TsncW1&Z{#w&^HmEA$w4`kJOnDNjG^elrBH zHMQrX+`_`e@EbM2nX=hmwu98Y-UW&|*=!5Eb#0ibG=gVnQi{lan^S*au#xPvBkCG^ z6epCT2`aYnb=5a$-EZzs_k5iyeGt`fwH0J=11nJ7a~wM+5sU@dYD zN*Tm)zb-hp!iwhlm=b?fQsm}k^etG7^|!W|$7tBaj94ppas8Yaq#A=@Eq1+Na~rBb zoKRT%<;Pz#?4YKpybPx|@RHD5oqg5|PJpq>du>?e8oe0s^CCD>N0*MGivj;LOI@uo zY6qo64f4fzIE9OyrWx>#<%QGKwo1gLS7p<*Z3!1d{dkdMEzW;|G80=x<(I)b3a9V* zPjFwIp6~CRdRn)XqVljO9hpVFW!;UIAzjeiTQO=mu3|xfCU|XZRTh|WFGEhw0^&;> z?JAs&KhtPg>Yr+Vri$@;HCL(cN{xTQH;I(nQJpuxI47p4ZNZzH7G*wLsTJ7Ihh`26 z9qZJA3(Y3==8k`BF=!k3tsvD=Z9@kQAuV-ZLzLZj*#&=UF*bz?Il{vnO zkU7zx#Ow?WSDJK8S)rK>|A!*&jN9~RK{O|+FfeOt@6xCJ`Ba7!2w}i1u~$F%TP3o} zWv<4mMm|lZKu?P*#c3ha)y=V)i7LeyuID;v*54(3a-V-{VSfaG`3CbwdVewC(^!zH zs6q1P`f{$tqO?lpwS8HCH_Y$V?w6J_ilvmiI=}pIn%1)2J*=Yv`dEt(3MPl2`5!xK zzxm@`^M~LnP@QsQ7t?+iKk!qd*itj0svb&!+J=~|K#SUVl)2me$$eTBh{OrtIMmb|=S zB8Vz#?nw#IPEund3lt)8F}D<{^tC$7L9c(z;u3MZ_}#B>(e=cU6kq1dYUcU44Gn+aqfc``4@Y`by5-RQ^(7a#G8Kb zFt7fYht_(oC$;D^UAv1(AK#1N@U$ZAtozli78RLv96fLJG+PQZDBZK?a;j?exru*) zd&%}Ixy~GLw@a;$bgRWwMZTE{4R0tq&XBl+_!63%b20!A%{ie9g4h$*U#1ZvCU-{<}juW&NiW5Bcj_a9-x5j{*Yn8o_^dC*}L< zCyPV6czTmQ#iiOlcnSie($A1&!E@BZ~!b zkV4muoqZl@a!f#Gq!ql?cgcTX0U9`UiP6Ozso+(9vZbYZkQg>p zYIgY#8ExLgk}X|$iO=mU#3B>1iGSvffV~LGcTh!*l1tMHpfX-#bPcksF|YbAX`dMv zR6l+n0HR=1-F*7)#=qt1xx2VF&zx@aLI>5S^`Jg+`8b|RV53b#{c(S5-~SJL{~guj z_U(=0iUk!F6$Pc*QE8&|Y6S%W73odsErb?A4Y=8eNK=vCR1}on0|^jRAku==03n1P zAPErygbAYnIRa%(dp4&yGdo{?4T8`rJMl z#|&@L=tR}_uH#!#@L+#Cpzhf<*z8M*8ArjY;oJ*$op$&=KGoB0C2urP`IUw!gpnVB zPP`USpo6Oh2U}~KVzH|YrdU5yS``I=EvSkG)K-IZjUcc=;Fk+I!}zlG_O@*$|Hh6~ zc04(vs-gn*tEoNN7;vlMi}bsB=K<$afm2X*rbVWspj7u+p@V!M zmB*${qYj@e6taJddJ(&AakLUP5;5!a+A=jmXBnkjlI{=8Kw;>_A{D-ND2a0Sn@ig+ z6q^qzz{WzTtmbmXi4}H^E4!{w%Y72d4E_QH7FXSK%F+WNpLLo*LRR`sg62K@ZsjYq z#kG0$mrdrq(dNaRJHF@I2F(HSl>#{R#uCyjLzz6uKURM?Gb}`JJu-^PNf+71+3J@ zUh`Qp=z5CsLj?#-yCeZiAAn9<*y?S9lQE@yJ&pD5geb4N%@q^W9cjTs%N!J}?;t zSnqh?7CN_BD7aMvT0eWAuEIL_QuI^4&`J<7~rR(x!nPn^>}13Q-cVuJW2M=Ld7 zu(P(<3T=cbcd56-+=u9?cU{Q*4=mD0Ud4QrMi{Nc#(EfQusY#m_fJ8dE3x6|!k6ky zg==-{W!FF{^SjjlsE@wAlGH;`X&lnhp<+t)=J#Q8VGMp%r{Sf*eLIx?7b_CDVq~=c zmGysot)icua%IC=`nd;C4Cs+jX5e}%e#W6_WMGfkkP{n_)1 zGo=?45f{k@wq(*H9#4;;eHU-3+j&f*IXDEwXdJZAkEa%=5pT;{Ro-MT1-y|9GiUV= zV4Aggj4iG?=&**D1&pi|_j<=#SIOAozZ8GRU!VTe}(2LFw&aKV^u+{-#FhF#zY4(e1?79ZyD*v)I5=|yE?8q;5OTOaV zQkOP2i+qTKrJ$x_Jwv*-9VoTPpu|s_61x&$gndI!*Vo%b8L@+NACH%{W`!9p@a%t0 z6w!?30F4O;X#AM2LD=r;#UA_gg5p|t+64o3&~;>-y6ksFG%kfzKl;RCa3a>4WI31% zIX0!nF9A*(y2_5ZTduhGBvkt1v;a2b2rY(doxcGv!`N?BsFN*O+l#m8*H zFX&MYBb`ndHD+O7_s^mf)R|E6C6t^xe|cx+$g_f}_K)Ew6y{z$nXfU7tkGjb3ms*> zlHxCAl33lTI$NN%P89@$~`JYneo+{;n%mfcp={I@7Q&asjj0Mvi9o&b8# zdQJgQH&jh4hdf`G#tn+I?kI}c_EPt&Udqk~kVmO+7x7Iimx}FH6i=}8&^KF}6YO4IopaN@P!6NX zb4t|3cB)zAb>?*s9{!%E)*tI(+rC}`QhX$LGPkE#MY<9P|&z8Rz3ar}Vr2a%hVYBRvG;pd6P+lGN$HcHcXd2stabkrpt>kYM zSiP$b!?967{t|Tfj9_X9%X;qJ|H6wQg-AA!DG_`mUTOJZq-eRd-^VbvML3O^8)$Sf z3y}daeRS3I0#q;86kdN!sCF=_=WEY1gR-%qo{bGwKX^*#Q(L_y3lrSntRs^qb6Dx} z4_F`*M;^uB834WXD5e{LN&^h7*(#8CO+_6nYrLszNLeGTpIvzZxNfB!op~7$Q2&H0 z!>}EautFp3uZLDYkbDg^#g7igPT#F{E04bhfoaps=(bx+JymAoZh*Vk`pQ+Epuu7s#(LY{`(1+yNjB+3pMubKK)kp4 zG=emzut$`_L>HgNYT$erkWPeewMgfL%FxOs1 za?=BF>N_x7Kp4lJ)9Cil#g_QyezY?>ZoT6yoynq|np8mwFm{^+#Ra*E z8ip9Nw)iTc0xo$eoSuyxTChw7+!fN0DHgfsb*$cMc!htr&BG7GHJ3`4Uguhsc+=Wa zigh|6PC7pMEXESt`CcwxjkWTbhw+%l7ji#H3u%CHTkKmuv%*0OM}=Eq6?L|~Xv+d z9;CSh>Gyo;@FHF?jUo z2hWl7GHHFvg+kF6;A|9f6y%vlFYS4b-qi)6RZjacW*XAod*6T9z(v1)%B#(ScAx2f z1DfANJo+H-p>vPlxq8#EiiX~=GjDFkvw434?Dmwwn-G9QzNNr}T4&ZdhK;pIu@4|q z5O9OyB4-aDa*rRX2uQ1Qec<*!Cp162tv8%kJtH(Pj2HV_+CU|H(C*vH;Plyqd5I!} z9Q!}#1xP@i#&g`*Kkpb4QuATDn1dMh_#_}$vT^8-n&k&!cSGtJCVnJxwD;YpBEx@x z3kHHwt*eh}i+A@=o=2t?Au<}K&-n2THSpG(hgA%7y?G53Ap{ndR_i@BXTBrdzgyOJ z-?@kPpf9txQ#~|af~-&?R;fwl(v9)YpjROpG`f;Wml@6lYG!7DApK*$wca6?Z1`6@dz6__`F2H^pvRpK;i+suE}cLc5j+V9pHVzlX(9>mZ3m*z#SZAGaVT$q(K zP9|te>h$u48MvU$Xdb=QcZ$Fr#T%|GdKgM!&qp4$BKS~NIm5gp>-lk(X^;*6p4F)} zx^{bi)ug0m*?J4tK?VgEtBB2ir+sI$M!{l2u+}PvRItCN)_fke&GGI`;;KsNI{X$8`SV+Ys%l;XjDg>LtJK&NE%(JJ}SIyBg<>enh ziFws{>rmQYtOF(zK)v_D1A5QZNFSsxVW46#d~1yjbOF{Z@?B99{Z-?BiS#`xJbHn{ zk!=9U#@d3_(tZhZ6O7xI)OUZCx36Iey3&M`jTJB%&{x%v2e|;X?_*>_*TDov=^cvi zCdMU9D)t+p1>S66T09!2<7m{hJiRa{YJZxHG*_B=!tU+8?>VrvjR&UlGtSC0G zAC6aok=pQ&3kQCWwLzZeEPC=_W6rebZd(1COyi<-$f zYRA>h&%06517OPe$myYXLJjkO_CAfAf4UTI?OS0KpJ*&@q$x%4ccN6IkGM!AM|IssCZG`xzd^umk-_79J7PEgDp1FZT7f-TXj*6{qPWXUg@r) z6AA8}R}QLnRtA3!(m(yF^F?dW2!CH$X5ij?2P$jzM8ud?Il!j`w$_ENR?5uwh9FRl*tXm{Ux3&yMt{d$)g6DNoc!kaYhlt=#b9D-|4| z2h49-LT*=-U-ownaxWAOAc^(rOg1Ipc4Zlh!)yfPh6Iwh)&1@f24sl_=E?JWLh>5F zolU@I*nnbJv^qR+=j-#g;AsM8xzuEC(RrDYVAJp6T5aF-F8_;?MaSmmrEf7xSZ(FT zn@J+(-Uxp^6$HJlv&m1QA}YEA`%Z|Zx!W6I8~XwQ_mB0&JKBE?v~qy8zEgRiSXz~_ z3ZFljE+_b6$7=%J``S4fSgOEKWw|0azo20^0oll|Wz22l5Y-PaaK8`tM z0uRgOk)*j32hW|&hi`~Y&8odWs7k6cEOz_Xc6fhT`qu_sT0NUm1c3^Z+8@W3t}Gga zHh4BXP^kv;TX{EUX@Qr;cYHHXWscv|Tt&3uFyOMK;?xJ#M%?Fs>)^qI&3PAQGXRet ztuqH3cYRh7HAJ2N*d7rVyVMftVV#IyWDRJPtwzkBW!1eQA08|u$#9oGY0Ip-4BEda zz5{>fQAI+9xM?<;;mzNd*nJ6|h)hD?a#a{cbwWo+4KBU6O77uc1>m3$m#&4H5uu?59Zn!tpyO`(3m&x*FMrE?@6Q12YwhA@4 zioKBTN}XLH`p7IJ77!+$4fDcP>S_9HPhEet6YtMkz0M_4vB$i-DxvU|3Bn5ZZnQKQ zO;W5V-lMd?3bD!m%tq}>K!9TDu6l^5tzy#k>Ec$6XZ<8Wbwj!(X*=V!33@3sWU4G{ zooo5}rGo4204dFqX(~HWRAVQKtM-QzyO%zEcAfmtSp;~xLkZk3TCLD9(7!6!*k*sL zlioz!K%b?nRgDhBpD+^mke$0>r&A>Z0-V^!AiR|l#PDA($d_@?sxIc2ESP>U3g`*= zjM(3&V6GY0X}Dqz#y$^ZCyl5dWH6vqZn`HI(Yk8b-%$Dbe$nJ(Zn2*E^ zLQt6)K|!6feNIo^UFT9BqF{je7gT>b4sl3f5;;Fa;CQR%usuaFOEhn(d^IpVE=4bU zPLHB*$(|S<%A@lS5+kI&A)8?m)@%idCbIH&yYUJ?Vz_&@m&%ihz zpaEJRHc+l`wN~ch1HX;BY7RB*+k$`f7cDFdm|qv1S1Y0QgJ_U)x0p=uP_KUtXp(w< zMRz_x(s)6#B7CP3AyOBT9X}t)(}oU_F2ZHc$H)U+#hi?W&y&VdtaP%i&H>AoLrN9{ zL>_^lsC|HiX!)VIK2AZ_p>_ikI>6N7!CcEc@fCyzA-pKM^*`NDdLr=+M4a{XQH2?b zHwN6II~|R9ps3Iv_4?t1*bjfX`gS`~=bpUPFbT5@sJsryQc+r|XMUW#o@p4kJ%u^Y zS@x|Y2XwtdE8M6tNmQh$`(dSr6cjni9rL_ker(lgwbxj@P-SCz4SOb;;h*J4%YBD% zr`4;Z7o9Q@vD>rOzdCd(D$!=g_e62*Q3XD%ztNWG)g~^MkY=HH_n3bjN}-%1;wxs@ zrhDSFBn+eq!Ru>ydf8Nx@S$9&e}2p`GGjITkNQM`SWyANNB6EyE4aoj<&RyBt;yGS zm!b~!qeTvO6p*Sw>xx&~_OC**jdx$SJ83}m$;+3!s!EoG=FJT--e`l@%UU^b^U?;D zkSdMPriHW9=MF!7(|><%i{;gfwvs=bmaQF(G9MWG5x5K(M032A^6Xs-8$;d9SVx@$ zgI$!BhZc()k0+iWsV1ICtp5_~d8dKICXnU>GTMQ`g{oOqzeumuxhnTjsoU%`O9M*T&G1 z5%bC-0>g?IuAA$xyb#+vdRi=C%zkyvS*iqf3xBddVz}#4^$WYr^TR%hg%S414?^j~ zel2CTB_`1Y0B3*nF6N90GL*El6T@{5H`+bUy&xu-dIwoJ)9-b(-SC5`f%1Uu9L5jn zaS}s)W5VM((9jDpQT|qY8-I>b1V~7X!+*2SiOR2=ReSAS5ZId3FIyN!TjFe?L4-2R zXy-J?uTV^%H#Ep_yO|0y%X()rUY1)D2TSW_<^3pa*?E5()=c9#-vuMHZ;QZF((Sby zlRMLKKut?pyUhwFV*OLhyqO3k%|uhQ+^0lrtgy~Bq-g1lLf0r=Bn2Hku-Xmz7Jt@V z2Y{yG`*Q*9eSyd&bBLaD*M(R%lqPDmt-2}i_&U^Z_euq1EzhT5^<&&V5$)-*NtmCd zRY=ZC{@#BJAq+z}%vI;3#5IROo5=nEbHuO_GoGJ8BWg>nAAB80IBVz{MzI3pFuXgB zw&b9W8H?qI@DJy*)9C414@5I6wYtvQjzoS{juX=Cug(X_F6IdVasl zxt5*e*4!zAO3mJ$$oAZAEiD&|-?^mOX6^$0$$F`C3^2J0igz9zC&jXy_i=_{H z*kFGMFRMd}K5lOGW9sIrBiuKb-~(AC^K;Iv667bz~Jm?d6_UHf~s zqh$h0RAFk?Y3>y>9|>+xYfaY${hQVUDm8!QPP)(cRYST1@1yLDruAL&v$ZhyrBAKv zC(P@Gka1n5>sLdxV7c7^*EA=VyYR4`{RvfVhM8qE)UG3dzjok|#QW>SYXQ;k$?36O zgk1%2%&34{xybNY#D)T?x7!;)NHT@MhJkIp7u!Dc#P$!JXYlL9@BYsi+7jP*(N=#O zX@JtkneXxx?a?@9e4>A!6wj;?_e=JnS(?w%0#*91;MCl1u8H>!AYbn|v`SyPX6=SL$37R_hbqox~8XlE3ahaVd*7~+f52D@A zsn9=ac0V&P@ZEq<7vplJGc#37qi+^k9TM2T>I#<+1UfIFkFAfg?!>=&Rlh?oS?qG= z{e_L?cJh%swQa71Rw`eWbjQy9u87f*uWQBFW*?4GAOnM*26x;8ab>Gca{TLOv~mO2(s z%mtOyuC{y%v9Z=zhH4oZ29%Tf3jW~rwU&i@e~S&VW^Rq@G5F||KVQ7bDMRV`Tp9A& zT)A?GW)^}%BzRADpL2nZ`Z;CYV8f%0>#H+6e9RluwD9OtE%GM~B@KVd0_#X+{JNT5 zzNRqn2(p(vd`fLQO)s?P)%i4Alqv?9t<#6ZVgZ{XJ0^qA$t%zz6?Nng%0J^&P zY2qWzkgCjA+#=j%bI-4bra(^}emCT+v(M9beb26*l|P5qU>bkll+7Qfr(MJJy&L^1 z0HGlAIY$iZ?Iu+S%wH!AK{n`_`K1xJfTvhfeXKSJ1V@$axzNq)1|sDJ8b8)|4f?P* z8DpN*_08>Eyn>C+!^otnyo*DicT4Y<;ACh+dc0CGqJS&oKL8LAiV18tGN_p9 zH;k26IOtHrtG~G8DzyY)mky|KAR2`f4|FXdHL*Z3s?C2&>T20c)JoSt5n_HrSXDuC z8wj0yo{^&i2dK6$^}bsu4QManoNa=N-+zQHaxTZ0i1cX|LR73Mu7MKUl_at`z^7te z>5*sG&{zNgBFFD(5e&Nva z{Wg_aOa6^q#9OyfJ#tPVH{tgz{YOCEpkbckQ^p`cWg)4iXapSX;m#kOfF zTze8UL1uXxBJk=St3w)NXES|!hNsg|uYJWrH+Czo^pU;9GC~v-yKj)4VY5$Z<@!$4 z*_?Q|WJ*h?YS+j@COIRxN$m}9R=}zU;KZ^E5Rw`U1Vr=fig;Ki;IPi+yeQu>Bq z!AyU)=1O1h!Z<=>-Z9m|k=l^zG~Z+!xoVEHedSEO3{RE}0l zri?n|1j4)~TO8WE=FVo<)s%2!trJ&}6@k~0V;MfE@v~XZZ>LThX_dbyz*hF^PtK&+ ztq!=--7l(_G5c=A%mIV)$@O! z0)~TAlB1E#r2c-g9~nY@@abm7y$e?S!P86znPrcUqt=>^{=uWI(Z__GR)uybO)f7m zD(Yk|nN-AYZaNkWrNL9(R;m;&d~c7l}8hBN>s z>Enui|IWR1HH)f+YD`IGq59j9f@cC+97{qbH#9`7I?lDSsPze6YSYPN%y}6>Ib!l^D_JO6oyU@ zt!!j$TRB2I07CEc)9-^E(-KR`PA(n=_v0QUixCVvAa{t4SzcOHCNZ#_01=*nq-X`xTp z5WD2ekvI&{j2ZZV0RXwD6XF&_>h|`aJ=YE*Ylx9`3k9kklYX?}{o62keKL+}Z%}rc zswDU1eY?gKOo0IK*Q@mDoho@aFwSN~uD1@^sX!>q%5vFuftSGPY<0`q9FcHs_6hh%LFKa$;S|77`rypCRVv<3}xV{AjsV7N=N(o^nm) z^x*Ck_YbG?T>RP;Gt#7y!Ml!M^xL6;^VUv|HODtlvB26Dw3QptWA^B9xx23`bj?<5 z5A(wn#qo-JCQC*2)M6YgN<7PdFF-l{4^Cy6$YVH}7H$EG1yRJrN7vGSZT3njxe=e1 zo#DH7Txg{o>x~gv;FBmhYRNdLEsEc-zHtsp_pkwBfbg51Sn>NlWn}( zOP0;3_6kRvSJ2i;+@rjIA@6fMcf8qxzn9V}oB95crS@KL8}~bQ6<+_jh$QSv?3S_| z`72TlXB%sG7^j<^e+Q@>6{PI%dGTdm9<)W{Z1S$X7FVv8CfrepQY8>6SjE*rO)@S^!DLr9_M8>_4=_Pe3lPpp){J)8xX#YcMt<8?}qVd_)Kki{4z`{ zd7~*0?Hd|2u^UPig~Mck)%P?ml{>bJ-XlXmq!o|Il7*z?ISB|Pj?)@M_k7HPrVB4G zkab{8O%2nXafzXSmlHsKuDE?CUkyDg2%AQ5f2n@Yeh+wxcfX&QmB$d|k3jDh0|V9e z5OV1$Y*4v)05*iak{KI=vDpoD-O=aKJ6lE<^L(m0Sx6aiEr1M*q^}&Sw-V>d5GQ7d z4S<@CQ=88g?sEIjO}@5nKU>I_ESAPMHj?RZ;>;eG$98vrHe_#U?^n!dxpDB}sRKt% z$D)PY7naX?U^i0y-$spiX^SmrQ0C%!Go*L#wYd^6Z0(a7o2g}x=G$|-0{OU$#0F+R zuz}x~#mC%YsLL0T+)iLk<%L~}U1~o2G_MzK<-GWHSz{-3WTv5XAvfUp2h^u%DcDo^ z*huy?exfCR>f$M`$$MjYd0Eq|BX~?ex>s(hqbHB_MVp3I+^B!FVc7Q?ugQ`97{It6 zecfdJQ`e+y(6*vytKY8#C-U%(S+-twM7%?|bS#)F?0sA(QaXvk`3hM3x$bc1yuR-LlKpYvS*p! zoC{2>L9wc{7<%RYTO0lgTb{+8S0&%K*t~x9@I^6oZ)O1A z%K}(`_0hsLsu^?U^9bLAD~cuhN^=ju^T+TtNKv{4dw%@OSW1XR~{bTy}=ySgoX1m z-oQz#s#tdPPQkn9f6VRJmhLy=p$lEok%)PJJEG9Ohcxb}|Ir5k|ANNqDdJ}0)<2!_ zsEwD*OxxRD-@JVm7#K~uSRQ@lkFYF$(3C6^PG4X}2d(s*P!2n0cW+o3Rb&zdA}^#Ibdvh!CcihFv~)y~a_92?G@?adfE8F|4f_Ra?pBT`IPO;o54jTN15S!~b?)7c{md)>=3f7U zKP{CU4w99GP@OZ&3tT(_xHrXR@@zbo&&1G z*o8BVwW{scbPo4$afkk%IM{#e%sH;;C$klWHRHek_Bp@O+P2?R%z>l5TupI)(4hb2 z&!5D8Q!!zUDf}E0!Rr2RFp*KUMw*y`1}RQ?{p-U0GiS6|rXiwE{9}m!VAX$*Rh`>1$V^7a zLYmc?UUU0H8_0sPe-LATS*zLPK7d)Ke zEg;PXROVt|FK3l$c{fgwj=q}uZE@d|)vfHOvn{7k60c?keyLx;7ZyR&- zrZM|0ZNWzfq+l|^s5$I=9i6yqIwyV~e>677DnSCEjJW!_yz~JQ6bIo9w_V9_=Wj0) z`eF;)O-V#Avxh!1>1+XiqT~#st_{ncYd!J18ptu-bo>4Pxc&dhZvPKO`Nu^5F_C{v zN@Ml!0-Fh7A_YqK7-}`igTy=;9z}?s35$3!n-qn6H@&~#1F|+m{qB* zsO!Tk;ru-&EIKTR*=%B2f*(0d&gj!qm}}DKKE(IC5)Eq{;}Y~=)hub4u@0OzQuN61 zA*U{7nl(~xwC?5=R86a!kgXZ3sb3wp79x(p3osHIVkm*v>m)L{A93Pu$yj(m)Ga7s~=M&$8; zx#apLIOQl@{%umeYp=~1ZpayBphCsQvJ%nl`zWlRWsKUAs19H zj>^;C#f+%EV_Yl$T^U}vbLEsWVuNggT^_np1<30g?5-w%1gLfxaEABRIT13g$pXpL zEoVzEAsAO>-*p~0`&~>J1DyFjRUrfWn48j3?^Ax;901lKIu{(Ni*-EP7;sBf_5E`V z`bN7)`E7B><(U&CzLz(DR~hr$<(^SOtwTEX=vmGyMN0t>6lW^>EAY*<983- zikiD)mRowCIm>fkY$R`XRc~`<7|R5~-=CSz9jCT`OSdA~YupdtyWzqa$#E>^SXL2t)q<+?+<;YwK^T~dTQ9K_Z=aA~*qv!hjcynGh^L+bp5&q+i%EU|G^EO9% zd|BtWr?rJ^$KzKis8ha(HmCS_a4MgF@bPVe$c(8yac^FgWV*^?1n z*=uyok!h2SODuPao}y%Mo~CsugEdkF#f~qqDBBirEGCrRNO$sTm#%o1uA=+zb+i8# z*NbGn_Zb?%vornt8cqlS*KR5@mV0+7&aVr3ctb}5Bhwv+KVluz-M?kyWq8%5q#LGx zys2?fqvL+VkK7bJMrxSg)tN#cY0r2~=ARahb7w}hldRe_L>JdFLO?G?Y;bzILsBXh zMkrLC2nz8e#v6E-)NMSY6E9@@QaV+^lg!nCnXJnxqGBh1A>X|wKU6jD(QKgz;>w7u zB;ViT(sl_@QKCN}<53#WFu0RYMB1-^?2K?e1Sd6Qt1>!#DC8~;MpGSSz(%8bEIv+o zq*HD(#v@iHX0O_QDWAUzfa>UaghG;@Hvsu;*QbF<3 zdDf|ayO;N`KQZDz0Ia2Vrf%AQ_&S>#`k8I|BVbkjesiD@;DyX;G0TZ8y{+iOs>T-9 z)*a0Y+l4G%9J+C@%wq3>tIc&E&h30^d;H-e@v{oHJa12)t==`5iC(yJ_v$^I`XrW~ z74pY6A5;>dy+X6p$+3u)R+W7@J4q8Qm!WDcO!YsFo7U9C`@L`;F1I>=jcdiv2$ekB zhpCn}_tf(%{6Op_d=)MGevL;9m!N-Mj)s_VeSB|9cG=B*4h;>M^&UY4d*oHE(1X=A>S}Cmb>%Pe! zS1nYSy)e8nYiH1+Ef(^B(waQtf>t8LFeB)y$FGEK*~+zj*MWa~cpRF0rCc`_Xd1*S}&=)8ZqqE`lSs>x-W zK}lpT7IGz=6l z%)Jwf%Jgd`>0=S!bw1gP$4fl16kyxCzRJQ^xW}`Vv|_`W$ZbE(fAe7tzq^jvVo^3- z4bSBZb?x}@*fWCD^YE|k9M{dgtgF`wY19mUxD|hhv9&_&otdkn-}>}6lwMk) zaD0rY^`n$k3z9Gp!SEPg#~t!q=#y&EP}H~()Ncbh2*W-BOymj`7s`!g5aV7DU%rrh zpvnANFN+j^_BlM0tvOgvX;|#2n6zbuz_nNjah*1+dcU%utz4|TkM*E~vZDc1hEg5! zs#iyMqyXcYJ$1%j0IyLW*zzb+)YKV64HI1)`(kH{3xPF#4bo~>#6e63&1%1$2Ol4^ zH2xy{Uq^npdg$xRxsLmngFd3xN{)DDc@?D2r*4{mfnh=$NE&|XM#JiYYNS@h%0Op! z%b$3gXPJJa`$`7;JsKAC2TK8jx9bQuZu}Gzp8>v|0Um!pC;YPZ)jn4|*mKkmslT`+ zQSzQ=u=SdsLw3Bz`e_XYkz)GrG0iTv-=wH8+U=K0J^r2dP|Zf0o86d;6Y{>U^ufuj zfSQYclw@m&d^N@;epxScG7F-VWa+(FNy3mD;vI_=;-)%7se8rvPie(HlZdnw69QIV z-CNImb}hu4z3`^-KmY41E?Vr~wEonhyG|3xJxX2M2ReK^m!7l#B~|%MNOoYLyYGB? zs3bc^cGsEp*03H&=1h91B5R#8?;oB2Fe=`ER0QRX2UL%$wZw;>XkgA@t~J)Wmnsdp z)aqxdM8KwAkE`v=p+CkxD?THh9lGdH>0V(L$aul6btA#ZIRBdyI*#5n=*2hj^b1V8 zBq#9jaJK(^>MUfoTh?nL-s{^|dedzrwODSf&aro2kH+;9xSE0oy815p*VeI}$=dUO zYM*2>_S@vtvIVEe&C;pym342M>ni?^QDKBi|G-Q+l~$Y$^NL z9wqQRzg@PFO+%&U=Gq!`#&IQUybL_?@aW3A_xOWa09(B!#^=Rdvd^`Bg3mom*< zb=LTp-Pd{8fZ3J&FTfP?gghFtdtp2FInC1;ir$Zcle(p{L@WXt1NzSv z++ZcFH$4GZqbX!>7(`b(GYU@pgl#TXHp*ye#(wp7t|qZ-N?!J$u_ZP^^Elu zhreFlu-b}G4Y0F?cr@((Nt!wm&(HLxzR#^hOk}IV!hE8berBRQlJb{-RT<*hfeNd$ zfmdK((}|VAD^=Gg-!b!$d)j3L4AcUF?v-a&94Xl^ml1a=4=ynf0qPI=~5i3z{wsBg5$WF{CF zKdLRpuO75f&=a7(x@IVUJ06j2TiX(Ai{2k9Wot6ree-W30dm)!OcnV7m+E^P%aks! z-hC!5Pc6r2Ntgv4^kW$%>hfo@65v~a>{|JW2uj<2EnHAgAv+wq^v3MW=z@RmQNq{h zz741V*)@q=mk_Ehy^VB8&fvoU;YRjWz!G0ifuDuuYV{*N6d7!Puv){oP-oV!I|kgU z=<6UZQ+RS|CK$amHRZIb;nNVBgN`t`$ZBtJ$*unvlg;8gdkJj`AWbl8eN{ULLEwUDCu*_Nr#-`MU} z(pb^&2XA&OK^&uhlQHtpQ_E#XhOPddf{^VDo;`w&98)D%TB)s#<+u#e*u?2V-Lda% z#(fL@1wmVd-x?LH`F4%xFykB8N67g!fO`P}URE7XsRU?kLqV~!C*z2qV9C_}Au2Ru=eYiIxKG&bUY_iQ3AO^iXS2sY96LeV9e8r0YQsKgx^ z3Acqqd(L}d_bDQSDJY>XHCYvYOV7t&S#j?e3B1;N4JT{0;j5VUk4PMp))(@(J&m|b zJH2)81bXCuz&3q#33Y_iTB$Q^VR)`A^p|1i%2>9{#nsxyGPgXB2(j|Mm%2?p2AYRK z^L@8V)K{B&yB)~kJ&Jt#M^(>;^gepyvv8c%^5rJK^|`Fjg|bQ<+pZ8BO9jrO>m^$# zq9UqIuY1=8M#Ge7dD#Z^z@4D$lT8V%KfZnN1Qd&wGMcHq-Qc4!ej^6hI@BXFUg}c@s)#sWDyBQvlSc^FB`9ZOw|Jc z+S@QoWb=_8RBf~sm3Y)CBk=z42B8Fhp@rk)Zs~-M8odcBx1)4xFF*YfB>cd+9X4ux zv~lZCbGUMQN4dR9Mc=Ds?Tv8ml3u^bBIbyFjxiUiYD2qYlPnFnu+}Eu`IA-@?NG}2 z82-+5{x(hKRPWkULf6}SISxIU=a%+Oiq{=QFjg%%cq_}=pxkWGz*!GbIUz)USGv~! zo?&^Z>NqXA$4w7yr8<)pRQz^z>U65fB}h4sR*pmRc~{nS^7+LH-I8}`K}Bl?K{@B= z3lkZZxA@cALmTK!y6?dNonmR4WcA?NvM+uX?@!>DLNfXOhA{9xo{9|v*3tYr*d)BW zyg4vJOx~qu%73Lwv4_G=Hu!vhI{ROLGhRK8Ft$cO7|x^a*v~M~kcLBivZ-yEsaM2w zyY(~hXo%*57{e1*oh8?yh^wnznJbbgOn0&Z%D__jen#JyKiKB^A>Oq}UQ4G$dGn3h z+whgiXSy=XNY*P>(VP=Jcpteup`G@akZ-I2w0X1)La{w%HYzIX&}uq~=!my*dg`HIfJFj0o_TeHlS*@Oqj z^@@}b3qpqH#O=@-4UU0-@7)YeeM^=5wEyS&IFLg6p^%Q%Z$?|#6WeOo+rBw5`OLh) z|LGu`CS&_pG8H8$-wXyN39uuoZr>6(DN;^XPs#w}`8(qbS#Oe*PIq3b9bsWUM=r|O zXK-k%VQ)bGuPNu>Je=eIq0e9wr+}?n@pU((LY@8*(LW;kM@0XBkBG=H{j$WrdI9_+ zu751*|GtP41jT-9S!Obvu&zSHIQ+D`L3EABaQwnWP+)ZX3hwB_kCd3)?w(*Jx; z$FDrp+TOtQxW7xO>*)iR^<7HIE%Iyzv)#?~H)eGuY6tr_do4S^bD{dTH~{!PQ5W*! zeq$CF-Pi*-6(QJvY1PJ#f8PGj+y8m{KX3nkU_C&hrCI+ z4t+ETH@ilp$k@`R1)KTPp(T~^(SI}U82QlJ3fjhQ?P<{D3K|y^rO7jTqssUEslf%K7r8I%Ho@^@kp8WQF{$E<0s=Z zm9qlZhnClWXVL+Ol5GO7-V+o5+ch8dCsirFrN7Z`amo&{_Qk+J1g&J0NDl5c(pYG( zw1v`Cn$SdLZkD>{K2dbcxLysoLk;-_nJD(J~tAvy+vKwaLI{2z|lG#pV~g z!TJ0LJd1~C$j%2nrC%&*o*1j^c_z_h9Ym6~dK?D_J>9vFIsA#%rOAqUq=*yE<5f|F zFlqSB)3AvqvFnWius;QVow~km;>r$QH}f3?b*(MH$#&ri1mpsoBrH)_YAH>m2V43s zI>@?z^h_yz5A*OM%4(y&-&yBXJ5`K$WP1AXM3PjxkvX*QkZ#jVI?&w3VkjC|M>8h& zz46&#(#dnvb2~B+v>D7Stn2Y2N|pQ`;Q%L9KhtWL=HpeDJG&qvuhPI+g`0*l_`Q1f zU2hvb31LnrD-jBLCZ4=e%tPZd!CA{-9wh;PS>>O0E;>p%7GRk8E+@?Aw&^M6?ydNJ znCd7(ajpB49f2Y~!6Wh;S)2TS1Od`;3d>p^Q=Yc9F@+c+DuS*reF2c`F9tsGNn+i`z3Ofj=LsToyk*rsiW^ zg|%p`t-i^+Jjz0b`IK)_?NXyO&E;i(3AamDJb)<$db4`jP_(@F@90y$z}iwliCs=C zG)9xTc3CvY{s9bP)!>>P0!JnosPkE^5|3`v*Jud#d5_1C2W4MoCTs6UoO$`sd!P%1 z#vR_Kf9$XYpw`SCb{no#4MYrrk~1edF7JGB)VuA~$Ffga>DrkppvgIL(Q|cwkILMU zJTfrJC;Q>@euI#7r8huw<@w!eCAFyaG844=TX0EMx937_4syKyT4QZthTVUiabw4B zHl@oj0Z{opiviuN7%EG4Y@?&1>n7V3_t>_&Eo{%=5#IkdoREK4ObvZhDc=O=+#OtB z<1kp2h#KgL-88k1-D}u1y-~DUJnWNMCwaxAD&MBSn3t_F_-X$pMBd{h1NTdsJ1#(c3fl zcKpPIulTWvGSRQ<-uW?0p=XIPi|j>MGM)A)B@1}z{~F*!p^Z@t{iyVRCs=-Mk)c!j z)z=zzfGjq;bk{k+D*?pb_@=kf()V2u8_i;&8HWi~gBP*8X^F{Mu^fyZ4J2fy02^p; z9ikm&sUtqqaD6KM^+F`_OS~FCgT4}5Y^8EjqtUW|KMI+xbkU<>dbMU8CSl6~gQ-@{ zW;S{W+Jvm=$pNc9U(SjJ^6)#c-RW6GXog!QbOlCrj~8bXn&}5`ftM4 znL}SApLQt~g%TSuD5Ft_%oT;F*^u?~3k`%BQ=iWM0%EdU-QtWz86!@lH!~gGwi2@R z(Pq=w5lN6nS z+l{q1Dp`MqQBFonW?&ZIxFmA5>I=9PRcR%X{5*?aYHj;{)AD=3$Qph! z*(SG1N22(EBF)@&N*vlCd3#*hbC+tM(fX^-MVzoa3JIPub6>CDM~6Ppr+nvvRe$l^%NZm;QS zftz$aj2`%p+gFr_ngb@-5wZITz^=Gf@bG#l6-#{)v$7CG)-Fiuvi-6x>tLtTh2|&N~qp9$Pk)Ah~!tY|aHSX#zQR9wu|sNp6R!v32T#a%(^(7)&H1y=vr~ zKt5A8$)>?8Vt8X;AXQm*p^mUH+e?3{G}{93JH2;QiTE|+xvR?+xk417#s(hWuUQzhtVSHWpbZ#xom?A_6ti6p z0XMfvl$bgojsXVkxeu$@$Hd+GWT)KTJ(xi9*<>)Scg%dk+C&_seO=8G{4QZap1M!>Z|Hm5E;Tx_X11KVwXE;HZp-x?a$N0<>)qUyk}n%VyDh5R zMZY|;wt7>)5g}j`ct&Gw?2Bm-N#1sUI`IOIR68k3Z~7B&BRTw$w#K)!sS`SS4^lIi zg`mvvu!9x0vMbfe=0LKI7l=3%Ks`Fh`csGCQA2Rh^D31O@#_Vy;wMUo+4 zYz61^lb6YOzRq}$c3o&6@ozSrHO^0$?}o7RuM4a0NsyZUZcS#)6PSI336;? zI~bXJbz+BrgkQvOR0E|!>APkH;{GeCk@3SiYBF@&$ei_wJ)!_|j_@4%>DsndD9g{=|e@ zIH}@D{$X{tr@cuNGGs@$(kEyBCQ46lXRb}g%Y6+dhvejVU}0s}y&&7H0DRQqq|MXQ zM@LE@eW{TZ2(|`iRz16apy+}C1aGwFb#n@@=4g4gC}q54WjthTTqz&zCCg3?KbVwQ zSGo?VHPBWu90cjnU^8e9D4b%pS-I$%ubT2AjKZtAaUjS~{rG4gJ=50--|ur3Pn$_s zJbTAqHdim5|Ms`Q!Px6>Zzm+?1*vYu&n-Fs#4#R7u(c_5@3!iHE~VtBCjadU6L*~+ z$DJ^-5i_w20Bl_JvVDVm0>IqJwu`$1z9DV7(e>!W?j5Dzg zkwU2v{_vlhHkr)JhUk3hbt^6ltaL*w>?bdlr0!7^5AV9ws63ZUR0|ouncV#C5wxW# zO7m=Tj{#voYI4ngE~r`1tuP7kU>f@CfOGjY+bpl&IYT%-W_fa$HO+QDNcrELYBov= z-175bvUB?hq~*mLuSqRgpochqF668M^W(5aYYN*X$fmYY&Ar%Hd5YEmXvq2;zVM;7=nzkY)?9wjBpDVhgZ`3w7oXdCC8O`IZDHyJY`nXR)LIR=jIi zxdM-pQ=;K-ZJbxJ2dEgZVZWXE&)ffb`#*30=k5Prc)Qs)6i)}b^v2~&zcYof#+Y4q zGM0TL;Z1*B?y34FuIpEMm-w2U#MVB|S=BG+TrhA2wrHLFpN;n8;nw!eqswo_mh54? zlLc)yOW8kvF5cMpJEI)Gd-_MM+-$D?9`-tSJR4U3x;YTmOepp@#<{Y2Znx9yx;H#L*RK@(MI+*7f$ z%%c*LXr*ebW3*{V)!9^WId*c(xuX{&qN;J0sdPTA2wX=VLMBu--4Tcjt5V~yct$7#u%2Zbu2`5W#0owkgmPP^7K z0wvX}^;DJ!r-Hf*@%_G>yArt4c~s7PdS1XLxXjfJ^z)q8bbqLh{O_uh*iK@m(R=~3U`#9_%p{?q;F$*|Cs7z0guo%7DyDgNZ*Oxv;65U0R>%Eu}Q z8mwnP1ij&NinuV?SJ<~v<0hISt~v9^HxcTHjCBYpcH{kLFvXG3Lc?bmJ6u_h*JM6_ z^+CZ;OV>?3TcdPo@_<^eayE2{SGU1e2YKcem(8|+4g11R(zs~`3V@Q-s#cPkA`_OepUXL&rbRI zu6_R!Z&%jLABzDtC>Wc2(XM5)s>0EKIDd&^*!q(}M?$Avbf-eAT0J|{Jn`}-?tQY^ zh5Z`NxQ@wlzditXjIbG-G|-48frn4+I|&@D1`*4GB28tX)eLtJ>J&w~PVS{uN@EHXUw8G2>MeUfQ0XAxD-&0a9Ik*bT+xM~(N_6ECB}b#g4y%$%^&-p#e)gtkKkC5_Jo1GU8cz7d1L;7!^;W6vT z<1B3v0q}_YB?VW^*&KH^YG{w?3!JNB`5zdlaca}Jwy2O>QElKz zXcyonV6$AmzY?%<$w4Q7MlisQQ@VeXq`0?UK2G(~l|o#3fJ(_*tzR6`ANc`%4{FMI zK8A()M2Y_S!ijC9QymqU7i8VaqKv1TW2U+>+2J>fa@b|d&D{jBQ! z2dlTaFr3Zd)2a^?Ju2N{7cCoXdhl>EqKnjUPln&z`O{Ko1Z4^wyu2trH{n0o(w-ZQ zj%c~WO-gP1sn zWHZ?3T|o<=hic4!smo12W$Pma1_JlLa9%Z(T#sy3Qn9 zCpIiCV8{)BT{h49jjhh?%o!-se6DzPPBV{y!WvX*6Oj(!*349TeSi zTYnUSubF0XLZbNGj${j!AaH z#2anV=+mW{!ATay_cHt~O9lfA5SG2s%wJ!>c$~^#BLOW~d{k={Yf-s3|DO-+OvcbQDKrIz@*v*hHMr5ZceC*-MMcbDI}w5QJLU7 z^m$`{oqiLa`Qk}xd$k_fpA&J1TctY@8;4{xS@s>8tl)?R+FiePenPVq6nnQOOY@cj zznL9Y;f4FRMV;bmVsS4`;}@4WnLjRZ&LF_P4~*Hoqfkpqbh8xJK6NQ(X9h*p+|qm8 ztgLrG<{h++`&j}dFIz+)!3VA^eqz*K5#MWnpnN{ba;YTqEDHX;$2XZ1+M9;miVnIJ zj-GBY@Gm>?MAX+DG20VB-P7BJKC{oQu)%eGPrbcKtUY(!>Ddw1yUDkh(M%o?jrNLH zDRjA8y!*}Ms0+47SL&D$lF$lcBXu+whMNZ}BF;nB(0I&PtBk8B0i^aa}+ zs!BOmrbxp;F|iH$m$fmHz{w%vNi#8j2(2I80~Pal9+dvb2*|auT+wC6sJ3{gDuLM8 z;4|7M#=TMCG{S29(#>e}@VbVxHd>XyU7RTs&vAn;%sC!c#Q7WD^I78mv|{cd%omz* zcrpuej&lU?b~REgKy7u1^<1trb2~KT=SlzD{S;1>2(;@5lVlv*rz8%)b2_?zp-Y2y zWjql=yW-`Ivp2!(N~>Ll^4$!7>#|qO3%IQG)?kx4(xuRwm#HdoKDb$>snStS6$hu9 z;>IBjROk}iopYKw{d55sfT)I z1bwnmb$ey?AyWn9N(6r!UmTo&G-x(Tp7zUScI|JYUGb)cQQldG!M}I}HKzmX`M5<& z4VLFX{^QiMs$Qqz{-;$h1FLS*YnI9>r-n|b(58^OaS5{Iuga1QYqqT}FjcMW2F6L6 z>Vv8Tt>JFXzeyg78XLHmpK^^nC6TNl4$fSjv=XCouJI3+tI z%dk4gf_wz9;ZpT~ff#9D&%eo`E}t%KUkF0_OlHRup6re59EKK@J5q{N-=&+KRu$uL z(L~0}6-+2I`u^AIFL2_nk=^7ESNx5Uddd+Hp{2Y9jrZ7sW2=RD#+l&@p;b4mWWQ$KRs4k?kOg9kjN z!|$a>{-BUE8N5m|?AY$LNDJR{ePf+avN-PsY^vzed#a68JLuEde_TMJL2`JCA3(_t ze(etjRL8V`b7!X`EZ6k;@1WhPT7AK1_4!S}q&_JQ$9`fB4lqyuGcLUd;UbB!ob>nZ z2#HwCSN+HVX%*vP zGln%pI0YnO;U zq`x7$S5~+jOYcdLT#RUdMXrTvZRqm8C=OdE!VQw0(=fMboI1RM z;)WA{#qjZ&Wzv)E5avx-ZbqS$$;w=x1IvGA$cw|$-dVkbo;+)}s#X`&q1vT(q;0c_ z3SlIAg4xZRC-2jxZnCjT0L~S6wWfvQ_2JF;)Oq9V`S^am9r1XM=SSBNrlcdzm~;Dq zRPlo&X^MGfSp&*1rfpA2IN&{GVDxw*ID1NeKCG3nCBfj3N-^=WL))twpEdO|;w4>V zg@ML%zhjKCQ)R5Hk=jO|=Gt0r7aPfED~<-M`pte*&b^bG_9PlQ65K0Ctm}SK)O!F# zTl}H0qH2tkbr7TTg3Xzfn9H_PAC#8AkCxGkRXNQr()L^R`#X@~oRs155`CfJ43?39 z=6xqk>Fwp}B8Y}J@J~NH8l|hjgqF;XVK4jfMsV3wYMLQa;>JB7B2OX0u zL(uzGn|9!{tlyA<2nK&Tp#Q`uYYgYr_5D$|FSaUr{q$pkzSU~9z2fXt!uHn*4}F(z z4`89gZ$p;Vy&ei<;7-d*Q;o}kMy%6+a2Xc;OMFk&d_`BiQb_n%zMk=Ia{BfZO7>K2 zxCMToG6DNOp=$NZe)WsS7Dc!3&;jOk%S#eo6Rmkj0tnJHi=8hRmD zFWvZ@uJsxJp2hl{m365-==?pWODfGNDeSOVN1gfemY-cQkusrM(zL1y))y#Eh#*cG zvXw;GkRn|g`_;};weHyyGTLpuMx~8%r&ocey1cCJo{S0;Ihs5r(`HTKTD#)LlX2|Hr zH74S`1(wY3S*(A~F6gh&HoXn?S3iL1JNsnouXtU$yQI?)zF+y$0I>fgN|o@+73sl7 z*UGLBbg^gKcr{K|4+Tz600H_w3cP+(mdVUFwHs^K%P~qL3ZEs6X)HEU(!MS^1%MC z;A0<|qr&(s&6-)MplR|eqQ^+(P51{J1f*_p>GB+yra`0FgfirRjjh}mxBOfnGf~~$ z&O;7MdN*-sOGP>C5#`-`#2=5R(w5ccYicHoU&z*P0U1qNfmZf;Ej3=9v)} z`jMap;8*$s4H%D7$avsTcr>@oF{cdB)$TyQA-|s_x|NIdqa_aor>b&+Q$k^!!yo?_ z64k!iito~3oiqo3lm?tGK?eCRTU}PG55{`8%RkF&iAfuW9owyxEFx)^vfy4WR8-CJnf zk8+R~{^b?>Hk$dasGTm3g)1lbkWUTBCV4-+DPn|Cwx^%U_nlSys`_1*hL zR$~OLgOzvM`1%oR7z2r`n#8Ymlf@&sf_d5A_lf=Q?S}E3Jv9ptxCC>9(lr=>B)xs~ zrB!`Q08Xk&vSvq=WMa@wvB(e|mnR1`it?)nSB&;PLp^gum)2!4~(`iS+{ zc7}{YqjG9jpp;@9r{mfJ0>8@HT|JV!HcSAv=*l1j)7|SH!NxMk&xu(th;O~Dq)z>q#gI=nG8NJv_>&zK(7sX|-Q46jBV+5bi+AwF+%0?t>ItY*sRp#Jo5_~P!ch&c zduq=DRql-X4^|)%Z>impi86X!960f?i&M=pD9Q2~e8bJZ?u%i4^PhczoapE^h*PHy z7J1eaua;=e`=<50OQhxk;kZ3|fKk(z3|WytF}3o&Q)j8TfnZ;zli*~yz-LjXQDTKed# ze_{&(K_6}b-iAvkHT>IK5X`^!=7$`Afo;=2b`W2|i+WkRb}1##-uwDj#(SSTBSASc z02Sdi-fVIM3j^78${V+n-0O^kDI_qb*dJ-2$r9#lOTZ?*h9aikEmU5`H;=VHIUQLx zkXC@_e}j(WwC>0fQPmJ}yMoxU9i|zeBGZvH(%sD^xZkWn{@Wg1e{!}X{D<_I;Lpgp$;CY!S~9~f1S<< zZr8iM2RPo21BCloG;5263hon>d#Y#?P+j`zW)E|NN2Xb$*6KxqBJGTjO{II8b>K!z zIxug#b@+Sd^%9+J&1~9E#)Yy@3D|1L7b6yvmX}WQ$Q&ljj54V`x+U*`vUe1|*PMV# z_+COTWxwoz-daY{#~kTb`FY>$!?_A1=zo9rV|#xPrzV&Q-<*Q(7A6#MFkglHeNUxmUQy6Z4i>Ud=mq2rr*6Uyvv{e=IHxe!wa1 zizs^;p3jB4ZBc$|`DS>@Q%+w&Wj)Q-g>S7#^LW){U)D}(_V0L^Ad*Mlcl0oa-8P)Y zdfCAN&=_Xzx0Scil9n}D!VeU!d$b#s?|uK}e%}?+t~==pQrL}?ySee`$`9pe$fhU=(bn;@Wcy^0P?Q=?byQI~iY%NUabjiMrivoAD zlopYup#H>DW^0{GxBUul`8{M#M3>mMo)L)GYL`xMNV%q0&a7P|o}Y{o{GIdZE3a?;=src|L0}Q!%~aOb4IYJ^55dr<5fRU@5w_fej|Zb7o?_ifu%x$cw$KlQwhs&6 z4sQM9G0&3fe`UI=6L~zV{S$t)0*s|+X)0Z&bWl^BM#0R$wYn^}J6=h>qLX|1eJ(L~ zzYeEd>sc=LcED{z*EGMrz*E0dTX3t@H@EM3u_T)y554bpGPLe_m8Qk};1?|}|C8M2jNh{EQI zmal+*N4m%Fo_+pu?#{>a@Q>jyIgs+!1$>_);#`rc(cd!Z`Z+}l!nx`+?PIbxRB}vS?dOCi~yx7+} ztR%-Qb2R9p|dd8kLRoMSdq12kh9Mf02&1X{^;^a-rn^=$HDPAGfe|EWfk3 zr#{T*$m>4Gt@sDZpx+Vd%#j109@e2-@q#oSCG;+1=fTrr2LsDE#`tm1CK9V;8VIN+w%%0Q{c%zgPC`Y`t|W ze|}~~z_a)T;vc2|qxAm|DV?Rroe{a0c04P1eJIdCgJz%5%uXSNMN}mD4Gq*m^{blg zQQ*-KnXCYIVkB8m1oL(oj8|tq2wv;fv@Mc{H*rzAOjLl8zkVN6aXW-gg4Pfg=o6fC zg@c;Rqm+)zChnz76{Be|m#WVco=@GG?F~@*0oS-7-R6c7Voc2=G>O z^+c5GYF|{9$I0;9;)RjJ%aD*w-sP%JD zwqlzZ?4DG~0k*CeLVX%(tEvmna^eorLQr!)^m|pY4&LinYq<#VJNS~G2P?A->-RtE1Vm_M<;?kgPAk|2i}Cv z0CEjOp&P%J^VX=q*Z~l_JVwQZ#sjJe2TK}OwX6eMT*S4YQi>&moz`^W1MwQG_MCNx z^Nz>_lAtDwrOTXkkA=tEW@mBVf8=ix4JDZFQnK;sf8SkA@|{#v{`s8csKJ%S0JZ0= z?{C;<+Q{0w74LpzG7~&neEMSGGH{Ve7vnef^}o!@^c!|ribAoI%yQq(>_^&-w65v`qaW0ZV_E7-8iLf2$*Q$x8E2 z!-YV-1Yq46{n;SbkMX;AWGD{X7x)9e(#veR>TPpV9nbamVOwW#NE_#OvRz9p>EU+B zOzP9@P^R*LwcE;IaD|?2RxLh;s=0Qt7|xveIxJw}`=W+4;Fi<*z_4HBS;)_Zy~xgB zk=tmzVtd$Fw$XLK(;AeRe>gc@AX6m%yy%M8$f9_&6cl@`qEo?5jWQJ$h~7}lKH$j& zG6_e8evMGpW2L9!qL^T;P=T-Y`Qo~b&nfQZtvZSU8&P?7@%4etrxPmUB1LOdXEMPH z_c3BawE2P(yZU|bmKv{vhIeEGM02z5HU@ljP(^;Yr`cQ(o)tBhf1B?27TBNSq+AZp zu9-x^ZR4>pZF0_jDLB3hv2h6FjyW|NK&@XRoC%Er4Or(CVt>)Rn3LGnB16}2u_0t4 zbSBJ4>r(N>>+q31R<=z~#U?)@#bT6aQi|U8eINf~H=FmxHq9khx{ym2tZ~(H@$;fS zKVcZ^Cgjs8kM}Q1f8?z=BK`!jZnm-1ULuPZt`IP;ODUU(;6fcS%lKvCY$NMkqlrx@ zUqMlW^#+q(nD@L_$` ztOU(yq$BQ#uOxv9qy!HagadzZ= z5!EIUsFj$uFlb~ur=b?b+8styDEt1Ir)3*;RQXMDfBUd2Lo(UMOE1H(rM3*epg!^8 z6z_RB^yg2?JoaV>Jnwkc@(VU5G_eG@P#AagDQn1dRSmI+Vn?bnmbvX9am&57KW>o? zvNz$?0J}sZ<2KC-YLBjr_G%QINP$pCLM$p%UGwkjRU1O3Efxc+j8u`|$4^Z(C-1yj zUEX9He=U6oZ3(*Fp)|SoJxp;_Sl9xHQE!Pk#s?U#j7l^yZ}@&5F}LRtUi_rR)|?Q! zpBKFd-5g9o8Rj4Kl)g9k%l*nmR$Zf4h<=*@TX3(4Xj3U>U@L8L z#v*)DjJ0-pCOdR=wzO$tA7o!yQr8llj6hba5(P^R&oC;E`}6OD{>*6P*wkaBv@`VW zO}6v3g|3ECIKKTX6CGm{Dk36-ShKe;kaV)(e~2=!^{11-lZ^~&tTmlh2A1$D~qQAXgx zkx1c68wi9rSXS13)}onP400IbQSY#Gq4cA)+e=I(u>skKh zmf$a)WmIc;^T9}w6dQA?vEa@sMJ+@^*VASJ8#xGL8of3vCYJ2MiAND7g2jB|tLnLO z2G;I{%@y{aX5uy0l~78Ax)GNlhvYa5k2=_==O8ah$WQs16em&#;I?cslhkdfe=ko$ z4ysc%Pmcx?NV7Flhq_s)bR8Wic2qM^%;%bW!%&MC$}H~{N&sALbkUel4%tu%^sQU! z>6Il__D<)P7DCCnI~n@-C0j#rY;;vx`!*^|xpTR>%EU&n$h@kp z%c($4b8)J-Y*CAVE(lmI6F-$se-vR=gma-BDt}#X`CU`$xm26%?6k?`vv%nYlRF-> zk1BS^j|we8EJH~S-2u}_gkXil;Of`ft5sjvaI7v`M+VAHzcQR6^>d-bvjdGSK`Se? zDoU3k(EiOtb+2A}Jf-vVND#lusDIAN#7KpK43D7VbcMYgj@9_Zj)Uy@f6pgR54}Cb zEm{gB*5f@KsFFbsfAv`#wzD6)SwUno^rsAVBt(gZ^cxvV@3t>E=~Qrr)Afc{8zEqQ zu^2H^UT&}=5)_Q6p(UwKz}%RE>`?ddfd$wJxe(OU_*kZygFtKsCJR)VseyR47lEhV z%t{Jl4`bRRx@H~t+DXIJf3e9nf%$#!!+fGX@BcL|!tYSPdDDYG#kqc(CV@w!tb^yK zYc@&*n#|q&KF(iq>_{0bB=`lKQ?W?NSqKiS3Zo?3K#q_@zuz0@d0Mn|5#)6*o|s$G zupa1(17)lE)`gbTPrlw3nAFnqs%zF!4j zp)Q64#%f=U+rIwMx=QGY%H^hA|D5o8$NT|{oSqeeH&WQMXkTTOOE&0(p5I6 zEpZt0qF^mn3VWi-$O3SK_}w6PQ>wq_-GjVG3(? z1YqUO;Jdp-aDfzgTJlz*y6-F@X&NEXYhVbkjqX&8tLTzsy?s;_=2O+ct+wpEOa$GM zY#2Q0)rEy7N|!WO#0ah!+S}agXn4jshMBRst44aor$eE>e^1B!jJW6c4tPpWuEmRS zY=^_5vb{!AyfUFk zvr9>lY-7zy)l6C`XZbO5QObeC3PtK&+RhL&U^5&*{z??CDu#IJ{boa#27BVx4{PZTj zweyCqglH&>!OMwQl->9^b_S#Bj?k0rFS~AUm*$#(p5yO!L(?XFp2Z*Vr+?{I=1&@F z0+xcIf3;&T2fd>^CpvoRM!~c04FM5}=$KaVY&Gj3TIvRLB7zlqDP6p4rrpHcg=T8E z(m!t%I6Q}+!k5eNJK9{z01TL&q+gm48+s<8Lq$+>zs@vP6!CZ62j8(nGDRf{E4B8D z$=bd(DKEAFX8%_l{PvQ~yVI#9QtTw2rRmnFe>^z|L1H=g=h!MY*FZdU|S zJmM*B==JMNHtEbo&oSLHJF5Cg`V#KcHm3EW#2@U;P&eZnwa@esZF*uxs3d<+s@3c zOedXfJm{~iE@4l|N|mB*0l+WLxX8RYe-ny(UXo^M51X68Ox9?KH){zM3kYmXcTCKc z;|Lx#%KX-zpJw}9Qd4|;Zd0m-fH0{z&G9KykD>RIxkAN~nQXOX(+NX_Zz*0abCdHD zGOgs0{(dyFH81N3hZkgcduLC)e~%OLsFPn`R%bJn)3(3PSfvccvRNBWg=MQPe;nS| z{E2%(pnerou=j);?ky&65d|mT#A#?F^;JDA2CV1C77#cA%{XDRkLNX$Bi>g4TKDVT zfO@1iOJ430CKSPqA<}gG*U^xPI92j9zV+p}Oz=ptZ$nMOO6b=|+Q_v}vn^`vLDxs) ziEht^1P*X(&`mYK)^#?E)M%N8e;6YZ8_nTL7d{S&>Q5|;2K|C0+~bf0);ycxFR?DO z8UD+x`u$N`dDR6aioc!bJuQ%`6|*K=Jyfh6{C!-z3@-|AdJ!r1D)L1%32hbXvrGb; zcU=6kOw&2mtEUJUm2>SgDC@6QMbgQkN3q~xjX3q0?(bFc~{p zdY$)c7pJhdjIFX8=^t~SiHn6E&&_TKSyWi;d2M1#RS#-5&}5t*v%Ft+{nyrd{;uU4 z<#dAvCam3V?tN0niSowpS0jaw@b33N?U)Hl;*^uMU#mC!w${^L{0vY%e3H|4DXK-4 z_|)z#%Zl3aAE^_T(-{*^e=4&PJ*_cW$+yYgv5IaOFygxH@@%))@c9rsLJ=&fKTkt< zp!G?*Z}f@l8c&T>N)9_^dEGzZuWtB}9M4?zB36*EE{MYmc6?h zLg@aJolwr+kJR>L3#6G-hHyFncCYL-OGJSfJ?Kpk;ZXHxm%Plgu)=B`QEySD@6DWQ z1G}=D0#R`;D0^uulP$5)MNpFH(6gQ)*FzFTgXd>5eBwTfuZ}+b6hSX6ThCvd zwKLeGlzjfzA1S@kf6lkYpyDXUDIrwervpV6$VDm(WD%5yr^2FKB!T>BpI zF;1k-bHLOsXKlISk!OQ+ygR%Y6G;~2r#l2PG#3zf zlv<7DRGYD_#&l)Hl2LNS?r!(?v$3k|WR#s4q;NQMvu-=2!YI{1W2NJ$()ag0v}!NJnTWX~zdRx$b(yAHH}a^g&OQ(Z=5x zwA4j5Z>w3#Crp&Lmqf8svR4<4tsJJ6mY#;&EjII?&y-lm@EN`7)%xC~Db1}gNoMt) znhgnR-dJmjV-*KY3==$1;mD!2+zjt$YO{4TN1>Zge$Te+Es!rX~IqB&HUhRWPs9oTJ5fu#(nt!7(}@_>!-4hbFF$uU^O3uS;`Xax(rh zk7}rbyD0?cp6L4Xrc+}R&n_i{IwPs#!=A+vu5Y@M4tW+I%>J#Ttj@FS^wfQs7W>b} z`ah>B|0w+*rT?S!{|lwh^o99gTrlrO@BhvWf5NWt?TR@{^XX95g=_X1{#1u>-!NyK z@H^j(`DVLgv+hHRzosqxhfef41snH~zx5f0E3eCc6n-pWGmm%{cdI^MdYRkT=e!2{ zyY7>}dZ;wrtccNz-ch_X_QN_i3qSf6KmJzn$Op;Il9-RdU?;YfF7JO=_wKyBykNs` ze}yzD&|Cr1_CiDMF zlbJ-TQ#Ya)TP%P`$;+KbEqt-|8^5~Uf3f*6hOi!k;vmtgbevE)4kG>}eob~J-DzRa z`3s1X*Xj$>2g?vhAz;H18N^axo%~v6N>H(~U9Cg%GyTk@sE?RR+Vv3TI$?G;gcv0| zWS`KiB~-}gR@gAVonRBNvf}UAn6t=EI<)Akq?|8@qFbI2O89#96iM3&7vwc4f5b+m zNg|JTvC)+y6+vBzt&rFmRi zL<~AIfUu?)R-lBPvPWf%$>lor)lc+;87qoq>niCjUti=b>1KrnUg?m>3-7a9Q2S|; zjjSIo9A?7#h7GZ2wqnzVbGXfCk6VCj!}Y4ALH#A4Z|tKnf3%`8FMD6K zo&rvq^W_6mP-3Djya@V~#xp+oSZ41tpDx5{Ug`oIUF>y-d|iBAwmdOi5o=5TJl`2^ zHy*+NR2@t{s4LlU%brTJcoZ?al~qs2jVDjxz5<}x( z)vUwPmyD927{B(}^7*#Ke=pw;E`C21k5s4b`%0kWAR&@ArZ)p^tLuYA4^h>>-<`Yv z%cF(H?LSc;`gGs|ayp3y0~Nh52Z^^47BwSV}iC>zOE=8gu)si->`mf44!0B3bQy^%K9H z^&95CSl7H1r>xouCL9=B4ak$QN-HWIAvB(vSO^k*8or7X->>@4`*WcZnL4PS==A&#bC|~srQYG}tdOsAV6=B|+^OKEuen!e zuu-&Aw*PS$0CTEgZK)vmCHT{hne-0}p=*!)_cE3%oN>ig9qQB!KBLQy9Vx-@qYs{f zNESlM9#*1n-8xP(^k;kvfh0ke^R{L&M);#J`*@h7(Q|~ zvm%P<{#n`L<9c2;BregI_@J(Encfx3ysRO#8W2n;Q8~wxGYJFlTXqxev(IhV&*ssN z@EkE-O!0krd)~@QSa&ssb1YS&uKJ>A=MsZga|&>|Q|WrM+plXjwme?v;#?mtENxR+ zq?0F4@{m!=e@Nnr^zb9R`s7EatoQ|0fKcADYOytaO&q0y|Exs*LzhQF5gf096qSYe zjG7GKUyAU@b1&2uSqEYHOsstDgmKiqLCjt`L~e}kWWX9tjWO!@PbJ()?dy*!gCZYX6P zw9N*(VB%FTDyj-Se+X?I;58SF%_JVHkd-O_er<6!j0s97{>_)cw{Ks^fq=x&@AqgM zM4Shrcp_0|f8d$w5g4a!#K>X?AJwETccIl?l-_q46}>HxbVOS?@Cr6J7*X@tM?rk# zf17#*PsJ54{!1QM6$|xs^5R-Tm;S}EI%bPSGdtYHssiB(#+H!UKx(JHJq3%;3cc-) z06>)}l?_KqTJ`zA^aPP>b%nvpkIJs$K}(CfloW5}{-oQ#E1++Q3&gR*2M6~|3cnf` z`_nV2w)L^FX5osiq!Qm3dNAYR@*}bee{-;qhG9x&?;SoXV}kG(?U<*JW$Ip)Xw|LR zMmNQxt@tNPv@-NyjtM10Nmdn26VWpgR`v{stM~>aHO>kKw1GLue#j8?9OZPV93}wc z`~aAiOS<3=%~Na1KKh+>)*|x73qAhO^`X*@(RS&I2a>R_*A{O)`a@YYS)G1He<16K z5U|GDMQupGdLBfzT5UzD&%H}%c_ty5Tw={w<1@fZ*xH{U7swzksz^b7t*O(=EOYhK z1aeh!R*WBMIB6wWgxG-IDmylgoEZFknxYZW>iFy8v|V>Hu2k>Rsy?d4e*?wUnpi#l zMf*migK=xRwNyuMoF-Tu(#Z+qe^qGQNfpU5(~%&m)JnrBxaYi6^ZR5QKIF27+$z=$ zz1XgcgfDbUh(ea5lxR(MIY<ee|ZOP{h`&p znmlXL6`utot}G4uuQZFF*fwG)2xWd?PNGh<4n=uxk=H}vHB=mO_TfU4Cpf&?@5AKyW#X48lx#@8Ce z_p2<=I`=x}`_m$jr+D<7e=O{_q0GeUV!-k4PJX~mJ(2D9KmjTaise39lZF~@>C zP6c{sY&qgfPkV{=nrci`a#oO>V^%;etaJ6(RP-SiSMz~Ivpi}vf2n%t?I#HVQmJjs z`dps>SQf^%b$_&?zg5|rnv2aUZbhQ?@oMMz%{?El<$pJ7bhoBuP^3{N;SG$W0=sH6 zr}Ig#7^AGlFJ&fO;T$sAhIpMK^~XJ_vEc8IkWD1;sFnGvcZFYAnuBos2`}6>-sTaX z&Vya*@~=nF44o7ae*#tm8V>ocCVr7Owvk6@HP(tI^8FM;cJJ8?o@bSw@{bI*Wr_v(;x@l{@jX1K`H4_rf}*?v8-`$R&56mGHdKZ83|YIDCc6luWK}=Yla867 znI4)uK2q!new^&is)t%XJ)fI$5)&J*c}%dNxa?lX#Spsx_XgE@JxkUGMJ6JQIx4$? z6g!o!loe8Ke@1~Jj%d8_zcS7jm1$yq!UH~+zhSwUZ2=^UMT&Y^aym!W>MhpU{7qUS zT{=C0oov2Z(o|()E$mj(|Nhe-Acet`qakEQ@c1($6SI$>2=fY-;4ej9BfG`%Hqsfv zORbX}^iPM*!ft2yEiKA|(iK25GNgE6L|}wGrrr0&V6mV*(&!x@DdF;Ww*2zs6j?BUZ96Fuzb;^_=urH>e_Z;xi)S zo~ytwYwqs6+RW@6-|pY=`+bj5Tk zVoUq_e>ug$i6$1B*(seG{fhk$43mb-GbNt}Pj*6)JMdZjL#G#-+-(Gmt4#fxC0iWv6U)D~{*sjqw*o3)# z@mdD~SZBm^fxA!mjz~kzN9#q9UNuyMi?7 z(rZ9Ol&aDpfQo>0={-SOs0l>~5JKpoM+gu?Nb(Z*+56o4)N{A@z4v|Zd++#1zU3lW zYtA{w{EgojbIdW1`0SK+C}XPEUo5v-h%4KEcv2S0$>j?E)-ag0JkO>5EeOr3f8R&T zKAJ&!hqnr-0iv}5Tx>2!gJf>T@}?t{e6Xwik6yKHkWTKV^BS0X27cV18=1+r;=gyF zR;Fo=6i7tjOUEM|uBSB%@s+SP&$cCe5RiQs{R?m6#E5gQ(sYXL!S+$x%}A%}i4Pt@ zd+C+KNW~WoltIMpYd)}m0aPAS7A8|Q5rpu`y=dl1{et)nG(!^lMKAiLSq8z8@S-+uQ@)Xc(p7UzXBKa2y1D@@;B zdFAEI%on=Y*5pzv4ZJgzf4z#---QN*MY(_p(n|+Gk#Q{ui@@77Ov+AuHQIUYE0wt> z8P!>h4^na(W>c4)W2}fEZUx6`J&@bd;gf^W4!;QP#55@rc{uSCfHTZ;pKZCH-h<1F z&edy()j#3A7qs%ep61V9a^c3>B}hp}X)6=1YAFzQM(o3#+8v^IeIPe!b z`^lZ)8##j2l9rPC>WivdvX$jLp2_ zimH$o4}dRUipV@+3Ld;d3vb*&Rt?Ts8G5&f;#jXxWfZG0fAzIycZ@3weYnvI!K;uM zT{-rp#%;zjTMrTWC=*Dt$7kj<9AYjN$D=6Q%jfGMzt~etPF8VKjip-Cj1p(lJbB_< z^gw!DD=nEF+pr!$Hd(A&8An(EPj(mA*aY=XA~Fi_46T-^7!*E2fQfFlTno#Nbi$P+ zWQjNKrOGbGf0tXvjM_$;dFKJb1G=d+XlkJ_)+$z61p5T5gO;CMoeH9P;Bi-GlMXbd0(o8X^KznS#ECz}f8%WgYNgD-#Am8al@?a9LN|Je znz@g>f!>e~@gvvfwk4X$2&QND!7vs9DL}Fq@j1F&Tk-OHjqb4+#X6U9AKa?j#w+N} z*z{^)hi+zO#gE?Oqm&E)w=0a_H_6z!l9oJedGBnKQp@J*@)pW|u5!9t93p1&MXuh! z3%)WKe>5sUCNFsH8UbI?Vh%C#drc!B9wsqlU{d^xedb>gDRu_MKNd&)-qU9)>&w37 zaA}7}QgcM1Am7hBQjM$z3#Ys%{CfAJ%z-mSBh>b7yxw4fE00fBs`DYli|eO`%M!9; z7Xq)1dvx<09oBtvdUuMezs_^HVet(#qi5eRf0c*4uMpbYk!lN*%^JCx1c>Gy5wU$xF*3mny zN1eRUIF%Lerm517W~R2ReY;6w-TaQ_uaJ&yQ!A3*aa!{Bd7{WpWn@y(u;O5pWb07` zf4Q<^<9e0+O*Rqmc2^{AZ&>`$bCNVoH55h2=pQ!`eZNS)6+%m=@2~AXu-=wrF&BaosWI{H9xHF!Kw=vZ zla`#GC47=zjvl67LpxNDkg(?_9ji!A3@jB-%Y@ChP`rqqrPp$;aT*zXI}aLUe`Z71 z#l?BhPm$8Nse?2Bh3q2@uNbFK;Opw` zALI5MSQiSNGBd(wRp-o=@I3}XVlnIUCKR@CKF@$Zoh{O?q0f7U2+cP|cS?!oM2xTy zWTW|{1j=J@-l;9&N?9jrP{Ak{>fXfvY6IpOmyz%Q>H{8)ADAl)*OOGjpOTxfs#2R)s zXep7?YYHws@K_3*->fpgkgjk1pg>R1jte?coed{>z;o1L}^S)Qo*_Es2Ivb|c0gk|Fz8Kt~QByvF5 zY=Ic$%zUQH!~+vHe&vPFSvckLl6~Nf={Skb3l!s!T95nt6h;xp%S$dem+K*9F+Fih zW(i>x2By}i05E}S-ecCnzTvLwXs25w1*Vrm*OYJsQb9@0`HAc;4Wq4V zhoL-1b>#9#nb4fDcWW}}Aof8m1X9xm8fjPKm+X)QWZD;%S~F(+|hS$d}E1x3-e4 zVn(MC*G>m4^ud>?obB-Xs@IsLxsTgDu-{gfu$YD6s^zv%f9<`bFfg*xfeT^kcHhI; zd2vO0t7`hX|L|uT8;CI}&hw7`!S4KvQm@mG(w{QAM8M*eli^afk@xPJ9XbHn91td) zluB>VK^Rn6+qK=+WC^Pcx1xrDjrDb? z+RTjhi+N$C%z)|A{=i%(0iVn}(bS2m3;IE{SN~ksviI39N~_R!6G)*g(#*#)1Cvnr zZUYVFo#4QkzEFIDF?m)8-X5d04GV1(KN?V2@xWbQe|Bfpn-=YyeLruhKDyL1ntwcZ zQ$egO;YvHYtZ9RGq9iSb3-)_DJVCWDhIU^Vrbi~iX_J*A%casrx>e99H(9M1p}zZ9?`ZsJ>4Q#4_nK#TYFTUBSQ!|K&& z0$eUjf5@c8NX7%<``$2Ql#8bQVNrD~9`>ZXbL~YHU(jh@8b2}j-sT`#B8#f+79L=J z5hppvP^R+D+Q2joPPB2aX)Ef^7lukSvJ{=Zu3Y|256r}9F=yi6EB2-+!ql5fVH+=t z7iWiN#R3xGthOXuld9lbxzdJ=-2;Rk{IIOS#iu{)zh z??5wY3v`xXUrfp;t+IRH1j2^is7LKcuo*r6G9a*>&kEccLP+3xBk>!0` ze+T)<=fM;*v6+#ZX&x?7TA3BJD@X;Q!H!+p@vAbedj?m=uDo47gCy2*qXIKEcvNoAc!yRn9JK9d8rhZ;% z-}ww-TDl|Us}t(SRPlFbzH4HLnGw}je=XC8_oNPnLIMpU-uu z;n5tCA`SOWqHlZu#wxtkpX&zEJQ&K8M%sJI{)hL-;`cPBK{&A(x0&`iC2S)`f6Q&X zKU`|v%;PswqOOMhxh5{yz6%)%r@bQoe|!biey?|AX}2rldRWb5ke8~^RVF9Vs=w0* zoj=Rnq&{h&IFM+rjCkt*hu7WzcT+V+6(<{7Etzgo6r6MVpVNPr!AWK#gMU8Izamlp zmr4Su6o^)hIgd|Aik%YJQYiHpf6$s?_%je5tIP6RZS(w7l|8Zc6f}KNjMCbUTbO1O z0nM*jmPFHstDO%$ul-gtPkhf%iFdZkr7em^Wi2xQS-FU-G~>5~aOKZYTm5e#r5`W; zTYc+S7yl1d-zv;mt#C^9I`ey?+EJrFf0NAZ4jWiw8T$BGozi#IxUY>KesLG8Cu|^!*G6K{BOeWKQ}f1FG?8xZck+kDXZ|?0_^+>`^X}g?f-O<30nbVcDUDPgMDOvNtt*z6h3+~0W^R&|m zg=je-750NUKBL!3w2V*LV)*Shy=^6siLw6dZlRNfT+%C%J`bB(B%H@ue?GXPo%u8@qt%Mx>6t5~eb<<-8fN{FR-^K;nX1YsxAQz{{?&6z z+CCUqg1ZyJXp!qP_&Vy(?=3Qd2UJayz0!>QvZKYvyzi3mG!F5=w? z3WeHT9CD~WbSI$v03J+qh0!sC)Rb|gwTW^>p^+r6XQc(pMIokBTG>oiPvv2W*alOh zYD?lHp4fyl1f~Ykc76ZZdTrv60Y!9B5QTGYglD*I+=i;t3WkWJ^@UFX$sFmivi!XM zd!0{FoT9bYO@Sx~E%`#Qy0GkpJFH&T&%#{|?1M)cf6wA4g_qLkd-u6;aNDN4KiXS| z^7$q`QgeBwXJA(QVT};D=nUPyv@N`eGACeQ!iN|i?h2jeP~2C3jQ z3pl?Yq%$;0+D6u2dYvy0+?j%|?{c}!V`&HfY}S(V4gHIxA_5sDT^hHvxJSTNAM=ml zdL+Fof5roo3`Of}jUhoK+UCmX?e{GkcX~XRzQzn{tsYj@BmU6GXWaFb<1#^e>$xMJ77vuF7noVmO3JXC2V4}Z-^;(Hz2X3Lt(PT_WZfCg z50KV#cCJ-VYw*J{3X8BGhGy;SK(oPzS zdU2GiooiGH-%UX&Kqw6e-g;bvYHf5}c}w8nnir$ujWKdnfT zlD53k0cK8a!C>Oo8GO=WT)zHFjeEf@f4a$2Z+ShsqvwQ}b0uPwz_v#?{0w+ad7qk4 z44o?Vx}xHgx%=lvYKR#`++}QV5CWR&5MOedmT%Vfh^PK$-5dWgK~&Zj*guiW_5r6B z0M~GfHGZ&n)x%xZ$)RJCE}_^soeo1MgH8qbU3L=Dt8^aR1*$GLGt+9=l~U+%NoS|NDth!4>{|h&iRFb-%rTHVj%~FAxNi}(Ee7*y8AAaI^ zw@A6}qx;#|l7Z|aYY1#LRK&jSX;jWy_b{q6Nj6Xk?JOPTp8k|8qJq}H<@Jx z-g<;qb?&CMi9_}{wH5s=L`qvNZ{*>`YAzJl?d~Q6$%N@`6XK1mn;L9Me+jUUX*o{i zO=~#uhWPL_Eccbq9&U>3vyyI2s^N37?e!ZGk(SFvRhP;eHd}a%ZbTKMw0d0l0lR%> z*^&JA%6e^fb)Cb$TQL>ni6WM$5(t@}x5mYecc9qr^%A3OZW=sI1}OnlocwZdjZc6Z z=y4;h8wuWlCzWnhiEc!2f5!rzx^Up{7?k}S&&oeH(E|Nx&S0k~4oPXD`jXT&J;2sx zxi4L`?ky|J!5c}Y$7xW>toRVJ5Ol5jDQeKrzDIY&YR?}%;|*TVp>rEs#I$-Lomows zOO3dJ)13sJ<8UQn#TBzmpXtQ~k68(~W;Zh{Y@_i_DneSKj!KtYfBn8r*G^t#j=rfO zcD(xNl8H@Wb2w>?_bvNg&2$;lMlK`USmTdbt9MORntZ2kN{y;D)xPks-3)fOwD*1( z?z53+J2TA$kC|7%zShXTbCS!YPBVar!n=_w4O-{ZoUW*TIrVzNBXSkziZ%0Qj?mLL z&l{n<6oWnTQ<49YW@J!0N}*N=NL$Hb8x9e{Hhm;csoXq#Rsyrge$^8T87 zoMWTv=&o(1|9w+;r%c(EzR(ZIes$d%b58~kk>&I$f%K<0ubTV!1|TaP5XG+kVeQxS zmb2=;Vp{B2UIcENGR<5?OaYC9sayoPY4Lwk#M&A)zt% ztJfT@87=N!nlyFJ6|$546qxI?vMw+)Hj!jIwWKM$@cz}0-1Z8UpmjaY(W(!2KPm@5 zLWfZ*hhiH+e@ZV(@Onw@`n}TdZ1aG{gb{!Az72IkdRDSu=xp+ZnDA=DGoN&@1MzN zh}hhjF)(PV?Io|yjA8(HK#0H0lOHDg?Ku&DJTvn9MI3)0Zhv;rs;RenUUl}=tzx@f zLXyRYPQSFd$PD288$;JG3_?r#jzn5{%>&rLmPrYW!AGN%w z{o14&n&lrmXxYYU@X{a4sd(GZHp4T>)LRWb@`&g(y`cezBM2+yaH|5RGz>0#MtdMw zZ)J121k(0`%Y`*dJvIiD>09F?g2Yz%2T+ZBKQ_nV!++w7*S^$ULc zAIe!C{V6IZTZUYFxw9gv9j#=%YP{&un&vKEsWv`TeAm#u41BIgmcC0iLj}kU=Zy>Me5K>rFQ-Ij8p1_f+m_c8ZW+7 zSQ{6bpiH(b_*-FVR3u?r*sKrO^iH-CVMI`x@Y`Ct!N`(ppTwrnwwSy1^1xBfcrS4n zWq%uqEnVHNZ;b_zxLaOKyk$Rno5|tIWI&+4-33-I0xhxg>!lOc!QB#*UL_ur^lY4Jht5Li(s^&oLb=P?ld@Bm4$0;+juaHh z70Yj$lmIN8xvZrZ_Whj3=^8yuLg71^yzYwBfFhhfPIKiQyD9WiqErwC1@F8C(tkF_ zrF*A46w94gB3;&G60dm3?Qst&``Ik5;fni&;l{wf0HJ+~hdG7{x?*B^ox$!x_nuNk zbVI1&O-=;_>4*}M=bcbaG_zTpR!;u0##!=JstVqA&(AZbPTva(S^Gi43VWyaD{sx7 z{6L+z)^(=-Bu%&BQTA4wH0M(?rhnenDb@*rw^+}^6qom|EPq_;Tg-81WQCBs%J1iD zt#;{We%QWtOHQxfXKGZvn<0v63d00bkffFGxq=ibq*zmRRJ0t>|`#5+g^MpVnKUqt_98&FJ{?8W#rS`K01WtwZ)c=YQ%M=kK;q4(E#%&c{kT_Z~$}=uE;0YwAU>TxtVmH zITn`AEGjuHT++biZ=;u;WL|fFTymv~b^@TAujT;|6Bl8-9y^2bMn~!f?oO%m#`?~u zC^)YH0}C4lBErNbeq7|?SAV5i?Gvw`e2%1=l_}UEIt&B5=^5vSsr(Nr>0_QBtT0(K zGuq10UQQE|6pYuv8X*5|dtbCi zXyszN$s7)kmb{>E_^9@9j6^Myc;iZb&1{*Q>Pm$31Q<^~Na%AYE`K#X!lC%C$0lER zvVqErrU&6o6(%CEGVEln~k{;HclDa(4Db970#y) zNh^Mp!xiJ-Z1E_V3V)7bjt(m$rEZq7I#zu}Y8$z7iDZ26HNKw^>F?*~I7><;9+E}8 z-iSKpI`McW$C@7YYPuhq5f#d^c&fZt+CCChOJQm6UoF8n*vd* z6zg`Bh?tS-)C$d=7W||rI4G&Dbc(jbTDr^v*gL%}gStZ5o3+BXO-G9Bo9g%X#3)op z-~l_IO%q=83T#EB9h_kEUn?~MdA19(78&QP64=eV_mx)oy$GfXdunX=d*Hfl_6l0o<)}Ej`WsGK#*lL?cGpAy-2G^mUy=WF?SIQ^70)1rwxx0?b zkrzPPbU(u7r%{-Sk}O`@gw$P*w8N2ZgjozTA8;9o-lcW`Q`p&xE#gSzX! z#kKPyV&I(tAk;d^f+x>!qnG6Ja@3ZBw%UzIC#MkR(_E|uMXS1ItIQY^v-q1IU6(Gk4%X zaMv@qaN8YWazXRfVB<3XvQr{D3rJM~;fg%h8$U@PKRc=DBu@Wb98?NHE2UmB<_>AF zynoI(TTHcH*6|sByQ6mgD&Y-9&{xn_y45l2%-DJDOpis*H|^ly$9-L88D47S($1Z& zLO!a#C1Y5P$}uu^8)aafXr-mfaZeByd=}oTTNT)(sbdSoM0s)s2y06IZA~W5k-8UW zW;37$V^~_;?|iJhn7*l7SPj0oOeUkA2Y-3~#StF1f)enP&DBAL`4rnVMyMRA2RHpf znPmAsj4djc7XnxsCRrrQhh^X=@WC{GVU?H7*|wUp7=V*1bOl}WjHN{u+_eGmXpiF0 zSwFAJeyoFmOeSJ)Y2&`{FZ^^~;bv}3c>J5&-QEtP^0Y)|S*lIT(p+xr65*q$)qe@3 zKWl@KcTE)cnm#jTUP@jgN+WZ8XB+wrD!ifTh--s#-(U#%SeRN=z-umtVVO@9j_)I#?H zw?{EXH@!!Pz)9uIJ&5tH`CQE9#vF5Q8q5i`HMA%29@{M+(YH& zUxH=KbQ&nB&Ap4fn}K-1seeeZx?767eP;KmNI3n6_qeaO=@4;uqPs(=KyJW zF3Ci?7`*iMiuSyQHc5eBIOa|JU3^id|L&VTva4I);U1c2BTH_L%bHuOZ>_b>ln%+5 zVblEd%)JiK*q>tpUFf&k2n19i-&n`2((F6c@V=xStEi=r?H|Mk{C~jp^~X^1yGv&# zO+O~{OH!Frt1{K`0T4FJX$3#axw5iw5uJHYpDxBbCuex=j`o*OM1659<%rnqBx4~a z2}HM${+>Wd7l@7KJlE`YX;_9Z%!`{f_Ux^ytwo(Ihi+$c^$K2Sn0>1;FB1mY-6**A z1$|Wt|wea{Nu^6IHJX--u&JI&=*#MLSoz+iJ@*aX{pmM7*C@8G#e z5m0z7EqJ=K(JhWk>&x4~-ByZm&#FDA;`H-{sq1H0CvLr;rGK4oDSqzFnkHNh0a_o> zD4A~)<-BTlj8ry>+tC&o|1cUuXq?b}H?w7sA*QS+uIHkKPu5wF){VDZTv-edF8mx| zesMRH1rVY8*lN9vy&DTGp_1{06+OcI(h=F*{=t)s+Oq5M4K?lK?r>J&k_yY71Rn4F z6J|A#iS^IhZhyK-F>TE}bND+jsd3PJ7sUTUF(NDEm+5T{IiEoL>vb3u1 zHl>yW28JQirO?c%WtRnZxFRv&qxuGG&Rn7y|JWMDWu^ZKf>Bt8eV26tK*fWa?cKDE zh|@sQmce^DHdEJ-5r1Pw>L#o?j z0c72kxO@v4s!v+@%oC~@Z)W=N#m`rp_-{X?I!$h7UYWIcslbREvI7xmj^kHj6t+Ko zy7V$&s(;O(HNn7)%fz|V-h3s?LoKflnw@09L95~4Kp}^}Y1bZnU^G%oZA;ra)T=$s`u4QjeGkO!PJ0zzm*w@(3SvW3TNR{RwRoTjP0Sk0j-L-9v%&`QlA8uhwRMfP=YxI948YLgG@&es;N9>3x{ zw|_Wb{o``O{(OGwjF&1|S`(c#MSY+|$rRAyqy3+MXEifV?98O%P-nN}y{6hD>RW-w zT#J=8&i~GOW=;;(=l%Ls3?fR0+29`C4kHx1rT^m(9n9 z2D@(=asC(^`}M_TDTe)^+#D~0(_z^pnSZ=QHag6cM^_L1#+vfOd|H7=&y6cwr(DHa z?g#Si2Nl2vZvEC!3Q~tss4EzKaDn7@28H>YF=?Ab|O3``hPSR z$%W0e7^ZV}(lkA_{oL=!N{J>bTeJSevS9acAU(`Lufi1hn2|bo{Gg$K-aR_{LBsF7 zJ=nMiFDJqxi|2Bkt3bu2I4;w}Bu5W>%Kr1QY7#k59V#d7EdU*cI!a_EC;Vf6WL+@s zuGcl5pO?M=xfXRFqILtRV^+`$4u8k6f{5Gf|2GU0wDi7%sINBwecCUr{GJ&fTg$;%JNK(w{G)*@vw1%y9 z_!Gc(XYDNtH^KV|-@13t(0{w2N^C6M@10yI_Ko=MP5;B~QOlU!a?J+!+p(RUF^PrF zRoCC?*OQ^oZRK(PZ-VfHSfeIiFFjE1&XLo<(Wf)_q(XEX3HYh=mhTDk z`C}1A*v{Sz2WI#(aJ#5GqoI87t5pO_bk@pi zN0j~!IItf2*wqWz+w)IEfh6ZY0ZbT__$RJj9PNJ^WP~Dnj?_xRrf~W4PPIt<) z!rU6vzsi3$)Boaow{?r!J3l2|#EsN^If~pgc4rxTU*d^}*N0zN{9jz?h}^5zS{3Ot zm7#4-=`0>ItvflCz$41zuAk1w?#i1+)r?P6&|wdWtdN(AzJKu>%8(3ZpiUm_-eWq9 zm=$!M4ilO0qI*6p^6QP?nI_4e`)N-xMaVB?2ID*fskSgyiu!SK(C~LG*rKUQQ|+$S zYnnMvxuvArW1<~g?tQ7HU2*IQtI^6ml32vTw zoiD?}w{9SEHS{JeS_3uKS?D8~8STX)Sr zxZ-}t_#WN=U~WDW$i>UlqhQcu4q3Z85}M*2S1DhCmG;_PDD5ZK;cO?eK<(;@LUNQ1 zVy1+beU(Nd^y1l+TFCk@2{<={dlKd=++Wo=$KOQ;wpQgDGy9?t9?p?mcYh}-nl>m|w3P5J2iM=!u>5mWk+X^h{_ddM zr^52eF|s= zg*{Yo@utb(oiW3cp@>f=xnAqQA%A%vDQ5mj)9B= z2NYEQq2>y*?~vdS%asv9K7Z(;@1wPR_-j6^1cG}jCvqRVYlfIsX!<|Y%cgjyJ?Xs? zLB7^MpVZCRgHmboO;jQ<|I-O2b(RVfC-d7cP@h&vTUuELIjGhx=hq7H$g6-dDU_CM@4$4%-WzmU`Oil^Qj%%*n3!`FD_ z)IS9WQ!2u8;KxVZsW&<+<)3oD*6vm@L}Q!>W{Ovsza#;Q>BuYBdL)0PghAUJ5aw5AH$&g zsUmuKhqlV4`#BXsS*=D}jWn{fq43HlGAWpXop~hIYONL%MLYLT-`tI`;CgS_&^Mbk z-?v~8?~1Vq@I1od1q8>jXP_jd4NKNH8(X&j(UB=lB4U@N=zn4C7EkrYpDy>A?v6O8 zTUDERr?w3V1NOWpd;#)2z5xMy>$$$UOmHl#sJDwbfidxJ5Avzc02_eTw{V)P znO};`?Isl#r!OIa_cj-NZIs6kLp@(HhgTI8zmJyOcj&GR9mXmat1O*1Jzfq^rW<); z!;bDEA0NfO6MtLBi9_`3LgYP{t|MTpGvCy<3o76}RqrtN7qcXFnK?!M}dH@9s||1{4* zhk0@!y=N+C2oJ0~EQ??8eZcFg+=#G`yDe)~vOUIyy`B~t69pijHdlOoz`ci8nCwuy zvzvx3r+);RIBH9#%U48Z#oAfqIN6Y>^Rqm|2>MSJM=qDs_K(3+0%v-?KM=%-^P#H^ zgH>*A))=jgCJF&=&|llVzTPG55^sX}z@!z6@JQ8p^7)?;jO~Y&1E{)^@A-p`F~HjW}HY8tc5)pxY1P0?|w%<@_t^FWkYTQU`t zFD+j0NuhTAK83cdSfzwm$|NbY{M?6b#7I?6b%?;fJALcNPQ&YW@^`$kV8TwMc-5^V z96p*N4Sn>i&LzlKxzNe?`1$T_lsiIpCV$?lYm-ariSD=?B6NLe+2sb~4Y&e)ytJ!} z62ViAypqUWa*C2UdVXIq*m?XsXQEl%{guzFiEaogNR*2=yd|}4>lD_u3-v!u;>qfN zds1vFGhZ9+PC386E%&aqmyYz@DX(zmGX0Vf3AB1$i8*H34~0wPPT&VLU( zl-W4e8Vwry7iMA6y`f`2=8a8UN4h|uuitw^6|i7F zCL7k5GU$MrksxBx2j_HoR?eIw!=WrnY%Yyk=3v0J1~tCL7HA^R01R`l3}9zC%YP&T-9b2p31sUV(C#L*Iv2bblv_*Y`Xq zbx0H!PH&Imh~rWujz_M~8CXE?q#)}wIn~Q;(zHtPpX=A)8#bBkJl^9cGlRB$!J{PP z3J1;>Q{jxN>#v>pl3oA6S(%}tWCc^6d8e)*=oEPw6mVO=Z?`@b%#XI!*r%MNO*uQ zL7q#(^6~f;yIT!}VTE;CX4eTLF}IeZ!_m5r8t(N~`W%asjf-(m@qbE(qy?%w4CN=7 zxHJyUuaC@VtK&0lXR|;)d-q2x5mnP1LJnBeo!M z*<|c8xiWPlLD<%)+s4E(l~2bW9Ra(sxS@MrF|2hkK1lXn&CLP)9&}}7$p(#!3FkS5 ze@y7{xC@}{xfeEWbAOJabM@OGf{UrVsb)wQ8$Bc@)d6Pc&^wgnglrThO?-6HTtb7^ zOGrFr5a43nZS)`(LXj9YcG$A3^vBu7vtmTk^LI(&X4UfZ44?uvH^%l`z#Zw)I&0pW zS}mK4fvS!fq#2#j_8>*c4|?`PR6%i6XGeZvmH5*nekgGHh9)LOfaKjib{& zIx&%n^=tPW$&0xbmg&(@(!TOBA>o^hiolJ<8o*wd)*=h+FiO7t%j&k+RaW#MU=O9G zc%Ij1=ar?&VhrqlKYS{j}(T{@B>HZUxqR}o>cczzCg=Rot}28 za^T<#*m0Yv+FdGWEhqusH}#sNE3zsn+?n%ISh?@W1%J#b`)q3u85$m!MLQEi@o-5S z0S_$eK4)H#_ibn?(ODtMtHd>^$Wt2LoD9*eaqJeAex;1CFgHz%!F=XXJwi@0nu0YAusV}@Ty*)n*B3z-|9SoNftq)HGk6s?v@E6k%kF*lwjm^#UMQL6i6(6 zV10YV&4b2@CIXt>%*y!oRS0|C9_?ll2VWrv&TaYbfNLf*Q;JfZ*aSN)`eRQ_P$sh$ zMhGJxTo&u>>oMv6cUA`WfJWW4*&?r`Eid^XSx!@C!u5vZ8`8^jnqgZ5$(J#_Ak>fIlAfBM32DjQ*0LQ zlRFX`-*OkhnPThf>8h-~hAdWS=W#yJ>D~QYXLPuqjwhQc(i>jTr z9X>Q~A}oLyq=sIaz*U~XY&Gtn4vN^0RbA%s>Z7%`*9uyNgo}=OAE~R_VW|to?XMO2 zN3|L+qYmE{r^6in^6@q9v%-N>!XI;@G_@k(FTH}W-oDSug1 zPSWyk+j;b5{on`?()a<9CvYl_i4J3_)_K>H0i^TNJ)70wQcS?!t6WxLSqq<_61x;V z$Mlh_d2~?SNDa&$Yn>15lE|C#}HzC8dk16BUgEuWfe~$-0MRy%n}U7M9_j zy9WriDmdCvEZbMpy(ePlN7rNdUVnu>lPjNv->td%<)H$0Z{|gr&a)V<-6Jn`v=pkxWm%=7_?7^r(8V{+{U=is z&GYXox}u$ETc9IR%DW4LcHrgbB}1D^e0|u|!YiI}0gY6xacgqq*GpMT(tk76*1184 z2~fQ#p113hA7efmPv!-UD>jP9_}Gi(0eN^2$u95%Le!e?JA4qTW7N&5VwjNHpDhkx#B(_u1{>e^zE zF6o%~7bs7lzUcOzqG{S?9X*t2*_~E<&MTX13j>w@%9%oIUJCNv*PY7l2jEF~H&3jx z8*=gyp-h(`tEZx+lV9iOwhg(O%~+Mr47emsB_TW8NbC%(eyw}BUIKnUdpe#41@xzp zXkYX-YsBVR(dDn}lYb-SLt@`~6v(}oqgw=Z35(z5a#4>oGr`~#);?=~4E{E&l}(4y z9j(&qesvJUgfym-R$P7w4vx5N9!My?=Dr>n+sGWnZqA(~rIC=1T!VNQ#_637!WDVX zw5K`894rekOl33ic;mCw>A!K0Fzb^mF~&0@$lEJ)ICt?2@qZvRf?}+knK`0iKQ#N* zJHvNAX3eN=%&BVPobcscUvo1;C>{{exUD41CnSmmr6=hg_JPS^qWrP`mtc@@5mc>@ zGDl@;F1zVf#VV1%)>!ww4rq*)89T$Q^wFsh`MAi3LW-UFYL01lD1ju1D_}u7mT)Dn zUSV|ugC;pXS%3c&)1(S=YjL@5n?KG*G-9X$S&0%TX(yGe_-R4C)H2f|S$h!1?v*v? zndEYp{jTXS`nuG$!P@QxD(!NXeCfp|4X2>B5#mkTSrzWN2B<{4T8`XYO z4aLTUmjlK_51R5E#iZx<$fmhes|S$V41AFXO8|)!#>acXxJEVSs8dhH?%ESxn5>=1 zGj`v#!cB2$U*ENEUg)>7L}_|OWC96N;u=9Flz)N`Zrj3$$T^G%4g-Gs+*9<*)3 zzv5*}y$Af-NbH;w$))D1^ezT*G;C`E#=G3tFcd-nEZ8?7UDndYPq>O5^8H6^E0?Vf z(Exn48Yxr^-gtJ!POvBG8Pr~Wxu1o9@&5%v}vwW3JI*c5nf60K&+ zgMXKvgt?^7me*W`=nL@sQs&-~I~Tj}_omHNX83p*UhRpvk*I|*M~nMehb&#q{xba?V?vGn}2S~ zX{&rP^$vtfa5KFbovrf}>4!`-3R*8|sI7R0n@W#)Rgd+(?7~N>PvwN{Om`P@`bRc= z*d8~M7&fr*y9&RXXkH-bx_6K5wKxeD!y(CEg?g_i!A%-D@7-S3Ey5bdaK$ja0DP+Km}{ zSgr>^XspUv&q-a`vbBTSWJ){LgJg$pDGK|0G|Lc1s1(GO=0^K^V)%{j$$x2fi`MHZ z2H<-M0gK(m{*|!^bbx@1P$7Om*rk=%shyMs-bzesl`;V7)ZL*1wmR^2!c*irLV8N&4Xl@hSD?{Fzasu{!wn( z<#eXIgTd}>t#?CuuIioQ4Sy05J~t{|Ko80Z#M^V&ruR!`LS|t8GcUyiW^vRO*l`a} z1&?^6^?fU+8r&L^lFWldjo13cT%6Lk6mXloGg~hkrdN%Z9W$y+AQSR#%c2MjF8d3# z+8ikK;86Jod&@6_5yN?fMLG1aBr`WBwn-lLBPAUQtMEWPfIFPm%YOsxpzfi3Ax3%W z+oyc*sTzKn4iipSbI+lK>4|!0!6PY<+}CS&3LIbCj^8Q3CoaYZjs!uKW>9$Lj*>`k zbdYp+%~0gYI(!733=6DY+T1_-7Ieh1TN7$gx#N42NfwKo@AKJB!Qe>RblayM^8xGD zb3)t(1Z^?BRHG*wvwtYPrXU>zX~qmlKv}}rcew-a!Yyyios3yN!Ncs7l$oiA7?CY$ zR<;d2us$XnJBNAqS!=5CYT*Qx!&;SVrrlhu9Lpm1zsbKD3&_szLI;qXCK|eggy#X~ zpJIY>NEbyOck7{1rolLEC3dH6>}%k-XA;lM+sJl?4yJAmJbxwQ(LSrnA^n)n@3fb; zpJ;tSZ6*T2R+TW}JzAt$IA)1e8~ z;^rKB$sBSUB{lD{?|ha!W-kO0e|aFPW^VBZI%n4i8G&O+8O#|l|KY}EcF$S`_0G%kb0~8J-SitE*X=U z=AisZMOA*Le-uH!>rFqX(Nud+wC2=8TatOSzCZ74nac4-UsEb=4G7x#2B2aY{v{kh z(Jm#@c57+Mn94{`*oBTfA{KQ&C%-t-s8HvzYQCaSntzg=XnZ9jV5PVL*-9T5R%-J+ z*gfAdu>0NFFtdyW|M$i&dW}r`w4IxLE#FU5f=HT|zQ<=W^ZR*LD>311A! z?hV6roosA##mI-q)*Pcx(hLr1Qpl?hPFG*nbhLN?i70-<7qlZWY*b>Td^i8K>8CU4 zocxH3MSlvV%Z}xushEUM%Qj&e03U4Im7&1FU3)(~ip7$czTJcIwg;=3`gI!7;x}n7 zql(Hzi1d)@Jb{Z^Kv0Go@-|&y`p7jlK7W$VrR^5H`K6PsTC5PAIqcO{FcRk0 z&yb73g(VFH)BK9oRENhL@Xoz79Zpv*y1n<+cu~ik9dk}u;L>`1*8QxD5#9Siw*A4l z86!pr>g-HV%0K|LXl4MXEoy#BFNH8W2)CkatG6Xt1S<6pWd!YPuf#5$TMyiZIyY*j zVSh-%OtRI=lP`Vd2}wBRSTxWL?+A2_M|t1Wmlj33osrqQk%8Si!nyIVhMc{W6KL{Ipy zcs?Y%?LB=)t=?Nnta|5qr0U4^Hjt7u5fHFN?0Z^!J0kIxh!xZBk0)Wcj%$H-FWGbUGo6So<@%hx?z2+ zn>E%K*MvN8R&6h|AEXib73or~XBs%ikln(DIfzyksq^A8NN+Sj#Prwd5E~+Vri>e@ z2CKBIOPGzQZI|=G{r(uF3?etgat7ZUs?k8g5Qo>8fm^VL*Q;vuu*StMc@>ebb+`PJfojn;C!c z?YzkLJ=ipk=&~7GX*3b3Y1*TQCD7Ub#jx(?H0{V(^1Gid_(d^Lk8@`h6w^V6>Q#eP zI6^q2CQW5u>Kbj&hlHLFdyfvlp{^wO*7KkQ<-HYa?n(_EEf<1wIe&J3w8eL<<74kE>bf|Z=DDr_26ic-g$&p{)EXc+SUcK+HO|{ zGnNo!_0(u7ADu1)q4eOk~LUuxOJNR!_)CyqJLf!_Q;y)vcusN z0qdRufAnb@fmXITJ89~d!LR#;P`6Bk+ULeH4dCThTSp+h8P-Cwxm|F3((RP031|Gy zw-)V<;xPC7b>^`dwl~q_(2blty)U&wKzFLf!1aeP4qM4L_Q77|ubwrVd6k<1AM)e& zzRvyCYdoppTzita?|=DiJIP`Z-}g%!)o9W&E=T3HSo>WmHOjJhw4T24`<|uxRg|*1 zzty<7T293?_eEvLs?Gy$R#>KPN8!Mob!g^HDDX}}?vwuSOl0iw@*dwV0poI7T*=qq z?h++*2v>p2LN~d^US9au3Rt(3_TSZunw~Sc8yWazugTj(wpgA5-}s9x>QzX1ZS*?;$MmmHklR*U{=k<}uX zKc;M=y>*6WEz0Th?S-B{$<=GB-3khC*U^scmSqN4YEvD@)d0vH3+D#=haiuxQQtwD zi!~FY^ufcemXPRQ5lN#@E|9InWh>IwuplDW3S%`>Tz}8kfz*~Q*6@(6Fk5R_;^cyM zyg6?LJ8*Qm(tx*HI6oHhG`yw0tr4#jFT;KfnvHle2C2xL(Wb@k18`f_zPK)7ig1Z> z`ZJ_pQ+iHP2JjNi?F#^Z8zA?Itv7|n&harJ@9^DGTjB%buceRzraNt|AXPJE?`90% zYA+|QQGfn4AcgXyH06q0nAI-#Xp##{)9H}Ybnj92H}!$tE+t8)B@OPa5&0~!Rq4}l z+5mV``T*&Z9kNa@U}xR7)Co82i_|x9=0=#`-%et|Hd5|iMbEt3i%gf-v~QBrOEa%u z++HVQfwPcSpVo~CCC4L$*DmDUxN>UZd3`2%axBy zkmW3|uT=Qg8@h$fSq?@cVu$@#Q}L{>Fph@5a$}5C&iDw0BDsITNj|BlJglcbrMchc zSO|y2BuMBuZD)S%H+T5B1%FgBnK;ELT$E({nHEVdj986)^Hg`au-bZW4=3LL?hE(& zw|`KzvTLs+6Xn!gAS82cS69r)lM2-%G8VS=tEo4z z6c?;k)5lLRsTkmqeKTNTdgo7t&%Rt|xUL)j$A_9T=jbi>Gqfo!=)HRRGUDZd=$rRu z8Y8$4l?k1zQHf%#xi;?B!l2pb<9_nK+kXLscd+-;pnT8z)W&^;Vf_PyO`UUwENtw3jdx7_pm?Rd|I2{-phm~~Rq3*&N713bV{ z4IUm#?q}(2tHs7{gFU^lfh^t0rGF=Td}7u{HFL(@be=fw=G5ifti!Je^96i+BL~3r z=RIwrBj%#w2&!u9oU+3+2vp2#04iZEXKF1beDm_M7PfRBY&U2}?!5I>oc6K_hs-u3 zM}YBeP8nw^V7g
1KBsfI(ZSVC9VoqnuU$plHWUTJ$oGf`2=B@Xo|y?0a+4u9>Q6d|NzV}jk~nm}mAF#D`Jjxddb$2*&`B%o5I6RRhT zL?Il~53X`M@tN;~8DrXS>pb2-;nUu7I)m58?hj18+gFH=7&!y13euT*rqQ0v2i{sX z=ok>Ny*0>R-?fosE=lL)q^mb{?*$9bC3T*Dqp>P`LE$1``&5p}T7NuyBE<5Ey!Q+K zNyl3|`{?!`Q~85#|FsimMz~L0MV$%$gZ|J@KkhyMLltxGa}8@q@}Ub7lL5ONn69_k zaIP=T>O5jCFisIj#kEub*-+F9su^!T=+N?I71*yo4-0A<_wFvzdHi;ZPl*#dcI#%; zX`yrlo(uk~mAN88OMgbS&U+VVDH1)Nn3&SA10Y1udbY>X_+SNktSbWLwR}|nTV2T2 z^5)MekDj>e!~h!hgX+269GB#Zx$p|3-kCOKz0xKcT*Aa`bqHC_nh-IzlC;*UFdIcT zkT;_`(8RgCxma#J-CNzgyM<2iU;S3VAaUUEL!cz>Uk((M>*V1+NbbS7lJ zvKYX4*jKE4!yy+W|26(oe*P7O1ALydj|PgJ9E@r=j@GW{Aa&06wXi5$&zbNKAHBWP z#Eg1SFUrE_RR{8EmKeylb~mcu){w(Cch3$P7^I}u?JGUtxz&gA_hh$F@)~zL$l(S% zHd^ATVYEBWbbn#P4w)EaX69M8) zbJD+fQXnc}#t#vsef6I<{o)-l<(U zs(Jua2~PVyS4`4%LW6cm>9gSaAaT(m zG<|8T27k~yy$96TL{7#H#Riv?=WoGhwBw|PUF2uwVLZ}?VzqWYONqkmcQ321etjk4 zvMD}UJ4ks<3PvMLFHV9OUg`WxKPT(|_+o>;E%@?R3LG4arQe?(D3E=IKIAY}gA3_2 zQ5@2U`}gmkWx1O!lFqXD(0S0zB(}P4`!q0N6OgSvRvW-a)kyVN4sji9X2x*X+O5FlkITJsM?doV@2L>BkuHO4 zoqO*d%>u^dCqP~^MMv2cOATr1JQV}=hADMPiAo7xVmP(4XSJPXJh}=W0i8!azwb%I zsee4dFBBAmRoBwwd%>VBPbg1S(hVltE-i&syJEoB<^vAD;V>g?ucFbdrqQCvZM_cp zG#TJ~5O8WPMNNa~Xj=FUvNmhhF*VL1k0!x_%O@0;x*X=dgxkzJR$+MSx!jMcBB3TO zjlxiqv$pyEAxP$5E`2uSA zR~v#xhIchNNCL;eI~&;fR32CIE+4uhB%Id)R*8m0_hjhTqw3|>GXV#V*51po*nch~ zd{mHFcJAGNA85X3f#?{3h?Hw*n@fFEe1|@KWp{(=I^{t%3rW?dn-BX&5a`#mBI}e{ zZLt-;Q*;uVjJ7O;A9t;~=yT(b?;scQZvP>fgBLZ-{zSZ=eeAKHnO%VhvK~Kl?MzoN zE+MNViO3hws+cJ0Xa~S(h(q`TiGS&jl(*&94O2XDmD?A~CBIS4*@OC{xnhPcoLBA6 zAk5v{_`g^l-CA0!jTJXsX;9iS_m%_ucU>5Ds?IDoQ1=8#utrq16yzP?aG40e@o>0~ zx%Q_&U7h(#4xD|lTv7&L+gv)bMkeRDJ$;Zl!B9Dv=7E_Y?}MMmrqq$`A3V&wmx{deM5vy^u66Kr9ZPd~&P}am*Ff`lDZ@n$)@rWEm6Io>Bvf$VHei;7Zj!}irt6DTc5=v0A zAi)^RXDVwuch$VH!)6Uz9eZ*gF)&`YSBj;dr7HlyHMKHey|iqdRZtkg7N#Lsa0u@1 z?(QDk9fG?P90qsy;O_2Da0u@1?(Vki-L2iN+FMg!&C}`AJzX;o)93vECZ;P+>tRz^ zq#02d{hb(xoQ|j|&8#tryU-foA^J{>Ey)U{ftyTrKzD*#SR|fOX^b;Aj&U<;T4< zkNiVHXbAf2%H7a9Qbh*(1dJA3SVJgp_e3AI-#f+jfLHC90#YF8_SruYSGuZT#}P_wl8r^ zaC=KYM+KG4)5K+60UGeI|A%m{?tFI)GQz?6GY!<0R)umq1n~n#*n(h4?@g%LnahVM z!V|b9(sYQp4~3j*`iuY^GuNhK7UGNqp&i3*Om&2N`c)&FcIJ*1p*SsTz!YUXVvTMF zG%PB1Ml{K~s~N`BXNMo0sJMJ-hLh$Xt5xqLaLSLSdEX8&X5bseX|Gl z&2&NP3Ng1Pqj2U%$8JI4m>)g8d!^q8}!%tzBn z=aRboymMzWVo|6i=9e-JkZrfLgpafSd06`^;Dr@Yhme1o+GToKxcnO0*U8Y%3CR|I zzf^8UUI#EH2Y4K_o*c&E&zjbbB$4Q?*f%H&KdXxLWr%M6Qc$^oiCQ3yyi!31$sb2Q zh-(gY-qtG$+SmZ{%PEub8j13E3l+pCb?Q19^m`$2hlge!mycz{FLwuewU!)gy}u8V zd$Kgv7+Td1YuzleAIcaVpJeH;wUg%tCuH{4Q490_G4-A*Oj~B1<|0kBi_je$avif2 zfT8S{IWPvvg$)uGLqq&4+-{mKTg+mwRi0)(5~aqF9GeFAX1<#RcVzn7ZV2OJF~Z`LV11CbVW*NfyH+=JDDMYGFP8;_vd>g zxbz$s<9q&l4%Uw89P-M0k-xx^wRDzh$>+Q3Vgxn7mS3SW{dsJ_ahx4cQRRr+dL-6O zU`*z8Jxp|Ru6+$m8B|%-g={>Xrn}Dh>#m=BPFRJ!#oN3I0T*1uq1#hN4=f3UDjLs^ z+6Eypt>O0_WuK2cU~(B)pV_Z(M)?I6QSrunj@#swigOE0Q`q;>D<$u?;RWx&TTh)# zIRub1Q0j79t>1~JR2ny$5V(N*i8wj_wo6OCl~&pUhsS3v=OM0`A)f!Uqz_W%pXasR z4#CIJT*sTp^3mtie8izDw$&Vb^2 zfG#llmRfd>fl8eJlHMf$rfh#eqG~R?yn+!oJ_u4Qga4M}#9ws};l&k2T$}cc0~mKu z#h;#ZunRY|9>C+%g_p6Azi!|fm$TYJwglrzfUxa)(Z+szG(4J;2RnFOlDp1$C)32SpJ6}dqzxWYoLtjLHR$Hj0 zY_va_>DO_H9h|tet6L$Ck`v|I!?Ok>sU;5Z!-fT^M6qt^{>N)?Y>h{Si zT;daND}~}`h#!#e`fJQ4L^#X>GEhK5N*8lv317N&~n*92m zEz+P>hMW7Ttqv|s&lotUi|1omQAGJ>Wzs$Fa_U!pW2$Vav+G*-z_Kt6 zJwTn+$pLM(Ee+;JjkW1;4vg)`ZhXrB^r6XV(g-R1nun_eyN14Up+;n9SrDp-QrKW(l52V{Ikr-h7dK&&fZH_{4R z;oBJ^brCeYx%|AgYjm?exYFDUqJ(zY0%3J>Z#WpJ)}NOE(oNz&f$5F&SvBub#rF3| z7dCNA#tVT@pPxtTiQ|e~N4To*2}<~mpKKZwN9(B+E8chpnJc<=$7;7fyOSl^KTn^2 znIR>DDW@?Wa2#q67DU)b4ncbRLv%i)xFL>Qq|lGVT&|cz#hr!dWbBbU=ofuTUkr&D zBeqOR9?pp#>dEg#0q<-@-RD+jKR7Wia+P|fJf~KVltF~0LwPhmA^JZ#u z8?x!PHgAdyCVvO%xJC_NWX&w-A#@7eR6G4=pX<*$2W3E-mZRH1}_t}dIv#Ccz& z50pP7sbuTSr#xIGcHmC7>PHfwp^tlRuO&Tp$HQDWGpDmS!04Do0+ZKmEZSaKbPHfY&X0zT=G zNznKGn&HqP;bI8@b9*lCU~OvpBZ8?e2xo z**n`1z(Y^6!>!qqa&vhsa}xFl3orcjo(DhU`1#C@;mR=X_N`Gw{y|A{lheoMIXjD+ z^et8z!@iwxoGQ)Wbol6k^>I^aT|U9`X5gV)W)o-da3~_?pdKXIH7YP}5-}B2hFQq+ z(@#BewjAG?2+hz&%;|FN)>_jFYrpS<4NSxmFi*~K6BT}DK%3>OpE`85D|c|SlEJ!; z>yhjM$+INl4K%=w*m$4c+GFAqz;{A*ckglO!#o$PP=*_K{afz|zHs~J^F0q%-`;!JQr3HUe=*lwKG8~W0fDdOIwY` zCVgtK+ay(njF3*>ssFPQ3Yi+h%a7w;%P@d9UFw+|b~facXfeme?-TSMJm&ZkfHMdf z>~2H6X^)T<;Jmc`sX+d-0s_44<_+F(aiXTp-$pGMf2|wnIKS&1=`J+&FfFY1joH&? zRh2K(OfJnG;DwelHR+gEM={2C+pAxD5Dw+p+Pi_FBaX6;RNB?-|cEAt;)+-`RBh==cQ1gQSi1F+h{zQpap? zySl`;%A)E>nUpBYec;qR)hbmsCi(j=VuaQ)l>_Evp}tpp+^zk&E3=gk0C#JF4asaf zi|l^)SE2|!R3I*$!3^WYcSWv6y?5q0nu{<0TUfjx|C06A~*&IWxb1O!Ud83VB zwI?c{t}0w!lI<|Gj~rRRMQE34;_pBagCQ*}(`6B%&_mBHoIZ;s!d8Jl3p84TctLV_H}NUx0cyQf;#kBe)m16%yupj-2uY*2&vx0&RHT#T=LG2f zk$V|@ZT`luU~$a$OO0J498w_c=Px2>exT*%LQTihxk*S$tx2&p%2H-{Y4e3&svADA zVlf-zF|vW<*kUb839s3HPBfZ}YL27GV6BYT0j2iGeS{P+(~Vy=M1e@2r=4{=C6<{pahw(neVykq0`eUsGKfbWunO=23Y5qMxEmHoUcr~s1Kk!7L zAj|HcAe8?lKKv^l=l_Wh`R+i|{M~_;g6Kct6>AB0enqe#RYr&)JpUzJ=4*Y}DRsf( z82@K|Q^DdPQ^68nNPZ#p#m_IKzmWMt_6xZ$FZxd+l*MH(tCmtotCsQ?DqpBJw`!>^-TY^$#r?c8HPCM$$6Mb(ME}cB|7*@) zdjbW}8Ce@RIni0z7#dg`*czG8TiTncs{G%lY4O_)76cUh9ux%RKO29h;i{pmIh(B= zA3Ft2*vX+7yMcO4=078(tNk5RI1IB)1Ia|OmZL^l6txyZfmWlY9cBql7gbS3Z@?d( z>{9CcQQPaeb{Eif=!u!Swh3giu$Y-%NKO2q=^*g}Kj+u%(D0qk@4*Er@C~N|^+O8& z0GR~|#{?5KOnZ0xVM;#T18RTZ`u@yINX9)}{N)9B;aRP8BrFozP*+>4=XY-iNAb&~ z3zLSZOKGOq$i_s~-}_O||IxizA_d7iwMF465OyaU#F}TZx7*X?2|?~H>qrQ)^hF9;IbzRz2^swU@d1EE@5w< zvsAwKBJpBh?z(Y`POJ~$e`+g5ZWlQ!Z1sGX7%Fg#N9GUxu{kV-oReBg+-hK|75<$B z(gYL+X&ZtVcQJw^+#Yad*q~W}M!Xkp!RM2~V%J83B3$&{^T5jB+!LunF`)5*i;1~e z6mc_lyqUucEGD!Lc2+_{e|8Q3=#gI|@DhJ~HLl-%aj>H|jK4W3EF!Hf{tVfFT1?DH z+~5d-HvHq}2b55nsWkJ1DT9e>KyD-W+@%sCRg<5gFo;*gFai^y@{ch`^FxiCrFtc% zIpH;`aH~8Z26Tbm&r(L;$Zn1_vZ~FAa?$+Fu4gp1zn2qr_@7Fl9uW1S+c^x%1H!%CN)QLCLF+h9~>ajTfYQ+$K(+ zMA+%S`<|kA&)Y@O@MoiB!Kdf5={(aD`x1dtZ(cv8ZsE=w$*`!TdVYA%_@?hs#N!ZJ zr=%g1*OHOUPWhkR%MTV6K!iLFN|{pBNI*OeiqMA$UIE|irUIP%ry0?H__{Xc7lEzc zM?ZbT0Y`$;sfm?*q0Q?j@w&r;T)sjKY>8d5JEX$j7lw=nPE#Iyv~o5nr_F^31 zGvGjD121L=)ZRXpiMOHSuRd6LaG0P5(gQVEKNfHJk6pu{;=zBC9(%K#j67N>d6#Uc z0A&on8G%xvXwr}bzoS{aTKHp8?#KOW-RzshnXe;=)-Q$Jwal?}nkTA%(^LL$uRkr3 z-;L-bQTArD?nrgxT=)K{kg@&sZxFe4i?ua3^(4KI@24vZ-;nS6jO{^`7s-=o&fC2c za&~Q7o=kQ;aC~^z6K(vcX`Nr(sAaWHKH|XcC@PY{ zxZ2s7z2lvV{yT4U&6O;2Z(U)11|s?@eL1<_>Yn7=$$@`+qUWV{1 z8z6akA$X=|aT3-JVf`CKlHn{QUM1oFy>^p7XE0Jrn^Zgb)(K z?yn!={P>klpvJv2!$Dpefe#04HTdBMu+|~EH3XJhm>K>~_E^|kfw1fR~OuA#3IR4AO(SrQ1xK0*q|TMsK#~i@Ko*UpFoXtWN~a`*31c?oP>jgB{hb$dM980@5)Vk7 z#G#u*ye1%%hGd9OkYgmAj+`5_Y0B$S5X5B~BZ?*GiD3yldA@PzyX|3u1?q%mnlEPr~=Ld3kPfe<@AS$wp( zGXK0_emc$Era@k5MyXASU8zfHw-9(zaw>!=bdk)AwT$;5pO;n7!I{-P61u~DaeiU; z;Ob2FiP{jqG^J#g!K#Y17ym|*Qkz_x?vOk^=0-tBXG3^Si$^@3kVVp|=23kIaV#;D zZIWm_X3}9yK3p(Lo3u>sq;ya<#-pjJiK6Mcd|hWzCv8GwGC9nj_)a>MSPvu`jSLL$ z58K3+CtJmPlF6rhsg%WLlV;QDQv!k?XqO_djr~?#=B`L5env{~4(kri5dr04p5}pg;2ebsBBdqY1SWHQ< z$+Ee!RcZOH;5VjQ>d)QJdC#HDtCuX62uu;Ojj;){GPAK;Wy~K^?JK|k5IRT>Q_K&L;DrdcO zi}Mb1Qwo~Zs`aXs^`$rU1$4VrKFfTX{Yr%bdvpT$AQ9oPg5`p1gUjLNaa^#Ev7^}2 zEL2RKCsS?DjB)LBjJ-CV^217{5=XoeuoGOA*_4wlnki<;yNnd8IXRgNyC=JMj zq(InUZRx?~Xy%CF*aUw0*nH>jH~4NhTtpNlLM4(Jaubpa-)Tc+v~3V)plEVEtTfL& z{cv1k-SKyh*@wJqEWCc!Yc9u(He`F z=F|K4`p~=3*|?iH6k-|@0daSU%1BEr9_zD;+LR%S$Xy_chc}Cz-_dK}sY>6=NAp&C z75o+Y8O2tXH8Yj%dWSShN-7~*J~?|V%j*}6$brXGUI70sm!Eg0)-}NqYZ4c+z_Dt;VQJh4NtaNO`K!9<86~a ztX}dX2fx%En-;W|s$e?Bw)Lt9`oF_DA<&Z|o)s_`BUe zTYa0iPJ~y3PSGDw`%!hYj2lbrrxyn@22zqW7&O~OTUcF==S>hs*%^xI4;i`}du;|D zlEP`~)tjoYJIg#8H%|QLFBfI>oNA}4YFa`S6ZPim*zM1^e~X_cSHHIL`@E0CXW=|p z&#eG0_Pax}OzCOt?RFcw6}lZIPtwrUdvc}v7Qg>UX>OyRBuCQ8kCKc|+WlV9(#%z? zE!kOR|7c1+&s{yG!K0yZH9yx-3sNss=iB5mFzchg)D3afKBHgHZ!??LTT1VG9O$BY zwLM2%0$zM){nnwP5N&z;c~ummROPELD#Czj3ohOZ``tK23AsgtYCh+=o1-S9mmzjs zb*`>6@LkajXaijr-Nli_1ia!MxyGz4ua}QMwPv<+A#?hCIQ9>{T$@g26P_8>?OPt( zOJXSz67*cUWvyrStuNv4Nh00LpDssbee)+)(TfNioE&rP(dN9{%HLdbgaboU7c0zPe4Q&#vNu-2Eq# zO9^%S5WcR>e-_@)r1v9>O&!H`^03**{N>io76K1zkM4aD`g#qWGVaUkai3qR7S>)S z4<}nzkBjQuLoQ97$Q{UCx^Hst>rKF;%cZ<4ge%`$>|5@uQ4htv{Hg$tH~%Nu$!Pt> zSan0SRX6;H+A7M_EUgSWjPlaArYB0mlb~yjIiPzTl8^_%H|&LFkfHNEJGJIR4-4!m z+ukmQk{rr+{pbyU#!rr;8T27J{{i+nB^v~Z78dCOuBFmO;PKvar%OpxWe$K!1P3DI zzN-FmHnMI*T7WM9;c~n;FSM?|?4E&;K@m2~Xq+(0$OD6-&v_sI`3MRT9&x%NAkMU) zj+G!;OCV3nV9q@GE@;cO3cNgGoz*2FvV!h8AMK{pzTVt_lVw=hQMQceK{ycaA4@1) zG{Eicxhx&jKoc>*mM$PE#Q?tMQzBmuHjoPmMeiy1${o~yAH0IMV!oB?AjH*R9o7&? z9l~bK93WzsQ_tsMb?1<73_(3i5OBfFOiLyHpvv$8?%0?ev%Q2mhye@;nBdQb3WqU5 z4j_;nVVIHGQ9`Elke-J|_5cs$d0NhsthVX2`+@5 z3^(!2e(5qapoj)x9L0WWJO7jwm-5zrsTCEf5{;8qFDPJu8#g}O!JfJ=w*eQ*5eK0) ziIcawm&k(vA;r?ZrgXqylnEF0#6FZWq%SJsNq$n*jo?O5Pv+!kQ&LxVefZz^xTNx= zP#w)GLqjR^iqsv^L|{ce^+!T!WvNYmsYL?X30m4cLBbU%H#P>eyni05Y81ZgFGyb+ z!$4y4ilKi~+{M2PPhhe8pyP!LrKYiFV0?|h{Sv@~g!5+E*ro5%!c__=kF`A&{t#KG z^{|?xub~wx1!uj=l?JlAU@wcq9?W2s1Rz-fyEGldjQmU5)6|xv4Y)|ajzvSvamz~6pziLoKR;T2W%!Noq z=7%rhu5lE9?nHic1|EE_-FF&7K0-TpBm$M z)jjZm2zkre>>qS<@~b0mp9KW+?QiPZ-(3V$c!bJ4igFV8Mr)0*cl zyI4tSOXoTaZ=-AXEPAwl#+HwyH9uj7i`@7HIsb$DAGv88WWf919@hHdf7YFq5p?~}AJ>3=L| zTUnaRxWR$zfl+|KWB$tX!1X+Qf1T`eMHX8<>{yPqFVs`_R^}!z4qb8TNgHj|nWpjy z*las(WYYGpP((_S31s$ngMBP*Op%Ap&TgD7&=2(rmQe}KyE-~SuVR{xkJ%qisk=Yv z7-@Q5xoQ_&t!}G{IN19pA~_r*QqFAf7lape%hh%~m4g7sx8_~hS=jVodJo0g%0N0F zI$W>4!HI&K+E3%l>fN1_CTB&rZE$#KqzqEQnym{Ye@}Z?S63KGLaA*eQn#zE3v66} z&+Ur~t>&ADjrGfo^~1A`$BftC*_V4U6x|9|3uRPByhf31L%8KjsP_~!%uQZDKi&y? z-zJ47zU&!!MhKv{Ak4H5#*cP)B?NX@55!pyuKXPw4FWP+y_9b%npD>3XluTXcZCu_ zAwTeLMu=>M)tZ;W-aN})mbkiTPqY~erMtu6e*?4a2zI2AA3jVtQ=d4QV_=$cn;2)% zS{Z7A^DsVqr*ogPGH-)`Ridi5OW|YahKld&uLFsLr%pfV>@F*G3TQ9 zj?k4=My1wt?BUQNt#rmMEp$Q{`X7YqX+M5$otb9{#v&u7CYAJF*I>{5tFZStEU!{ zP+YqG(EPAj30n*aU2SsaT*t{56*u-4Hv^YSdHt47y*J9*V$0}*-Lt{p7@h<fJ zJqSpVOWuz|ygu&!sIeF&K!J&w9GxIFzS@BHXG<9jJdphx8?3*OHV!+*Y0SCO*#$ZT zwf}M5zp;ep8~$#+$>(bHkI>6EXUoyzX>Qmq;r~2`;o_NwYo_2Hzo|sL)pmG1= z5YhpCV1r%61IJuShP|vK(B8(cwUs%+Hc1%PSwxC!oeN z7GFn;)FMn^p2!rGXRE9xwA3a(N6z%9vM67WT??oEV3SZwPAWV?W7Gg zqqvinh@C4QDmTPQ0?SPsEUnbGBT##dY_qJ+1&Pk`iHUlvFIHq+(SY+V>vc1%N_K^< z5XRj53kzwWj)aR?hYQG{Z+^W%Zrv{ z4Cu_~qlwNE4NSF=>_G366?7?~3oeMoSJOvyGMiiePvRs?<8wJj3vbpU0v*@3Y96V}d=={4yS zUTHeIDl?e&wCg8vg`9ky2nb>Fv@?hl$n?#;xuUg!*4HC32@>m{U|?*GlNvqu`ZI7E z(5l0sh&Pb6N122DL)J1`mJzB^IJXKZ6(?SB=vupuUK&B)wC U-v88=od`6SxJ{y zm~IF7y8C6!`Zi)eaOGaNmjBtxl%OmrZYr z=Bg;NJtD#-HS;Xa_QTEch@K9Di1mreCqPimihko zUYhk4>}W8OV4ko1KbG6#8Q7+maFK+C4k0{f@L!KNvVr%7$kLH-R!o_L8*7e*`*%O? zNp*Hr=i00`$c-tIW&5+6SjwDr5v5w@B@NrF#JB6HNU-xDhZ<7n< z{0TJ3Ivi-!Y!A^W+rl@N>ebzIPl6oU$IhLtUtQRVpwXhh)ui-Il%av3Grz)Cg zr>qa3+JV?;%WTR|kK}@wg_6DWH|~EbK>CinIz;?Zp*82DboZVD1@G*?EPa3KPdAYt z%NSVu6;FprjimL5ql=`2J~ts`i!{&%&~rUF44l{s{muQj4g4EHQG$NSW)v+U8cLBC z5bw+Ja$=1eE;kB<%w{2PfzmtMa@`WhWSIo>Dium; zPKHUm%A=3P>yzI4UPdbDXSx$=Kq;tRKXWzTv7Ik7+{_*xR^V$z1~$3O~@ z;|8cMWHoI73S-ve%mr`H)4yoxdCy0c4da zuqI41pE=A_bV@^1?OahA0G=4^Kjr)v|3Y*LJo^7(nIz zW{WBc;8Kn+H6jU}(D|uM4g-^Z3Zpu}iCVJsX8z?**+_(H7M{aty8ip?snQ3Viy?w$SKK9EkL(*emz@s!>=}US{^q8n zOugA<$_xwP^~sL@n40WZ?9B`~57 zRX(1{SA{hfXzG3isV8k`B*RCT<(4xo3FiDxJx^5;?0+OqL1QubY>B8R4p4Q$S+;Lc zt=PmG*98eC%Y-DvfbCC!s z3;g8jiHQ73V)6%a{AUQBhs=o+A>@D!^@jnRD^$5bc(4YpKiAP)8MU(q#$=5DSQva>i#vD9cg! z664Ml(Ua_26QF_oUAP28=Vyy$-Gn|Y@NWrr>_g@|8wE&*D%4$}@SLUnlC_Roe>>V) zJ?MJ)VJycFMj?lkZDAHy!NLEm~LtLQ&Q--4^g@Bw!HhrS&=1`cF7 zm>qoll&%5ig6h))+j;#(e`m2Rj9l--e3cS?+%8o!BE*C5_p)4WvrS@8rLRrs{Gg9A z&>ttTA|tXdlsoWSC6B5UPp6K$`&}8jP;Prk0Se#q{IrWtu`)Kc1OC!bZY}*dO%0dp zA@dM$&R&WIl^_%(Pyt*~+eQs(%wsvS?|6T;Yr}p&R-%>Fy}BMr_K8JGt7x#<8&0*0 z0>%hL4R>aJBrZ)yM>z<8DestxU@ZBpo`f%^1e@NcLS##)J{0X4gKTZckyY7FA4H<$ z<|0ZW$a)jx=x#bA>1lK8HDiWG%C`KguXaohp0*j&(TCbm(FGngh!-C%sT)Gqjz`MA zx&;&(y2&6hPu)Xf?zLCC3^kdD5Uiz0{^AMi&c(4;hQhl`8F3qPc;{OJWG-3prarG1 zT7Zu&_TY5%7G6qCNS1}g3`hrg zZ^t9^F@uf#duL;t-s)2*Dn5T-UucIn=6u2UOv<4jp#e(Sm91+2xn@gBqa$yC7O7h`M_*37-Cx%*dq6q`iOqn|)xyqM6_Dg!>$onbSd zrh}mDw%*} zxJX`SwBl7+JShbBL-C&BoP!3iYAYG(8%!}AfcYPqIwWlW?2|!!PA3lAiR`^FjH~Jk zc~$yT)#XS`dk?*#w3%3Qb~W0{T8B4Sh=C326CLL>t#uqD(L{zeFQ)L-(x6*)K@Kd!QjWq4E|wq-I{~qO-L}mvdua zP9z#}yF%9GX=nQoZ}0DW{G~#eTxnWE(G_*#-yfGxi-)$$2+-)Z9yg?Fw7Y#qwSW~7 zCan*zvYJ#dPdn+~MNz$=JzesDpR@R(VZ5f)V9B1;s_$qe`Y!q0MhDR06XIh>S=t)L z2SIFMq3u1;AD3MRI+i?UWt((bTM){pEUn{ANSR97HP5*gQ8KJ8s%9*xqbF|+lie!ECRM4!D*QL#Do)|Ie~F!%1zZr1kJ$$IcgGnF zgK|OYqq#~U{IM|)m}XEXW&r%E#Ue)fUFRV0G)O-ByjPyhx(>&WbHiE0H3_Tu!FJ`A z5)`bhE{EviqUdDbtGvR?MWN(c9@t)g;GAF2x@x{0=aiRmhYnyIjg5wu#7=#hJZ!vE zjT-%J!i_^UGv|y9%XtYKVlizD4!$jw&6|@jiwKWr*`ecjX#Xl545%8!)&8iq#TG@| zt{P`rX2VmCc+ltF(s#w)cYEVp%T1`h#Z-~*?%C7My6oB zdUca}j-blu96-@|os4q2Kf16^=X#5OTUPWo(0{tFey#B)Ih5}fiT4*9mJHtxHl7dv zxGPY;CHM5Km?}{G4b=Mgn@Ditp+rRo-^?leW3q;lW+ZtPSn|9`-*ZdL-*25~%)a`u ziUo7Fr$sMZr!mMA=_EdU{yUgz&tM$xvzXbm%HCMUVsg4O0l`zeB{ZExS z!v|9V6a5KG(u~ws0ZNi0hz0NTMO%UHW!%#rO|`2K!4JF)7ob~PR+sztfx^FEzf#n! z$vv?kV#Bw|vgm`Qm}K~yAgb?eHKn1zw?DicL(_Izdc{VAG_}DKHaA3s+uWu%Ww811 zP!)Dm)FZUTug1HU_Ab;ogXtEsnfE{VZsv=lX(jN{>Q$4eA;yIK(s`fWyFYNVN*qb5 z?aVt=`*ibnmjKtNUd=z_%eHdKD;ZqV`9&s4u1Jy^1{>=Jutv<&&> z?jkSpBpV5ahfxX(t4W;xxapdbB~5e>+TSSfHu7aKYylMS>o*FW?*(G6SMWz&Vi}x1 zhYJM6TQ3#t6)dUIOw2DW0PYCOpCI2wHVxuPGY0e}x9o1cKENhsN}s=Lt$Ra3gmYA* zxK`{&BV2k_-c1@r43g8}>m*Ycg`burnZ3I&^lzOVfHI)%EczQXjm<655S5gwxJST4 zpx!>2Tmk0~GWmpSY%P&dZUMX^YK`ZQoeEQHy}Zu){? zMa;{*?ho!RaXf03Z-lklrO1_yu4f~io<&>Y&k75>LJeJnuN4oU+4(kEnE+ztNR_-=+Wzy%v8*8-R2Ntfz$i41;LFlj}jRTvWz z&$CrgX1ilnI5|}N_tzbEGzuMp20u%+7!pzs-QIQDFrn64FC8yxC^8OAx zq5w)W{)8O|Z5?ihl9?sDx{m8qwJ+wXIKR77+=}20unm0w9K&>M(+#mJoTI9Jsbj?| zy=8ji?44)?(>;;n=3J>0UedHsyCDg+3NS?f&}mZ?9SnDLalRWzF^ZTS7WNc4ROnox@k&3W^+mZ`=pr#J9|%3-f-)HWCkp7o@l5BT&}j)34K&AJv^)~hKHt{oXA)1 z+IrC#k`?<5CKP({wys83J7nL$U{;>>UN?MLuI#6tRG*Qg*6W;|$T3qqIUi_VBoXMX zog%_c;#ltRVwpxaEtVgy<_gALvG5qqZ(Hhe4Mjkv=SUu~XG-(lB2?A>Vb_lSb_0kO zO6JNBE?Y*Cu+tD+x)b1r5x}-p!?03vz(6^67b!$nxJB2Afw^*bw&zr6I|~MFeUN3N zyb6DV*;7{~uRM^IZfPLCWl1pc4-sngsSUvYhe&*c*c?vD%>4u{lO83Tv>$!7C=i0F zNk$1u`dEgK$PqIUV&@XCQB)0~MfGKm4YFE2Nb}4e)FZl3N8NLgbX^OIL>P5X2H7nFZ~fXKSq!QLb2uHmqD zyPd3c{H1=asg~XrxhbjLruw7@kA1f{nj&uOZ?w4@%A;muNPoQW@J!Lhg@VvdWv(FlUs(F2R~2ie zfgfiV)*;Lh?{8c*HYnV?)xf-rghOY$dUS6=b>1s7yE zS`OW=6|0*yZl!+Vct;jE1~fzidia3M(6Vc`%_gIj$z3D zJR=~JgxLEyOpY4_C9~RP;u8fV5Vd7Xv8xkuo2m{yctgOvY zf0tTdE^I?~!HPvU6X*g;Xt;WUV7D;elB*evGH9N8-L6?oQW85uD@@S(63}qvi9&6n zZDxvHv_LV3Vt%jnU#Jh(H+7y(yV+>(5`sV&-s6bt4c(27=@Fa*f3>*RFTr&vAglA_ zJHlF>EF&GEQ1K4oqV47eoE=4a4kx(hg@Vj`{DToE zr5h=J@iGdBE4vtPhrJX02X^H@$-R*&9rvuTYajjMQzyxu&O3zXR+M_`fo6JGkTaV6 zgc7ws4eGk$X`4f3pbv_RFwqd(1K#;^eP@Hm@ZKFrxS#o)`A*qS^#;Z@_^3-2J12ij zre>!SpO&q+fq;1m)&{yrwgKm-7yHkVr9mnT)1SS_bz+9?_^IP)msd2pM~%%jH(67i z)M_a0tFlfop1Y%|sdQ?&8tMz6I=BxBL!3KG7{{1WUCggYFF_Y57d7Em*;vbctq$ok ze#!L{g}iet*)YH3EW*T2TG|t2j{=ih2n<m+gZ8P%D@f&62rpl_JnCxH?KU;F#JaS(r--Nuw>zw$l(xsxV2>+e9`qY zGG%BduV7I4!}Vfk5co-^ z15jOt0EO73F7K{&ATNAJFUm0PHAR_r9M-?LjzAOIoMO&Uz_&`ezCR-(^>^^MAYZ}L zLnah2yTDKpv=Xq8ZoRFLeohc!k1Y0zQ0tXBoyX1qsY`nXAgPds6wAs~yOY_1qnKUF zC1d%GkN8&Yv&ytn*SQ1d7HPwsCjKcMopqZHu*?53>an?WA`>o`GTY-{_8=+s!%nQ9 z!ih(8U=y=NE~Bom(8VaLB}{&a6W$ob0> zU@hN;->mWL5c?JUnb5Bxd=F=bQ1r3Z`(LI_6M@BR5=3j`^Oa`pY=18|7*FmN7RCYO zLy+UjwXR=@OmBlICDSF+0D6; zLdDkRW<{SEImxh~QP{ezCML~5<;yoa?GE1kOX4VS;0FxuT!y`5NBN+h!6NJZ-)3`F zjLNVylMt2_!p4-B8ij`Miu>28u4h#<={5Qt-YRt*@8c>u?i1iTTNyuo!{6`uG>)lwJ8V*Jj5` zzHufKwf(w+=%`oVg-aG9ejTB&aqYa|9@EKl2@oRzyRzqPHM?Rnx_VjEpcK+|LJC|3 zR_K%6*$|KTv$~7a*{_Fb`f=n8YTt~ewT?P)#ZhClFqyVGcP1@7x5B^VfiQC!<(?bT zwG}9GE6paki#1*!ObD*4^8aoiS>4YMvyPl7Iz$WHy2rINV`9#L)!W)~=`HC!vmZjR zBJQ#g!Y16q&8mfKC8kk__0F6~rL5}3$QLPjepw3qX$6Jdq$2*^L<&f$N3gb=b{*H8R^RhMWB0>qJ%bQJLmfN zlC4zrDV#-?T=eim%7J}EEy5&+*!sm_9vMbvL-X$_*EwyvN>Vb97%&M7Ncce0eF@9y$7h zL!rKu8I2OE(661YdlbmS1L{%`x{zRlamDtM^?Wve>Xn0n_sD!or8mk7h;wv!Yr zj%nwxQLkHrzhBn+0LnJfnA^Dtz2tP1ThDT;HhumCML1swt>#;lW4_h1 zC{Vr`T9ylUULzO3(ZR!XB>BK=l)PT@Y(QWW=J+H8tu0zseXYi)K$5e6v?ls68G=`p z7+hP+r!0WdWPbHMPDaB<`KxKHTf}JkF)qcec$bukMF!8C9S4^e4$~L-n)!MaYdJ6?2l5ZQ^&&?hFbBIF)_WheZv-b9N64hOCD>!og<*F)X+@9;%SX`kJ_(BDLcgix$?bc(P0K% zK26&J(E(e3X5)mo@MFamf$4gmR@LZn^YM8lgZ<8Xw`wMcCJ~vwK>8soUJL zM#I}$GvXG7qv9klPd_{HpXLH(!h~L&8IZ6tHpboz+i0fRVvww_xq85eoh&VJq)C^? z*|mf$oO&bA?M%tx(|esUpR{05hw`5-5suZTksaQDw|28LQbXzyJvIuq+u-*KN_^P!!nM%uuEWvo9sx{u}TUt zFsA;V&)ZQhlB2**ZJmE4VKrJ9=Kdh^Oz7Ype8Uba!~$mrE{R!O3dVka{No6;+&Tz*j-71nY~748xRzlfrW#%( zlX*?mTxOlxY)Eq|Snqd76A0X&x!;6m+}Nd1N0A_0makF^27XJ`RS@yms7;>NDuY}T znC5O3mRj9}Wi?8kc!nG_rtQ$#1g*{4{3`hFP{W7^r;e~+)_Y+2_@@1aGy4)f}Zj{ZTKvXgI%_S6A$*1>xGSo3e>)>|( z4Yy0d2d?}2EFHB9r_d7Lo~DiqC#o`eM?}TrqJ`ozyi_zIQkjFUUN!4?^%QWZ2(X$kI@)ktE&dD_DQbu#1L|C z;bIBU%ikOv{2;dFoN3&xYgXLf;w{5_7JRb^nrc^-^Ly+X!b27HEo?mJjGNA{sU8fJ zal|#4K57WR95f+Lseyt*Hh>>@vylC`IeKpvWt~D55e`ydIjSr4nb011(v9^) zLNwGMBCj|`DwQvcb3yV`tC$|&V`hnE4bIbUp%5yT$8W@HD8`M2;-!Iq-CYXhhY4pA zA`4WKyOU&}%xSeIXRX-Iy_+m8xxyB6wFqx+k~G6>MI`NsiAs&2OUOVUE!@Zy2^sfU z0pxyd*z)DZNFP3Zpa7=PE<^|4%Uh12V@z7|(FUiu^B~G-@h%iNVeEtgHn7kQ(BSl% zU-2!ny9+k_@`s7p!_#yt{l}z%Zmqroiq`SI;OE>ED=`W%o4;(vV+~lzU z=Yp<+ZnV_?CTDfttg^*=nM#S^JbqCwM6Wd7q)tk>Ye{k`LJ2%^RG86LL)EzG~o1vwSsdtn;)- z9$Av~p0Bn75-b zUSr%^XdfIxZMQVoo7B6{x~i!o-R>k-yzQX}3Y@>Yia(ft0Zsjwl7<2f_cRDha^rD_ zGTPUrweR#vkr!I}${B*oikN2cSN9^H_*5yy1k9b~xp<`7qYvPMJU{qjw`WUaY=8F7 z?WWTk{NT_N;h*K!Z3Rq>l`t8$*iyeUq}X31@i%t5g$g9Qd$LA^v0|yIML?J>^wruN`)4 zHpo`-6guNhuW5ynead$-_#LpYImUGtW;HUJ`fV}XN(g&z(Atm9@7@~=eCWRm1FRtJ zoN0*BdOBZQiObuGh4DVyaD-X9L!iRxX~_qg`6c4R4i8&)GR?DW0DFHUYw-#zq>>G$G<_P=6_ zFqHX^2UPRd0bL@D-Cd-k@s9JRgr`tCD_rDJ{fKVI`1m{`ZxwG67GWJP!(0uW7$_iu z7q#+#bu~iDBOz=w7ESO2`L@;#=*|Ncl1x#;JP|Q~VeP4o$a)(~K)Ij+n@)VtKBE@7 z0Hda{ro2%%ra#}5qfW~QD6mHQ^71Jcxqtd2rbBx>4!A7KLJEV1>=;6b0FqigQS+z3 z)BRTmcoaXy37&g_eKQ$qJ{M;uz2~J3O}#>Y2zc;Tg#wR&?9{3 z5p8j;uGqe175CkqqcfB8DFNzQBwN(`PR7`$?K!iZ)EH&O61k*wbM4HvDXS>QukL1M5KB~39Y;vzb1I!4D>}_rJ*<#B*7n*=Ww|K<=Q)xi1BdF z-7pP~_p;rHu({OoUH$OMygw!?LS*)TEFG)(g6h9L&p2bKh%Cayqe!Thx5!f;@ zG|(jBDekZE*Glf)&2b6EsLR~zVlKY4C9MFLBpMD>#HsSe?NjmaRw$rEx*|8&@G;S3Jz(<+)_&DGeBnx^<$a#_&ewb!Vr4 znAh2UVrGLypfE*ZB^QmI`Zf*1{NexF;v$}Nm|nqYjQ-{RJC+*obpGU9Vs6Rp^^y}j zRw?4>yf*NWumCM^^K1Wq^S{C2?|*+7Tt-Sp82zu?+|Y47i;XxiAphroEqSuD(y>MU zyuDO-%fO3H_$vBI$KdR&8@J#zd?igf+9hYH-W%HkTvjftA7lj8-Enr*%urxRV3Fyh zI1Y>`LRyKV9ilOe>OWgkO^^M6Q(!_V;>gcZqaxAzpgJ-ljQ6^K(6;glsVRS7e>_cD zg?hNiwDH!(o|4rBf7$^{K%Cy{kF6|Z6aB5pnv>USmC8(TtK1x_ja;Ub3?!p66wPcD z%naRzQKR)lG&mB9W6R@3&HX&4I`2=G$b``9kpZ$Vm5QuEIcb>5r$abKQA6*?w=j3P*0fdbqG6- z{#Br*41(S4lpRYd@A1C+RytZuoZICvv3KlT?%9nR&Cx{Jj{*LZjwTATYwN5Gf1*!O z>0yAXl0S+tl7ZB6zcRg)?d$5chm($J@>mykMI>TO-YL0Hoi?>Yue-nBjB0Kl%}DV_ z;lFZUnCV7;nFUmDXb@O+=@w+x##Iw}R1Ef7F$7RM#e6fg7Jf3i^RvN)9VZh_XZW-1 zooQ=JHIM&19SBAFvKEnTqa+5HTM_9NXMro$Se*(RnQzn_j^;Dz(Z8A{L{&bs*}rkr z#lwY5;rxmSaTp^g{X|zR&SmHbGltvhMqx&ujr2Z$!AG=%qhZUtxHXyh{&Sj(Q^~L8 z#VmerZO=O|eWD}X;?n0Fk!`0W99S2D2kK5~x%Xi@;TW6n4$=0`R;d+UmS4g)qCp_i z$I_~M$&&+q#lu2jC(ZT)V{gbTp z2zz2-g6-}qDdaTFrnRX&ZJ(<-EM}JNPW=pjf97HrbjxC*9YV?)_uB=fR`ya4zpe!f zyT{E}pLr-rJD`Af#5#B`!=H9xdUgkY1F~kS!0#Y(j@yrpIJV6EI^tza+5*XHsBx}z zix}WsvfQB0+-nUJfciq)04*)QuhuzT@1{v=7L0p^w1zYMS((nPwXgA=v}oh~VGEmo zzd&3bjUR4F6}3nhm4Lrm+Ow^T`A1d(#?KI`s*g&(a$KI#7&u)cW4UDGR=je>-`DSR z9Ztf<=dhaUT~;c$(nVC*?*&OY@dM#IHAC*EcO17+vh_0pzPuV+PQ4)Bd`$+t2?%xG z?PRz}vCivU*C}@Fi)tg6ZW4p%={TByC|2KpD#y$8*6rF_@1%=(R=hlhp4h;rmY@yM zIKrw<43NOJ#+7+;CXDMxqADXNNKBiX__AY;^J0)0y9QyhWJLH!`8nyUsdv+(2NDqn z^&~i2Ulb`=eHn?v_$|lyz5tjr#I<)q*%lcegx_cQ3(>jaVtg~tZ8v>=(xH)m^l@RP z3+|#7PyX~xWkcG&Enbk&z0V`NJljYB-T(v#eMm_6~4go z#)sF20TBsY@$9;6e03Nbrt%0hx|T=tbJ}w~8XR?0!d*PwdIH}sCpm9|>Z)tWX}2U4 zeZ_P#0v>CNTv$8-_Lh<1Y|P31ac3JW4jED>-@U1Ao1KJzQDmRKK{Y3RZVy*S-qX}t zQeHLgqk&p|nmKvHU!8f-U)P1L0~qyX^=E^CH!m>&IxU)XGrX<<>(~3vh|$`7_lGq5 z$VJ;Xx;cr&JxD1Jr!}vwrNM}%`rG*R?_ZbH^C%zme8iwbmqVLeaCD{bb)>r=qa75C zg(`!5^fY(^{<>d6ak4Lqfu|uU)QOpa56z zYo94bcjND8t$2Q5L`7S{XX8)bQ8jkR`8(E-+nTlV$P4T4wD*FsFqdZIW3X+llpS^Si&55Ju2L{} zCin$^h7~+btKRf?nshAJMlLH#v+6_(oQD(fX{|cjOij(XqKRgeLO%^XmI%GC1J+8i z@D&`GFy4c}cu%2z%i%}mRu)^bAki-+_-gt!0j?>DO1N8|ds|CA(d1dRj4`c2NDY!> zT70h6NJaoE%rlA=?^gJz@B~$coTEQwddHW483{()3GP~$MV)Q+!BPxs_Kh&Y!nyM6 z;tUqn_hH(wqkPgG(bQdIqqdHVN=N$*yz(m86eE;bIR_L#9Vq56<6j}a-XLCo zP;EhKDmNy5nuOUl|P!R;?=*y%+w4U$X- z*49CNT-PfwT_J?HC6OhDkvQ>#r(2zqjmH91%#Y5qVv<5UDXYzX8F8T?dswqXzZlD0 zI}4{#p#^syW2rn#&ie4lc-oh~Q}P6A*|=~+5m;L6b~R7>*|=w$W~n4@`5m=?9Xd+i zMCQI#%oOxD%Fggd>(oV?wW0+IJ3|_Lz7s# z<{bBbd6;5+w**z_W>pE1MG*~u{vLO9^z)qPyDX-(2JcCV*-jqtGQ|Mp#?tDRnp=*( zb#IuvO#i}%%f_K%h<~I+a;90oub!6%E-!@WgD{?o0o*X0CS>R?uo$%a^+7dPsUO!_ zd_RtohOy|-JVqpRf5=oWh}y~_BbogI1*Wn;Q3jnAXt>AVmZn1OYn-v zih!}&wY3&aVP%qBK{bK$ccIA=6}C3apJs%^yd(}@j>6&I?d1Gi>VQx5V5eWK-JwWj zUGlv6p*`)+2#Hy>aX5E>q{NdCs!ISf%CbTZvRrVOrQwSd85w%VX3-es#qH|)FGhT^ zw=J$8k^Ex)z{1YQ9&QsRnvAA#o96SH+H^u_vx%t1FDzc^_N*eRR4O8?rDV0|;Ga-N z#rnhP1R@J?=W1-LL43yK;B0<6IeCyvkNuH{_UgTAAjS2 zGyY!0{^5Z~C!+W}N;M@p55}=R(a3M&4)VZl6=!nl^I8qY>;_SE|4ui52X9}95pN=W z(ijJN?Kxp(k=$jEphXg3r{7N6QbeIYzp(qJukAv_{-Mqvg?cRN%FhLEbVNtx>F!O? z`kY$443Tz^3bN{dc1*N%L6`vM!NX;X;l3mI_eTMK<2U{laGT?mUn+rrt1kNw4r3xS zeFCjvI-ZgLR1Kqe5K(J6?*+vQDOXzW5;8mvVg!0Go{v|5t8GjU5}Wa`r0YZ&%8Ebx zykBNf|3$)NvcuMJOOl`<5h+^BvaED{172L{5POSmvcLoq^dZioO=f@XZ0Y?y-9Xl> zg-K#3Cu1?$2tzQie0E)>QH3db%Zil5Mmqb6V7+Wk?I0<4ku&|KIJe4#z6{3;POWl) zWslKr(;S0;k=y~HhWgIh%%A^Oq-G+!Cbp1K4e`li5k=3|i{il_^jn7XAm*=^4#MaG z0o{N)k8S!czE!WKM_B>sH2OkK<*e_Uc^&;hYHFA>2?oY8z_h-_68mC`rh^z)|5UNgDnS*!LftoJEuwO_9^DrycM3 zuddMf8~@EQxhGH*wK%{`h}6W&d~e+V~L-jj>t^-Z4DvrabYmedpPp7pNhUow{52%!MlOfj*1W#Y}H;`;%MZ%?CVxy^XE z3D458Ho$8>-(3XcC%JRnYwGyWcm}aag?QtC^}S=8$3{o08!F(o?N>}Dtn7<&Yz7kQnI}oIOf@n z5Pk4gwa&E_)ryg?tF`tpR;p6jUyvl-LMSe@uNk_pWk29Ap$Ex3%T|mry4;r9ZZcVa zN&Ji^iS)KzDKB}qqW`3j^Kd=>JyN%--qqzE4MJH)?YxA1y!iWD@vfD3#uQ!+Cd~Za zYRyI8t8x-+n6XoB3&NnpJ}fxb;S>__i=7`>L#oCvm1a z?wqtG7Ldu(AM&w#JAFT%{S1KwHc`fZ?Hx?BGqvz;`PFKtSN=`sb3-J2Ww;iOt?8Iy z(huG;{WLGTW)`hY+*WfpwK;w)XwPZxOyZA9ntqT<=s#(Xid3kvI>y2QOO(AjTlx$KF33<7`)iJq@jomsFQe3la8q{Bv-GCh=%ThDqAjRp@3=8%-(*n9xb)p0J@#L zWZRwK7ddlV$Jfd~zk0|ku7&#PO9LNO0uEfZmm&^XNwEEX)cKOj4wn=T_p~(Yk-9&2 z5~$E^GSRir(p^h+8NSBr}D{|qk!Vhm#NO^K1LM86e<)t!3AwI#l(6tWo zeMWMZdnB-{Hrh4XCB!(&LLlJyz=3oSW0?1NN(tGpYq{;71|jt`R%~ILu_>PnZrvgw z&+G%90~F{P7usE@+N*3J6?!%L{KJug^3C|7arLeGtefAh-N1~0fB?&=KO4Uk6nL%) z1(wKf9vAK#(PD*IIuZ=c+|^Lb;*IIdin=3c&oOSz5|Z*dL$WkRC?4ls0}zzbRa`hd z;$5ktpi#+YO~T&i#=F}ZT93HYYwDvzCFqglXAthdwF@L85CyiL?ovwZzF!-t#yD@k zsfq(6aKa>AfvF~cULSn9WLHmfJ;{cCK?#Ur{My25I;F$L0}-@m@R{YV|ro(+h9JpP?Nt-!ZmX3?{5^5kxFgZv_ljsQnG`^ z_#%_{TJPz928;Xgt8!&tK+>3dx3$&0Y{OHrXI`QzD33pwW!INiHAnQ;hw?%JrW{^H zq##qnI!OvhTmb~qW(q$$8^%9KVgLUXC>%u=MGozXV^F}|m7L$PfqV)IOo)eM zR5gGstjGn%PKWrM??Lx(>b3qQ6&vPH<|=C+3?ssSi1Xj7=l|Vq{Xf`B{TC{2`W?ai z#y`U3ZRL)&%l|xr`3JkIh$fGnc;DIKG2nKH)mh9{G2p`CuV*@qS1KZE$A$I38xh8_ z)8b-~5d(f-{(brP<^Ovx*S_fL_#2a)^_pU(ZRDEu4$oAFOq6#l*a_Pg2dX1|;L zZub9w!S>rfNPhUe{q`IGN&IKF-~Qod@83Li?ilBIv}eDsxPQyvZS2@1P&3(*X~igw1#vsAq!d-Wv)80aji0@fV@N;tJ)q*T8Nny?9-} zx$}y3LO*E+PQdJyx2mP!NYRe7*6Sj?qLRK@EOf4GT)3j5g@N2@j-Lc`s3hS$McloA z784N;^ubewuf_`9+UR@jU%c%l7W zZPV_!a`{W2%g00uyITo_XO)++g3pY7NWfa90o}cxgI<(8hrQnJjW%8gs)X%+x z>*Xi@t#)LswAwp~__It#bl)ajJ4vW7uOf0zh&RVWYCPMjBc)F!!h7&UG3W<>6gWYN zxZiki9aco4KzeBK7Wkog{lXwJ_b2LM_sRD(l}YwJ!CbpA>^H}5cS83n9_|l<9N0Gp zps&So8lx5vT4dE=)C+bXN`~b4^$2NC`|}9t=(-a0h7^{S-F3;H=9Gx@oZ`>yx49Ds z$#wjwvsN_w;WP-jg&X6lt{h{3oK7Q$^RwF#KguilzN#H;%6a;neMBNjPWCc~lxL4V zhXNl082EeLDzz`|t^+px9Y_87>ghQ*Ap7jc*16o802<*C_g(!YL)T~0ss{O~^1_|E z1$_jH*-?MqJ}`)@7eqm3HdZ6R2B|oc>1C>%E`D`;_x_4Rtq3sLs(){P#reik_z5lD zl4B1P*i|ihv!yX3;{posxH%&)GMs?|9WDB)Ls{-J8@c1I50&-TT<-XZZ$SO(3s(<2 zP$1@75xMIt!_NyA1y}zQ%xMG!l-zCgUAA{M&8dj*6`~9Y;gV|gFMbtwEF=|`jgYCf zZm_OXvpDiIzg?J4$9k`SRv+E35Vt{MC(M$XBAmAt{je$>uKlvWVJmD%UU-l$IM>0h zk$nn-*(B@{!qm&Jc@@+_v5~PaX*@|lZvMR7J|k$4jyG(H%f`4j@u-|{g1=D(hv7y! zzXpFwd&fJ@;^C+8T83(F(Xz`Od5JtX8_*pq*{pAFs*WMHoq4l=b z_4byr4vr%{AtbD^6=;qn|IB_B*71yZkxe>J+yen{VCj|ycJ1D7%y$+eHu%Y&4M-R< zcIKn6U`$_EpIzL4(|xG6&b54|^y6T$!L2_z>g3`sADw=&+{2a0Yco#h$1TH zZpBy0=k!|^2-z969pO8t+D#Qw zi8RgcI=zV;wPlCffnv(aP@)j++t_8ON zlfy_4u{@tVml|ffS$2pPa=*GvM-%YU%Zx*Tv}_stB9N&=9OS2T!HXZ#Y3_nhU`<^N z*Al}T*BH$o5C{@=H@EeD7UhyO#?vh?F@_75R;WK@#BdktccJ)^qO2%!gI}CEPS`EQ zT3tEtVTCGxxJv{}VQ~He3RDOasi>3YSw?NU4Y=F7(+!Ye|AYeGvv_5%zHwcUUxZrm z57BV)A;6?49W#+WqHJ_^ya<24#ME-yB-Xyxmt|kO^uB&f*4RZ&b48p-?~24r5jh2~ zbGw>lmlFzD*EPoG*sLp+K0PXqkqf+lEcP}4qo>7VTOV8jtmJPNHj=f8t&azjk=7ZV64}IN!%6PeP7xM z4MIjn-S$~Q!|r#==306H3aFDvka!=f?2}f`TM2Ly{v4M5upGYf0pS4`?mN}pcClD4 zhNlAlX*bUwQvS7N&FcL_6BM}0Z^)Zo)R&Kc%36@iBNuG{NPS0xUxim>i$Ze;8%AG%6iH|FoPsly zso0aHz`pxd=-vvL&8R~sMh~-M+Q)q3iOX|8=a`Iy03IK+JtyY-%#h>EQO!nN){kv} zEKKa8PIbHQcH+WT$r;Jn`go}fH#~ZaOOaiP0#ikdpC|4}e6lGf&Bm7c-{9~FiuEQI@?U?W6n~=5))#AdU64iI>T!?&s@zKm63`L}T`sB&^(OWsx4&)?6th*wc#V4=L zDFP|%*M;%(imesu63H>qv#;0QI);ZxAOUadi+Yc z*{GsDDB-{z*{5V-oIj~Q?@rIZOq7M_VEX3F$uz&HiyaiD+3ki|yAhqq0?&&yERMCe z-&{qz?GxbXxX(Nfi=J$Fe^)aTR;2xrOM(I71$HcoxnzPj^(vp}5AH-g?7cR(lRQ|c zvX`*j=b%q{C57>w=;NDtVeZv`^K`7#GRBtW=cluKeiE^Fd=;s&HDZ9{WX;3NhD48- zCtTXqlRex+0zz~PHNMh(QYI}Ycb$mEg9-5EmE?bYB|&lE?t_5oAMerylt)UaTE;T# z(}-tzx?#q3yqMzcn8~=y^uq{S7LCSBaktG>sfRZu+rcaCYB5rAYH=@rQ%y|kC?LNS zEc`3uL8GXnImW;ff+%|E*2-tw_kaP>ECZcvcDPyge@RJjEMrJwD@ zUW)|-o$RVy2e2|O60>bcrFrt_0-DS*z1S1P#_A`Xip)0ik2Z(Fd)KvWkGHWr^&Q$9kxL8+zokNpd3VRL*GnZ!$ z2n_FX)w=VvWQfa*mun9`4aJiqEONufJIUOVcI$l07eU{*IF>wV8W+KFhwhuF)4$Qj zMI+v-=w(!}-S15PnTR-NVWhp1EB`pq_gI)T`@g}l;{P+C@{Eb!?ckz^OjuBrAKn7q ziylF2GG0-?VvxUoA>qn&JK3FlS*CDTCYL48_c<ANx8R$u`6rv-J)j2+1B&Om389c&20ljp%Qzb_)CTp8{VF7<4h2l}Dtq5`Yc^HX@?i{Zh${e_)Z``G zx=&_ismAnw@{QDWv&+34XLGH~au4a}5SeFDCG zDhrfQAd=|*U8VAy(aeCNIGnQMhM1V}|LmJ}B}y~)XiJ{)_ShQ}_1)YlVWVD0RYQTV3f6V{T$p?4_GTD);J zmX?nm*Y~nvyrky8*p`oJhwzuPmmq6~EBu^)S6EfEbxGkJsF@YcUucrNlLO4AwH2~0 z7=`a2x*{XQyU8QrE6(QGpx7WXozRtlQt3?dBMgw|= z<-Ska3w8h>o}-5+8>kk?~C(mo(fXs`PlGiwY zws$y8HD`EnwkNSdgFxNE#U^Hr)<*p$MiMWJ^vi_iSrKyIM6v1p^6W{RO`Jd}3VF7R zkYUyB^D-KY=e7{fmAtleOU@I|vi2dHXved3w1;Of3E>AFa-9|j7eFA|Gk;p?{58{pVG3 zpLn|UH-!m^_R2ZtvqPi}0*%~Dxd-{ozFXDXea@Yh+)Dr1|DpehllN-wwyy3(O4Xj8 z+$f5^f3}6ts=>c(45Z&IsQbaa^g_SfBH*0A>mCuE)@zno#n@LT7V%J2G zU~BJfGSY0!u*+uA>uR+ME#Kg`KcB-s7;y~ux=6=z)(RU8vvejHiZpf-DAUDtzlfei zuZr?%G#89DG!8R<9B=|xLpIN%vD#kG9Y1h_Qq1c1pi6_m*6xfRvM=tyOgiz=4r|XZ z7OxR!mndj-$EoJPgf8-+wH4-nX`}5JR8&{TNqv(m;^aEtB<0=`LM!mhw456Ctidt8 z7ooUTL}B6QHLb2Ow$G&@*4}Iz(lIcr;+J)!$}0MiwHvdoRkn3wBThl`LYA5~(1ZH= z>`G0ii6;^)F!JhqEKxkUt<06XJ#L|#@-=+`=H{HbT3x& zNBOu0yp&3=C%WPZ(K>d2FomAYe#YrG1rBja3S>523&lC>1T+hM-FgozH^X1Vzaf>V zg@wUtow#rjIo9|!lRKIdR0vof15qDhwcky0^1FFyYT8j6F!P`Z9IG={%CR2b%`6iS zx%NQ;y3`@LvQj7zAsP5|@%!BXVC)gW_qqZK2uy~#X%Zcbc4jkwi?HxblbDlMXivd9 z;f?ge?qr(vL>%=-$<^G}`&o=jz|`2-cK8>#zW+exwDHv+LoAr*@il(eu_6lI<<}PbF&bH&i&)|AmHM~g`7vM?9vaJkzvO!*>C3WEwHv6lz=InN- z!nd8Lg3*^;S+yU3O6S9+^GL=w+r4;vA0(D*S5F3PvK%`Jz#M2`=0DQKA}~D-{|*IoTME>bRBiW)icI1(o?6Sg^YpGk0R`B9ik%6;TuIO$Z?|fM0#T{m z4|O#sRttfH5Qvmn5NBmKqm!r5*in`@pCx1T@fd3tb5TsI;5-e&vrhe!M?!~1WYn!UfaIZTB4K(5SI(Fr&@>;fq^=z5{vgUQ)q?bhI1a^QM7lSpv*_g!f&e zyzjR)ATl9;^H!%yA@^z{L}@z4cl7d`Z$=NYPqRO62slU%t!pWYU{<~%c0mmpMgvy& z7`{!Krs}GxnM!A9=?j6#N-iHQJb&aOv!C2Ee<~v9b+;*%?;TUkfby} z`JOTAazzQ&$Uqwoe2RFSws|#-EW3EZ0h<~uh+{Mt&r zvboKajF<9dUju^4*;%)A@TF|oIB=_C2>xX<^AN(P3eQ4NjROaS6KbiWC6(?9tUJ=& zh8~*nON!rTWeE!V+xFPy@{sZjZI71d%*^WUJ$uy9Z_wXMK(Q_Sv}I06daY7}XtUj^ zHO=#XGT?k)*Aoa;wU&|bCh!SSo&Gw=s&JZ)5rMTJ%9tFt78H#g`{hWS?v5y>X<#i* z(e$V2>s4aEuX)9l0<0$rydKTY>t#szSw^$Kg6Pwjsv)^O{!HpGDVM zJDqYzE6v7yc&wfA8eXvWIDFF#7p&!1JKy+!YMcjmk=Kqhd`5AM+b=*NUut_X0xT}@ zr%mN93lpgZt)NTU*}RQgWflG*h5)ItNXKZv)Em4(iqcgrJAJ|Ze1d*nH*tp;lcq4b z!e=u-{{hbmwt~f3r=}n51f-)|`idZy7i#F$qtCzYct4o=v+%!t%Q~$mGzT_p*lv4& zh;>&Pw|y%AB@4XqmM%cvIcgtLWfn;PDU>bG#V&o)g@`QPMxz}SsZF_;pA3~OEeICm z9K5VLI#YjYZbWa6z;7pzkC3(HUV2|Y(L5BD(lhIJFClWHE{ar$Ke;lA+ab;TtOR9O z)4V;APin)H+@BJ=pB;`o_@J1vFlirubHme4N#_y)PrH3of~2OD4++G7ms^tUG9pI8 zQmFK+m=+44`lvEY&^xvv`o+d~6bDKtUYB$X&n#WOj=+`Gb?^p5K!mWk(2q*jf-MG@ zZFlDRm$)K3lARWFkl6*XwYhYcGoi83iv3Dbx=AH=#k=;L6?q|Epu3^QTOaj*`UwWJ zDy5fcTc^u~{_!UEy|exr?G)#*-kZrw^m@)N3HjuU?@IXW8pDPv07_{(mh}(N35Na6 z&SKBIUp0jP)%7dtm*KM}XiQT1LwJP3=>F{VVYqEW{9nV|U%37)p_3;QLc4hnwz^1D zcV7Jus~|cXe6`6@8TLn2Ief`~YbT&wy$QouT{PyUP;!`md1d|}JbbSW1>SJzwyaqx z{3`TMv{+pgIEYtMt^277MXyE6etJJ_xM}8gslojZZ2^8%!-ynj_o24rpvK=|@?k6Qx*_{E?2^iQZSv6446GZ4m~Nr54-uK3iAXERHgD4wE>(dGi^y?)5X!dG)rUYJNx7~NWO3BZVYG_ZEDD6drdVO) zWU*1?A6-1MPVq@r(&v>Nqks9Y-Zc6@2+mrQ z;3*6aOrcA6rPFSI;<9_ty^2LWy>`t-@VC=>A^y`Ep(!XZPM`gk_nrRlJBIN)clbB{ zA^tPD!~ZB#{dey0@5{e0|GxbHHh0*8)|zw0V*M@pIQP=A`dZXW-J#a>M>P5Uu8*EO z)on+C6Q%HF%KCP>F-=~Uz{I6fsxGgMg%-f)4eOQhpD zB9E&I_~ZI-Zopy+C~!ywt|97sWz@}oDtC)0c-HX`OT58}IX@%1?{FWTp4DXEV9eI8 z8MDdg8iJ0?t5Q=3X{-RFjBjz&oRC`u~eL?gKmG-M)xP}nSPoZ1UN@*St~gQM7`e@?)>xHO@8Ay{sM%d z@0PTG7H}4SDgE}3P6D7HbMV@qST-UA|=Zwg-5z3~pMlE|Hzx74aE6(i(-biQj1{{$K$j|Ge^xTYqnPqK@WbZ8?} z+RDOFNHJYY*&!)k$`)pme6QuQHbY&v=;KEhHAgd|aZ4=`>TEX2-ruMz=gjNP(gubZ0i%n#mr&rA zXrOMaM7y2y!WxG)bFIXIb;8FAG7W&tQiUK;U{?{|T68b5C^(CoxH0F3{Gp8FeV%r8 zHfMGo(_I6e3KZ}^0*UDARai5A%feHCxxc{saf37Nr=!d+z>_{I*EG7kIj2L`czls?JGSU-YOv6`C4@0;HX8Rb5x+()AU!iPn{Ls&21**#?DP=*IpEQ2<#$roaC6 z{OoW1#$WOGXQ}^TExzCQjeidRWWg~6Dfk$Md&sl2#5Cbx)0PHt?c2&58o00a2P1J0 zfBS-9zzp^Q>~n<-e?r&?g!TtxTj|KZ8~$$iyW#JKzZ?FqHNC7>)-f)1?hLul1Ldz zVy62(9}UNqEixh3NFSBtG9s+k@JYou!2%_b8m_nynu}(U|0b8S3qr^6f-=B#t+WxB zmS^3!m8iZj0sisVT|dF>tn758zdmw-PX4PPOKV93w6vBq2*Y`QD!~6sh;x2Pe={PC zbrH~LAE9L@`6GGGPk#Tt8y`F-+xM2&c9mixBDac#EJVz6`6x9fWJ7}Y*%@NQWRZo| zz??d0Ml03b+$wGd``h`D)oT=IF;6HEN;yC2Ft)IL`GZ|c?*&&gi(Rm&R~i-9`UUNW+-5?5!p+V!w~D9_SN6$)SN8S5mC%tmldwrLlh_ZUJxg zYZd8uOv6zVE#=2WqJ4urXBVu=Ue;UKd=AH%CIN&XFB2jpGaC2r81h?lv^`^!or9

!t;H%5Xxb5mmxkqn11twPFfU=jw6wpe@|&VY8pp@5-z3{ z7pPQW8A8mpc}I^kBgua05^5%nCUEs@^kyMN^SV~=+C1s5r3=;lc^{()N$;t~jMwM{^Cxsou}`iXQ+Re3lXNPm=?Cpzz@L^wj>SpqV0EMh@V2+&bD z+L8COKDU?s7s-701n6Bt1GMM}HUpKbY4JYoWf2wSJLc%(KmmSzC?JxKex^>ibS*Su zMnWcS!>~KWVkgEB+2&r!~w}nS}(dbi0T-}-m zn!0aBe|iL1g>EK>hcPu^%@J|cI1aJ}npQMte^8n0wkLZ*fuJ{#qB{iNfhZMH>1qcW za@rO#QG0%M=E9@o0xUW!#DDu`Ij6QqBGD+nI8?bTZpVUazU!sW8Ck58Kia=~HGctJ zzV44p1+%&Z|EJvqD{(tJ^xU$i<`%6h{$*xhf2+to0vGn3zu1b5RFj&ZoS1|3-p*eI zC=kE%)3bwIyTsHg6Hg>!?Tm2ps~?BMzUTEPKMDTJyLaGv(Bp(kB{7#z|0#nt^-mcr z79UZ1(abs4BS~~9;QJVRTeJCbz*7aw<5S6d-}v+*T{q;qQ9BCxZ|>Mvk5F#ZnBE_Z zf9#X>q@@r$kng1(v3Iy}gt>m3(OqCTJqGnuOE}Vk?NkODz zptsAuFbT}6wXE)7Ut*sY7iikP)z}|<>JrvQL8!E^Q}Wjy?Eb|vf1#t&_EO`jmk6}E2Yo5wKT=7 z2rOHm84|Ri>Pk*Bir|s7pR_O6o70I5?60aHR_2W=1HiBS<0wJ{y6Eyf_ytO%7@c3} zgks0l3z*APk1wly?U2u1voEnxr*qMNY^fRDrQ65TVGadM`QIU=AFWtL&Q=nEe;*e_ zye=cW@fDzt6-(<#Ea>E4&?=TLNk<_-2)E8XI;djIDM9KXGch0uXBCWO=qGeRNTpp!QzfSv8Rs(bZO*z}o% zk3>N2f$|=n=jTpR2A^0`#P8$lm-~-)k2%S;r1yJ8q-zcKQ zTdG~wkLdE`&?8c~7Ymf#HPOi_%Otg*iaJ}fJWs`bgSvlK@PGjYB9bC5e^E{g7T93_ zwFX@do#Y_reCLAj3s7~jY`iUsD5n2BQkfK{!>(r@k0?Ri@>9a_3~x#Y3iRX&?am8D z=09MW+#==)9>q{^33s1lAv0!Y>RRwI?d4F3bU6mL41@P2?2a+sf5hCt2HR;3P1so5 zf+)3sKV~O<#9%cl+AqaFzh*LCkgySKZRt*b!e>q@LKZ7feifpNY?7Si*}X8`B8*?* z8urSUt&i*JBR$=DK4Y>LEdw_}YR6Y#O1;n?iB+A0m~RRN+OMac;s&vng_`e%x0=;( zz!v(M5frEf^;Po2eylQA+-5!!G!9iyQ3$l z7t|DYLTJEt0bwN{MIOi!v!8Kkj*L~HOq*j6))f(AtCksu4*iYUMl$Z*E!X*>A|e{n4oaD-idAzmZUm01`@|MMgR zNyR)za!}zc>!=K~NWOo#>k0}q$JKD8z+*~-cxm4%NA(DGpi}|I3m)>?8vM6}T2fER z8JR&CYO9u0iPcef#aTXImC6+`lTEPFv^Lg0c6gzAX2=bfe4EQW5)h)>JM)#eeb=KC z3u)aqf8aIoQbb9y)RNv&AJzQIi8`Jfkoo~_-#!-#eDD^NG_oDVj(S8o#M516aHans z>P~(j^zfm|yUXAje_)yodsq<(ldrvEr}V!0!l>DtzDtU%3cctssALR#L-=kr-Ehvh z3~zp0HX>)1-IqU`JhIc7jLqc4cJ+bbCkO)#f2dih(PaPNvaTgYvHe~{OZ}ti1cTbf ziqYpVn7jtu1zSsKV>A<1s_pQX+{4%S@WJHO8uSec42J;@=vdz@yN$W7;Y?_pY0 zoC$L`t#<$RZ}#Ubl|$GH`H<{^@lFZ^8|%;`p~yUA?KCWFDdhW+I-LuQkKw5q4Z0`~ zBKV63>eMSO*XXk_w&nO^VEV^tciu2se{BTCnrnC%`M#N9JTRYW5+4ccwpel}t;d!g z`PE|fj}|WV>UZxMK)Ybj)=PgLIj@U#laFu8j2o;9@Rp)I5Fm;C1U{ij5eut1i}8YK znYJ{f86z#;wT82;?)nb}R9z%#YrRac7tcz3fAGnqk@z0M@Ovn*Q&fTs>aOa(ZNFAcdj?ZH2BF=2 zD6sX)pB)NR;jtU$OY)qwb>Hf;Tu+haZo8+xbZ=@*$Vc#ZOl!sRM{9C=bmXJ4#gdk@ zGjmiyVR4*)u4dp@H+0eoI9=Y@9)6^h>E{V~!9B<_^o*E0kiNv6p;ZNb_FPF3Ik|ly z@B77#;G82j9e#nNNp;8PCuk8qzAZde_a~?`R_G%;oT&2A*^O$0E<{$BrK*eB!Q)QM z>Wb?@fEo(;fan$`XP!#ce=&{qQylZ7wH?{tc_}Yimkw8E4tWM^v>K89GGL9SOtS*|8jlJuqR_(x)uP2H))7#SI_+d~bggc5n60RL zzhY?<@U%X6S_$hIh(!5ah92Cq&V6PhUOi3io}6YI444Q|JUY8aHhx=TAdNW z{QNRPq|OevQudQ%x`-3{fi~;g^;3C_QNXv{@kVX)GviFQe>T;!ik*<&Vq~%1E83UL zwh7NS$8d?-EHvjJzPAJl%yqymmQbvhK;^cpT|Ca2V_I&_vYtNuv#5mXy=doMUFzUN zLNn`Ew}q_D9g0u1NKbb5M z(wL5KiuM;V!qu8=oo3Ug$J>x5H3(_j3 zau=3*3FZhuPsX1pc;@N#TAEYG;PUeYL@T6!x7jN)e`q$-{upPXgJ+y7ME6m`5O00G z;ON0@`G5vKCz~-hr>$~egFdS2I55ktNWg5x+S?-$A0nHdOP0W)+)fADm1PnDqh(TqKotx~S5{H{t0vn>b(2#B6L-#`1#-f|7}KJ3MMi$Z}&9U;-X z0R%4`J)L<4&-)aEH;)^#KnD>bs+$j^qbfR$P{88DV^YaB;6~NF>WOuX7-b1|!neq8 zT^=N$H0dhmZ0&9wF7+7HL6o0!AUcFM2f2Qee{pX-gsFN|sndIQRkeoH!-aV1TUbQ~ ze*(FX;yEg?CRNoeHcgPQf)5t2n$3-GR|p^;3c}Q6U_G8; zMLloha7YeT-XkAZ5=U^=3hJxd=kbGlxO88-5pTvX?mPmQygFE*K=^p-bJ(OA->Y8_ zKbycWG)RQJ92+!65@%(p!AnB7;w`gWe@WUKM)=U$%Uu+#%n>{0^+>K>X^htf7Xm3C5qM|g%<9MJOevWl+lVg|fRm3vb2=0FS2yI!A#;F=pH3Cz$W= zA*1P$*|=Xaiqkqi8;ywJjXuv2n$I00KNx;-_Urlg|KKFe@IUtyeVGFre``T&E@_`& zHF=K&@5T)l*VNPCYzCipjz&<5?|7MrqwI#N<-A|Mwl%x;`}VWtN8KRXT$|Ljca?$K8Qt2E2YD2_ z26s66tzz_Cqbe__{&UhMf5zds^N?P$de(?AN&KfX?T{&%Sw<6)z$PpdSzr3sJ9=~$ z%>x(iy{n)&k-Hne*e!NN6#C`)Co-s6aYvXZVGX|cFe6@3HGsCAH8$Q)?06E^8a_j6 zJ(8D>&7DgAcz2Z2{c%E+f-6ntVyc|`MfJ9&Eor&$^f=8OT_-M}e~xaNyU@L2R_~!9 zwjt?m9QxQ%cLEo#-6OH z4(6K}!h`6{YqwU9R0e6R(F}4I7lVP#+?td|c7|B&uWK*&x_JOc@Y>#M*Lx1XS<9yw zDMf@NOD@--IQlV*e{^h~xTy}Uy-)FyMqvuanE3+7<5RkVQBSwYXOu|OY0!MTdG29=qnkedk;5vvBvqdniBcvkf9o@0cZZo&@esP_Tr9kxQsgTy{qyGvAUo% z5@gzja}xD*C1Db?j~NR^%M%A0C{s7d`x;qll5O>WbmQ z^?JmMr9uJI`SkghnAyG(Hoo+v+MViUA;{`)g=o!m8@;Px2AlyGuI$q^6>TFl>su6n%c0IX-*Jub4?>vgz z#m8F8fA`eplI!>q!oLc}7Kj0!6J05VmPDcPUQW-{2hqaj$U3~$l6e_N0Q_}Pz?3n!`3hL zRdwQ~oJLV5<6E#pW+`&e~9!a6Su&_>=cRgsYaY#=y^!jA&BE1zME?x~2DB^k0vu!qx|K ze-W#XD)KV|RU8D3bq8=;*Y4CUom~;pZHXmS!`(tx!HZ)`nPdBzk+!Cv;ZVu|RR}Fy z|LzTLVt{r7D`F2z_*aD7VQ|m0h1Z|`1`W*wxNL@YxQida22Nz;BaB7G#7g5>ls3!G zF+=aH;=X5DsdZ$TRd_017-J?exNgiI+R=ik#V#cpQS&Q2eg>hEF6Iu7VR)%QD(YGL~^0oT2dq z7%Y(b{9g~PX)#uPG|gSzY?}w3e@fN(!Djhjr9O$nq#}sGpwJ5StQF4;jObMV5X5j6 z;%y}SRUCCY*`;ky%+$A~^HYmCbD52?+tBW4!wqdfuZQ>Bx{BD6l5i?V|GFW)tlLSq z%&eU_q7W-Bjn@QmKTgEGLf!EeVYBS)%JkXQ1>Z?KVR%*Og|Op}oST<;e=2)FZq=Z6 z0nU!KJd7bev+tXWHG7_|F@{B=g)P$VNmRLe`NTh&d3z~S{jPitk7{E*|FY;<>mikj>vw3KoHpXNnl`uDm?F zivmOTXWbw%i7z*v)R_#3x!0aD(r$->8?Qs)1C*hC=yko_*)FFdcpa|@gVzlMX%ZKRf4{&iYyT8-PCBDe z+c${FbKk*bPgUcSh;}@;(Jp#=XImU9d9PLe2l(ODm#!mABHx35JWN z)-|(npuiCGoa(Ne;6m@32P^+i$7$oy5+&gSvjy0YqaP(yI?;5kadQ{g+mlp70kinc zH=N10DrbtwzTs)4!OKq@=#7JU5Xh2Mo)HN&juC&FS#}=pe{xKmg3VEid5-o>Su7WS z*u>o0c#BcUewtnHVvQMhmk?c@7@Skle4%?Jv8NQ;^lg*J_YYF*o|L9;`)y}kbka09 zU(K`IjcKD%e6x%Vg=oI-7iLs>>aj|Ftsz-0sljssqc}6}O4roI@1qUiwk(9Mq*!Lp zLL|-lrDz|xe|Us~ER}?Hayh;z&xtv81b^fc?&7(l^UTm(_8<>(tXy5l&Iov{EgUyY z)#yq09)UvuR}oH{mR3wx=#wnWx2%7Q+kIPkV|))m6ofgPhW~?H=8=ES4gLGgzTfyi z1G4W%q?P0}kVoV*$}dD~A++O4dx5JKxk?h{=wTTte_~`kjGSOsT<_=q6_+e6Eeb+$ z-@*h;1BvBfUoZg*xCLPUDe3ax6}|Wehdh3#ZvMu9691{x%`sOzwUGl|sXE16e%1Vc z%@y9|IlUj`DcE*)nk?cI7x?0(<}+$}B3h?xld#3T zd98?iCemR=^}Lj1B~7)kBX9<>#hd&gze3XJf8D!LMU?cBKdDQ{ZAO}`LOtynD$W0` z8CZV<&G=`dg&UKjOdR9z-f*(wDR@H=Otn!Q)EW%9K^g4h0wIfumdy#!;kO5N4NNck zl&~s#saJL0J_XC4=D53a=$#f!YJ<$=LupE%D#=DoQnGT15ekV&;eKYfe(h6$0Tz}L ze^ZmGrA#UMa<@IMoLdW?5$$iKr0FgK6#gYmJXmTKEV<%W?$8nh|0?4TYo}Jz5V-3e z?OgF*IE4LQy|(;#uIkBz(?4qLnATmmXVm@fS_;1)#-ODgNUb={2c* z2_0olO#*aY!nmr|A$bhKY@BSv?swm-e=n;qDGMbwNL50K474-8C3+OCp1!ipr>i6i zD?E7lPLwPlnC&bb{Y;s3l#=NMTITnB+I&quL=15c#83obB6cO!V!LmZ!lxwg4lI)B z?8>U|X$*%$!LI+(K{_7Fa9mc&H+ctEeAWqek-d+4P)J#XLvU4PLT0^tvkj!>;<~{4kv&ojecwW@nm?p3AIyBphIosy8Nh# zHb4L`{=UR?XYsPPph8Csp1!x$2)MH-eLC_-l&Z4NNVMsb0M9(DJigil�uRJtyW& zp7e}306eCB%3h$ z3-#0S=!iX)_Iep}t%%Tz#@Y>^Pkc+oeHwGbkz{LmpAmR+9D;nC#|`;z%#+8b{b3KS_#_p+;F=l-0%!0ekp7A6}1e|LL)s@B$D z>K03J)~gqM^Ki_-|K{-j3-Gjd))`ba+xQZg$j(qJ{zY`aJQ?v0p__&f->_db#{S$K zH%1IcHFBx$$M8$7zZym{_L12x(=W?nGCF2O9(HyZ)|V7ZzP?t=Sms95j4D#Dg>$lQ zZMT*8prO%n&G|vy5MRqLf9ghHZIW=g0~wOjx6V&vZ;0k9CgtC{gD!_3s)m%L+E}{e zY`h5C2yxZ0Lfkav3SdEU-@O36R~;2GAE}bNI!rXO(lhegJdrv^XFL2XP_es96xP-9 z{x<2pjOrLU39$sB!Y*{5TNzTMxdZV%wWQ+UMIc z|5|ni4o1TYIYRU4qJMK-=)lrhE<63rMm#%23(y%!l%_#|PAKL#(jaV08ozPJL%gRQ zhkf^Cb#OmZur_z(w=?>mfjT!ueT;Igwk4u3t7GcK$fQ`>-G^c*uq4!%_}Dk%9TZ4w zxpPsv5RiHBg967Re_Cfs4rQSi^%*Zr>o5bxEWW$gUmYoYe{ju$0uJ}|T;fn50`Vqo zQz4zmRFqbE_*hd*`&w=E4HP(OOYp-h(wV|Gw7!lk*d!E7iKy0=5r=*O1>#Pa z?XR9mk2Tz%-&eJm{QP*dE#Fk#m9$Q+@dXD8z~xnL-m3nbe`+etYl&y?kfR#-IaZ1P zLO3GWZ97dwzNRi&;X}5q!(WTB1y6mywNh@ua$f4Rwbs*pP&qK<*M`+#r zE@$b+YxVVs%SDBO5ln}CE;neouj7#)Ch240zND&uLxBPlYOe{68tMYB%zWGDh)n15 zZc+-Zd*3);e?ftpJe(Ra_Ig^h4tOs#S1B=4abu*+*xP}VlJQ9&<}ls^JaRXa>BU1d z->v9_8$Jzf`)ID1(4~u&%%;i~WCdd8`H*5gC$xrnuho5aUNoJS0ev}i6$ z;KW2-e*{FzUt@kw7ou{j)IiNpV?p|H^t(c}oIFyi zOE_drC#nV=5>YyA)xRV$Fc?i{;)vXCOcsNBrerVoTajT(Wc4ud$PqC9nE*O2BQNQXv^Mszqk=T+Uhf5Ji*!APXIV}u41?s1S|zV zeDFVv-cmGIQyJ+*pn<5{>bM7fI)4_x8k`=X{u2SSj18M04@t3M!GFy3es4?j?oHwS ziQ?U+dN@$Zl*c~wQ-<4ix2=OdKlceJe@T06ZaBrO)aH}CE$-N_@qaBT!g!-G8~p~H zj^8H@J)lo#Y|sg7t|xE#02-O1nXHg>ad4te`pq)ns9rrbStM00R@_zpU1i}6&)Wnd zhxsaMA8JF{_&ns7p|>{E+F8hAhKU}w2p!SdEY$Avhhn{>16rF#jh%xyj8}R=e@?+r zDUAKEChd)ZOhV?7-RxKaYR%nnxH8^^$WFWTqzXK*aR%+-6&vlqzk~zIWxo|c-3djz zGnx*BYCB7s^K`ppNn(*#@d@zAe!d4wEXOg)mIw$NJ8qJ8Z;?Gzt23)7E7ofZ9b5jo zFA|aG0HsUW%k)=!(T$4S&Zes{e^xV=P6SZqw8A`sbjP0!+h~P9v?2c`+yLMG@6b0+ z({#P;ilU?u^umqUJ0sFwgmIH-s!toyfk)awOZ-s089q{FFCBGkcKzjjeee;m86UafC$ zLprIbU!$Sa34Z`zYJ#VJ^V}M#!jqEd+tDy=9ocoaJAujvU0gVzNMTA;YZ=WJGJ`~$bYu{m=M3gx2xMx4_2 zfW*dNN!2DddMXnTQ|8jIe<-8bP}kh(9wZTEx>kR3tHk!3*>jNEg}&_Dm6q=mL_5`a z*|)$yg@yX&`cd7!&+2i4sjT7+&IXon;@eD-rWVz?-=nlZ$HUA_dFSI}_c>>|Hecay zyBty-gB*Exf+2q>ZVoh>28a_?R|Jpmi^CkUO2$rm(Yg=& zg?CO%Qmt*LHx1cGgbbOEVl_+~n!#!NZC`i48_tE77I=|^6qqEaa8`XzdGk=DwHq4n zapVzM>>ds!h33oYe~13e`<*ZhLA;j6sA%AeudQD#xb|X-%B|{ExT?wEWHv?+uqRu> zqO0=nzc`xkPZX9FkC*#kA3;F=dkV*^Fw*XE>Ym@jkD#c`G}Xo|>62L#;O$=a{FFmB zH>+*iPWTYZrbr#_hh_jW((9;#7Dh|`Ml?3RpbYhjiM+Z*fBfS2C|gn(*A~+is!*2C zYLB>rQe@|;><&>DeeAD!&P$1GtJ6@y6o>>u^CiuSM@EedeLFn&;!0mt)-TU%bnK$u z@F0as_n#L--}U6A54jG~Uu64ZAv>wzKZ?XX0c}q}4O*MyRaRl=JKfXTi~`LgFlMli!)d1mwu)kC@@o;|e})?Hx%cWeQ8MW~GYm1M$L`=* zGXC(;W7xZm5dBf@E61VG#8#j+zPtaoY`iSS=Ezj9Gu&8YZh%E{GOKu)@`feX%ZB0O zcXwETsI9H9V?@X2K~EaHS^vy#Z73u)Q%)%1b@Ro#w@PV`4+>WjOizx850Y{o4+ z77e$we?cGNZSn*^O*%nEh$P3fc)DfqciA8DGP^Y*G4gY5F`)2;(*<$MBb(nsS1qPE zWn}rKpz=MF?DtKJy2Y1y0aJC2x@m5W2tZWk_p;r$)67a^2-H^TfP^Ah-Mi?(!UEqX zAcM7bH-KIH_jvHwke`f^)q67>1eaQ#{f5Gue_r!`CtsfwCU?o_JSyGP>v=y@b%&AJ ze2-FG_SIa${Myjto3rOJ-P9(>*_izNDQ_$q8sW^oRq)}=hyQNZo#5>FKA@O#YUC>5 zRxWEJE9<58>*^$U)UyPXg3eT$i7Uf?s}#^;-w+rL(njqkmY|=ZnM}(2lbV>Sb|fPV ze+Y+2--mc@d6Pi4MNXpERfJg$y!1J+yqDnX?bgCZEYX4p39>$!*HA|BvXOR?Es*#n z=92z?<_0XCz7LT#X+o||j!l<2k;fgS%;ZE{G?PFriIXsn%@yu)x>0XTnqBItq=R%V)X+;fj#q1FtVa0D^vn-W3U4wLKLPDov-w_bEh$TODm;{80cL|thX7P*_p_{T z(PICQKsX#J*5Zs0Udt@bp+$?Q8mltNhj-(G@_;oYc}g#(ykbVf#vKjp2SNFPf0t%f zWJ45T&y+-cx>)RiLe>d)JMBmQydu}H{!kyelOR2Qu1)uN<=*QLF*@s32`Pi^aeZC7 zoLO1L$CEA)r#rH0xDFVg(@u&8&yl5+{!nuTe^iTrqBr$H7Shyu zwVx{M6vZi?2S`kfpD2quVE*)Sd^fWno(cn+P8N1<(Y$PuVpBINV2~~-c#}f!LU3;c zx3kpA%+4*C%{1!Ej|aLJU6X=oFUhor9y#s9fzec&WaW@5_Ye6sV44VxNL(dEIz;7x z)10-eBl3J15h|2?eOXJ3fUfQ9|CE-L_aeyV~rIDL*{Hj{FM9QJ0vHx`e!2dz(C>7G-j`HfN@P3% z0v)Q4ocZDa8_y&zP9IAN3lep;B!@YjgxUq{?g8RTC3xO&fwKR!#!F9N6b@*-0TCPu z>!gMws%H+(e{OD@B1Sz3Co;9=w-r5fA&qF~d`A)$ti@-jH^o4a$qW}!4eFX)Xy_%t z+@CEQB$5f`1QXn6=E z9nwuRf7(b9#iw3E2bXCSr+dO4A&IsoIJjr?{%$v#?nZ!UT4tkR7HA7(T1Cf$8@4n)HiRy+gt#B_{?P&; ze;+nRj^NF|jghPmzmMJxdV`?#2riuFfqNH%UCouKmX zmDX82wW}aDMApH|!5TZ}`x$)(F7`X4J(|TLBhf zv{J4t3rFx)#H$O)eaRMXK$zI`Fwurh1be}fJa7F3`8JbH*Fgc;f9VXRr3_+z`=-lV zohnbuf}cf@%}K{q2?6H?$1?)7+-y7zJHAPeGg2W3<^I!B*f7`qQF=-waAqB2e^oD^ zmvfQXc`P;VVnS(vLA5osaRU(Q@ zKR0Q=jAj!!4PR)Uuz$^;{jlzrd;A2T4O%XPpk?^9OkN1ROm5Yish zz<6$q8h6xijJ+>C__~kjKspD6f2-=cafHD<*-!+)1&TXMFD>px*p}fXyX9In#Cgp< z4|>dw%^Jol7MK3OqT6lrJah@3@*KXVaJ;tVYs9J88ModGxO*g<2@yP|13M=D&4!5A zj&S{Qw4EE-_jH%D^_P|=4f-Y0#%zi-`x>x!gxENcb}vz`O9L3rUI>MxfAn?Ri68xb z9hwYdA;611Fi3<0g;pP}j6#^GP_YeFuTz`Eap9uh1?P=1$TRYK@Y1+o`6FmRUBSff z>^`aq!7x5MOwEv+8N-!3=_ODFS+#Ml{iUNs>AMGbfjE(SK8+Dvzz8RJvRd&=Z7@yB zrXL~CeT}Y~62jl;HGu&re*l`>J4^u#B>)+T>0ghW_)qpoy=0|jV)q(tp!_ z`yX-qAM!tx|Czr1H?6b(p>O|?|DpWP^zFatE&UID`-l7wl`s9ToSwh`fahlr9EC+b!1pg4LVDge z;dmQ|Mezo1Y;D9VY?2bl0koF(axz6Mv8q(@KY{TnXj`}Yymo%#O($uf%9IN{|> z2Efrh#?hCj_`m4cnbzdeKt)x@_t%uE?}_{uO7VY*{HL}Rpjs{WURgGrRpgoYJKEm= z{(#7TZ(wj;e^%b|QMqoP<}9`HdN5w_XUVL5>26kUEf!FY84$z%et|b;6=NN@wBm0u zS{2B;NGLi+18OFgwR7hGY}1k#_xdV5WXN_;t zOsOrH`tcG36!Vf0T#z7*#bAv(p7X{s4M8z@#7^Zg`vBvg_HLesWu^^v)hQdaB5$~| z3HGm*khcqlZq9KG9tFvBVsh`o#&Qn>BhKnNIqWVLX0 ze?Dc9%NJhkLZ*-F5m}FQvLA~Vtm)-TH!Ibt5dHAR5L|MsJv7;MIZpJr)+jL&$+C7x z88j7|CYcXY566u+?1=@qgfeImoQngPk<6j9%c z{08kV9jZT*X)Luk^)aN3SCElojZj_${{~-Uj3bxNwiPUlTzWW<@_i=kHC@yPe=qOz z*}l%f?9s4y<|2!cl)u zv3yV-6hxBNYY6M`7Yxnp#~EWD0|!P4>xk%n>q(XSiQ1hw*lC={yrZL$JbTq{5uA4)$iKmxP6XPcJ%RcadB8|vbn!U?fc8~1}y z$lCTbjvgh_gm2eZ28(8eH(B$afcBNifH&q7`4f3gWW#dzzcj4|C@jbmZFPJz`FUES zlb#1u_`KtJU=Z#xJVkE2#eP855US?yUMn?K5SaOjCCIgNtAreXf1M>I&E@quuy3`{ z&aRt76Nhn}#nt|^YlStb255Ovr?*fqtg}y`V@ERdzQ?DrwV3jYr)Odx3zp}j8eFbN z1*G2jJGJv9R!K0=W1{o0QtdS5nH{+!3C?(QSCfyl#qoVj{!+~#NtBj6!5G;^Q~q$? zk~GtHRy6C$wi^m;e!z_Dmth@&#g9?>N&h`_$>8Z#SGqmbp@Nxjh z!Nn5ar81&|hS-Fb-Plx}(rm{j(_e#d$7Hi&RS)`-WIkXyx@EQ+KJafo0_>NTb+vkT z^m)V~PrxjE4aEsrvOwvp?e~jxW}O^R19{3

|UCuDp?6e>A5ger_S>w>xVEMMX#_3N&vF}}|B#BMt z01|bci&7qFNNB0`<%WG1+I`lSC!i#NubSFin1f{@VyB+r*}Z&TE)?rG3R0rNga%Db z_=bdyJWoK>fA?sV=+Wx33(wm94^=ZvWY51zd*WAc#gwwnsIhBsw42bYvA@V*uqfDM-+E;{X7XJLY2 zcdf?d*9M4p&)OYQwHLl|`(ewJmg)4qYARfm0N4f)e@xmZc^|9-D~k$J9m*c(u`^n( zv>>rj9r>2}%b)DZ+@Dg4UtOqbJlfLsQCA1gwigLjcMi_&ryCtd5a?F|ppl{w{(YgK0^QzyA z!RzgLyR~My%?FOeoDogj76#TC;xN1!d7Mg#gma%3QvDae$k}#knZK(@(yk z{Ydfz#4vSCbevPEGx={?H$MRt=ug1UaU|mC=O3D<#r1oeUu)|5;|q4s&rNSn9-h}W zwqb6vB7d3$X9kt(`wz@mJpp=(k0S~vhWA0(cl#ow$i|^0o79$Vw*UhM%xbY4eDENR zbYS&+so%4+_15m?JtgJm6wDFFJ(-pH%tMlRC}uN((YeGoE*MU_1jmVaM6aE>CM(dpd+D7bJfA4{~YMII7Q zHA`ZH7$lmkCX7zxYd*l=OmN+MbPBIB6z3to%Sp9ruemc4az9!!DUjfUd4Ub(y1GVL zE3ge;64`g<_^~xD(A<4UZO0Y(2{0>iv!rS~LHeu`QLG|$Yx)Gp7+L9F2g^34>-N9K zSbuz{7T}k#&?=cP${LcMQHWTwC3^b=gs7kBS-O^A6H4kAPSx2wB*7K*R~ z@y2qZrvJR&{2uSC$|Qpl3DkRfo8dcoA%9^mNEgD?^aGNS4QqKp`A1(=(5duRJAZKC zeaT@dSG1iI##Q&H;m2|bX_$-gRziTbiWV%%RWab2uX$9kpO^SC!e-vue@^{T;-}k# zcx9QNF0YB2(TBKqi@pr9J)8Bh=wzofryOuo@*El`goQ?7ZPE}elb%~?BW@Ohw14o7 zIuFgqk-N8UiWedgSl*X?3`9GcQG0!|KDvaNY@gv(t>L0NLIBrOMO+Q2tAdI8O6(Y5 z5k8RIqoYShoJ{juzY}hN>@iVH(RlgAr^`^~8~f6B;)Zb2w=^Q$QqN5G$H-%{F=hGp zb+23ZHJ952S+CVJK^aNu*-Ju`OMke|im=(TX6gv5EMJFH%Be@GB;SyqG|?R~zrA#7 z`r-dh;0Zt_B%ci6w5=-_dQ^Oi-;=fUvGx(qdwXwrluB^;v*;phPAfeMzy`9or)H*~ zx3oQl74uzmW?-JvUbk03Q9!BS31C8Y?gvi=iVz)p#!F|4K9Bh!h>n3h+J8_QHSI?0 zaIic*ueVt_Am{ITSq>#E@O_8ii0arP^)dKY|9LjLhw)1r+D5|Yk4uk-yk&RKT<_$m z^>j9MDVW9#usL0no`7id=-bUF;1<)RzT=dZ{Lk;d$B`qJDJNIe7&niU=9S~%AXzd~ zHEs3nU&iVj_ot@F=F5(-b$@%Y;-85V#(J<`j3`ek+g*&j2T|Y&X>{%ik6I!Wm*D@~ zky@pHZ~=vkMqBvRj26gdH>!(h1)lHH!I_WgO;wuR3CO{sm7a;P%zOp9c85=W-FF>n zlw|sKY2>MM@pE6s0=-7pIR-WjCT6qF_>O8TBYGs7j%aEWs% zySHzV%hT4EPg4--^j)2on_z1LM)X!;QcFG3z&)Q5ilRw5uyzMU zSPU$E?>by${nI{DClOb8PxWEqzC=evwZ6TfImz-hjVp@HNtFLU&i3u{q}IppF;n%^ zokzXKmWHPp=L}nDbY>>=t?}sBnd%5QmElrdd3!rUg{ISiz7^)c z^K7gi+JD+&bci@Hl5!fOhiaJ z!dRF&#%{04N34fElcl8vI#^<`xRLm+zl2l19Dn=>w<^%$FMtMVA9lF(|7rUtoAEIY zAmoI$a)q8}OaNG#ne>)#qtS?*(kxUB&UGP2Fg`!%wAqUAi zQCZvzrVU{FA}i!cQK|@URfLujkL3DG+Gs~9tY63|E>FN-Y2vVFwa0vC20LH$_$2&6 zCgr4Y+Mty;X?{>f_Jg(Th=)S8KfNa7Jbwx*dmtE;Rje!0EO1N@e0^z0uxdx5?ts}% zWOm0(RI7_!(kM3d?vES!Muav?k!z{SWZ%k+W#XS>XW3X9~L>fd3#CZ2a>{(mpU z=o)Ms*eFr$(T&KY?uBI!HfmMU9Z2ROl(c>|~ ziGYo^A;=oXt3J9B3)@&2&FnJYqL9L-5S$tCC;Zh*Re7fF5d^4t2FLJOY=1ogL?gP> zM9`&?-qr?WN~#EnmLRzPHK+*bo1~^fL+Q<~Yrmmx#0Lh0MtS4&m7Zt^Gbhp>IYN?t zm}Tg9acyc)5d!N0ECR|;0ELzCvf+-dGM}V%)Fyoiy!0wuc<+hiePG@b;N_p8a)SH2 z*{-~CbU|ATX-|X{2`~_%nSVq$Sf2k2p-v2vOnfyq`jh1+{;nvut@2HD-H&Dru7O` z5O3DJBHfT=?{%6`H8wp>#8Gkku)s>}dlJf;xkQWL?rYu!lFmuF{C^HC^;e3}m6p%f zeVz9E_@aI}{ZT#E&H;CsuEk4@J1JaKFluJI!ZY_#WYWL2V>IAx;|W%>9dq2FG^u-q z$n9)h>+#b28gP~C$6t5`eY2>HCJo}m?idkn13R4G)?%pelZLM-GlHG~W{3!Wt=g0Z zZ%2(N`u6jM3-Za-<$u>xui$uHf_mXM;w0mF9x50bE3iH9vsO$Ny?(7x3c7VFHm-A z;WzfZt#!(G(h^KUIyYm%xcis}{VH#m(*CTb!2LzNL zX-)CH|EMti{0#JU%Tz&>h>#j%;do##U}XvyKUrni(<~!pa)u5!lPImMb!I5Hj3do{ z#(ntp5`}!gxVaQv%8qfP{;R}zuFzvGT*yaC*>&Aw%7531Ki;E4BADzA_PD%VKmKzL z?~*EC>z+x~8ua=z&L*}Tl&56OGq?x;<)ZGO(`G>IXMM2m2`r3fwiK+$ZhpAaJJ%o> zn=HCcK3wph>zXyM_Ibz!PHih*c+Ru&MicZ;R@xklITW^Fyh?dSmac@NE&gwqTK_l7 z|BPMn|9_eV4%a=uxNMcInesDJnK3_5ji*ns&Q`2^QI`IL>)mIJUmkcoghx}^6+z*~ z3`@HCOSMU3lj01Ieu+btuRKP)cJ^?7`X&j#73?MtyTxLt$qC?v`b~OGE^jXhOxaPm zkY9bd!PD{8xO`lKmL23vn!QdH4QO?VR^4SNfqz}=gYGreb!30?b8t(B>$#Xqx|4=& zSzcQAunGP5r5uJ(f?v>HhY1l9->BRYZQX*5H=m=aO^mDV^03l3@PvZxypSHw^oxfR z`K#W?kGkKX?OA4Hae-tb(YCjF%S}_>#xxHSEX8;uNcq<~xRvq`+GWy8ybt14jw*+w z9)Ij+WBK&FYGCVkaI?-d0ZWk^9ikL&X^1|8r_92&GFSXGu~|J$!-36mPJSpG zUUhp-;%_nQRxw8k*K&O}JB6*y%u`kj$p<~tPu=^vxJoD-pv?S_Zj2uFM;z3T&^5hP zZ=%p4^t2_5JMC<)2}(P-P{Ofd@|}D)SAQiqz4B8%tJ|^5lFm6C0UxYU8S=t4eRbS` zHMi18G0hsVofZ9uGVI?S>}2!G^BNst*{MFVZlPSxHFU2f99euBa&x{Txd@Tg#<~_W zRpgp3AD~tJ`UQAYq=t()&2n=$zaSOb14-wqd@B6v@(U_=dO=+7Jf)P093}L=LVsvI z@ZR9_c8d&MfX1tqkAjP=V-}C?3k`)G4Z95P;UX?nI_;Ull*EJ@s}Y~NQ48mU67fTx z_aptbb=ELuKZkE_?wPQ)XB6*DhufCXciOcIjJA987|e0GV&=6PJKQ%Bu_4&3`s15& zTFQhj#oNU=iUHvEfuSPjI|=mnOYW~sG-@V9J)KqLw0~ewl^JG-I^AclYibi;2KCM%u14Wvb$_hs=i|bCqnI|C z`0)yOHpy|L<&7mdTVpY^(^BVScmRp0Xpu?0LH69&H^YSsv@rAkn1gOf=r@@in<*hZ zN?)??A%RV~p_sDWhl+km{@AdlFOLiNW$o_S`5pGV4Y-VNQ1b2=WMLUF1q;158J%Sw z3aQ1Loi89+Uc1fA#ee$UW!7|y>Y@sKm5%9y%FLyW5K}l!k|IR2v?<}q7Hhh|(SQGf z>X8Z?s$x167!};183uY#YHz#}Dq2v;Bb=fPbQF-H1Xu*PB7K2;XPcn;WK~b(XKKW;tnxGJKwZ$hS>9)#^76(4=dVji% zl@(kTxJz2ThVIgnWBlT5i}=dNVfwoVnP3CHnVuFQqkkpH`~)vyFT&)5krsA>jHviW;o8&*C{rs%aU1iErSU%n3Bgj#dDTKsKr&Yi1ONqtfl_#L=f@RSdpVR5@ zcO^-J=YPWwiwaO^i1hb$61ZL7kL&6>2lf%BL9WsKa!GrWDpG%n%;DVlpc=?zDKQl1 z1n~G8?<>ueh3kN1V(AbM4O7|xWl-%WQ3nB#9BVf*KZ;_agd%W>HwB9smmO9D5C<3g zOso1B-srpT`|xt1QXhZ&%UO4o+@CQc6rBdV{eQ*hcr(f(ByAzBkTs76&jr)dd}7qs zO`|;Jf-iU}C(AbyW!NeyB5=m?tzlmYEqlovcDQ5uk@!P4myl}LxgdS*gn{L4dc*8^ z3iQw`R`rO%m!T9E*z((_FBrbZjS?~$siu&L1F^esg?qCeRvuWDM(w6Ck!4xShSs|< z6@NC6x4&H%e%se)_WdMTkNyeB>T2de?83!IBxK>XTeSe?=fvPSBxVk|ipH530k;}R zb*=krCiW&DX_7X~u?`FC+&Gkw?PH=}r8tm;t0w}mhrQPDjd=OC!i}0>-*y4YE~C#x#PP}Z6ZL{*Fr>#@a_8^_icG37%ZSv`YAg~WaCX6JAW9W z(@s&_*jOK~tG;|JMQ(YBk`D{`1n*Awb<|E!4VDH#uSBCHu(v=(+SUzRN!9ZiLJnM< zsPNGV8xh-otR`ukqAmRLR(5`fBSgXJG_%9?3O>(HQKhOmi688d`oiD+xWZ8o^G_RS zIy0n^!UhcA)8515?R_Vqqtlu#^tOBYiLmc@_)NRr(%#S z?q%%FI2N@yIro}Yb3_Bty%R`RdyxG1vp97SK3O72S%iRZG|NfAQE%Z)uy`^13BXqn zmMMQ|bDy}eDjeB4|7Dm!UYYNDu7s=tT3mvd$2#7f!a210H$IHP{}=67-2fu&7Z z?~5khiISF9z-(!shZFwTan+N z|Ki!DGt_ODx)1)y404J>8O+kVg+(nER#!nGU39U_DD#wtj%vZutP>3U=hg@ZS!d5h zUt&8-o_AI(%dc~>0){A-g6yCwCXR>qKIhYJ#wlN&NSw~eo`63ahJQ=7&+HsXTm9WR z(S*@5!|5AX9GOg`P`CBqh3s>w1IIP(#95JC z+Np!Aphj9a>0hT~a(M`Z5+{_U(7eNIAXlitjK+?@6X0HAsXT}kkLX-10c=VABFUVo z6bD?IIbMqU7qx~BRDaUaOLbChL$sSe%@Lu-jrvHN4JT<+5iUVsQL=O6hrYy4&#SAs zS5RC(Z8X*^2`p~1fGnA&V$c%U8fX92v+bCcHxX_O&_`c}e3;=|x#`P1h_|QMF=9#m zj;nx=-cs40JsTuV?JK8~PN+r@z}g{Y$jb>p+y2DwAV4J6%YR`T+0?Eq7%#D3D)LI;de)?6_utBnZ1#qTUjzm>0CahH=Qb{xue(E z+!ofq{ek~e?%|rt6osZPZ0zeD<=QFRQer>bhT&D8rhg;tdaJclg=(J{Qw5^eYi=pW z&y8z9ap7=*FeNgU+#8Ko!HDA{pUfmg&f;1^*?9jTwSb*ZyolqXj&CI*QsyJONsL@` zyIJHNOehs13E5fy>e>y9Pc99Pa?6xIGn6ANBhcYaqli0X%yZ3r9g5yn78J#XAH)fz zkgXJM>3=hsShHBSyL#x@=57+O3NLZMeJ#=3YqV%M=xkjN-lP>TrXk z^XF)$NA*}I6UuT}4(hcnp|?!NzXbXsY~xsGlEL@4>Ne#HS1N&JwoGp#w}`d#h}A{+_3dCV z>|I1;&J$qvdP_6(39xjz-wUaz^tC}-SXz)fli&*amT2p`-P}xz5IX^_+?cI+$E*}g zZGW_holHYcE678lHs@$ym6~Rqznp*W?g|Q zoMQqG7wxt;7ne*DzD8E@#*o_*S==*Ed z)rSM)3d5o#8#nvN{+)1lob2SJQ7}%QbvIkip)r6LTLcy(ZS0m4yZc=EpPr-ma(^G^SElgYuG+*1MmI}yBl!}=6F`5Kfh}D(L(_!M z|0PL+E8UF-3@Y0LiHs)+N-8A&!oUrF0%rS3Tnnhii^}*j+(DzWplw%|%R~%#xPIV@ zYPN1W(I{lwp@k@k=Nvxf#e_Cjb8p?!=jPAW>pUK&&)H8SPDUA6TlEW2yMIdtq#sEz zqUUfs_bkikTE#v(a5#>Ve(J$)L9j%$n7dWyx!28aZP$K5dA=ypzw07GDvvwkYWji> zNos(qfFZT<#5@JDfCKxT@K>OJEshRuYC_GHlUQDcnQNvdD*IS+T3t}mplXVI@Ox4` zxf_aNrKkw8n#S9-O07L-FMn2hX?_L_3rKifk#&-{>BmFe_$xIxVNnqx4oz;G^$fOc z(UbbBG_ z>S5%XP!%fpD6uJWF&|ZK)1DbWo<6#DUk87fD&zRGw~N+(%opElFMqz)_S#*Fl9xc( za|)-eNfd~UpUlv!uNX>QPX8@};6*cTT+aNpHZ*?t3>77yR{;d9q(V+ zLD1TD=bb=^aDS(Q;2mCR=jPz`JaeJ@$D!~To^fD^zteIF3Lk@3z?>%aSfk+sKYwS>?;xMT^vW_4L!aGhM4WZo;H0{vew$7ek(ZJ}pb6AMU$=Xc{%%`c z0GZx>X2d9T@}0~nh9m(jTl-EC5tUFEpUYMdZU0*Yo<;f`PyG-5F!68nLlCe_TV#|; z{5{E&fRFnVKd1hs6~=d~81AXSlB6XJ6f- z0d?ZWx}INTn-bnvD5o7B5saOee~G0Oqy9`^G)NPG1PF!ibga=#a{u%Goz$BB}ER~3LC4#`7)z)jI_sp&5_8eG?Z%S}sCWq?i}mlMFfTf%1js(Qh< zm220LS2?qglp$q;^LknOb+dTZVCUKSY!^t~rYgp5whS-Wx2aP3mw!Y)A<^*>jpBZPv)UdJgKbX#B$}%YFv) zg8n#G9xux1w`57^%sPc54Qo3Z+5#&mXpCPJQjyZaL-9T@Q%+7IW_eM{L6wB;c_5l) zZVZCr=pZ3zn*_{mMDTY@G#f5cKlzzWeSb#DKJ*JJ*51E0+sr{-EX@q;V?T$&+?JZN zGzn{pnNd+Lb!JfY%bsT0!L6^Th^wqB_-EOe<3rxhofIlex;yqSuy3!0f6pn3AmR1u zc?P|i&7nVJnm}vbxNd^pXlmn4v+JT)x%WvNG!Wo+)`+|G_uPv3w#-+`|55-gg@0@9 zM4E*JQ(*-0sldz7mhD3|`sERzib3AAn~!Q36FEQ2+%klf{OM7;qgOu$U#mwqfFD}! z1mfO+8!;Ru>)_u_DCsSVTOsI;I)<@XGdJj>i++Osr2O%7(a~E8Y3RKO%x`uYzO-Uo z^I$7pWmT=OutDUCb=HhA=INgd;D7jqXzH3@CeO9bFOP=)wedsP%kbGC)5Z{*=)>kv zaxsw?6KbpN4NaLgmS0!k;2JMV8k%RbKGMXqZz~?s4rXIoGw^k*|D1Ma%G&;maLAA~ z%h?;O)DTMJrj_?_j1a_g4+)%~P-i@P-5xv^%6c^*K$|bx$1DlinvM#(uQ3A7Ar|DBZjihQP(v>gUjbHdl zCD>T+MO1OdBXDA6^`>tnG7grbkOeWCN;QURego(@sSA+r6qjB!Pc-uu`yY- ze)2Y+F^FZ$*^!_a{bt4i(-%B;du?sZ$Fmi#Bqvt}_X{NY>wG=0OdpS9Cd%Pp{Qb>m zE?>#>oPEP7yEiQSNp5@3Ey({eJz{-{$CiYE^_!MT_A?OP3C6-u#s z>!g5DekCO=(?6jqueJR+g2?<`%wpPKrhEK?L6EP@yxG>i(YFLydd(Y74hs}MTm>`E zSO09BT7MY%nZ#bwn18%oM&8CT__Rd@iA-x=n|Ia4Cg<=q_HI5}wObv-w(*}kcX))? zM2Id3lBiyPW;;T7a2>RZK4Z4J1{=QM@ zAY){wtbO(O3`2A#2Zi?d!H}WB1udgT+<>{9;-wYc1us7@s(%rqZ-I-mwwctECJ8;J zJSa#(akX%5f4X*yQk$|>!C(m8IR!Fx z;$xi#v_#M$lF28VCns~2^e5IHgu@FjO{*t7RTbejcIO=Fv;s7|Mi zxQ82r5<2H|a(^;S8$x#C53>jRvn%)&PjKJg?k)<1(tO;)ut3OMYEeS56IX9(t*`zz z+U)oZmbR23nTxIqzVp&x`W@1!G8|w9*CMXo+8WfgxVUIax$4M1Ta4+TXwzFI!9+#k z$b&M)0aqz*ZfJc;k|WH!H=ZDWH#+N1M@G{m34ii-5+Qc|JO5Ml|@D@=;xI>!SR#f!)++Qh$`{vANiI)B{2H!yG{`g z4Aucuaauq0Cx2!`l6dgmnAm(PBo=?1LK((lZJ0YQP4-B-{bV_1%D@uo$bb2$2In6B zE4jy8jh#Sy@DDkQ9{c9Jym2#1YgQ}F>|>0ioD#xr&QE5}HoiVZN=+u3D<#u#U@oWT z;D7H3WjZiCUvkB*ap{^V|!{>b8e#Z2@|s8D9cpqZ0I&y3ODkv-t550)I1G zRS2!qnnb~|Suz{vq;AwBZQzzc>W;S(j)eqdGsce}zV+0{Nxd)W>WLo*h`vXag#Y@h zhAQlHh)XA99)fPfP`Gv&fDu2BCc+oKv(f{S;Z-{5bktUZmK)ZCiN!uewo|nHY!3zn zFio0?qQma9Ui$>#d|>e<7Cx27wbL54B>#UZGpMcNdoq?Uw!sjpLu%g0qeTo(! z17y%Hi|jqcakNs43ahKh8yr^czw!?$%y8Ng50BYPN@0Q6Z5Hv)LrX&EaGU2PS}>O< zQE56W45DaAOn2Rx%7^Eq7v_3R@%iTw)oW+==zM2BtmZiLW$4mooMAi_kAG?ew)K20 z#|pG@IQX)InE}!edtQLtw>7#LajW{Fnj`d$*}FD1oY-qkUi|w==Oui zHrQB4#CZZ5MAymERsz+p5r23V)k6M26;DJMR528xSlKgmJ?kjdR2IQpZ2B{AZqA05 zp>sVAs*5Zibe-b-N_Ds%Ur!#sea?`;bOqC7Im{vY#F>Y-_@4MUfo)JBiN(rf#Uz2U zTA2g%0u`ido2UVfW1X6;7>5%#{+qcYJ0Jdyng?cjKcy5gXn7Z~8h`cK4;QsX94||# z(89*)(h6*2d&cKf>wjBB_(R-8XCGv?xV}vDdTePCb!;-n|AiIn@D?E!T;FP;!;VPg z9$m@M9w{ol1Po$pIbX-ieEt3Lcba#CUf}gO@&%sG?BVhyj$r8*F~}pH#@|Bxmz%K< z%JHzEzBHWCM0j3yM}Ml*t@O82PFS$lsu{uS_!=?Aq9$t_XB9ZBN*%|e`l$>@(%F&W z)k=?!eoq&YbJ^m%txis%pNyEnq#wv>ajv{~8e~MQVOV1_sb;&icFK+__O$v_Q6nL= zAEH)x@+#4RHE=Pm5V2$v;ZJkqY)lY6Pkxl=tfdwk%K+uiBY(H0znsc&R7o&THfA_| zWK_OI(GMR<%!Z;}$u-R)f~6r;#E8sV>$dVRbhXJWo zIzJCx=u|8YK7VkGY{uWBa6QdH6aklIxaHI9e5_Mq6 z+nZRIm3?1Pu<1Ge10H1<@Z26)f4pRf zLk~NN5*1_r|Y_&GZdXxHAJrH7RYIa ziam;6{eKeyvgbr1OAwkPZJ(d|G9$MFyKqbU#eQnR5A#c9o#$Jqu{V$?&(2;zI_C)( zsP=gR?1azZG+OB_tU*2sC#Ag3Y$5^xPm!x-o*||I)oYATzcE~{kgPn_$po2gidDC{ z5ltooSUG(x4?oE|SZ{2aiHK|jW-=7caDaJh7k^AnXWU=Y1t3qN5LBod-ie$b%3HG2 zmn4#Hye()s!JE0qEClQP9HE$BI@u45{*%qChh93ZNh^`X-H1E^KOLPO7>dVh@3dSS z6_d&)v(2dy(l77nTlrpbyew#f*Sr}#Aa6MBH3r!wfPaB z02)1jW;a@10wIdXvJ>O1@%;?|q*WsOd$jPn#N{Yz9Z_paz0MtOibK!H6)->UaesXR zo?*l-eXx-7gJnPnHy))dH@Dk0NC&STKYu9Wy{VqF8q>p4lxZsg8(4W4Ud|-=8M8dB zM=ct`ydwdqqS&Dbd?RgY>_%l1wY&(|>%qqn&6Q|Bd9e_jWiOoX!hZctX&TS&ywpSb z8KHgS1MAJ4q*tMAkyWXM6W0PueugAsv^~sfz3FKwPbs2Ed+TZSG7Ay#m|my%|9`E@ z&+b*T)Z_XCl96ZTv%)EXMw~pm{jP|t0Zox~%IJ@Bl5`cKq!u@8t z`dXY?HI$ZWtR4o)==FtF8-CVAEq^nhN7xZk3b*=NLklwJ3`k|4{fZ1X!C(%75j<;Zm@C z+Q|7Y@Xba(;0jLWcV>s^&B7Gzb%x@@)?=NC)YM^as=lmpKZs_-a<>$6U!1D;QnJ*V zO)8?~phcPhrBVj?T6yk+zTh|fa>Q|G=QPkHtN&KRTd%GEtDQO5kzE_1;|5XyN}aSB zQ(rLkY%2%*O{JF-T?Q!hogJyqjT zDA%mxk5rh{Dy)qj4RE5Qz3D*foi=Nn+lZP3jY>Tl$;DNSH^LsxkNydWwsE*U>=>9> z?_Dv0cm2I5ET(vZ*Ln&9Q=?AeCc4yG1n`}xc*Z>|tP3=faHdgtAF*n5tAw;9145eE z2~#FtF-z)~=&$O_T7MR&ulZnzfg z5)h+yj|C4R{j9Pg5)E>iSATGy^tfYMYW4G2Fm#IWAaNy|&qmV({nVeAMQV!}A&gi{ z;mTQcyQe#5bwG{t-}#0;5JTHM!|aAGZC7@%C1Lu9yo=0>Fnx*lrtP0uNC|&qApzPy zUln#wpid$SDSs`+!=|21Eqo_=m7X)B6_q-+qRN$2DIX7I?*ei#51Aer6Rl{ zgsOO&KPx{lm7d6bYfiz^Pk;-D_G|lT+))VlJ^ z+yIj&%6gQDdRSjp7A`^=0`J;0zu&4anGbDCQ-4gCAw$Yxko^gGkvXG(0bVd+)tQ&0 z5x^WiEx!G8P7WOC6St-x30EAlv^$51WB=4~Kb}Irx<363BPYM)X9|$kJ0(~`+Sg2C zZMZPMm}LWoobF8BY6y5Q&3_qasY)qCK%~Ei-C{NyGlSLk)ifzLnJh_!Mdij6M4OBV z0)Gp@hL{?k1V0Lfq@f4kl1#U0Q^S3KJbz4p*Q2c;|mmH#Ns!`impA zAdTAccXf+Bg(>sWK1^%=UDbGjzx6ICEx-KU4mZ5BWQc3^4hQx)rJ#(yLMuj9kmQaC zPb}`Il@Q}Swndwhek(f*YJZlu(wdff-hWCcR(kzkQ!Cbh!00q=iI9X6u*~8bhqk7b)*W9dz`fmmK3Tf zWM+Mm*|@-x^X%>-Xz$gD?+?4dCGu-Di~p!FzO8uC+C8pUFwf5w;VX{h$hF41T&BJb z68EH8XUQ{o(?s`>{(M0H1pEGz)ql#xamQ~}&N_VW^3W{i+k!Xq?tj3r8fd9?Qnv6$ zY7%u@r3Ygb!`W%FaJF{;TjkrsoW_JQ-&KF&B-W2z&|@4Bjr{G|CzLsp>+9fExM!jH z^V_I2zHN0>s|R_^j|*R!D}P%)yS!LdDwp4l>bP)*I&rdTc3hKkMa`~fC}QB~*=83T zd%MR0B`;lfC^UR{_WY*OPPu&GY}E(?rgkqWEnkr)k4a+tmTnUJ8D&l=0S+a^ridH6@kUhkuh_V_jY2R=Kp- z&7l}LKgK*7oL3k(wL6owtZyZOW51DTs7{p+Hhksw3C59IrnzFGQQr91z7;`=CJ zy!AU2!>%lZSI1vz9zCn>zruejSGiKEO#+RMs+qqh%p(9yVgwaGs_248+h1{QbG`6V z;4(Vg^HD+|et#5azB6X9k%oC8{mKVt#ZCJBy0o$EHwnQewfv6V1Uws*KM&kURfd*0 zAdJ*rRK$2+mMC0z2w)iKUz!`u-3o>rbeMhJZge@j5vd;Ac%W;>tlPxD8gVatr*`-Y zO_3LTKwqBFK!gzwRX7jeCP{6pz| zhU=TBne3sqJK<7`D8}E1yqz@04J`3hy&O4v^F~^|Ox`FK&~Flj7W#XTFF--d3x0Pp zEPqnx@Bw`ETdm@EHwPS-JW*bLev24Rw{xmmI?&Hoj{SmJw|C~tnU5>k*_fJj4RMPg zHf-9+79ZoLj%S6*4ld;SH8?kLs+iWdp7&w4X1YPFXBd=Odbi4JY1&YV#_%<5#Z*VB zJ_){PYd{hG3FvQ|d-%kCiub~*{m-{nZ!Pci9~Eutoq#~ zMlR^RDtc#A%7yZhjs#RCa1RJ6BK$)DvbT}uJ_Lq`t$)81 z$5%C#Nt}9ucoOI<0W6@xLp#bQc`7fQhC;5alkP^6)R!SY8_T5Y2{<`&`xD;pL)4G+ z%M2m-pWz=riXQOIJONmu9MAQh#pb?sr*iBkAY~wzvXhk2Hje{IhDxK-m!1`qqLTi4 zOn~$6Q7*5&OidTFEv0=g=>)ko&VTzi?RPXkdQ=*8jv$Cy*?T-58&%(`KX^~>DLZ(X zwrth;{_>S1UlSqXtGo|>*RlWY3E1V7{IjKxXn67a3D6m7ZlA7`rL+4{6NephWPwt3 zSHi-kC^J8?<~FewWJU`{{X?EtLQ+KcA0QLHO1qiX=s4xn_n$+00v4>9(0^pxgzaNp zY&S6G6R_h+y=O{2-x|JH998~ohYDAH8h?6>Ed!M2^);l9TpVn;ynod5dVJ0%oG-in z&GinoT`|eVU8Rig0s)Vt+Q9NQp<(v8_Hs_(59R!#SIpzhP8gbYH)7HyI52jTfEue) z32t5Zyp(}-(d#G8qX0vG(tqSq8EDD~#}3>;0vFyl2hERVoLuHIif2Y+4-m>I&GBy( zInT!O?;6#Wj38e#M}Va6lGeX=5KaAc{F_}gKWs-LvY`Qk#h-pxum~aPxK;K|YidLm z$5hj>57zXOGiU8IV;zCS92+W_yywr6V>$KQeq?dFE(l|m&}Oa((|^ipJj(p#h9WU1 z4ko1X>ct;_Dm6r2T<1OtmUry0`NvjdHm%_N;<&tAN3>wzi^H`*K$&c&2N=Fj_-`_o z__uf5ak^d%9JWkc*rc?NP#-Vhz0M{4MXp#5SnZu!J+EcA7IXRW+h$NWX5*}k$wBqb zJ#{xpl^z3!`dckCRe#}WIqCnn!b#*6X94qCg87~b)qSbIVngWcDD4lWY+A@lc+U*e zPJDOlFB02H#4tNxC&K-*A`19Gi%p<>Uxbiyd~f68=2a}iPM459b{q(~??C&~n^f(k zq#uBA6Zq_cU8@|kR z>nrc(tqCbDf^09l(E{jajU;u7n?e~r3HE57ozBAQ9fQ>znFeda@CKfb?$mZ`I31LA z2h4Ro=I=1BE)3mo%QU?FwCu-ccjIz#p8ZUKb*CcpV}`y!!)vh!(pDVhYu!gdd*S`H zj+pXtVQ7b$YLVSpv{HXt8k*frwFn|Z#93%ea&&2g$s1L%#VF0dZqqZ*1*hUM-9=dG`YS(Vy1|TTe8H_c}4Co zt@F%w858MMsjTVI-jn0D_C|N|(*qQDvi4d}Kr=54SVb`zs?;zc7g|wAU=JNvs6|b~ z{m388*CVt)Kd_I9W+y}u{K>D$oNvvV-`jgWTcgMr_ndza-%%L?eIzI^|MwoYWS5x( zy|}fJXnoRrS)e+ua0(Yr`hpyYA%ex-O4L4uFmYJ@RRA=YfP!CFY{_ulhz_R}g z;r_hXXquD%uJLiqe)e+s3Zxt)nT;ou?^3+5C@ENL;9JkKo4Y+$4yY0pM3%^s;K$ZF z=fZ!Z00(~`h`V0&?qNEu!{ODn<;5b|rkLrsh#WA~!aQ|v)+wgq?Ow;g)F-W$10#i$ z49Q(<|Cgd1x8BsKlPrk}xE2xQG6K?wQ=rZ&D&CjHe-)h4B3K+Gt(e~th;IFHLdI41 zv^@}A><_UCzI=1CdPkuYsVfSm8AM!X1Z$x@o!Ng_;Z06OmY4W0=(L$ui}#O8Ok>ke zfXp7lx2f&M@efHZAlY|%KL-g$k7NK_d~D&bKN7%o4JkN?+(=c})PCgo?lb-GnIasMdwsMY7=pFV%{;|ZWLd}x{0;o#oTYH4f>`w@%ZFbodx zCc&b+4cFJ&e^%0sphy@!Y?L7h3tG38m;Uo(msz*z(aK}C#?|>VYmK!wHvEKnVFeSX zc5=prvqo7;)eX%-lsa76a!kLdNFuzaovp8uW9sTM5{jwlR33zU(!B~ z`J1r+=DESAU)cc#a4=C%{g~Qro~*9H*u;2P6~tD|ckw36#mjF zYl*iMaP+*VzpQ=&UPm%NTs|gR_Z_LWwVn~Zbg(RqkP>nip8PRqclc>v{{1XF^TPKe z@jcOg#i0N`KKlH`iPnaf-&09W>iV$p>|;hvx>ip1o&E+pM2mI%UDkDZ@(F(p8RPh! z8Yc=km*d^ue6F7qaFK7KR!;188qsSCGCz5fw-;fe_5BsX774dsLXYA)DrLM>ns;=p z2Xi%9!miG;KHA$3iVq+4Z+H#_>Fvz9lVuyD-Z1vE0EEJcW{Xx=_LL%{>9;uwY6$=Pgw3!o84}Nt~0%VZuWN zxu|Fk;I-31FZI{+>Ix z4WnZURA#jV`#aHa^PNPSi#Du%laJG+4l)S|r92$pSCVW*`G)8oFS7m0 zr*_Ge_>~q)Am~dA-w}V(u`#>x*iL+dMgxPM71+11balN1leOYJt(fsmee6c6-5_N| z1#y)1=D6G^d5ksU;pay6LP~!2?2`eM-yS8J{P!3AWf=W8&D)=V3ak61{HSMNzUHOp zdYNM&_YY_CjL9&rrhF3jVW&b}c+Oz5#e+V6o{i}PZe+Ae-S&UOlic=!)4CN*pBi#Y zJGonsJe<4x3i|0rSNV=p`SKfYeqDnwB|Ql%=@>z;?XPk@S+)}z?7j4^F#Tf=e%)He4 zKl}Jj7X99TfgykLr7B2$+=rgXF}5`|B}AglpJRpd;C5B0Lg1Y{iJ$*a^T)qiJ~H3e zac#I1{YW($_C;ZVs!kK)R*>LmUwD%u+k*i; zzKPlVu|MP0`!2)BB^Qjz=R>GHp3Y$tv%vW_jV$~2z9Bj0dr!9 z;sMUsdZ}50y)xE;`)WNyd4;>>TcNk1j*(koEV_T>w8Gqk5D~mjy0~^`=o{NFLQv*| z`yskIA+6*Ks-4dfwf-#XBSiQz@b`?%3noY-+P!{B*A;OR4AI3O8Z+Ue7-X#iU_p>s zoV7g=AA~(v9c&R;`$9NT?ccX0^UaSq^EA~;WE~pl#$X*?7vaQADs{24A^keFeP*H* zKzo0qz%yzI*G^>4of`E0KDJFU`C)M<9EMBFr;~N^a@kt*MTNTn66`ZjMz*LA9`ro9 zI>8DpePc_|4|0g~;hRUcft(6*=A`={H1P!@LOf+vKV=~9=lrvGypoVCw91}oz41iN zh8Ck3>wqkUuvf34WrmY@ewin8c2Y>!6E1&i+=P567;)t|{@EpW@}`Nqklw9DK0rnr z%000en_wSNtO^+*AIhFfA6%QDa{EXq9K|5kgCorqE1>klo!_y9&F}btKB>|{Ad^gr zDl(<3|1>bZk6#JN|3$=ZO!Vc{6Clbt*>B}QT2=FxmTyF8V23g7q)CPHm{=G=5|w}B zSg=3p3j)ptWGUGb6;e9`UR;l8m=?R<5gJL$R<)T{X4JJMDJYqJ^`H3dsNFT6FZK<# zTdak$<-y&k^GjFO0<)^%*2}f8QBr-Lf??UHKQi*?vRmrwQjr{0^apFncm(|LK(ly} z%Y{B(+Z=C~Co~CSJ{|by+mEBLmcV~Jnuq)o5U7r+vxU_`r1$=;AJONtCHooK^?j+T zN|42@>y0vDEVsQ854oX?yvF1q@owEW#4|#C)7Yy&<({+%!m87h(~Oz(^LqCEe`_r0 znM%k(%~!83P({5G zzE!z4UtV7xHEoiB1&=cfi@<@)Y_vz5h7en#4a+mafw8K7T$>n4r!uSY0k$#)$s&UF z@@wN~<>zDXTrI&~YoXZo)7F1Va<`*d^i*tbp}$cO_8}s#E{Z7Qr8Bf8aL}2nmHQaj zB-uP=UHr!Y;dcxUY7+7CjZrMx5w-wAe7|-7oxl@t{LMFO*nxeh@As!AnA7L)xWP?+ zZ0vsme$KW>+u;)Fp@s{2&ZI3?ZP!#}XSBX!n&o2iVvP*=v_PQ1q{HwxZX-c-zW+=50dD~XemVTro-|!x{uz6FM8=8XN!+o zy<l>ajxq%i~p7RaSqMUf0$(;x^&}Ai8;6BPEpo-;VWw8`g)+#d(IKZ$`6Z1ND*w zWA=5^6XNjy*OW+_?`d4sbafb|4aQ;}nfIIMs41ba9j#mCvUsudNcE?*6pm}urxX^X znAaD1n8f)?m!r}CONGt<4bs{44d&;3zTs=HfAWJ|gQFD^wab4v^))=x)ShABusASb zNmI=>htq|@E*ATD=QaHQQRoagPk%iDFD@kFsgBP3*VN84xOBysMmA6y)3?HQfn6lq z36Y?GqoDt5(Yrsfsh?96Ok?U%@gL^iaQLga6QT>b(F`a5_ktNNaaVI(tan_&XhyQ>1kPx{4%^4(rC&r5_W3Os#j2ch3HJJJZ5T;JEC1(CAIUJs~->RC* zZw@A&vS*))Lo#S^VQO{e)(}x#r^VsKRn<@lx(1a2oY9Jb}Fl-Zwjcf##8sV3XkxwF(r0KwUN9Z^l#OpsC6( zdihF}z+r#+@O&d5uu|4;PRePvbs8A#`0YeaVj}!lGWkzsb_|@Fs*jV#+Cu#ef%n?5 zfb!?^zZ*yMd20sWH$B#Aexr|ubGUH%|Gi#(3CxlKbQt($r}0(m?o{6Xifa^_(S@Ep z0rr^u(F{W?Q|aSW*8ej1e}4zOd=VtPtm2i5G)sT`6z&iTBd6I)DhH9}Y$Cu7?_18w zm_P2A5Iade@u$Pe$|6-cb6EROIu zQ!IbeS9tW{FQql^TiBdmAZ>(*M3($V?imlW`+NCjtar`}VK_BcnJGx*;;^vqa@47Fb-m8=UrkhbsF%`$R9%`p#C$io#+t}mZ!EGW_B4M2 z3yT(~TpTYAIj6?@C1j1o2sk`&Mnd2y`sZmX(I`7+nt!b}e*FZX>+35PhpoxhRp{C8Ir~2>#-)vDdI$ z&Z$p!pAF$A4iXXi-V0NHwYq1^vweRyVb#qX)E0%fRt?y0A2E9x7x$j^%OL9i$G!$8 zTFrR(w~=XzPLaQc?sE|UG*GrEOgk=bo`44AjQ+$e2gs-;f^~R*eS50D@m_EBR#%B; zlox#c-_BJ255AA{{;F$ZwEr-lwvk{}J(AOnQ(Cw-h2Zdp%ilea=r1K>;F5nAH<@0a za(AJ}Uz#@&>_GF&VIC zDd8C%x(%<6 zkCeH8>q#kzPk0T?64LVb7>dDZpXCF!3b5-LWd1(5!GeY_>mwAN9A~9gqimm7@ z|LFv!aNvJ&woW(XRRd9~o+QU9^)q(IFNm2uG~-y)yT%>gRD7SkGaP>~!)|@JFck6e zh9T~M-BH8+_5@Jr`pt$F3*2234GfMj1eVyU`U%}EJzkKQ+%n*+5u*-C@#{PMEG?qa(R~iW+IyHg`-D-r7hHbKw zzGun&dk;pL!PqH&U7lMdB&f#v_WC0e*`y+k^HFTSQhQK|KE!|7${p%ks4w0Ax+xL- zI$*z7tP}*o^rqu{{TR(>he8c>Lp1DbqcOFAU4Bj8f_|Y9Ri7 zDD<({LEFDSo%~|NKc)=Rn1}Ls_)QVRtLH_$qv#0B^;AtkQwrfep{U=8oc)3VIOusz zw2dv(a*I{Z1yq0fDP@U<2{FppP|pkWq>ejJ`!Y=^{sg_zA!@Jrz#;XkBUZw%U~A zptOfV~!KI5q)RO8+z-M1sYp=5N5j zHl~oK`Ppz!lZ`GuK?-cC3`)yiP-On{4xe0b&_ekd!_TwG)sFprM~xWe@FX5Ot}Pl4$|JB^^zuK$>IsH#i(o{M%~Ic9m~&>aus%BEi;SBA|G@uFIH zv`@RqanI=4``9P5r{2g5>$KH0@nalqODF<6G~|`A%nV2uj1gs;ESN%Iq=GXgzXJmgn;~EB~Pr&OX?nfjM!bgg!oIAI& zw9S#8Uy1_EiY@(kH7@4O?vZXJ@1UnOW>HcbB!23;J>(}B1y8_S*Au|1lR9=ERN=p$ z@~8$Ufxcn3Hh*NeZ5C!CIJl}VEC0%7_tlwlQ}RfRJsdHdC?qi4w*JUvjU(!H5>6@H z>{}<6$vnD|xwylxX`@`|XUq);qrs zU#qhH)bnjh#C>-jVr5ZzmilIeW=0@f8s0s~S8Cr!-YO{7!v`L)h6646a+l4u%~I)O zSR=byo0SUeDmt)Se#Y$y(8GTM%*q}(v?_nU5(x^xz4)LR_4y=qyzU8zE!lr;WPThL zL5jYQm|M9R5ssV47f4F2TTWqkQFv8tF*WX=Ijsfzo)KTz_e|X|5$)cQ&J z!`?aA*1fesP}TQR-LGml!Xpd0y|lgwZ?xIZw+s+ z$R`8nahqFCFt^fOIF#+)_bO5Vt;_}1{6)+hHoFv6jqb9o1=RfYnY<6y5C4BM#*;Q# zr?`_H`^AqoB6ysF^zv-edm72AgvP)4!LNTh=@LcWkMD*)E`A((O$M&MFZvr#1m>iF zps6=E)J#yM?wSkJ!2uQ}fs+Aj<+YYaOt1DBAO`MAVqaFd3m6J&bQ9`7w3=CobmQ&% zC%SkL2}RkBJS23OcO7ad_~lz zV7Mt1@;nDyMS>RoA#!pVB<`9EFOEW%nUm$_SYJP>*C91sFO*S|`OW^aS?{?N%P(ii z0nZLz?kc@>0=nyG@BHT>Je-{GcPFStOs?xo(Qb6no>yH_t#;?1wq zFZzEx0fMFM5>p%ucN~9g%tr~v;)|jKmm^k?*=AgwZ<0oZqhGx!;&KH{1nC0Bsao3R zm~AJ`!69^*V{)PTV;@ zBbRDxWBRhfLv70TJk7+vKw+GxZw51Fc^WQv-6KhmjSI7`wHL*KG}3iuaDQ7j8oSqV zKa@U`h{*q@!4DMP9R^^~YZ1YpyE$=avkV%G+#ec!peGk%Dk+hyREm@1qp!>fEp ztrZhf@&)lO&gFj`&!-zPgVNdaVdDwdr7ZWQAs#a)LmWR~GW29r09raHC4ACklVnQK;pVvPv3EO-e55WIi)F zYn&;>R?+{I$}YAK<`nDc zI7NhIu>pVnyYGwl8n|CIBVyWv(NU+X9`P7NP(f(aLUtnur9$l64xK3X^uCl9v#!R; za$*)A%~nr>=&=tX;X&ASiw)_(=;fy!NE9aXXa7xthYs(R0VQPnFL~WeGgE}rWFNp1 z#DpNl<&D(C{e*GTDE_7QgaiyJgTP6jqRiedLhOHuS-PF2gDlr3=!eVc#reIBh3r&L z)i>7G=xzFkvviZG$>QrD(i5AVyP7w%y?GPz4#XdRg-%RCL`clPnK0ez^_n=0g-d=}pYubh)RN`S(+E{Y#=1qUO6@b)|#2{ej zYd22{a9W^%-Wn#?lcYcJW@E*TuIhutopZ^IgQFF#Wrq`MYU3cGA8gY7UrU>i;OKlv z$}ZD&&feh>xGK5rZCSF4|3xs2x9ZE=50QVukY%Fx)j;Z)u6TpV{QDar2+v2OA{i^+ z`V_3+kcI{WJ29C~yFOE{47=}K=Mz{kUm%)6B6IP~r&Jr-8k3e6Up?Po?pyXjGcxmZ zD8G96#3^|0Q@XI3rrPw842BoEaIyX9g|SS1I}n6BtjGRm9^ZPHsnm15g#SXZfgL~CxE8f^itjJqM*1mYR?Y0^}MhKlOZZ({5E-!D-;OCPA8@zA{ z>=#;jum(@RnY4eWDmRV$>SL3KqcDFmbTWXUKUdL0Ko!!!s4Cnlt@}C|j8M+6)6(?f zSAjNnGE<-P6R@NVdGyf`eE7MmD*85=-CPpQh=wA%5Tbe=zKjhemMxtbsV39iXed-X z>jh-FLNBZ8Mj|)cy)jqVO}Y4uE8mqeHp5M?D{mtTxFb3$UX-|4E)xYUE;fIXX%P;% zj7_DE*C)$YO$trGSJn3&JeU5p@go}}jg($cX2KqHsgV&Xt+9q(AXu6=wb2qjL)GpU zr@s30inNUZ={)kq9(Dsr`tnmGc%lkr4GihkJ6>8ES7O;T6x_{ueO7GFAyv58A0w2Imgp_&X7VRL zYo)BU=s^+GJ*3sN+xV)l9Y(hI>i zpLQk_0y82M9GxJk+kvmeX zTYj1E#GFFASP~hKeZGNLU7_1Hn<&7FQ9my71h@u2&~ez@`ILY2z10A6kesFvF_e%n z->vV>DaA{)y1biNd6}8-i+ux?h&a{R!}nGzQ<&P8%=A-z6c@9j+#{p*DdG(WF*7^v zDmLugsx{BuVm=g9@Td!^WgTW_C364`m2bA&988FxC+%+!MhOKxW@WJ5>PE_{^2xpU zPWe?#X80HH0lj}B_UkMtQEhNqGQV%EC;8g7hWbom^gT6OL6NWrUufh{kud~9TsY&? zQGKk}MOIf>`-Ehs_6vw)^!;Do8Q$@Me2hh{lp>tciz%Fx4XpeKt0JWN9@YW0p7?XZ z!s{rGZwfWxuRlxYs-sk)If)Q)>4*lc8tRQv*R1Yj#b;${`DNCoirj zXo|*$CCWuQf|Nb!%tB_)n zGwgy**iFEUy1yh!IA8C0t0n(yOWK3k6e0Hbc{Wy1ldto(VLb@4ZDNPGz)Z1v0v0SS zDzlTYqt}1lm~z8I9N>Pp7Wf4Cx@V((%^9;|)Fd(+-k|axTF@};ROLLnegY<8c%b1z z=l68E$+(;*k=jAUC`x>KD7j3_x|YWFhGK)?D`~WJZ_-*g#w+%H59EGDUa}#ZkNE{j> zkMI1O=g)7I5DUAvOwK6Vlc9nhuQr4gWip4>?JPdT8g6|@j)?95-G)sbBKq)ct&$2V z%$yu?xa@SlFZ)4WK>H@u`ezR`D@{<6bQiuNzG&zDomwFie3k7JP(5|tYkqBQQ*>ys z4E=vbxUL;fQ>WB+##CW{E_IC`*Ee`TTzQUHR|cJ;`8gf)DJF|9;1NRNrv~>Xw{0iP z-VEbYsX9&|*y-|qFP)A#%sk1l-GtwF-6 zyh-u9ca+Y^S{0fRCi#GMVOEYc*D>P=XZe5N!WUlprINl%CpW!sM`4|KY8(^AGJPR7j?0u2PJaXXx^!{p%!Q#s(w*QCgsNVsol&v#F{vAyKT2 znpZ>5rH^kT-{5%cKG5h12=w2TD4aPmYM9D@6KM6k1a4HCfS^c#{RyDZ5?(dD#mRrh z+@mLWN71A*d-xM@lu2K`RhtfV(Rta5&)sZVCLJz62s1X=aG7@dIJ!-I(t1dhPF=(} z=>(i^)XS|+^7g(KWv^Qruty)kox&p3n<|kkf{VkBb6f zyPap%^-*0>L-)%GUXD(EzhxL^bhCeuZC0miaV@WKZ5?rjR>Zqk)R(K~c`bi9(_@c~ z4X4&XX%U3YDruxhTVH;=Hx!n=gjiZ9GjV^>fnZIZ+vXBT>bE}N$89!D!@x)~U40tp9-jW}_ubkqac4t*TJt%7 z-S%9Sc&)Q!>A3PfaOz?E(G?Ck>oSsCi%)fE zvYXmxV`Y$RTy_Of6Pi{lA+uDc@N!`8o*bbx)+Yj3cpk@V^buu2Ie4cefAh~U=g zOZZ^2nuA`g%NOAnVMv~rSgp-@eop3g@y&}!!t5049#`@FEG?9RbZ&nsnXpf%I*s9R zt8vaf8?TFA8~f=edz?uB94zTOX8@;sp5-oN*QuXe&TTj?YUMbh#^;}}r!iuO1S(r1 zE*UKES+nAn;qjYdpVA9sFN)f{*@`%PA)bD(jP)*36Nz9-DQ6bXfpf|-LQTaJ_O@SO(nTW%vJ?XHsT=hUoMkbr6lz|i zEn;=aLYH&+0A|aTQ#4qpuCTc;Sz5XG9)8^Lr!UU+AE2LmY^r}eE%op}*=N3sye~pP zc;BdNh{ZH~9>H(2tB9xZ@7AHR3czh+AsDdaP_y%D(UHC&vcOXF($ds$g|_EgV{j}q z6_}cO+f*~1CE1wpJB_<%r$P`G!ZesUhQW8;JBtKR+)u?gzi3v%D+_gYm93dbgR=PE zYzE0hoJ5?Luy%i?wikO~#gQ9t6b5y4Q`fZ35+t0Mk}lXqzVk}6&kZrYbTt$ zi9rtH;7s-aTKLgA4;So(bMVW0o8VgRD>bzjo^I=5!&V;LAdvy%#g`EIIqdlhVp943 z!VbhE$=+!=Ll;>h{x*TiGNFIVLf@ zP6>g_RezaX2wdmxWitM$zV1yqUh~lJWkZbLo(Mw0c8K25vPYpls&j{J$<{Ya8u5SU$B9UhccD}k!tV6_6=%O|$CNrj zz?4zoN@D44e%-@-6fy4y*7&et;|>pnl1ZkfEAKVkZOLI)4LxE zoMa6J(C=w5m^E>CZ>et`RKfDxRXU2(Ov!0++a90 z^Q_;#0Pdf0dV$OEE8`iD0QHxgi(%*c^dK4t_mGE@C~>tD9cDCj-T-hRT3luGZFzrb z%@FqD8Sz_w%XbI$kHAfz@ub)muH(bci*mx8Q|{HkdAbJ$adMIO(D5|f%l6ze+DAan zb3)9E=(f$pXv`J5$)aDJ#+}22W~yR#Wg~9X4ptDs*;wU$hse7G+5ETM`Bx= z&$(~r3V6~-O&c1z8+yXu!KYIMhn9r=+lS_oQ~kluDv(&Pmt%P}G3jVzqUKXq|IWc& z%v27A_1&mrs8`BzjUJ6#Hm-|NvdfjeNw~u*|MnaGHw0RGpYDQTM&1B`W(R-l+-cAz z*p{9=D;YD+^yDGfiF@yueSN(Bf~n}Y169}fq!7-`BS089$Rl=8{^#pQo}YQ6lK&5>AQzPAenO?-!LSnatwM zc0LkXd)E!i=0k%0HjjHYn#zci>?Eng5*oaKYa$OcRv8->NR;R{3^^(D1$s!mbr*2h zyp@U>Ep{iqUSe7}lxdd&<<#2o3MS#MNXv7aPdow!*V;$3_h4{Y3tNAY%jss;Kn*^^ zr*&Q=c`qj*A#RTeT>0EP)cJ-F*NUNS92C z+49XRi=yt?8IZIWtGHv-nNib98k*4Foh24(Mm$K@J_5$)gAN}=oXZBBoeGz6zdF(e z#~FK<s-1|+U&Uk;aOZl*#?v8#1Rx`1%0<^|B1golDtbE>2%HSYMM>zcq^gXP| zOc!~s*POU&b$>^1`#^XnA_eHyK%CBRi=PN2&Q!0&Zzs1m^)ov)9kAYbHl;gO9`Us1 zu~fR%0S`tfp4_YIhu97Q;8nkzQ+nLVp+H*m>|Z7wdCs)9i(S z1I^d(jt+opCZT`D(%p^bi}woa_o)qTZSlvLw90PZ&&zX6A)Sp*pbND)?w@eWcAIgD zJ7qdrO9!2tt$QOszW*>5|J*Umug9m>) zu1m9nG&u$g8SV|sYPke3BDx)nSGN~T(>@K9Ag1?hE?=Or*6vT=Zs>uoPR@`oQIy0By+Nc*2dqQo$9{^uD=YQA5C?rO{Mo56}bk!IqV;zYU_2j%vz)QsmMZ_mY}up-1`1dcV!b#Feqj zOlxl^gd1NCo-L1>`gVyX13@laC+1v*1DS>w_SApZy5nxsL0sI-2cN%ggWGoev=;{E zCm)i-@}Y)y4b!Z>rz}ric1eZ|#Fk9@YUf0_xriJWqS9omBsF5@oD}^kxeMKflg8Lx z7$EbGej1z%!5U+Eo@co>q|Wt#Z+AAtR)LyEy8}g2>icrU;V25|$OX#fvHCLZ_fkY6cfZffUyC8fM147{v=LNs4mjW`9b7M`7J}`@OUuI# z^B9xQ6XJ?I=3g{4fJh8wpANpyy~@SZn)sdbAU=Js^Cv%aQb%kf@ES?s!sH%Z712bm zXfCqyaAaGe$&r8jdERCDDhVWCf&X>SP;-CihvsX+glVvmf=!sFWrhR7BLI|lkfzg5 zQ^n{y8Fj^OCP9O^L5HRdP1!rb$4yiXzAiAs_HTiXtIMnxBLmw4Z^KBPk*??M4JQ`Q zSv^)RH#CI|ev`&IKY#&ah=Vf=T_SzPc;n{C%1CwNbOUq3)K2JSOYVP`2(xr~GheQ*&6HV_}7kN z7QfnTls?kB<8wi&_u}OeEKC?3K#S$4oG##VG5TG;+~q|rJ<{Y(kDaNClJ&+OB&kWl zQyGpgwZ`=tjUx2CFjcy(W=S-TJ9aq0X3L=hcT$5y3eu){YDiwY(vMgRO9+3CgIsIR zK9t#(NK)xXP@4Un1@K}Zu)3RIFRM zf<;X8M*L#1@<>N%^883tMhbuRo#89pq@*=+G5J~-IIjl54O#B44{#TBz6%9)9wY|t zQD^zsq}KT0jD}Keo#W})BJ}(=QHqicSTGi!8WvZ5--a-wb-U;U z7SXw-UX{e+V1wy-(~CsFlzvqv6lFFXcFET9>U2s#69rxkS^Y=fQU zt>1ioEsFH>gP4{VmyCZMZr?baE^wYoqu*#w_mNF~Emj@uL8eHi!$g{lj7flliWJ~^LA#Hk?o4n^Touh~h000r< zH>Y}5wo?^B8|io3)Pgb) zr&$%ipnlHaNM;MtIQk2iwpCq2y7#e_;t=k#fUxxvd_{u^oLG*k(&@AKhzwj^rE%BQ z$FRKcxSwMcMB_o?y5WnLAc3VlS@|@kG!*c~7|!T{zCC}vo&b<9pt5Mc2*6T^FGE&I zbwBd%;l0QuNOf7if3>5*9e+2FR3DG~gV`<$$iqpqaOPZ=cq*f{z>(>AJ|Lg$07`^a zYjJQ@O)rX|3b`L?m*!%WSv!)CtWuln0a%N|!I0RjitKq0Qol$O88C%oZ}f8;wCz1N zn2`~?3`u{jhw77k@mzEkP|El;vnGrr_<<+7mS^CTzPYG@&708UwR>~>g8U$i3Xx>+ zXo7;Ela2D001eu&8?AB%V*E)-H@t8UoR}GsYO4~c?cNmhRi8y7r^;DHy()t4ydY|_ z2V=HwfW^EmjaU#UodNF+Ia~LWL~nESld8UcZ%TjVWQMv1gSrpTulhU*hhVVSWcV!0 zQ#_}Z+c#d(#RFqWgXH`$=%+(al~t5&l3SN+1g+@_?7Tu#l@s%`x6``4AKZ#O$OZiw zrP`iV_pa{aDhRtujRC`Sdz0LB?rs>-Dp8s(NO zr=ovIpD2mtC**?eASMQ4x}Zr55P|7<$5Eev?t%tAI!v@J0lv`x#dJ_C*NY?VjV1ccjs$fhGh zL^QA{uoR%v3~p@uJoJ~3jD_|KKm(g0@g}LhYcInBy$eh=%k^|cEP#n)nPnmyVj@f#Prv*=?ugu zec<+GY0@t^5w(>~AA|}es1m4#;4=}w!zF0^+wt?X&P9^Pd`C(B$N9RQi9% zE*%9rE#4lr@>!n4?F(*{RLu<`r>G50MpF%(g6_LuR48iWEi)B|<5K-8JCiR1@n^J% z`p-Ter+UtCq~)jrQ0rM)K$RBk(BeLNPfiuF7`SH;HgHH`?z*U!n}olY-M^UBfqR1kl}1sAxjqqDJsjyKqsbKON^ta{d0sGlHzLY0LzC8GWC2DSX{nEF8T zqnJwZ+ zL8Fh)X@B+NZ5n-!{w12l_}aadT$0j0n!`9E@1FTh6fpK}n)_a)5!y>o8#9n}Gt-#v zpPd`bwuytL-0P1HcL!9h?9auzO)0BXWbWp2G_dUCJZ=s{X|puR0uQtz1t}!x z=N)ve*q}t2?Q|$ZnDDm(Cysw;mpDa@y99C_N^paR^wSHL%ufbG!n^(LNCKe>$Z!VR7ZxFLKTrJ<2jO%Mlc9tALyZ1&tJ zjTYO!?4f9tGkid7u*sG3vp4YL|HwN?Be9_7W{tnB?MlRbz$j*frM|F}P@NL6T&zARl7UXweE+`w+1aZ6o8yHxBpUpc!er_5hY zhKNz&Z0%@wD*U8=bSWok=q~UNN@qG%d!g2!*z$_CIdics@Qb&R983{%>rR?!0%$dP z7BQk5VH3H1qQNK{Z8i0s66DKfU9g3?JObA!R7plgfDs;pn9sbq(Q5hmkog{k)kq10_80eK8B{<|%Mq z+X^6!6=Rv3rj`!jpJ6KPLJDF$SEbHOAxvDMJ51QI=_hFZ;V-Ch3}XFh#(Txsj>8Xw zz_e|0EvRE&#+ZLcJ8B|`pCc8iwNE?nAd>7;jq+phixt^Si4)K1Yv+_?&DEk<^!w(_ zf&SuR)4y;gI||ui$wpAC;zbmNEDdc^kiQ7iWE3`UU*e+Yz_U7(t=p@L8CBF55v)p8 z$wV&UNo;`>ImOJLQ4EMh+hIj_=FIi0<*ul*yTAH$EoFZsNpZhK;~lBw2PZqS995d1 zvTB%H^KU%@f+WL>$B7q7SG6(ai^8}ia1a3`h4Bo#Cl&c<{-VNsds3bHTO|5p&#!ZM-{VYWDdR!I40z8Z@m5qrq}qSW4<#lob9Uj2Vx|49!5-XUd4I|_3I9n&_n+FO`9V}ojUoFq*xhK73v*VT z5$74U$hCh!>JP>mz)c?=*^!3ovXb}LYV@>EIF=ux_+5o(6cEc_5YB-wPk?WV&_(d( zOa|gyc;;k=^^luzSF7)?${Awoi8q8*+e?s(EE9hnjdV0!WY28fqA;hj+Rs~I2}cK} zS$y=0KL-0E>2C%*AYXtQT4YdA`2G>7G%8GV3td|J1%fXtgIVkV#CMA*W1H)=oydk;qcm1@Yo)O#%W!Icpp;h_!BqI&tK$-ZRN?>! zPIpxQ}T4!y`dZB-I zU^{(TOG~w!9Sy+HXF#Zo+Ea_bVh>2x7Q&Qu#E!O&lx1$DFYwp}8RiGaFD>=iV=Q&G z!iDB}>GU{MsHFXE?>C_G-^3JRg^}Kpao^SomHW9Egq&#%A3>Uxcbt^V^wVw1Aal}c z<@bNmbSZz-i}uut$o}8e3()|%kj8&&?gOC9@|SD*fa z3o8Cd_Typb%}eZQXK_gjT36ApWh;blZI-QcP2((7@`IR{3?!@(mkihj=W`F{v(FbC zjzBvp8~D&O)8?UF%8_4iP(;C8;T%v!I&ukTo;;^H!5?9zSAmQl=!Vp8x@>*zIW|}h-R=pb;9HOTbw})?dg{`%t1##3YF91|PtG}aAZK98u zKFF^CHh;8Z*o*lHc)t&nhvWw_=-H+SdW%^`z7$c>xK2+!^jx#Q^_*C9Tx_K;DHeq5 zLZGsBj8wVvFKfIC{(jG_9)+q_2XUx>|0Pu@!8LOsmR2#NCQb(4Xa1c&MTS^Rh)nUd zM{Dm+s9@lAdJWcJ{AYm5%|^?_hl4Ocoul|xLlXa2{j=!53F*j}QBZb!+`+Bz+T8+=@GD2>acKfk{c4!XLUHLOio&1mb`nMjlQ=$i8n5P6OHoIaMeHf z7w&w3tzZ8Yc4<{hAu{RfLUjkWq0%%OaNN<}gTuemA;`Jlw5kB7y}jcrO!J6aiL+mV zwT+5@PmZ2aaSjYe!+#MyV6O_>>{ESSmn)3^cKy|2{X3Gl&Cgm_V*^5?YX4b;{H2XKS-Dw%*@h*!pWbce(PsK{ z>7V~oiq8L_E!zH{j%fOSPm~b*ztd(?i}khR&t&Mr7L9|jMS}Jo@J~(9{%Mr?_&J-A zjJSGMu&V`Ut3Y7lgnR$U&?yhOTwWyw0(~nwiH`o|9V}Y5dxr#3N@>@(gkP>W zIgYE<#d1LYb+eTCDX8Qtv_`uD-O1fkg-BfP?w}uZJ zE$!Ymk@;$e0r(o^bKJJ1VQNX|1lUD0d<_R~POdSTF0q-aDCz&{9`Ik(!An7ilgJKa z0{jW&A8xP(*N*szq3Dzh6Z&AX&n)&#{&PTFUM||y|Ex8nmgti6f3la#>?kpF0M`6?}Crrm}Zb2&+MaQFh;2!r!2apu@EC+Z)q&3}h#M zzU*I*P0Ph}|9h$Se_8+nW1$frw!&|cw}U|}9Fr7Z5S;#6O{T)ehF{Blwl8~=tZfF0 z0E7If@D1-HQ@>9EhZgh{AwjS5%NJMr{2fwNtu4-f2fuGdf&W9wa71u=g|(k^ATE?O zR)ZfF`;R;S850lF4J1JjQD;7<<4pI8V%gacE?{iZI_Vs~9;$|hQuauOiti&iZ6}9Q zpBA3D#v)dt5>+)0Hypr*-#Exo1~X5BnKXIfx0W{mt4 zeedso8ARnNUJX|##E!9A2F{AEOe?@Y%GLi*hc;IFeN8LxdnuMf5IE{Ja1Po>#{Wku zAY-^}PvmEy_kHv3X{l3H-nT~kjL?sY)<`1WxtQDxz{JGfwWOgI!4V{TLxhr2u?JX^TPVdV1;6}}3s^=TzLm38${-5qr=IH0_ zc{+eF6xP5;^qs)zf1EuCsqpebiALS{aU>DkHDOYutBsYHnn#Qe?EkCh|6j=xoBS`o`s}uuZ)6{E zakN0o;mGzK=NFw&^XIw%Y+o@+Uoi9t8<97V+hntWGuU3vzWEwgSju-dPoy)7HAC zHbqm*lI=HAuCIEfU_;^cr^VB6ZQeZV z%Om8ss74k#CgpLb%&#u*FR)3bp)|Kk{dhUQaaZqMeygHh#1>mY#OkXx-!{>I#~sHq z3-I`xZwE|Sn|&2>G9`*LQ##=ew4R(p_m;9W`HfB%_)>%Zj`lk~T$}!|AZ<}a&CV$8 zDU;ag$KcGWj3`f^s`8Vq=>)enH@BTW0@Lg+6fOxDRpIeOeCcfm-tPB;YZGbGx=CQ| z49ALP(1z}QJUWHSg+mFGe_5h`UE@4oBGi_h#)sa1=w){(ELe|ksI01_gg5^r(dt^8 zqifr3?koA2>jMjw?aA{=<<7x*HN)!9wf7kmzCI&&G3ZZXD=HK_ZEC-*Ip5KJMK#+; zv4y375zOk$c?y;2VcajS&Q;B&UqOoY)p=d8?g5gO>Ge_f=bCr~jDzTZ{mn0THe|Uo zoVnk6)-^+$FDnECD(L!p7K|mg45`MKxA2ZM&1QLw6x?8{(($3ev?Wy{VHynvGbh~U z*Hc^UIZf=qZWPe^@9kim9s%tX(!7#|F6+#gUTchk)pDsvfX1-%;#jOm$)$W%?!zUC zAtp^P>Xlm<@lhW@tKYtV^+FB0->^Z+zfP{KBbscXHQV`Zs}U==uu_szoMRGW7i>BE zw!yiyP4_I-t{;~&%hrCM%hFv6R=E(nGj;pVx$|E)<|ODd~ZAIpP^{ z<6X{7yT>CYpgFK;x{`6Y^$&~JR~uQzk%sv`6Cu*##zVwF*q0JG%-K5X|n zeXUL?g;+=Lx(+nQs$%#`Qf0>UR=C&7XSi{?`3N|Wo|CRUWw#V+8$DhLhUvR5*f)A< zLstAVF}<))ozTdCECaqEfXeM^VdYUzN4QJy`i16<`b)|r9UcLwX9z=<{7-4@`h?&F z1zy~!^=F=;TQ_G0zF#Fv3_pqf{IWJ2#C7-YC;vccP^~a13eJxY<=)6Q+I-=GTbO8f za8G%|(=JX%m?~P?5_2Hw5Pf*C!}bI5qF=;5f7T=Luo*OemE@|_Dq7$qh}&Pb*LnJ4 zR&34WK0V0**oW$holSLK>H||XEi=0b>6Z%zeT~e6^e1;B#t&>@5(NB5?8(NG9m+RJnp3E z+OIMymKUyn?SzXo+B^i1?(9kB48EKYOs}-Gb*L_Dng-&Cz0|J!_qtUGcc;sjZMpMOcMr-X-wZFTg6rwuai3U-JYHbN zO$WW+*IBb{u8SDaIon!Fx)VW_*;U*39k_IydHT(NNls#1@MIF0lbR1;UY1>x2;7qurk3xIFU9Hi<30vG%3s;UsCtLh#B z;f!k?A$OC9r{_!ey=br|Rg?vAjP@`8{O&<+^3UujwJ;6X&Dv#n?wmbEpY9de{g197 z9g5k13UeAlRUZ?#5kr3t!;9kY8zZ2cvw@uIA#rjmi+$g33*4S3((js5kEQwVlpJ(e z32{Huor#4b)Sbi4d=i=OCMAzNddki-pCE--%l&2e20t1(vyQa;=eHp;>)OIOxn&+* z5pvCI5VO160I-f?bL>RViQ&kVMI1dog0YByiFUh*nDRGNioOLzmpy(EJisRT_>3TK zB%KCaAQERc5q{A7;-qq}R?_)-bk(Wc!1@D$NlyjSzF^e&_hrswn&6!DAg-g`+5UDH zYJ-=V{YbCYgG3G8jI{I0*s|=B6vGNXFy&u4%a8t3`P|7_4=%H9Ni1V`NG7#AnJVgk zhxCRzdrGL&q4-A+XwQ3s*2|H=+WS_o;|Ax62IhVg=5O@_S4 zI}R8=R^Ly%^trAQrm7MSJ$qBU+&r50oy=FqX3hPcIpbH(U5L!>kvQQ6wh`!9yLd-_9F@BUtYY@)~#mh>vY zCD80OUSAsIe%bv&%em*h;3Cgq=K)*hU_e$Wf)h{L>d~Qlsp#a%{Utp{$IO9}xidIM zOe6I0jQyJ36{Pjy6+%tz(5+Ce6Uuu2fiirCzzdFSi9QY7iq+ryjWqLLKQuJw+H z4#k%v4XrpOP1SI{BK}x^j@3Hw-&-9bQB3 z=x0z2(bigRLkdnfuX0_${VjhSSBOOU`v|b<3v_X|pj)jb3d*}JcN-x3IWMg3x_>iM zc3J9v_kG5y!(24y-p={6!hICb^!=-~h#;9Yv@)~MbFSL)KF=_J;h`2%XohkfO&3m8I5&eLYKU3EY&MX2cJl3mFJD4#=|pA&s)LkD5AHm5uLr6UoreOu2qpw?OG~AT`a723}T~|i6ctMU0{87u*MTR?Fx^lOW7DV zc+e{8JPR#j0KnL;$L^lKx7dH>pX<93#3Bmza&x`Bk!s3(`+j@?Gsn*Q@@~WfC3@~b zb9|j#ZXBV1A0y#=$#0#Jhe4w(uZEx1k3a>1iQ(dXU9rVa=UrGr6mY;SranUralE_6 zucc)wzj#&8Os_Tb2-MAS$BOalWjNSE(#5SsD?kPVmlYXLUy{X>?jYUW>PR{Z>!{N8 z(lcyQH&=Y$Aa*FW4@Ks+_b~a*L z*wDsX2abiANK9zN4Y$~K&UD_ET^5AhpBnAvz!AD+(NMclR?fFOw2f(tTN9|{My z7?9FyS`F5o`FxkhL!Hx@)uoVfw9{`~*j7+~u!+8UhE~(*KIp-HwHFSv6?ZXGe%2ez zx9`9?hfAd7eAcnv#lJt%#z{8D8aiu7^9our=Hj%}TwJq3RN@WZk`oij%Jt64$i=7st%0$r3U>9N|GkP^Zv_X&D#J z$wW2PHJZ&Bz7nI+7py9|&#o?mVUL00A#gU+-QsZIKgVr;Q1axbl0%_g040;J2|lD5jaH(y`Ot{7sT;_l`WSw zkv2@J*)5GLg#R2gV$ZlSSM`$&0lz$0WfB4Mt32R2+@oY40so5PTrA8OM?Wh^{$+b#B(zlA@A91{tsij<*wl40UlCv0FJR^}uJQ<{E0sG9^qFE>&7eo8&IPE1*1l_JL?`;tEJRHqe zL}y*h8W4thu)F-lZf2{lrbNGg-h3o-$EyqSkC418KMP}!ykB-ZuX-&@|A6=T`yG`+ zBdj4vkXh*WlDyIzExf9+KAl{o={lNOa7N*+QdT5;PHS_Jxk#9pN?V-DuPq(d=yH0A zwM+y#{0Hi`7|U>e)&jHSwZw%Z2wWxdaF0og$ghn{F${aQP_EN}twn2peO!d&ZFyJU zM;=*QfZg_(s>6{g)>xv6KQ1RsX8V+6acM!E=rZKoy|*3v8HFgFPZ^M8lxDVC6BQIn z6`%Lb`bPVPsv6rr15MRHOmE)~b1tuI=M?Wx8ySj(ZHY?tTJZ0qUseUm3Vu&Hsr-a z%YSOf&)h`8WF(2!n}SKdjNTjA4v+QpgJ=b-%AB&D3aq!gyjd!r7~Vt)Gs4{=_~BOj zVo=hND(U!8;Ksb>!{7f=6-{;#>Ul9hkXmR*uijv5gzOZn_QkV*&u*S7DTYRK2Cqqu z=Jt)hjPrQVYrTVz1MO7h1;pYXpnV;Xe%p-k>cd<&J{DxFAuvoPaMy5nDN9TR0_M6( zF`A~lw1SHUb8w~1lCXq_&#dFjn;hduU@qwOed`0gq{`|1z4p=lLyr8eha6b!XvCb# zXi)bJA8w}Kjf=j2_1_N=+XbFdLOOq#3+K*|=A&$B4CQ!+!e zP3f?cA*m30nj0^G!4K2F`FR(@7PwsI?txjPAGUd=0FVAem^F>$$T*3LJ|4M?bdEFY zH^=bBJ+)&JG!sj9KS$^ra(5I~aGmP6O4!hPYrMOC7tVBlbvb)qI8_X?dl{=yFW4Y1 z`n%YGL6o_fzMcloy>c7y$9oofcp(15qvGe`tyHG;H>>XTmcVSL@q$u`;w93wj;Mk% z(d17X+3sIIJp$RE?qb+F;~HThDi;CHqG`h$L9|9XIx(WNZWaw|WzZhi>MF6D?kX5b z&Sk#Rt4TtC0gNi4*naTu0iI!R73;<051p;MWd7_N;>0%uM>E>42D%cvBRh3uS>*)) zR!&cmv##BllBfFt+)KNb*OD3$kP15W*1RT9)Vnn;^ixc3CI#JYOObQ#p$=HHp)mt(rd~BoxX$ z)C*#blb?m~Fn}2pBa6s!<|Sv93>xr2^>1D`;d$P3^l{2gGwgU1XiKZR<{~Qd=VICN zWRx9$8_&x&p50ZEFRV+P6PWjC`yA_cUsBdBGYlGT;T_sF{oS5BoNua#wtkFC5c~>UBJK#qbQZ@sUv-)hu0F%B!zjbrcaqcG4%&gc^#*zSk zoY&N_T*>p_nu?2FkjLVFs#eVLrapQEyx^h!TvdwfMzW`3`N&=uo;Y#pP13@u!P`MB zuB@=m7C)`qijVaTiQ6~QMV)N38P$6C)d~r!AEp9MI3E5WMJ*U}ZkBtASvGFMw~JQ@ z<#r7Oxcgeo%r~g5D28y+zmzGlOFub(E(ii=n(Og*Ck01k^UpEN(GqC_2#1X4LF*s* z*b4J{wfCps5lx$a7Xd62*KTcw$Fs570F$9(%C$$xEvR65Rtz(605tGd5vdhYH2s-$ zl@6jmg>HT@22&sCcqY;6_tgYj^6TMmyZzL;h>@yq)oRkhPQ?t}3%u7d=p7+{o+@5j za$2jBdS}0s(Q{d>(x&M12z1yW=}$ibYOT%M21Y54nX&^rG-PCc_sp*G04kb#7+N%1Ab=aSP?2VJEE7~Yys>vyje&9Tu0PD1Jo(B#R z2L48S_{bQnPOhNoXEUBQ(AS|b2uHc*%;m^m8FQ*?^3&pWj13kj@R;Sa=^9-?yOR9r z0UaC3z|f9H=jiz119`=pjUJQs^i_qNx;Sd_&38Y%b8#AKFZ@+*Szji9*)W}=I9yV+ zY4^E4&?6*{sYnV&0eTl%@++cqm$dpf$E39jn~l?jj`|N+S93M$J*m zg|HG)hr16J{V^6Guu;|bZ?6Rb!~bj@T60xLuVlfW%@7CQe`6fpT`icZ@c_vc-mV8x z9-loRg1pl#1i6=P9hWW!b1?<`^PxuN%S_=TVc*Zi9?HZ2r3uh~8!xX?f;s&+EGF*H zj3NOnoRA<6hu2#zelxLj*htBSfq2CIflRxsY&$89f5_`U`q*ecE2;kcV8{#hvEj#3?tW91jbn#nZ@g)%$wh zGr?X;s{rau3~x2{QBmd!RqpO6cWnM6IN>Assmxk7HThu82S_&B6XvN5A^12S#I^$(;7;B*+Q54P+fj=J@dAoK6NH!tu+2>c8+*+vB zr0`#S_Rw^zS0LUzt;!E#u;V+iZRp=oHOR#f%8?&`@YgX2e+?cpMW(V0T4Sf&h1P?O zWNoKjE!-{AIYx#?r;jxMX~J%=r?~nS?@$1Vt~PEsh_~~CsrmWc8lbCksU$8w5Rg9tS@!cB=Aw%m zm=)s{PWkvZR}WywB&v2vIa?;p+vpMQ67Kqep(DBnA$W8}!G({A&B>}4ZE(AFut031yHnplrwmSZAky{wlSm{CPc zberOiwC9LJOGkiSKimhf`rHuQ6nsZ1TRwmm&X#w%PF!m?vGzE;`vry8#6J?F)mV`6W(hf9ct}=zj(7X ziJb0P?zH14^fmeg9@ZCOcx<5ZZN_;2=6EWep^*wvXjg1qUsO}nIRosqNm|1dwd)?h zlcg9~-#Hh=?ZGN_NZTMKzWO1AD|>}W#6bID#P9vcm*hx`WrWy}(%&EB>R}^m6O^NB z)j`hZUA{hRvo=U?x_@dkr={c+6$TJUt4!+fpcYi_E9~0VNa542>#MuE95NtDD$QbVebYDW+cK2AN-7;FE#4RG zT(nkTmTkLDB}V55{N_Z{8|60i!Zs^Lwp>+B194FFAnYF`OSyACpkO|K{Jic1rop!Ur|9d8 zlwR4+2ouP)Dr(j}bYXGCy+>64mxziuLru0p27lEn)fnnys1Oq5yvj|ZiKGQ~$>H8i zEi+(6E^~HQQ3Wsn?@<7M_0Ggo20q5$ zA}G2)ABoy$Nsk|M=CKYf;~M+(|H+UDE3<7>HyS8~O*1Kff369?EwUFplb^d#a7|6@ z{b-v1>do$KtcakFkreyO5H5VY-W9bsCx^^gHIoz&P5TBb_NvWi(c21s7UayO`SAo* zN9OGfetHpuo0KXl!!^|PC*U(RR=SZsA)Yj0^nnt|wUQhZ>dGz}>kE*IDbBzAE zbC`#aW@BXO5|u@3EI?BON@qkE^H1s}EW1=-%(sz2pUB!5!c}ILl%IwuHFBh!(Z(Mq z#JN2XR7kN4=(-;{E3Tks(5fKNmL>@p-iu9u0G9V>3O@ZxUFPmG_ zIZyb7UcA(jF?5*&^RSBAk5mZcGZC&T9pxwsx*QpBt0ME^r8#bWa!Gwta6(jh&!xiK zqHcSCWnh93>bNKQ{QS^fKT6zn5(ZuUirtaLqArp1MIp0#`@z3LQUO9+=i^khb>{Ty zc2p1$!dT4zN;cD?ooDMow;Dsu@MU`=>3DaEfVCl~Y|J~gC3EDqx|vK5^Y^@j6kuTo z)iCk#BAnPZ6nBkT+OvyJAz97CYn-9|L_I)%Exa8G%aJ&*%8mj?rw+Fv=X;=6OnO{d z?E3{UZ;go0fU-!%yC`#- z6x#GxVGjpo8s{_wqu(tyFp+QR3TxVEj=4gUDZ@N8)sZqG+>C}0STdRxbnPn$P3H32bfuur?bR=2*erWM+M^=f`V1*V@&Rx88ph*jXq{+(h5Vf1_9e#0O& zqhxDNPU$kTPn7Y9g&T&TO227?;KN zvOmH(h=F_JcT!cVtIRS~{``*FILg%Z7t}j(m8UM&2Fs4O@+0|o8Kl@Ortz4iq=WaUNe9ODL(%N3}z-n^~|FU}l$YHP$!EclnIPts9+xwB@P2_WTjz zis*df*uYq0$+|W}A%fv*PO+W7)j$$*0yxclNoAGze4|C40SWO|zL_sn14662+dqeW zrRz8QZu&w=8BRTp?8<)~3+P^k=gR>}MBfm@#q(Jod~{-e#x4BRaVDEFt+0 z#a*pGLAq9&y%aGR9@ncQrKiew*0m=7lyOINGG-J#_ODfLoUV3yt^Nq;BoODr|0D}^ z_nViIV&|xKzpmn8R_JAVhINcqEt;uTVXNf7y=)+s?1C`N=bZO{2f{|>%s~5ir2cPY z6Iono#&u)U&MD<*{)Ib!MRB8C)KAHBp>5$B*p_!SPs$Ztfcvhv=>jbE&DE`fMuO#6 zwi(Uaa*=Xf_v?bW;u$#;HrZ~3H5!TqI0k7TX5D`g46D!|n`*&Zhz+mVY&hX*@lR*W zbw|QD%IXR**))=W!VemS)Sp>7Opd)wE2M1}5jaFEvXm&XZ?VfT5|uT^%IbLM+I*v< z1&7oMa*ES6t%u2`WUJ+`7L&)Kp+sF|T3I@ZO)cr9OD}YHL-|)GL!pffxmctJr1z3C zx3v8n?`4OGm%iWl`e|G^^7`F#IOH*rIc~SOoD(!kQQ=8{ShB-s52fM;#UWPtjM;O|8`S%*aGvuz|COma^M&Tr+@RR!tViu1Bh=ZsuWyHeYEfe(Qow zrI;hCwiKqJcr)vz53!nY zNQ5qbmnVBTf6Wf^$99Bk4k=+~7tAiWP?2H^_-I~~A3uuyYCsdRpg%_Is-G+5(&Oh@ zR(XC6x=^T?Xa45j3ZLvjAAGZ-9qm;z)Ei@oFv{e##*b0qR9#{Rbx0k-Zy2Y^MqMEm z5sf$n`7$7-rVN;LhJoNf#6O&H+qNC&aD!`q7@$>Mq%UVt(r3$o693fmdRoq3j~=AttczB8e$8CS z5o}9<3B_9STQC@dbVMp_rtNg%<;kd}3!jPp5{cZbtqO;e?wQ{2*c~owmdQ#*_Gt!x zp~yc~Wi)PiUQU}Rfj2q6A^OYi+p9;JniVH>zBOXn4u7-hdWwKXL(Wv_xk{%=SZwnK zA&woXm0Zw$%>7&ymGEO!!zFZeM;A%axgm_P@b+dN%=ZPw;@Ppu`LdT;4`rPA*1u8Jl=m{Vb7B;!&x zTAn~w`F;!0e}4y56eG{g(V5Im@9qrK|AR@l6s_JlRx@;;UJ47u34QhT#hN{O*6qmK zYoM%IWBxbe+$%;f+-c1pDHrAx9}eRLWpkE55L*Np5pn`88+|{d$OSjQ?+A*2%^A*} za^Hopw57?~5`X3(7b~%w&ky40tlN25B1zF1Ic>1$2xQUlaOCtG2;Uj&bz91;I*t&j zaEby3(&rw2%K8~3Q9{XO2gg_2JkKlU*m=NTcz=dW<(Ca8qw@dBtK%uo;U)pio+U5x ztLAzTU2HViQ)1Z z*d1gwe>l@7yoi@H^kc=NHJ#lrc2gYUaWR#vLo zV8^3WJ66DYFZ~15vEsXcy8fF@XT_%?5k)WbKS1^y8V5CLfpD~Hnnr>;DJqPog$1NG-r?Q0*Q9M(!Vty@ft$|&Zj6xmCZyuMVoZPYc}{^-~) zhA{(IboIQVFKgotQ6nXBMC3-%YaKXH1=BwQw$wHgY4L2O`0{|-s&bSo6ZRqW-lv zXf@}h&h-^Z-aI0lQn4Q#P1igk+MTDwV~y(DH8c{%jv}HZi6>GC11x@~cY^?5E;z!cysf_}1PzHI`X~(h~;fsvacR zo(S`wluKvgs8#-qat33OaAq!L$WC2bdsBRwK?GK=2Vh-IaZx7)xVgb}=H{LK3I@w| zPER^l=w~!?Ar}gtn~pC8zOWd=y%+0tUP_7`qks1Ny1K9FW9exTj+Za4B3#tR6TJT6 zSHTD>=(2r(S~{UD#ii8XCq(3EMOAMf+E`=^->3R^L3_rkRr5kac_Z<2frGBfy?+ta zcxY-MrTwd2lJz+;TsU&#dIVzUS44 zt0*Q|AL zCypnA8t3qQxpI}){It#1>WAa|iwU%oR{AOznvzV-ESqO!QbKfjSrr=vm?GysLr4@L zuZaFNn|R-{v9A0(_Z35&gSBt~<#>ydjEeom+vh{*hPxyDgCa4yb#)=#D~_OkJlkNq z^u2F?FJt<(!qG2F)MTyU>a!A-kq4oR`SDNc(>4)0Mt5`=VvyEE>kfDlfTcKypzpGH zzRV{oYJz%;W^g65+c3r>l;_UJCT+jH{q;=#6Ib#CEQNWv zOG%5;96=l!YqmELOQ?CpiUgNOmPyA6M+WDAs%YFT?~450@Ne1@K7Ksb-5BTknO;{4 zoBw7+_mIUdr{jN_igvma z*4V%DjalIEPQvfGfw2O6Tb-km3=Esah^Ji?r^@blQ3Qy(<=A>KIG%o>-y<7~h%by-rJoSHzn6!RC zQg0c#mnzv$FL_sU!tCHPs^>NS3$YS^duSE9xr2_?eru;B7d-;jpXk!=7Iy@^9zx)h zgC1{Z6t6$^7@rJiM08s zj0y}2-pdbSzYyWw@%swK7L|(Kc-gpfTjb9y`crY9N5M1pR5ytT`YHa&BVeS;=$)_ho!g0 zn#+KLDGoc6qojPc0(ho=5m^U=H~ue(Q&Yq-g{4;&z^&<2`epR->Ba6CZglLmK&)nPX=(2Q zeqN-Pi%dg9kdSYCX4^mwbuDUqetN$9i=Y<%Shp)oFz3Mf`Z7X+uV%9k2e{q24R z;sSrO&C`4VJgLm*Kht4--ZU|F$5NOW%Kmr zpSrtI_$$aMh-3bv^wNL5+b4qfzOQfZJFZ+Q$B(;BN9qIq)-0$y0;>J(1~f}?h)DMn-qGQI@*XPpA*2%F0L@tlC~%$DPrNB8_e-znI!y^VhoX#{W0|h8SgIpu zD?DD-oH;sr*IaM^X;d-pl!1TQd*ho&tAU?T25GDA9Q|3c^7rVt?n?W$)2DwgkPYn7 zHir;Ysc8lQ;-}Fp)1jJfFc}C!W$9Q=cBjIR)vKv_)Fv(Zwr`$7{>5_0HPIC3LWZ&1!&1L(0(}#lr8Zb$5 z9z|O5hJbjg>1m^@0TF>t@B2Q~Pqgpz^N)@30m0brqB*lq=X?oM3bc~8RF>{Am^k72 znELAJqL?p;A(}hLb2FRizJ2tC+I7P*xuB(=7CFU#D*E}Yk421DI{pxV=uMDXfLmji z>Z`W?Glo*t2SB43}L`?rzJZzL|~M1`S+BL>4t4kSAW-Kb#!g2g>i;0t4-RYh z7e)GhXa22dmgj>eLK!cpcxh-`3BH-=6DlKrMaD7Hn$M)ae>tZqdJk0^|7?%8a3TBF z3+d(Pr_8g1SmrW_Bda58)Pm^m%oJlUgd7z+AaPRzazr`Ol!e#MHLD?T+vfMi!v{t=%viezpio)kVtT{F?crLr}_u zCN|b&&{wFk_zw$$If#6P{|_8^KtDZy=P09@q?dr@f_%D;79s-@QhWpwLe*~QPyc+E z+5MpQ2vFIIt(xp=|NN1-E9GrY{4Ube_XiHMag?|l6sN@5#hE;4cnsN;Ga;@uSgTF4 zqu<`~9bOXNLeH%1MoT6Rc*Iz$%>Fs+mv|q-vR6i*w!2`#ikPdEU|y)5aL&PhtNYEs zu9kr0W?29UDTvV)J~`E81LLIw?LEs==^=cNr3blu=jZxEgkwDV_7UqEwm)tkfs2Zg z;xMf(fJ+g_pf|wl@>xda6V8l=^9iuR2ES$E@Ds!|nkXq-JijXD;`|^1>3R2^2HqP@ zt0JqHY8p0_#tZn508V{bZg{YNB>x&%g;)S)#I9wmiIDCAxJvet82<9{tv8@ECgFYK zbGcFUYx#MPvkOyRETf4#!%BoR>{>?&EVnE^SfYx?kWKZ?(%#?h^(!q}eC8%Ui%_}s z1r?}M)sN^tTSf}~Y_8ZJI+|L_Pk_xL!5~ft53Tmrrg}V1Qn%>rh{kn){TVgc|HrM5 z|9kxJJro05Rf@${awYfv(RCU=d@fk&SOM0*!vy8y>c$bE*0o?fdUQCUX{^3n&8D8|1+1>pP}{l|J&OrbAy` zn4JG3K#}OrOzr&pw1NJAICKm!xOLp9&iobk`eNYvvclQtnZf7S+;ht560gX&MkjTY z;mnk(mfq{$?9b^N(6(itC>?FQ&YuA>dP8WFWTs3W@omK3z74Q?0gK80&-~>lD^(5aW4k1jM)#8@F!8uJzH$RB8KPAU~uurU2 zCwrAXVvuJ0=tx6NoH%gtb9Wk1?gPoBrtS4@cfmN_!bG{!M*;AK8X=p8~r}L*ih?oVI$ao0&kE*Ot;oyicMjlU1 zOW_h@mE@5#&VCj-&dJr6L5_Z3fL@S7b6!3dzyM(@e3~_X^}K5pRg6QNoSZ;^e}8ot z$;EXEFK`^iZDiw!eW<1~s&kdf;+#8ZzqHDtm!zKeQdHBb(eRapJFH)5_gc#t$6if= zOwb|QLgnnk0J4{eTXyzL*CEq$7pHIN`fqXXm!luFczgmsk>}5o=#CR3KJ`aO3=(;h zMw`&vx@{VNC6pAbs%J!ZRH!32Erqt{7uVhNlgRiHD2Gfw0u7JAAO>(ulPI7mtVpc# zrY*|f?Sbe|9|=i0rM6o_$oW@|$(w z*W4sV0cuGOL7_oS#9~z(I`u2Q0~&#^ z$cLzZh@m~iH|AeSl)UVcKozq2+IwlcrxgT4t1?BDc#=Dnoi;+U{k)d$69F9h(AYYQ zqIIVyY|?KSh6YY^vF(_VR8A;zN)=h>-JJsV-m?dMEvp8PunNK)*(4E1THhntklpS! z7E$rqRN1Zd>06#fD*yiegA1YYTv)^En&ds)H%DLEHjI=qDk*bQBjiB%0Ky(9K8yThu4C$R}4QcX?|6F!D8azrfXm> za~79!<2yC9Ye|?oTJFPAXnAK`vUK+d$fvb2e-~c+sTmQuzaJ1=ztO;(){tcUf!AlY zH&OZ)qH7ZFOna9Y|9<4+vK;#2?ju;`W?w8r6-`CT(%9bDBRczSmA=Yx?ht`Uc$rmc z=%vn(ak(`59V0@XlxoLJW&Jj``>mHdQeq&2s-C^C|9Qt-7H#F>9<*QZePhbnz5snR ze|Cg?Bu6W|LC|#MIpRGwb0#J5yv^(9I?u%p8E%)w-`{Rinj3g%jaq5#X#941cPf+q z{RL8UG0HIWlQx;c%2m144nEVPiTS&z#4{sErRAxW>nPewRF{iF)G=<(oEC{bNqY~J zE$WTM#+Owp@||D^^g*EniroVUCY>XGeNU{LgUY+`|i_$(EU{Q8g8?G2E_F6Z; zj}%H5*Da()Nb0I2YEvkZUFZ9T`h$y&u+qNs6s`FYa1G)KMz(Y>svKiY5qY(^9Htc} zN1uyem*kxxrwt{JiseHQvO&tt&kwgL`SdOaMR z@$D7YF)?~lG5i7L8iQV_e8=RW_~v_!dQ~(DDYYCma3lCd4e;Sg1vkK#E?6G5Hu$x_$+Gyh)Fe^?O)JRX7w zzDV~$ttYMurKaVa;-VeN0mBav)wr?x_g(L0-iToUtx{_3GY;o_Ju9Ni=j3uX7-iN~ zu*T_YJkfa(g77~^lm5r7+CylTpCQ#q1TQ}VH|3QivteUD7fj8Mxdtg;Q=B7!{;#c< z0qITDCOF2$bcS{P(~35uf7l1$N3572tn#O6N9fU0>vt}L1M;Q@`89O#)xr-KV{Z^l zHVl67uP|TT!EE|?c}$M`h7m!8Ajt;=yX5qF)?xXukCmQ1pk6nl3a;37R{irEi1&Q1 zyH3*njI+bWLu8`#jVbB`gv*I4;-YrIK75(({#vF{=IrR&yg>&4|MiE9YCNTa;$Xe_wQsvHOh|~>6l~TzQh}U1Oy&Xh`ABUq^x!K8w6vu z7f(lKW#PspdZkteiRXbFO-Rh-{ZE3s6}1nUu|UuOBRH^xLY?^SW+6seuOUitP2 zREHJA9bJn1*V5Z}e_0{a7q4n9PIwcKpZlnKV`+vJqTUne@9dK*ylegPa3n53_RCJU zP%dAt;I=9+VloO&8DKkxB(?mSp8#e+=g&ih_w zSThDMx5k!-@>h5Y#$@M*9hrEp7UWTe0mj90v@S0bwb2X8e}oYXJ$5?Dd`Djz_2Px= zy&zYiDRNj-DPFZj-8t}DjKxJlt3o?a#LkrfVkKY-df$xoY+F9KkyRpS=^zPPL{+$K zkh4K8kecbd%`4(X$`G2Rq>!f!-`0tLPY9PFNb{+(o#VWI)@&5w)Ne=fE>6Xxfytkhh~ zO>fD^EP6JKUfCiCn86)`T})8ic}hyFtu4GC$wkyNe;NnXhA7Eo(O{4|-BL!ZTNdrK z!ecj`ZakH>XUI2P;ZyPT6T{$43)q;a4oOiKXkf9#G5Yx^tbgs4F>)%u&i!6}UR}rD zVM7?H$-GPf{zHsd0cH*!OwCAlx9Ug%Q6c|sxJ=2HwPq#kvFi_RcNm;DNa|62aQLx# zJG>RLe<5rUuc8>&U2SAyuvjn0K{X0T; =|LbvWUP24Q&>a(|P?IxgZfEU_!zK+m zf38(IGtO?WzENc{q3dY7Cswe&c52G}l+Rohxqqu49yFM+w6e@2OB5}Q&cApKq3L1| z4a7NO2&pbQed$VjSP8d9z(Gw-NtkzvxWqKFe_gIp_fDy>_2a501rota8~BqK`2;#! z0L_^>Eb*KLBEA6x@1J%N?D?v4*pzF}UtYN>AtNUtK$l%|PMMAXNw_IXTV;h%9eWcn zd?zLtvtzwnae^p40x2%I0s~coxMv4OSdbn^ymU|W&ok-B3Q}$-zcXZg;M?+-&4c_h ze@a1)V>4+Sr9(_y1HV;K(wT^?qr3?>UJbJkiq0BOs5W$b+G!km%RfYfM)3u~H>$_^ zDa;;GrENp;NLAE{AEtfCW5D{B@>_?VZy6B{oCismMK?9-z)ggZhB*e45+cfna;Kbf zs9k{dqL;a_WkPI6PBLJtrpx7>Bo`w2f8wi&vwA31V2<5fLrY^@q>97(PIEnT{HjX`+$v_ULzD7oB3wBgSdr(SS-G`56CJ-a*DEOe|Z}& zhpwfTr~+X0m^_2A(x=c;S(8S^*CVtLs{pZ#vm<-Rr1y%k*sli96m9kufOOLy3MYgv&D`ug;wWI5#oQc!lp+^bO7Xk1shHd!2EL?00 zp5#y`Y*@9Ke(sW)Uv(HK&eaFMYC5tZ!p5-aKl+}e^iVU){Qjj z1!w3NHkI~DeoAd$BuJCprVJh73~l%C7MuITrYb<}xh*an%$aZzqT<0|mXjVNLyl`a zg=Zqb7@#C&rm(@vh95JMw8QoT*uo8zpRVfwcYYoR`yW6c@O?ALZ%PxEdOD{TI&^79 zF=rU~LZqV7pw-RpHKs#kf7M|l5q=j~%B`pOM)}amv7OIRC~#ukU=K!(79CRgM+b-t zM)fwIpr+Y#1ZUR0Rt@DZ+n8gaPa!h%?|(0E8t7H!FQ|%IM4)|qIoJJK5Z3)#;^#R* zhCLiH_th|(rJY;bOi6&b6mQ4N>(Qqi7+ePnnzLCI`k%)HZ8(w|f0-}HqT=4R@j)%Q z#8_iM7sd=o+EGj~#|73URDLB(*`Z}(;ta&TQT_Gv^-Du<@tSo+2;lPWNb63rbimq* zu9I3$HLPnBQf+6>Q_%S>#ksDJlRIYx`FWo~l}eARtztr-BLy3wa4W%$b@~A#pr&^P z%}PnnT@ju3W!TE;f6l9&JHNda{z~~Q&?6joU7Tb_^Ti8x{kPUl?<3pCEa*Sgr@k** zl=#8Lh$&8vf*%yZDX5LVN;94JrLV1Hwde&^fY`b?>CEJC50k`=ChxrVRs5%;OhJmx zyA#9ku(CxM)IKBSyX}%)aH6;IOxMv~gOQJ7A6woZ?Hqd(`>)>Qw5c6Fbx)P=iVsFMhVY=zBI5 zlaaMb@Q$2`ZDW^{h%dr{;&5dFjQ!Kf@4I;G|GleN!C=_Jc&+ zX-doae>GK2BOf0zdJ`GeMY*H_1bX6t_gQv;3xo|sL!~+< z5E4AjW~!z*qOIqUKR8{^^v_S+tMVL%{c3CrC-@KYO{!_JIh)ou+!{u3oD84M7jqc) zhgKD)HMV;B(57w;j^l44!L@cW{=fLV@vmQpe;HO-v3(mi3eE4WL-ckvb>dqa4@>8m*dfO6ndnbR_NcnNH z48a>THtrXe;J$jF5f=GH^N{zKkbChyG^?Wsc>9e*wj`V)I}^b`@XL#I(Q=Q^{C9(e zf0Tl;tL=jB<8}gV5~ceFRMV!GAEFx#{oK1K^-UGZ42<@TDZz>CMw`I5D}Ex_f!1^V z>30ne6ElujQ~+nLOYD6l=*aYdUAK=e>DYZB11zo5Lx z4_cX_u8ZqQ#1tk9`4CZ!d^^)Yto6_ze_X?L=Jp66@?en|4gYq49X){KGku@AqNx7b zDd2u3Lg?rDrnSSOUoaCd3OKS4X=_xdCb&;mX{EIaJSDEWY?_Z!`K|a&spQ>i)8@Ti z9$=O1(ko2_T4$geKYvjur$JX)KBTXw#Z_zA!&4P9nXEIU*yps}I`Yq)CfpGW%=``m?K_o7WXJ-kMZX~6eBu|ST64w=XAH&8 ziXgr`W}1k|jZ>d-`guGf#EhMLe~Nne2vD&zYe+ZGy{_77X+~u>{7@>h9m}Q1i8+C7 zr@kzMmon$9yt>*O$OP>DtY;QlENCJQm%Sb}#k#SMot?0+DH{01`ZC^1PZjT|X{Mrd zAG$K5U>@h_RI(GOb}_JpB3CX&E2sKAp8vK#h+A0F+WtO{nZ>SHf0&>Pi7Yl* zdvsA@-Mh1X2@jA#Lkp)uh}+h0xv7J_zzC&B3ces1Yshi)SeK>wg!B7r$Ed6H4!IZ! zU@17BLEgA<$bFgS65527Q-x4Lfo-w1LT{?M`AckF&A>ZW7o?-oP)EJ4$$(VsYfV&k z_Y%LhXg9<{35!J1^Pikg?zREZwM@{jS?(Kwa}8+4}p5 z92p;?6myE9r3Ls^HGa&7!V4`ram9FkN{(Dq6S!wAIX>(^0{Wlz!u`l?>_BMHmsBnb zrWO>d`IRP8FHf}UgOw*M#`%A$+v&%GitLQt%F@RohJh~1NOGCee_V9aj!7QBQhl3I zuCnXEkI$Eq?*%mlw>Ty?6lAT4r8SZLoP@Q6KD|02KS!Nj=q(Mf6=~Xci50?_nWau> zQ+l^wSgN^&BYKjHa^eVX^ZlSMLsG#KEDeyEp4@0X0yR)voWs-92azlHUC-s+v(gU+ zq(jC1wPhANvA`o>e^nt{5oqQxAoi|b><2zV#yX;MO4sF=;A@P1LH8=1?q6b;Z)*lg zIH%dqX_(LmYKv^w7E}7yMme`XTb@K;;v7+%p+%hE2zj`dv!EK-lC%*BjInd7Bg$AoRWzk$w|cw;Xfp`&f8mXO$-M{VSovH`iHiNss>xF) zCmaKEd`|+x800BClFx(DQKun=(TJi>Ul7n|xcxAj!@j6+9|r_75bo{=7O91-@1AgT zW_UK7QB#HgXh7JC>p3s6Ju|eSRuDL%OJ&oujJEbF5+#^%`XJ%#@8UEivmIYR>2A8| zviIaXfA`4+zY~bZ`@-f29|!V+mTz?vfiOvIOu12CyS!oF6>_NUJ^vk(!oiqw|1a5> zTu_338cNReu}nkz)VXO%yI(R$Xq>XViKzUP$$R-4 zJx+HJ>+cLu+`SimHlqYBZ*6uAKjqi5Hc4cNf83C?e;uM{^bfy{tPP@9X^z5Qtd|6n z>aCG}jrDB%1<8ShUaAZM6Xn~44MeZDJY1j{ewx0ac+_c%}T znrBEfEpY{y6QZNDlatKui;7oPR_?R5!sT>1!>(I?4GTp9?HiFP3@>N$uiu4BeZM>h zVH=$%hckLe#FsF%od$$fpH_Yj_1YlLe-9G>r7yj-JjkvrgwYCGT4r6TDngcaaZ=tJ zF!p^aHemc~2UU-b zH~sKb6-=52wLO0Xx;ZXgNPV(pNjc7}A@~U92=6|Kp?6YAZWSM|L=~3uYXvS-^C2EEZ=HMv47`2}jqs!$CYh&kr2sUlMeQF%BpHh$}Mn z&Nk|_sDD2}7MH4ZJNkkw0L;>oT~g}U2w~l7q`jLSZo)t-v10TkSgIy3+L2+CTA?4gOab-Ix4J#s+(xe$eMvm^M1mehu3X)?n|e; z2QUNuz@>oU64j+$iabAGxHdNI|NIEN^w+q*cmx(JN~&Z$R^uuR$1ZK@!}rRtiN?`y z{B01)eLWr)z>V2_>G&#KeT7oIR~OqWY_Ei?UR(vbU(&#Iz@M3`>m|w; z7<@p#8Sl4wkPcGoq|NJU)2xL*9E6z+JOZ=g$h=oxtHig=h)D!X;Xm#`U3cPrIt_DF z-@ppk>a#yE@^DAKh;-_&e6_Psx}Q=3hmm&}UcqI*%;idNJgHATe=Oa+iMer0?A{Jo zflQCFKDGUQDi(o8b5UvX1kPqWa3HG-e5vD(`K10{#I)&lD}~B)!fs>=`cp2gz8O?# z=_9A7Y4zS^{xlb-rCCV{<+#(@jE}vX7~_&Y=T}u#^O5=R!O>H@2Fw=w0E|E%e-yxO zO|Qqt=5o3esuSm|f9TKr^;_2NugnhXSKRKD5|BHl(VM=bsnIA(*mOC+us+Xw>s`Pzf4(zxGD%P2l?e;);SC$fS2)KJC6nQkBv zQ5w(7@v4uc@8@MS($KdixlPl-9bpSf`RUe>GsT%D&AP}5{9)wMPH((r^@fTv6Hduu zXq#cov|^Jg10wG8caCiH-cV!B!SmtteY0?8?>Qn(YKAVvnh$cwXV9fehG~K zs0s9`-#^g*@O?<8Xm@nonY*0hV%8sO%H5YC7h_7-e|%V(E{rL#A#OsH2*EhNV?g2^ z(QK>(E7;WEX61>CwI2LbSU)}R_pJR+OL%HFW#swe>he#qdC-O50e{N$&G5Sf)XAyn z@B#U`>8+U(y;q$Q&uZ_N{kyJmv4~bTB|BSEHWqiHyq4*rEra(02SG8so3b5Eej_YT zhSzV&fBXftRF0J7yEkO3kFss90*=-VPhg4=(6JNPYc?Cnkzy>^AlzbK3h>|=u=AW+ z`JFyZ9Ii1uMA7O+7LdF!Pg7xFhg~J%VMY9kp)T`72}+0IIWayJ-vz0x)1hUE6cN5kI>jo~Ya*JALtE;K(f0>0C6ZKH1joz2z8VBBMwHE7Pi}o{% zyQ;eSPrM4^IJ_|GAm?+uT331ffXq)hUG+Nva+q#u)^V59=3EhYQENCh`=xV9Mpotb zRa1gB&4RVqF3ji=P(}e*U9*W|oSylstECmiN8jM^QzXyMzr=;!iMX(sqh*h4uNHG8 zf0n=zx${+qBi*@g1T%{Tu`vS0MNXNa&1GusvJqqA78Z3Wi_1Ceyb#6IP5D=W#V=RF(24pwkr=P zCEaU0I}~ww1oo@K{S^E%ENq!@S)xr{;!28{=?!X{bA~!;sY7;VxLZO(lH_Oq<}b>1+QsLca79mI5l-Pf+s30Qx_i#m?m0Vq@I3h6w_FO z^=UzheLl#Z5|ONpMtpI40vnFx&{@x#LQ+YPs zNK!muBdn)mSz2djeZaXph{=BKluhb9?pjC2*%q1ivt6lhdlL;A$@jUVf8A7xD58RQ zF57qIPlRzLj0Eqe&U0}Lda6uoTBTsf)%E@;a_ctF4_{q(*jm41S@|u(q=LIN^3166 znC;L-_KF0hbkTCc$Mc>H%S`RF&-&UJ!{M;NS(75d(ShbQ_ai zdl-rCUn2_@b-%^&43eR}f1URdC!%=u=iC3y8U&)6JIQN!RLgoC6l^k)` zs0b+E;E2gb$h4<#ymWSPrp3l!pbJMBAX7y(iD2E})0wA5+%V6*!L>`(mWnWhIMD#Y z(Srjs@5MkWxf0y3&c>ycAc?KvoFb)jROvLD{#<9<&0T#{e%XY+5#~R_wB;J$l=wvaK=^)aL&$=H| z8tni#D#>G^04@7ov6GXM!1?Y+fJ66Ywr>U2(Cn>`PN=1&L-dZmLkJ;i4}8n~xk7il zo;YQtH%i21e?b)$w07q^@~|1mwCzyh=Zv0KupUx_24RiLR+7<}aa^5drVma-_j>;x zRbZ)W=)CMagD|ycqkmh`swafVG&(7k0ZB-1CbzaWmf_Ip2@@vgK+p3v(d^_`iH9D| zmLaRURA~Y_Mg{Nh883aq9+EH?O!VdGQ%?(e`4EKye{-h6?n4(sWtQ2;uc?uGZDhtS zxDk@gobfy8+LOCpvA6nvLP}5VjjRPHLxX7LW{{5NDT`D}BslF;Pg^((j-b7!6hb7z zQpk`VhnMhZpiD^mJ-a^xVn;)#{mh~JqS!14ThBb@L;0R{ePeS2Z({E0U_Tnsjj#DP zk5~WoZTGe zs!Ns+X^2j5{6&-9db+Kr2{G%bsx)voPA`944iFfDmjgUtVd2gN2@t1m^xF#aA;l!M9GVAgbZR{UVsouFNBMQ%Znk}vv_6N9)Y-#&>bfc%|`(J@(L~pw5zh*JX}iH+hYpS*|i$6tdUA{u{zPAy(y>% zNTZ%KqCuV|c2l6@Ix{~2I^CJ_`xpo9f8s$aE!SuTHYaKpyxSO`C|sMO&M!cbT!iJ7 zK3;--_^Y09ZaKfHhwvuABpG;BgsdYM~Zi!w`q}y4xbF1 zma>_qwaM8E77)$7LJ7K-!Cot}+vl|pw4%%7U*#A5=$?z?rZRJMSh?94)xAk7e=QsL zlTzRFqoG}@u_oLVnIg12ph#FqN6O$?=lb8;<2U0v-u<&S{Xe?iGwUd%YBZhEcSvAx z+%E_AqO}9&p>FwhUOHR`+$V;->(p4u6gj)(%sR@=)31-yk0;GHok)dwMYEXd?0l2M zW@=ZL+9s238^hd!_v}a+*ql9be~#V-Exr9zk4j`MS_lu1fsy1E2v3>^jnn+o;an?J zJ$B$B*?0I#X6C21cmGl$5kmOfz$y1qmZUo5r^$G53D}gEC^h+JmOFM+YxYN{foDS$ z!iX&U*dB*CfwL3j;ThlU+h>NXEkNYwR^%YJxPfc+{i~reeIo7tbt@u{f7-a?gmV!2 z8|dkXI!Rx2zPN!vX9Yi6=C`SsB(n+#xbluP11Gh6W8}W zPBFoKDNxJOeLU4^!_m?bL0TmYVzN-)E0cqVQ<6&?$W%hn_+cti^gAEaC9d`3M3o-~ z5Zd$fvtP^`6c|~a%%l^+f4mOs`iMb=UbWP8{OINm4MNLSoZY?+QdURH)FBu_IMT9p zDd;I7mA@cH-|Xw`ws^bvU1(hxU(D|Br02($B$<$m6GUuiyO9;!CqU}+n7TJbz&PA;A9OClV^&IIKV*4l1iItmWk z<|Elab6zq0N^q``fB$IVFbbUia8XgRJAO-edGn!ymmLX<`a$$d2T@~FLtPr1T0(U1 zT@UdOS1#yRa#j~8cBM*%-3vAZn{CsMMS8K}R+1m}u&N^-(%VvMuD)H)!B1B$nNJyz zH2R*F8_1N(oQ@$?&0U_CKhQQ|(?#P%p9rk(Wf>54;TwY2efB~fRwVvM`odw*sbt25hYtUDy0pT0fyd(RICK9wHAUwuU~oTY286e#YV zp`Dgb_BXE5e`S@Brp)3?m(*@%YG()I$HSgVaegr+W$HoxFaWk*Gg!5~$+>L*^3F2B zbA;q#v*|7ja9D#iK%)jMBB)}8c(#o>y-f^bB89?>3h{rv_l?O^(%9oaxe|RcjS3>Ch!W}s(MRcE<9)XX> z-p5+W1yNE6;W!+9^3bEre)EuSGcobMP3(W#=V1Vh1m9Yd7O^$q#J@UwgQ1c>@ z^b`Yjf83W}?eNEmGPx#!%dP6b6OfV4t=OUUt;)T!OzCa|hn`@emv(x+VExnjlga=C zCJ*w!{;Wz#m(m4;mG5}0q@S=@ztg~@2mF>cKEfDBYYXlKe5j~WS+oXs#;2FRS`|xn zVl0=A<_Wmd+qbt^9j&dHWQR5DLoqscYs%b7f2ALKZ|SBvomY!=O@G|mAe!Ll_o1fZ zN1S};U^`e|l*5Ysgfw92r(H^jjO3)1)nfNhl*{zz5t(3}BD>OI?if2PqHX zO|Qi_m8O?fRn2})uerz6zzVX_Q;IDDrIg_!7-Ytl$f9BkO6T#`4OOM zHM!Zyxj_eE6cp3Qlo-&2RMt{a{gMVMW`cg?c1X=tvz0WatS8*npU!)BHB&~&e|uK@ z&gV0N9I9xV!g`p$o^&1c*(j0k#FdWuIX=l=-9?!dJ*~Mc3oj{K8zb{1oIv&&8RGn5 zF^OSnG`l2E2zym_g@B@y-6^39jV*6XcdzNO;)pZ5PV+3^>|ECYFWHCkfbQ^oHWokG z45;G>;&zuC@9VTAXEw}fyB4b+e_Io6k_8DRO{yYUHK$NMp^XZ#ASw3$G;j*KXPYJ! zLeKjoJ{+>T6>Bd#uVP<~Q}nT`zmFrFh$=LpPqTjZb^UL`;mdKIL=Lb%&u|Sn1Cr;2 zQbSFmPQ&o0Sb4Y{5bXh*dMmHKbmkdHJ!3Sy0niAli^%1r5;xVZg?nzBf10jeoR()3 zhuHfOKXbPk`V+o^}`=>Q2^oY7PDOD7;l^f06#FG3iLPcr^P1Imo zg>yfPBYSKuIG6F_(%zT*f4hm}%pE08^duTc%jtF_v{vy&wKIGZ6vP$}#!bm+iR~e0 zDz0$$U8VG<9latlJR(>A;SG4+sM|HXc;U;!@uoSEDyHTn6AMoq?^UhCC7fJpYSV3M z>w&L|{tA-(RN8LDV1 zhJDp{F*+s}-vcl8VFy`_Ayy}>quWH%lR>1EjAE~W^IQy1!V`rJgGEC`3(Urac!9DH zvY$!yj7kyTW{HcgxZG&oxIcf@9LQY2B)U>5jj&l;(5tE>f6UvB|Mz$FXwyN0Oe1rh zThITLZyvML5_z*yRr#(#88Z**AeIpU_4A+l&CWMF;7>&ceXlwLg@P0nd2g-Zo36vn zT#+2Zy_8`~ZKFXtc&*G&#BHmpL;)%tqUNmlc0_AR(aso*7CB{+(tjlWjAftePuL*2 z;nNvtZBJd-e>=jLV7{+D_Wv;a!a-6Y2nQCwH#ML031fR+cCg>w* zLiiU8f1f;sB1{k+9T|}Uk*XBz1=p%*Ts4R@X)7Q3-ASzXOqBTgy|my~RirdT zN|&Y6GGvmJPhMP$5|7k^K*j&c&|;N-!shU*@$*kX0s_w{prBZS4)lMt_ttT7by?cz zf58)igap?F2?Ps};2r`5cP%WqyAwPl1cx92g1c)KE`dNHK?|+IT>=4u1S@Xk?S6Z@ zXS%1S@4erh?|$?CvCcVE{fe{qI!m5)_Fikf;X69$3!>KA`6vB+%M8;C8qCZ~#}DiI zRenU&NdvEJ(mcNE@azahvKh$p1$B66e|nO%NE6>;`K$dW|A}nQbAsYHzv_qD1xVTerR}xvtxS6GHm;a|Epnv!uP3@BT$;Kp-m+n)SZi+Z# zIgCmFhfGP5{pbNiQyuR3>N@^-t^2&tp5xjC_>p!-@1*8sY5Qc`pMH{h5f)lcN*8$cx4#$S+6|Pbpmy1_1Rv@68H(wf}gx|CCH^`1nb^ z$Z8dV`|^g|+aEygs{g4BsT1|$5Z9~X@Iy;%SuZlh_KP40AV!c|8E7PVSo6B>vAmvrI3C2i7*5(D@UR#F7ph?q-DHhM5sf4!E%6UaHn z-lfRg4S=KQStc8thRs2u+*py)_qPzIHn4#b469cgz`I;8|ZyVN^YM&h#s_@q9%{xli z#|R!@mF}U$Uz1n3yJ)80f3vJ(5K1Ey$AW%MqRkd}D(Afl^fEs1Tz#fz*zIrMplW4i zMX{RG1L)^Q6u1OpfxqyLTfJU;)43V}v`A~UHS#4V%mud$s?ThXC-kl@==HekIDO1s z9SojPGf1yX30)o3)H-MJy z=-*~5Gpx)jQsvNH5l2H&6MX>p6 z#goiBzr2l3O(-J^u7?@OZGe%Ph*tIw6U0f9ifkWU*cC@$UW&Zs9x*Tvlm{mN@)SxJ z=BO0p93|%0OYGgZf91?vo!D2J*&4p8x@=2Cy8lZ#tEFB?^HSDFdc91EX0~ZSleCoM zcq>usZte3t*BIz{;a|yjfnOk22*b!HVN7zGA;Vui(%D{4C>PnoYBAyKTm-&o5|s6S z&1)^cvX9^P7B=j^Nq&fUeV^&eW062lX*{tvs+jfv!4zs%e|%qR&a|_xYn}_5rFPps zSH*(=Wsv3bVt(geI~?CUC#n9p&w?TU+jxeO{Z1xKLs+ChJnQ+#;+rG||6I!eEdFbXi_2pUVoQac&=_kZ8c4*;+H}zIdEFHnWhg>Pq(>lD zdN=j#*S#8Nf8kM{Jrevc?b59u(6Tu0m~F8_QoWtNnYFOh1$ZBSh8}0%IsSddnoqwb zABY_J&rH-nvkT#KvR;lUT$^T5QKMK9zb4I^baG7TPG{XXJQv(qmiVU>Xl+f(_vxCA zFbMEKOSo?Of#;Q-_CI*A#6)T1a%;!2+*|-cG#NLq@ zch$1awB^07?S6>Yb_0yq&if&(_Y?Zl&xU!3e=$Y_LJPRb8yZrkePd6R)4Dh!Q!RR3 zzOlCTPu6oaI5<1*9bravV`9u{9G`-mh>Siz!ROLeaTXOHv29wB%ZrptqNDhLn>Vo* z$O@FLckXs_X~qq} zK%0Zy34ADl7hx!?F!`XaIcx9k+}h(;e<8PS0Q#)5g+&ncL|7<6VXF z8=%iv;Rf*UIMsi01FV_DZh-l~HEZ-cGahX`7{wglv+A(M647Fvj@XJIsg4e-jYgOe zRCH>Z9v$bHD8DuRRC%mpjl>QWeNCf`#P)`Qw~>#4?nKQ6?)sjF?v&Z3O%q={fA=+< zelu2M9CG5F>A<-nfIO56J}!2@y^@l5`o@Ad;&m#FxLXt4^SG*x^-!$~9hE zTI{Tu@seD(PLm=HTYIs08}Saee|2ms&Q!+ZhM#mzrEu7%u(=N2b^BQJd37r5^)$nm z(q!)+wD{SXk>7u6E#8j5{tzoi9>^+%Vyf;9P~~gNW>i^|C-Ve3efn`CmYYIi=@6b-b%c^_5+(ffmjM>I1cDe*qmLf!4{F z1Bp@LUD0`ntBBDt@a-zI1ge5NqAm8X-A{StIf$5}EhAN%%wNU5$SCcNwz6V_LU z`}XBF|6OMtu?e&`gE9kK%C=(&gy%YQVQ$X~gRueZp*8T+C7Ghi-h4$X5h5xp4-8Cy zPy>gsfl9>2?!PQgFvBFee`3DSF)J2b^DKA&`Q?KSO1XK2h(pvdI-??pfXJ#H1FSe^c)X_L)QRgTyo3 zDdQ#!l7;=niDVW7G(uxd#7CAjML2rE17~FT(51&tkJ{TNhm)MRA<+yAU-LE`Gg#2u zR&9On#_a7!e`N}3=7Ji_PfqhvxbO0MbGOUf%0D)U0e}7ZPN|`$PRD0GP1}U*tNr03 zdh`4Xn-P1+#d|cSe+Qi2n-CTZj^k@z%*z*K?e2M5?8UhLVK$39I4ZFAX&zG)h`!E) z*Ki&qhG=K^>o1|V?~bG}Gth=MY^&|o_6II+J>lCsPnsc^Bf8o`AGf)nY zVZh}>=?CjVx{q5ie8i6X$T8;;p9Kw}DQl}fy$aW2#vvLEvzM%Mgme zOmRvKTy!5YUj;jgFNt5=tr+;LKk_j(KtnFsT@OEYL$H`Jif6}5yiLb9zd6tQ7qnKT zp9!p%&;;twf4s4hk`UjV3EUz9bz{0dDJ+jJw- ztB$iVf17Hj&BDHGpPfbU`$)V7k2Ws?>~aIp9tAcBCYgF0 zBPGWCLFc15lARR1{hv9DD*G8BuhT{WXZH(VlO7J4V${l35~?P8ztGj47` zl|zE)!N}>UiG@3%HfRFTn&xrBX=5<+NruLRK~-AQ;sHf6w0A|+YRj`L>2hymWXz~^ zeTnqh{Or#hOruj(9W8(Zu{IWrOqgZFf1>83WoigLPlC#-cDZHyq+9m~gTVCl-T2z2d)I&_P(FXfV+%0z{=?#sd1^g7g8{jc*wmW^$!wJssR+IN%*2aOZ z(q|7p9HSS6AoOZmqI?$Z8UpA%(NtViI_kzxgmau2_L`$zajCgTQWgmV%Rc_9E1Vmk;w5f7 zeQR^WTPnYM+BTr5Ko42*@p8nD-e_+wl*zRJ2JrDldL$JZ1lsM?%%6blJaDehcCYOo zSz70&cb9aFj^05zMnwGe7Ymw*( zcJ|7xRuvW4*}bf>{?HqWTN8v+-J?4mxJLHNxqdju1@dwi<_FG5$=04teZlXJ?wqu# z^uSlY0MQ6ZV)9+dJ>pUD#Wmzd%5X?+W%X$*XBE9aVMEw(T7*c)RqG-gf0z7z0gG`P zFvq{hZfo0sdY@UR^H}?U|IBjAw!jwzzU4n(cZY!uK$&A@>{Ze?lGuF`DB5LaQl( z?J_-!gWy)MER{KC&>su)bL`yLA{J5+kjj5dLzF(j-y#S8i z;a>0K;3hxb#hKQ5e_qT4E-JSmF25fUz5mgvrfW52DOZCbHTD#^FKYQ;MOlsKG>xeX z+b~Z&u}Y**khFiYE!71MrLkAI27duPvyRNfsge&> zdcSe-DIR(j{+ckSq7B`osb)$?4!h0z8{jyW(*BMpe z?PI6k7gxoM#u~P_FMrg%Kb)zUd3T>Vf8`IzS~lcQrAyh^r{ex}+5?!(85kJ+4DjL1 zI%C4(4wRcQ8s7EoqqcJcBcRJ%zg3n-JXEJ}I^HVS4wLq3Mmq z6fef8`AQI}sJLNc>_}r9*GnIw(&(~1ooB-5-%S#P%YCNy&3{aVsJ%*^w|*FO)-5h7 zA`-c8AvDbcv~a>+7hGDuyVgxvzX7~H1tbC>xe~hsB}N@%iwNc!B5Y912KcyVlHP{n z1f5A5n!?h_F4|*t16bI=-^BVC%st~&rpOhs+E&P)Uwo9u7w`8VeZX+=0lC0x==mt- za|bD%jz&mB<9{W)_D~%z&k#bqz|XKI?z$$Q0{Z=Gq0`hrNy66QXyzM&8)nD%&d z&og{A%LPe75dPwDPS}vJ*g%{tRo5V;EgW9*Vi*^fE^M$oUIE8oMK!dN$L!G z)~2VMo5m|P_{Q@6W)oiz!3qGKRtJ^mMDuN9O>e4oWPiwaw{>32`G`h;(g4*MFBA*x zb9rC0;d~KtXp2s38DSr<1KQ23Vt$g8taLXh1=oC1Yz^#G0+dc^T(`x>r!!wf4}2** zs>&8emNw+>@GSH&@+&}A<{<)fBgYlig)!F2-r+3W4=#0hv#T2?z<2sOCZC3>3sGAJ zJ>G7h?0+q^gv10sHjCR&0Y16o;w)<;$BMTa;(Gk};SNQ@Qt=^$c?b1}`z?^Z1tyMI z$E*T{6;D0wsUE>(xy0ZtArzSID7Pge?WGdFi{JCPYcn@^p;*bHa`w67YTJ)g?2YJ@ z^#`7GYg+yRcReBnnc{PzCkl@RSOeOcBU(7$#DDjWD+Ia{$;dtF6=ww|>f8C7T^A6A z8y0=aSzN1h8RpWjD(9KAJiHy|j-!cR*-9L2Q{EXW68{vz^XA9IDnv+A8?Mx|*KB~2 zoCVXj2r37)cvR#we-F8`ZQ0QJIc@gzDZ#s|m74PWDPBQNhup@a1VLr}&dYioSxZIfd(9Ry3 zoe>ycnHz+=e#QEHyY8to=FnA`2Y*SQw=Z9cbn|H*4ih) zKJ{Bgd}9*X8CgEHSmwLF9&NF+9sZ1UB6-(lR_nBjNr1clMviErbGKbs!tyDOW8`-& zBd8G=-OsP1zLCM*9#9u;V^h&k;#=B7|rX8vkGS#jJ=Ra`hR?t`t6!amTih;ebd`1ih?E^2JaMmbCV|+?lC)*-LEu~g8To0q_Bcl_|5e}C52V@fJif89GL z(T56jvhfVPR8y4ebwa7uoP#nWGjD(bb$@ZAV6ic*?-?wcgRJOTuk@Qg!ZOjEisSC7 zqi@FenO0Z+oe)ie%IhK+cKjXu_2UNs^9Ha!EzgIYoUdNKXjst9H+(8~b#bixTSb1OpTgUWkHL>t2yi#yrCW1$lxdkQ(-)&SY6fTT zpK!8QJWQUC$P$!5G+6x{$15LLe*$;Iu`0NNWMj<;j<26!q#w58L) ze_OO-8=s@@8@uog@JYj+tt|3`*Je{7r)gPc_jvYb#~M><;Pk62cr5-&`wdX_HXB#p zzPY`)PNIQUU4QoR*N=l{wZXkWS#klotGPj^k<0)MzxBuPa1~A6I z3ZqQ9ilX!yL-`#QHxK8`hhVS{J#+MXqc1^=%UVv3CvTD_LGQ z0E1H2wTsi*1l>6$^xAI}3)ByGAb45mgPbKUve8Bhuzz=nNIso5fHe3^7_F~T!MPe0 z39|jgl9(gJC*Xn89!MDoukkt7eBf`DldaLChprN~@Qa^&3P$nML|bpyiZuJT)>7BW zhrQCYJymPzIxcSwEN)r)ot=dwmGO!08xD+$D_)T6h!B`P%KUw=`t zcfFT*-+#x3I_q_iXVnAN*uZUd>-&rsg$3+ZtX~_&<|dl|t+E)Fas4hIS+^WdWT?=_ zV{$X65{7nQTl)KH<@@!8w%Mj|lutGAI4t?LF_pWw* z_j zGWZ5~=nh%pets>P1jejwD11t<7qTY{X&z*Y$S z4buF&!zyHM&xShzG^&34gobjTyvp)YseZ4To7}>2GNoX<%df1(hSM@cca7QJp@CtV zB+n&xR_2gt{dld8ki+ovAfYk~y1(b?(fCmh2f^#!S9RHDN5NUcIW6F6_9efQwtpX# z+)a6K^d*lTg^IvF(4XIq!Y4Y`Xnt)nF}?vLvvI=;C*2!28WIt8L0^hT9A#}kAAk!6 zK}-X4wO^<=*qyykD%voR5^;gefvKgL_ZcpkX%o1RnrODyypicOPS#xub9kxVj9C(F zqH7cGpcaqJrMb_-_QG#*vuxekPJc^E=k#Qqr$!O0ZDpZdj#?zts9ABtnw}s80Kyt1> zlyD{1evgpgjuWOEXcpRVIBsF{SLc~d7<3_LQk$W@1ubx&KnpJ0AyXS6NG~$myZ@u{ zSnH(Ji?n$m<92Jr!JLm72Y=NQOoGOOW(&xnXe-P)p{GV}JJTC!oZ)19zsJGi*?1iB zLzErG1%{Irb;c7`#?73X6sYlzF-i6oYF~5PxI36Old;p|EjB@EGwh@!->;041yUu= z_!G(--wEua6+cTqAH|V2kPUa>3|O>A`3idlu3L>afVTIV@3fOaM}N3?6YS$L8OC|U zUCJj;qL3sPnlHVcGn+2tt4{KrV-+XD^E^y%SHD$5`%o;9v}(BZvsWj=VSfJlG#hT2wnFj-gS~P1>5PU3 zQl{b82F`=>M*A1=^?$wkhI@z>n(cMSe0SAqumoA*2hGtm!F^7#N!za^M_hxME6wn= zs%HuUx#&owbCd+yViZzqlKz{tU=w26fCgOBGE=JBeJ@&~}ck{PQD;*dnXYv6|t1cVrz z)s*|hC~FN+YYMTeO0G~9JA7%d7|ym((#&cDSRI?AJ5|+gzXH?tfn*ERZC-igrbr^hRW&cP+m^ zJI`^9v{|eo(X};NNzv{P=;?c>){>WfQ7mxLj=lN! zS6lmLE?qb|BWvCD3pG)eEl@;!k=!H_ZNoq7VCs;QXkJ6&`gLb2rEBE+`4W!0XXBxmvD+ql$*lB>wC8K> zs=z6U(b^8290Y0P_dG|JeieeY|MqanP?B<1Z#Hs(g!tWB`NA>&w2{y=P0f0){i zt9TOd)VM8~hS5;3eZ0rHdldF<&qJ0fSy%Uw5^|(whem$X?_pLJEt!F3WM06xPvTPs$1Qtb+6|>Mfnme762Z_ zy16M-au^ISc$Vg#OjkmADbl79Vw6nK+yD#4n}#%QnU7I= zi-J+D9m5}^5l%nPVy}CUT7$bI?7oo9QGXU2F_Y{A9%r0oL+OSqKj?2MZku4^bAs)T z16Cm7XxgAAxN7`z$hvpqc{V4Gfp+5A2;ZCOgR7`lbKEnl7g|FaNXZoAlb##k_`WVz zU|(+FZe-w72{IkEM~Klb&#w+R)6Bn zLfk`cV0L>*6WL|weRWNHxNP94#f%#DPB#7}X^WpGf8Ra(*PUi`1^Ros7bI8BF`3Mt zfTbjXEtdrdp}d5PuhcGOTh`UYq)tr; zKO1#=!xOWJP{Uu_6jOd}Q!E`TMX}dQ=>`Z^MYla`n7XJaC$$nYjjz6*JI1^gn&7YA z^)<&)1ucqO-pOOZKp}L^vm4-u$L0pu?>OQus6hSv(I&Df4G2|l3;gz|9e*NB5l``Q z?~|6H{Qd?Vp{&RKq{MtH=d)hhJ6Y$%vezyu8;^ON&Pj4_fJ)~ZKpse{I2lZLYG9g9 zKzK1el;Ms+6YB_*wcsqPMBGjetfjx|c+hW2qhe2JyUlO|#7d)nXLhbl7bd=Tx8pk) z5SM9by;SN=4d{|>_)G|UaDM~DE06D2KAL>8oHWR&n1fTK9|A?n*xdkU_s=;VB@WWI z6?bjoUFs4#9Z3VaQ~Q-18#TJr*$E~UF{?4GqQ0m5!+sBK11{rqvK>E?fWAEKbvoX4 zmO139p$&b2n{M=Pyc@+N8{74D$5H1GG5q+1c9e`_i}e+DO-bjIynp6PzQbmLL~}F* z^WwJ7l_V-Yu!}vD{Xt@f!YXkVRVX%GW2#~);UeuWY0k)hFql!9wR3~CersV;8ThM( zNo8ix-|U?FukDoj*Sn@f`sBdpRZffN_clcJ85+d(le=Z>Ve^Gbk7r()hwfe+4(N+&i ziXd+{upt><40uG^EZJ6~4$Ccfe6*FuUlzL-Z?Y1?Nb<}IKNrE%Un&qYW~EbXZedxg zNkr)TzY+4N1_#xs%AMC_1_Wumg5`vq+O2W4n_yr>;rY7gY=33rk_o-$*STx>QkS=h zNupN}Y9v~muJYDy#%ZS78x0qmd!&jaAOfiJuut*)?(pb&AKxdB0!V`b$a^oZUw49K z!~i!l0-I!HXn9vVsSuJ3GQ9!3U(DbfRE9)gTG?D!RHlVFd}d$tNl$*Z?f%oU{hg}m z=VeS5a;S(~kALxRoDTk2W363nbPbtbi_PdEzIu1i1453xyOg?A;MlOAjp188DmzxC zFF5(6b6RD?{g93VF&#p9Ttd3=(2+yeF^WZzb_q}OSoLZ84e-WtQcjmSlLb-A=5keL zIWT}w2?D_@Cy`bYlg%u5wM9=O?fhAq<_^9JEK+=~h<{+g$P%Nf@s5x`5>Cc5H@)LB z>MEpNv6s|#g}t!uS`@!SSX}0=iX%$*jw6`%HB2$`Obx**jyq9Z^G+E;T20ED|0i}E z<_2*qhuJ=9U%byDvIU&VoXn~!Y1of)7<-p5<5yo0! zv$Lq#SAQ|x3>L26@&E0b8t)3rLNDzlpsYb3h1)N4y6{P(3BtRU7{=BCtsHf?F?c- zO93HJwP!SL`dsX3Llq|7w)R%GxoTInL>x04Wq%mOF$PJJT@HNK@Vaz%O>$)a$B`>H zj07x=j+;|#1A!+#XA26@NMid;C*L8B@ze~)pd?zq5kF6US1)`%^ru!C$i2Qn!TLy> zyWXloH#eA=9=05TG?Z5G`M{zsfCvtT(oPcOoyM^Y4 z7=KglG0*crBrddehs3Bh!`9jH2xvT4i%(Xn#w74P2*LfD z9HUo`Np#h49L!>-T5y78=7JnYD)KB5r3Gd1fKzgd5^GVbNYCTqyh8At*PcWk$+D~ zBeZ@APmiok4jeJeS>VITWEA~)SS$%F4G_OU=kiq_llVHsFAo`yA4j+>xO$l#EfKKz zKbkyYnk4wnE5qHp@Nt-1-(8g+JYVmXJYx6wp@GTXu1Z5NN5yJGER&gPo~W}xNpZHt zt!F<6HLmn@w`A2&*Ae(LmSzJLuCPZ%zD80`B$Hf3s4V3aN=OTa(|uwsgiRE zO6k_G%Cs-KXr52N<>hmr^PSs2wcsph#W{B$Y-JM;WT340@WitgB{@xH6j-ZlQ~ZEiO@q)3LT4sA0ZENmSUMB+3p0 zNtABjWAqxSa`j{W2!*e1YZ~krin5PZ*y`@g_^Hdja!i&|d4JXZpoME5x*X9W z4Y=7Yr)!<#{rDKTnfyIvyK`ILlhUI7lLg=Tmx{*FD8g4k0u_`f!p4VT zFW)!Oy2R(4R@Od3W4{4>5XroG(!eW&EmGJVRk3-(g3M(vL)2tCK0QpuGFfzl?IeC! zAB=L8wI2Ai)2be+gMYHDcsNl@tK;C}Y%75UirNI8x_py0;id+WQc};aorHV(h+t1aeoDsAIMM3qCp}!W;Is7 zMiI9m+4F3a$}efI==ehW!NE|&ZH_NTFN9=3>qxEfv6r{Rh8!J`{nb7o&%M1^`7UET zcvT;BR{sr<32HYkQcfCIlc1QQG=zNmAqtWGB+2R$9Mg?aAT+4*wtMJEfllx&V`j&n z3+tCrSlpg!j(?)3iiagJgoq_2XHrAvI&9MH!c>|Q_;h~j94r9IC9wZv-NgB7?xQ^lgX7MfOjBjqaOn(r(-UJL6^Mx^Zh zrnTxa14ykgH3_0twSJvb* zsW1GQk4yh))U};$1kkBMpj#&U(1!BmoY;{hyML^N_x5qc1&V225I*=DI;$pwDz%x= zJ-s{2EoNbvJ{!@#GDAnGJB~>Zk(+W|f66{O-9yB1!^q1Cf66#$CWZK{L^tO)&j{0O|eAFmbNOkh|X7mgZJPGC`Yac{h`-g5z$28{Th zT7MTT6vPJ=UkDwq*dsJoN24^gwc}?HO!+U8qD-u+wj)p`S+yRvT$kpv!mhvEz~>S? zT||4VXLB&)(=KPWDfXjw1;s};&<>7$hM&%TGx}J+qboo%$Of> zEj~G_Oo#x6>tGz^R%zZmcrycs^$y!XSVAzBx5GZ0;Wc(|=Vp6xj6V5IqVD&m*?*m9 zv8TAkiHC$W+su(bwy2FWr>MIGbe-lp(1f$zDH1^9i&68MKFElaW?6l0g`HPFTH#PcrY`bf zpp8=FlRujy>S1^kb$huuxji(QT1&zp-=y(AEz{_#>pBqs#pq)O5)NX#GJgvjn&?{n z7jnA2*uKb*fNVBBWtDLh7-xF39%oIT>601Q80+^&8_ONR8`9 zo^53iA=WEnA~3mn3&mNS8=*8|TGcg?$`BCBO&a_TfP=%-+p8fHCV!WreFK{hVs42| zdPV?45SSUH&hH729I0Yd4Q7xGmp~aQN@?^9@);8qSG|ms{Y)!Rp$dk|8(SQ8tk$k{ z?%ufoEz+EPeOqCD`~3H3h61nj6IHSCJKpV?raE#mCM?7D&nx1@aTic_(vGz@@l~A6 zt<&;IntKS9#-|98;D474gUss`r#<##5)@=`-iqVgrQ`^UU(Yq&B3B=-a$01S={WMR zpH6-@H;8QxA?1SsGT>Zu31Q95D(ERQSs~9tF-(WAUzSDhlY_Dy zd}QZZyP}lCUF($&wFPmH7MDAr5I+=Vy_NI#0A&QWs-YOk1b-E3V!cWqR~-dTPs?H{ z<=+Yd7Oz4G6*ac937k+6K#R#F(>u?xxB_}qk^#Kb zd0q`4Ppo6+u5Cg1v@jLVRDL-&4<7=3uyr=xyp%k6;(2OqWtIo*Es64C_i2rC1KB+} z*X^^{u-`Zp#DB6wdUr?r2C)28TPb<&Oh^}Ihyf7F>$6b&7P5|<=6i)W?h*CcaxWw9 z-sfNtYK&`oP)9U+KPIV7nwz+)oC8e$Ix?wm=hhO+;3_P)m}(lRJliUI<9Qy)x>cNy zMKv$Ba4|aDpQOy)5M@>%zb6# zSFF~yf~URT=(_0&P-r?rt_Z;q4*Td7&N;_lv>XAYrD(tyAiEzPxS4qaq@3$0p#G1O z6Mi`xR)6yw=`GYii1BiYY!0@QQM?GMG1%Pb^|?TM;p@(AJXl=f#G{2POXbWB`vuj| zSq!%i7L+iLlCkF-IOohnJXYb6(PC7QjZdhny`R?OnPnNVUK67|cMzkfdeve2dFfYF zE7q!crrq!5csg#*9%=*CuR(=HS&AV_!k1jhM}HwHs>geai z3>4?So=_+G@!ViMqOhnJLTC{sMKe-`h+ila^H1t!lpCCgik;qG7)mNQeFS95S21El zOCdSAdOP>t&(*sY_x>Elw?`p#;)9crTh+$y>bLpLEy7P$i(n)H-&NioSwBjw=G$xY zTz{olt)iVod6x1fvVV-bifkBOey!|$Krv#kim_GRB5J16k(v$m z|C%T69vpl^Vmoy@b%K{%Nu#+ZaFQ)u`fL6qw;F?n9YSNdW#rE=wlz?*`_-~Vc4K(? zu9paI++7KVobA8x<-bl+qF+c$HVg!!ea#P8ZT~i_EjK4&<(46RQdj+_2cpPCG=JW| z&MqRMu7zS+>gm!yQTN5wbdM)}+4qYtil_7^iwjo% zBfqz|A}3%g34(cgrVr!aO425xF@OE12~k=$P8=rjuadQN0!ZZ6PPZjUt#j+vC*8Lg zk1R7t-q|1D0Jy+tZd!R|&S*$c4h?JD&`Cz77>pI#lKm)%LrqauPF0Qq4~0VIO&Wu~ zI)sMOM>sG3n6Bj)qlW_Ch{<_D~O+yb2VhBN_IFRy;nHawom15{swP7B*Edc z7YX$vzk8yr_09$eZzGT&cxIa_%GrBw9Q=&<<#D8QD1{>OQ4meB?dgRdJCnX%)_!m3 zqcVhDjhr*2!MWd#;d+G2gwDh^If2x{O)b;Ni0@W=cRX~BExV}>(tm=>d6%!d$>jL`kq)UFopIc_5r#NQkO1C zzqxtJv0@GKLGj#NyPR!>x%P`Qb=2N>aIW?$$q z)0VmN&N5%JL%AJ}T4Fw}fS)wZ<@ARSg;bYGi21 zC2{Vzb_%U9M}-a2H7{VZD%3bX`^qlB^U?V%P7FUH`52V46My{69-W@)Hr1v|het3_ zS!DD|K#PdcCS=)NztqWe=HZ1RNP=EJPk1?K;lMHQ@UuI00EvtsjR;+n(7Aa4R?StJ zpG<^6eTzjn9{FU9=*#|W+|%auiupOuxW|Kq-!l1Dqw{?awRvf@5lXnV=ttu7(b#h3 zlEf}Q8_rDd4}Tp6PI|O#_MzeX<{^l^CzN4ojhth_eM5k~WL2-_%O?*KCi=Tlaj=_l zS3UTqc-dsY{=$yVQa%yRvgbV>p6yn%ra#`&neNm6+z5KSbA*sA>gSj*8fNb2wgZO2UA65dPNMDCl z+79`xhiPo#$==FqE^&4rDJ%2L2$M=@lA~p&ybolAT!nO`Eja3@N;onxU%bwd0j`8q zu@as~WdfC6$a^G7>EB4}3~~<)W$D74ln(ChaY;H<}P@7a*;Q&6ZiK75e2& zwYy71a)!`Iyf^6C6L7HuGpUQr+5J^v1{$0;H~4-L&0d1USnGYpK1EkNYLg?;GTCn1 zT7Nk|_ZwhW-?V5(Gk_!Fse=I6KYvE%O{bCotd zeSc@q!=MjmK+G|#K)fgG2w&x5r=xE&_I_X=sngG2J4Mml6=HejA6_yb-b6cp<)Vt2 zzkgOvN4#}Qx9dokZl{aKW#75y9rE<4l$jIL zbV@fd9m|#NX-27J@pnz_n3q4j=S(8{L$$5&njd8vi7xtuPoz~8^1{3mEL)^wWI~fP z;;#g<(G~jaU(30P%!>q}ETOP7P=R%bD@$LnR}UH3#N8LT@@Gx!q0Z!VU3wjlZGT(| zRAe}^tdCb>2E9_PK4m~ffxop3xTyL;Qe7wQWLc;9S?%sUe}PKn6S&H)FC){p#2JD$ ze#Xz)Xpgb>b0tiamdMIe+GQi(>dUg&)@uc`ap6tE+e4il4Tf_$j+E^KhDO@I*9i%QGal=5>d@s zWe5iG^Vfi8c@NIVbes`$emcU%Ms{R;_d`t|M!uRGUWx;bS5gEI-d)!s>(N|A{T&(K zb8xIBip_t^TUVM{ir-$S*r9A|3oWPmv-!`;qn9lJVUlC>iod5Q)oD%wx7`3A9BdsO zyMEN81@BJ)yZ#=u%N&kN34fDom+4R(c%?1nCn1+NiUO&pR@ugGdj^$Yu_7LX=z1+8 z_iIal$xv1|*yd+EsT@0lOOj0|10n)Qy^_T<7fFV(<2f_2guRnq_#(D?h$FZr<>%_wcz~7zC%hjYRlKPvACbf<$oI@tnDz>Ld=gn zK1IQ4q0-7gjhmOv9kHl-bOVG~v5JM%9aB}v+~JslhA*Fnf*WAk zX{R2o!N%Q@(yNxU#J*^tmP$5`7}`#)|mGdYv8??hg&fFRbib2MlC9~ zA^kfqb3>KRNw9Cl%Xsg?+5pEHTT*zGm`BZ$v*c3fmL10x3Y5HhX(We|A0i))m@9V@ zf_SWpod$j#(nYHn4;-pss<)!zluqx)#O6!nV$(+$#ActNjeqGYo{?@pi~pGVj%8k2 zH#b5`FqZTN=wm~|pPZBwOiw(e0_j!KONG)Cro^h~zw7BZ!o2|?k=J(FkO%>e)rbe^ ztU=t{iZ8VJr{HhUS7z3I@A`&r*>Z@hDg5A)W?O5L%KTaGz@&}urwVx&eozrSN-N%* zwcOF>|7*vn1b^kUPz1;bG|Q9V_%uWaPe*V?VS9Y#mO2FxqxJBe?9%Jp%AKquTA1wR zg!hMv*z4CEfP?P5oEK5TOImaSzQ)g6f+hP*)(>p7@WNzq+f>F2bK{1|UQX<*U;MbH z)D!8K#X#yEd5b0GRK1Mb@wKnLTUx&MVFthc8VC8DvVZ8>6Ir1+ohRp^#6fML7!AGt zq4=U|g*Wo77Oosen?TZ8`Bso2nKzjSm#dGVkQI_!dg0$y-FK9JuEUQv&K!DHvT62Y zxy9)y`_9M;3tUEl%EnPkfIW2micLjjq5{k^_&HAFtqeJ=*8*(|6?@7?}GQ$?-#kD7T(G%1y;VAWZ_jrDCWyN5s&SKtbECU=8B=oo`%`g49VUlndMlO-XQMP zj`)*p$IJoQSLEm-ulr%2bc{VpK)U)&Lwzjdc#p*FR|`eU9>I_@=^Ym<9sRb*3O@^* zb$|a{Ix%hWW{K=iu`2RF(G75$EahEGT^e~TRo{}oozc;$u<)by-eXn5!8vO8N+*q$ z$$j_p(Ait2h<-23B|jH+1L~g=fzymg(;y)~AL%bd!+Ge^csLVU3Zl|aI5|1Sm&-t2Dl`zD+GPD9rJ_ z0W8u~xKq~XK2KK_*m^rvoL#@>rF?P~cA|1NeFIo#6J*vofmyyW1v5R@_MB~W*Z8il z6TVwcz^FwBCjjz;^Fa!+$Q8 z@Wt26f{+FM2-%1ZoQQz5qZA=yVf$f;ZZEhLw~>Nupl%HjLo z)W*!Z$_8Oh`AZK`dEE&cxaUmG5t&IIP2+KKD_g6SdsWa2=mU2gnqo90@eL67y5nTF zBd)^#sj8APVvoB-GNDmyGJe_k9DhgQ1~5!N$2I;-1|GiEgc>g+j4xw2{mc_8zN|Fi z0djc8+IOe#2KXh1wHr4;d?n$<*$uE#CRo%$%g~f?$Em*6mPBT$MFxlA8Cnf?2Sw_G z)(`~^W?XY>$}P6&354!M1?A-NUejm%jmMwqHc2Mo=UEn!!{q*vjee_XKz~_jzY{?g zdy3k5O)-Uvk+dLZg~Rb$i@o%%Xk9`HiO!W$??|M?G@axx*!UKNQQ{pyr75i2Lt${- zhaNG zj~xoZa`p%G4C}aDILq)L_J0JRe+hR}C0tv^sTw}y`u?*!-IHa|Q7_Z`6%icuj*mUA zp3&vj&e!o30WkFY;5?C~AFyBeh=w$)7_`IXb9%nCRI*fZ%$wy=*@OSN9{g4zRpa|D zD0+TPM_qG$+C~qb&|32zun&E$Bb+^N%cbWCuKCej^GIn46&#!tV}BIFPW=4NtoiDn zGV9KntTgSJsWlqvPjIwmKmP{Y0GSzLdn$EW&65>QMuUAvNS%ih>niKx<-2r8cD-HME)}G@M4qn-WXS{wMzSA%@ z4lg^Pfesqp{Y*iUQ7%OdH|!d~#u}(BA6IcES)v_b1lN z`$+U`*fHBJ`!UXkdW;r>b$Fk$R8+dM4^2N5n7iu{XBEkB6D3XX95c`d2}V5uzFIM0)A)Q_xlU2XaJ=#7LXildK-sy=`iMZoHXd1rpD1kjqmf zE`W*WZ;g=zdw(L6k=rjeE+yYJFY0zCQPKA3^WH-nM}pn=!`oDgur0Q`22eJ!HrGZl zpmfG{0g&^vLrT`6s9p* zW!gAlMtUIO#qQJ3yhu`#Z!=OP%PNfhgOQVahFTlUO`g`C&byf8X8DWy*09GNUl~gqJm5G`^gGP!75McKQVnt1?hE z_L9K%sDBGNttwQQjG1ZjrRILrYGXd4V#iH-vuGw;swt#huLcn_2~$(V6E#-KTwiIc zG5M93?j?HbxYWsQMfl@TvpzeXH;)tfFN3yQiDDMCt-k4lj33v6dk^=A;13tL4Moz7 zq<5uQ;M0ZZPL-2{so= zr+)%l2fmSSVn9U^q{%-%uV`K>W|~*s_F~9pjv_SS*7pqy`=-;aHGZ1Px0k@T#8pP= znJ`t?wl~-A|1}h=CI}~+1S~<_-M>P0U0qJ!@5(RNcpsWw`4JBgT1k$OFyBqCTd;9O zAUHQ28mS4Q2m>B-Iyjk+kKd6KNgdNUrhj2GiU8I#H@`gXoGR?snsMMEqw-l5sEB5T zUJCUJD*Pdgz~%(#OTNEe7J0a< zW2BC`UtTI)mV9@9JzwPAd;5Fv4DNPuZV*>qU6b)$xz`$Jk7$Bxg${d##_c?H_fU!FpV~GEV^72hPWS7ogHq<~dV9Oc^wSflG;K(Obs-K1Q)WTI;-F8V4D;In zBle}x9oEbuvx=R*4{~>72VX(%c7GnuzhCla1EYr5DrMp3Z(0p6JiW(e_P`nKq(`X| zKH^iT;AG#D0a+TteJ-)vNq;girbwW? zt28kTew1g8{!Oxu@l|*ytE^fNnL_1G|0#0=Y5YAA=p0q>jA`=#3kHr3EdPVdC02wZ zTQy4`HK(YH_LJkWVw>Ed;a)WPEE#lgaX!uyn#K1IRqEQ(zAl-(qz3$du}2${iuH|e z^YXbdkG9}y?j`Pc?nf+VUw>Cyq8+j`U+;S)8A&V5RBx5J{H48;J``?`nDI_jeh&K* z)+{zD0#2#7n@$+2s7^#xqD$d5$cUPBug@4HtT^U~*za?>6 z4oWwa>DKt5R{h@bo(CFCfPsA5P?R`m<>Zc%o95j?Mm9?5gD;Z70{dSMM1v~)Er&OD z>>f2Jbv+K{e_(>=JGDpLIWlRIys4adOJ>ufljbqpoCmahAjtyX3L-+?plm`N1Ls9; zWBsKLT*E1q+hUEnYJXYN6g*ikN5f7}##f0*as#Kpn+F~Thgvod^(-~F_0+R+Jya(M zsBfMR8yXGS7{s=-S}66#DWWTI9Fw46BfiM|bYRQvgLiJ29U3d8r~Q2eUx?e%A$eQX zK^BiJX7)ly2W16qoOkRtR7wQqAD){TH;3emecN|%e3I!7uYXh26QqvHlAq6wBX5{K zySt<9$+zsnr?3`y>-g#^NK0ad!B$GmvS0TKY~U=otFwi1=L??>8QeQ$(hG7T&__*J z$y%LaR2};*9>LM4lEgHJL(@QLyIov}{V2L{$h>yW5ONtA+SIotie<%HSXHp7B9=ip zg8z9=Gma}zet(xGn?UwbTsvK}e!?-4rfIHJW@ftjtU}ba%Wv15ukpt#=N&?+n8^ld zKoz#pmUeIYAe@>Sxb+~)W2I_L`r)T)$mFwmd7p2~ddtlw&^TOVpng-y?A)9!-~F!i z1X8A9iO$t4VeLKQFFGwU2~)bG7OCWD`_h&+X?~UAX@CAZdrW$|F1>QcE0C>x9A<~> zw)~=*ajcg2-ZFT^!84RF82zMTt)ZPSqBT|TUA)3KP&MQCsiygQw-lbN6Z$2&}wu@UiFW9@8ipy`g7LkZxVEUQx(LW>l}+; zDBVrJB7Zo>v*9u1PiX3nE0j2ClwS9rFjM7=v)i4iYPS)ig*jQcrd<#1K6;qIt7RXP z=j7N{(%cq3mqL0G9i)@ID9+&0NV%9BB*-?hx2Ha?5Ln02yqfj#-pn^qHf?;-wJ|c~ z#&hzX>SF=%cLnxp64XgWSB#MA%FOEC=lErh8Gq3HT=tI3^hq;3ibi}+gOlQewFS;= zGM$BVWTk;u*2!Lcmb-hm8@_(5JtB}@)?7en!oY(2-yZ6%G<{}xa94I3WjB8L`IlBS z7CG}$WtNi2cX#UJU4(11D6}U%2(yGVrUZNo(xh$m z!>uc|nYP>))lh!TaJ*r{SAu-vB(WbM#F zJ!fXBc?TuM`DnV7LrFuELs7}wO|KH49${n4R>8ndV-h4)9A<6cgel{e2T&O7sL~bt zkTmO7Ey%s`_ECO%&?-chMjRKV;CM=OO37d4+`K>UTM4=f5x!EMbol-x%72p^AF}WE zES1hZ@D`2Fh4v6JT70QZhb;_0339iX3dboWyYxAYtZF+Xb05tyxruoLzbSk#W^x<_ z#4=v&bb$iDt*fVZ_0h@`@H_3 z|2^ryCU5^|Rzm-S{=cXHnt#0g%ay&+A_J>z;IA_kzv@A?BtHep#{hmnfxirX@wNCC zLTpCW00{I3aT$ILPfJJE#IcZIC z9LH0n>a}(`lZ~43mQa<4*D}ihu{o7yU-+Gh;2`W6w$M-aSvTW^f-~7vG$(&^$1AjT zs4{-OptiaJ%2_asB35(^r(dagcv*@K`!3y;xVixftgYcgQ#n>HEFEgh3-iQ?O0b@X-zXj zS|~lU2r~q!H?vm-I!{X>@Y8=BCoh`HIv`x3tC}3TBO{&!coM7W=|jp0k{p@h96gus zPoswQ7;iGS(XG<^E#sxZl--)gWxmV#;kh= z*`y(?5Z76knm|;bvP6#cHaVmFaTU)f^W-ZV%7l=HoS;qEUZrLu7?rVXN94 z_SX(C>*F3<9sHFA&hier4e+dgDFW$RbbZW%o-|xxXX8imCa1OEr2!*G8Rw|Dinj$s zHB)Y`&GASOBRCVwt+_S6e+6`(3Y)HlI2a2#rUw$8n4S4J zEY7MR+*nnSutWVVAa&`$<(e(Eu{}lYpnh5d3(vdjK){6Usp7;*a1qp@Q2)DbEk)n3 zvRm+YB-)m{j%d!VLrF@=GNv1Jg5Zt&HiBey9E0EnFs^}BiVc4`(>ViI*=AY2WwI-6 zOf!!;GrzS4q?Tt$-cuo!h{;I|>tl#{+f1XN1k>-e^fpr}erb|T6uhT+XPT@Uqz66} z&eNJd`Vo&Hp-nCBE7$OAgCdQSMs!>^@_5A6ed=c@D_(C zU&L>&hnskGcAkG_6U5|Xnl}V*e%S9Vh!F)Uc|C>#*2N}mtlL!_Z!ZNfhzC{hPBKio zvy~*;TfCo48Qnu_qE)>L#LoLwth+%t;8jeLZxbQGyre}AZ%?;_v?uZtZd1OkesT$e zvf0ZD$*wgT=!{$Jc#x0~+as8kLd*8I7xsmOE24x>H-LYZojUM`!T&#t{*i%}<^Y&D zIEaBn^)U&$f3YIChF%hJi!-tLHI`6ZkPK|}P=ee|Zwr2uicMWb1q)$I?|?aq2tsZE zl~3R1h|2;EoTc@I9@P~nA6buMP~8BxdtDF@4tw}7$qXmWjrV*!Ws3Hr0)o8@ABt#j zHNna&0cd~f4>K4mK4)8K`Syy_i7~TLmS5Jkej4(Fi@yUVseEH(jX-BFmP)1W>^z*_ zP`=T*`9CKslSBiNgR%^3TR z1_2${w(rZYK+lX#<2$1cKx};%v1!d2Y|60;^`w1XwB-whL4sA-$1N8{CzT6LSp?f& z`qo){5`)>BtQrRe`S{-_kr);gpLp z5NWv|*!<=ALVd^AaB`}~#_ukzfv}g`*6j9iHr@hZREPiek=6AuTc9z~4PZ9slfi#? zc+%ISHtjy>H@j?IsLOfrNSWU(b1ur0_kHnp^Ta?1Y;cHLB%-1z6K6lp{9E-LEF-f6s3AFYH7d{LrSrjS)>f!D3n zA6nFlo4%<^7Spn?!a#^1lZwo8`pxp^5a!R$Zb?FwKcWW-JX5Hul@%#A8W!XZa(=pa zyvn;v`V^#MLk;mh3O|=vCEtIGwH+8cZfF>XcU^0KH;WS_7btC3PA#$FkG*HAz%SMs zD-}CBilZfMK8vvBHa1Xl8s1=x zl-NABMzj9&TI9c8kK#Aiq-|7^kFc!{-#FJ3+a*}Mww=fg5{O^6+cbZEWViB)uDhQu z*~L&PEUZ1lk)r}SAjo*aJXJZSDEapqfc=@!!pf{wZQm%8_wYnv(=O!Jl66gMkdU2K z&_6kd${J*j9cYn%9RBteJtT}{$*?#S(40_v`Rc!W`}-gC{|){9G=r00aq92QqUioh znL)(=d3&!v=nwjZySIO`QMT$I8qI%_#PSFILH{b@WT}suLgZ_umOJDVI+QeH9#A=HFX)oIUgm6$6?d?ATZ2x32>vw;`)-EXnZL49YY3DVy3=yuW;~@8tUV_1^NMsd-sw(%G7s4o}>UC|L zLldA2UW)B%Mk~h}Bhz%|C+S_(9u{XuChFqD*}@<-(;m9*b7z(`SGge)&{21iid-XG zie#9E#z)_mGH4i~BT{?PLE9{qB#e`-QX;Vq+zpVZB{P49a}QpQwyA==9W`Csd*@?H zLFGQZ-qcy^(m|6Sv@>|WjCVar?yd@K$trLs=^42J7+xFi-vHmgj^|!vBKjy|=~cSs=!v!c{^rweZj1yKjhJh0ecFc;PZ(N473%^S zhoU-^ivX(zd2Ni!+(u8>m2{yT_301rf-=Qzfong%slv806$@hR)5yrk7+`4+nfi^T znudv_t3Ute+bIn1hVpTiISm(4VYkm0rj?Zy9A|&;n4571)41!CR^wx2i->gzAhjok z+BA_afHn$JT;gDcgAs!&Yrt$C)hd0QJ+Sa4HeJLeP(41SH@l}p&M_e zJeq$O5|v+7ZK{XmA2rl)UUQ+HpqK?qJqg4gp|1Pt=rFewB8T)RED%O(tF-;G?FwRK zDSyOBGhY+)@nTxafInj1JVnKQUAKl#CAt1e5;gaAyOXfem+sx) zg~do93haH?Z{`>-!(ZJ_i?czGO*Wu*`95bFiDo4|xiPWX7@?_kHx7Fddc~GgB|7@V z;rMKV(~*Ffq;p>x6s%8ay5l=EeH#5WLDE)i&ALa*Y&%9&Nf>Pw<>iBZd3b-~e|vvF z^6$~*KUvlNopA8oZ-7}^rc2E^DS_Q@GK6b^@_R38d;*+3W~27|dRgxvsO)~ej!c?s zYm*&s%K9m_G~d(x?YQ#>5HH{+vp;EaI?=9&Nb_yoUL1DDS)V>G*KJ>{S{$K+D-!i| zY8?*?;RWzc&)7hk?Txf*S4DMC#Pxss+7G^x@ni2SXa~=%Ety*(3Lgi5UEDsffoIfn z%Uie(7>hQMSNPi@W!MOR;vnB?=N%NZI#Cx|aYi?OB5=`9@qOXtbzoeV#ASwygM;(T z0CV~VSR-^nZXr#Zd`xXb9^N)ya)oX;y*sI#I-PWKFq(`RU8NwQ$;=`t2(Nz@Z36#1;#GL|%zeY|)5>YF(cjN%5MG%@i`K|Y5EP6U5$J+B8B*GFE4 zmviQG&jq4k6#CK5lr_JNCCzce8f?Q<|(7cF<)LgDtqwb!#Re2wYLmbqMVnoGTBd;;Nui~+@=y!V5wLc%jeLgPfWQFt2Nq@jih+kdD@fMWC z#9Q!bP35Pa>rU=#mkS*U_w-kV8JyJl=4PVEdNhSDl-y&z+xfL5w1}G9uWm`O3mv&4 z-NFp(maudGtux#oZ5yFBZ%d)=fU_7yCd-LkBX!XD{RSku3W;Cu(_Z#m$PJJjwH$N`M8sb)S#Rj@w4XT)c%R!U z5Z*z+=kAoHDs@+}#Rq3ieI#Fa5SR*g@#;*#3gTF?JR5y*79ao9LDl7u= zjfG+;5@T~s(xz3gdm3+Z>gCaOkoU*lf4;Z5Ux`EQH$?+|lbBzQvN8NjNFGhKk>t#0n4ohy_FYiJ4*J)k2 z>Ok_JtYUxtLH~0I!$nRdHyAHBOB&#b@FxZa9v`b>uoR6+{3{eb&`*%xC z%7vpi*<&wnJs=t#AN}vmA^&qN>z7?4s0UQgl4Ij#(&iZV_H8OuJ7-QB|LOtZ!gwnaKO9V>7#AsXf zIvJ)>syMl0MjjE|610)hK2t0`fG-Ehrf%LJU;mHzn)UVrJjBGFlbQqrjHQbEs8W2h z0V%}t`&`ASOcO_Y3sFrtJxxkLNmcKM*$n4YAMf7g&Ex;ExIXkD7pg zruK@Q9Bq(kIcpf2rU@>3-?8s+4uI%?H`{;n|L)f*sIz$(g$Qof1eoQj&dJe>bSZum zccVs`7{K-Ta}vv+5?TIcK^=>UgBXB6w$8?0iBcbv!0#b1Qv34}595PEgxbXTD29v7 z;|bM2j(;5gIR4j<=%&9E3X8a26;0va_Ee$$?UsVS1WwW%On*_)2Rk-4I2sE7Nija> zKj;toozzGFefWfPGy8u5P)h>@6aWAK2mp+QJ6s7Gr4xez0RZ7y0RSlg8~|=_bY)_1 zZ)DO=WuB?L0~`to00Zb-d1>fdd53_KEVqD?Eu8@)jD6MtM>UD7@ZlLYq!3r>Q&LvRo7?he7--4h7z?%KFJ0fJj_ zcN%M;ahJDq&&>18IlupV?#%gez6`(a-MeM&+N-Ko)paf7n}VD;3L+jN6ciMSq=cvv z6cmCm6cqFkJS^lBc|pq!C@92o3lR|oNf8ke1xGtG3u{v-D2Z>0$$xMv$_MZKU%c$N z5$IpNqRtKd*Z~vsP8f-#BSK2zb?`e8l9GkQJ}ez1nvW%P3xb$Bi65=RFp(k*#KbrU z^KvR9VDOXN0z~X@GH=#A;#v1zPFC*Bcf0noohYCbzUb6^!cc=Hkxrofc6w7Z&`0^} zOa$tcB@~)7RM5!C+kd*Ap1@Z$%Aj@NDj`&46({|G@zcY@3k{985GooJ6*ADo7nKnS zh*1)C`ZG=kD#oW|)n2jEJ%ce{CO!uyZ-sNbG@h4bduU|c*HZqlYyk{R&_zGqRg5AuzhbQ45=2ML6XZ=nb0ofI&egN_+A zX23%Qmw#x$Sn^8z2lFpvxqyp}cg?V^-8QFCgFi^zQ9u9SVS*JxC(094k-(S=`u+iF z5Y=EGobAw3*nb7!~Loc*J9y6~DhZ&67 zn<4_`Ad8-QV_TzpX3~@@4Ma73s{ofi$7+WKm7|Z=9~v>epq9UOeC>!N{Mp&NX@B?$ zc7Gff$3J4Aa64&>W5abD&wkIIuldyq6;bs3PwkD>vs`xuejNR;>)!;v3q*7U5dZWe zp+diecZC)Zek~-bAwoq?@G;}V3JEXy_qU1P`bE`^sAA=HgjYmY#I4EoC{f=HNYIeM z4rOABNl{i%B*q>};z;@wa3Ufi!Xh#tZhs;Qho~Z&B8p=%&|cDdlL6w!hO8YwYD%q1 z@JsNEcZ&Op!l<*spb)Ok+qTRLO&#p`{xD~{f`0>SM| zUjZ)LE(eg`!%o6lA>$zv;l&}RhEj(vBQGP1f4}^``<|Y)jJ4HHhRqDG7=IVcbi(9o z*TeLUUDJHZn&sy&1N--P-?=|$Q}@kGvn|&zT_)jau;M z{Yev^-c|c@v}fEgl{`(A4^&CM7-yPlu4%q#S7?i6-OyjQcv#pF>k@c{4MGJ~aYS*X zay(cQEtD<6E>+gpH2Bz<+kXz!0jk?fBz*j@X#1@O#w4D0R+-AJ6@E{g1yf`_ikuf2~<_xo7K6pC1XIlJB#Te!xP*SR*u#Om3cj zYRHHc-1`9xI2PK;8O%A)8R*{aj@y{yI?edOrM-6<@;>aVaGHn^6{WM)elOE!4cCRL z&F0hU)9_Quw~cQ{K7Vq5ER99_h$hw%dOTCb1RM7gNYtM_2 zj+yREb->D*pgl$?yGsU^BHFC+;BlY0(|fdeGz@6V5j|z=<9{n8Go_m4bFn(KPczrD zSVH=st-;`(NED-&`pi$twr$YgG`cVvN++zQG$$EbMkO-Sso=J`6~bqPtxQ#9_I)ykyz_p@Dzsu=q{-f{YtWo;b)Yb&PJAUP}$%FEu z{5nUG_0iH!y~TvtirM5Yk2Y`9OhQS*PBOA}W?h2?zJHJQWMk7Zj&+yPp@LRRP`ya2 zh#I!eTNwu4iB&Fqlm@%n(UUYYyY``tvC9p;i$uwA1#-xGOeu8EQps>;N~Skp=gvuC;VGZ z1%+$1oquIQ9l5tupJ@2*kCF@YPY95>=Ug*9+Mk2ZzUfBJ^yg8OSV&GwIBC_DlXn*G znQRO-sx{5^>sR>lxl6sM^jS1ar;o>sV`kjBa=aYvbDg!|>?u#6)zg3Rvj1?>Q*nKI zT^S*S!>E7W=A0We>u_EB0901jsmf>{ZeQ{qJAcqpE;csbwu|awcjawCa zU1Dw98%3GG-?O`4X?Kor0xccKEc3bLfiKr&+i5|-8`Ra2_M1cBwWk_5zcvt9)Y#=pFwfPbXS%4nnfiQtdN2!EBQKuA1!9UWZ{fKm|7?xi@c?d-d+Eu z>7$c?_C|rc4#64Om#U>1@lk|g?-qRcyob1>_|o`{WS9*n#L^LF3xTiUrW%rFva(P# zkaKt_=vR19Fp#rXkSG5u{Qo)s{E7x3AW!$4*Z=x8f^g28f1N`g{Vpb?EFvihc~v%cG&Qw#`eNt&#JZ3S`2fLQLemKf z3Y+}*^Od9$*(s#df`y8Pvxck;x3QfKgQ1C?ktu__js0KpdEL1or#7a}h9vGb*0xUE z?tCBr`UN-S{P)LUVdu~QXH#av1H&zBaM{`DIE-o%cCKg5(7JA4p=$$-loekaTZJkK}agl#r zN7U5G*wMn?*}~42wcyT$)XvUU1rwICH_{D1uoBQpaN z8AJv@GYbb3FXR95tN&wBjsN|gg_ZOFQdHyre^FH@ zQ%4ay8_3tr{QqmPe-{4tFaMa{E6K=qGT$n%v2Dt`vz5jzx=Ae5x2kc#`Oqg8}Q&ba&uA9w~iyR z5ByGU2b~fKKsmU?^qo$-IM(2IK#4qpQUZib(U$1Js~0~3<+`z>@e{inx4Yx2`J1!b z7q?xW_0&wC@gu5-lMC5NfO^#n0?I3BbU~6yXtm0FmCl9cMb9oZd4E2NgK2%R7x-dt zVnkcyvlb`>5jzI<%JQ!9peL(CV0h&YsiL#B!AdH&2S#`KIPndfUo!QynHIowWQ97B zcFokzu_KG`AsHE$f9a4`r#WT@pxfKPH`B88=7rl$Zg*6h>f!phsjYl|s!&Q|mIlkm z2ZvEFp^&&T=9tIz{(t^8j*`SvD?9K41s^yVOB(k<1lHsJTiBbh)GPs1!lXqfoWWA% zVx8r5P8%|hi=FGMkPF->Vu24%yNq$YVQ3H67h26Ht)QK(A{27vqVKQLjayxxd60>@ zGK zp175&1%21Kk>E-EpMV&$%tceGTv)rtDHe|T z_83f1o?Dqe5EYTNUihLLcR8=m$eB{6o68^J^b)MW?{Hp_@yS6+F8f*8w)sPc$#9C0 zQjrYqoPTyEw~N+Pku1iX2JDBwz1c}RL3En5Dx$rpzpd8)9Fm}fsBEDjEd}b;A057W zLiC~Nhj0h=rxM^v1kuC2{_yg@4|#h2Bq}rd!u)?;3wZS4EuufI0>q9bI3(~`SIIGG ze_92|0XnxC?vK?ZNOA}RFD&*%zVqgf)#MK`#ed14tLeX{_^&DcYl{EK6oQwC1bnVC z&$mz2Iv%rFan#12CDS=%FMZ#S6F`bez`o{9j_(%RA^o@b-~6NHVh^|xunrtutaH?D zt&by@#+FQF5ew{Mv_r4^;YR^~)%EhkXom9ncY$vt_Rl>qi(N0A>pZ4y%m5t^U}EQ$ z1b+@H!K3c;J&4h&sor8rqW?SYM?UE77m?QhfhV2L)ka<0p-kz+nKHBI06Nn^-+MwH zx1r+fieyNlplq;GO$Jp-C+C0%x=Ar^UDkbD+|Tl^kf+QQm26j}V|6p9i{(dhd&SE$ zYt4q}AoSjsH<~|*)qMO`PV{Bea{AuldFvNt&ufTQV%%RR zZ?||KOug(R>-#nEllfI8y3cb@xp7-RKbWRB5xui7;IYke!CL+)!|xbWot+585;VC7A-GlNmH`il2JbzHRjtd~T3Jxb`LZ zwH`XuNnK?%A~RkVMI6DLrh`t`t0c+ zrqC@T^V=w$HXjNIf+L3jnyFUcw&b%nSN9Hi%z|H32A7kX*=Po-coY$ezjy!2%X}(W zsJ`(Fg1E))*vrkh_uYBxLOut)*usztp{5X20M)xwG~q(obhRvx`+pk$0?SXt2N99l zCbNg~J|cHWq{J{tM6dBO@VU$a_w{KkytXFcek0q;SS?-aKFa%`gp@KxPU-H~ce=}@ zx~bf~k4Q0hRYLW%C2C?)n2bjB7k++7N364(olEx>4N7Gu%#*`>-xh^oMBZKS_INO! z6xt^|6JqH5onfv|1b_QYFr6g}uYxyui7{TjZ{HA-daOOFKiBg4hV%+(vE5Su&4tMJ zKBLn-KCm%=KJ@#Q`_kt=oKPx5Rgr_)&X--#`LRLAX>7JAZU5?<17bVe?;znXLeiN% zLMz++JRvZVLM%EhRfq^A0PpCL-ELh=lvzRNZh?nS!DazTfPZ;!3jEF$6PrP|<%+3h z!w!F;*k+g`8F7Vx+vkUv+xv%-9Lmv9GFk|Xb6}Ju^TUa2Hh1+{qt6f!-?|uW^nI$J z(|-QuG6=!8l9M0Z2FJLZH2el)LAyKJvrNpwN?K#9clf5cflXd|+$i~XFKeaqJ&So1 z@M5P!9*Y{Ep??CtU_*>?U~VxrsIuF_UOUWtuS3A;kVe2dJ--l1l0e^W#q=r@)ogDZ zToE%p#<>Uo-t~Nj8kgNHZ?+1^)9~F~X9>bjOhKH>JE52J($C&aT7?eVj_o^*;A_J5 zp{>G^@@TB|JwY8lhy}Jhk8anpy{K6xZ|@_DZ+$65uYakx+?-`@yksT@Zu!!G3TTl{ zq<=v^B9!4gGp<84eMoUflJNX1`-Ac4p=8Fco#s2U zA?slyJtrGHNnD^4Hf)EH^`ECl1b7{mmP}T^tbZg39gNfWd)9P4zL$$4;HNvo{TfaN zkA)!_hb+7}o6QELdQy3=lq1-k!VkCn_Rl2*z_=bQFl1Lfj3SC2X!o&WCRzt6xI#(A zhGWuGuv^X+oyQyf`r~9P@$t15wPm09$PTDNU(>umx$(rAy6U`w#H``cX}@7r zb-fRL;n>$#%Se)W9}`%i3MthZS9FzfjmRGeJjDJS)R0kBJ0H9|n7oj^I~&O6ouE@K zn^CHDe_N^rT(^(sJI&O*faEdX26aCDe}A!Vn{<|AT`v36oC4`&xh$Kj;Hq`K<|84h z7xV<{GY&;~k$A?f$n1ki|9|)M z1q^v+!Yqss(?7;WaV0#0cvrS>T>gJ?qR6Ju`m%yS^-piR}WwdnwHV`FRnc@ zk(`#JqJTD&Z$?V_TBa`_IQ z$kPMb;ALuE#q-Aya7=3h0Pg#@C4Ul?67(4qT$aIj`3B_MYUe-4hJsVrDNMfQ1nkqB zcTL?Q;v;t#dt~P01XEj`X3yzkGwx@~Zx4Uv_QvI@SLwYVxrad?!ea%~XrC~YD%2>w z?qCi34?HROoyNlfp61IP_xvWKPBRJVSiPJ1L{{Q=T-IUbXn~!jO4>85V1KQL#rvNK z!XXfKWmn$$+(@73v>dDI{Rd4{3flMaKs=S%rHnwpLHKv~k9ZV8;;NTljL*Zhb~Z^+ zD!p8g0<}_sy-(gPqVSl0^QX-qh0Wlc%jpK+%E$uA__-?~Nb3FLiWM-PJ`np261ric zc;>mHOTm5WRe;najl2I510VCq*C%_S{?bK)?tNFL;+QynidF+X^FQ^nd0*y8Bm8_qVSV zu2ZAZnPj7gc<|7JYn!rpU8FuorhchNqEnQyQ~_AqA-ZS&-{OVULtY>8Pd{`Tjt^9P zw1EnhR=tRSELf2|M~KSiy57z2=_(nSka>#2K-7QW#QV8>vGK8&085#Cf^ z@6WyJEB`AV_(Jna!+(-vp3e9_>10h}G7a>@%Y$haqf2Nj8Av9V+z!1eH?4X>Qhj?k z!D_z62@gc(@W}jLH{nMoL{*GB^2eGJP4;JnZ+K}6VliJ6=5jk>1c^W+2_N*wnpw6h zn}xT7MiboVvprRm3Z;N|-6Al#&S97xo>%wTds#L6muQ?YJb%ISp8^%S>-WX~xX~OS zwNGBKLwJ|}W9tj&Vmr{vN~%49T3O4mL_w$ojsISw^Uf^&>Q$Zj1O<%Ugq>;vjmoKO z%V8O<4>9j#pn~2e#k%G#y?Z1%JTUAh&7WI~EQZ|5bzre5#?!@e@Sc%lD3(;5Oa=Y$ z*Cr{8*U4)DO@G2R)Kt3#=}-vFSl=G!T-?5|MBiG>BWgTUeBGy*1Ydf0W?+Ul<5_YNv#9QRb!KOzJDE5l+2__ERgr>U>DOXIUBd3 z_Pjd05-=Ap7;%lEUU|yc&K!Zx?ZGl3_~V*M%IEm4uXV;dhGu7sbiVq4{gKd)IQ%yv zrTug5{L~^T3sM*lm)k`AbmPFF{FX*XsQ>S~P5*g_xrBzau)=IQ4%OT(mAAvVCWy&% zEzuq2Re!pzE@(IC-$nit(DAEEo>AX_2-n1sA_8A$`op{@Fq~1#d;39 zqC4c!Y2JzT|1q3_XekgT_fm!wg}u)#TLUKd2!BesAY=ULF}ckdM0BwEKKhUQbzVz~ z{=f~MQvS>4u^F~x59C^q(B?gbwyiS{(M_z5(SLAX_0Yn9-4_D;Vm4iz4}%x_FskU= z;95rFsy}aD+5WJ(^^2S253|o?&mdR7l6i^i7D)8tc`P9A(W8LJI*hAD`SZL=zd;;4 zRe$FKo0>7Q=^H0;{p}OHDO_?7A1>h@MDw;?hwlRw$^`$|uSmFIjz|)+ItUGF^^fIu zx@>R4JW<@TMbU)0dgZTcJ6(MafC|}RHh)}Sl9*ta`)l86Rwy7c3s-XQ6V>KP{05}D zM<&_LPOFd#n}SZm|JXmB5rZ*;r{uA1tAB^o;N5tIsI^BmX_il(He+-cm1hc%w zeD*Mi4XJpbO^Kbsl@-@UD(w^?k_6a{$Mm`hh*E%6Ke_)@CrUV{XbgSi$XM9}t$#YU zo3Oc;k4gm+Z?7#vc{s$X^g3QoO9`#@qcXrO68t)5Ut2V11X;}|>!B`Wqy9OruNCx$ zPTAUiFWF6%8>n!<*n=SQHdqSL?bY6&Is$sbYX=0DKGXFA*w5(0I&W!&$s3mAF8&lm zeibF*2q;=O!WiNrzg~E;Mjz&)&wt8eW!E!W_iQKrQ(bu@ioUZBH?j84G8QJ14pLY; zvN(Fw&jj1pIbDZnp_)IMrsG9J$b$3$$DcOAFPQL@8<^uMUmlMXF@kG@q&Gjp;oIFj zbSXiSgXSM62Qt$D1)*Yv<+yWm#J!b|2*L+vJ*DS6-s>s-VY+XJQ5m?Cl|#}S zbZ($7|2+gs2Hd{AeSYAp^?y9(Zeg}kD^;wh)$zGfJ##XM3j6~XQu;xQoJb&|5ROg< zU%1>ZETiYc&7l!t@o0(y2~Y|{EgxqJrISsjlPYQ*XYm%B9U=&LZ`%mCy%$=|Qd!Ir z$~EgGN|pKwKhF*;m213}Ol2{x2BlIx@#%qS1)d%wz#Wczotk=HEq@VqnJ^Ha69_OJ z5r`d9Hfu&*=#rNoN@WcLFf!1RevVESvO~Eb{OGhRQ~Tvv2DG>hGWj6)I@_mHC^Y&# zh`WDel}9@F^_{MdgQn#8#O_#nSKO;8FO^JgH%rF82<#{%+=stOJTYX@&k;gYA@jfD zSetir8X3QjCH6(E?#EGrUbZNm+b)ap%C{V0Qiw zu&b*(zKe}EVc=h4h&XXD7i(t~zsI7_D49&}WezNQ!p0jfpST>op1sE_H5iGx992%} z)MwR0Cis87x#3LKt70WDJ$^es9|C4f#Nl>!A~f5rM^Gw~p?_A_*HMeRsPniR02HW# zTp6+uwnL?7hp}{h%Ky%MHm~(QP(84c4U{Rkicvj%$J26zP0ko;4ER^gk|EHz9J-J) zlc2c2&HXH{{Z~yoo!)uwyvw(ms=hZjgP^JoXU#GZ@;%~ZtmmB}<*6w!+)aNZ-k{G7 z+qbh;(QSXpqJO1d;zfrsb+Xwka!Q4*?TG^!kXSo1>8Xps4c}ljZ(BN@qY35EP?l)& znBoTzg+b0C{TKifku(@VsPA|zfgn!g<8>?bddltU zp@~hW^RAe5u2NO^GcsYyzG2A@Q82gXWl92_@}i4Pmeo_evwLnNfU%{#eXcHACjG1u zwC&wxT!$cB=y)aeAXTq7AB-w&K~;CMI%aZn(ore>IQL_1s@c#-otU=f_h_3J&uTo0 z%F5#c{(s-l8wB@|oTVlko6IjIHj?0YQPMq-I-O1v{T?WQEc#RlNZD{Q?yn+FnEFZtV`N{SE$;CdtTA{$w0CSBwa@ zh8M;y-oXr>I;5So;!Mc)d+u|AW3yNhT-%@2t?angQt#?;-1;eLabu*e*GSYsl<%DlW@P%q^!8 zi=X)K`}^@n_a{yg=dVm=96wHp>_Y+3x4a{wu+P&mhs8>L2&ilxpe-GWhP!cip#>qi z8EhSU9~L{edFFRp>WU@58HwrmAhZah(|>$$!FXAK&V?5yf7ZQO_xelx?jMO4A>rtT zB-MB}{)!t-^k5c-A!>LaJN;UDYCLc}uCD_O)wM=FZ6-CrdEe?8hI~TlHea<3xtkW} zgSg)#OxEk2%C6uwS~%2ahQ+#dD$ex`<=^9s!-*}*Q$@1d4FE{I6e?jgA0KYL(tjTu z2fJ$S*`M!74mnBN{*ItQ{6vDwYL-C2RV$r1 zj8*79*&~G3Fcv(LRCW7vsdDd=?W4*30*`cJj3ZF0?ViYULvJX7I{$ZYZo4^3tm;~% z5J&UaXY@B7eF`o^6Sdi-lu4cp$BJVo=IAMO6_3)O;*Es_gPL zzg*}@OQxQ$gY2!=c&BpoZ1&J6OgpMLY!)paZw62J1m3?O$MM{Z#f_{OlaJk*Ph^cb zb)|4e6AOHh7I>6!Jz6ncZK948>GW<&vmm>WIG8Hj>8DHOLyN_%1nhlFm4EZ8q`xz0 zv0gh*B;a<|h$7^%RW%w;q%BUM)z*?q_mZm$P!}sf!=6Zdyt8#8^c-87s4!m4oY(Pb zsQ0{guyWbntug@)v(I(7+@8J#RC!*}oX>2dX^#K|bSbqvmn^}S9gP-G^>&^=H}m^p zWXA3u>neJlE5}ot1h0=@#(!KMAJRCPZ>}RN`S*sq1yF}VQS^szcC+rsRmxfqhiOMU z60~i{J>qvJ{L(R7@?)Hu_s-XdeG|kY1nh0dy zYuXnQ`LHhJkX%^#wAGroH@T*A<50OT4jyOhdVi$3*O>v2M{=4a z;>`j=37sQH>NF>Wvd4H_k63ow#<((Al)2UaH%jQ4!)}F6@URYaef9V**Z@U+%*ni(kH}XBoVs#T5oG1@C;Fg;MNn5L2b`_~kmE+*_73AuQ}-eMWeX`8h0f8aDmGs22xcB#J?Uq^o_VB7enH zn8m>c3osY!^RUvLpfFIb3uyh0_nfqcj)Fg$#;R{NmPL$S;&_sio7L)KIXJS~CNak( zLgCqUP0Xf}lz;ML7!tTM!Y=5Gr4pse)B$uA*&b5i*sfv540|qDnK=VJOD>>7yRN3! zWo=wxmY*C1UpRsI%3_s!pp(<>u!ucPA*Q|CH6OU4+i%J~x(Cz6$>d1pLv<5wF%|lL zAW#zyQFB+zrRNitRfBqV0E^)in^*i*>l27^=MeCM8h_5a-dX2@lJSNjnwUC~Ceh8+ zg8_8YVUkLtwlF}Hlbaj~Qul;aWc$Uwow;~EG^{%3>~qckToodVe%PSi7K)7EW8I@K z-{h$NsMNMi@nz--I4`SVSM_38;nJaw&uK|g)ip=I?_8M06T0v1HxxE?62egV7B86F z`?3U*Sbv)qUa`;n`>C-FE;cAX_*R zLxJb4)UzGFqi)sSrJLChCBL^a9 z&ws?+nn~lHT&xR(O0?JkL;G)o-mHBN!l+nor^-$>kZ-0J3B@;gT-a19A0+(XmfE%9 zbMQQjf1X@AgCRG~)OTCcTVQnO*twa!-Fm;qJJ^%r5MQwF-xOr+;->F3HmlU7N0@3w zL&KG>4nH9V>o_2Hh0wt{;JNN z)gO5{qW#8`$L&lf>t^la!3+?5Qv=6lwIl(1>9eZWpm)V3;Jz<`%ilwY>4I{uJ&CSU z#i@samvM4K(qL053JvV-kxghNd)ij!BL!K>8Zo5ys_;i0qgdB=A5~-qT-1WYso@KAJK{`fhQ47?`h?9QI;L}!pO+u*;G19*dDS+Q zGf#JNrSp+z)Pe7=Gi7E=qce038R37@iDqcwv95()a5ctM1n~sSGz!{azTY4U!yt#} ziqNuluva&+wYFvL4N7E<;VDehxX1bqTt?EWwo6l_h8sL5US&|u&Z%^1S$7V^a-eiY z;j*)loMK6$A0`rYl$RNe-^~45fp#M7Ej8L*iT=AG&@Gx6N`1xk4@-?6{iAPU9$InF)95;`?G*#H_IWgMC@Y0(J_29N(3r_l;+^?L;tgda8Xvz_$No zF?Gn^KEN6rPrn?QCsYiBd$DV~e%s6BKwFPL0n#o>Jo(T)#~*k#>>g3|b>KX!rdvl0Z$~;sk$;164{>T9(Z3 zrKCPjEV2ck_q9eF^VuZJxw!8CG6cm4Jl`t&N8fW1qq?nW3hPrKU*c!_jwZGs@Kx)VTlq%H(*dMj`DL1=>Um3Lh>J51q4b8jee%)N^hSDVhr zKy-GDk|}Ll=dxy&#P}MyksNyc>R50 zJJ8O7ZK`r?glW9!jt4N0RKSm!5{I$}h9c^8B9b zt?I{-o3OdI+g+2^##YEEr zoqHLIYgXio$7>S2*O8x?a+qQfP^Lo12!phCf{5gPf?j{%bX($Zo|Kw2xae$2j-tIf z-~R0jRgW#6%G3oer(rmRbXhhOGywP$WKvmL?|YwWo`jSuhF-*NQ3ftKm5maa1F+2> zdV}?r>Om$j+8IyWR~d3bd+VR0Qc2#xp}4qMDHJIWT%Z5UN7L$t_sWO03PmL*%^&3- zcfFcbAG&{UAa3)uomx!YzV*uEbe+&=aLpui-^mrgy2v=o`_a%NceSq4 zKgrh*hf$Y;klO_ZrJ8GbA=XYh5>GW#*X&4qYEFLz&av?~ZONr@j-L&_$vEDhG(ImW z{j@_ImCEClmO!QUlZ&mp3IG~=RGZ0hsDlIIV}v?#JVl`qu1p|ieR^O=zB1Z7skNH7 zHE*(8q=g3xuxIgo8jxKC%Ch?CEKZU8VEsC>xdy1a|KwuNic;PVm0j&lHc+Nm8n{YB zylj6l*Om&qehEEfb62UDTYVMK7>+e8na!z>f%1VneTjFmR#wm*d^SK7r>^I?&Z(z- ze^BEbUO%sQK%?a{&h=CT8?42@yoQ+1wO%K=-Daxr!kMZ2mJtpK?jWGPC(c(5#esVn z7GulWTsY!6_lLpkImBDNR1c=ik2i!m-ei9X+_xVJeEHhWI{ie@IVVHkYP2Zaa@Joy zuH2Yq1!3jL%~^Q9vj^zxCP3?DS8qz2CD;1EFXPX166Ev5D^n=#wRk$4=FB+-9X zc(fP!S9GkinXnas4V(|i3E$$2S~$ACyhgH-vF~2&xvTL3aV{4~vbtU2S2(j+^^;9b z0B6~%FE{e(-;^;QVEPowqW9bCWy=TYvKD=>E2CcAe$?;qB_rl@tez=T9Ajt84N}0{ z*dv*Czq>1JJGqKG4n0>p zysBo(=6}u?C$_Pj?&&{~=T-s0(7f0j)(cc8pPa)kljD zB0qNRgA@+NoYV2hF3|wD#J@moo5aCWK9;jKA0g!W>bO$%d*{+^Dedj<@)ztApAF{4 zW{ZeAX*Ivv8C6}Ias-3xH~t)k*bFR#dtWK#3{Cv3*fzKUVwZX z43e;i5BS)}_YSCjp=(8_r8<9Mi?^ZkboS1RU9GA+P=Q@0F;B4we4D&kJ=o4G#stbD2UPn(xv(c39S+LGzDevC6_Qp|MahAKN-r-XB zET`+E$-%N;HK3}tw20kxW?2nttYI@daov~4ptDz2HxYsTeXGf&_4y@7^ zdf@q4&EJhA#JwefH)RlWL>4|EJLwRucP(ta`LKuF_4^dkd#=+F)LaJFWW7 zrZt@ZRDMy@ne5D}>L7yv$7}{zn`8#vxH5LUuc63CYPF)wv^mCAAgfOTjW>j+jo%45 zhnmu#NuHhq-NXP5#mM#~hUX1e#tHy2$WyM2vQME)o|W(~`E=de?PHjN0qn^xOfgR>R?VLm;qTs@zb z2;S?jJTdz(M6E=&=$7@V{IxGXK2(Z)3E}jixZ^}=QYldy!1A6h6TC2!bwRP2dK6l` zziR5_&s;m{Ed+l{;$c`L&hDlt$l+j&J2X0E6+o*Xx<^LC>G z)Ba05(eF7V4%4K9bJOHNS~ZS@0O3qNxnlLz)5U@7va5eG)!q!OrTX(MIVT+D_uwnN zC)4u;+)m>IHcxF%pIxM|PzX26;MH9_I`)13QdkvBVX#a5^|IsD->!SJ*Mx*9-0K~@ z;3=L%2#qUT_wPR6tc})vm8sySeX_!tekw;F~S_*aS3ISeXw5nXoK;+R+6TNl?tgXr`#g+ib}4^wyZOvQoDNH-xzMiahsBgmkdel> zF&CdT8EI^bwtnyR`phEgJu)uGaKk~} z?8xh)N+*=BhJ6qkF*^RDptsp8l{h@Ib4IEw@|6AdpTi~&k)MELnT5u=sB!-`%5O`vy&N2Xmzc&XOa#YETb9MZa? z1opVw%K~`$eMHCtw3*QvM^@VxRffzoD{!>H5Sl>B%hQGD{;A<&gH@9iaRqwgluG^928Y}R1l>1sxw=@tq;S1Y%A2n$2tXDnP*F1GOD)oN}nSiE2 z{Wss_-dJ!}JU2|r=(?@C92_xdwqKH3EU!HwgQfvtJE$xud3FPS*jZNdlVuxbHy}1A zWAP{Lnqi2C$X(?xI^jsnC$G#FVW*hOyuB#@W^;ik?81^;F-OskCD&Hfw|x%mAM4EJ zj?9#c-m_-rOJ}z(=;k($*wKGmsWk<#wSnL6ot)R+vF;;Wn8!BWU^B#`X6iZG5DU0A zs{qdJeoZzo2Vzx`s1Pdm9rl{i>WEK1#G)re1pl3<#3eupQ8{rvBMDLM=@th>%FCjE zxeHIMwY<{0yOwFHcZ|-UV$O8%p1blmDagq~D;|Da9qZWS$z8(x*XRaT| zdAhx5U-hjqGE}v>*`CQts}}s!&6_ninkzB%!R4)P>cHEZmw3bK;3I;9<89zOs^9dC zvFGR2v@=cI3IO&Entm^X|SK8*wE| z-!ionn8wk<(_h->{33r*W#tW%%Z6N#OJY7hf1=HLpWdtD}1fBD4qYFg&=JF|1Y zI*IGdE`u89qm@;cuS8WY@!~}7>+{7zVkmUn+I$##`20Wld=_^~!Gj>XYbVr*>Dm{S ztlT1Hq^r+$yX};#tvB{4shsAmn2cHrvp7S;fnw4-eNiVdzutdiEFroba=iS=lOrsi z5c6r!mYLWEyZSJ?;5Wat?^~6zuUeIHgiJ%YE^SD7k^_p1uOFpM8U@~<69|}eC8OS>v2dQ^{%|x#5LRmQgtE(X-{2$Y-XPn`U0y zt`?8@sH!=ql`MZFCcwYPNsh|3yGe$mDI#8;PqSo>PAOwPTkO=h`DQYlpva)p zOmc^q54aI*vb+>8Z8_bUP&KEEyn#X;P z_H%#F*H+iBPAp|zOC;iZXy(u#CrN0b=F?vFC;Ig?&S>w;Zm&*bvMZ9A6}Msu1SfFlbo*-uEkjz7P<-#UflKS_RvhtY z0=fpvW`KX_9qN8fj_Op#e5XfUe`hHP95NnFe!q3u`(#4;>%4RRlL7h+Iwhn5> zTt4^pm<+zjR$e;lu$NvUY3vsJJTfu@jbYj?y>5R$n6UYE?{Zh1l{o<1bA)}vE!XZ_ zXJJ5~E#dD`|8Hze80j}l24#d68b^BX zOt*i%;gPD&O`a?UciBiBNma`E?op|N2iN__+V-X&eDvGqfMm%=Yfi1O>+DXYUT%E; zho5O{zwXvL9#G@XFp9a`w(2vsi;il|fN&6l6p|#&&$v68%g4X#3%?!boawR0c<(g= zJ&-Zx%dq4F31%HVq8J0wF*J!Z3t?Z+8Z>`O5q}Sst~cLPknxNLn>Kjp%4g20^h!F| z=0z|EpT$+x$sOZW zZiqb#7wWu9r%PoWqW5ro=ofE2l^LNdbnNr%Rb_r!=)6j|nJUmknqU4>nDjlBrTKsU zqb4p#bC>E{shi1g1ufiRTNH35r5&21`e;p4-!3~xK&r%=crkamPDdvJzc<;wt;JM9 z?BktXytBo$)Ng7i)ptGxSSQi!dX`@2=W^POnI&dCs~B+Y_8X?>8;9_JGMB@dtT6Ab z1CGi=6Fd$oyfErsn)Hcl$bIpyOY?u+96br~v#Tb~r5${{x_I%}mHAcEn)~^RhGkBM zYj^$Ked~Q!c1PzOpsMh$18o^FXVt~e2P9IKq|DgG@^8t&%Dv1Gc<>J5eJt_1uVKhs zk_HqLC)+H$e&r|4px2mftl6fNUVjeC6e)B+%X0)KsyE;F=niF2$0HLyf>nRKB!7?O zvS+C$mK$U@6sHQ?ho?&YT72v(hM2PJ{$AT+=}~C!gNeZlXQ9P*o32l{+U3Mn``s#5 zIgw96Nc*)TjulQ7=x#_HEPb-#kc~G1(t_oYKd$=6z>-?Y2kXC^RTwZr0x@Y0MzG_$ zM`GEWi)sjsxA+ImPcQ4lgS3C@fW0)OQ2{C-zt)Zw#DdSuc5nsWz@^GJJOH&8j-N(i zBBHwPTyG_<0D%!=!sbuHd= zl{k1|7mow>R93eD@a6A~4y>1|mzuI-knsGM>knur4{(zG+Lh|1($s(PdCqMV<;|t{ zL5JRqd$nr-&k62jWqIC@j4sC;M~JQEsx{aQj&N%XZs=0kLvE^NO-B&iYHM*TF%^E3BV7HA_X5m>~c1DHD|;4@N1$Y3_j z@V;O4^Z+(+>MlD26HhXMXX2~fRsKSMyXZwQ4-t}=R))yo8{B^n#HSj4DCbcwFXN-7 zniv>c?~SxdH5#IhmT%JSG=8rP^l~!4ol9STDY0IAOtjn`$h_ejbi`;x?d|;j4sYA-C1Gphal=!Q_A8IE%$*Hw|?s`D?n7?1FlPAHqp<~sU;De zjgEH@Q$5hGY{(W(?H;zVK;mcojJk)bD`jOMXmVinLxz8SC(^iau$J7RPzeN@4yD8f ziTFTUEw?!B6-azL&#Q8srjgdN8Co>Dux2pp(Y=rzq|M+kU6KJk>=J;=sT#Z{^h#Cd zp9s8mHsJ~eXmvB&ep)WRXzfn8rORZqsIizmwoMs4pst!sK1k~h$R*Mgj;3*uA05Ey z^3?t9JUD-ZkC0YXO_5A86D{yeH*48aOhThq{&h5+kNzlq50b*asd+m#{vL|l2LF$bb`G`tVJf-}d ztBnB-6Ib!n6pDB>QCoKB!9-it-t1>7n!iiNjQ>dG;qjf1YzH6*a`$@nHai?YS92kG>3-hCUG;6D5$@>P5wW=GSyvuh z+$)_BLwT#`_vGT5ya!MZDKnGBAM)?4YU>}^T`Y%#SF0KEuz%8<>u787et9A?Zz_L( zHYwd1b%?L3vK(sz7H9ijSD25biK$g+(5dTU-7gBvsk{YQWuc5W8Z8R{o^1mdfsmwN zhu%UCs4LZZAhHE;?SdzO&Zq*(}?F^!bdY{;50?B(x+`d^5NGMOY{O-f8U$qmM# z-tP8SfKrv5tS<86%;w#7rs)CPiKTzU(^&N8*jV(I>vG4k3b16aayVC6mOf2@wVya2 z{JYoiOWDzvwmE33;Y3>cp!=KOV@Br0{QU>X&KoKiBDjpDTwmjX|T)TiDesz!WJ=h=eMzyx{&nW zx;ARpN4uj~E6VKQ*8ejnfR*oOZ*)W4#9YMq?;&+as*^LrqFe6`>^LWdlu^60@8;#^URxv2a_xjo0K-R>pAIK#oLd5%oV4(bj6 zdQoEguWQZuPkY067khtW=GF|5{Ips>6(4GTj4fAU^YXl2Ae8^Gvv9V3Z98X`D-5&4 zr+h^NK~a^yYJba_&S2n$aW8f-gAWah&eAN4a2y*GsFkJK_OwUpCU|k#Fmnaa(QrAm z);I-CTG>}$KzN$e`?3^`xt*{>~Hl&Of|i1_!C#x znYJnNYj=#gWZ3&!3t0Motu&%h=K?Su2xT!P8n(LlPF}#dtrm}65CSi*^qbeJB)8jS z--Be;^?5L;PMfH#%Rr+U{QN(J-Buc`npT##m?zN_>nfhzzkNX;TC6`GI(A#3))E@F z)jAUe#OpZpVYz>*m+QB=f*XB+>!r`fn0yi`_(GtTI%(+$#v{(AG&-y zyxd_i|C-70InoLBhMqDcT)~q?Pr2|;Zi-HYdDfF6t&A@>ZEc;N$YYpYw9EYI{xYGz zXem>Dx=i#w(_XH-c-uMKEA7c^JrpDBjJs1ErpSXrLU~1o6lZx?_3beKT=j=H1-`3V z*JWVU>~DY30+ZY%)=9rPh+|)z|K68R6T8aAa_L?*QD63F7B9J$_r}zL!?ab-uG%Bq z6;=;8ZC`!EGHmvjp9+OKmc|(XMhR+B=fwClE4x=R)>F*^SUnABL?;a^r$7pi)6L@7 zYcJi{H^caRt;=ECSbsOMogp4di0VytMxX*Wy9R$b2KxQQguokHYLuM>dSxlwwqv#5 z>+1-3T!pXG4vN((wMduUym6THcuu;SQZG3K^d6dseIF$cSAFCKPlZLu8?4lmKZ>2i zth6`|&F+eWb=z(ur{`&Qm6ej#ol1 z@`NIWl*gRCqAY;U+I$$i?`>bh>s5GGgyw~g7wu`+nW_Gs4PKg*@WDWfW^X}Zgt+2X zG|S@}W*!}-cU86W`@V_~dzoruWx5<1=SzRbO$DOCJ>jX{8CLLlnM%G8W~L_yVu=`V z=Y!r7u5u>Lq^(wWVP|Ts%}rK^Yw}eYmg#Upr!q|fBdv#QNET?SC{aW_YA`S)0=O-& z*4BR*uCDMIkVUWAI?Ap#Q4Ha9O7kjmfG<%M>nK;a7JFc(L^A^{ zs0)H{f4g!d?f0stOo-ol?$N!tC)WF(gyd`n*n=zcCUGVacmZhk2rdz^gXo-UN8 zmBt=>TT#9>+g`-t6g6Gl!>NBg=b@JE7aGwz`BF6_CV%pVRxc7=<4Xe~trPPWk&nnAQkyK>3tcH;+lhXTMG-}9#t_0wZ&q9quqI zZC2I_HH9tZB&y+Y+Pzhnq?zH@8UszRntO}vot8ukPVyi@z6m%pm&7}sff1tJ1tD+{(q%7KmMd5BgS1Qh#H90!_J;B_reNq1U`80nO>ryouFl}ow zTlO876`HUZ9|C4Wy`Ab5d!t3dqGJ!hqjB2<^3a5p>sLfYcv`zMsXG=JR1O?f;R<%w zxKw`_l(J5!@iy9=gY5G<$A=I~HPWedAXPkY@j_Q8YhRgw@JnpUxwTEvOsfu;b#Jgc zsB56vN(iIh<0`RIsZ|AGqDIT7N)_ZqP3FFYAQBvRXRWBYde+X<*$*L-}+Ehvm6Htzg+3NG3E7>?Gw|A8>*Cdvv?@8{JOr;04!$u~C0S zD>Pc9=F$$imt{WngWwK$PN=AU_g>24%|$cs8xqgIT;(8K(|P)pVtt+>{(6)4VqqGa z!DU-ZbMYGzTdDx)xQ`V}uCJ>V)xM4pv7JD}O*1L$GZt2|t8@%wy*$M0np(FBv6@KgU72xl6c1zziU(JRT$qzx_3 zW^IZ}xqUJwy$;om_U}G3WTp?q<{;}H$G)*T3Ki`Jvx@JyJVWd|WAlIdQL`7l{DZ$& z%rIg2J2mD8JxK13rj}d1tmXC9Zbrze!H}Y0V$<2T?TokqQ~+9w&`mVKSWDlI>onhd z0gq?dJ>A)#-UyUaBVIkvnAgD~Nv~@vX{GKds5w>ESip@k52TPBeZD}RHLrUHhtn!a z&(Ud;MSD=JujWLtSLA=H|C;g@r~(V&U<}ik71^WA@@t)aD^1`aHTWvmrQT-N;P3l) zGnfTQ41c@y`VLX>c~{7#_$gS1YKulW_8=60N?SlQ48!bH02MSzo-?cL`{a_H!8VPs z%QBnA7W}7c)#D<5k9m^6SP7NvI9{A{{n=_=$f5f8eIW!YTpxe``16e)CaEFmT!%AI zVbbHmaqdq`!7sdq&ixK8MD^gh@wfEje>zz9F-hP}EAfR&_S-$Utp2dnD`@LifeN1k zNfLZ`{-gEu=)o-Wkt8b4Ywa2Af2@v1i0(;+(1$w--T!!ZBNA?i;!{DN(P_Bt^v3c2 zR2{FhA^9~HqLF`f)IZ-53qAAq%n27#c&toLM)Sx&Rfj)4#L5!r!L@d9|G~aDauCJu zw!RGxA5TxN`~0Ws_<;?n6B)elV$+-dVE2gP-%h^@FD$k;SL9F}^I`?Qf z;LmqU$Ww+WzQqSpLWA_;2(2|E76uyc}i1HDiB@U0+AHweEHqj=jZz#Y87Z z`bxz``HK9Te;lg>dQ3^+$DAEHA;)(xuU@?pihFI;lQDf*>7je`Fs?nHR&lV*I_*?! z_uhTtVr6-{wDo>I5j4YfAn6~jaKBmtpP2P6xrXA^JCgtGAq^=m6&S9-!DTP~noRIN zd;C}Ue}#YlSNMO0|5x~bh5uLhe}(^7_(A7bUuIHTV0eDN?|5TJQF<{gU_|^jNDBz;O}4tq4Y`gRAQfS3%#$GjXox+wH=I zF)<}XT1wmJ?dX@`YeJr}{Bgbq>!IwI;S6`WY@ye=Z*Nc9z`%`MP8s)dgW)nPq{1H!>TO}S>Cngl8KcRx>H?I}kseA6lh1?IAKvV2MKJ8B4}N}arN2Ps z^MHmuQ=uj&aLl&YU>two_X5%6R2x{(L-l{btwM`~-S3*)R+;j&PhEe62W*niH?Qwj zk>SvPPzsk8`hg(wIFs#BM%*VMRW3jN#{;o9RkrtiDwoD((MQJ~gaL8=n4d(TXEFow zW@j+q0hs%N6e#d=7kvqymzKh8_Pn4k)vGhyc`oi3ph%=u`+7ut zEU@McvMJTd59t2Yvi&sGvh73tp;Wc$8^NdB)(pRAk6N3HrG%d0Z|Q#bK%ACb1%!(d z?gqffzIZ6`r|})~JNU>S{j!Wh9X5a5DW{)NiRn;4^QiPk&d5}w_oQv>0;|I#N4syy z9`mU?k{Q$``ClH9S~l-r*BTwPlekYk1$Q z<-a$&fE4Hw_`)wG@K~Ar=Xrl0XWU*Mi;hqk-E{dS_rO#MDCwtz6F>fT^#%}o>=T?GCo@rrl){)QZg?CYv{XVR ziN4QMB!SzCgHn|(eW_+eErH(_g5&N%@lav1o)OXg)^4(j2aCw!v$acdlgsqQ?)}lX z$w?dKO>UHY4v~c&BM)jMTuw@{glrzSo5I&Q$}0K`1Lxy>du0B%=L%oC_=Z<@u(|(dpZHAk6FH z%xk||uYuhJUHN}Km6f}U74n$)yYmRf&IEI8de$UzOBH}DGWo6$EVpC!~HO9a_ zKY}qAK8N+Y=JN{e;?Uobx^w=_U=Cj4EltoeM4dxg;arYj8ui>X@5eJrv+j@B47*|o z|D(vFkC+fOD)E+gGH5m``Kr&mqo@qw+B~;241I^$xpIFqeEi=?zVU?=5;2~(#KS+T zwEv7i<>9W23FoHfRiq4?mOM8qCe5U%oj%Zh=u7=NdY>$2RO82_7e9Y&AE zPz$CdkDsVz#75>KL}1b{+E$Ms-gEvk(dQRLx=r`nxwn5iH;a!qi?g!sK%P_KIgmPK<+akg&h*{eaf8pNB;L zqN;x`QTncL2ka-n>Hy`Z7m&DS|H^IS7ykLAc=jNdc{X{np654}dDZoo8^2;RMDwe( zhV#_cK9OSMxg(F*3m;Fnh{sB$nm`9xbT0I%=%#gkI()qqqfve0i)7=14U~E5*d7=l zjgOr~-nS#ds-vXe?R|c2`jLU_dchdRkU@X5Q;N%%?6(V>4sQ55PKMy5SXCchf$TaV z=zw1{jjhZsYLmE9w`o(XPJg4->UGPQ_SMnFc}s05o29~hGW~eR!`j0xp@3p?xWLn8 zosLHzaQA7r%&W8O>DaFMr@DdRQHu-yDA&k~faTonoeMs2cK&*#Ded!*0}7Wq4T*o2 zO#7a*&P`~(_tI}IU%Kak4fCkO-ScQeX9L8H$vTcTDy8miI_{fy4ncCRMt*zatF=}W zdUre16FboY&sLq9xx(|(Ei8q=6yvCG4|L!mLLS$lla7ZSV^iOrGfIi>K$wD!E0~>d zT88+pMlYSk~0PhPT=9lX|dg-+2|6@7J z2%~*h^JZ&r*m^t!=1hD0DLk+Hy<~fIAp%FFo*&ikVK6SyV_nz!+0Gf;y}a8Fzfl%wHcpPYn}&e2hDwT_c!HqP$yYwZ3>PT+sZd9QTnX;g~m zCu0XGv8}s9{L$h?nfKKdrxw4?mz01^{Rc`ly1F{+3$NXz3dWH&pDIh-=hBSA;|~0K zxvslid|$0*KyPmGW&Bs17cbwR*AKqnDIF0jz=t`Aodk>G3=$CgwBGaKwXNH!th#qcx zV?nFLnib~gq3cdvbG#ic4|0nQR?($Z5YS9LZ;`&b{8{u~JRN`DrkZtvx3|ErKI|6< zBF)h5hu}7s2UO*kZMQKzRnj}0hf42KaE%eLZuggm7WntpHGG6u%N6=*94*Rnyl25B z7p|0v9QDK?jrC8d-njJ@*6ZDnE4WAE&#@lUfY}$q00TE`FM_B9-XePQonW_hA4(h@ z`vB_rGL^C%TjGD0LpSZDD34{Py-c^tgEs6bj!Ota$N4n+-Cl2|@z-(@z4MeLWvZ(e z?(qfYuUhV`+eDqx*)tBIC_1se<<7a&lnQzzAZGNEbBSDd5sR%w*K z^TDU%xZflzabHV7Zzv&OlN#yq$3`k3Tia@BRp9Y7KZk!%qk7{?YYm&NQnFVqSh#cH zaf*YL;e2u+dvHSF<>TI$^t`W)XNu-IHt5oionsJc5k#kdRelimp7=v=tT53Y52< zee%7d`(Be(USHU2$+%K>93i@GYTGGmw!*Faaq54K?JBMbF9)shE8htq^S`6XmEVcl zR9>uP0RcTTJxrYM!^JKmCi1vD%$u9lR%fPoLcb?K*E_hAZj!<3*!=v2)Rk&E-bV1Cw9wHp3WdDKF^zE zQ|^CFRIbw{RIa<982JotJgc9&Jg+X!1-pb4=gT!4Br03DySR2a7r$ey``mP3GU|3S z1v^eInJ0e^*<2MG==?HtzU!syJsDDK(OZ{FVZ?O6x6%E98@|1-z|Zczf*d|?PILmj z_LA4|QCocbM)bE**vPc>ZeK9UZKM1|mQjC#Py1fri5FJxiJe#UrDD>EDs?|?ib{!U zT^cgc4Cgdw&3wO+&BQW2#Vi~u9+6^M^}z9Ut#`6iS0~?_LIIgkR6Jf@mba$8)!GF? zlJa~f00K;BNAngry{rA%b(!`Jdy>DRcgVQme$YNA3o?rp$ z6(LfJnzKTzY)o>PZX_xK$jL(35oJ;=Xv)F(Hz!NW{A zb=7Dy{v1gbS{hz+J3D`A^JyPjnrOY+AlgunYd7!6V8}omDUEt4=5!x?YX;PX7 z>sZSI&$qpoJiFuCCE+Eh10KVcscg2R`R6N3S5w7cK_@GkZHD(wy}{?rl$5xICwqW zP%bKt=gBjfERcV88jHR+#;tgQ7Jy;3^Bk9YzUj0|u~{!?@iSTx^)On2wwuYjR&?dv zpqr_u?EUU1S1_%z&OYmaJVHOmLF?nwG%Dy52_xpVHapcc_U+ufd*#5ay_*qrt=cjF z5WVUi5$1u!g@6Mbp3cr+5VYrV~66 zG%vvXe7h|_RitZPXVjzjidINN^JRRqR}tfie3oSdxO&PYMf{NR^FfUA z31>VzRpBinn)PxepuO5R-{aZ|fHP2U>yK3J+3M8{HXi)fUH}uFyGwF};CO~IgWciH zUa;ha+D*FZ&GjePWrpSrUe#NV0jkM`?aZsSalC&}^%d0H+nRin`_W<;+Ce6&(92g= z>crl+h9-Fgo5l(sCZ&kMPDh2Cs3#r2>?$(u)n*TzM;j``grc`7;ygl{NQI+zA$@<)?I*j z9_BU+co-kF(@?Za#Bex17(S#>Jr7X&o>)B$VN2zI75JRU-bTXqOW(P69W_XKUfM5a z9M_tP2x^8u-cra6TRI}rF}B6g)p`fBDh{!!Ypi5_Gc|l;Q`$o*c+9uicrPg{ikoV6;HP9|({vvOy2_RHn)FnM4()T2XpeJwbvQy>8r}yO7FJH3DWKxV7Ij z?KVBvf-v4O#$Afw6B_x7wrz#=(rbx`3zv5@iPh8kbRiM#T17!(?~zy7o~vmzzC+0A zNZ@x47P<@}2OPxF8cf?2yTEk?JrVAIF!m;m+jPs|#d0HGg)A`nS0TP}CK3S4DDo0K}mVlc!Kw z$2f+x=B@A%*)?%oD4;X^QV9NBl^HiIvdF&a5|_D@s8Tduqi-uT_Xf`xoURANAt9YY1K==7(k-y3l%sr*|XW!CF|S(%NjX8L9Rz7_=z$r?RMH9;ubcIqc3bbOVb!$?T^ zHCn9SlG~ACVAOmT{?)!hi`;T=49Pl?i>$J0ijX$SJ3w?a%|*l&Hes>o;4T=CbKmxe;91O6r2FHTXSAadRMH|06LDFCF}N9@&&wCZJ^>z1>| zHSYH|f>b!q+1ECP$9*&(^B6#L4MFu*%fp<_M^xs^O>OE+F8j9k8H1qrRqiiOXh5&y zWo?V{q>(T_wjU>-8x41VUL*|eNk<#)Xh!$O^~n^lgPTp`mb=Qeo6L$}G7b9$il;2p z*4fKBssaf1^dWHHHnm0p>r0z6$hbT-FEgAu(eLRrXvobq1E|v=cQe|K+v!q>uLeQ< zw){v=<&(6hG3nm#2RHM5@txP!(``i<*Yuw%}DRG$C619_14cQ>bg zV7S8bP>p#uoaxXM)-KtKV!g%cRC7-dITUi%@#;gG0m~dJ{Li!uyaL(?a}c7jL`Q4G zcd_sh#*r{x()Vf{o0PNuu1;OUgEj@E`AuOIg0jQk6EHX^}l6=ova@6c{IW)KQPxtlfj<)?2_ zyqEGAcslvo}mjo zHdS>v8nP##FT>1j6_+_N5}{HojWcw=;FVh9CcaVa;6wu&raQVD&)ZiKuqkY{*gsFd z!P~a|2(49J9P&vTwqsC)Ci4!ZyG3qdpkH7mbc<>RC_ywpiD~ZXg7!Oi7M`XXbK3JJr@&jvh0bE zmL?y6V9&a_CqMLsL5xP1f$Cn;U(}Yubqkiz4_ourP8=#4sw?{fU#e0QD5DU9Rkg9i z1VXdBU3)s_(TiLNc=RaZ*4o65QtbZH>scr?hMMFgbxpH<|npaDIa4}$^vGfr#V+;f~x(UkEM){DW8 z2kS4-tIesP{5xa|Xd7zqMQ@q;krQ)&O;jsFxY&NO2GFVCZk&74Y&oxF`c&^2Y-fXw zLqaTQk9=&h{Q2vw844-driUAcg~k1+4H#)nJs9VL$d6(p!xMfqW!qV#SS)eqEqtVOivggPNC;}$rQV`KV2bk=A~Ji&t-?yTN1mwDcwZiidQOLty>TJc2M z%+(m^ig>S8;x`%X;d}MYse+H#xlQBYV@G`#s?dwo&uXEfu{RKww6rV|D)6o*$19@M zI-^1jzvFd;lFtJQM;J50_vb2*7JmFn^_Fqs^BE)x(?u$!^^e)Mpnra0`r^ef*)+gu zku#j+bz??1+!GidM&wOW7Ttb-Gy60py&N=XR}xlkRNNGuy|2lzEIQxX9+;4Cmq1Tv?M5 zk>JR)C|gL7WPvxPWp1@-=9hPswj%;XX1w5{BU?+Lo38E8L~NYeHS4Tu|Tye1lr@ zOBpSx&m7hv|AWx+^iqg{!djTBg8IrO!RKc*z%f3yQ@d3Q;_UY+?FQSBlb9gXI#YiLA4dsPZVuA z^$PNdQSE`=?iR@U3X(=C3o@9q1cdpGz=p*i0h zZI(y+54*Qu>U=KV!YbKybfl#gbPr$Q=xbnMWju_3a%iH-Lz$CA5+MJWfmSyv?0)tW zpHqf{L7!!ReqIS*El3TpNRmhL+D*yDQcSyWF)v(+{$!}SJ+wX&fmy&5g(^r@RqQjM zeaFQx^i!J$#)F1oyJH_~YJ$E!YJ9y*!{g_>z@eb43C4+?HF8Bgp-*aCx}g3Mo&}h` z@Mkns_F*h|tI5C@&EY!3WgFIp&-JR)*d`B2E{Yg`aMx)$LHX4>YMa-1pEuuJrCa4r zgoe_X*+ht4dO2x0qtwqg9KVC!Tej!PjLuWFz&(S;P;da?kisBnam!e7(m~5?_t% z$&67~2whjJ2~`yQ8H|n1KufTjfocLnOUWzF;Uy@z&?2hJ;*1S_?v==+Y+bfm_F62UcYGv zcnhhl8mzw>CT}=okJ5<$>0n1JFj;4r*IXTO$sqq@0c=|b?c2RZwBVi8mDL`5m2T%dFG1LLHIXF%`mE8$I;JwD^ zS)~wGZqIi=+k$D6HTF@Cs`RXXttV71P7>s%QUAEdqty7qm?Q+f?$?QkiI!Cr(ifBd_;XZ8 zWf>(Ht??L{s65@MA=&0CXCpkg2mKb;D=;iyPQMO*0FzM5H4_5e5t%st!tRhb3CmZI z)ZmQ%wpMhNXSJm0u;aCV{_ztmGILlMNUHNQ%Hd~*ZZd^~TC7McT4Z+Sa87b^&`tY}{f)^lTG?^qi>}+-GMfw>er7zsZeZIoXCLZV12?I^MXa1@ zKiOV+*tpM&N1YQ%W3wd@fiuT+zD{#5U-x~eR;kpR&_Q&AikJz1ZUjJ^^;T=ozn&*> z@qvxdSKAl|7$tI>;l!J=&Sv7mwmv~)px*#Y>-C)zwO%#U2PW348KD%|DOhRWfXKKh zxN`jAtwiiC(4~6@BWilCiNX?@3b+JJ*}aZCYUYg0_lkn;f!QE0|-Mhp4VOk+L@zq*{6C)iE+#1t=*;vGiPXCN)lX|l%h5Vqx zLiK!zuHSZp}K@K$iQX;qU)A>cxHNhNN1ulozh5kN|NjTEpMH_|4ku#J1j# z&;Y^Q7^s7PrIwBwJfNFyp7GG{?i=4b(>s%n*7k7n!Rag2Qm8HUq#ct*Sm0iI_&SY! zNN^)OW)$<}d+8;&!w(asV!~guKPHTHnntvl3a)>b?9kX`|1z`Etk^HwXXVu>rhA={ z6norI9(<`3vh|=+vp{QKs*-CY>(s0^W_jK$m!~y zzM%xm-6(zfy9Okim7WswmA!E#5O+1r2BjJ`+MWW=gKYa|hU&S9_QhCGTi#NJ;qC`c zzmO>x9+XK%-WOGb^7Bl=_9K=z z1BKuJ6^42^?efvZd3-U@ z_oA7eYs@xqxYz(NiUS*}mw8f&PMn}z?5TB0;N7OG={i08dZ?rw*0k?Sd3u_hb++t( z{CLKjW1AR8>*_KHiy8I+s-^pp;fn(|v>qeJ6wlAG*5%D~57BinV> zl8-7HXA_amX-93Li@OaIa`h;XP)g7=`V}}g$v>f+@CsCK?{O6Vzg;14;M9`LMBQT; z@Pb0hQ(|?$aW`cuKVoI^|4#CPLf9dH{A-G7fE*uI?sGDgVG%Kgp$5$Rk>KZ-%5jZj zjb;gIR_1jHdLm@?5x7c~GPU^>aDUCBCDsA9%Rxdp^s(~~)1~4=GxFFB*Lw@IhPn>a z`2On30HID{n5tych`lA{PHmZYYxc&agIxcJjiy!0Uk4+{(Rdb!rQaf|)mhGe+Zdhr zTQ$Hvm{Q$;68cKG-?wo&Zi7!+X>C7N$@lP`1=hEhz~0Nn?XA2H?FS|83Gx?4AgnO12rp=7i^=1>{JK<(di%J-%6v#2+VKXyU7k!>%mN zkB6^17We|Yc~+USA4XE1r*&=Mc+IV<(Uar;v1XVeC1>S&3du36mAPhrYj;c|iQuSV zVe1YCDVgmmXFdGD-J=f$u50@rpc05ex%ON@&APKGK9C~aA5Cbw!sJnf#{J2&Nvm_A z5}S!d&Br!6&HFm^n~&A2H{rqx?`z`-ufkAx*04{8Hr`=u(d*BZg?ylr6dpnc0ySJL zSfzFCJ-4I0a>;@b8Z{7qd9K^GmTc1JKkG~wuG3a3&CbbAh--dDu0wUaFrcHQ;jm4y z?z|2)D)J4h{r|D|*I`k&T^lfJAt(|`3KEKRgLDaqDBT^>(#_CFORIE)bPhE%j7WoY z=g`dzJ-`rq#`}KW_j%ua>|4HF-~RUT{Dwi-#5lE>0LB=r~-i}vmYn~OCMCl!>uspWg zZ?xr&zO+8wo?>?;^qnx1rqG10t5?f5z38}xK<0jn5IM~fuxZT#G@V#i?6j=|Y){^i z@+tj^wv-g#d&KsCwAQ>)9Kk&-8~dpBi}67sb21y>BXwK4-qtT~j>ElJ@Z6 zG)pxKjPp&@>3zdZ5*z&94dpM@! zvsPh&A?}ZJpX_Ei?y~G}jhB*hx@tr<0(I3L=LZgds{gIp`c8ncVH$J3Jd+Kh z#u}1{YnbvbH|hx$2Km2t$9GZRaf--WD8L{{JzqF4kvcwKqrCsEVZ);F&7HveGTG;6 z%IVxjk&j5(d+zzFGGbI}t%fS8cmdHKaa8DB?k_l9Cv0&~>*r0osm9|+G zGIAI_(BYWNT!TN~RXHmRW-y>;EPa7LOVgi!&dF?myU)Lc1Q(mo#=FnFaRPr*#w;mB z;?pZl%A8DZ&z607d@uCA=404v_g$?q=Yu-OHX$mVg4a(ezXp6I!V{tzcYE~kr|K`W zt@=IxD3LhbLP~gX(q+w3YTJJE;YYICP-=sC&GIy(s0RD6c>Ueu$$Dtrho=Me8FnLo z)v2|PutUUd9J8J!zRmn261}>a!X(?fL`n)<4%B*j_P|z{>qPzj0k(@PUGx2%SYUgZEAdR|0{PUDdmxiVo&9)t6NR5g@WG7OOaBeG~lLdEPxKV-q)* zQHt`V1#$pES!1ajqR`|An#J=$E#k?43X|lj+M3iluimf$38v2H@=FNui&>q8+RM_i z+I!nL`)wuu@vNytCeF`)@ptw4t2(cqyXv-(%ZJ9+%Ql~_&B&NIHF`9z`&EnEr6ZXy+yNiy)6aOn&h%4HIW+IR&hBxEa;=^f@kB+au-jnjMA*F8^POnMnL*|+Z z$Xt`WN(KN{ErxWjaB6d-l>}d=CC0jU2osGvu$w=V){&soXbLwuGus;sj`+^-b6-v@ zI5oIA+ymNrCf z=p65QwTH5`IF;bvNV3vfTphbV`(~-S^NH_HY3%3wO++V4>@hs(&V)JXPY8$A zcQf|^h(=Yo;-h2jy~QT(`DQFy45NVMH7{FmzfgQW6mwYQ(RA9Ow@1x?zyixli_x;Q zM-=BD1fQ*?SQ8m_&U8k>nQVWAlsdanGf1{FVGi21e=iMVtykc-)me^k$-RU&otvlW zTSo?$U>t+AB+oU7%Dr-TFDJ&%@tIbBmXNi{Yj*W{l)c016)$gdHh}H22F}fg5TL_P zWpuyneTjpA$;83>LKDG%(J?b?>37jz>cIU@0OB0ASu!zulp=M7Q;Dj^A$0xd!~&oG zuEI51PFm|BcHYcnIN$yGKKGX$Gyl0}@4sqo^TO|#Y^vp?or`HvSaQz#>nc=RFPDQH zv%338(a;N%{C{lNpa_c+6q&%ud6uTbH#kG!7sbav0+r*TUMib^2G%OKh+l(6b*o~V za$-jF_4rD^y&IGZ7I;8<%r+1SsCdj61XDC&d8hAeUXy>$+m+!ZE zUoTck+3e}ErWuY^#vStdII)(Q~HKWdYRsxCqvfg5U4g0FjIhbO#485}(Z%bNe zoijSe+{LDrUf3$Mn{;C-hcR0{Jv`(Yu);5y$HN821h45U>sZ5TovX2Nv)dWGQva?K z)6vUlLwrhx%Bl8y9@W}vHXDNMqnuCo8WcBY5sfPNnICl5lD5O#{U0^rM(Z za{Mz)+(*$LmEl!2MP+=WPo9;`(;B$vWuu0l*{BBp*)I-d@Tg%a+kIHY3U5C6W{w{I zlt;9q@DAHnJeiM{nne23xTy>cEh$}5krI|n-SXXew&_6(9y&{%?QzV>0SoP?aGGy# zyaY9W=$a)%Go{LG&=u-bu$;~9{9ok~d*ks9NQ=GR#!Gk*p#Y70b1@88@2 ztc;6P(sDZK=Bq>iRzsgo876Tnk~M*qsaqHEAk11@ud8Zl5|sUEotjFI zQN{*cFxIy;!`_Y{voSio{HI-C2T+G!{{#QDl;HUI7UAqKgHK$aM{P%S*z6q6&GWkF zb+upAOqQ|K@7?hwh${{i_yQ%mf6!mmy?lpJE9V>0Wg1Kq3zPtQ?Q%n9c zmZmq%mG~U%`oq3uHfx;FxM1?ox-)*E$zR|0@sknO;M#YLhv|Gf~qBH=R>uu4DZfU7W;HhMJ906JwQ#7k z3)n=5$P3#>w_kyEWa}c^q4J1nPEw8Cq*YgDusVzgd>{vzDT@p(EgNgj;V12uLCS<-2RrJ08?i z)fI4WfXF)voF1tWQ$gllwc>BB!4OAV9`K^p;|q5e*h{tP6mD7SIf~X5^%^UdFeyL|?a`M@8dqr+WJ)EY`1tL8p?;{u0d=U1As!fHIGHM=1>tfPy{<0&vw`%}=@vZy#E8@a<__g5Qaon9&Xs&h! zvRR0+np=`WH#2Md%JKXSXi{_d#nD`SsljD`kJyC7ZXp5YD$s|ogXbC8-|X@d<}j}5 zGvL%CYcy^_FuiYLTV%+@dES24w-y#>K}x5G3}PtGz7)HeD6I7a{hK+AR+&3c6BWOS zLlvgJ2D=}TQ%c8|;A_xM=k*}+5wEp|3&u1=&4yW>W|JF(sZmdNfkrusNM*aH zH~v>;3pL2O)iL->aI1{oX z-jZ&8h~L56QS33p@2!_zz{BrIOulfWN~Jk0HK=jht;DnTAIHL=P2`|@goI~*sKLY~ z#!Ly1gF|s=i9kg|(?Tvn1z}a)DB~KKMB?;QE*naRKXP>K8vY;ZzRnrg;5Ta=rN%NgDzaI>(&-8b^2>3XwG+$lv)aF z-ec|v-?AMDCxQBaNjdeE(sz*rey z0rc71Gf${PAI9hu*BTgPJiXhm`4})rrpA_G6&O;4BYgVKV$6~WH4S5bAHwy%q;b~U ztN%SpW(KNNGP!t^yLV){h`5k{wMY!t0b&>z1$P3hPXT}K&S|v z^ESl9?wp>o4$su4E}r1(ByB(#q0^JEO;gpnbMMwHOf(a33G~ej)?08C|GS3wGPz?H z@UBXhCY}0!`f{3oevRm~bC(7iyHEcsK0RI08P7)hIfm|d*tagTt?$RXQY8TCc!uYW zrgGncBDA}pLI@d=yO&tN+TD4TnL!MJzc~5Slv+pSx83DMaz=$7?eC!#vTzE~GotE!2D)uT&Q z@phI_HSLQnAlZ2((dlH) zw>G!@%0Fp;`5Z*EHlJ%r%{esi#ioWT7}NkR=G#52=Kb1-uUnq4*+oH9vr}1`dqY=D z`L=UgcPcPeDwbidxz0J zGt%d4A!*?E|F);1PLs=mbAd8zYm*;j zt8EABw2;b9hfyS4W?Esg>}Zc3BccNLScP>ZDBQ<}QPM|Jk8E}viwb0-etRqNNAS_*uT}fN^)uGBc2NtKAPy@UHJYhS_WDVUm z#&%ZX5!M||jUoW-Sr7g|&z{na*rA8)qRtAq*hlz;x4+bIEPrDAiDJOEh^1X2(%>~| z(N9(|Q+gBEruFy61iBx&%+o#9y8_`k6{)O$wlSQeW1>t2dPN3rZgU`0?-Z8GS(q=t zV?R*Up`acSM*W$hI5y*%KQ#{>sR7rkdxS(%1(pRsCl1Wat)&PZ}3cLI+@#_<%JO*wh9qr+t210i7MLfyz4smIe z(^Y~x$vuj9ChPTV>OkKTtw>6o#mP#jbn1>zwho3QDWro>tXFFEeWK|{OrKH@H36+6 z&C6Nz+%|(EAgEli*nd`VgQspILcN}UL2T7&>ZG0_dt%Vx>5^*pAxC@jHcO}c;~tP6 z(dcd&#IdmxYB8>*H@_ch4tx&7COt-z7%`LR%VMO7^|9T`s};V*pVMS80SgYrveETy<3Nshi2cXd=hK z*UrAhjPaLqY_r`>APd>E!lF_;dCAMS6t_cJzu+BtWPv}lgEF>N11cwKXAG#`7W9Jo z)^PDR7lJyZI9A2JTp0{@LtP<%_xxV*hCI93@+F$3%x zu;JU~ZPs{aAXQ|PH`N@>)g{sd3mYwpf47aSQm)s@`P^_=OeeOh2H(dTix6Cy_egpV z5uP)8&bcpkK{-``LEz#+yN1a}tCRDUAr{7lDUvdh=>5I;>TEq+0$`SZV*fT9mXN$^ zB}XR~ICX**R(HuOivQo87CZk3BuX>a62Hh)&Rdg9&F+?(Y87hHS@R#J#&g-hQ|Jk#X)}j3?Wdu-m%ovtz!h zV}0YNwB*k6!3PZ!cki0)e|uD8sw+?vAIY!yOnSB(OJ|wqU6Pd~U!q)o1Y^vTwf@8x zplwsSuUH!s(puaDW&I08fNd(W9%5<2e!}8}NNVlsW~-QcJ)1m#O*tz2bZ8hgchf%J zVd_F62Jdggh-?`i{BhN^yD6is?v|A-H|jL_fK+{{ykl!;DaUpsB<-4{=+!`p*3Nbs z`l7+73X8s=x8=7GdSOHfPCt6+A3bUL-Y=!96vlOWwHmMgAO^o|cv65j>EkB|7lqd< zL40pTyS$_qYa4TaapUK*rI6H&pyb|oReGwp1uyB%0FoYkRN`MNQp|yAK>Y;nu`oTO zVsBQ+u-EovLMkr($x_C1s=k-PP&mVgTdNZ~_)@LA-VG0ajEiZr|Q+_b1y_+3S)0!z9p9!HmCY~eV6K@9yDjZlTsvHX_12&s_< z=&K2j>k_?6xw##a+IQyTRppOJ20K088&eEGuNM@I=p_oT9}-xA(bf|5Bx^n zbe^oeUO@1wwl~1}t%!*n&qa!qk6WD#PMBzg`RpZs_P#eQXZM$XVTg@>izQ#H$t{Uy z%}P$4WKp-L&iS!BJ>ug(F}EtB<@gm(n(FnF=1Pti{D+Pyr`t5>{7&xlOq=v_;^)M% zq%j6!>e*&JoeG)?dYd7txjg;mSnNwR$Ik~Z_Rn3*6yV>rOi-sht?nzfa)LUoPd#GY zE`wx$`)rhT>DV=hWL{DY6uba=_K} zE+Qc&_?S)OluwcwVdb+Km~RRx5Eb(zPhwfv@X$7Yh6kqO4Bxt!Pq^2cEOZry&s{~D* zSn#bq0>o2AdTCt2@GHvmw7jbT2|aa$XHS?p97@p4@?PNU(fvA+yQknsVaFKq-n**Y zRK`Q9Cx&Iy1HEU~g1_SvJMId|T6uGF+8;6&eZB7<^s9FVC7@1V@gtD=5(=25!5a%y6Lh=w z$H}pXxc`=Z+13YP;2^DtZ_!{OWbmt3rypU%UCbj?Uhibym%= zA6WvFM7`@Ujkl?_hoMEp9|q2_ zzR`?qzUA$95(6|=XeZ6i^$1=a^d3A=Yk1yYcK;5Ke0kfd=VIV`QCP;5VyNYRNi*t4 zj4Z~jjnx`n%yKajNtC>D0b725KTmmg?xe1~T2F#bC-{|SvreK+jUZx{e{+&gyzVxm zkG@8OQvcZPH|kHyp?Rj;>qTeoT_xw?o99d-n^WteniWogd}#4b(di`7n;bmF$xeG@ zrA3YYFgx&|>KvJHgK0^fPjx+iMV$lzX30yEh0d3UP1XY4xqYmO+6)C@MR_W|Lu8{a zV9i3ep^jc#0s8|dH{(&fwb5@9@`$zPkKWCWHG-aER2yU$Vi2e$nk?hlZhPZ+%)L$( zv?`7MtTzRpzFql{#Du!|b%_Rt{)f6g0=fdx%v+%uwWf=WiOn|nU*j!*HtzKSExOSs zd~I8{F>6I7Gy{Gh>rS+INN@`oGwdRap{$gy}EsCDw#6>S zTLX(!l9I)D7hI8`_CW>MGDp*;C&b_DWs&)lkuJUgzGbF!H0fB5pQ*#BzCRoH;aj15 zi2z#ZF6CT(RXq0Y1;K}ZxsZ&lF^1mBPKxRwwwbc+sQe#zI}4;SDm*XB7))~(u+YMt zc^hnNSS_9o6*{9gSXPVQhsQUN@O7owLNF7YqCsrbXa+2WMgg+WA$B&mM3i zVa;xtUToEPDya6CN?>@f3j1BZ#z!U#*9fq3Uo&C=lpnP09jd^83uSm9wrSJTM$0kg zTZ@f{fi)&U(<7xp6=-PB)Ejkr)h->w0}Zhz#W0W=thT1MUheqzZ$KtE&a+QGz=Eo# zBxaZxB2%J;Xfn4hX8}X7Xt9g4HE6OypBb*zk}ZfoH^-eT6kIX`l?7pu#c-u1=j&k% ztG(qFeW^z=AOuiG->m|FP`EjTHx^ne)n58mu(koC zU|I?O+Dq@JuN}AF*D~3Tlvc9bo3Z}+Lj+tc#nFfxlx(V0ApcdJL9NufSetTIqtp(E zT=1o{swR$Su`x2!J+607kKizvUa@ap?M$f~*!UUx_^SlRR0)bHb#jE(=pOGJ4%bU^DVcQMy*l~icO z`vqol4hSj>tDrG!7$gbUWvFp;-!a0*4B0mvQ!Hr z8EglCX~t$2;r|@-AId*&AGlS^!=xkz(NRrjhp`?CrLDDDm>J_z~km%3Mr(i%qGK9m~UN@TOgxnIZLVt1^nPJ$NGTKMG- zT;~LlUJ8YbmufO7;O{@Y-^!>~Via>~hizVCy-ldJFl{*|)j{E~H<--vJ~nvgC7#-U zI#3ftXYq-5Xa)Oc`zJ5v)Z(8X@Utc1a>J$8jEAJxKYxXNeeLC4OEb&_{c1D!U5ABROia9n)#?w640+y`p6#SQBe z?GI^U4!QFsggw!9^yNy5KcF%8=+f;8RoWFO^1pVCrV)G&P^l#l$+}0_vk*EU=ZQMF zi9b+_ruv`&KlP@1Q1&l>=A7*8 zIO(#33d;du%`7+4(z4lCc+!s_vVPDdTHuMX3?pjjrqh3wuTC2L;#+3^_%Dr5t)dz+ z^s0pkF@u6WTl9Hd0dLD*)o?v;2raXW676(w|#0EH|(W5v8{6~^IQ_4 zK^3}R4NW4xampFK@MtLI05J1^x!X#q=+#fqDh2!m8H|eBkRK)MI4uW24Cuwstp8?> zr|rrCKhIJaX7-p?I{A-bks9TnzCYHX7h`|6c}=GJ?9FA3=v)Qi*vUftw9V@#)XSR~ zQGO1U`H2Y@g@GR)$Cn!Y)`sWaor z4{m$>vQ1XrWCg};et2@<-*}`JQnM!zryHtZFwQC3+g663SwV$kVDNy(vH10-9gI=i zlexMsF2+uX2?tw!VhG7|xSbTnKTCgcf3B#_H!k27)w#WF;CB+q0FreJ;;()yd5hs+ zX8~kZ`%7D1*En?>=6XthB%4qF;>&#`QEO*?*D~RwJ~d-$g%DK2Bd?t0UHwvp=Rsc$ zUuNKNiM%D(&-Spd_q)%c7U9m<9owgpC=;EILKvY-tW=t@XK= zUIh#fGJxe0ki~)!mFY-0<^sutA{@7_rp-=%s;sd$@LU?WC8 z`IX+mI(V3m4;f2egO^ld&}k}iPcyS-r_BqgM2OH1S8Te|a7Q-Us_#&LqY4y>!5^=K6f z0;P(Y8~I*vq?n`ePEaC0wkSP1tqGkN&V1tCxLwpx^ANvJXX;|uW%&h;zw!P(=h4pw z{^5Q}5NCsN3h%{EfojoMZSA6OMFbkR{fBeXUxC+!e=GSPuV?&dblCI+_rB=#i-|v3 zqD0#3sOY>P@A+2)?uX4YNK8QTz{ayOgUgR-g0?S_L~-#vg>J^GG2_#u7;QLCfuRvo z%gWqOH_nwS?74_x6($L~*(m<9*j)J{g{O=%#TG$N(kD_&AjdPR><3WO5~@7E`naC% zqj>e+eHm&qMe6bvm;2T4u|G-pu?Xjb zV$fJarRK^hzB8JRsP3K4=(6cSQB5iq(e1!D)~l{ClL6|jIz$LfZg&fF zCOSNScRiT;Im&FGb)zO2ZLnV#v0+$RV-=HCt7;E4m-4gTsNy4abHk82V ze?g2lP0`pp?HaS17x+t$T|8*V8|pe0(E-qL?eQWGEx{JzyM%Qu-zw8r-tr|!hbp+P zmrXYtum`^|k+!CQDYi?g5AsN!ZN(yyfpqiL6aXu$Z+n%B$U zDi>|RA^zqTWDnH>%3gF66bD$$SCwnAnv(7*5VQ?zGe-(VpSzHS8b3#M=DKJ{03TGknJX9RK?Hy2OI7YHL zeR&&CP>mm`0En^UG^Q?Y{MZ@>YKK1+G|qln1_1g#ks5%euG)aoUVRuW8zjiwD@IZz zp8k1=H)NShS9ZQ=;1=gl)P>0^>i95HXLdh9KikxI3vZ2nu9&}7Y{x=}f47`Jq*1d` zw`3@9W3^=&CK%6B$s8*R*ZTfOadY-c81!oi=(#0)SIV;tzec%NPvhICAJ!I^?x1YR z3G6$TO#271`i--D`=ll$vyfd8J2t5h?1wS{o;qds;|=2jAsV9R`^YdwUUaSa_QQ`J zhSk1gA=SNkJzojZTCjHWf6|mpbV%2>M%~rxbeqH&{HDbZBH`z)th272AL=mU(%3$H z_5^#Z=O#p_mNgnN<2xx9DWn09!AA#~9?1Nr^ZY!`gG9Y6J zyc2l&&KuesH(h{EF()6e+9Y4mTiXcFD*GqkTMGCpn~ z8TnXL{@4X*;hHMI2;6JUH?vBDE1$P>76%~z&DPp4ZqzCNjom1nW9Zn?(G+smtxwB3 zz>-TNGIX_ic;4Tdf7p5>qja>bB{EU?z_0#WN z13Yba8@=oOO{NYv+lRRNMN9;z&M#Xvv&F;OBE8=x!Q}QHv>H~sv>hR=wTQv6D>QWQ}emefNfl>Z?%tGvW}lzO@`>c@FYpkI5QBCRo?-K0Ske-2kO`p^X*Q4|ySaKmTJ`b6MPtAmEUwc<)PmtR zC1Ikqax856V&T&s;mELDZo@0uafi)iR*7h=`RbhXe-6O2GFP50E_TlDH2L2Sfu~$Y zbx#ebVZ8T{w;O}TT;%-J=NfhSlCMicj}C*KND@qY6Ltmx-)>V%I=$hxQ@su=z6;emspnm`qoQK z%0i>Zc7|2+$xi=?{N4_Z$oV@0CJ`Q(OGNGz>(;gQJ2Jo>?AUV0&7qB26$hfdT~mF;n$3uR1#kD~e;b#v{5Z(*$c0qtBxiwL{I$;IM)mroJa>gY!v%tIil~eW%JV;JJU9e{uyZ zX^h+Ui{LUqmcz>njHq1Ja2dPgV;_%x6jfYJwA3V7ydFIE_pY&jIGX3#F^@wj1L6M| z#&tlQ4;>>ykm@~(AO;gT95u6U{CQrBti!QB~iQ1l$v)G zlyJUhqox&&2+qSf<>G||cxVyCe|qwN7lxWR7%NoyMX2{E z=70QZ!s}0=3J-ll6@ESX>oU;?3Xd7+==CK3YD0bKOJ(wtir{`@JnG9z)VFHdm2>-J z^44Wy4@WJOnS3p?gJiWx06})|FMl;YEoWJ^|5uKnct=tS{KHukv4mN)f1(8XF8BO@ z97VtL;OZmXNXG-!dhsrQ z_3F7Uo^fn_HBf5YkWJL{o=Ec=E?34mlhU1f-4W^brck{kW|N=%zqaNmzd$ejD7;S< z<4SaxqlAR@u8x5eD>2G)e{mBhR?$Y0d^9|E2Z{e)RF`~kn?Yj{(}QVm!3T9iNsWbDAj2mewT)}75+R#$iv zd^NHMgvCwwvc`H&f63hi7DGyWCiCyg%L+&BI`}I;-lCH}_S8jv_|EqS41CR!MNVoa zR(o~6o&6TnLmxm0EeVEr!LBO-6bddX#}q$&rsj^`Ncnxj&jGx{2>Fcb{(hqgK7X$u z=)2MjvTSvFLjIRWa<>R_E*?LQ`Sj@%-N#P>U#h2MuQ>7Me-EuI8X8vvjREuPsmslY z0sv?-q7G~qx$ZfTTD#o;bt*$Vdcv)^x<{%KjvoH@)2-Wg9A$p^jI0xbC|ZP~yq4kh0e@NBv^|kj<0|0o^vF(O|OZoe2_Z%1~F=CmDDOKc(W~%<{(|`D*@a=0{ z22Bgm{K5hN9`zr;j)B`%@QG5TqRdZ$uf%5L#SJdVC%G+}Z(M+uS(rO78}rAnQ@*Cm zyW^+O{qv#NYE`{#*bOeou()?G^IHyKKw*+QkK!M{fBxx~Z^}aqF=|XkjM$MaZswa@ zQ1}=(Kq^iPx6op1?#+W6$n`&p_>Us~qlo_~;y;S`k0Sn~i2q9{!iwBaVc-IaQ$FAw zTvaE-Gv(?}%x@gcd!9gH0(aRS=XxvT;=@$Cz@zlG9WdBm{s^%OSeU z)>zD`ZlV<>lvWmtgfN`PR*BlN8+OQ8n~oe#20rvM2eq{mrOhpggk0kAzj!NjJhOBD zmb8>;LdVTMoD{BkL{Tdj2FGV{+8U*oQ)@k{`EG!8?9pvf`(EaW~gesp#tJK6>Z!nZa}T zf6M507KARv6e7ZV@P+rhOFYb4yk@q1(ripsEfhM}Km_db`~o!u!R8wVW_MCa@(FNOBjjB@oiEiBb;ZEaU>Gctj7W_<>%zd{#O zPP3mE9L2iAn&95!P8a2p?V0GXc(VBWf0^eeyLba42;;@-9gmCCf#tXvSZL+InZCyM zlo@$MAfXmpgI+z=s1Z(7cut(o_wGx%o47Vc6%En`H!#Hd=)QJG=d}eT+Ynq|!Gkqh z_c-MubOH038X^-q0HYKysKnISH)&5=a_@!GcK{zL5q-jd9)4;cZ2p}M*Nc6{f2qw) z5h1>lN$b2F4ctP?mE;=;X`u)M>0=gt3fGLOud-u;VuuF=4?U^8;iIiXm1YB{E79^s zsmDBF@1tqVCZLVSxf))%6Od!1ZMl@xk!4n3cBjuki66JhU75SoLz+E0=vyOuBhfNB z?)%(i@y*|nFm;C(W30;kL}+2pfAu^OViJMaX+p|(1#hVqPd=hSayA)cU3t7_zvw1n z=`fPF+|mBY*T1rQ+)n`z)yNttuoy#Qlt=5Q@H9l^&k%9@j=mz&uzm3YEVP6+i}uGM zuD$s{VAS+<5--9muYLbLF2iMZgSgP)ia$HV^N_o)U{-xO^&pyac}*MIe+3oVrKN38 zYk;MYVTy4rI*I++g((N$B7OLt46em^VcpV}e%l$54o7d`h>yF&PWg;nWmP<0!1C>9JAn@ht*~T=;(lNg3@eQm><#zzs z^HZo)^AQ&px2oM7-@ig!e>3Un*i4x$wIHCMb04;|~vwlXpHvuB%e&)%nSA2r)e^oAUp3y}HJ5q2b zV z?FB@OH%`}ObJA)g!mx-r=vPJ2=oe;}&qKD2iF4p%LSB5@K1@}h+Qr!I2jz~FM~C0jlovNgtWxT-Vjc!9|}dm&C;a@fx!&s4L+ z!K?@Hy$R%=qe46vQGiAB-ntrY;aT39{$J8~+f3q7IZWGdQmF~D=wVPY5tg^hpE8ynqp~CL&;j@7H=rukG4>c z$?m*3e}P8t8$lFC6(>p&qQf6zC{+ZCWTM1xpo`ZTSa-})ytV9W@_v@eZvv`$R5q~{ zbByAu;<eQAmgdQ1nOfh5m`lyK>^ zf05G7d=kLWpy%}RCNDL}g2D`w#{m}cpv}x1`A-ZpQo0JOLKtI{yv#HcJZ+;vK_A~m z=E)}Ox%l2_WrOx;78d*xy4JEzrA1*jx?A75`TDmdk#;fFct3Oa*64clmXXxuXyZ80 z?Iz2v63wuT?;k4ih}XIQ9l0E_E+brvf8m${cJH4nhKYbp9U^CEkfOFSp84>=O_&?y z^_|V+R+2o~u%Muh{Pbys6gybj+;@+O{`Nz`XDG-_sG%GUFeb`>5S?p$oOnND_+|Q@ zGWk_q(%nDbeW3_5J2U^N&oJ<{$^cY1%dA2`MNb$B^-^o}7^?MnD)3U>@6rc0f07h8 zaws(4^YfsDS2@B0eIxw}$lH8dqyDWLVRD}5j7rb*e~RcrZgL||%$T6e^Is#J7M7N# z(BZ{?w9IeRz0$ZAv+L27yS>8`i_WmPbn9OJSug{R%%9b~aqCVr&4d}SdwV*^@z+R^ zHoEK5IE5lE!+tk14RzwlpBHM7f0UKY8&{uKy@^*B5+-n4yN9Q<(Q|gSmZDawwDuf> z%RtFXCgBzJ&VALfYcCGB8@O)ry@w5A$2;>}?k=!NTS`2n9}DA%=9e$i(rCwWy^YGI z6POsQF351Exq%1e%lQ^yMX=Kf5~~nCyG4n z&vwTqiS`iKSyZrz@3p?mCF$ega-Be_@*)X&FCXEj>0=#&WP*u24Vbg8H&JH^X1MhH zn>g=gJ|Eb)K=ok}Qebx;h=C3nj&F3melilS+^SOfoh?8{d5f*V^0uH(BBGHprMmRE zbHeFl1i8yW!74U{1u^xwe^8et>?SB!<$=W6gSdCNg@SpZ9uLG+Wj@}zfyPvkKp0!I zdz7;LA|GJ6~gYvUDP?5mu7E5Xw<$e;;=~n}*BM9Ul0!QH_Rq z4nZhbzgv`KOUtLV&+uS{eu|Asa+a`{xj}|nygvR0TKbB5w&#C8d-#E#`#xIv(A(`A z^U{$#`KZ1i8$490ale`GaX>wM9(}3JAkP$hrkd7pTGsch@njmZlqpFb0h&11E3RA1 zFm?ZU1I+owe}^*Q;=k1_i6K@38c*5}vjlz|(Mk6@i7LPL4d$xL%Rkl9CaK39rIjpG{iNN*dCGEns#0yO2$`Wy*+m8w7K;RzIebq z*gC?W;&~xXB1)>CC2?J!00uhp73Dj@SCe683}U zQv?JO>b9Oro92G-bz>UT8{dCc(F=_11=jL!1CCYWJ6eLDC6f7Q;%?=LG(9s*ag%Gj zI^Qt$G3w(&Z;z5cQ)DV{1DMt||CA~k80L0tf0bsv60Qfp^sMIX^MsbhCay>rfmMdH z_6XoI(yZQ+WHWd+??>3SIr_L3+K~ianSU`K#ls<(F%zym?0;?t(&Fe_u7vYX}4Q;f-NSL8Lyajb>M8>%DsPr)TSU zOHJ^3)_&VWleoxUmL@hreXAiuee^%#+K<`C?|8iUmiGL?!%-sF7h&ul2g9w})JZb@ zl9BPM74I)p-xb0DjYQ6L)^FRA;L*y3dta(WXnQ*v6*+oe>?_)@D0Z59MGCCte^2^c zH)WM+D@%*!1Ln^ptt#hOG~dyA5zG&X@CtHIewbV8m#6UD1J=T>f|r)0Sq3^T$7={9 z5zsm+_oEH=-pXtN{u}{;oOEf{=8W#5eN(6ohFH#G{`q-9xy-@CW}_$}hVznYizDb9 zBQhpliPIg~(qnKLt~A*XcZ>9Sf9C>V^Y&{`MIs}(bwq|gsGgr4jg;G;@U5ExC#44_ zv@H8h+d=Bt!NPwQjLwruk97Z~Ga|k4RvdxaK7;(9+LiaQbf|3V_mlX9?;`Q3OZt$a zZ*Iu^bZ>Daxq;15yS_;7--uTDpmZI0X#N}oryTW0KnX;ItX*ea5V}w2f4UymNNlo5 zb9z9177X@xDt*FPbn9ONhdp@Ll5OW%vBY*k1HSt|F|XsXjw=^MBCKnsEY=Jv4=sTW z&siEbhId?}<^Unl^85yI-ul!SvDg*U7Da6!b^W+|8SVodrkR-_%F)U4Wy6u@+MQ06D*@5f* zL@~#GMA7}s()VHt+r>?4_9<)FT9{&U4%=k*=TFT{YAzSL#Fq!fv8`-A7w@tHn+eTQ zoxNQeFI}`dPiAaK4E2frOe=u$lwKus^9b_M74)75;(gEjq|j?OgDlrh84vDdCNb%T z(fTPc5C59SBzuk#f7|amF$WKc*Y~D&KiPRIALY)ANYh-JJ!AZ-=4{`c!>Na(?mqC#h^T)Ax$4x+;Q>y0Ard4xsj;I+Trr5|JHZRrul7`s zAW+h+!I@q~7%*)H@n5c6*YC19o8ydtMFRH_|oyXpMnVQR;}yKK5b<) z90E{3k!o6s=zDentpP`k3jgCvl^JpwhOo6@#NjTc6|037?Qn-akA~mk- z^LEg@&LlqME0v8U2YmRJ()m{75rKQ4yrpRVPdVklX6rK>mM^O3&m0*rVyBD8efoBW z(1C}SwdaM>GaOi6e=OJex=`+!804Q2`bl*oagXuYe_uR(Owmta^UVNwVd=E}n8(C1 z?TpJM@U=4X>1Ju}FOiU?ScOD>P?u)#fsFj32r{R=$m;*D1JHZ#BMajRGa!E@=ME%D<4s++pBVP{hZ>mj3 zk8z9)e=p&EK~`iaDuyHlBB&n5@Pwoy>TTtt8%n5s^t| z)Ul2Jl82CsDQt2+5bJB?=u5{Gm~ zLq_DTHD}#tQWN`{sBEK5;xy|+l#x0$ibN<19t~z`UL#-)jB#2HAvEynV`Cd6Z!^;| zn~N~|l@OJsT~oThbO?@`o<6v%Yn_^P^=D9@U>wvUE^G?OK=maYY`y-A&^T97jz(vt ze?_)Nof>e%klc;mW!@Y?+&G`hZokxq+FCq429~C`XCqP7)~1=c|CzU(%fOU~fKc_k zdzdEFR=Rcep%|IRMVaefi)4jl6D&=@a5E>~V!wQZ3?%$sFE8G^-es4t$^F=>9Ef?7 zp!|{Lw>K1oBE13h{BLj2!e#KnnthvWe?^XJ#Aa$fH7p;{Yv|lGOH1WT0YLXA4G*7J zCLp0h3ZSTkOhPIrdP?fIdcIzvBV}b%8Ls%TcCkahJuf{n!g*YeQBB@Y9RvaeN>9Nr zaRl)@)9oBwQkuWBda3G5r;`Dusr0;df#ri@m>v1FEX6 zY!YH{EwqOa5Xi9WVQ~=N06GY7rvTi5+}ryOv#EGsisld_`FL!=Un$_$MrGq>^Qe^7mfYm8K;<*EWeli zuvS??$t5WXiD3hoOBfI2!oy|z3b9MpN#-%i)aseCa}PA3hu((jbH?-(IvmYHH7#qN zoBugZP6xjx36c`=zh5#WOrF0&kD`=XeE0ztXl5>YTc42&CCWSIdG^Hoe>888Z|{@W zasmapCa?9urHN?%{;~+tljqrmFE+AvnpM>``Np+v?M^7Z*wQx0rLAx!ZiuQ&c`f zjv!2DzuY|<^Oa`W-z7fWf2Zv}2Di1x>pZ3Kxv)P$G{CAIQydOLrpj;7ir`y~Di_ew z{B4|uNFP69?@S}c5xUUezD`O?ay?q~F#0OdJXYKpkCr){JiIWR8B9iC+jQ(wX@pNQ zLY9XuaFoHjvT%thTj?pXx6sPIlbzl+TA&;@F>K}u#gUf;>IsEee^&}18P3HFM2J-K zdM_Y>*WPd+kR&i^1`}F3U5<=As`I=-2BR;5a(1qN!MeY~0C>-Er)fZR>H-}{PoT8< z;2R-x&jYa>9-phTfYULcZODu*xTsb@#bEG=z!jN`5z?kRi7_=DDFv<(g7uO7#J(eO z{Uxx_gVV!Ve|o=zf3!3b=@S}_>)S}m>JN@?Jb~8h&l1Dut`(x$)JT@JPGm4{QqZ_y+*#68I}mu zLxY^SW;4A7F>vBcvcPpkb7~L{g>X|HshLf(a8t3(lr{4)a}^I{TRLwqM>{r#nk=14i*^SkHWzQQ*}fM3V>GBEf!AEhX;X*v z@C3?k=6z&&P9U0pj~hoc|3UPYk(gSH^H7s^Jg`9Lf06VJwD*P9_kHwsbRU4^yIIAV z+yInJ!0`n}tcoeQpF%{$$kMVsUWJdDUQ~SE^=ty>pS@#Y`!QW3AuCzafS^!Tb@kfz zStj4BLQ0i8slk*gTCB^$>Ccq6@@TYUrwA;tFA@8kHxMN@qrxOsp=EqOyFAc_%!ND< zohU%+f8_x(>6;@c^q627%8}MJNcFl&>lQBE!_f(zwru+2DK~N#e<@nzS}9kkF=AOZ zvgwtLE&{qY4KTz_9$no)8yGPdF+myb@SQkg8UA>k-P&MOH_$3QMr_s({i z#AX`Sx#r^BTxAz1&Yz|C>9zAinPhIE+)??6fAa#Yr>}qS!@^nWT2Pz%9JWYm$S(~D zk{TF!t)i=IJvu&$A(2D*=R(gdUk2QFt@05gxf0^yM(np@J^wO*4DC6wtJc1sQWB&3l&PtDJ>kl#AU2MehymU zf0aEyyNMWA^fQq18Kut5>vwh@Dv{P@xvLPUijPt;c>n`bwe1bu?e-lv0#pZ(8ZHBH z)mrZ+7mU3?bpTaSs(@k|+-}qXgzC1|)D+EcjH|iP_PT1Phm_6y7vAiLRClg z(wLxS9%#+4-K0N6ff?0`C50)bDqWXMe|r=A(nGaX#1aJbSh^(>-Q)rgs`r?Y8AFwM z)h6vfiujKr{-cQhDB?eg_>Us~UsnV-G7XX>_u#-5x(II9@-iJ2FW-0pVMwcSDci+v z=uFplLdAEU=CnCkfR~8QCD=$I3@Zymy z54a6=3HD7=hiO#TRK*K7EU`WTSmXXrinQ?Q>1nlXs3h(6D;u!Un{s(`e_V^%+JW!a zKb$Kpot>T8KOB3vd-st&JN6eh>JznWD*?=s!DkO&N3)FN%7zi?dD2@Jm6U{V-7lVB zad)o(RFICECRtLhM_01MlbshXde;%33 zJmb8(M@OXMrtIb{YSbg)f0eybBm4?6G|5}SJN|=9ElS(R(aULTv|w`W28RAR!wc0m zzWsFqnQ2%M>Ds9soY!VDtIe&a%YWt?*J0h--VXDU@>j?Kc5MGWgwXD7EtDYWCRJR! zm9bNObnF6bJhBQCk7-^59=Sc_X}li5S^Al-PY$0Ipt3sNwPhvGe&ijjS5 z6y6(J;qu$e)2=CH-)S>43f`?|fu6rI_!&U#2J^QcpCW2&dmmlly`JK%wNrb^eWQ}& z?K`{jC=iKmyq77}Djujr`pWKgb+Y23=Nc^8+gUorrDIAqN!nl1krCr~V_ClAdGIBh zN9pi7z1N^qa6un8e?<7|YJ5i1ag%XR>!Fy?<&2pT6|kYlaQ?MV z+up^hHwEZwL9{>UBv&Q2h=aZQ2wKe5?reZH$Lo6RH47M?t;VJ1q~HHD7Txl#V?|-M zsGgrf&SK{9@U$abcd07<5uamKp(vj^0RZCE#=9sna`z1rX0X{Z{H6g&@bU1*W4$R(?W ze`EY;%z2Y>XIS1*G#|AWCSEhk z-q1YyP5R9_gV0S{bbUQhu)UqoNjbV0OipaRw;n>9-p}BCnkW*f3z%-!02y^gM)*Qz zC`m|3vrrpaMaiWhs0p|RJ(2(t9%$u;h+XQER|(mXwsmYRB2vF*(&S);zI>_Bzhb+| z{Wx4*f6u+Ty*VedlxyM{u_T%IyXfX?*`M<9`u4Kn-phT(O+MFSOQkd3nmZ#pxLfi* z=Z;q3Lp{y(7Jo1Wj~+mx9DMgC>xe`(ei!TO`77MLe-w6!%_~?mXcDK%K<>5xH3aFo ztb*(pCE`uiH*FLUJ5AbFG5OX4+lU+&kvoj*?2@UAln6G9q5 z)6oHAd8?d#+L^q0sLzltwyQG!ENBHV_2~iOv=S_ zD}84=rkTZd@4k50EcdXOvJ11GZEvK)>ix{M ze^j%y+Jg$rcwJWQ@ z^`c^M=ht8gl}0t*k1QBmCOt742C+|pYuX=toA+m7>2yx;8`Re+2U}3I1nrk2uEow| zIs2Xm=^G;>qio<8kNqh5b{ub4L&4)C+sa`fo(FGtzue27*2XP-HL7}E*nE^pf8nuI zz&Ce^ZDhiYON>tEO#A`%3EJx z>E)ysmDbX6PkrpdTqeDfQ8B$@`Dc2ZpMtucmsbj)rCqZIYnSo~3eMV}UMnu*>~ zE+Y7%s7&N?bD;dP-=j~-8XXD-JQA?)C^|y*emgudDC_NI2J)HGGztjeZ6^(1eUG1T05}B3Z^#)267n zxcD>s(f$1mBThZE+D#hLj@WqJ$?Pb+n>hFh70SUS{)Tj9bhk;iY1VB#Tho?hfNj!} ze{Lt#eq|}GPZZzG&2kxYe?PM$0>N~SqeE+%ZkA@-rW&ssLsoNShU9?}fkb-wQ@zo& z@*xE3WHGwZfGqdb9JA!zc<#9VoFg|WTi%5Qs^xCQ0b|+ik_I^)pIwvoUS_n+VQikO zgEijEUyh+(BNJOWKy)4FvDj#`^!ndCl@y-y(kj+__{%6+%_YYR{l*;i<=Fd~p{d$@r$Y}!!u_!) z^QOov1u%~LMdl+ScxGsFWs~DB=TmFu@r!`-$iT|{d(uMgvOCd!dl$fkNXdyv6#ztH z_LGupKbotmnqt2Rf3Fw7L(+7@mq?l(rxy`FKAZhM8i_*?O5tnI4#vfQMNFdJ@)Y%D z0)ZL_yQO`}&K7j&j8zo9a&FthqF|aCUL0p()f4?x5npE@6WLkrgsW1}d zE-y=JqU{E#G~2OAcqzxV9N%u_EBI}Ks^n$!l6aTO?q=oY=GxJBgrD(xd-d*4omKQQ zk%{xn@CwKQ?CZCtx5F-*cRuXGd|1a%4b^iK5!~F|9XaYc(OiJtVJ>VTS8!qj{s6=s zky;D$4X!*ye=;za)1`%09Fb6MSZ%{(V{gAen;1lg654c~d3rQbu8H8*wkF@Sywp0s zI0gbcD+iizH;eXA9H5sKtW!MX1KLQxxX~m~hSBfX9fKqkopgNiH%S~{e>g$S-R1fz zRD#RGZ#3xjK@K(OHSMS1Sg~?Jcms>MsN ze;L)vqx~~N{5}oxPP%SiH|p1vhD;W|Ra&2(aDVx`Kq5=B@56poy`FVd!oFJq6}-v5 z-47kyjS3aWM$=)&M)U5cy1i|EnEXRA>Si}&GD&s`BL*x|v7hC@E4K=?J?A$X_XOFk zf5v<~{pk71n`0@Y;{AwpJ%Pb87lb zYka078;!F+9k2P!M(~hg{^IsfOxF7VMC>^HnQ0^De#X}M3CsQOrpkT9K{v5E|A*PP z^Mlxz&F+2wmZ1VkwZ=xQ}1(EddfwwzTS=(0RQ+BHlM`ffdsFrEowjSDc z)^bv$*lR!eBz0FYS|6ELz&!e0>M+C~h*9?@22-k>^EXDFTuD^pslZ|00yIK!8?}7% zTZM4n?YDhA$Rxh&>INvj_T>}hf9NwmmP4~)1(jAKAmLrb_VD*l!nO8tF9j|v4}Q<^ zisM@RYT+>zcx@se>h*Pe`!K8TJL2SSFO&1!O%|#C8rR~)y}xwG;)KA9*f6f9&`tOYyxx zbumFeOhK97<}$Vj6KP52NlCrrNJ!>t$OgSB&ZX02a2alIAs=$R<6trh<}&SQAk}vl z8J<||b39PfuQ+14fi*Uxj^F@J)C>>h2_3}maJeBf?C9uU@8fgMJ9un`s)h~Lr!SGH z?q2m8f5e#X(Zur{~Gr zq$LZ$_jT$9j?*T3QxiGYQxe_M-G)WZ-BSPkd<9=kYDx^QqLr|@f11OX0(O^F<-?vc zDPzqO<1|c{&rn8Itzv%{HECQ1(l2kZ#UcdY&8G`Vo!=)4F*^gm;T)tY0tjEXXMb|2 zJrhFQ0E_D^{Un0I7@^IE5sM)+FxK?7zq*^4X#S&|ul@=~Cyfm0NyVKNhaGKZkjUG4 zwxmMin)N*jj5mlxf1c&Y7@=PbUf`9n$S9V(reWPMFDuPP&8z5xKez3?e*YS`@a2#h zZFC^DZ2d?l4~muH|GXKK(rj8T+X%YCW;r_Ptb z+i3SPNsJVbZxIpwKUd)V1UH}%FWcsy7UJ|836&khDx&#@%P%ohe-*50`6+OXn7{j% zjTMmolO)KHT&|G)zAA3vB=3X&B?IJ*`9-(>DW1PZ$44z)l;HoZOP|QaqLod;AGqFb zC%!!2e`bg4IJR^2dNP}C&o#)x3LNiLN+Cm?T6Y8MKSin&WWqRh)k!CmNvTq9|0IB6 zO~sK}G(Tx`I(u=nKzaF>V}DdH3z&vRE=8lc^FBnEhoyOM=fKT_kC+FV#cQ=Uw0|@* z&?fJ5usq3y7#O!lOGALymp6>$!=r>rDEy-Yf9a5^|8ad|F{O$wvGYG22A3g@%r3Y( z%8*7zN{dvK9M{B?fWpL5S&}O&$X%nh&*)|wh$ChB&n5T0e1Z-%b?wH+Cfe=A<@zYH zEGJ6$yC@{vK3zEuY;`(Zb3Zwq5}ty)L8@Sm1=R7MJz({9ECA%~Fth%#SnKrex1Ff{ ze=j>3Yg+75f7^aeYLf<^@p*lHI^?3=c|IkhKoLPo)XM@t-rz*+ID^;MclRF&r^itEWHm3%-X5rbz{@5 zY8L^^KOF~(rNMp*&IY6Z1h2l>$gF*gf4G$aL)9wfUS=rYm<}@iF|DuXDNfNf-&;g) zw`qW?z&%DATB~9pmpH=s^4FJBo*jklgu2=DHG}$CcY-P0oMxR4t=)R_(p`DiHUuQG zntn|hUz|R$@3jVWbW%h`S=UUxSlgPg?U3S%!4Mn!l>3jb#2SLkMT{O*{VzM(f12O> z+|v0bshidsj@2L(5V|FIk+WG=f&YrWz=pFbZe zQXfdHzgpd=U8WciOrALF?iYA7QL1Y+8@#+!EdY%q(kFk4S7Fi?r|&(bVct$)SU^n_ z)t@RbFbJs}xa1`F+SzZs$X4!W(ua5e)89h)a+%G>sV!@^<-K*9x!sSpf7z|t?w+0T zTLPa-vy{7@X^r@gk9uvV0+y!Cu36JOpl8I#(;LLR89q#ndw%6pG^7&geCMxr`9;!S znJekI9Xp@dJ0SY`Emcd@87_7z2MpUT!BOki$6}59=gC7J)$`>G7l)l%mL-q%ucwuA zEL7pKu!!s4~YaV!vI-U_VkaczntbtCS zajK77LV4%bebLE7)hl2E$6uRMeebS!dx1ee4^)$zj!$OnC+1oYE&IHr?|)XXlU1GN zuAam$$Ny5(HEBOR8j)WfpMgcvreA0`3Yf|iS1g5=!}S_gtr61MojIm}b1B#8CbA5#oWu0jj@nAc8PJ zN;FAXPK36U!d=<(pu6qqe1C&~NWR#DAb~>8_A%^RV?1 z-&y#(0VV5gBn*{bH+!JEC0@0z?rTn3PT9&~Hd@uSo7#7ni<{;vGZ4Ek_t}pFS;N_3 z5%dJIHTuqDu}f{sJ_P!@#*3Rh(zLNKhcoK+cAn;PD=T=buHgzJFs?8LzL^MYS{v42 z!R#2T=OTl2lKZ^yu78I6x}6qQ3{1JhBe#J~jh;Zxeqq*r@WOBp!{d!sQ4X zfdRV`l`n$8v%bOFolwgq2n?*xkY)ouH4N);0}ERR-O_nzbNR7TH?L^s$QU(*Ib(YU z;Z=`>kWV@8Rt}Jt_u;r8-O00H_Q(UG&C0?LE52xSc$DPrVSlcJbWH`%guvT~lT<2l zVtdJ7V?h17^d5gQmw6#_FDI>AY!N1#9)tmfDh~g6I??MVA4e%i(TMcE$?}&=%KD$S z;mM=w;~L5{nq%*f*wgq|?B!CzwV0KpXixLF79{uB2{}NpeCeZk1!+8Nn}IbG*&0{! zqxP@C)%!uDZGYx@&nS>k&oIGECU^}bTBdD)k;^rK+T+Nt#9DHdQge!0I#ne**fuI`*resYh?**DI-+K`T$rWZnWJ zNeH58<$n?PmGpUZ0PT-_GjlJw0eXDR-9DYEE>TBV*uR=<+|F@6$Lo3X(a4;>v=x@; z)*d0_(I={zbF)qH+)6xyAB+K6zIpjHEw5U+ltl?&)p{mNv+&c7>%#|?vAEezs!a)THjrJihs)kv2$js{c*8cKVV$xM&{Bsbmry= z8U7~CL}-@o=1?cQdYY3l)1|H#0=f2{z8nF}@^I>y2imT+6piL1&Fg;V@G+^6hx9Y{ z(@7^{k@FUA_LbADgXJEP)*uLl2MnN~7vrsOmTs@4K>}3j`sYoZ*gFx#0n^Sk=@Jh) zcYi?>&=PF%oLggoM0!c+nWeQC-=#ubyqJ*j-1V<`Ui8&ZA^4e#$^~>>-~b>iniVZC zm@9}W&*1qyxqa*gm31?g$k=C?Nr}|9T9&-|y~&J?5FNJg$7|w2 zAE;8SXKoTz6o0eMdPX{o2(fcHa&SB>YJZ}T>r=fl8N*ckO^n9et=wj=eswI(?Ie9a z%s=XLykzsYblB1I?s7s_F7sWrt-LCci>|EY++^Ec`E>84Mbr3GOZsr7E!JJ_LXThj zdB&W$vdPasCYMxyPT#H5W;!1W?A6U)hgd@HF8A?3dLWyb=+GWNv%@%``hD{xOn))e zAS0y@8>dE*;aJTLHg!=|DGNZ>+LtT?rlL~XBw+>s2uJIVW-RPm<8>-?AmB(bxHE%E zMp>&K)?}5~Ms=S859jJC%1xi1(ix`9v));9sm#~BnP{5cB%k8ZU$9DjG3}76fd^fWGl)@Gd}d#AkLGk(XFE z9Oq}3m+GWbT;5T5=5`)!3gy54-n*1yrB+gvQ#KhP4(y3azM?;8R*lq8nqufPtwsjl z4i*7e!ew=HT$J0xkFjYWhew5InM!jMyS3VaX_q+eTK%wUh%QAh%OU7sIe!NNxj$8` zNMDs;8b-ptGw#WXp-SWM&)G1Z?^{fJE=NSl8ynYlm9ACS8DnWQ9{BAlK1WT^PZ?ns$60ZANR25BUyv<+hz_1?z1cI zvq!p&C8vG0<{M)LypVoA*ms97jD7cFO#`HoDOLIx{<+kxdVi0E86(!zPs$*_r{o=R zp*aQvNd9^LTg?~pxP=aF$s&XS7W26*>Lk<{Ajs|DwmEoYm$>YNuA`>#wvyv5aoa{d zo4VuvivxDee4|7lYbr$^^@!T_3BX8(J9_Y$<59QS!@`ugd~=_89T(^3Uz!XM08D6! zNCaY5&wva^_m(;_LGr0mq@D0 z)#IcY;fsaVNjl8IY|Yr=rZm@`%*{J8NJ`qa6xP1yVSi&@T14#LYHMUi$Rss`HKG5s zsQ9SFbekKH>EH>U-_En&^F^qZB$?VH9n$$U6Z~1oyWVlkJ(g1#tLY@P5Mt+mgaCly zSF5QXYt^qRM?>;b67_9L)Bn;O^^0`(r6_iURV5c-TTC>6^mNuL2w2%KECO3E9+3}c z=7zjNoqusO`5{a6tk?>2)~{vXPQ4xrL-NC1yBm0JPN%Ikap91-!e%O3<~Olkq{@5W zrWd`Xx|gZMYn(GS9)druVKoHo@M_IvrkI!}^4Te;HAJN5GTXzWbzQ=O^#A%v?q5+R>3=u_902q73$OcK9O%|SRP^*pD!hmG zq)aID>onb@E~`}ltyXKg8p( zDYI#q1d;yZyfk|`JtU8n3xV0zE7p0hHe7s|VLn`4?;%QlmYp*-+%yu`!50Crtm)Zr zI)4wvG%B$LDuTkKD?bdszs91`br|9+zjB8ZXxUEgA&g*`*ge45pQ5Ua)e1<*Q*Wow zOO2i2zn&d2TE^l@$dnkeEpWI{(qIbgFs)XNZ28sP>n;DWSU!u2zkgCxYOfxRJr+~g^!ndANV}d0J5$o$lu%}) zK{K|PDs5gn=RFB9w>BeO%HQz_E2)k$OY>r3J)dy&>5K1X)CE53c?4b`*D&0dkd8N< z)aN;>u^$~8F{P_t>j_()vFs=2WgyX`CwcoGXGm#|vGhVV#cp`>>@fGKZ#7Akrhiq% zKHP9MRxMnw$-RG%To#kT|J9hE0%LKbn&+3kqgN2m!wxE+gz6pL5n+Y_#`Ky^CYjiY z$UalZxji?gy5kHgy)vbkegU4XN?)6P#SDA~35J3h3AH}91B239vD5=XmHqeqX2wKReiZA_ciW`EX2cH^M{2IEykG2^go{$L+$$(ODt#I zmj2P0B2w(S|6A$>nB$YGf{b~rMT%E<)?ABXFAA`LICrfIciqJM)2T&@## z`OINbalb(u*0p`PiPO5oc-W_Iqx>}Hto1IS(+`tHp?Ul$qX|jb9FUo%lPh6qqHuPM zm|`z$jZ86UoJ8}~YFXo`2$RPv=fcxeG${m+wqqhrn;KYiGHuzR8ZVptxX42Cn%Pqb z6H(j*r8%3b*2W1{mgBG1X@BB+p5l(4Y0BLUx|i9{YyB>JK?Agl%=w9G=DpPgk4H!L zm!w(OrMp#&;(MkYaT$alz(^tX(k5t!Q-UocuvScXTV=#&5-n3o?BpNot*;0^u28DD z$FKcV*qlaDLzmUv&|cQiSarQ}&-r)gehTexi#D(I=c`mvRd5+=Eq{$&!9FP^C(BZs zg2w5q&n=P!yUn^|mI0KbNcjJJiJed#K8H#|v~BC$WP>EhIA4}FT;(?IQu#a+o*UNb zb9S9hT<-`Z(v8z^hL=Lxm84npN~a(DDdfc09YPK4FR{RVSGT_Z=0R`ITNV~f_hzEB>c{gfb^?O`6#q(Yan)il*2p3ZvSwkKBo7ik!c2(os zNGyw|4T(HpJ_?UWU3~OltUlgs_m9#WW^<7u!Vx~|Jl!NmtPac$E14DyC%`+}vZAiJ z(%cU=T~~&{ZGY0NQA}W%peC-_8^Zg^S+HlT6^Uq>!&b^ji5QX>OL5+w<2@>LXjsG6 zL*n2hil_9>Tz{P5U7Xa9D{@((9L-DjCegO3G`cuE^I3(>@NkmH^bsGI5HvrQW(65D zv0bb8NE75vHJn_ZWx6$;JPVM5Rc)@#v?H;BK5B8NvVV`T71m@Dg!`RRC9&)FKRT-H zJ$^A5S#g=vgNJ+0j1;wi|*3d`V~p40Dz??QB^d*N7E&8z)wuT z*f(>nI)99{i9wEUxA?;0E2YYa_i#l(UW$v{1w`q)ie(q z+-e!nJg!dKT322A$W($ny}f_TmtPBe{3WGjA1?;)+RLiMZ))ASe1fovJjv_2bq27< zY6}@)9#2GM<8x!a5PD|+)p6kd_F?X&TFs>ANPq2NsmGzhETFkkO~=05Ef$FHr!ZpA z`;YUt8DbeI_ROe?DefGAtK!Zhnx8#*K!Y(>F|=Mdgdv(=g}-QFX^jhOLg)!ojb!=M z`_#WyQvOucIPD{R%Ek)iMC=Sznnhil=EUoVz^n!7MCMU>|CCwEXw|0EI)$lMaqTq! zSASD(Et6P+h~t)-@rTbn4P$Cl@SdGbe@>hJI9{K< z`DV&5^fzJJCPuj^n_3{M3mQ-vzZ@BN?py6sR9~{Nv_Q(nF_gB-YjR zu&GEs5Y2%<36OfoZ0o1MOzKri5`h~a#p=3aPa7|ek1(mcN0!h>I6AAQ37DJ%hksF5 zQoc2xajmpo(r0ZX-fWfXT`p-pXULJ}++h?K>?Gslb(!^U*xdS6h`j%2+mZ^Gi**r- zHp4J)E@;#C+Qu|!+i4(9Zr*(znGoKGM9B*$Cg@v|1i0^lhU_7i;Hb`uoV^nW0H|Rz z1?(G?WOyf9wh9Kh%Ovc*C%1T-mVX#m9p|UeDk+HTu`9rw51O`MTwjZ=w1_=_kGMvx zgDX}SUAkI1Q5A@&EBiJl-80YzfdM{KLgHi&6Dh0OZnf3bHDoGGcGkPR z6qKABF@hwy5!8hGRp078XNON?8L!|OUBuZ7OD-bQ)M=hun->6=)?@XzT7M+=v$au% zmrS1%?nF2qZ^tu#1z7$PQU^DhiRR}H_y1#G58VT&Dx_S1_(4)LJyW5gWF~k44BX%T z+Ls!m6U`sW(vQO8nPb2?&W>8D2pfoSuB5I`vKbIW_@L@=J*}_)mBfsw{*DpIY+%*o z?s_UMv}t)Q!Lh%2Wb8dHxPRd|l6R#n>)_yQEtw|>;wb$~c*bgEEIC#*p8{54CU~%J z&MB0$x`q0AkWDdh*5*@(oyHlx{gWA3OnRL;ZXw5L1v0|a>}SIO{_+t7Z~(LvF2wsQY3M8jVcu2c;K zdkhyM7MWT9aWVZg4=aSZX0liZwh0!&)3z4a^}-Q2T`ylJABpQ`RPVx8qTJ08N0-zh z0akGl1xP*5N6GkCoPWHP`Gpjmem5+<1vCM{2lJc0n>TK-zAr&iG^|#4w78aR{R~x) z@_vtuY{q|zY>~tP7Uz*9yB;6HiGTM$=vAW*;S$zVSkq?o z^f25&QUJtziUrkpeOyZ~kBH6lceE1z90gEkaUW7iGA}Xy@U@>D z|Fxf#`g=dAD=n|Cgd-(piwQhiQ9kc(*=g0}-i`zk0w3=qR<{wMir2EpR6lny^HsxO z{kAP(RwY1A`+wn{Rpa7$`5H-)1lTG!Z72)!8FB@!u4Tcj&VH?oiYVi#3W}%KW~2AH z^2>LJzbdc8OKRqx)5xYh)2(-kEV*w@K4Vj9g4z(#MC_jpdzhQoDrqqEk-s{sXUxTP z$?+g~u$b*`Zi(~(lCW4cgq(1%I>)qW5wl?lX#6wkEPt$yNHS9Si(^~+fwIT@8(C43 zNz9CWg)!A@ehRitmoCM1CqZthHgomurJDIxUnK}5&nosH)0QQP@p&Q=9Y7sEfsf{o z=DpVECO<4Q8=*?YDVJ;E=s|Pjmla;C)rX}bVZql&AN8EhsMlA0--_l-J@5QSR;cn2 zLrm483V&iGnr~?IuiXfIBtW{&2gY|Knx*oLz;}nTBvafE?DsEac?z{y`-G3e#G}Fu zTA@JqCY6M9zROWO!;uKEIzy6;2&2EkrYJ+aax%1~BD(5gww)-AELFfca_(f32R!=< zBle+WGY2xDE|;>eHL<&;u3bIa5CIU@^D$$t$$yA2>km=(rIWicoy=cnoYHBNTbpWY z74Cg^90(V7JW9;el761jb04)^GH~U}eQc1#&cN(8Ia|H*M7ph1WQtwsBa3w(k_CZ` zB1Nc6&-#ux%Nm($b~x-2%-4*J{N6d(S|t8GUl!?U$w+5xgz`tL8YyC#yK9GXwL9BF z@PD#i%}gIK%3zX2dM?6N-T!dJ)t7tsGD(&NaQ}Pfee9!X{?T{y_qAaxbrp@TbT(Jb zV4NPFRLM^_zxbaFn7-4>Rnc)4ao!rWs?M5O4rK%~54=$+oCH};c;*)Tu5C7rKFL{n zdQ!=c5lg-i%IJfKOhrYmuA&I7oDQf2Y*_2 zJ%H24+Q);X#@K}3Ks>g#Rk?&wxdc&2oN$8DOwZ-U z5OBTshZnDnj^`@lVfzbBIy2MWVbyAD)=oSc&u_sW?_)p1MMgXI5u+Soyvx=@V;+c&<*4N`b^Tu02&49rT?^V;R3M?whoGFjEicI1E!K#3$A4?MXX972 z1VY`TqVl$-Fo}UQAq($7ct>~ypmOzSZJ-&(x4xt8C}KS}IP7XoU8Mg{0*FBMP7oW> zp*G>QdO#RREVimvqe9P2@}eig!2L*fTQon2gazlH4gy-hB)|uPNY;}=l}1v z2RZRO_qI^mkhJgS{>j8QK7ak*AU!K?D`_%Xr&l)G;m?b!q}5%6k{x0R57ds_Og zXuj1;5+Mw(#c^7roe;Q;^jg!tXxmBwJA9E;0$8GM!wN}U|Kt1LYYsqy?L!-*FeT2HvLQ}bjZ-ou4RDN^j@ zuRg4Xzklau9Q{!Q2(yvyn`kDljg_aui1l?2KvI3UJh>-J$1@k{s|PFQO?hwETk=)D z+))$)HyJx2ragzup~6E2((8?BH#(K}32uM|z-_dvPxo1Bna)k;qKIKPj#ovG8kg^_I%RH_pjGTbVOg zLm7879hocm>4A)5L36~PpMna<#-e#Fj>nOucM)&ar`+4jY`xf+Z^brQJbZI}*&c)i zet)L7n*Tt|goKP4LzQIVL5=X!WgAaHVJ|nCg036hgE>GZ z6qhREUR6$J zN;!#iLuJ1*5%%91O>6BHfO=7Xx`ZNvfc?P4ilJOZ*xK+ZSWv;ERx^&ExPSTc>3mZ* z|I+9QNoG4orvGt1Ru<@T5zFF94!7VuzNihe+b=xw_|L!A&%k;9HMOJ@LqAFXchU#i zFy8cB4J6|bE9PsHH3>BG@G>50v+?O7iZ6cN9i!T^i6Myv+MA4NmoO1DDwr(V#> z9Giv;^Ae`A#EV=iodo&8k*O9|g#Skawe!P*VkW~NGvATFS%K)hN`2{b3IKTu$+AU;fD^(%sQOc;*Xgk z@IumE<8G=w3w`iX;0O}--*5XkH*@t5MM~+u#IDWB7Gn`X9hBeaX7{SsZJLBxG%xH+ z!6gNjrXT%I;v?=nZ+`;HaYLxDM2-LF<)(jsx!{U+fog;oPKtLRy?3%uKSM(4At7EC zXpQt6yRM&?69Aiyy3XdjJbivTyh&ZM|88fJwquv7pHo>k9R^cMC*NZN3p__w=u6feoEh zJe{j|1##pBp;-@6R@Rd={(j@=_L*6MlDJnSgJr z=k(61JMLsY_w!S+X+D09xsddXvZwyIjF=06w#zjwr`01b83Ln|P@J0tzCbvb-;@`g zGQKHi#X5c`Kz|bne)e#fE|BzgTFp?a?7&6NBjzW){t|kSxfVAIoqN!7E!$e)Hcwy7 z7@+g(Lr&<(V!|(Fg@)CQgznO9dJjOe5WNl>iFL_bmJHBTfZjd22r$MxzbThrZl0Op zLs3Fth-MSd!M()sWO_GW62i}*{^sk8gr5y|W@6nhgnx+tC(X#TRH!tbHrq(8@FEWP zJotGXvrglT)^1@=xTlNl%OZ#SbS^Im9_P7-*`v6C!-mkDR}+JHuD6Mk(F$1FMOP^Q z%pA~e7*(tTu`r*jP&xYiH+=~bj>*4fx)|;F z(|Y+Iv;AYXf6Vre+5X=&n|;+!``o!tD!fp3Ms!=|;9fon6A!3Umw&#&WdTj^OojKu1)uR=MGr7&$n1sE?wl-Z zgQGNLutI#u*(QBIi&HC`qTCBaU)R5)FFs+P^@gSe#J8=Phr%e14Scy2iAaoHVUZ?WcYlS_{>7Ar0yF{aIc6=s33G0XzcrR3MCZbuTC}gqR_q7k8Mhh+c3e(=-Voxcz?_CM zMTS>`ERI~ESBVaz7QEm>Dv9gdGJmS1H_FN6@{{Xv@A2t8hZ_R_)lqk?D9?wm_`gHo z{GjdF)pWjdPy$KENMr8r(Y*UhWkhV|{dRuUcnY4BQ_ZVNog?>UM5J(QjSHkl}MZ(O>B$Nv&thyVN5qODny8Go`drj9-H zgzAJcrSUwi&bx-ydU7{;9&>(vqLj@7sH)Gp$E8wS`+4hiG+R#XSGih)m~zd8Ej<2| zAsvumzvG6G&iH;PR;jl30ym>cfvBDM z$aJQdE^V1zkrfGv%3ic>j%9yb*JgPAH=6;oQ;#;H$yO}+fwaQ%;YiI(`jkBIQA(W_ z#MiVEId1F}!{IBlSl(gu?EUM)z`BR!;Vrm%wR8C4bka0cDZjh*;2#Ij1Yntp%Nn$~0srjixS zDk1LWz>&V=n3#tt%+&8mb&iOK9ME*5*spoa&&xVDwd%W37TAq62BY%l9fe3o3I@JP zOv5)l0Bg!^`m#LZ6SlHN{u<;lKjlMjFbF7__fYrcAy`dg4}YryE?psJ{_x)(c&W<) znhQH(vPJLJU82Nk12w~T`-ZKw5L4g)S*xD77^NjR@u81@=X;+VeAW-g3PBM1(A6t7 z!8MaaC|@nL-9u$)ab>v@dRu>JU;HXHOQ5C(xwlyc`0e)xlHQNs9LktlDJRA($Y(~^ z^mk)53Jt?)Hp+NW85G5#zpVnk7b)E?;mO7rr2+Y3<3(y?(KEO$1?$>!=s z1Ry>Wy+$k-+%#}4`M@Au?S~S=KaU=X12f@XA~mwMYi5q80CBic*nLReaGH1H^S;S`&aLo8zY@(E41B=}2+2a+)e48-59e?>)qbbK^qhqS6G{@ZJz5`wwZoJ%= zsz{ig^W`-cP{_ICsSz9xKjTGs7(A$9qv;fLU;DZSp7wl@vC0F^aOl{25ChE3EhZMm zwj~emS46@WJkTQP?R~=FORq4#Qn;@(Lub|vlRySZ_wR>5bY}5EMUkPJ8H3pG=?c

T{r~&R8D>;O@<2=jb4*Fi$jBCH&z$IK_&dX}%ttVjltZ#E9m?RdSc+9}G zi*{c1M50@z$PbJG>>fRXNfXKLjFRodCF zgR+lj7fLG=JH090(e6JH0#;|nF;6Xskk6bJvWf+CPtH_cnAfczdLvr|yLmTds(*f| z!GBO#kjS0Z8N9p*On!{JNkBdhMGCGo?U~%WQp97;RuajldnU-n+Pl$uk8X~zw5c0y zE0szEHZh|$0?Wa7;BTpWlXvS+Wgms`&aEhRAAjJTp#8@N7Xt8)*X}SAQ|-JY3=eS+RrG>fj-hYV z+ikT!G1a-LM{s?3-3d}DPu~RpKAA4yjFW3h6LF;D1Op76O$L74(K37A zg&h4_8Zu&-M_WOd7i~tssZoR1^wmCoUCNKO7gE7r{5IV4dB7cwmkH;7* zMtrpM{j;lc;!ehAndarp+x_$u9Rcc9h1%N?YG(?&-U*W7r7HhwHyE$iCrk756Uy3N zUdYv6T{_2erz?Nsd2*npczn7#05N6H8me5LqPuPPr=(7es^G{;d!~|Msv)MrRNNO6 zOeV+h%zH4Trd=axahu-yBY(5beYW-sfG_ihAA9!*qN!&x>pUy`u9)5{TcpyhpU7kE zU&AhrogL;wbkW+RIl(rCZMnh|pValN4;RF%@bOS4&K{9%q!#bd`Cx*`G%dAZH?!lfPeY5jp|Q1J{`S4 zDlQYa!hSu5hNW;Q{J~~rWIVUcZNp~UNU&FvN&P#u$?q;Tro6sRZ<7Tr_U~A%UCdti z`bz{2qW+eO&6lFYr^i#^UM3^E)ckY|Z7ojuPv|bakxM!*a zFGyQt^bjD*_MdB1vorB2Un{leBco-8jh1igS#^@K<`JTNv%6H5s2%9m zaVPU=KydhSNC8hX$au~tRdNdJ?>r05iqB`)!2Z}|Azpd??P5Wp`0;ZJt_SRYJMUzG zW{-YZoHfss_kYND&R6+iD=V|2=IZYbHbWv@0r}j`t?&MXb%$!hBaEm@4g#f?x1TPt znKrs>!Bul%4Tv^Zvmq8Z?rnhPbiHF@c#qEBF<@ezj`pmV65a@2h-iSv-(d8e*$lfS zSB_rbD0cc~R`C~4O%>KvmpamcS($sFp+ur0iC7E@h=1iIk&7=iP!VyMv5d+_9-W>= zH7`|NOf2G}b8i~B{I^^LiZb1Rf0!!bOie}<0dlVHO~HPCK$w4SohfYLerF5U#3Yc0 zW|_velLfh|_%{W{@i$j`y0R{V9^u7Ko28F@tOrR2-VDl&Y^1xiuUD^)(8h>qvpbt4 zF~{9*&3{_9s3Knxi^e}kMvJznJ2+bO(=^J6iJxWn7 zuwsPut;@&9O0lV|_PWmhxSagdEz*7K?CaJ{T@P7RB|+{n$T9n=^|}D^*$8_?bPQV< zaP!H9z~Ryr0xOfc`Y&p>@q$$DkK4y{KCG{J>wh!d#L!K}F}txu(4Z;}cCrytqDI z+kce%Jw^IOOvn6wn3Mq9Ug0Ei!ERx|IMSI1;_j~wp^g&A6Bf$@u{=8#Sb7Pee`P?T zJM!jQB|%^4o*KBW^hR&Bj>x2OI!es&VUD*|JkzEP4rhJ|SWE#GRkLYCi0g1LGFROe z`WM`We?|fVG>x8?Oi>-JBpyN8fkiCAq<>u+GL^^ed5?(e03+7LJ|1m94xGlSA(F?t zSK)hx7X-3`zkn?LZ+KRfeVk5I>6Ff$|Fg~`-}TSU7!k)5O-6J7x)(ckb4s;xIUsfuzL zMn-RSkTCZm%Cu|^qLpi8MgOfEtbg8G6gqxW16RDH9N%nUH&#aHaRzSzS}iUE1hxZ{ zs~t{J$5KaTIwb~XJGTBcCZK&%RvukkuttNVN~%<{%3~l$tc(DFr%?v_m4CXU7wXPQOj6}nSZ09fK^P> zblrZv0`J=C7ojBt7@a(u2PY6x|KB+<3a?1pw%Uo zbCi|%7aFk;Q;E__TyQp@_j&}~d`_}PcF6?N-h6zFk)rb}$q z#S%Q`56?dk#Jv>vfj2<&@_*yvAq5I9;d^347kv3I=)Lrhu>T1AkFfs;`;V~y2>Xw) z{|Nh!u>T1AkFfu*3j5R}(q;a~Wy98Y`DLc4@a5iUQ{L9^-!1#&I6ZFnzrJ8{0yaD^ zU%mNP`6_-*wrI|Oj*F0=8jP14^KFVc#HES3y&7T(&Rc^hL>qR`#Guu^+GiJnA&fX#Z06zuYmw7vkIi}@F!awk- zMp|ewgIj~1@u^?=?|%|8SN*j-#Zje5Vtji$o!??c z_=V+(M;V%loJukD)rVSlP?u$Bcz>i032Qk%_>wjp=XX+rjlPyhJe%m51Q}Or- zh6dsI7&hJnztjEh@$%_q#F)K9evy)P!8DhixIvp{C3HW)HaLf}__UN;Y6|NQa4bks zIV>3H4HdNR(hpDUjwpDiJLE>nf#wXAhITT??L6E6BkSt1ZIm6AeRZrzIl{Nvx6jBK zYH=s1#D8Gh&amm22C^IQ?!1n4F4C`%h}7jiuXd&EE2k?{_7&2A0B6O`U$_g1rgWWH zj_v4E{Ywh2l-}3J99k3`v8SYYN;P`NVZ#M#tWhX${>g4ZBeh~kvFh?K_Ve-%PUBEg zIohHLm{<|jX^ebuyR0|)>1p0gDhA`OQJ;n^=6|@5Ex0ep+r*jQ3w{OQySZR9x$Vtz zBYQ@8H|!#K=6dB?cFf4B4ZUm;_=%PW=j#AX39s!KqdNL-zD;l)Ow?iRd9gmCBnj4t zgOLsIwy)+?POT33o8P2*Lg5B|K|6T*lQBp8vYm1LiMH6%ml48D$HgM<+zT4!%k-%! zXn#L4sdE^4T>$16{QOY1=+T3+3-~}<(0L_uR_AY6dMsO%sMzvQiVl*tOU1AonPFTU#vfDF6-JVhv_2mvCW!H1m_yPB_O zPUGMp3~*Ya*4ZjR(`Mqx-5RRLlM0-6Z>%Y1>ixDUT&hc*F2HXTZd0c2bKEzKT%LAL z>>03!bs0ijm9yop_a54KoVQ#Mwf^~d^@imMckXumK{$1Eif{>0q+2=$g-l#$@_&EQ zqvz+`3y$wi7F7D^wy#uI62FqONb~*e$4Z=iu|DP|r{?^FAylTd+)H!+mKXjQP^rnA zopj$sE<-s1%V#{h&)!O4WsAZRO}R*^d8?c19X*;P&omP?L(i9F;woK(A4qz?I_n@l zFtN{a?w>iWo|E`9e9%a zMncUY%I%SKdARBH9W3?*ndRn$Ia z%f@yK7;700v6-qU*Pr%3-nxmui0SYBW%`%@Yx+$mTXwp&LC(EalV>WfD}TLr&_+14 zeN-HzEX0<%_|4C={q-%tM~vO`m5)17T^%*jcf4kCA1ABRA8t?8#vOajE+Eyb%Vc9| zLZYfcrLD$~xLeHug$L7hK(82=#+7EGyzW@G*{g2p|ye>5|P=9VLZwPJa+nf%JB#$qzERTJ{li^|uV*m)@sOT{M4dz2* zYKO512Ex5JO?}YG538+uoE0A^N(enZ)R|Mvf3+aj==w%%1zb#H_l71gY$3q`&0xhD|pqEPZOIi3N@wFt147VRbo}$gr8C`Xq%=*i2 zk}mr~LjXFgV|y)X`4?%}BrpEtx?2GUiPaHk!;p6W7qP?+ya^eDk?`7OYB$Dh9q{G7 zUYg$aT~|@=&U4AV8h;o7zkQL_h_v^A*-_wMc3in48+x1JF4lQEQtJ5r>A@p1~A9*Vsp78D{R?>M&Jncz@_?GNKt7;&!$WPeNJP zJx@b!o|}hwnFV>IVL`R7G7b(V4^7dK_7)TU9GQQo`gPGiE>T zEBln}D&SN5+%KZ++Fyw>w(+c(337Hc65H57i*{G9%&DLc z*H_v&k;+x?6v&;I%`dC*#;d4c@q(&C+ptRaVjX&c;Fqx{JG#vErJ? z2pQS(PQ6z*MB%27b1?#JlN#x1C(=@&fl#F+1AjoM*@=(FC6-i!vlX{_efF2@N6N-p zs|zH8%Hc=72YlV%?o>`mmtPXuZ2PRJGq7*o;EQ)E(A~sN$PW(&jm?aLjQugDjglvy zdQZF~r4oqptkq>4Hd@|x26r;+)D(!mn!G8j!0jK;qus6ZRD6qi`~65?9Ec|(y*{+j zvwwQ*&puV_M1>-$TpjH6eNppNv(nuP>Aq(tR>1zS8l zDVCv1AO2sCnc;&<{Lz|WQrgg7C%;Hg0R2gx8AfnIwaYX7Y0M&#p zF%sC%m-U>)9!#xySl*K* zWO)4p$Ffaw-|g`h#o9JDOK4NblHr|hrkMM}@4lHBmim?$9RYp6-eMG=VXV>mvtPfc zGk7(B%x=RB_FJ!g=+fX;6njd^>+3OT;W8Ai?X-nG6Afl}%CAvP)|O};07F2$zj0UA zbZ~#^?!WT^FD`^hMK47vQj|FSNvVJLdO_~`N^Df2_gfk~1XOb<0XarnYZbV2Lh!A} z^>?GE8M%q`sj9j3OIam6cWOJvZ<=o~j!t>5vBVarHlzw=iR>1)y`>c$X9^ky9`}y4 zti05oeO$e+0~WRy{ieMNP{^lJp;2B{O=f=y{bLe!OAa7hcB2UuH{{G%(Z7<(%e(eQ z&DoL74Z-tOTSID0b?=#Wb*4|jBbuOLE={`o7)y9z!Oc;KfXmFLQb{NTb)FIF2o%tT z4coZlwotPLkIvV5R|qK^oKfmmDK2jL1Gw@e6GNbLx0{f-&~2T7#kI-D{_IJmo5}Xh5o|d?yQ+GuttEUf5fu{O2UksPktz{3!&LE)JWe= zFwrV`@j(CUnPz%SscP`XeroZfJgT5tt*_9=*UNDE&=wXDn(W)GY`oXm$FI{)jw~Rc z^aCGa=ayM1RL}3O5r@atV_uolB@%z|b|}rcaWt!jMY41k7Z(LU`!XX)g^X0a;qI>s zwVh6%se*h@>qPTAazWevUdhZutk%9Qw%buMxZ05t0Q0N4%~U=arMqwTZ)At1dsURv zQEeO0=?xFY)T7#)xu|$~=pAiGaaqixUOjr9x<^jmFf+bAYWpvcidRrDPBedSrB6h< z-nX}b_LjSX8WS#E*Ry9n9-f|MCZ}kXnkm>f)HyhpngIsw`%33JEjbiiLsS^oZen}o*d~!BdB=ZaciyLizAmBHZ}uc9f7Ts9j#7n9!j_SZ~k{u z{X>cMIR)2D`6FVge$JiOpDBNWeDe^L0)$NQc*X`Sq$Knr+Gu;^dS*|A=-gX!F9btF z_mUK&>hYivZRy0Hm6b~9@x*%9YJXacWwQ*$eltm$Y+qp3{?S|5p;9BnIK&7EbxDiC@Gu$_> z?%#obccM(iXAem6z{V<-w1!cG}TPT6cqn!%TnUxphBD>cn{9-I#cY^CeW1p6O1_0Z#;vC)&s-mds4?FNdDz zsQVOan-5>0a-nYa7b5ZetCXS~#60R27}^zkHKPWRF{~^-#Cx{#+WA#f2WGPGYQ`3u znv>mq$0C?HVyZnyyA5F6vcLIg#QXH>!y4(d$42>&jevysIedS#(Dsu7uk#kS*TG6G z?6+*cZ-6Y^0@(h^pr;#4M0?n)1n;jrvS`1B8zD7`qzdBlS7IAd)-dw1A1V~Q619Ll z3{4lWb!Ibdaz7NoGyslx%70OnV|SMZw6??0o^H!7Iop(z1o4HP&MYB?+M*K8=9P@Y z&4x>lns#QJ^t*rc%Gd~eq(Pg_{}A`g8K@kf>DyelIo^Xlf_85RZ(3%P_^4%|KYI24 z3159xktWvc(eO2fS;zG!yO#+;ul8p=4GqzgZmW}*W(bk4XOT^X4KH@OyhP6a>bz_5 zp^w6UF9|&(xqeq)pPmkzrq*D7^g2+}C>^l_Wco@-(1EagQo1k#p0gC|)l~}> znZxvcD+|mcE=0Ie=cyYmk@}n^B3QzYZOZ8acs32Z&P#qw!NQ)Jne}NpL3H8 zF%@^`u`_>B_VM;KhZS!rAW)sgOtV)Vvzne~xO*eteM2%`ZlI1|w&>~&@~PdyYl88M zT9m&=0LO>Y3ih8L1Px;hM|2oQ5Wnk0&FR_4cq^CnE~FKuzqC^IzgjVpdRf|OyXmfB zyzXa?`AaFYzK1cemrx5)dk>a!t$U^a4t%#$#Xx_}Chu`Ra9$31if+^17nF@Y*uJOM zbWMB;h1b3~ocG6G&D`RIxFM9QIU&rW&QRxgt(B6gJtL4xn@ZIb4vWbrK&z)gdTHg? zJA4l^*fk{XvJvMah70E|(TZb zcKUx0P%TlU{2rIi5A~Xcw{7gn2VI4p#w<6o53g;CHY&8DeLqXV1P6NAGY2rA>@@WP z!51KW#^;0&qw+t5ug(p|%V}6&Vk?||5T0&RFi6LQ^6)zyyE|D2Y)Eqr>s*QF_QjC< zc^2KeFEm;K74!|#DfUZl;F*#u^+a`HMx=kMQ(3QaaL7^EwRgBJ*C&LwSyq)iURQ9M z+)wpud>`$%_XoD$RY+Fcln_$&4+TexehTlO5hBKjN&OiRQoVn3 zS~tK3na*wV!`GQrlHRpRzxGF({v+I#RgGMfd#FX9X6+u14pa zxI`7SuiqDz=G!L$fzpIDA66fN-esARQU-<1fCd@JjZvRAH)flgfW6VpLlr zzLMJiiP3f^ljbl*X2|Ke(nbXbppUUPZr8UVzhkL=ma}2;y|MosYfMqh#DNsbmr zSyH3q!WWZVIq?y^rJCyPEmB-v-w5h6LiPAUw)UrvO?K=&n#g(|i9MZD9x*oVn(-lZ z&>&!Q<5G-sYkVV2xT<4t<3((H%iF}GOv0oS@$yRO(w6zQgy>7*3MPN0Iy@#b@@^g& z5peij>N!H7x6xsVrW<4`walZTg4RmmdJJ2qQVwz|4RF8QI8O66J=XAB3y| z_a3*Nj&$0-(`Xj#4vl~6jSJ8f5{`4a!o&JET_FVQgYWAJ3H$2-1KYyqeg8gZLlsIT zPp_NKluc&73Li&x5+KPjKI2d3fVx0L>W=p2Mo9G{M%Q|(S^Gs{N2y7%++emtrp)h= zN_}zwyv!x0s)NADIb2?^cUOZjOxE4CzXaUS-t-wtuCCTrZS8+*T7OjM>TL$hnm7fB z^GBLMyF(qGT+QkX(+cIy`Lf>*RN@k$xG9UPPsF7%>>cbrf zEb%Zo6TfbgqfLKd7hG)XxYXXKRiVWr!IK$D$h}Lnax>1FeIu4T{R{jKRyc-kbL>fW z_m4o_(QcLU{cg%gK;ZQe=MSbk4jB-2RDDg)S>XSP_a6R+RTw+VsgODR zd1zV*vQk?=G7#eNsoW2(9KQocRe#re&&->+l1I1 zbviH3TdUJE8TcS(T_{)JytCUl8-$dKHAv`nwPkOr$6p&dzBFu#N2IB9(3C?qWE)KjOE-tg-C_m@yW?dhn8QM{O~P-T zKW}O^gls5IqyD{ht9~;8@5frn*7dw5XM}d6)?w;z_Eg)mcUI%~NBO+&(G;%dXJ=pH ztWa-r1NumV+#Bv`KPEt?c%D18@;|J|0&K^S%8`G8T=LRGG|5G*&Qi6|F)`8398hel zu2UlH{ROsAL23fJe_4w z_sR$-cV!7MPzro>`ldUHwSfGVxFY%@oK*5JPD=YXlIIbSyjgvHym1>;ZGlh+@|<_# zb60DrM*RvOX-_RRO~0&A|7LZ!Lj6uZ_>VzQ&WEp zE!vE}^Y5w3+NAJjsf2*T{cb(ZYwA3AeroqP(cJ8{a{Tv+i8JQ#))Qs8&_|i_{XD7~ z5oh?jj+dlz5%e;KJL{u;kwb%-dU`!hCGvnD8vzFcHCVJp*_&9=U1YGq(!VnkMf*P- zaN>k8}ba|9`UF`P~6t{5p@YQBC<8^qs_6~4`{e47UwwY0yTZN zY%;(@q++LYDv2IwNMfiI2S8nkwa=LGcFqf;Z1vT z2@#36>l?r9~fjXgsw zhf9K*GR}i}v*ZSuPZ=t%6Gt&Z3G z?#h#aFhD6eoeeuDCa;sFP8@&0xHOH(luP9l>Um9>gmq`yCwH>N9U7j%C=BVLit%owJ(sN^IrhL7|uM>b2yhD7-F zmA+W6FH}iXuP6q_&p2{n*5X`-j9fo{?F8W>nB!VI-n-$n3qX&<6a9ZSb=ijPWXRoj z-V{>aqGU9}ZxYd>hkFF9x2`#J$$8<7eAb4m^YdZ3rfeD+GZoW7GWwul2E9r{?m+YV zoJh5#MYXqUxAWp5Ms>FA4e()uQJJD(2zA3;RSKQPob%6L3a;`aZ3|5)!X@3CeB>U5 zF2C)|nB#V`V5hrFAV+@%Ui{^1>n@gA)%@bsK01{K#+G*kd(L4osMI%O+Mddtjs*cJ z_2*j(fCY;Q_Jk19s8PEo&U0SIyOqD4KE#@*ywbDfx-4_PO(7B1{E`3;?8yRFamgbb zuRfmrT&U^qHy)(0`08zT02gF?-yZvSQq+;UiCg_ z`8_`WkG5<@qfQci?%p!^=Jr;Nmy`iRmsEdVIZbBdxndNkLU*QJNnrmR3O8x0wMk#z zB5f&=tpx#5N|%2v(jT?~vQZYy7wLK6Jm15m9`Sc4J+W-jcfK%&i*+?zz7`2NUlw2f zmnOoPb!IHLF4$?oOu{9wf5T= z>R~*@hOB(AVg#-8)i4W7O0WwZf z${?>_><05?wgvX=2A6*nd=u<@I^0w&Q ztGj=M4EX&%UsEc@D5Cy)OFJFdgiCJy^Gl*^F5L>2u@V?@V@S{zD{=hShTvGa3o7o} zx#EOvj-Q;mI@A{wY>Lb*gBSp8a zlWS!d(%e&&wP?VR)HgyHm!uaoOzD1^1z&$^Tmd${3OQN4D+`N$586pfrsG?1BI5|U z*iWTtUg7Yi*qC-s>|Xv%Sq2G-H+gfuF}mRUb~JWq+#*8b(>Y-!3h(dDFf~do(FELj zl_MIU`Po18nast!bN)BQv*b3cN?IW2~kV1mOY&e*STC92eI2(yqUYHseF5!99KM9Oi2#OyD1y}!a?M`72$ zWhelKb3Emiy?#9@03Flh^0uABl0tu(PnlbO>}h=VU5^yW0L?BbUkCYrsbG$BqRPI( zVWwGEeIQ0@f4LinY`LTOU$%VD-zP~sUwydkZ1f;LD?m&t-L3U?Kcq(D0=^100dhsW2 zg?`6J0%EN8)Wwe>oV@Z3uCD zaI}Xjzj`elF(heGynTPO5$1o4Rx|VS2(TPT(A@#}goLM?%FBaa)udKRD*I>Yz>u&V4rTd&Xs!fBX!Lfn5vVU+J@wiF%7uTr@D-k4$e` z$%k=tqh!ht;Pp0%2{EDrRT%iIyn&&a;Bcz}I4B0&E0|}}bs!aMK)Qb^>JW|8!#RZN zJ~lP!eJq*O685I2>AUc*RW{$sBld@yqy zk&dsVl-B;Tq_(fjL^pr1KLhNOX3PIbrQnOeZq@D8>?4ssCf?R3_YLb^>)e|UUP4aJ zem(6dbpJ7BE}~p1TRAzKD)L>3Zn=D*B2~zHz9rJkpDW%n!UO zM=JxW7%hG+OE*^TIa(;~clhYkuj?k{VAcKwI8I}f+C4+@;7fmgeeP*v922ap0sOq5 z$NTus7p8v0M@aQ#D-6@-=&umd$ouYBJl)sDUf}rOcHpF@f^WO$S!PmxWuMd@=ieyu zD(DJhf)PdbBytp*fYm%WU1XzwYM@w6vn*D z%)NnWW;Sj>ekFed|6~NTh2%*lb@#i@qP^;qCSA|1z-{N~oZ$W7DL#lSczcq2QxwOZ ztY#~o|FiboT+96tj{gx7J~MwUl6_o$RttowPd5A1lo_@dCGeTAUe7Y&=iIVi2EE6e zqXW61^a3IOaG?`-{h4c9>3>bMeS33`+Hn%9QTtRD+n9fegYDr$w^z@6s;`qtWMgu! zXZt_ZePQjl9~{T~T2+~fj}9ixSW3Xh(b_1QsW8o>$`XfCqdy)wocC2JxcqCgKD<6t zJVl{Cd1i2e+O;HzJ$C0G0!HefWO8J{4IJxp&uyFvyZ{E`X(;qRqT#D6yu`w5>+oaIIsWX- ztWDCUOmG@_l40cNdJECfx_=dVwraMN9yxB&5<<(QPk>_nj31h{RTXnB=%`KfStZWO zYiJ6A#)4#fDh66!?jX?G`y2gB8N9RGJNSHN9p&a3=?($n75ITc%iW+bb8({|k8!Nn zCSZTwu@B=W^C+~`?JP1QMpIk)Sj=BPsS()>Q)N9l`Jr#wgPdloOb*iu)qIx476i%khM(h4X-(;}hTXM`nHc-9a4!9SCCU^uN5I!fc`j9;t zD1Mr&(c<#?UeLaJ*!BO2)iRIg12hMYkZylEvPH~|%U53D9C21skrf!sOf__F4sq;) zpn&upHR(6ZmUE!!Jhz~9gb04?Cc`P=KLC%AOXznkMBBL_+-l?kQl-mv*~Q%yFELJl z{{U+|?y74_GWDxM;+u9x+iwvSvz{aChJs8_&rz}3V>+{*rIXegSn=TJSXqv4&@g`r zcJNFXkE$cCl>^$ z4aUhdc_O3xpw6+)5Z4z{H@sBK85gDR_p>`r)tX)U(UKVHh6GEFHQAB@sr^B4`uXY%XJd|L$V;TJZ@^O_hfMG$Y@-G1b%w53oyu)jNbR+=Wd13!PFN>+$E zy)3AkP=7h9wxJSdGvl-F7C)&9pe+adN$p!bRdLSvleP+cDE|d{9}mo1cDuJNSvv=@ zk~ugm8iD`0DZI@;(2|kpufHLf#WOA~{_?6kls5Eg=4z!vmi&#!Fx@3f_h78yN|6&5l4bt`o2<&>5(4D@CHfIO>Sb^UQ@rZoO+W zY8a=Fub!qR@N)<(>**9C1)R32vLnp4~B4&OMif+JKc3r~m4!4prk<{bc_o;vp5vAPKrCrWT2II?W+`yWXa z^a5qEzaapU?eBkqu_*R>KKB%ZvPE3yoDrtt{Hnj>>{Zi@-f=f<+)|u8f-yGPJ5l1k zFQm{N`(Zm=d+3;f7H5j9lK!>GN-SN7o&VCK8m)GXOl}W5ov$xF zN+Esl%^0NrH`q7-5ofc^5Z;O=wCh20++pjJ*_U2eFSuWX&STRVj}AZGt03#cRuANa z$<1^(y?KA>C`qYd+yuPd=PIW~fvfv=X+F1A_ZJPT^yJl{I(KV3eg-$UK<<|L#A)hn z$yv8eaQlojr^o2?A6wVWdv4Uo(4c6CEv);}UYD48em2p^xMF6SPyi3Sw$szDQoNY^ zKnolXC$Co)P}di6K%2sog|rjOjkcv=bSYf=8&`h-evWDfaHT>Pw+@wc#M&pylP38< zDvP;Ho=nhb`WF9Nd06e;!E%_n#|x{RMwhA;F@AU~?M`KOuo5B|!YSqrkJ+kVSC ztiRsKW*+sjOX}ph0x%4er!uSrVDR_SsWdP$^GAS?7*!~ObXB6qEGi_&Bo%5~jD@>x z;$YQm1BoIHW;RV}y0yqVD7Y-RM(EYa)C7OI<)0GaYA?J$P*Muzy4p0KuxZb6%KM)E z^RAeyY>|CMk!|LIN`&GjRO(FK$uM(>rgv`c;AW-;o*nGmiU{PA`*VONnZwtrqoKPA z!0<NblUV;Idw?p**Q{i&qnhC|m>1EKi^F}nZ?oVUZ=zV15g4vPOtZOX+ z8HUicbHZMd4UC(@TPHy`<<=D-%rQW+4Y%o;Ql_>eh(5l)bA{i61@+kkX0ZnL^4hQ# zWgzf1+$$gG=~H9yGsZXmX_0B;uQY$G|K^^qOTuF-C3Y*@rvR&9vw<%$bG=bJye4sc z&b+5ie+KF`n9`b3Ou5lZ)%HXFhh*rV!`n;v1*1~CdxB$Xdid(0Yl;3f@l5JZ5iK)k z^AC84Jz7X$s%Tr>Le*+oD;8%nq)bis8&cs(<+EP*S!+H`G6tqYWJmy~SzCWDZ`-CE zJ28L>v_OIkLwC>nJuL4hEg2n;;c?(SVVPStf-<$Z-2-3DQkh;GU@j+`jQNzjXv`?% zzT}iyc!;&oC-;(KfrIqokg5TluaNPeE_ZQG@TEUc@L)COvN>vOUZu6hO=)Lq(sFN~ z>4iM4(ROYG$zHSc&Q&G*D^`EL6-IS0N;TFG3SrsW3JU(a9I0=D#qA^{G;U+$2``K% zaY;yQO>_dc`yE?~MaBk#Ajc{8W14{>Jhfb-TnGL}dn;O$Uy{Y}r9YVC;4oR%xw^5> zwcSZ@Ahrs4)JV3FP?vX6ZUDwfosd5 zexNx6HBz7)Q~F(aDV9%l4PhZ+ChtH)#ce6?)zb%>aW2WH;`FX}7tyt;z4_UGs9$i6 zb_5jJ?y^?8w6f`}KiM$OSs9j|YCevuR`M5#oT@0&(xXfLkY-Q!Rl-)9l~m39RjS_Z zP{HM_#yYMkE7Z)5>(77clev=yG~KX{56Xm9>+C>18#{TYO&Ye z&g^OJi&b;n+l`WD)qw66fa;GR(D~+o0s#$?Q^0m*^#;P+AwEM0isC|Fpow<0k7`&!3Y zAKq)JV+{HUrzU?saDQ`-mFyveIVr@=Bwk9&2Af(vbM5f3NhKB7O1dJ`LsU{bV5op$ z+=DjAG?(f^GlbZU*p%dL>BQACMbgQxZOl=FCzR(pq)`6?6FB({XG7OO3r6PbO}Zt;Hs^e~3hDxc)d`dt05RUa7F zn?5>ID?G08DULZVg`94$0*n9PTgMwqq_025JADW&?TBSn9jgx<&RufM!OO`1Xa1pc znI?!%@|$^BfTk>rA^G7eC)Gg%7|phB6b+=WMX?swx$Pu1UMgYM>jFX_ikYNvj(0*E zI_Me0MBsmOO!kwGUX{LFxRtCUsD|j_Gf`={z7#pO5(_fjuIS(*&UZfE+I@Q5;fs~k z2MZmFVjtGH4g5U5VTXNv_+_MKqR%NM_sXiGU%FAJj8p>A`*!to$F`SJJ1u<@CbdkU zTW?6ei=bm5c<@sSlqel0vc%6}`(nb9sM^D3s;7=}ewYt77xiN5I^(Y;|-Sa7gadq3#w# zp<3noX=2A@n2r%cK^;j11kmPw^C4NT-|?7$gb17LW9+=$r5w5B#pC-MLMQPq$D#Jy zgG_&GNRfpnG~N^Y(<18!Gn@4+!=Q3{6&Ff6m%iYk+5k-(H;{OGoITx>kX!vXzs?4HbgAxf>}=8MBAR;@F{BdH7+8_em*e4YyJ9unL5^{Yt_u17~NB zwt(n$58ss8c6NT8m=wlb@#iN0cNf`8nTmhYvDQL$C@RQjoa}8(0hRlqBATuM-P;#Y zQ9CVkGT0`mLB}$*-pUovy|CRdt&`~M@S{hB!X338kvFQK(6Xc;F_2}mW&7ZWmTqtS z9{q>ScLj~z&R0HE04Pc-Aqd~}0crTtuDEC;rD8Q0?wGvqTj7pZlXv9~BM+XS=6I_fN-lPOVR)jx*|-%0EiAGLAB6BrcskSU<5<3#7%G`%M$s z*V@Hn{MNl`w!&|HynfoWPVbgc{IIAXM|<00v@OJqhPP_K$=IzN$DF43D)lK+ zg|q0Vy!N(Qo5WP)vCpuDP>J!_HBx`Rkw1_6q}--_a*MURbn8@FwsBde%ZNGpdsVw# zHYyG)tm!4A=A8WR40^}RRVS37J*cf|o7NkZvDIw-C~5hPy~A+l$-!YY_B8K&p8^v% zu&BE!_cc4=LU613@dT8?9_Ae5%iaw8q4V`Xl+0|swuI&$kF9(d&arW;Y>|IV4uVjI zj?^h9*o(7e5`{qp=MuM_?ALUO-3LG{euSh#x$sxo?oUZdCnFZJh*VOUM*AUs?CqVs zL*lL!&REIG-11ZKC$Fo4nw+QjYf}DY7)l?caq>n$wD$eRr_qG$cIv1K$2Se&0(t`0 z)m9|7BD$(=_gss?3T%W3mIr_M&vl+Bgq&MG!{--{onGXG@;HT+FqwBEDJPNUUOmy3 zjg^}!PLnN)%lEbb`*4Z${aofmV^XpG6BN!d|4S8T^_vA$}S7*h!gfwnFx^vl@TZ9qdE%Ql`_? z5EJoKKzprJ<=Y(h^)1(bP3U#5N!w4WDcszX7AN@RruS7O^nQc@e?8kBU7S*o@(H@G zqbvR=Refv>stD|QcSRaEj_q2gp1c;wxaq1GGpM0C|I;*fhe?QM(7fPoe@m1!UtvlZ z{{Ss1>f`YIPm8PORS$n76xDQ7$_!)FsS2p#d{asu_%d>m$N^SvrMLFYxo;Hz%!SvO zM*n=U?Ji?9H?w=Q`1Lwd%?)4jy0?sY@)O)rqn%i^3~oS3HmX5weo2n9q_I-hf^TO2 zST_(p&|)qBB3ia6QSFWOPyM;%;=7gg+Vc?5OsQVq@FjCu$}NAN37YzFsyFX3ia^lB z3`VWpLB-i`E`dM}QF7;*&f-E=4A!YntQ zDIX#!3&o`(Bx+=;Ek5%MvDGin4J~{6Hd>witN_(j7e|y`EAT~HF|;tD@egUi_obbQ zI%7tr-LUfU-1vVE_q6&r|Hv7aKr@ac&WeJR;);sq=Cm2hf~5n|a>>Jk#)xCADX~(6 zBDzV_CeooWwU{qe`8H)pm_syX?U;xC$H5?F+Y9dZ>mF!%gGSTL!a~(Is@zoo<|>^% zu$-&nfh6FuX4^+u!UQxrT5r`8Wq4|tMpCIR`cSVt$E$xX+4YH@TP8-?iZ6y5te+lM zr!nYBwgGs+SE)_qf$N< zW9BwDP4|E7Rta%@3?W%x>S&bNf!lP&y(W^J!hJ&0N`=r<(gqDXG9@FIpRP6;LTiqk zvqLH}MYQ(;b#08OS^wWPd6@rrF!x7FC1yQ}?iK1qNC7|QTEyqAgUY6rPJO%&TZ38$ zWE%&1#}hh>FwdmZ&=3-kTvmtAAbe1RhG-S>+#yF(zl&56NDA$Wn(YVv4@T-Bhw#QKl#vbcWDfeBp zdtYFNM5o9TOfgCGUB@z3ngc-9x|Car&2@SUfH_cBIy)gacH?n7m%>ZNgp?8TnBrtMR>BCC0ehko|sk-}PP=?%|0DFn%fE z0b$YzKh;hj6Bw~PP1S1~{*a9h<&oh%RS32o$*Lt~_Y3I+ohGdZlWm&pp0eHMr5hab{GE{n=5niD^rKiFu__ z*~06FO&im~2zuLAiE3+gI${qA2$FvU{A>(6`$=!C3*@z~1H`Juvx38^YF+g9h_XYV zVusRqUkDi$eDW}*XrvqB<}v`Xq}cHGi#$7D>Z;5v?nK2F$I#G@jvz5Ndgp>n3ydu1 zD$2JJECf@V+6>y)5`4a@YA31*6|GH%%WRRj++c^sKf~n>+01_ zA4uiM{OcRA>AuxJnNTx^Lo$DUkuMy*ddn?l6;BMxq@B!tH@y)zqk7&`C~JHj`lgxl zbjQSL8!Fo{P#!Clu;z<%v&=L1PP5-^SMXw!}o^+;5y@qC!;En8IBXF#L=KG2k6W2Q!V!aSf6>r_6b z*lJeQ9l}xT8)p3OI^&{LEPd{#Q$clz`p}_4INCL=I4i#=i1uK)+)E}w!nuq~-;6FT za?q0>40$6L$E;IkKHq=DJbEW|rCRo}7ej-;2mGDSr0Qp#Hr2QKO;honD157lQ=xm~ zL=Y<-2eIOJe)-Ddh?WvDah{o}Own#i_GinL0gDxdn@{=r>a_V!12v5lN)`6`KHs2B!rpotYo{7mGSW|p5+OOk3tjtQ6@m||eO#pvs-|A!T@g-b+kH+lK z3~G*zEfawnn4@~wCv6MNhX>k;4nTu{bQxWi7H4hB1`Ko*y?9LRy%+2cq_8ftzNc1L zYvV~}@E&wl$9ZHW!H6nb3?hv>4Wcy^h$1w>=0!^BfV zmx$sAKBa%f%w7Mn6Xe0I^TnjRx!C9c&9{Mhf$~-UB=+j7?M-#~Gp9^r4xlRL58V>M zvJ$KoTI|FtjK7(D_nkNQ4yUTn-w)8Rg@cnQD|h38eDIDly#4mU`bBbh(kg9B9W(6$4 z8Jk1aUfu|D=R=TxH5sL419P1EX2D1k;KRIw;P#s`tJ7|#zC%Mw7PW{(UD={Z=N66; zYo5DcP0sO_je9!cRJsSi%_;LV`%LzzUy3P~&82Z?#)5srsUjxJ$MzpK zytSp2G5&jf-Ky8KbHp2gJ}XCFmbhuAuG^ISZaoVS*RIbT_v$|uGWB{1mj-B(OHRQ; zm?>Kh;_b(k$SI!>0`}aeo|!IQmz%+Ei0gl44T2@dk0WJpaex|{uFwLxSiNs9;p(Rz zj(=1fSNIw-fqhUcD1E3^aeo z^t?>A-Z|=V)}KNq0vXAEj+g7o_V21eXpwGU=2pc8=23XJhI}t-Riko0Ryw@&UVeO> zkWisYGqX+{o@4FwSdl`xa*4}qqaNqL_g>X(*785Yati^LrC|8tCnR#GvN0rg|7=jg z6N}Q<3|=Dt%$YZ|A#O9F8pypVl%0RsAE@p(V5fK5v6!>z9q%$Xbkv`zNUsb75EkQ} zh7;42ee4GCDT8Sm|13b{Mr^x_()}9Zj|{n_&gY+l_gq${fv>j^OJ+O=zmJFpmV;D^%jvHgKoKhQ`Q6l$JYfJ~CfU;jw@I+qGly zc!kn|-$o`Y5r=YDa8;jw-G0p_2YMIa6sV}tFv7BN_-QX8QX}H4N5K^>jdZT!H~!Oo zAfm<;&Z9`aYA@vXC9=CfovR7}sE z6vkn+P`2h0+#6@=(@X9h-NiSM=!5i&vmzjT<>!(QAlrh12X+Mo&4)QPD{y1`2^a+* z#Ufbob%3T*vwD8fh+nw@L{C~E{x>0C=3M|R&J0Avsz1|9nbX$Kcp!fuE0Uz_r5`7y z|Ld)$my>8aQ;;jwoSiN&-wNN9@z2g9@=CXr?seVro(+QM8U5{f#fYh-4_UbBy}V?t zqa?qdAF}XMjpfMsJ}+qzW4w(_IHcERAH@Jp$vnWajURg8_9zQl9y;@!zMz}&D>u@!PHK_N5A4}CWL9FB`UE?Zz&&NY&Uj7UHQwPc>w$IgfWt|k~|6ON82kPovC#ro^o<(wN9Zxv}zPm)2sAn>=l}$1MnrLupVK2&< z274<}R3>T%hd@NncA5XDJomW>b!Q2490sIWeQytqL17wZea?S9(pPfE(_&a+6^WQr zcb5js0C79BO{wN=Rt{aNwvXZQ@6t!M-R$ub2AG+>7aan_9Y<=l9j4ea+_vV0-%Pvd#fRV zaWG8NxKCpATVRNdctaz{xZ~7^wyWEj(KCtv!7tCf z)DVmqMU}#?bIGkQ9JgdNT#mEpJB<|3ZI3Y>CIvon%Yc8X)(Mg|bY_RXG8Z&W)KdT= zumj7C2P*rntBET(tFuGVt$TLymER4VS}St!UuXS+>o(SCs8})~YohH?*NFO>*{cI0 z7b;>!WP*HCF!a)Ux`uXgbcVWjlrwcRAxhXO*99@Q5fcrs=640=X@G*aWi|!b!|xM# z_pUrVo|k_CTM>egOq5d-U^Y_njDHD#Ku%I*=I^*tR-zy+NVdmL=#M@5KBTI?94lqWYBR&aA4d)`!v%LqoBBY zWhMBwC519~rtPJb-%F%D`7{}h1uRC)FkOhMoo;Kd2Almjj>R;Sfk)vc1FJ@@b30F&PVx}fdCn58JCs<=@0htyl?EfR8vPt_k2FPQ1`rR=nLW*~cJ`h-qoZCUU=={0B}~ z+EQ$!BK3M2 z=E|$y=1D4EZ*xau?VKEH?O-1`j5^<1mpnIp7n~yWxZq=ziI{8Bz@DSljASbKvv9lA zKvjXbQD}l!S2X*h`(}qWskmO5&BJ%_y}$f8nc&CD6y4YL4#gXinBTBb5`TZGvlVWH z7yQa1%5V}HX#-HnHhf9iEonw|RsX2<+DWR=Kl?VWx6@?>e~qu;X&xADy(I^6Gk;go zEZn>#j^FMym+(@$K?xs5M#vD?*~6S9q}KT8MO1{fWjQ6WdxS)RH}Iue!(4>FwG1n!IBFpA06gAUlY~em66Qb=A#04av#PaGrIJop z->srPYT5iwhVnnd?IamQ z6L+?}r`Tl|oJENe!`W{2cRnrTP! zd3`D%;@86cbl&2D!LIt@fa3vG!kmd%i?gJgu=PZd)yIT1g(7=}^fhEehxDcfK?MKL z4?%~+2B0`IcnfX!0A~k<*C?^QMf67&M9b8Ha zn19n7?&jbvSLi#k*~m6@&?MbeHp{no1l%o!z`RG;A^aIczCsgA<}*)c6O)Vg1eS_^ zl9IjAX+qzHRJD)6x3?z`8p<^fhGa6U2pP15=Z>J(IIj6MT{814T1w(QzP&3C#2E~b zvP%wBT%bajZi|13Ys#f)UOF>+u~T7sr_qvUX3g)K)bKjQ8R=`*afL%$=@oa!Y2Y)X>~CtYpCaD(D;Nu? zK7~Yr{uqr`_y&&_>GWst0D3@$zrhi>=YCO(*Ry91GtlaXpae1e6B3PBI&NaBBRt8? zTs+ElOrcbLHLfCRh7g>3Y+7c_Kg+4F9}Q)BiVC)5@-i@gIWc!IGb{ZNvKHH1}vB1w>KXQ4-|@ztMMFAyVc8+W;h^I>-O z@K|1(luWX6RS%kmC+4TQl7bFlw(l!GoF==EuNQVQ9=HLLNz?$})k;l-g@4kMjX>DU zD3oOa>2^oo7Z9Z$9%PN>xqlXY*~W(rfm(u&fTlSHuPGZJGjH}G7P&1d@LQ(3JAMOy zIs@@znH*Au#={t=-=uO8DFViD?7SP0j!X{S7TKr`jioftuqpgBS}$Fy_iXsmOo*EC zvKM9Kh(h6S?FIvFQ#5pwPJ~;x`n&B|v#ve3UcIV-LXGcMZYJ^u6Id^jUw4Mc9oz{I z#PsM^4g+TKH6g71dm0M?_VkmtgGGx={TV#BGUtJzIZi>1!j12m_Q5eiJ5ammh; zmvo1gZuYFN)>IFjNqnF2b2_w{uWHZtSTbI@f8aC9HiYm8`$rv`5$5F~R@$zh*h7ZY z0z~wV@Xxf@xXi32cNfGca^0m}?W!jsnSwE@AZP6h4M2x2(qi7u6HV>HD|w!Ob=+I0 zhm}05HrOSNh;R5~Y4ne602kDUVe&uj?Wq2EUg4x4^~xm5KD+l>?j@~MLJ`zVuw@B{ zah;BCcSpi!&%!vTVjc&@t$kUK2=|hN`?u)jSnQa^tS_P2o5tNXGOscO4cp0=H{Gg^ z_ugqNPBHQCf}2aW`$j_8iH}Nu2w2LOHtSkXA~mDikaEa$8gpjrADuqQ9G9VANR%q< z{Q1nT1e8;XjIH5x~Y=J(jBG}8Yw5wSX#egw%4Rl-WeS7s~Lvjhj z0_Sx~DJXsBmM?lsL142FrTUGp^3Y-K*$(-LnsNQ49iPnE_x=!(!!t;qj;)8i8IMn7 zw_~Aqud2dhFX(FqkC`=pLUPU=n1OtzJ+{evfIaQ>^IAX(IS($RU31!W(mBl%@$ z_ET)uMM{Mr$(0#lDR^^Qt@Zb{vLl^MBC}y0J|O4@99Y zlI{ZpuJih)Mma5i_u69ny11b1V}?rI5%%6Bn*sv=o9=Cb@yCUr?ev?5L+FnX2~d`} z$RUNcCbXM;;18!EiXwFeBVK24$ZjULYUg?Rr&~?dYRDhl;?Ts3^ZmY zw#DTn6sLHpVSidII6wlG$u3#^uz8?&G?y;xiG{zXP~(ArFYsB5S7rLUBNxo!Tsd#~ zPF1!PLC7ZNX%dB--Am-cyCsvSm~a1?Al~?hy%VI&ZV*?b(QL7IEP*NuSF+}>y(3VYrD-EGYP>y{O$x#J&WYEQ9{NS5jl9G}xc6^2pSw;Z6U z%+%AL{N_@BT3k}4tu7QVQoGUlv(~orT_A}9gN2;t^~h7Nk|ytgB3*G$6Q+k6l#C*;%S&O8wqN0P32c~qEzkb%o#ziBUUsi%_PQy$RmX=SAnW**Z!#S&~Bcg zvBR8nZeabb#Oqgq$UQ8f}(~laZBCL?6gNT@E#hlj2bwkpH+O8mPw19jMv0eEFCEdW4d@ z7M;KuaNug3f6y9ilfksQ+%~5zK8xj zjf6&7qE9eQ+nq}Sg2EYQlF6DeO~>APMrbNVfHeNn5_@a<@W-D(gNH0#)4 z3ebG3J9W`ryIsVVRuo)zpE_r4e3wQfDiBHqQB7vv>BUva<;3}-vB}J@U$%3&rCXwd zF6Y)_VME)%)Y{s@w|32sQ@&hqM=P#>>0D*+I%_^sytbB;r5#dT74JG6vS**>gJZ3% z{StHt!W`TPT%5Jd!FEMz)SK(gXjh|DD7d2UbzE@o?X|lk4?Z#Lyu6~=b*xrzE*Y*G zH4ql2xw9&8SIgDnZ1c;Pfmh~uVRs2Zi&@ASV$hLGPK>*@|U1tl-j*6xNt{GnrBYb75+98`8MtF*#u>9ZB7u+`d@$lL&0w=!d zx_Q`cbF}xxM3faoIcuYr#-S3Z>086g9JkgUD@MH{`MUj)8($UryZ!qA^Iy<7JG9lkg`J8hYvxg`yi*3~T4R`3yOyjuBwd}h9ms=OyEzO+q<0G-0 zm55*5=iy(wp< z`=b+YU5!pN^l1g%4RR$%{*tRmpY+hn?+Bj6PdgmIfT&14?M=^%_Qxc&7f$JY|s zBG4T@S7&GQG$4VA9k!dk>6Ch*SBW}0U7D?I;*~A(mfO0(IT3$7UcF=<3(&-{>+VoCHxWM0*B-MHUH?Z+z=hxsVIW9-W__6Oua&>zlY}o6j&W@C?GAOm01}@zU#m=Fom#%&gmY=F!=R z)M;ax;#at>ec~C8|1!Qds@V7cJ`3P~O7}O337;XA$AlAwnAmofYtK^)dXQo^#g2g# z-NoOTqwQ@P>YBg-nv>goF3^ja{J9;;gEE5Ns6MzlAABIB<7D5T6TXvJ*W^t2&Ps$! z=0(4A^MM25I}C(>?;L%&k$w2mq^NUN$J3w*$W9#30H;QMx>%^ltN8KrUXZxv!^fZ| z&O1S_OoBALuG>0hO+Yr{_-^OV!2j5p`@at@^Z7T#atxi9;(%?OR6(wyD{?O@(!=b% zqON6Y?QOr;xS(9ecRKrJi*$&eLqyYKWFLFEHF|zz#kcBzOn9OcD7dKLM5v2FFR`Wh z$1Un0<^0`Pe0iZCMfYqODI?_gqgfN?A+bl++%?z*CslRBb59bve7j5CEm#Lif?!+Y zyq}h*Fu?5!4GKGOkDJuRFl2uI?yE6oost`wzcK?H=!0&`e1dMvDdRwfE!!2c6kMfN zVoVpCNxO4@cb|QgE!v{?R6gu__Bc?pigDt4;yZ@$>VfnqW}SOd7rS+u@ObtSOBE!Z zB&X5e55)~!%RVDdr`d{R35DJ0!POUS%Dmr}$MJVU zRAty@i#`Qi%VzoY^up2M{PdC^cOV3nurLtbet+wKf*wdge14zK`xj1xAoog<@b-hY zi;{pV=jXIMh^8kDbofqC7;y*xMdQL>HwP&$6A=dL{*Rddi20A0|A_hjp_r4x#7DSL zrk>48C}&TrA@7`jlU%*SELUOtGhKD*(-&d~UqMkw9bR?Ab^%>;1{A6Hd;C)yDlMr%9bz08Gv_3Isi zT4{&E<`S88{8EDywWNrTdiXY1_QQm)WEX^{ibm^^OMTsiQ1khi*M^PmNCBJu2grB0 zg)7Pz>rgB08B*Y$sXCc1G#P!qX7&|-*E1J|<=r-N4ezdAqHNEd8BUpUjG>6b#G~t~ zR8Po0Gq}81v}bR8SZ7>k7jcH(taP0M6m}2!?El_s1)`h`TUHe(Td`i8x*Mup$*hXK z{#!3D`@}S@{VOol@3@T-wir?55CH9Wbks0hIQJ%5%Em0k0;bcfWh7$% zgvRVW;a^UQ`S#8fQlY&0{o?X}*6OLuS8tHVF7SPcovU~Dntd@aRt2ILD}UkA3a2e!FRE=t8^@LU>y`#$j!O| z>c|*Fy5=~Bj<*|TA88z~#TA6h5XY~+`NBZ=7=uAV`{I0{f1aU#kri#XB~;-)o5B(b zb5P*qg-#gz7^wT#=nelKz3|&%mEqN)et4|Zu`M@XYR#!KeKO+kmm2PXV*|6jkIZ{F zHnsSqj!vtOPIqOIw7>3cA(LKtGz`J_H8>ns{!+K+0odsclf-IEKGT#+AJ{GX7OhBk zlS3)BTK-JFRkLS*vH*>8{{u1Hdef_ox#>}f0{WF6Qx;ur-EZ135CO(|f%dQGQ>F^d zGa0o|A)D?a#$5W)*9oAb2-r9o8DOl~RKB0bR2h!nrKA<~jIj+a-*t@?^T(8Xs7Z3>lr2?SQ*h z^x2((B#${~Iaw+Zp$!YWAnO9kgA;;bcg%ND86 z`u6NbTeNVbV!JKHP-!(Sp4OswgHwMY)JIXtmj}9kMnE@w<*l2a>2;gm_Z3@IX9Lbt>xnMZ#$(5SSaBbuvxTRE2$y|;H?<`$6m*w>h&Jjs3Q8%A4O zLtwWxB4`F60mZ&Y+YQ{!E+gg-SQAbBvI*a8n4Q$5ZmIb#yiyYdPYAex_& z5UBirmwT)K?cQUpOOygrs$HuIJ~f41hBgWAeX5m|1bH|#M!k;Hd|b_A6X0JjnuF~{kZ z7}YImBa^`LZTfV-Tze2qP=+Ggm1CLV*6=TXnI}2+-7yMoE6cHAzK$XnA_4RB_n3*} zg^}^hIxjkOZp$5Q%e+8K3U&gkKU;)$#U~cRN4=rQiHC_MdWPpNKU+!^rW`M>%_}<6 zt6TOyKaSHad#&|_O0{eem+(e+Voz&*VR!VB=IKTaYAeobufKd`Ki$JgY!Zh-*7tmW zR?zsGk32Je*v%BgWm1%Nh71gTs^PK8K(?~&g)cum+)hMxG&@aqGyt zOG=x!HfyyQu|e{k(vwD30eFHXRw_z=*h}^CwtTPiI;-~gfX*R&i3O%rb_}Ph(<`1( zcuzGRYdraxS%;3-^I`>^YXr)1XYk!E0?WF}mTX%5AjsDOcLng?ToCzmrwO6HzVB`P zzW_L*Us!4n%P!rTs_T&v>N_T5?;R@zb_k8d#gNIxL(xWt1S%m_sb@4Crn1g|BQBhw z9$nZ&m<7?5YE+yJ6hW?c|6|MJ|8hS;!YmZXMIj9WH1S53w%qZ7n)ITM38cIbBX>Jd zQ+4sGqGr_b%q^)oc}MZi4Pjs@$NG=gZ8y&8%Y@q_`b)$8IA)!k^(7JKNftr#bZ$4Z zOukHiT)xBfacPU*NhR8AhDEA>Ns4hI-GQRySk0^Z29y80j#aUZ$ET~|`5zB{+nM+( z=o%g@83Bua7Z-NsoL2hI#(2E3lZ^UuF#`MPAt6m6QVudDSUYR|E5Ui!v9o@dmi?VY zaG*40yCk2!>1#9c{H8Lq8oh2)!?8`8+n01J{BxgRHt2z8Nf!G(G)&Ncr6V%uG_Sq* zoS#oj?_SUTwKQHi_oi}hV@q^v)C;$i`2?gADOLZiftvs_AT^`zEwO`U7ZvKh_*f0D zjk-4aX)xZj|FP-#Gzc&mZncs3+ANhhPVwi(z|c~TfF10+e)1%8r~Crd$s4jodz9vJ zic?;DjO)v>8gbVvp-)18nf$*u@WDg$0#o*fZz10=g^=Z0V8+cb?Z)(-Koyz8DI-HA z?kB;CE8MM#rQGI-?MSMisTN?v6A+%YE0M1Z+PBxk zbsZz{;>?{f1*fK!mjuWe+|WUZI4x}Z?|uv!t~i1^252&pU5tc(eI+5ddGe0~GB)Cs zw$4s#0uRHT9q<(!l-HLT)wKhkloJbYL2OovW*{w~g@}jjPY#3lp5%4RnewZsb9z?8 zj&RoHmaF!7^*WSgl)5X%QPiiyL%wjue&X<9vY=OM!yHr4RHH{~tphgciP?VZS*t~; zPKkH?7sjWZo11fgYX`l|>(k&*4fdK)j{IYfv6kcCT_brOO_63z$k^HCgwzdh$A-O; z#>{-xQ{@UkCIFYV?)v)bgA!0OX9gQH zIx@Qkh14<%j86X0(GrjH<=*HvM$C4@cJ)aMN{y;0QK6K7`807;4f+$CxsQyAwqWmE zO851~)TWG3j37$5`D!r!r&cZQYs1>s$$;%M0xQe+U#To%7bnX!Zu9`~dCmExK5*@P z=#=QUx1#Tv0(G4BKRRxCYF`p3zIm%s3fULz3Tq+v-Y$vf?tV{sB7Qh1-vV-yZ!^SU z#0%Xy@^rx$=0_9rN_HBQ|WXe`rk5EIT9x6T*oqSnMtOf{c%G3dQRNG276NXb}< zsjev1Co=?SqMwXdH*@hA)iqG1oDFzRybbMA;>cizxUR%hLJfChmK|w$OZl*MLpnmRJ7#$1opk6|V*X}-0>Bp5lW*12`8ISVhOhO6i{j;Q zBqBpw*%~Uq#&6vg5}~zP-Cb=DtN7vXnC8>IeoQ;_F%xXYr*Eb`qL=N8aR&f*%sEz~0bbC^OGFzsQ-3=7Ayw*IKkYHz8|2{74`U zFLoS%XG}b|PFU9wBg#sAx9FKWnDig5CRHsJz9Qu97L0BZQPSd353B?9QbnArntsU^ z5mGH}r{=nke&|UWr&qt~Lh8*@P6AB)hE4(KBkz4fS|LmC6l~*CsYc=}c(Wp~qjqBq z+h^2qOWh|OH#u0Mg+(8JrMH9KV!9ik33xSsaX~9xBUtHq_|qU}om+91o>qm-anb_~ zJj*(VP~ll;to z$1Y`Q$?QH0RYfbyH}gU=htb2gLv~tw31_DVUGx4LzS_88%$=Z{;TKy8`}mj;#d3=B z8RRLr%9XT$CMSw8bfIkSjqUHTVapnJXCF6e7n|Y*mR4ZPh@(xbrS`u4>ErEVJ>?8D zIWFDr$`#EyY3aOtlNbzOc0j=Xn126%G|{}PT`l#zx)Dn{AfxD1rro2?Ss0&dGO^Z~ z>#7nPLiR2LU4GrUXZQ7z)zJo%6&6iaKk(UlqqRmQofD?>U8_XDG;S2cLjb8SbP`=U z*;5V!ufILN44ADhj(g%`l(M-q0~xtiSkN^}D`>T?v1z#!qUQP~ndsXZNr%dRwh^rR zJPD(_72fcLR&?^dQ*|h%)b!}Opw(E3FdV?>m-V*SCK*W!i)$_HXnXS-#w>_-_BpX`%fkEA2I(C^B*z)5%V80|GyP;FF;dqMOt!^ zmNw*K3-!#q=J!rO9p=YXic&Iv&u@z@Q(zRm)K;-r_Dnc_8p{J;1< z$RD8j@T+Ai;S2FO7wlvB@SKH^G<|lDk$FkM>j7y zmZjM8S9um`pN(51Ui|Op-2{AlpPg=2ro$t2i{8f{B2l$4_4tVG=?0aK$l2wZ0q9rs zk#i#2Yg{GXxaSbGHBr@n<`19wtenWprpN@UwYT=hp3<0iAJ5fKy9ATc#EhtOM=!YL z{rX^7gP0mv`^-Tdug);7GL*xS_sh|R`~o6zvJ>L5y%{kf!bKLJpGxv7S1~Ga(#?*bhsloh0KxalZEm*7D`C@rZtPVJ{yFl#sDI4i z&TI$#Wq+EO6`a0u^25}}*2=Q)xQ032MEXq}{g&NSUbGNkn18sq!D*Hqrc<&5^0yx= z;gxFi`s`kDRm^rELrN=KPI$7~qqHT$_sA()ak@_WYs2rl3zV9*bpBc$x&PPd$Wvlr z5Sn|q&RAopcrnI*usfccd*BeH6)DbzQDa55k(cI;dw6VHP;=>MIyF~C>Xg0DKeI03 zO*buxTkdkx5x9n_8<=*6a;q`fuFoE9y4fgKm_Uvx}MPhFQgqZd>n z>4eqMhuesVX(rjCLd+p$d5)(^G|Oob+;snc4^7V&KBC8J*NFv(S_R~ao2GzFrSNqRo?0Qp5Eztp~vw! z`f#01m&<&~NpQz^wlQ<1f{083*4TW!AtZLX9J_d|QFoO=!hnoTr!`}!KzxF z@tMYdtVS90=)=$lES3FfDcJ+MJ)u*bXQ!lBJkJTJ&;^WtTkfxh94!(#1$WhW=FyYA z*gSot^f9LZ-TfivXreH8z;Ir+kmGQtQwK(cTZzYDx1Pil`)Jx9+oulEUmRW3g19P( zEq-EMB&)&5ob6hiE{0oqVj?A+>a_Q;$J5S#`y2Ld%NS&9;(1|EIn%BF{C0I$^g7!M zZsXv@%2u>d0xx!OXJ@1#)9d>J$O|Z2B=7cs>j+tfU>I5NU+n|-)bl}`)7IiX?j?LV z9OE2F;K3*-Bm<3VRuwr-zJE>g8&>i}*W^Y?i?6bIBH=ts?P}E>%5jf*^-Ww1d;Oh% z5EZ6P%YC9WVIg1qjT+Ql5&b+08qdtS6Ztr7G7N`3!Xd{pD*4J(TcZxXATnCe@g$-e za9}M~{(=NDA8?T8zOLnZvJl!KsZlL~l`Rsm;*BAfKIy#eI5>YBaP&SrPrb$;OH^!< zUyxou1dg)zfbyWJuNuD|zx#DIC7 z1AQBi%ix*Voavw~9`4eYh^P8GGEcm@z%~)V#IXbwJ9iOUes|Lf&?; zAG91Ta`BwQsL*unuW=|{k9hxsC5oszA9?6c_9cid$bY^Olbq3?S-0X z9V<7Ng~@n&gosIep|l6U=^Hn389KS^vaM z5_tT7vG-qbO?F%RDEwGZ1Q9HNf)qu1vrwc9h%^xdLJ3VoIsxgOJah#CL5g(gC4`<( zLlgw0cLE6nkX}O%5R&YF@T~X!e&0S=Yq`I({rmnWIlJ$BX3lF|;~Ha*Ip%uqh}cO= ziKy1+1a9rRrcI;n>XX+PoA?(AtdvCwxkNI zHzuw>3Jt;j^(Xt<)tUcfIjzr9m5t9nmmg@Km2^L0OOkO<@_g)nCc|>RL>HEf2%6OF z{680vRAL7EXC44^f}hVrZ0_)>qTsp7DBY{uD9v+iu;M! znFV!!!ip=Hldnd9u^4Jnr@qGoUHGCc@0kKm^6i(+%MTzEHO>mcMEMFWIq8OuQr?B? z0Oxu*YNeOHTJ`R`@kYH0t+(l8Ia4WzwJgupXOMBGK0l60Jrf|Z`?|+RtIk*3=iPy> zj?4XK)ed70CV}dgR~lX*wgXzDjO|5!W3FFQVhj2?H=Ngh`30^Tw`KOc?d8X0Ise0n zf$E_CSF%;bnWa0;dp+oTlz<|*}TkehlwZ)LoNn~J#cK~)m72H^t#>iM&UG!x9FH- z*~NuQI7cmiz(ngpShJQvg}D`OeesT{21OOESR+&fRi^`b*mnCQd$WSkYv~T)SqE^NSB9Uj@qnCgYxw96VF-1n^h$*L@PYY4UU(8Yr?DHu-X&$ono^&(XYf&TSKwc zZj1DAvkI*6u3gi(m&|F(tJNYs!s_mp%pQVMzdl=mA#svZf52S3C+y0XP404JOW~Ba z%LS>$#J5=P1CojS{)d%Or<0dJOHE~QqA$o14>I>u`>;Nr}#!fH#c^ zV!a`MhwEMM*B!5HgTK-r1!|hFZg2Fz%T_ zC-#p)YtM1OJCoFxbJoV8Tof7SEEtsOjg@-ryzwinXoQ~tl&{6toDM3}qu1UpXP9)v zeq;_CW*9c1WtxZ%q{!W%MFToOa+yZ8Ts9{iERmj;}gz^rjP|7%P&42ILFU zv5$SbHFtmBM1W*cR|OHlPJg241^;2-#H%dTU;t1x6pog@Mku6;a&o1VOP2@2~N2S`POSwC|j21tB=Ec$g}Q6PMxv2R+HOIOR+<2ws~yzGCduk>TgydPm0Q`8U6RuT33? zo0tN>1L@oxKH|%k!!<0ywR2ipgQ4_U@cG9Z$HC?Yyxt;YX#{8D;pVmv^OK}mh}Q%~ zn`tz-`3qXww?kOO*FTiQXc7%V-EtJk;uC+c=k|;APxThtKh;|_ip6IR#(jXUPWz!{ zE1y)+{^a#+p3R4u2@s`rM=kDuoIZHSX9JpDLrI!AAInjrRZJannhmlWtH9v4F6_Bj z)rpSKo`Z4tE<8$=TlQp)GCsj7nRs@}91U~?Q!APIy!K@slwL1;V%@z^HD|Jm2CA=b zw~f4t|7yY4YnXJoDOSgVY!&pgTk3>zu6=TxhJ+0Ctg2mkd&ff|te z+|@)}AmAEmyfPso+2fAfns38)EA3QHKB_82hr&nDDUZF)!@H6*0eQ069JAg%=;OIM zAk4XTgl4?>wU*z(h7Ez6MycWE>hc~AN!G4i12h**hSEnHAcnu8csWF?*5A+YDHi*0 zHE6OsObP{*KEse3;sDowHDnwIOU^a$3C{-)jSO9jJ@OB>bZQe zU+WKw!e?i%|5K)w@K2c*wbH@Xv}#98_Pgxa#XEXV))^n-zSv97WH!rxfhFB zh*xEQHPVSx4%M?3MNqWGp0(F<%Ys(D;6EFvZ8P+0L%Tq)$WqdkYS^cySaFLU&gFsu4vyQ~>Q?AQ$ zyQ@XnJ+D<4laKd*Y%rcjQ)hC965-Y~!Kz1&m$Mz~zFbiXA7F$`=iAfzE@!qM4(6)3 zOgQMx1-%Z@X^W&!u1?v20vYAA9eHnt4Y$;G!~z#{vN$AkFI~O;Hk;FbfjiqLP62zd%rEs@lF0w) zFy;zK9eEtH7X6B@sHc-d_}a_)Mf$MbuktFj8n2tJyAw->H$R#mR)R|(g2@-Ihp{~I z3Xwe+eR3#K`&JtY94pXB(s0{RVg7Yp_3j)J@fEX5l_S!0$}*&(4fI`)LE&E|G3`Ak zZ|p~AvS7V`>%?5Jqw8A=g=2NKG=|OAi`^Uxjq5Jek!{N#C3%_^DCV%#YdlP-td3-8;t`FXOzI|Ui;J^qKxm9G4Y%t!ZED?VHb@mL)xO3^O~(E3@VF7r~5&c6MgSw|dL zv0tvJcKX%6VdOQlg+;@|OFrh=GJ3HJq^A}BFV5dPONOk;f-s(+rQEUGS(=h=X*#2m zw%Oc&DM_NdD07o`gP$}WM@8{`)CK3VqI*nj{LO=mrS3d_hZ75}>YqO(?vF^906{L# zpkAB6VW^|3x7w*<@qUOC;W0JLefd>G;V|}_UROd9#(t(bU5u6VE%8-mk={xJcPi&- zmYly6j5kxmb+Wp`daQH|9+e=Y>XX^6$QPJ@4bvKXM9iqSUU9#WA8@Q`_C@@^r>?#$ z3jCAi5&9?1!$A3GkANyKQV2MtHoG1A>F<$zwnK{FwBAibj9~0N??ZJHvBMQ)&O7n} ziJKA>k1s{}JI(hlD>HPLX``ViRg zzbKJ}Gfo^#3djvsDpfSTM*8oi;@2F1{sD>K|AGXE#66r#?amzwdnE%6dzmf4-=@8H z#u~9~G7F!Tuf=PMB7U+e`25-{6c~DlVV8HROcXZhkYRF}pNqtFv+iMQBin+%Y%qiY z>$gGnhqG;mq!CDNex`);)B}!yqxgW`uA*v>)yum?yx^u0aebmS-1=l+Jweic#oWar zmoxEFsYrcg2Gqu5j%vAQWypUM@#B$%LD|u(wW;?Ix!k10+S~*aU4Cj<$II7-c}sM; zGpT9f6-ilgjcCm2Qu?AUzR=Ihq-bk||MXOo@;-yA#`%8#P5lt*rCU#uOveP07Ml0^ z>m$Z@IM}!4{;F59`yVyJs{t{pl+3~Us970lzYG`p_U{7dtSs$8|hN}`!&j0ZnHt` zLr(Cg{p9s>q`KI*+;fR>3P7PqcE4iDg=bVbr+}JPy$WAR{6NPq=8j)~PX?}5Lf?hx z%{t@UK{y61{~oBBC##aYXTw?~_Bz4(rH%pm^>T6P%A zKJM@a$k;1faw}}uzmZH@%9W2~^ZMGGEvZ?Hj#v1We@~un?>ObTVsVagWY2d+We^_I zXInGY)^`>ekmj&nDI(KqL^%bfXpxqX3_D%r}FoVcg)vqqfB< z_ko?P$Ui71{l6cupl~GP*V-FjL|8)7o?7a2{++n8a|+0o_H+{UTp!)U=fYJoo_+^9PVjPN&>0mMT@rRnBQ)Y%>RADUd%bFtpNRhQaV;%Q|F2@^Hy6o z5S|h)FqU+fdYLR_?0{FjqO|&5w8e<0Zmg*%OBr%AhE-U?Xdwa)_Df>b)VCn8x;l2-J#GI zOxNZbBGqn%IZUj7ob$yY@Z7fj8SrksmH@H=)@^?Jq%hqD-)c{xQwRLY{4UzjeM1<( zMGP~2pAvmD?SFd#kbDmbMcJt%8m9{W-sdIsh$PojxV0!DL0oD4vrZI~$!p^q-7U2` zCAB!#O&I|im1re6pHkF2egXY_p-F3AiX@q_$XxC9wOLAkwUyo727|$dHvJhmDfcDM z{)}aBv%Q9`rbL5U7IyKK#lTnWf8b1BnNoP7#C?leiF@U%L5Pmm-RVaBX9`C(X%E9U z#oa0l-tQV(MeSd=?LU3tyy;nV_v!oZfYnrakggG{B&JZM2VU(o=~52Dep4hC81$`d z?wsDsy3EmktO~Sn_ng46S8yh>{iI?x7r@lv-V2Ysh&%ZnZrgh874um>0A06wf`PP=|q0CWTnaHz}H)cI~y7ct%5S^vB!Lym1&bUrW_Fm!))?-QYnsk)}p z?EOy~B?|16mwGdofef!Qr61MX+xgCSXz3yqZ=#Y1%s(-OIg~;P2IasrqE4=`d3g=Q z5s;9uJ^_!hZ2@#wM;S|$$rWUJvnoei`@(DyFeX>$Sl+3HYFa-1gMK>icdsynX=J)n zbdh3zXYGx!>NblS8>pDePzT^{^%*Y%v7T%DE8oJ<|`X1CBYIr>DJV?rmaVbm_|q*jd{v>BY2v zd6%>w8p@BZ7!>5A-g~`MZE`7;hApKUi2yx%{_VA7wn>m405SQ_06=t_T60= zVnBs`7>j*-wqX#YYCbFxCV};L`Ji87+9ii4*I4&3t(FG!4~&ef&?-&4Uonb1US)<1;g&5T`_XcRUcqX5H0EdoVQ#x)_Ka@`r0DSiI6isxRtnNO= zG5yMr+MB|7uTGht94HJAi{56h9_(M#N^ANar3HU)5PFhQPa|)jvNE86`t`MH-umZ< z__x74d*sKIWN!WHJWW$E_Y^xYPg@>0nSAVRv3rd3I)}9@P z?e1L~ypc)6J!R}wTCd5V;kOJ9drQcC#kRu!&e)+BsPfDOxVhY>eZqA?YHP--vc=4p zsRlRoy(Pm&cipPDzOZ96CLyEZoZPDN?k|8|bgg|&I;$|Xc+lzd26C`}>eO9Yq8TJ< z%x$rM@w|a`cgCuD9XNqjBAkb`_n_f;0J$ASaro~gF#__&i@ao`%Gt}S6D3U$J8R1d z%QoyQw{qg1Z|onKaML$c|}`c`^~Uwy4W!E)|d{a zymB*?cD!`EASLU%Ov;aVrcfQPPb8+OC3*Ybw<$8U^+%g`4<+7zmY74}3%HvKou1Pw8UJ0WI$+N7?c_(_9;vF^z+kSM^&k8j?SVs|HWnhnz-APiF-Ttfk&I96hTLR#v}~9 zakyIcR_{j$)xvIn4~so_R_r58*Rf>nfHcMF`g%_gax|V}Pd=g9%Pm7Sr%7ejrnT1v z8E#B18C}fTklQ#adYQUr+S~Qw4)oByr%$rcD`Y1uavW+>!T1$Bwx)ek2}(U3yUziU z)rYQ4x(y%@1vSxfco+?^tesr%Eon zFM7Gghd`(R&QB8h((#!dtA2U@>$IX}To+D|8*DEUIw=MK4-V`v$_l~)7Itx*s<9ix z5LDkVFeYnb?`5?jALj6T@P*xsOOP_|1vm&Zz11&Oly;Tv@x$8nMAJ_A>)K-j%U&fN zrikjD@o(pUbRpvnTfW=bn;7q3jE1$hAZy9ujt0liFX*X6Ww2*-JpO~|=86y{y77F) zu2_7LZN!10o9ueHH4TAIjdz{Ut#fN<$-2Ghx{<)G$wW@PrBKK8-m#V&8=EOp@IFTg zT@AIH3NF2coP31-Zfa0&DojPQzZsyx9E_5Tdbp);pd~ z1)2bJL^)FPMBm~EK95=j8)$vpsrZ{Jg-O`_zJG1tY^rT?a)KX}BPO0|=+NP|{7I2qu+>?GoTtteitO;YPa)4!JEuP`IO8}oST?5+oh#Be!Za9S`~I{4XgJ#O76{RVUw^rTx(>jf>6qiUhx{+Dc2A+I%j`<9E~lv=k5^{P^!H&KvrVB#l0kq!0e}!uN^i zFVnLn{J_N|Wi}nEMUHDd0Ar_A8#=A(j)N~;UM%A7c{kl5Tpif?swaR-v; z#SgbIqI)eC3)q%-J0k@~@k*{7-4#6szdHI%9+)-myD4s#y1IA^@n)&bq+6|iZ}~4F zsv3kcvAWB>6z~nkpMlTbqlUcd3I!s6UWL-G=KBmCU7)w$pU+=yL&F1$B0{{>%SYV0 z?TePwcB;ya>t36)iM4{f0rQP+L47rP&2i+np~mh^-J=ytU0X;zk8jh}JCD^i-38Q% ze&XCF<3j4=9C6G+#*Io50Z*|A&$*D}{5K?7U4)wyzhU75*z2j6Mc~F{@NQjy#Aksd zQQ=J!Gg3Cfdpid1G%YgFd|2+^5Qe?!=V8y7al`fa^QD6}55|mFH7eocHKO zxg{L4CL3L{S;@JYd2uFkwP~^tQF(2=oQH~bLC!q#9_cLJ2chcXTf=vMdHrP-gv&h>>X%Jq{YooelN`U)Z!eEyh$FffUC2s!S_25h^^z)nvABte0PI)28 zMSU=n`E`_9iRi7GzU!T~H+}yOan+;Fa<4};0WB`TJK1Fc1g&h&S+%ndlM)7 zNRba{%?As6V1^nEPc%<|Q7+j_SW${@r7`bIH_g~^Z8`3@I}TiAkrahe zPwqUi?&V8;!V98WWOMf;HV>hrCo}re>q>AcN;AwJ73Uk~IE5TGLcf?%4y|F!Jkt2cIvm9!T(IkDWIE$aRAfYh)f zi@!@;?u`X6_fP9S6Wq$Df8F?54svWuQXg>1GA(KpFb?KdXOg3r(e~Y2NZ3ceefac{ z<0&Tr&+&Pnus_+CHs#mxvcVNVSas>EtZ?rb-wNaK0?&1ct`;I;@zV{JDQE9>m_ui( zoXL1`r!B6|y2ncr`8N=A{L=ZSJK`rgk_}Z;(IV)7V6>iPsFfyGf6uw*Au}+5XXS}J zAiUSEcy%3F1Fs_7HSzvMp)BsYthl?AH7}-Z2m5)@4U^@Y)X+xV2Nwq1zR$*Qk%mZX zJ^AlGaxq3$msDi-sG(f9Ih%<;rFp%C(#hy0d1VozP<;7K+>>`*8OD!oWZf)iTT;B~ zC!w3mWr*MQen?Hfe|EOXZcMhf+$5wcb>gllwe?uz2I)=b?kt;GZV(ETPL~e<6J(I6tMiK|9(fo8XU=4e19r3P|r8&zfrvEf4yf6Ykb=Q={Lf5E%O)b zfz(O}g|#QgmOGDEOBqc`*t!9#mvFoK))&Fe4U04&7XaMDm09}4XNtMn;lNMon?fvx z&#bz*wsBXNxvbKfFm(En`BD!S;lG&d3@un$rP^40Fdu%d%{ni_2VK`_N+G+-#*PVW}J|9@$SwPM9 zWlGl3o4CvszD8#`O>#iY{{CrDID#)N>vRH`o&G5CX9o*AnYiTm#48QgI10K9$vKaH zYdh+)=yB;W?3x3Q8%^Z6wLf-K$)h#<`W+eU{Rv*Nf5nz&-%~c=VIs$K&|h}0l4i00 zvq2wA{dU@9Zz}nV)7W4X9c`Lf(CV>U)}@G88=qL_F)7tjLn+M{pHeqMRJ=~5#P@B2 zGaHu;y)7R(#vMk(xiNWjw5*iqe?33ZTr@ty_oYsp#aJ6_s@PovlemT-D&lQQTo%1w z1@&xte|HMQ<2%ueIaaaX0+*_#_CwK0z5}QC)dT` zjakjyvP|6+>2TKgmB3gk1>#zNOkH(i#E-N!e|&-3T>i{!7x0q3)$pjxJ7PrN77+F< z4&di7Qf6r+*it0+S3Rgx#075|=RM_-NxR{Ta?Qu3)hU*_?@rSzL}!9=q<5H!Zw- zf87J1-A++S;MPcvJlD`z@G0$~nc$E75zb$$W9z2k znQvLV)+P_qbd2V2f|d`U@Po!mFLCBt>ftgIq`Eo#QO-LQ4Hvzg^{t8!?f2%jf7XP# z86{zf4I^6SZc!w$MFZ|=eV_KZ@%tW*lb!Rj{B%mjf6v9_b#2LfW)qd}tx8c+7q8?)>r6;>1qCM7& zrJf?P-p>T>go&Vn4b&Qd=~pAfmhd1rkIfHzf+WV%4o!%=76ysX;wEgTe~QMw3T3#h z(}9#2kwK&bdgIe|Tscc3-dJ{>*3ru1Dp;4BrRh3JXsag9hvQ9yU zUKz{5t9f8*@~lsQ)p%u-nIHs=>5i?jN-xw^Qb0=T`XPKUDt>pxHf|no#wjemr4tt) zf3KI@R<268+q3UJ;hh&2f8;G(d)Tv^)ITAwN>8`CsAomq-e55|HE9|BuH%eYigM)Z zDJ2P+!H22Zy4QTyMh~L~@%Rz`np$ZLcF?zK&jwG!3g@6rqgc&-#o__6Pu6qXO%{nl z25QR5RA!4M+ z)sZu#xlK`S9ymgQdhBJJ`i01t_*NEl^Ja%5v58);uG=io;_gQI6UeqT>)9r@TEyZ@Fsw-$x? z{3S4@h2)j9FhZjjf4(noQl_R@#gGqWK8NMgfE45Wq@os#I zBfm6a3=MZ$IQ;Qn`~AIlx_F-|=WrD_HLlVTK_vO-e$q<&IML({PH2P8DAKOFAUfSB z8eKV#E!K2;AM;?VN(X`t*zS7QRVX~@JgOnZ z^tRFxG{9u{m;KML{353B&Iwk~ne%=B(i){>ZuGnCF?T`DCHZE{%^fV;qaOg#wQuLD zqHKvKL){^~e|Uw-nQ;EBpcNaJFN1kh0p^D3p6JK@A#a)7cFaVHH1)rvZl{RV^W4#WKj%z0$gO&p9eO_u4Z zF_Ulp-X7Rd;vhYXgw{nFxFapf#;FQS2sSwe=3ZN!U{Vt8NoLNyY*rcA)lHH3b?#c% zS@oHee@jh&nddT&G*1VL9nw|_Xc!Pn%pg5VbGK-_jf#Jr4`=eIdYA|k+l!bfFM%&M z=|BQ~tq|r-#LM7~5NQ$)QHL#ep;gD`OPDisn=G4ko)MW0vNM}?0qAo$I9L$w)?8(J z!ODxsJYLxkoi>R)pNh4mzLe8qa#RzLpS0<=f3l1)M+#!)ejeK9Any4M)O=q)IAPnn zJl#y3Y)<@o6sqIBZ^tXT&%u*k3$>spik#qG2(U*kKi=Lc0j!jDo2b%>0t*`6FmxXt z|Na|q%DI^)xUP9nx~PmdZi;K(Ke;SocZdJ<_WJ1r!GGoot%z51q54XT^Lq1K-<2Sj ze`XB>adDx^-6*axd|BU24^X)KJPhFvr)}zQChq*#7Uw0r4E)pwJuVQ_IePyTn;6ln z#PT#hoS)O-e&4CeTcp&b&Jw9=$H+{s@Vg=Fx{j%oAT7b3pW0e@O~_}C!ywSMI3mdD z`N`{PbVuJ9#};(>Ba&vsoroQCP{Fiwe{@#`8rdzS8b&n2L6Vox)4}B%hVTVB)p~;b zo`>qTWdO;uPsR?!;bwYrae?6_w zRci0|Ls4}*6$xxI-s;&Y5%sg3%HCeICMgM214HK@&>wEGcv;oX!||1W#{d%S(c7(+ zX8z~kZy2G+w8_6Z=ZiGN>1JT|*oRK6CY3bLwUM%;fJDFjt##$#2PVue4BblBZ2Na# zu*9EdW7_sNsGyjPjc?kelhd~If56Zg*5y(T&xWni!IS6Z%;X@%vX{2cvo?>&{T-}{ z%nPR<%Y88iD}mQ&zZo6egFIN`I5eyMHt7PqQld_jE{EpSVQGm*V@2gGZdYSJTJ*Y& z&5MluoodMaD}7!n--e-^mSEgoWFEYZ!FnI0(E+arX%{=$UQ`(RJLrOVG|G*l95Pe~#i}Y@wTJ1KS;ieHhc&zc~|m6E1O#=rXJcfcU*z;Me!! z%$o4OZi+Tfz^oi{ltfT70Y%Wk2>8yoY9GpjJ@4JJpX8TpzCTs)jbRTQW zc4`36^#Hjqb<$plXc{;9v7O#Wuv%pCozEOnFV*$!WGkuuTt^PZ1>Ns~F9CMtcc6OLi_3J5 z27{liDu*-3w>KK_cTN@gaO`gZ-?r*k8)tc(Crg9mPDj_vf0J?*o~~?)7V!MwOawSp zk9&LJkhsHsOU4AD!7**jb6`@r3=EMG*kPww{NXZrAli#uNl?pw?^w2Hl};g}y&c1z zU1YW-@9nT^Da~5Yq70Y3C-ivDVXVoLi&tHW&ckzPTj0MIuf`&t<#Uc!pzR%2cgME+ zD(v0|8O?pMe*!-*!C!N4vAFLj=`{WdpdY3oH>Qu(1JWoRU$}tn+z-)7ePW%^DNfCJ zR}Y!S`vrMwAky$HLTp&V1zYLBJO|s=Qr?f7^n?Y9iDJQ3lRHNK(2VzWx=> zLiAL{Ze%;l&pQVV8SY-b4NIjbTc8Bm&pyGqg>-9ntTk)u*ZOFc4#qh3Bc3sqKKPQ? z{5d0!Ijph3#L@6d2Qr-zGTyzetjc{w$#=}9)IdORJW6obsKFuAX`;E^`0Gc>-*ieU zjpc~If9EtFpU9_@k ze|&THYF3#dUmjRb@GoL{(%+4wk&dlOgJ)G5d`-16=s=c6YWQ8vEe|b9cpo5fz>13M z0lKxzB+zdEq_H)s=u+3S?RvbK;=8>)H)z5%lB})8V#;Z2$IzybK&vD-=>4}*d#XGh z|KT>aGSC-b8O6-cCkI0$#u#)i-Dao@b;y9wRy)*cd2QK3~59{)CU)Me2mk%x6Pnyvji;^(Uoi4RB;e7BzB zLA$|TrsAc^DF%k@S3LYWySV<_yx*1iC(cYS*UpwV$Ows73jQKREe`&1x;De=}nmT=8hRcC@jZVlf>nIe~e`$Kow)9SzwmBMp|m zIGhRLNa>C>Ycd7yo*eeU?)M=rix)z4M3KkAWg{AT-iJ;_0Faxi>;^}Q&=>rRAf7Qu z1~)Wa{JlHH*fNS0i&I7e5hpW)Ilk45R@DwB-^+|2NA|`77`h=qrV_xrf0e#3zBh0O zH$aluW7#?&Rl`Y?W}X%;!Wnv5*wo!G=TsLcm^`^HuNuw6tO<1&2wytM-ar%U;4J}~ zZkcJyTTM`RJ3wv@2j<7nr5tv!^XCncf5mf$rX{v-fstk1U&8}~%-~8wxpqnDMwkne zLn`y?6Csic?jDC$Vku*fe+DKwrg@s%H^B9~e>u&h96C!R4qj4o-5yM$qpVr=<_SVv z$)IAI&#a{nje<|gBHjn# z*{UBV;-%sYRf_8(;^M9SKCK?;8HppzC2NyT(btmxyW=(gPQf>MfB%aJilOYHHJi`d zm^R=88)6~US))8-5_Z#wX%am9%?Vp=M#k{uu71-yIJh6ar(XDp%mn#RI}k{q|Mg-z zg|WBlH^X7*mOXl0q{um&P4{{rls0*OWFfbu$2O^^|)E*{P=VPpSI_ zy(91SS&8w;G}1)&e{fY`r^qT{m!YfZM_7~CHwN61Wjs8MmDoT_YwCSqtuB#{)VW|X zU)XU~=~zXz~#ov!&qO?J$D)M)qsQjk87tW?1e->_p(hOVODwr`m<#%Zz ze0(=Dn*yNfN$KUdUm6oS#UZc?vC^RbJ#$hMZur~m`Z)j#9`rAa9Jg-g18+1RZ zCned-Ps@*sV#Od1L28&g${ExyY2;?t(;K<9v4k*N8FJ*Z75z!06fQByYoG`v&&HYY zt&DF$58^=Qf9j-%Z6~VSQ}%uBKim@dZ0L;K0g|1S2IyDugdauN&t*XSwYjFoNb|T_ zF1w%3krNcXE|bctNHhfcZ>KdP=&u5_|B{&X;>?Xmk7 zeR*5Vxc7&r)%9Bni`*7jTETd3C$0z8#mN{ts8G7f=eVAv%|90TaX4g4XjECX2bW4iApcfx9Y@Jrr@0bVG$!Jx&Cw z)Kv3sAyE6!yGMJ8yeSC*}z3AKUC&JD&F z9=6xYWu4_kAizZcg9G|otf`N=XI%_A+dq3{IplsI#|HnM+P5$7b2W4W&2KNC<(!M* z*Y@cl*u9g%*CGtAUxJ)ndp3T4NguPX;PWNlm~N!OZ7 z%ngEjL%?#4@PrLN+4v4r4Kn45-?S|%%9ws?h}$F3S?7^zvdiJ{4>eNce>1_Q&P*ypa;Zu$_MLj=IO1*gC?P`(I@fJ= zH%YhI&XQR18G4Pi;_9i+c=^nO>Dor!JJ9lXTxi=B!tfhBaqY$;`)}BQj3(y3*Z8;( z_5)nOB~a|9=#=%3Z`hMmgzK}LBiasp>)%@f-3|Z;W#Ro3-e9;sse*3Hy?}6;q6>t&TJH5kmZqA6dZOk4MVen?_tZ;nf3-Kv?E>^K%o0&NYM_IxDPG*Y(L=!xAU#8rbFYY06#!_K>rxU){O}_^jeI8cw8h3jVYLMl> z(7(0S1nRuL`NYTaf4L^_%tVQ`5lq-|M(POR+F#??1Tp>Gd1%drCdHO%`Ks_?oKKf6 zx_YG)>Y3cev7)-HZEA{q6){-+vAgt@d8RQnPSQ8a<(7$ypZqV#Fz$a!L2jf6ZlXPyjtTe;8cRFeAT-U$6)=DIh~V zT+$La6G3mN*c>vGQz@G+k z#`dHwUyRgENLQBI8~G>03n$cA$02M*EHEMVo5j9yFIM@ z2=fhhe}#i?ojTvu1eF7z!`7z2e!b1~3*Y6oE9D{a2p?d7hTo=*H(&_vjJX%NPq6Ov z4maO}IqF{nEz4C6SSmhmf^WpV;3slfo#UkC+Ztj>^ELS(1jbHbdA1%&HRWAVnzUs( z>gTQYCI@+Mj32s01C4H1fudhLR&4v2{QU58e@@x34!KSt8M#le6<*h{uk4ZUbOYV1 zS4Val%$Ln^FL$*A5q+#x@}Y)Wy-Vl=U46{zWDC5wWD^1kW3VbCP6p+sK8H|3JD(Xx ze`m|4%PzQt{3ugO)MX}fmfmhHC|dfJbFVR1L<;ft0J!`*XX8%wyH$?sp*rJUuAkv% zf3>qo(`{wLU&A#JZeNF5ZhC#feh{;r+y(3=8)(#gc5tB+29{!2&1VITtb0rb{%hz# z3Hcl-9Q5l`c;c&}*2^SIr12}VO|X9nu;6jIOd>c?c3`s(KZ$l5=r~?BNDc{(Rf<1U zWDrQuVDaC4N3V1Tf06EHAT;MbMAjV-e`RYEAU$xCE-OIk)3F-EI30hyYYX~WcW9sD zXr1YX9u^p`Jg!+LFDU7aSNE4LjE=w^m@fJRecyhd$cKppd`l}F0}K5xTlq~RY1g(E|juF=txybddC=g;Th{WFcm-NQ@g5ZN?jl zWfKOM8F8gZ);}|0G1$XOZwzVNe?r3?K^ML4Ie)P-<;p;u)D4q_bVJ6|ZG)?JfPnSO zS{U}hH;Ix$6R6rbs_URHHPK8bRe`*Q*SnXDZg-ZxhMG6M9=87qslU|ipf6lO27K^) z1;uTNQTnoP*x_Z6p1adGe*n5v#ZEG1 zijx5YMYGV9A&w53yY^?*4ID@D#q{0MpP^%CrWk;2JX-K*m96SnS2RuP)I zQ*u8M0-vk1wmDn6FL1Nzl5=5Z>fWei`{(UVW+ccx088IRQ`n&OwNlhQY&Me>)q zz8H9FdidZOns|L?Hb+~6ZpguX>S`RpnFg|zf?gJ=Ji#iQSYV(K)vS{js%7&Ap}1D8vue!8=i zb*mTb7R;x1XS}<3&EfBja8lcsRL^%X7_q5Jucv&c1t5OC;_>&mue-W*$h@*nT8R`vD{q0AJ&$ER|2OL#?N`I_L@;(CVHom)FqTAw;5K8pN$*1m)1sIBV8+5RS zomhyUoSfF#+5S6bCLJ5i_U7QL2J8^wt=L6*^b)Qmg{G?WqOo{Km8!#St^ zf3X_EizBmZEhNx0tQd~H!%Lm{pr=5E2%$2K0e%N05L-3BFlDRET^^%RjzjQ0`+9)5uXY7}FAkYu&ZqF8M8celu&UR@q}-khNn zugrfPh2nL5+m+;>9sf{DaCj%wzf37?tbE?WBZzCxjZ1e(kh-kq#z zm>n&Vs_yXt<=Ap@EApiy)$xc8yn;2Mf4bTbv*F8Sz|cKoFtEfdsH@|k&%X>xTX}V6*$h$G7PI3SA!=jnksn!F^>GiIkrS7#9T6#8E)_x^X(7pcxpWH3*B`u`O* zM$$VKT-1z6z(lW|V)0ul_wYWebAR}Pvq^(UI=Mq4uLWog7icb%YM3{qxPs38KSPb& zv;^^gdjb3(1pU7le?c$LjzkIlsW5oJcXo`a{=3b^ob#MJA~(XK45NJi)csXmq*iJl z{s?AMEIxv^+S~}EnSb-Au;o-dcXoD)N`f-#zdm|XbXBu%@1LHA`oK*Yb@Hpzb;|c$ z7$v??MA>E9KOOazvm=@eB9(W}61vUNl;ui#%l@adDnuD|f0f5?3snYzAV_8E*~u!S zKZG@dn$F*6XJdl<{*N>GKc@W0rrhZ{H~S-WcTZH)dTdRB+X5@ATcS(Vl_iXEmE*zQ z{X>H>{QJzns+K8>xOvkP1f*92ksPA~@I&$l-djxHh1^8`~MYs$4`R=U_ z7et>-zDi{Df4Y_xxFhcCJ}6V|GSf0>De3V3<5l(}2RkXK>|}#`%9QJXa`l@Zac;_W zvBL<*$XYi$)xf(=_>H0U&B(!^QiGBQMOkT2YBPp;X~9S8Ud1UfAym8S4b|4VjnQI{ zz`gY!w^)QX(>4Qo&?=_<|}k zxG`2_tu z;O&k!bJ^g|4wM(~nrqorvS?^*Pf|Bmz=}4@y{nf!JsR11_}HeGZ_idc>6ArB=1v4) zp%iQ!J*e6+4`&~4GK?E^)5vlfd-m!8of1|)f0Nu-@!8NKK!CeR?WQ|DyYz#xeuUbE z7bep?9$P+{vd1fd_P-}WE)V56(x*yqLS}pTf$|lH4Ucz2@A0FTo@dDVtoZK^T6I~( zF}*C-Aqm<0Yth;dm6a^?rp`YGaE_~veD`HvJPpaaI zf8<4;d-9fMLat}aOv2n}>59F(*52Uh%~cYs^vW&ShJ#qbyHVcxX{H{gfiF&z)rD^y z3-A92jsh~ComaDspipG2{UhpwIo0f6a`du1;{qaCukj&!?*>^#HDfm>=Tox^myA@B z8&f1aI};Va^7g1D2m9irZ)exC14{K1e>0?p%~#%__bc)PQ@9k+=tl$T;wq1=yAumd zZ_0gr!%(V|bN;dIQev3E@8)vF7G8ophjb$OKJ`I;%gjZ^zsJmNr>|HZ@6^niMn15B`72GzarRz|o0LIw0TR&ODc9H7iWyT97f4uAo z2ErA)$bgsG;D>K(d$XPlAUa31VyG!zv6*P z)U9M;6L>d=ph2U?>v^64o=AEQe=m38tKv?-*55Kly~>}fDBu5e!Mp1K?H~fBZ?fS5 zKQe2vIQmjM90jmVmgjB!_HQn9JUtUCF33l0KCP+mr$0H35CQ%+k2~dVjs0&hcSkJ{z<8#qTlx{+fD7YICOXVkWD{d+S1qDJA(wTzc% z{TvDlI7d-6Hk#>P)6=|FUN7V1JhXeepBQMBgc|Vdw(@{X>OL=L2gxdaoUfV8#LEjA zY_@fL679&=&&i$V^7Fv4@KamgYod=}yE6E*314B;pA6rOGws1jfAG37bQ7-f?&gKt z3+cRVQ<624F6b*Y82S`v%^)xlE9i6e^0m9~@%)Y-?FT@z?=4(bT?3bANA)dUZ8GST za?FA&+eRg*FXe1-iH*IPO($A>SRbLNg=qdQ^)9NM-`JrmPvbd9ulNi$>3l6 zXuap7ZuRq>N-2Badob|qtDfwG=TG5kip3tQ!(F|r*@Y{OQMbf6sXYTK{tD5VBaxzX z&I-VVm%(*91?%2=F1-b z{y|}Ou*s1Aj9MRwmnDoYlvs3UgUjPVXi~|q)$jt-)L}Q@`AZ4nj>%s!K{B>2LPxIg z=_3et8B~-@e`jpwW53>u#wQrJ$(m)C5-%@nsk#^9QV`SB574|IVC{38{;c)Gop&?K z$~bS$fi3jms+wc&C$w9sKk8bJ=lIsf$N=(vPs%8SYy9y;v|5*`W5eFOcC__SZuVjv z08*es3ec^7=L!OQrDpg!^nY-1-JVsvBlfP|#Bi%uf7h+6-IJ?-&%BoAakf?#k7m#@ zpbw>SxG|aUGjw0K!kMkaXimrl_*0c^Yd>7va!b@)e}l*yC=X~c*$yR~BMZrSF9muf zbtiCZE`1<;#H)ZA)|%1F-5V~;4O%X1SatSha(}&7O%U-#iLYXyl*dZt5&rssCn&h( zjzwPrf1_W^rf>h<$;>2xW7Pn=z0p%^ES`2+fPAo^fOP-}sBZ$rmC{ukW5H&PFJHv6 z@FFXWxCw{2l%gyjov2mc>bdNh^1iVIJxu3h<-pVZX}mQu>!xGEc#c{6*O{F7yw0vd zuoAPOg#AK`zq2a&hXRT4{61q&*=80DWr8@-f1X76$0d$A&Hz`&nzjR&X`+_7HVjeR z3S}UI~Ai zb%bl!xs>Bnhr3G9*`i1k{IwWXa{(9Ru|}^q24NfbGWD@pvZLn67z9=C|JD-JZiA0; ze=p5ubH%{c$9s%Gfevk8Ednl{>cE^)u+|C>!|nU;bas#JrkK0??k*(P><5?)<-3mh zn{2-+J18Y0D*fS^qKi;YsyOP)aJ$ z)om!jBHwRJz-^9dS##&D>J|u&jo9@5ozK68Yocm%3QUS(%zxdcd7s3lPz^Xje||J~ zrx+fScNLY;5!AdA9KzxF(k(O;UjGJn1=@j4&)k zwHOI~*3W(Z=y<6|{dyhj@9t@Oe_4T%Lh$B!>L2P2ZkarC&Km-9E-N=o8G&$)MZ}vB zfasA5cmnPfbZ7;BZ)9XY8mauid&|@ka(XVTlCdTH)14>X-PWqD!iR2d3W+BYF^=WL zH>XC~%RghehkvDVMNSQ^^E9-WbUuIxZvYB^H*6g&w94TEx2t6Fy5Pxte+45|7N%Q% za2%KJ+NXqggUG;S?-_Q^u~I;?jbWYlT-fX-uU-xW8Gfc{>jk!~!3^CY&pBf5a%rZJ zmip(sYl<|Mdb=(|%>#V@Zr<+x3!l{%*(|atJ+qM2SzSHYCP=*0{UJ`_3XFY^qc)sM zNNoGWalroRni_SNi3`pee|N|6VyNklb)StR^|v4v;(WAvUP1FEd8ec0fvg>4BR&YO zMpToGnsM86ZaTiy0Kp5Nb$^m_U)GEs&z$NR6Z`bgd33w-?_lCLvKvwoCNL$9XWPGA z7edf@CpE;G$cCcGG7$?q>h30slbIQT!2{ooAn%~lip$w1YPp!0f4cBe6cx(kYsE~@ zfZA(1^id?U^bV3KAWMy#ksQd!t;k1vS-ifv082=2_-i6=CNOCst#-bK@bj}b_KV~t z1Q$i2)3H~Zh~!y5cI;%j>IK}+vrd0?y>7?67I&QQ%GKOHI3E(O(C8maMZ-l$xaPe& zC^7jaFWl0(xxt(7f9jp7l<+D*JtZ04#erKHcUXXqnKKaGtt#*eGoDFrVm& z`i1a5)VR?v%@X>+<(AJ6j<9Onp5YUuE2>^ghTNRCJLK=ud6mx>l~2{s5i5Hmimo}_ zq_%RHNH2GXW}9kxB9J_E8JDan`tfez0U285Z&YF8k|<~be|N{fbFqkFwZDyv;j(O= zJ@wyOKx)(T3>;}R-lu(TuVDna3Aw^t9P3}@u(YhxXvcDFHyo4#HX6*Wa`uCCF z(|AfeTF^hqe?^`2Ef3$;B8LWNmMh$N%V+RVw)xBK);zj5D4@)^DA=DYy^L~GXzhSm z((o96FECqXZEw|)5VF(PRn}wD1(u9?pDO4|HkE#C-c_3UFeZPhpoXUZ2(4Hb0NW=q z5L^*T$8P1)q2x||LKU`EinQJRwJKW!DT(#!oCEk&e}#|GG7m@T6g8rrSa%mI?1C4v zFbZ8n9hS7)ol|EQW!TCM>@A5>gKQX|l}UdA^cz*V0r<8pEjbD2&2Jn_<{ou`dGEvXIcii~>1|r8LjH)^Ufsb!~ za;j{?#`JVs>E+>xIFQ)J;jm}X2}iz~cDX`Tf4S>ekcnUsIbE3do~KdToNm$<0ZmP1 z>Jiujq^wO790a|BSS*IR0m|Kp{9j&XZMzoV|G<_V)3oKUk0rm)_c#CcEGSjdF^Z7Y z$gDe;{X9fRGey`NI(s@}-_)@|c#Db(4{5ecS}D74mcO%8T;+x*#x<;uji(1wW%pWt ze>5nWY;a7$^oC?G3GG}~+(cK-Aq^+0$)~T57D8_sDr<^STjM~%uw`?fzO2U2YTN#n zU`=U{&61pvimAJV&hzGn$7cXlp0%gAsrunHmrNDw>yGf7en($#NbQE53S?4dg9MvsN2B_|*O*mdoWD z#UoaJnx5G-ViqPz-P)+7@xUQvbR6sfxoT}>xtaqwhm6ai_b-lXHMprnZPh3_Kd#po zTnI*SKR6SoA!zT_>(Ul+vhNKii1+I&d_r`rYBh#H-tr8M74HU;wfdjtg+*jOfBA(2 zmwnEZi8I~vLi%sS`-^9Ifwn=WU;AmN(L~2%95^gQ+^eDr_q$6ygVY9M6Fe{b(#tAWfx zrVl^NJGe)DjU+<`x_!}Jy{&W-8>;&z&Pdg=q~w@>wu&#j9EbrqdeR6XEAr=g7Kvn` zQ=VI4H2s=ek-Fs5g(dl;6}@uZXpno~Sl@yT-Ni%^z~fq48IrT)p<9yU#*@G&RtdEB zzBeh>tV)YF0(_Lg1^xu-f9hb~7^lNt(24AL&6Rp1_R*}5x1rusj$R^#+U)81e#zvT zuXRMblZ8CIraK+P7fRSrU6;Z6P5?HlGL4^gqh+`shbR_%fhv;S9n5|U1Q}a@cFjv5 z`D}o*2Wp8);a8oez&N(@>Ud_B4>n`Q%4%PbU$@{@-$tGUvR{$me_;Arzn7)Du*W|# zm0<)(>4%^uh(~uJzauw)g4t|)lEl{t`e?(zxgK{kIt2w`f3(K<+a`cD4hl7n&O7_;wc$iLWW*2D(WVcGG zQ=8~$H)YGmCb6WV>3jgS-$_>_vFZJQu+SS&zLOEvT}YU_*t}hc&Nz0!$?16`auSag z=%1|I9@ZR?<7XZ%*-YN#fT-a$$iX=pHcTP4vu63%adN}`f01loV2QOu>Ja&9PePxF z(EKK<)M$EeFLqFh?a?W;a=zNvi=y(}V6>aYYRfqUKg~TGtu~VRL__6lR^~`GI&EZ2 zQUawQbZVo>$r?_}nCK6aj)z@<`DY>`?;MV3ILwpm&)Go_v3^8rvWU{ayf`)JbMF^5 zMLvr8s^=|v7>Hc7o{7G^Z*I3dPYDxUn`#_M zR#3RDr7>j14xjp(TsAvfMTlvv_^?=K6*}(!1PC5AMhKP{Q*wOKtg*?%(g5QXloeHc{xFg?2gvP16zH za!jp^DjpD9zkZliej~mPOSW>)e;)_%E6Mvoe@`1ZLEkrS(>28vlkVK`{ohx+=)&>W0@|{IL8HrO6`hF1#o7dAmn@CW`nzYxw$AC1b#K$7qRDV%${u z>E`eIn@u3X7TU%EPWVBl2>M1EE^|B1gd{+4_4f%BI&H0YxWIa&*fNJ>gGi`b<6nl9 ze}w0jF=EO@c~1-@vv_ZWbr+4f$&Wt!fNa&VizLhLSm4j94mS^**gc>ZH;*s3T~eQ> z?XM{+F9l>qi+aZ~l~p?E91>|_?^uAjB=L=%fgKVkG1PaRBN#(Q_V8ljrJPgYYr=n7 zqRjIj5iq`PpiWj(7%#1&@$K$b4(xX1e|Zc4ReSrKoHVt(JOiB*bqJk!891Ky+Qh*Y zhdn>_BJ#2rqWN|4{W!J?swW#VPhMe!(YK$t8I3yj5LzahkI{a^=aqj@Ol>!P(%tN+ zTR>t`?)Ty{;iC6qGLg{(4R_`rP#_vj$ToSS8dlu!Y2hdB&tlgLIllr2>$N*nJpue#uE`JHw6TINlG6Aar%+?iz6WDtLr~N+6LQuUT z{QqMoH0+9q~W4<90Ioa^e zummSHqZjZ$Y@f}B{SyWj#r>~qYtOTSCnT3KbW1L!TLJ`c*+j+-XDQ8VsCBTIo%@c4sr!f z+eq4dUa!h0b%SWwqwE~&G{v?w}4csK`ee%bK#fn8#i1zi8IS4kNou!DV|){ z?_c|^mwCjdzWa$_ittO_;EpBk8)0j%R))T_Sr|m!$D2|G3&)VbxaBHA-}b zeY`+%UV`lUcm*?U{c~0(9)G`P+@AA6F@2;42wKK`cL&Rb*epX7!Q^6Y!S|LDp++MB z=4u)zCYu9#@KHO^+c>>mtqnFavR!F5|6>lxOvirYU?ko=Yuc1;Q0&*z#4A|_u5!G| zlJP}1boTJaqMagNItqd(K6cxGTBlxn7>tz@)@4{j74(=|^J9Vm+J7~i849;oyfz3$ z$|&|6PNPbt6tYW#=3zO=Zaj@O(^sPBP)6}w zsaujmL)7c|az+=Vs(;@xxxc`Rp_@a3Hi8K;uidvkMk|Sys{89r8G$(Htw0USf?3_8 zUd+7yA6WhA&Of1-)PDm?=w$_@R+2kC8p(m{SiHK!%ap4IwPa)knqk^E%S}=acPEs| z2EiZp*_r<4Gbt$~1ka$`1I0a+Cd9y;4-7n+F~3^Pn0Bil9goDnWt0fv`XAIHXBL*s z%aXp7dB(@kN7B228c#92_^E%0TRVCj&1Ab3P{&HRCvDq*_kWT1qyJ~j3`3X(a8*WwdhITrw|8+dH7AjbyAbm$bHgnm*8u(UK6)v-0F?g zzDH2>)4N5UpU2uw{gPe#qDZeA@CnJ$5F}7+&_9*J-BX54)UyBUE`NA9m9k1Q+^kYJVy*b{B%>d7KIXW)M@N5j*5vW({#Iww-1G=#(oTOYEhxQ#YZImpk3qFfhb*Ail(@l}d_*2MsvJ9C zUKIgup?`Yc5f_W?u6^0=K1~v?wVP|VYeQ23>je$CTkJ4?6};a6z|MxjVUAuw{RtwD z?dkjPW+q=c;yxVJ3Bt+^A*MzpCsgGJ1ypF^33f*#O(ohi-%!p(-XH&=Jf64ZpO&kU z|Jw)>lK++w;Pe34#f$FbklmN_U8yRXT2oVa_;r(e zT0mte0(m183HS;QYF2p9i&tJHS9x>Hn)H!5I%oK}uG2k8tFGbV9)1-?&CQkY8vcnv z9bRqw4i_J9R^6p>&tLpjLcB1EJg`23bC+2rW%*}r?Y8f^n|W5a_Xlou8qcgtSF96b zMSp4rYxBwxLlM^$Py@LFta9F~Hq)m5g5hEw^DToPuuiXN@WqWE1>{P0#|1?{3D!Qf zpnDEb6ub5G_4b2LD=x1OgaLEi;_{~}H!DE7B?m>P6j;tS4JR3Fn`A; z5Y|C`a)x>Oas^`d{`ypj+2I3j#4C5_s}|qZU$hmbAB{UZ1Q+j_0TeQga}o%9R5@Z$w4A`RvQ1MYiA8>e3HtL^zrDI*u~JXJc?t)Z z5}H`$wspKX3*Y~O*=O8%heg}UJt zVX{!*F^x2|ZVTPAX|msxcQ2_XGW)m5O89`X9VbR0pj;I;sq3>MkAKqG(PD1ZOP2!` zndUvdoKq`@X)LuuYq!=PbU;8&C$f1m3#UzJI{)GC6I4#_;>H@$ggZg!tKfT^F-cRpeY0_K|Uo)T?Yax&Kd=!>#bAM}{H^6x8E5c43$7`6g zj|I0=alo5PG4jce14|V7s9JBG*J7b<8enYfPulg49M{filfz5>$_gZ@ARD;`_J=ucWfYKy1dPc9DCDFS$hYjdOboh8|llWzA*AAyl4)c zexlvz;p~Dkt@@^wh2`@IecD*BawlvUpW!-A-E_LPet&BC!enMO#%{eCWs&u!#}Kkq zGy=XS_F0hnV9TFl!rYmhEl!B;m7V)2@KemmXvR<@ReYV+|74Q+R5eM8x`6SjCAFM| z?_$!%;&RrJ2}IFePUDJ_$^dIxU1Ajs!tRL&(H?$T39ty*DKK+uhe;gfsBS`}EL0VC z2MkSJsefjMUGZ5mhDR@5^b0BMy8V9TZ<55(1hTZ=-VCr+NRzCI<*hFJswZ*!ALAN19C(ZEO zb-;k$XU%)%8#z|)VAR%YWZ-H+N`Jszxyv?KA%Du>QiHvpt6w>i3DG$rW;SikVOs6Y zazR_IQo{VyV};7zgxR*ZK<*OLA2};7BYT;CTZ^r3E)yDpW0iHh!qn|6!~hgaV3wRl zYKy62vD=1wc)5GQ>GZuXH3P_BA`MQ8pP$H>W`GR(i2wjCz5M=dG9!(uzyo8uyMCir zNPj+~9v5?H?05b{ryKI!b0Ip5lLi-abfpa(cC|< zx}-g}snZ#10?JGUFNCM0qp#Udx$cR>wIEEX18#Dw3bTCvNAbL;os+>|be+{mW{SnG z^8RB-yI0} z#8Qs;|eeRlTAIANkRd5glKpNIA4;MR$8i}s=$2vj#u}Up*kkQo^FT^mq>l@3-f?* zi_X0PcAIpkX3L><2t!6`=K_oLi&2s70z-3?_q`3J!V2V5O`F}D+E3A!*lWQQH>{f) zqSKiz*6ZwbPa|eyHuk)jnfF#^Nq=F!d*mTeSOW9YP){%Y=U|8)VD*}m(`SqR)=N2C z{@W2+1t%*7Jq|HVfC8HdHN7l;Y^tDva-m9e;UW%&EIh+ya?}WdoS1Wm0Q z6v-jcqZzLKRELM7-b7Ugmplnc%Ug|;&0Wc$0G(&&K9sl!6~DQ)q<`i2DdJwNl*6|Q z8m2-)3az~?>QrG18}<&9kT3mf??hAk#6flF?*yd)nmrX&YPtthIU@B78ZzL8f*?1T zgUsSjMFW__*vdxga_2R6hdOLf9oUS*v=#%zj>wmzN|moN(eRe0>lFe9nj#JP(_p+7 ze6o<$HfZ2?kgf)UfPddcmgflr#zT5BW>N;4SH0*Gj|8psdv!00e4K|rY?Z{ zKrvf(GeLw*dS=Ao@1qGRw_IAE&1ynOvXKd>A-rU|r1goq@|ee^#_fa-3bed%BLi7r z%Up}`MdoZ<8PE0K*}TNMzC*J>Lf6Ku)^P14j(??~az<*|!hZ?v#e%YMFY&ZPbts^~ z>0&eehQwi9*6G@lRQ!niaZ`v6aHglCdEA12zuAhgx!L$SWr0qK?&+zK9!xQYm%XKP z+`@&8EQ6_Jw0rOl~8HDVPt>Z?tKckCOKvcdSnV7;;w~z%6MY>MoYCp zcRASni^Lr!IDgYiTG!n+D`>8k-v1gAdzp>-ID$kn^=}h?0@A;tv^etnKU}>*GCqGX zr@ID0iD^B`gcSJzZ*;kN=)%8I0S;cz-#!-XY)NCH(%KCs2x33OWou z!5;*tZdiw-5cn+|BzeU%jWcm6U3O8};>z;@K;}xYzw?lu0?o9Of#Y4+uOT1jR%I%e zpIKfVzFr67l1Hm!x-X&at$hE&So`%bU04j!v_~}Nib* zd>o>KIDcGo8EA}p1}N39>Z*|4G*J#euI7;O+DyGt1Q_Skuk9PMX3li%tD6{|xs8&q zIYc{7&RgNS@9cMz5Q{-(6m$U#%5)q{+@Q-?0(~VfuMl5InJcv49B~UtlLnfD?fh$5<3<5%ipvPk(dR$E>TD7IK|07meZ8Q&zHav4D!tpXk z)c*TbI~1#_=rR29G70K6omT;gIB3i!oz`gADDq~kW%B~UbOnL1*S5d8B%^GB57#=hWIk7|)YAa#rGFA_eW;4QFDDanHL`6nW|(bGQP+ zH4|D;^=sAF;du}dqh3A~sT7$m;oJhP*ea0tDJn>DX}r}xJr=uI)r+yVa02Euh3z(lYp%Ww74i+a6uU6tEzmvU-*=_Ejl zkjj-M;*o0S(e}%u^6jUNhs^L8$C<+PT$V`+u0Q!~eB=!jY|fk3S~g9p^H6rrw0|)< zc@unI$RUpTMNWEO=|%yK1xx96PI6c&b)uNQrv|9DmS|vH5NmCi!hfdMdMTLOJ=ZYK zwz+3q6W9d3K-je6&3Zgl5uxNogRzbaP%PGY{alzDU@3t5Do4=74U;3V}BM0ZC^rP-szs*n+Gy0v%E$x$w@cDh6PHhKZYa< zRx~Hjp2s@Hu!@gjJ}Lt$*xfAq;TVt=#%ZM0&}t}5x`O}xZnLA@F_?@m%5()bTfgzJ zOfhqf$@JX{CkX>yub~U9_N_SNX~w>c2E7K|X14j%LSkZdLmLNOpIrW!zkld3qcEQl z?=@DuPVA&(DYKtvZ>dvat+Pw22~bp{a)RUPztBH)2&TxlAq~5m?8upMXY@}ncwYfv z4+MWnCItZawppM`Nw>mu(*b6NSTE3=NW=CV28udn znqr5Tx;Q!x=?Mkx6YXoIz<-J2*B6lFvZhzfxfx6R9h~-)4Ol@6K&)S;%a2HVOtz19 z7%aO%YhFqXv`cR z-z4GG`A0;Z7)mn;ra9}TkCtcJ;@F*NG#!c?XIG1su*;;*BwxN|l7HX6tMW!{6H~1u zyT^Wx*Qn)X+ZJ9zmpt*(*C7Af7jr@`{)bB$Pxq)O)NIRoBTQq}$e6f}8b{C7 zmL>%Ghbk#=B;u~iOTk8G8gg)- zqVb|(o3&pQA{f>5qrVR(*poQGMxAVaCkH75R^?`TUW8|Xou5*(F%A8m9lge0KfjaU z({Sj%*W`;5#ed?JN0l!SV`h%XbmB3WfB2|Cxn`Z*=?NsW5FN}fd;3Momj6Vemn?P!2~E6+N9};Z$q{7W@Q&)_-+!S5Ywh20scmt0GD?IBuDDU< zj9$}KB*VbAqBCG4MH;FwH` z^9a0Ab??W9wb@*V^~ozb*q)vVa^rQG;$c98c?wI|uv>1YwvjUh z=h^%d1%L2b7Em3<+E@Y`z*)qZ*PU5T-dF_ZJIFe51;VU=G`Xj zMH!M}Z>9YI6PQzU`pK06S$r;=&#%`KxJZsu(SPBX+G}j^I;}L&?1k6v$m4N&ui(E` zO_P#DTaF&@LE_|38Dy4LO~RVcBBW0{rU-0dveHdN5Ga zrhiN$Uh6Q@e%U}Q7=7TmIsC(LyTsVhI$Z)3F}+nPwbD^~UTYV;Lc)OB=*IVYM!@Y; zPFcH4`s+*ReZGr%gU<)dysEKHxVLny<8>TnfcP#wEIitj7Rr#UYtrDqRWHx9G3?FQ zs90v-YD8i4ik7tb(kC#y{9MyzcDKJXeSfAx1dQeo+h0Yg3!;o@UfPYC`dZtLwgKy~ zXbER{PR;S4x$9=JvuChE_#6F4>ph~#-H{~`hjzLlU&=bS2uC#PKxTkZcliw_y5!=7m*sDaJsJvR<1R_*3xBHs z^k`xN33Rot+VY#cdbdm0udzW39H;AtBGxg?&L+(#=m)H>LyRqqi9BmKfiOC`Ww^0s8`!TLzQY15?-B|6 zLd_{qwR){Jq|k@0Ao|%b#t;LO2GIDFfl*Fl;EIy2Wc^;5z`nf@u77^u6G!@OxZwDA z(c%x}TvS{B8?ADc4P2>c#DBq<7=;2$?~cLe`z7}FjDub~Jdv6GP6TCoj=K?l>Wb8! zRK>VqG?%^U*qF5&94tK7Mh`36!0o1vt+jqa)+cTh5e_>OVT&?k2Zi-2LQ6|&?p437 z;c1q#hKQ^sHsT>hkcmlXnI!FT)F9E##>GPOVC`h0T+0y|XxsR0eSg}DeH(05wh|kv zC1IaDNl&e`IQ=f*-#2*quAuRcBxC+Zk_l56WaqH<1=7n3jfn8hXq&jUGj2R&c3^ow z8+6RG2mO6b=H#OP1gxT4GTGSXPLtiw)QH7oudTYc zlf_^eL%*&mts>&&Nq?hWn`5Y3D_xSK{c4O(9yevAqY#rD%AgArBw*=>CIw{;zbXBo z2lZ^cUb*DzanjC62_9dV@O2Zzi#V#pQN_$=nZ0`fihNr6&$Yp~$uwks=uMc-|1qsfJ@f_@kjo69u=My>8BJr!uuWwyCCorXd^z(ZKF?9vexV+zHsh#>_!i$h*|I4!hR z#mT-`T@nVsq)v|XoO%PTp5Uw%URSP$>_uP-%IR|Ynbj5f`tsI)-ixH$wJhGSes1AX z?fQh(vz@AtCPZC%^FF$6@!L?%Giev`TehE zo24&aW`FG(Y@k+c;HB^PS6V_V#6FLn=a*n6JXezKi}QmQ%Ok*HRp)UdV0R*Xjn1!x?j*=t*b``Kze_A@! zA{_6`t{N#oFFq5XCqo>)8IWGGV(Vw#-M6Hzkbl0UbQy(4h-5GTZ$~VaJT?RyCcVYU z`yZ7`%n?_t9wE73BA6>y#*P%$xDuwYRk?C9cTPg_~Sz($I4l z68HJFN)@V;rh4xVmCNXjT~7#VK~2vPX3Tu4^R*=-$LN1hLj_6S|95#s&(R8oLTW$D^x)AmjhLf z|Lp~^oz1`CHC$p_Pos;oZFTpamPP}e5r5V{eDiMJqR_BXYH8FMw5b$k*V0fSmglZq zrJR&>P2uFZH!0{zO;(!W4`_@3LG{g8g<%|?=C1t$t(*H^uAv`! zLRzu442<&aJ-};9;SFJ;<1=?Sz<;~;Hc1|*Vj%QS5s%>K0NAH;1;yg4H?{wLCH7J0 zKf6#o56+Tt)t7Q|6;uJn9((@Dj`8|E(+@$z_3^@WQ`JN28bReu??+5En{mi&+k>?+ z@#zPm5;}n06ye-E3%QlStu!>k`FDUFwj+h<=rR40>O?4#)L6LoV3FM+X@9?qH~Sc8 z+#mjCw7gW2C@xs5Du~CEMv{wW6GFhdy6qVqF)SyCZ|%X6A$CaON&1-g_$0JxSj7Ex zf04$02j-$f$0S$2XHTt`mJbgy8u~rx92*VhGse>0ph$Fv{T=s~6ZEKn8;F@khoNRP zM<*thKl3KzRcM*=^4B_SP=BXQy}QpBo|q||nPbkO3YEhs6Cf#UH$2j#Z?H-q)UfDe z#_Ze&fR%hsk5w05-y2{y1&>>R*=9P*2LmRvdlWnK;KX4O*@AhKi^xj@oOCmlru@2v7YEJX zSfExZ>8qDq>VGO4*3ej$0@{2{^{{{ro6Lf-`u3AT{Q*)?z=UAe8T*! zbHYpN`09%XJ2k=jkCPMajZV8^cL{fyPq*^^LPLoPW<4N@Y#rDPvLn*I0tX zZ+%rD(>HCWHEj@l70};X5@U*#o&Uj^hp9cuTA+EC2DF)Vy3NlCwg3(t^7S)Y}1=? z zJGI7#bG#$axinAEW``rs%O58HA-O^LS90Tj>E4t(bAOXe8raAiutDF@6tdJ}DKl3` z?aovI-OV`dw(9mw8SGR{8aimW$;I|hY%r_L(<{sC2zP%+j6+RAO4U=F+xQL0;b1-U zMr3wS$+jmT1=tM-RaJSe*T)E3fF^0O$!sn!eV4hK-tEhgAgE%KP6G=l95&%}!4d8} z-wv0Su76YhaN0!LSNN!q6J0`x+9kZXBQWOje8 zyGT=eYqvSJmt)5sZs<(?+FPi~8}uzjr(s4E@Io1JAth6$-6D@@t4Yo!MU7SC$E! zqhrhLCjXjQd;@DHeFJNjj3>Io%W3BESe)NNS4x0Nu``>AF>Fyl^)ufo!}JkyHA~e@ zJb#Cu1TBmiuk1^QS_xLvhV;45eT_4lbhDLBs6v8*U49Q=_%C4VkGdEWk^T4Ttz-;+vmI!AaTr)zh1hc9^`)@On#a1&5@1~h z4b5U9$}AxxIQ8Y(BR!K+ZnFxMW28SIeXm#e6^K-|3 zc{ypL^yG8{qhE&Q=s~KyEipOgs+~r!LGx+Bmo@b>)WOkhgagwuz&9p5PUw(O9e=8W zflvHA#KNq8Ob#^tmhdOfuLd=CGldD%?^>sEZfMfpL<@|R9?@vgvT_P~nNceN5`UwC zL+@U<+$>iZiqeVRC3H79h3G7fzBB&!MG*S`6hXNCPZ7lEK)Kd14mnWgP!2O-L^)_G z|Dc%;sAw3DVhozZ?JZ+I8Cw_&<9{ck1cf*VrvnP)fuN%uUb6Y2?a-ArZ;=`gfC-O^ zKaa4cC=Q?PYzE7U4WS6bYs+^2|+~_ zKUWzSj@P%eM=x7Uh&p3ES$`^4i?Wvw$H?v&ukf`oxAN>3ZA+$@M##9^dTl;93x|Ar zm=dr@-ku-u`fY4DIpAz2142#tZPf?%cP;l4`(Md>Ct_5|wKbPHoW#u_-}A2rMIO$f z1yS2bDSK%|3aAJlb_g;d!8n%X z$D<~BfGMKWJH?v@BgEu<;)BzO#blF?QiGD*Y#qxm_YdfSF^^3~L$Bpp(N z-sh`Z`@Y}(z2EukoZ3~Zx{69x_nLEzIiBYkbBwvB9phcHU;~+SJ7W($ql)vw6V9$D zSjSdbvBkc?p|%JssJ?(uG`qqFV8aRslXzX|;q95x$<(pYD1W@lCx6~Gnw9!s?R=%R zqEnXTm=rEJcMgE5rnwG44P=hP?6i6eV=Iq#5m6ugQ0{SKX&q`AhDarD~izt{fvE?n?<2wtId;xT^xj9P?pM z@BMbo8osk3Mt{G8wp*tulRWOM%ubQ@k$#fU-8~G-tJ$Pj`Ex_q`}R|lT@|JJhnV~w z4bbTwL$0gMr8~sFL*Ks`-}!LNFv~E^m$mibk)Fi;5&D!a<@)tQS6uOM7!ynm9_;Pcn8U!cuN48dsE^ywyF z_v1f7tX`E@NEmI?A@P2o_?Y>A;;P!T)hc#HzGL0=Er4LyL4ExrlpLaF6MAiYM`2rN zE;a%kKM=O6VgpP{R@y)yXd2Jgkt|54VT!a#9u?46v$CTOL!{H+V@YKHRfVo^2dvBD z5^@ZNU4P;2E_7EqXB!Cg0+%mEE|=%q*Cift-0sB0J1qI$ahA|%i3}^Msg%OU8goHv6@`N4R>Di_ap? zOW}T4S7~>wo5(VZ9AnV9HrTl6^Rn%zG0dk%8h=uqeofZ6(#;QA7{p76Ur1LIIiuqC zE$%jjHN!u`@j&l-#lLr@uq6) zk5`ZuR=gt5PUofiA{!R>QJJu7t5+Cp!{B0L*~@6}kY7g*kc&x(y~U+_$cJe~+w82& zihm@u-=AYuFOv^E^Q*JH9OG7;j&Wz3yY}6>e+cu5>zZVp=y|)W&bnF7y=_0F=sEVd zftCvYA{169vb_Q>>A{0dxtNu*h_8h28qh@i_O3J#y?C^nzC$4^+_MqK9)!;_V2ABB z*Yg|WVlX`+cb`!BhdpgH&dnuSw)Wd3p?}@R4bmo|p7G0r1yD5i+UJgwB!6`9`WGSh zhDsg-k;_!WYv*5_N2GrM$|jZ_t?tWxjmun}%)aDFS`p#fXksO;tQq-~D-iq6?E3EH zqPb>hrz6k$&78qdf2G|)GokaV+=hzU4a1tfZ2uNx$t!RSNv7J>a*O!Eoy&b<(|?jT zg~HI|3_9KDNLh4$>UyZ-jCI;6p|ksAqXuuO4iVUfLHGA9(58YnBm_<$qm}!n^j13+0Gy|Ww>qFA-LSG!4UVMrJH`N{UQ{8yJUZVho>t*&{m}@JrB{V>tS+43+jjlK3SZI2b*&PlgZ5Dw?tL=R5B*)bbz=6n}w6%k-sn zFH`}Sknplib+mlFnPGc=^8L*Z)3GIIHl;=i{a@1l8(dZ3Rr}o@Q`$o0Nu6fl$D;RaBYuD3tP=I2ACjS`4DDnJ#X|x@~-2HCg zn(I!Tz!Al<_{}=}7HA|;qkqKfZ8r$x=>v_@4|4nI++TYTi7L=#b9(l2n zBQo6uQ>6}DsAjck`LzDPs>hDFf4q){45nK^VfvBhaFP?g)U1;jwuY4I3~D8~Tak{u z)19`$aY`r_mcB&RWCv_($&t8B*#<5`iacGl0`>*9xEV^*R4xT4=6_7HJ-Uwk@Aw{O z)5@zqL07;{YY)?%@OvcxdSzbKCU=d%?q9CrY(C_`ujXN#c*Y-g&h+3;_>Ia$75l0h zKF_QgVeoYOhL~zE+@$<(^AuIY19C^9HIn7qBX{}0lB02;Snu18(-HUcg9@4} z_S7w~1>9W5cI?NsqJK3#5xpRWbQPn)g#w|s(YU)g*?7}M7eYw|SV1NQ!w@vHYgw1QB<)Y>-LdMg^tlgPZbOZN3GLO0!zQByvf zU0aO_I;|XPW+i+eYwIO2egBdPaXPJ)@&_ugli{eTBu~;+i_iAAUt5ZwKT+Wkn@MLu z|7)$4b$yX~vVXSG6LIP=QCbpIrGT zDxZ4(aJdvDbT<7_3vP!9*DsMmo+&z+^jPk` zy$jvNfYmUEl#)9H@JG)We2^LP>i1;&Rfrv}i@tvy6I0W!D$nK5geuZp-Uk`%s|E%p>m z83LnW+2bMv+v>(9%g)s|HEDsUChhAn8$(qT8%9TawgBz=dDj_7uUY{`*L$sJ^6Q;j z$ws5u*?3;+4yQlGuj;s$x!>~3E+j`7B=4S~34dv6wAQnAESR@OcEA8QStsma#n0|GONZzqd8c#?+Gt6TL#i2%Yn0aB%@)_ zMc)AZ)3#r@5u;)=Y2!(9~fqd+pkP-DXu5UPY8a z0U*Br*UfbIk$)r9%zs8G`W0I>(a+^01nNpE9$S-;qf;tWxZn!9;PeU}dRJKWwvVYr z^Rr0ZDKgjJk5ue))|n4LVpStd?K zr_#K~8XcBDI)y&rQ@aFR{Kg%<;$^KCzM6k3z%P@|aK2&bj0gY-2)J8imj28mke_R1 zlq;*sK~;Krxw60}bsp+E?Yz<~)PJWDf_hku=hX5egsGdzMfzAC!n;z^>3x9U-5o-U zw>S}f&_Mmc`rb~@_D{L*Zp#6M$DNK*i)1za;D=tsWPlCDN=3WzbY@lD{LbtQVz--H zZVEfz?T5LL5tsaiw(`3MQ&>{!TE!S1@dU;hW@S^&ih)GjZu6LikB8-*+<$#h&Gr!a zdT8b9npv3Lio%3P;kynBpuE&eGA|?i%B;)}3rL@1_#Lj~nbmh-e~xe1Bp?*r!UChac&eq~yfwzu|*lVUp)&1aUt`5Q;@sitL+~1Qhe@?39)fBOpex37xi(K$!-U>8XOPyK+bsd@ zwL}om#51iCFWwS+zd;OskZ@jgnu|ylIk6Sy>f!bWujxT(Upx|(t$&PI(G6R6h}7vG zB>Zi5PD9tGMS}CHjgVPv=X)--BrL~F_4@tVPVtSuBBs|`CyMY$|n{9SsKqBXA~J~sqpWhV?R3pQ}v z@CAw(O|p{teWsPpbXy9A_`b)_?dE?cv<)f@Nb8+EbB)MqwAgI-K92E(? zSchclLxnX&(SIv-@u!7$>hw4R_yd(?^n$PcHCJK!8wm>kD-wXzRJGsotNZRuLXM&K zG%M6WwfeB~?boHh-^!#VUSgP}`NZ~P z7-3`cwXYemjOR^3`kKpLySn4!0){&Dc`v$s)_JOpUzP2@ zu8hgOLR+c8phAKl;bI=fw*(Y>a&Nb2`WLs5OqQABKv5tiR-vt^BxU0`WW`1vL0Gt@ zx<9cAUxmlm8x#hcW~K*HP=}Edz)K2tx>SbVIPT}BlBDVW!m8RzESoREQBakLz&0ML z%7ZQssek#%y$w0W-0d(8f!<84TnKe({Jfb7Ib$Vu^Pv4!fbvwUre0*1!qpwUXPG1Z z9E`hi-$@ ztKjyGgSlG$aWPuY*TR|AF9twkSnmCGi`rIes_Fd&lnJx0F*ZhJ%DSqFP1wT9HXWJz z8h_QhznXlx3_j3K=hb(hQPM`UD_k^BxS)DvIo)%!{%rbvS2%870xZ&uVBuNH>Hecm zrM#o&X%XuKxlKe1%JjILDM||B*OC};^Bo-2sx!E5L*#@)^Zi%UwjH@NVu z6<@O5{RQ#yknZ0?ohk1%*Klx2SweDw4}W6bk1kC>6}%5Ajv0qmju?ka5zkcMSHBQM zd+cXUx$pjv_(~^Gsw=O*{+GFaF#X?Lelta9)~ z;fE<6DLhiHl<29A0m`^34?pH$6qYlNxwd%wFIY`q3GTW>D6zx4u`1y zqAqW|$o<5*knu;Utiy^UIAz2&z?edNFV69O(Ig^?W)$4Pk*;JF!J0WYr9QCVp30y^pk4Flwj&ETu5#h_JQEQM*(Bv z5_Jc%G7~r5+>{K+_?fJ*n{L~y12YMuFvi%{dwqw+_0(APXt}wDM9qq$m+`qQR_1{f zbnD0A1&neglMa>2h<}6B)G+y68=DmHe7^&$ZmTSQo8|TER8;MW8zr}Zc{Y8@Lzkd=j%6jLUKle@u9ci&!S(2&6PZ^H=<=vWVkV-3pXU8KLnqkla0DtA+0 zIde$8SEGvUE~f0i?EzFu>Z#ge(O=1c_cSm$7wmfq;7h?z01k2B17 z!}jjHdWqFQzr3yyUoX*&;M>W&W#chSyLa-$-b`~C9qcO2RB2?R%4PwatStk0@};fQ zs*Mw`Y!Mt+Qs(3X-!&kLO3>BNZ)~q8j{-&^EohwaKcoA~l zmTIGIOfR*}vQIpb0S2CzIjAO$>r3B>E#H#?^tBgkuvZ=>e)U^L zkQQrUIVp<)!H4k?fuR-kl`)+L7g`fMl3>}z>K=~2D1We_ICra9iExfs(tS74&jpin zx!*k(SLz&Be?u#9ho4zAetsfwdobCxXxm}+XLJ3!YO72l)AvtTsuf_DF4>?2uW7c}f(3)BP6Qp7O-oN8PE2PRt zj9x?N)?;yp$F9#d9ojp`vy*=8Lo!mme4`~klczqw#(!od>BBOZUC^dqkrhy*Xv5ELDa%P=7?8*m>Iz!jvU3R+v^b?JIM>sYmE* zc+fm!@RF|C=J#rwn-hy9uVMxZ0){hUEG# z_f=$cu~UE0)$F`NHBjTWFKKv;^en@kt|qaS*uk>=$JV~$P;QM=wXN0zmBGdl*A6xp z@@e<|#lpscR^aifoj1kTVZ-ExIX@75`u@XR^GincP4urPQxNDcJ!}&={LMs@Zu5axgE_8C!K`)VF8`ktgf``WV`QNwa zngy4!TW1$1DG}eys$@T?1u1ulgLB7!Rk?vm0+k6{8{O@}SK(x<)fS)Hs(c_i`}c7+ zI#hpOr;5YS-h<>Z-*2Vywo~{?eYo$2EyR`y7uFax+^XlMue1O}blp0CNaw<}&v#JK zcJEGqat>W zn2|?qcu*~3l8N`ZOvhBhX1OMdAt#+Ic+zFj{@C^Kz|b}OVRsP{qB4ZIhUccK@r!@G z`;^dd$9vk{$oI%GRR+~Sj}w>RrXa;xMHQRzz};%Jyk*pGLSvLyVUbcj%`P2sTm;w= zgC~_6;5TdO>$U68P;)Cha%ZX_sq04foBL2N%jHP7HPf-AyQaL^$bf%czzdf68xh|< zMZ||~%4ugdrCO3BuW0~xjJxplu%v&XZ&-Cn!7NK5=E^h8!tVucN;gS~PZ@rxL^efN zjaPL{fRHihb$nYFtcz|+VJ6(r%eTxIvUtI=el$()JWgJ9D*+!a2m#J!@tpCeuQMua zm7AZG+4wupj9_U;j_}s_b+q1^5w=>vlrR7x`m^VDOeEty)FT$Nd9qu7#d%lE@Jy&e5{vsM2wL`)H9yO~*iUyZdv6^doY z2M>~~Wt*xqc_}T=?P_*_IqlWwEHoN&-^Icz-7!?w3=iq;d;o8obBJHVmq(-qz-qTe z_NV2ZHryL66InYb$T;;iKv;jmnI`rfuy8bH-_5d_l!p~5=oBah@zO>02r5RCWr%}&|3 zmG!dkb{z|^8vTzE`#Xx{PywZdxi7d|{YqLB7j1b;n#yKe5m>aDAoYg1mtV20<+DvJtu zjAYzKzZfdKJFO-WawmaH`u@E5{8+-L!%VC>d|gdff;Yc>?OM!|QAAiSQ!W=N?n8;U zb<<~ESrp?XODeRiErqRg+W+;7XMTui{B=~)|8i6|@SV97R@;Ac(~C7s~+f z+C7v$#*!Y6Y$<<-0Y80pv65tcBITn3NPa9``6MQRC8TA_G{;4)4p#SDh2f|4jWL6QFBh3`V_??ANM-u7vMI z4Uz;$C|2*piohdQxZMI-nuaIczS|TRON82ZbZ&pJdGoEiKGDpo`F_BhIt$%n0A}Br z@u4cNS-mNedBnUnyD5ZR_2=q`H)YN2+bCg|wcG|DIOn9j55cQs0%L8ayOdW=8832V zHkjuouQ#mfOOO+Td%XX3;li80g$vV9g$wuTLCSAi9%;0%9GQku%hoEL?R~cY;x9F% zK3#w9M1Xr#xnVvkQrKbka^t~Kt<4iv{T(L7QkH+LGAGO6HEe_aU8ldkTF7bikk}Lj%J)rLsX`ePGMX#-&NGqj}o)OiyM!{fS^1-pn8<>JXcSc+3H0Dawps zKdYp2R$^{>Pfw8i=u#X+CD-dgL9;Am?6pc+9?vQvca{tGe{=t9a zKd7^SnW*{0{bUHCqSHEJNGaZ}PN`)3+nCOcKQ_l_EYn_YE2I3N1;+YofZG$yi{}22 zW*1or3Dk5pN!=y7iYkxTT)Ify88lIAgwJy;H)mDE&@@8pV zusBVQwbo7eYaNPF7yRd8t%F}?3_=9Rk{1%o{O-53)MiLPg2Po$?7{oPk*+=H`{7jm zCq({`BfRcGVDwcMnr-L9gbPLEu&~IUJE`#ITcg~AxOJ(|RX2-ZpTxKBz)Gzq#iW>< z4egmj9{pm+gJr$lMO_my#8rP}m{OHZnYq&65bbIb!&0e~MybANM^ZIX*|a!+h?{u< z_1omfV%%h`KHi4!7SH+INNW3)&fT9@4Xz(nE#`J7_LJ9Hgn)Bhs`y}&ddsf2N`cBZ zJkPTKWxeDbFVR3K4&Zp_r$eB{mxr zplETJ?C8Xwx)FcX%43bDTledl-Cg`ri|~VTBE`zLYTd#J|H|>+A^Ubnwv_!>#sY_f znxR8FcTV?X1A&`CT9ZtT_uFOZLe z3&QSQm>|2oRjYJ)2MK@t2FbP4Z}v8YvT843M{z9!TDwJuBHYM@Nph`Le2}5)O6PPI zyJWJ(p8yM`R_tlJtLSss*j_($%vLQmlt&I~FcXfu+?jK!HSrx1YUh%s&s(JNS{B~| z@1wvX$Wi^7s5mU{g*KMRjVd!jvYWq(!*M2-Fv!4td9^7jsw#iItOnirdA5-sY^7R? zO>0J`>npp<#?9*{@{P>zjLqot?iOmUxx2~|^cw6s?v<~qP!|+aX9HVh?&j6lmO5eu zO+i{!>*b1-&E03II%o@9sBp{YRP0X6UxX^H^0_&N(3M<-$34Xw%Zv-ZB+W|A{5f)B zN%V8?Pc8$04Ele7)%IhFXs1*5F!69YA7&1~*DdP)MA*#uPEJ#2kkvt>$i>!X^A>|D zO1lTXYKpZ-R^C{6nMv5)vURf9gM4JH(1=IvwlWzVbwzid!V2 zMa%_|2)$use?zv;*^&IXx>&ZR?`^9M_$4i%W6or_6DWVq3Y@o%VF$f)YcU{bhVVHW ziQj2-qx?)6kIVRd6~;zHsssbG^GK!XA#YRc@tjTAyhR%QTI--(;=Q5g1{pXr=0lE! zE%^Eu$ZE>(bjgSCgLAGLA|2P{fm|BfTI=rlr?63~O#N< z0dpfq{4{^Ax*%nfr6lh*CFe?7d|G|Cy)w7$aHIv)s!w`E{GF_h{S_~~AL-Kgi`w2I zh#pwT9lgH!ynqp)(9rg%hr-5PFfphRj9u=?jz?d+F25%WWo$Iq1r%`_v@McnHi5_% zLC6$0{XTQE^T87B^ezfBZ{q@$y+jT@x#t}7%|d^_+oy4fl{75z*cN{ZROa{3K4v7c zHVSXNzyLGsy*~sQ9qP|$UO?=@O;x~$-Bm{tipDAT-THdAwq88VsT}Ssi{B#ux-6*V zZ&{GX-_C`@M-@HBHwwyAO?(zDR)#ZK;3r#&SRc|T=2eI-PfduKdKf5i+|tO5%6_rh zVi12aNIllhwE%40c@KUD+MgJ?)~Q;I2i{$-m?Ay#U3}dQp+{QhR*Ys|QmE81v#+g2 z7@=IU{_;u)wU|EP{cet7Z5jS;AX2u+D!Ipm>O^a(p&jRa;s{JnZE*5x3Lg>}G zZ5|8_?3mo^h4}IEKCDME`D!bOwI$kA{APr>)N<={BT&&D?(q zNqpoltFqJy))NR#z*<=Y#N)nA4OSyk` z1XZ8U(3F!?>LwZ>WJC9y>9f=Qz)ME;?~gKK#C0Zc)L}jEc*U<~GGfNgfUrbo&~7bO zICD>k$Izc8T@c#ylQLnEHKu*vjY4yRPd?EH?;ZN#xDy$!Za;fmktYnUbl88b;(q&W zf3Kg+n5o&6BMSR^u&NG!ao4IfQ$NR~ACskr`M!3%Rr1Onkt)48!lb-Xb*j?^_EP|u ztwl*EaFS?lG&zW;n>-&u=#e`P(Mew2d|9hu5CX6hsRX0rqpE3Z;p5c@DMwnj4ei+y zj>G3e`qg`t6WjN9eRdWjeYbx{yB2|q*^diUIC!_Qh4Ig$7&@J`SscH2FsUkKT`u)i z8zTO__@n9p%xD)tL+Zb{v)^{^D(;Rk$EN@`+Ig-aDKMeYRsN?tdO~G)kvOw){M_L{ z;^p|l`Q}r1{Zq}2OuxdV`*u350yg7eBMS{O{vzdm6aDX`j+Jiv={0|4_%cW^s$uT_ z4Qq2TB(G( z*;r$?RzdiI%DSOt(XfHvVG0g>Zolk)O|?Wy4?2Wx5+l?KIeymZIImJ18yxADST!S5UTzoGCiHh5gufFGjcQ*wq%)3$^-)8ZVeVy<;qGM zl44&W?{|-YHZsTBP&nw4J(`lpXO4%Xmi}x1C?ns&LREi)ex%&mtttAAg)=z4IGM** z87%MGp?~k!{QyrEie}8i5b|&E<-Te|#lPShVq;x)8uXl-=Vh(Hw%aMT^SUhLT@sv= zRniCo%oOdqZ|P})Z3!7w&A{RDMYp1gDO#1v-YyF~wXVjcFjZ~yTvwDS{PPd}I(zN# zOYiK@v|4{9cG)ZZFziWecZI?K+<(G1Z=KN4n?A)Kbw|x@v;no)q_-4eIN6att}$JT zkX$^)`|*&I?s3n$uC($S0;`2XwjQp|w9K~9ux|~kbN6k&jz>`YfE~%H6o#8E$jz5Z58!%X?tZ>c$8)v7hfPKOYk!Mx<{FQ=nW z8+?DQ@cfiqyMjT)*6`{GL6Ckc$Q)8E?i@&(CrHi$apjl3)Avf8+MT+X^v)4F* z0}QG~v3z>vVrzPb6k#&-E{iIuUSDYrh<1N)q&ESkpLY zD7!x^c4jdHC(mWHtlidx#=r^H@SSyaiOPPb`fy!jrliKporm|}-)_teRj8&TH@5;5!j+IwvY|F>ixA00Uxt5H zRtp2C`05Px-L{JxK00z+?v+|XGX5gh(W(1nxQc$-4A9)@YjF7;*a*-01F{Q%WVvt&iL|uY^ph9k15LY&Ub^u zi^^X1qH9zwR4gp^$`5iOBDiq6{A!uQOR?_iZ+-9kNM$3_lEVL^()kOcy+U>iQS?5fS~hNv?X2O*|x8acrKrn}5P_AeQd`%8u@AIXrHBjWc1#ksoOg>Zj0+sNoS zD2tpb$3O3Ld;UC78SzWRT?(c8cjbk6I7yJQb%)zDL-M@+xj!$6Rb{BCE9}q=WB}oP z^$)-EN8anlxgyv4`-stEB~Mv*Ci~{GBz-3on+9`(rBY{`r*7<_+gW8TZ}YH5>S-TC3XwN5aV_ZUgw_L_>vY7$Ri_ls|( zKiHRXsEp;p>+X6=3Fx1mp)*QLVrR)z)eP8nEmc}gxhwKLZs*c8C~T~(cbifrRADFr z9|wfNg%^!SxDPTo!*qY)+z^cj=qH+gFjb7nf7Jrk-!a)>@2^}5*Dy&L*k6fUe}!Kz z*@uIxQp;1guA8Z!Ub28I@=)BqaQvJ4V4}ZIOM4|!GQ ea~jZXXrRxi>2n}QPhdoFX7Sjo~c zscC<gH^`x<~64)7{a^^6xIwtkm6`%uiFzX)D^5Kv{LDV6}*mXBQyOkYZ7q$$83WXKlB3IOvQ}1 z_8hV9%K;MpvAf;AHmcaNRBcN@VvWhbY|kSJNXLIwpfBI>wGt*EWbMcQ=G8#|>*MAz z=4?$d+;cvvNxj~p3TCkq3ctIFMb~{3PIJGbx#RsIDcw0%GE+#}#i#9^`ig!aEQOpq z%YWa*W>LL`A1^#9Wp;K$znv&OBaZK@B7b_s()wi z+$tp&kW)M)jHJAu>VzM8mp6GC=oMSxF6BBS=>c2JoJY3T{H^|!^H>e)-FrU3FV8Bk zJ0JYAR=n4=y&rmF4~nz>8Wnhf=1=E*&l-Oem$I~M#^EPjZR@EWwh$U!9%G4|-wp4C zVH_L>Ds$gK7=fMl@XG<8A~y>xA)Zk%u+4x;ZRm%7euCA^neAY@{gR%2hiSHa`u)Q} zCQ=H%#A1ku$*uhU5lFCVN76QVM!C&-5n!sciadaPHj?dJe)60p@_F#sFOX-lPK5=iY}Dhis(z< zsODxlTxy8>(;JC;B%&7{HCwetZ!v$G6(6y`5a15A)}V3TmXBFS@E&Ijww}+ax@-Rg z>U25D@jL5Z1HV>hr&G*`^TKL~(`y1oaz`1TaU=N3`RUUlO`rF<#T+bfm17<*?luTW z7>_Js)OoGE@*Paet~jmKIrOtq2FL6;75lV1&jC}pJVOrz9HguI`-fo1cPf7or8@`~ z(6wR<-huDbZNf2Jm=t`Pe`Ywv6#es*l)%U11q_^<4PNFBWJ^sG$62fEA~=-K|JfD5 zCH4hC?@mq~qHN+6g-4}c4D>|^#AFU{a_fCNK!{7`5>aK?K?@@0B1DdpO`xY8A{=vT zCwP&#cE8S6=P;rAey`DMI8%QhS@7yW)bl>|`D~Bj5_Xyok4v*~qc_BQq1HLsaP;^r zk;(AB=4I{VrsJA)mS9WAsE&k@Asw@Tii(PpgL_~zVi+Owok|u_3gyY|@9bmgv7=0u z&JNz7Q?&OJMr-gUN>)Wysmp`2#i5|Bz2Yo+C*n`lP8$3B8v#FsOgn!w=^yTQ)%bjG z<>MS2o&j?GJFXfNST%6kWHPu}sKLZITiA*Rj#VcO?;$B6!DHH|JUUoKeH8jRnIN>u8s~38mw?3Ka!&dGbSgybJvw z82{Xk>QaC0_~yxuT!EB#f2!A}P~L`qLTv99FnL8WZH65%ZL@!@y9Rd5zvVC)))F%y zjG#l-zrFC+qznxV8ZRlpQ4lL|QPx@GbKP&CYnBoKCC7?*DCJESUy*W>Xo?{RzZLSr zN~quY1QIQBDnxZU+hh>)iVC7f`a@doVYYa2ycIu&JNeb)jaIX=*x01U8;*|y2XAnP z8MH!_Yik`Rjf8)8W^87?yTde?wJjDqEV4;Ic>JQq3nD^(OI(hoiHwM5q?xPXvMRW^ zI9({GbsIPHP!4$C^>4n1$fW{kuUhOZyGgJK{2t~*7CH^+Zg+Fj-LMaFe^IJK*F8!eOY znA8-E{>UNY6S^Dff)3x-j3S0LB(kW3?=jqCZE?UG_}C z1B48`#|c(;-e&n0FxFn7cG4sQW_URh=M3a8SAMyE=}}SN4qeh}uTW&s>s0lF!NAt< zk?!t}pvjwsgjsz>$@QqxlteAA=Se=dYyL9gpaXv*1agOT9{YGgBY5V)LYvX?m498+ z9h}X6&kiQSycbpo|J`V(ee@cz0M|hj8vr3g`@!vOYf;iQ2s0f^`Dfj~iworU;I~B`taLj+H;NXR>9Ye$?E9582|`zAlkw=k8L`#R7P+F~1~;xkCmw{8*}|g7 z`;C8}+fKCa$y8YAOku2A%+y)?KMWlg*5j0?7Qz&>`IzkQj>b6r*Kx=Q5LD5&-?2lN z*?(cQVV5aY16AAn&;oj^7eQrAOp++eK8U`&v7$;S8&7CIyQ_D(Iy1vfA9?aa4pi)mEr4~ar!Qfy?j3YhoXq7T&1_tupNP_#c3-+X3dYf2Eoj| zrfj;xm5xjkYJD|Sv4=w7eQ>V|K-Oq8L9?qZQ6Ipmb6J;Y!6 zOV`StOZqx|BMP9`{$;(~P^sVUA`^dxfoAmkX!8~LI&?UNPbxc5^svXD=BrthQ8_m1 z*g?Mkg8{kPZ1e%992{hMUSi0BMUd!bYe=-U-=FZl5#s6MPj+qj zQ2fISO7>3V{-XZ#aEc1cX|w3r_5k&LuyV}{#IE_t>?9^8WwXz>n<|F0J;Q%OtcLCH zqVd$e=6$0}X2~*B^n`~}2SBc z=chF*iFxgd{Y?;2XLF(~mA!xeI}j=UH?-3Hg-8UdX(S3?MvD6@&dk8$VqYfiGA#^F zZW*?oIDODQZ#?H<95yp@m+F7Fnbz;{2r45#>f_Fb0O5nrp8br-o^|$0p2zn74M^(? z{zrvtzb=l=ZB&MjD$5&Y$i@bKMkinc5}A$E&>G5CjdZqst{`XLm(5*qn%l#v2Vped zu=gR~O}CWn!Nt|yk|^Q#6-jgFTcg?FoMeq4O604>2Dj4WF9VFhj7EP|QH{h~Ow3;s zDRl$WIZT45tM!(WwX*v&gO&A~4!?yhx=+nr9IhDLpI(Uw7{=1vrZw#5B;z04St8wR z=KEXvKbnA+a&ycGxJIDeakiC3dfndWg_t|nwL)@%3%%z9J=!5+(w-O=-WX ze%Q^VdVn{%cokIkgEf$6%B2Wz!2a)$GX=?+Nt)@7Fv51CcF4+{aR5|NF$Ug4i>@Rz z-i@|(Zd(Aj4HkdQ{urL^^Zt=23HnZ2ma4Udu3}0en_&%Eft4P*>MT!HeXY-13l|p? zR*XJ8uEGQB<&sVz`f6XB&MTVWZp5h~f3`7j+(`_4D;7MnF8l~AFT~y^wf~$OU{aSx#U~*x}`-{0N=pNSnP0Yo? z>O!*C*l8n~H*!*FkU))1#~g>Drr57M&nZOB2%i=eQ`%qY{_@yI+R3Q(w@N5f)8o~W zgGTrFY-AYP2Ya(yuRRM4NJ@>WV?GbWEf5R!ABqA>2Dyn*n1#B|X7{(w$$DzY8zM3d zef++)-}HYCWlhaG*K{H>>31$?b-)U(%B3!>ygzhS)lxjpL7E^`z;h4O)NpmRb`vi+ zQ5-f`I9F|yyojMEYipZf$aos26EC#W^BO?aE*=&;IFaAK`J=0+4iMhn3G=I>8^$P4 zmhMck9P3;_mO|cAn*ZDl+J=eVdtZ5L60qIKp#6UyOkkxU;Q=Onl8F4(-7@vgW|o;& z`abR-y6-cT0tw+irL{f-3_u!Iw4J6J*Y{c*5=UCZN^n zI2S14k2y_u4~MshVyFH)wA>RxJ^FIfUvz(c(PTG;Ty~Sc4$l4!q>ILRLOLY5|3o@O z$orbNFRvEmg0`Sq*;r5U)cX-afBl-+fTC#A1t#W+9>M+nnX}Q~ly$JAaB47W((kmq zV?oZ9|5U^@F;Y^CJI&!{gDCf6(cs)%VYA2oUg2zKY{zvl8+(6v!CAFqG-y$z!d`!@ z9q-plv!l3dQgvwu$hSEW9mZ^uDoFYjj|%V4*%6H5*EcqBSCMvdXHTc=Ub|~KflOnC zj-z(lqnDBixXV!_KE!5!uwp&ElPpwXqmd(+8#*8p@l#D=HmDktsKZ8m?NcCUuhZn0 zA7k?4N;$h@yXtyfu1Wh|MS;^PMud)#av6y&RtNBTYHvY-<`v zS4~`T>FHq0YS&>WB^av=G)vz@<(#-6lv`Q>_a{(n&qP|6>5F4g88Khti$MV4%vTrroCL%Ai-& zfXAvT{U|_sI3jB}URjJK-{ybZ5~9d$bfpnjls1nwb0&bV^xqUeS7JNrd&0_L5AIQearmnu|8}C(&=O ze{tJEzl!F(cs_V<0HHBxb;*)bl|?oKrI{Idjo`NYj8+0)>#oB3!@Pgkh^P|aw=^QG z@$L^d-)&bK#k0L`OX;=_Meh1GMzXO{v#h6Wih4D$mF$+j8F&41uI+i1v`cdLhvXK$ zBA|;g#qs;QU_b`B|Jj@8*->ve2V3#c5NBg5UeJ8Krrds-wvyQUtCqX9ux!iYm$%R) zk%xPXC+Prn8oRV#y^Vj2BIA`gmMX*EKWNeJg5i>o(A}K%N}GYAC@Zwr$jtNe|# z;T%6*_6_{^LyFa%L7V;AezQ{DuL0}b(dmuSZHhv!L^XI2yq@e!Dy9ivLh@kL0ded?#Fzn5P7CN?-*j@jNG#c@SgUzlN)XC*2C) zr%ad4>FZI~N1lJkzr91r^}2lXPA(bFtfHb)=Lp3a$P`F!VIDXZKWy{-Y4x2k*lT@; z*KEHbmH#N)?~bihueiwX;hV*yMq@fRSG-`jkc~4Y|EIk1IgYd|0)4R4&UY1LeA<5L zM-(Do@y1Hn5~-eKcysb|<7}>J?V_%4GvTL1WU}l-u~B~<2`|u+Nr*&~z(OnJ1t^mwr&+}7GkoK3nj9P7UcHH#NT7Ue|1cQ@w)i`j%8Lg&)B0ILA~_^{3S zQ(V=$kB**Rzj4oCA`b?1C~pP69v+N7`n;##!dYS<4YG;JV6uSYpM+_vX;ro%B$3Ww zWh^dxHW(IZ!fw0g=c&cQ&cJx(v-dAh`b3z`x;}pqkR}zA@_Lo5lAmqyJ_Y05!#RyC z@598M3!aDX|HuA|4S&Rb#)--N#Xs1WOASD(u2GwS4w+?+DpLQPszla}c<4_1n zUiK3*UOejNWRU~;2VkS%m1_4S5D{%V8L_^YUw1Y2gJ)~|Ss%p(Gz1MlK|(}6`Cy^W zGlYLIit?vEyY(a`-r)nGG6&hjyUXz)WD_8=p3SwRv@`uPfil-SHuHg9XrzdEQ^ZxZ!gaI5Eek>K?G z`-K*prq25-`%c<15zZeU2hP!IHJ4&m%;tZNI&SBU9xenttwoK-JZ-f%Y?LCCr9eSj z=(Wrw%aLPPalC6DI`&~__OKX($uc`aUle9x49(f5gCmh_AF7A@-~PAk6-i>G__&!` zv-Gc%r|R4u>A$$}gDji)*B|DJWK7&P>h)`iqZvYUD-;yF>nK={wj2YW38eDCy~%(2 z;%?TgCiLT;%>+DJvkGgIBCv(Pj(^XcXz10aa8XkfbU97o>8kV?k?Y}1KB|GU07b{m z_y1gfP5(`8W&KTUMZKfKi~B;el0|-3_M}LQ=^1^!tX5iErb~1s(DFum`v!!~PR%cu z5ufr?N@YggRyFPS`{!A{u)ZG_MoE8s`^3#SG!X+sVudZigKq)=H`^Ge=EWV9PPQbABD}0sZeqt$c=b2pGo?ueC? zm@eqooLqyp{xJ`G|6-4GkY9~y1j#Yj+wVl}47NtA-6i|s4C4O;vp>7z+#Nj6H&vqe z<>F~E3cY4v_~}A*UL*W($`OD3Yt9fNb&3@;XZ?Ipa%q*|iiTc?EQd&WBHEfmLkgX*=o%!Lewy`dPRL|| zW<;9)dHzb3?r;7I_&0ww8LTY*ikSrJeJjz$0SwR|&EQLbW*2X@6vKaDP*jLGTjyob za2h;^qNoSIK+-w4p|yun7CF9GIhfs( zI(<=Z5Ij*wld(}>ba%p$n4Zsg9gf`vnK5ZMbFe1(MA_WkthJk^>#o)m$B0yjb`j%M z02WqA>^mzfniXclGAsNS;Lh3qoH#@Orid&g{}N{vY? z0&goK=Q_mZGTw7qjKyP+G3n96PM?TyL*KLSbtG)9_#O`q_KnDUo2Q&k zn%5)FJ!R7n@Br(JkN8M$G6TMd`&K&`NURU{a{U}p*KB{LPk%!+cYPLX_M9p(kO&oc zw`+1Hy79_u>7I@CsGZAdLhLav$j(;iJi?=uYI{n($FLvE)xY7gNPk^c@vqCOhA7W& z>UVf4U{dfg_Qz&_+udzfEplNPh@+{Wk^8lz{o?4GF3xdxNTh0kG!=tRUNZty3F?ix z`|b;et~h@y?H3qBWg#2beO=;VTe}k*vQY_!j9OLnH3(acZ3Rg^OW=o`P9*}Gz<_5( zA5TB&laIV2?xsP0_)!kHaZ+tnm+5H88POYW!_Mn9icS6#;krGX8rT^_{?A7vO-cF) zkRjqMd)#;c8Z(^f@6(jQsEkw>AO9)A{x7YP_e6hhkVKy1a>1isacGbF$ItK|*N;e1 zr(or$jXG+n*u4{{?cvv#``U)DWeWu!pYY~4eK0C6DI4`aCS3n z5|e-Y&soZUTpq8b6#yk;VVKG1J7aR4KJPhx53K&*|I)`0fhINoG}`bOKnZ)-DaXGz zR?6pWkH77?j_e)wc>4s+YXF9q*k&iJKs#XBB=pzNzh~ty2uIv6p7Jtkj#?6aHxi*( z@Q(vZM^pa4zZ|CX+7FHzcyC_p)jOR|(hPs3^ETnJYAZ+JF}!=jX`ayBe16tjvreq{ zpGNdrnggQ|vJT$AM55t23e<1*P~k8erfhNENN2K}>PFCBBRP~fl*P;HcFkV$6EoN0d> zJJf3e4wOz=Z2lZjyH)agO?H(hfhJd*ClL&w&)&Vuu0hN7x_CpcQ)ixd@J8ZnGnZ&k zL@0~q-}I+v0rT@8TJYGMpo92tL~fbbsHT%Dm&{IMhc+YlE!-b9cpT7xUsnD0LJTFB z((Bf<*ZPSpEO+`2k!LJ|m1GkcsWN}bt;&C@$UZZ56k#!97)q+HA(d&}d)FPpJi=0Hox0E8HCfVRw`B22$I^qaMHDi5)FPXP-$U%bzYSpsu6u#vN`f_X z2ihCgLB(ff*%~T&|3r)Rm+`2z2jLJj>mN&ne>Ic5XgAKbBHyR0f@En* z%b8VmRw6WP%m!jbJTb*C610Eb5Dh-6<~nKo;tFwt@tw()WW`pOBT6&b#?!4Y23T+CD}R+RI(bM?(2$gP+LEc3jjztgPN(XyVkqts%{ia#@x z3nRkmi2pZN=W-;1+u;J8j?-o$v|Q^~AG91Ky?3&n|40(?qN#mv>8*dJT3ggh^<`M7 z+_QxVsy`~%95sB}vL471YSUMh+wQu-0VrhMK5ft`!2-tj)@NA~uo+`l`@!W})q@8u zXrHgLI&QV|UqK57USbICRFY(4kyRx(1tDAii0`N@Zf}ew6QMHg%q*x>Dpcz(4AbR~ zFT2vZ)O}_Ds3bU?XZ3$sx&uG#-*hR_)Jldl_zJrO6Ns5<0?OR%bY%u`dK*$udD|<4 zO`akl2*NY|Hen7I*%9dl{fE2lU#W7f7rW)d(WTsI_WDVH5Y!MN;(vdJXb9Od<_0J(p_XRv88q#itQX|1=PHiAn)O{do56oB$+Dak0VR=yYy$_L zXq!I*f^afp1fj!fN?zR{AOcgcZku0zg?@{qUas|( zgAGeyU8rdz)$Z+S@s`)scQ?BNr>%(Xg!`2S&vG2Ak+~(Z=^Mtz^S<*(X+|H+!QOoX zLd~sNY3S`yEwg{+Yw=>x987lPnZY0R^}zV|#`;Zkk-Em_*l!?CDE45jUN4C`BfnCN zOrVIFAv(PTH?r_gtLxq_=lbdC=^6BE?IHkIkPjtyxL2!6f2Pl`&TPmUljbHQ#qzil z&QA{ZJd}XjT;WwKH zZb$THzLrkF)pi-G1i}_<_X9vlMvIeN@{w=Nw>BBDAJTfI-fVwD4-KDHa!p@qMhNs}9&v8w#XtVa!2<{)O9n^&8nMxqR9YnoJyVgBFjZPbE3@leqt; zLkb{J1`eiC>jIfxPoEV=exAEz(M530o=lfd0c& zzF2YS5enb2vMR1MNkYpei(Ju^NNb*#Yl(x&X_8qDCj0aA^9vlwS8mpvSwe0Ao3mmc|t>VFgUailr_EN=0pgG3?oEr^VOoE+UPL@LY8Doma9$MeLe7YPU}F$cJOekLo+Mp2KmHt0l!!V}RJQ6_$`Cp`*#DTJTN3-9ZMERIo&e zP74_kVQWN^`yM8iXduwp2KuYPkZTL~>SiNz@Qcux*FB)$e2dOirMbKYTX@!5*F zw1+1%sL`V4R)}`5@O+|&ur)ew4^n@F_PM^Pnhk}Ja8>`igjHG`;qrKa1u|Rl4Q%4k3}B8pfJ`C6a%M?|Z;w;^aUtL>f_liquig;R55aFbly`DPva$5EPLpjnk_*9IogYkow<`&EC9j;o_?GLnl#MOxt^5F7yO+=z<}Xb;6JV5zC0FaN>~T8Z#_&y(+jt{8a0iOW z!N$8BF{)bjEL2Z*QZ12I8K`Wm|2z=5oIWgY=JaoJ*MP6IITzI0qb#gP8Kfd!A1{wr zSaxQ3g2lG6PWu~gGVfQI7ixcq^CmvO?rUj5_dV|tY7Ja!X1an!-Bu8^d=9sg{mfJudcB$jlS<3$-gf2URk zW+C6(8?9S)b^yIMkrv(>w+-*Fh{STM*MpdJd@#jN^uZP)SsC6Zx{rSkxrv*?0ry@2 zKYXFfxmOab(x}RMNd=6`{-6*;Dzx9S-@EflrZ4J^5X^idLv_KA)2U)1(N?E^uwLov z>cEstU9EY8a+dvYyefL9-PiOg%a_A++jh)EojmB(1ysjGY)fF-=^>6rPI^vCU>Bm4 z#m|zT_+q|URyKU;XkCBPoOj@OrhM-6FG8d84U}y(G(txD4lftznLFI8om9|8I-(4! zT@QF>?MzS7Ma{!A>b$*0m^ER&z0t2UIDZ;1*st=2*Sn_HVy*Le?{VXyK!HnXTpYK} zr1t{6twY57=!LI;Fzj=777i{h*)+@Lp$NTN@z}KAQ~LEvt8{S*=}CRq8guG6no|Z5QecDE#gh ziUrG@%0((d?!bRkJq@z$IUF4eu+yX2DZ?cXrSDhY{c~faZeGXeb+~ohHM_KQ069R$ zzcv`IEKU|UzEoo8?98Tab3dMe1-%l?#lOyqH?c+NdDg|MZZ0!NR|eDL@>aH8uut9uLJ)BoCyy zl3aAWx`gEa+-bRJgs!ohNnYOFoX4>lH3&|$6+_jvmk3!66LO8Br{`uXR1|c%Z2k~o zHV3llFY&MxIu%C5+%F4SRzjYToJJ1s9MunYnaN>u?nQ@ z+e^O1DZsEo??;S(s0XuuClpchJWbo*-Td`kTp6=SZU@%$G6nhG>zYXE+h6$o^{uPk z|MGzEjqtu+M{M1jn=3^D|E)mqUF_(3G9yKvKBIQcmy7+?8T-X#Rz0e6&FYx}h7=Jn z<2m`cY#P|x(^FkwXL|8;ElS#QX?}=<7ebo=r9WRMx1P~|e=2XWhu!kZ&2TT_Lelul zFeavzNV>I%wmsOlUd{XG$}1Xm`sg^7-SO7=Q$W!tQz=RPCyoFxCPu~}!7F+H>bo6} zfUTT#Tb3?H=SiE-H)BlD+~;twq136J&wW=Yb|Z>pOEL?Sg%zvN#&lQ9oVw4D#zT%r zmhWriZ-?-Ib1!}@6s0?;ehvkfZtRVdi^G*~aJCpp*aWzKE#N#WxX{96;&3#R{K~rniQ$Hvx=w0Z zD`108apqJqHHL?Fk=?`JW~QPF>Hb0{%s)3F5;GZpBL1}fII*TUy%eIpw&$p~X0PM7yX1{si`ZJ|A z{@uM~`#yz5TxVIQqjk0?(0ak4O~@8yamSqfxUS_x$^{8{)?!u)%86QXAL{7d*6|+R zwoqq(aV+c$d#-euBxv38(g^eT)nbczY4B(!f5}x*aPpPt&Cj9w*i6ymSFC77tf64# z4cw9kHK)ah#?~b{nQe~X zbEQa)Y#X~z+2!6%QqVelTE&2U$BnT&zxyqJUN021;?s8<{cF|fj{UxYhf6n+@qF&O zATr4^9t6@(7yDxi9}K@n+ZEPp0QqRCFOWs*4iw9CpN z=yZFb!@Zg$2jzsN3_+{YZ`cz!hK)qfU&G9Q63u8uPC@M-=)7bfg>c6oyA?i<-I_Xo z;`cqyzgBrKGmt-RC??E(pY424bO-W&nHUzZPvMMp2V8E~*k;qNz;cC5s>tF!xZ3s~ z);{yaO}8z}Zn8Se+l zn_v*&7FN%4^ldM{2#GxGU=akdCf{X$5;9j-ha!5--@dqCoTpvgp3}-S!58SNu7-14 zv!ciib_v}n%guy=jULy?)ykE#(0Ssc!g<rP{%&`NoWY9swwU zirs8co{dUfgy!?tm{Ah9&z)*{AqL?urH~6iX*uPZpa?AwdZZ$ziAh6^FC@tt3I*tX z=3f@@S%OvLdHNEy03}k3BDQYvch*$ue}04b7TRxuROJ-<1o&34m*Xj`e%$bsbvK~e zyO;tL(0fcU=IL%PF9yuk{}k4L39Ed0@h}VPMxc7tA5$LSQ9u$O>uNn0-1H8Zkrwr# zY#5*TDoS^a$NSROix4*=IpNnc){eAy2M-_FSk5MpY#$m+QQ!;Ce&bT6XIt@-82erd z`=)1BW2+t3foEmvCCalZDQ0LK1qMRvWx?7GN{qnw@863VR116z$G;VSuhc6GUwH;7 zi4}oqAScp);^a*OJA~WH$Rl)HYCjs5P=P8!mku?!kY6llVUhDvjsmJ6VV14^FD4*#)nAoql%e_xZW z&?R|O)q#W2?Nhumk_nc7ZY&Xd*-SUnM!s=UB_54QHO*qiS|3PUa4_Bo5+WMAza(q> zJbU2jC8D(2&hcBuo@)9%&7URz-#I|iMy>S^;cp#l31HE3Dd^?7s|!i{E8LhbCpa}) zUTEb@y1I_)=m(wuW>>=ax;YXrC!0vWPT!3d`n-Fw)h=vU2Cr{_Q^N{kFGq4WPc|H- zsK`~1id&SjT-!P!i|l%~DdS7umL?jaJf89ablY?Jox1;XcR?w?0^iIt>flUG!d>5q z%$A>b+{)p`8MKR?x8_l|i*GeKm>jpy$1odSQOxqZ%IqZK*EiCu(%0FW?`j$D^tp*R zOJ-ltRaWqGJH9P{8H9Bz%;z~fZ|%D>7}Pc6vFeX=evjJeuo&~>D;OFvYCnq>dXC-7 z3};v$ijh)pA1uGt$>e5hvRvh&Ur&d6K^n;$t~b$qOX6?ie_2nH55xwzm_i~RRRV-0b*}28JH~!{bw~z3r#WCZL28Ld#O88Fb(P z-Jdzdas89hkP1Qlud@L3T&Fx~Fo}}r`9H$;WM*g`aCA1Z=#BcLvzeRxTcQnCUY+)e zdn2E)%{!TYcI6e;nTQ8|Xj199om2)%__RPtntpYML_+sO4~2|OntNx7?{Cf{>Tyg( zcUQGhT$D2T^MhNwT~hh9!@p$wGyu-#-VfsZJRg3PA>GJ$iA1BxqK;QUALehACqG;- z`Axbvt(>mMUk?<~24|5?V;PdJ^hU~lE?LZ%ynd*Eux=7S%YD(r?GPkzI!sDkh1vQS z&-PC(vb-keSfGyRk0=SCH>+Anvrs(~lH622$^|=%q2xuDqZHJc$ufOE!y&W3@IJe- z0>hYsHvWtb8PYKG6)e7wCZ0XVY1#dO`;f~Lg`(BKdoC#j>5PW9@0FWS;@GhlyP>Z( z*(b_>_)Im{&9&$rB+>^v?Z@jLTi8)(ss)u)>haWPDPkvFm!iGciuB;FjS4WC&UYXa zXoz>(lJ#EcS&LaZH1WUU*r|~Bfs#b6iSF^5MQ*>=V-HFRJ3>}}e!zn&BjFw8v}ChuqHrXeY{I#lq5kVS z&*>DDEWx#UDXor%V5lFkeTTMPdGYN2tQ<7TtXWYaSkyUunMg-@B&L1+fvNl!8_Tt1 z#y-bu`RB@=Uy(3-cfWqf3=Zif79@(o9}MgxyPAlozD$(H+y0=C@yJ(m7kye z{yMQSjr79er&UJxh!2tM8-fyw%lf)6H9nr*NSM+_vHkqsprPUOAJ>#geJy#yyX>Q- z_O_#lw;i}!uJ1th6?<26M;HlMFK0x5W?{DT4H5)IR0)jvl>qu{XJ_UgUTd2`Bmu)v zSn~vdlk@)tYi9xfe_^fi*6g~((diOh*sV>895O2tQGQv?W|4ViH<6b*U2()Bd^XqD z7`9hF%5G-`W2@rruZiAy`I?fnukaPUNn^F+glWH`W!LvNgpb0@x4aJnv-JIcC*(r- z$aiD?VQcK3N3lI*LQSJN(qb&Zg_(-o(IJF?zRK{SBF2!er0{FVTs71CK4oWJyKv;`#S-8HKW^X18CUB0aIMyg(8s^Y$R!!pZuu=fR(NHc{|tL*n4^M=j%)}Za5 zZ59;V5o9z#>?3E~d(E5E6~;z3Qt+BK>HUugs97Z@hMKl&-EEKr{RaDgt|c?(IuYX* z=HR~4L|#ZVAv{vC*L_g8E1HB?sGs8aCgnD-{L5JID*Zz3e<^KF5+!Pxuak=TRU=Nu|QuJP-#M8OwtDO@yO&h*ZFZeLlK zYkR}F&Nfk}ocPy8eI`)rwn8pLXUj0b^44+oK?2KWhI z>Cc-O`wSwQ zHx*-~1UJSrBm82vfl|VpZ(}m}63yP=E0tU6GO10!t|7N{=biqDzDoV|)o)|60pZ+; zfl`yToQhDyZYNm}$kL14%gr4MT0i4O0?b*43IIK+Gt(%4@=$ir41u=2Sj$T>`|2!% z!f$ooKD*DJ4P|(5^;*PNRA(tGjm#MuqZGQ20m*BtS}?)b9I-?~hOCUItVd{K6&l8N zBkAi&XaQ#;ox^Y7hwp^mhkW1$sv9wC&>wpqou2Z#=%CvWHUU@E{lS)C!~}_w&W;Ym zsDPu8ljt6Qtv(|xib{8g0CMhfMS2m#=~dQ8G0!(OJuaGCh)WF5Ckgj(K#qJ zdcDzc1BGo%z$I2^TsQ2CSCv*QFstz^w>~=w->u%(anhwl6Xqu<#C}PkaTX?85OYgx zwm4&>Dja()$^h^+3(UFRyvo~8X z13Juc0qZ`p9sSL`%V+r3s>(V-JpyT*7K(N>IEQWxYj8gQtvo1cw-*4#AYF z>98ArmRLTE%zW9kZZpLi+jxaDlSi7(;agNJgyDb5ucarh2A4?)DU_wr{szL#`a}Fd zyG79&u}*a4EZVi*@`em^-U~ zUqe7ZeaC#+(LTyOHCF!x$Tu*&ockdMvsX%8kwSJoUk2ils|H&h@gj4?YHW8dk0wNC zYdXT06~4QUAA@Rt$`M4C>)O@JKQwNN*)tbLFGNS1#tN!=_6G;~?l-$(uzO?9?H3HY zmp@5x3nOPdCQ-OC!)gy8vI$5#)4a2Pbb;6Ibb~fOq4CI8Yuc}`ZLVGoJ}EJXFJ&+X zK`x~xDCVXoJKgj-GS*gyPRDaIg1*9T*RGi4mKg#LDA^zHroN$Jq0C#^@Ary5#Dslg z%Cy@3HG#gXKxp~P4z|ndYa0@hrQ zO*0M9kd9Iz-gy@c03-A<{~ni;(ED{%rQcE}@{`}0pt}}fM-G`=JQ~qXDTBgSc4X+I zz;IK(@WZvgk!9pW;Y>4-7vH~s*d-4G9z^hrE=p@YMENVq_Nd}JKcBU|?`|>YfkcD> z+d z6Co~R9~EhR19psw6Uh+%jZ+OWw_V>=O!52ENH)I{vjvmFC~h*Njhpp6B5pc!Ua$_8gbdoJ)OtfwKr6U94ItyDvL^aJrbmX z!!Z}JKX&hrNnHI<^fwwWb1F1J;)0U_J#7oshApYmzhE-~5oInvAUYwJ3~lR*ygCIJa{*7W1Zc%E<1;gfAzJN(Vv^df~oaH?*cyk_Kf+ zW0VPtZT8?!blj1DplJ?H&0I9almNvkpy)VVr2?c{tK|_`DnVN%4_^ zUd=dwK^LVzhFpO57u%!2RyL8e;k@WifPRpdCtfE!!wG(Wei`bJb7fYm%QN(KoO09! z2hioUd=*a6F&-X(D8yVGPGg->5kh4bK-#H1A4iWVu@GD+@%oA}k*O@;2t8})eE>?= zS5FFM#bgFGS&=EhK>+b2*7yEJR}!HlTYe^!a$g2U=5%&Fl;`V|+P&vpb?zu;_uPDC zSs0!7TyIN%oXm!PbI*QisH(wfQPG390F8jNqdR6>bf5ewfQTijyPnwemX z>@P>vbL$t|wfCF1#K=b|!COHZSxlvakrle6LCv7efu?<0pUS8+d^zSW zlV?Ojr|T;nh5S=QKa8rf5YO)68jRjQi!Luig!}?7#Tj@jT&a@+%Jhnh>iE@US(1|t zSWEeT(F2$8tx@L7!vocfhF|m!WVlCGqF{k%uVzTYxWRWxg&D(+=MYUY;u7XAIAtC$ z;L|~}#*h~QldAU2knBNmh2VOmBl;fzwtX<{lK(k0MdudhFs_UdoWQIjkCvxce&?(n zQfqbY2Io!(Ri57cE+!38&YiAz|C1UJw9@i_SNSa>1OzK%FOg79L#hR{3k<%g9@nl1 zb^Txi6Pvg}*=Wx*gDEwY?gu4mLT9iz$G_I0z?rkYnP-8T!!hnn{DXn$BQ?RWRpPv&3DwKZU1UEmJ^)<)!^%7>zoA zDPxHjnxjmzaOTlaS0L3YmCb4>Hfpy|`FXPW^?5Q|h1rY6kM!mF(_zH-q_m6U!VUbq zOFgp=zII_}tS~vSkdIGDE7!zZyd=Ly)@X~n)`?Yi6U1B4W$br~a?Q#L7&G>WX$^@h ze#Mw5cElxf5=}^t>X4$`#?Ag%D-`&Dd$nt}H{gt=yRP*~33dP%33g4ntd*5e2+(X) z&=yZD!N-f4ly@(u5SW~0`Vp4B#uSJH?S_yBtFb$yH7ROxu+qn4OABs^5<#x!J~d8% zelpzXK<9=(yb!hwM5OcfXGwX~hh$)J%3xK3+=vq4=?&Qw(imr^qvj?M`WNhfL2D3S zM)NvZO-yZ9D|1CMq3Xt`HyRIQf^!usIj|WbwwI9~paJQBfp;Zqw(Y$l4!VfV%J+=- zsmfb67?ba7rta;dd{;*5`Fs$TSf zsl@ZB@*a>Sk=W$BRztb1|(-c7HzkJpyrG?2T&~=YgHo#@w8vwnZSm zJ11*|U#G9WODe}FM@gOb?z^3B7CqN*iU9iI@;|1EAEF{P)b0;AK9TU+$#Jl5%ONAK zr@T%v87*wImP_YL_IVF~zKpB^^c)F!Lw)7S_(Y0NQGYJ+8?L_#?)IP%4h$CejD3#* zyF=bi{}BkFkKP--7q1-NGS}Z_QLwv zqP(9NatI3k24!%!)={6>^lNEkZr9uD7Zk`z*pifCGG0bWA}57^w;w>ufvF`}c$!da zjhVW4t3lh%uP{QtTT>qD{`qK>lp6F+tN}{f-8RmdZ(e`avPE)1(wn064wr&!{N zp>GZKe(c{x*zV_eBlsVq@jE}2n5g(|m--oQeUU-xMELOqleVQ2lk7z~_LLXv z_~jdJM?RB(Nv^SEXrF%*Yi680fk_;m>-fUQ_j#u_552}0F9-9KUQ)=lnY_Ro+Cg;M zy?C6DX_enWHYP_TTUvmARw}OcayBN2^cxLN;)4T|G-jv@vQk36vB_1>o8i`DxXIronMZ={XoPc}y(D6g3b zZ4hdIOYD&@yq7cF?p(_;F^ZVG1c1!1K1jThqw9V<#t#av_XKC|BXG7HR#(Sx;MuvG zuD^UKtO&*Z^IBF2a>?oZ=Zh&2R9xZ8mn40$!`|Fvy2E<76|H3?{WiEeJi#Nj(ve&M zSog_Uog&a!Oz&1Tkj9@?x#es1wn@M&@`Z1INC*!}r)Gm*GxvS1&)&}g&h~vX=SMwz z)%_EsP3!z%+{!PC(3W&) zB&+>&*kl_H3sJiFy#IT@<=hqVTX8xV-vQE8AJ5!PIp=S>`P#1h{UG0qD{(SHyDH(5 zIsu>I=i*5Bx50`VpNQsa1hn&i&bW6zb@2YwUyssiaKzr~9&}|tr^W*;WX~EbUnRF3 zed6+9!fDH+6%~9L!kFvUBYvy8Bx@M0q#ESeath z$%oIhwA-i`RZbW?msG^DTZ_H*6L7LQcTC1h|0_*|@(p4d_7@3E%hY1a*r-yGn|W|e|A;k1` zkm<52u#dZLBlWzC9#TkVRX`l+)nIOgF`gDdDP_zTi6mn({?FJHloYyCU)s@_gt5v7 zHaZ5qCZba7UBBHw8s`gt=C~}kLo_sDof6ptT4)ik{V7nD)aWg7SKejC{W|1JqxD|X zr``xl+_uO6-ux=_RRYTO+cjc;ABktDZqvGeT^YGn}20CF~rI*fmAO29WbfOQcCoTnfsNQCewQIN(9B3 zTc^)N8XW>#0O_EAW=eYRkE((4j#l>>PoZ(pwogqE0kC&rd2WrZ6Fp#y%pMYivfWur z`ubu^THPPzhwLdM1eH zbsg!J7LeNm%e^T?xw%-+6?ag=XFb^B6N}V|nNWr9k5;sQpB$3(-}I^Q{*YU@1h^i* z*YW!OgUS`uM2TufH`vNCsG6!b%F;+HT;dUt$ggEql zaEtafie`7kaZ@}t>Ry+3$k~{7AE(vNel%_3MoB0PdSm=(HJ${o*+UESKfnjqd?B;y zp~I`1BP8N~Wlle}KSDfLl5k^YrdwF^kL!MWb~E-@7vtvQ*L~fAC@+mTyS?Pg39$Cr zNJbos*WipsRVy_zdiH^csvj5%PoL8l;V3Ba}ase|ej5_*b%S7rVcSKZ#$!kw3N!QI@(*%`IZf&*f0 zw5oKY73_qF^L96zmHUR4t36;uDZDf_d*W-CFSKi~kEdDCH5f%=EI-C5Sr^*ozV;k^Y(GM*VU|@VYe`1< zbe4$xGVUjJSccqjmO*9HG(|?}U&0E=AtJ%#5&=*p5c(ZG6|JHnD1S%;Red*<-xP0H z5A6x2PWY)kiecm@hVTM+h@BxeX8SgOIvsju*sPUM*Qf$Ms>YSGyfAV2*_@KnUe?WG z;TC_FXKIQM5Gqiw<(`>qZ*x0L#7J-2t^QKWMgHeYwW3Mj(pnwuC5dr3v;1wh?9-`T z2_L9r(exr1RM09CewTwpF%v0<1AvqWMc2rA0=*)R#b5<6vAP?l%u|wg(xc-A@6t|^nv%JLeO{YgR zh+@d`2cu#XP9j45TIF22o#4xV&8J=xYv&vfB=On?)9&X;+Lrv)-2twsDIEQ6WHD$- z&$~lYp5;lRqZ)NMDS*M)4PpfwLyV!^isl{WJ}eK*oG%57m3e;n2xcKBQ3PvrZVGBC z{h$Ko__P%!MgLT#1TIswD@&T(5C(_Kyf(rd^6$%xo$`EXhC6t-+~QP!9`o(kCZikB z5p;+6n(3pPhm?R1znT)s&#K4`$o(t0@(AyxE74mM%N?kf+ui?eib}pbTRh=Y9}wKddnNP5%XT8!C+%r`{kf()D+?^`yioG69j5J z)Yi7p*Oq>D*2}klyzxB)%eN4PU?2+rvo;i+*M6}FjJxHkYqK_=y0QlsHGTdlO(x|G z#F0-Hhk^`-&~l@a0|H3wsG9n+mMct;FV4Bwk-UnjF7Hjc-X4+-tq^FPGYz-`ugc>| zD4gLd){rz<%P9$iYfd`cwdp)89g4OHHTF}eYz*R~_IdVyjd*)p6?X1eFMOr7s)oEJ zS^AUxZ*v|ZvgiqyYhSh81_(F<(0Nvx_Tv3I%Tp81kJK-0R0;L=8 z68~_qx6^Y`N!i@T;q?FG*dMD%QFgd^!SztJB;)hX75B>5g*(}bde}$brxDI8g5IEA zd;KJ0$?I2tE$L_tv*&}8clpi5zZrIxEO^l|bxz1SLGi;>XHCn{r;od9$pERF;*Tda8 zMxvx$ojRA)7-ecuuxWfB#*ytN5&wTH(f(mTQ)6|1uGw)WT1{@y5aMsdE8=z09QB4* zg>53jto>=jV~8nMit;&c{|jT#UeE*jgxt8P6%F%{JRu*#vQKe;Og3{E7KKW=YeG_0 zCPjG0M-JMj=lsp_95q+})!^@^S^f=torAp({J!n`_*XZLD3yBirUw~Hu%z+q@=cu% zhD#5BK9jEcQ=6<4qw}aHQhbkIBQT!toM2ZBn9}O+rVJKnM2LZ|F4O+^)&~KEM&J?^ zr^Uz&?@%gd@eu_9lkU5Lkyuh8T5Z(Bv#p_9{eDclLe<_gnY&ZbO&kYj=4V3;J#X=r za{`oh3yr9Gx2`mYm;H)IaKf=frd7+d+8$Pa0|%xbXMS1jajsj7$$4?*UcCF3gshC5 zn3P331$bc<|13n=>b!+R!>sNOG6Dtt*G{&?2`Z8aZGT3oav|)`?IxO9n6>*-&~}m5Diq{hIyWCs&*HD zGdZ1YTJgEvHSdsvY-YYCjZmOBPq*&80DeSxG7OLX0&V^F9p(Fp_Z&6;$6uu>sK1N; z7;(S7yFoj<8MnV^yYK!j7RPNv@abs+1GwbA07|%EmZ|KIe>5mQ{?TwBoGOeX4SyW3 zEPU)6gR~F=t3hi)QaeoNadxX%gl{5$JhKEn3!IBDakmo~`u|Pm7?s6@E1vRVB zpPssfc>_QY9!5DX!gtQw!;>{)E7O@oeOU;qC;?9`9$7-L&4U*0@aX77$eeviwf}h; z*pkJMQLVM$^Ki=h#DdU?cmHiD$jQ$tMC?>3hBDOPJCw&AACOa@{rfjR85HS%4EjF?{eP}OkFW#mmfQy?&6wV~S7l6Y=O*2M@-FMGbiyM^ z`F*sFvV`<_!o$NcdFfmHgbd9`;9@Zp0#vS$X;pvj9!!W~=gDU26o>u&+iNg|zZ!<|sBWMZCrxA2(1t>OM$`P#H& zzHF@b&6DyE3iG{Vi|s6ozB6c&=-5Z-pPV$f`m zewU~C{X=2mPWs!EqdyANg%nR`Wn}K_TSsvy9WA?m7rkz3)jQ~aj-{%87tkQ_V@;cE z2X}I|UU6@_VCz0B{oI2Ap5gJZiTkyog>1M#ry`6=JLk*i(F}&0aCviN_uI@|Y{pcq zugOonwqI}qmYV)Q?fg@CWkL5Z3U_SVHam7YNvGqi6&oFPI<~EjZQHhOqhs4PPtx!E z-}`&^xj7f#`^CC~?tXXr6A4ZKT%Mq#!gllQ^xneaD^#-$}9ri+fPK%XDBtgRL zSLGs2i4So@U3i?zSEkk?gh{WS4ak#9&`x)C@G=-l{yRh$@%XPle6h{A;tmB3jSTGh z_HH|+;EV#)^`0P7LLvW{yuo}%y;!LX-{IH_Z+F9)5a9d>A0Mhn<)e9&oc^Jp7%B>} zaQv$eGv)(-71{5e|1rt^L^0nb1vl1-JV+ZX*P|hCk0oa)!{j0ETKX83I(5kcmOfBo)(42I>?vlA* zzsluC$*3jZzSw!b8X86qUV7<(h9ta@)2USa^%zc|qf6thyFbpLIwt_NKol?d>$YcU zwYxS%)0M}T*i7CSc|b=2vr;DH#yLWy(@O1kB}#dVZ>+(5%U9z1HUA^l6=M2zhQ?Aj z?XRGJ_!c#;c0(jk(C5pw6ny;zk6e46OyhgG;>EJ+}VnMECGJzj-7s}(z|PX1@TWA=@{187m{F* zgS#@4L!YKt3LKX+yLp`4w0V1R=5LG;^3WL{gPdjRcp-JYJHtv$=;)B72pUfrm^XJ` zK?3268Pjn5h4%V4t}laLt@`0|OyM3CgcBr!AT8& zL3)@NcN zo_#0wMxz1@g%h;gPR~$CqDxg_YXjq71Qak|H$FZrNtwEzr-D1g@F7M2)ghhVW>Ih!ll}>p~-K~kLi=4 zNDaFD3_2`nB9>U5QCkIzl^F0dWltuGtK5(LzBkwt*AEK+NN?DzX;nyln}0ku4FkSB zWdue$ok6}W2HJtT;dN6%?`Hmpr+I0Q5E1k3HzhI+8nI*YX8zc+!QBecjk%kDT8~{Y z*oX`jE#HriIGqNU*W2LBn;CsBU;vHM> z8Mj1|M*?_~O{vWa!wb)(kt%6Y};W{wjDV|9F;ki}V zw@LzYj!aFFaWCf9K??KZ43=h^H=$Oph^C8h2Q$7duVv!$l zFZuj2nL4P$e!Tv`acB1Ej0)S(^Rb+$X{1c#Vb3=>2e&y->Y`^e>Rp>7SsMNGkZ(2v zM+**jqVWR?!QUizB*wS%W@q-?eMvs-8*}VI;6FnA>OH1WKZH@2D)nT4cMlV9cDY+= zwz5e!aC1oN4<|WA>hJCx$8I`5QnuV7K1ZLFck~*^|At5;Ksz9tI{VCCyjqdq-8DXy zGq5{Ywf*$4MIm*|^u}(w)LWy~s0%`)T>8g&T?OpO57u)&-Q6cIrgBs|s@p$TJW4p4 zRSa>W?k?;TdyneyZZ_S2rkwNa4x_cTOE1ecS^{#V5@gGDIv6z?PGeb(Cx*lDIP)Td zb*8fYDRiW3R~*;fwdbjfsk9k58^rSVHO*4hSuT7KdByI`Xv6aqmtZNW4#?-bbI-kf z18=VXDBKUnvxlJ5Dgs$YqsUtBI_#Gbo9rHg4JP81iWSQbFn(x%t~f6?xmE}$=PSZg zQ2g6Wl~AGicmMZ%Ddh(PYR@k@GHNzbFFT&i1*WA3U3wnRF)ZetIX{i6jZu{Z+qNfS zzm9@2SRce<=Kg#R*Fr6bHQDGI~<3|Q_z@(;dJ0@3Pmw-~3Xgrs zIQp~3lfqMn^i`|tfLm@wYLy=sFOdx#=4+%>fV8ON%S#`}wouxVjntRzw$5i)#kluI z4jpOj+cV6$-J%YbZgonHW7L&_`4)%oyR17Ru?*A`7uq=gtWm^NiOD zIPbYfP>~s*H5U6d2S#AyoI9?N#K&jqc-!;6gZ zN`{pbGSHQ7t?y1?ZY2uQh&S6kWp^X^%N=djWk?i%8+Y^h-BpxYECX3{$h9Vx6P>(?tOADILLJ>YT#)p(8KEH`SPVxqXHvb8C)C6f$eto{j(os; zC1FFnZT@MH4U|3i(zKI`ZJ{HLPW!JSV0-DhgHmR#+Bx1}g+l9iAqV!7Gc}%NF^5!& z@)-YrDJ9kWW_}MYpvL{6vN%EMF`R>C4XJh>7K=7BK7uYPHj0x%o+(DNKc zXN=e;?dYu3k+n(`TGJ3EY!U;B*6v&pSzxMIadeH^Fsvb zQP*L#1a@aS)}CS?3Ntfbq2W$_Llea$TP!$#YkUBYUVCd-be{!Ayk4X^Bf z2_`F4&To05a*iZ_0*;cpgE`)>-tOyPNXv!Ha#_nv&&^ogs5>{XyeXHp*z39dF^_Mc zSytD%3Lcd#NXpRumx430T<%Y$2Fm(j+_rnO`Yg^i`}L4XC+3{bz|Go_9-NFiD%{d& z-Z#JM?b$axPu9dQv=+ydf1?L*icOiDnO2Bx@8~RcT z0dFn4#MKi*V3Eofk`uof3+v(#s}a`UFBscs*cxqROYhD$yV2E#lFngTyrN_4s%f{{ zcLD}n_K7pEuLQQdNSe)d61QdOWb9CWm-~Fh@0zN0WX$6S6WsT4o>BS(5pa8dCX@X= z_z1&NwOLbutNrL0K{>F^B# zckB^?wJSy3WPWLy38{{wF6UOmEXTK97J&-qWM-ynMoSl5XZ@-?-BrBDEk+q4emd8T zD`gy2F~~=IV<|4}poR-HfA2E=*^fFaJRXJkLspIa-mNv}TaHY3M@(YILZGMA(w*Jm#x|T??CIIyhj~7=Ks&<`G zT5_ZWGV8a8go7DY<&06sxXJrM6`BuI+BogcgjJ!)$+OM5<{CXnO*l+yV!++$ z;qLAG@-w1mNwC3c{}AQs|6}cB@>(Yg?j^PrWJondgfREe*BcIh4cL@ z+AJ`3Ee<&xsHED5n;U zZ-ZBIt&ZjUZ|L)DX{$k~(wuynE?!8Izz3^=)e%`t_jxw_&o6LB$8%*Y$!&s{e&KuB zaN||ZLYqd@H43NsF?c8;ajlo#hofVhnrKZwk-B>6u61f8f$i<;8Q^dt zf9C2a40{_R8!QJMu(|aTl^Bz`4j)OIxqf{UwCz7ESkh~MysVN7dsshPkjqM}GoAmM zpu^G4#@lS-@92aa7{@#Jl(srB>SoPIgBEYjj6tP3lFHiXc@n^~r7!fbv_)Ywwq$Q~ zzP275ld1E|bCR9fP9{xYrX7lMfLF2(L*Rzb>iIl-J|P){ZoL>8?dE;>YtAp<((tY= zv!ffVvHY=r6uMsD>|SR!B#{F&tV(K6=eH6gFq)9ZCk&3QW=Gzcn*)a*nk;t&Pjf2? zSkakPW+Dn6R)Zvg$syx{19k^itr8{BagHxT?T%~QzF|_V29LDc0@K0svm z->1CCW*wXRZhZ^K%LN68!g&3_GdlK@5RXuV5W^9Fl_|}-^!^iVz?7+4GuKJ>b?dNq zw1XYb;dkDS;qL~5-4Ej;JqFEk_tE3t_q_qVvxhaMS0&G5n}I|YPq(L9b`RhBC5koT z8!eiMrPHKda^UB+-2IhEyIDCs(p?7mzG3Y#IE|I_?%1|lPG)-JEOc0s5(J>0!7+`0 z)ez5r8paCGmKnF5!1;v%bU%Mo!;gp!pxU+DD(5TD^z&FAn|IC$D2`ydl$ekS=?&Tz z8*?Yw#~!CU3VtKhL-22U|0`;VPUAkrC&cu8b7a(vszZb!TqpMDO)2eYy9W-Bm}xA{ zSV!V*DvZv5whpPx{H*=jsG^z_ee%rpLarr$Jiup%fm2&UTyom>0)k{ug6So8PTf_f#k?LF^9dfA{;`(v$ZLeT);1S=PJke5o_POc zI=ZwNd<)PB&shg{waDTWPmOy5&O*V$QVR+@rUB6W+NA}RE9cj};Nm|gp=#R%V$1h` z&~FnUQl52n(U&)9+$7*^2%n0`ac$JfyiE9e!}uz=MUeJ&?y)vp<>>wvPmBHJ5e!u5;elbb-g|%*-H3-50^<7Q%}qLa}CsmQ27VZ!$A&UEW}ouME%~>AHQnedaVh} z1mQ$5lEH*w4vfp`4|_ZS6M@?TR zwaPdC?pjFYjw>!qNNHm;RM+DJ*7|uYccE^lDZ6Taz6QKpO;*$Ij|cp3dtk{2&-jQ9!2#uXk~z+)Ir&6 zbfCeV9uZ$ktw$5Nm*=ie+TB~fx8Q{}><>B9n#c`mbtI8#x1RB_34*U)Sy)NpGi=5v_Hhnm$4-^*JN$*1m@76z z>48ay|H4p?tgQfBK%~D{y;}KGfAgzsN>?<9Y>BX1P+SgB#Vnvj*$=_$-SZA;pe=PP6vDiM+#KjKH6aflTmFwp0Z%^m+7;7##1pd)n znUrC%C17WR6N)-m(|-3?g%76ehENb}fj3H~*W{mm7!K*Iu(#8ObxiQx)fW3anOW^r zGk-b`ALS@{La10McZ2&g{yA7gx_Awux#R@(^c%nNX_(k=^1a*@oQ2V$6NAeV421P9?tUK(EjoC&# zd9Yg3qJjb|DN?egu8$E&f6QmG(pZI2e46=n!2VlXIX3@v$pOwM)LTs>DZ`>nTG-QQ zZ4COD?Rb*VXr%9`HtMe+TP}qz8bq0(305nsMq-xIoRQ-OeGP~)Y;&=zu>(iR;qZkE z`o1FGJbOFJID4;M{}nIl_ABVXv;pA^uoIVii?rkb;`n#wA~4cNe-ty6N*R(nmIKzZ z)uL2M`Pwj~oz-wjx@F4i{$M20QkMa)z%Cd0uoq2&=BBkP3Y0FDG)*>mhMw$6k4>f)eg=bHcW3p6$79;T{i53i~YxcN7)3)@+W9@>k_ zXom+s_-dlvyr1>NFrMUTEpJOzMq4}kV;045{oJu`x^Wj8TTmCFA{lG^^sBq;ZZgGq z13m0C<@ED~QjGO4v@^IDnw8NmySrpEg4X>KhpU`j0+EqgwVGrgP;f7i9yT@`!)B>!hU<7!gP~Wmkk;J9B2ugoeU*uM6 z%H8voXM=BN9ihbuB1&(tk(UY&c~H%p`a<26cqt5WQx==O;PZX|!;Vf*Anm#-YFZaN z6JY;y^j1};f4iHO4xxuMWP_*nf6f9>zAwj!%laSI=y7URiM?!xBC2xJU9LB$PjlGb zxWCRt9yW!3ArxuNwlwQr>QO+;5Jle-1+e=5!L*9J;p=?~w!5xHdxAwzM9!rL*C&8I zhprUnz3H|+<$sT21CiY0Dui%>hxn@Mydr8Um9PJhf3GB!{Qg5Z^_{B?`52x%@S6iz z3CN1}26v}Mw=l-u6k8U0v%=~Urwl(gVgrqnz7|xhn&nW-UEFI|1L7B%$Fw=wN=B-( z#T*&tf(>T$5!y@mV@>MKZ;;R-iKAe3WKDg4LYU~291)uWhDWdxNHZWfNDSb`p@P3N zO}N$(e|vTU6R*<@@`|uk7q<>}Y~N(&ZY^PxzU=imQ{oJ#0@Hu-h05z1g9w1>*Am z_EV{6+;{rlWJ9x2Wr(;6mr0o7A-v#-6 z{K{w$mkGxsYJO&@y&P(u_jg-D!g%}=mB{s*Y=np(r| zWYXQ~cn&|K4)iMJS>h(&^(45GpC0;cF*G$O6gcQYW#j&vbv%jLYH?g$_ogy_L3u_N z0yd6SHu9PCO6SsUB6TJ`h1Ov&10h#hf33j5E;vFTVrnCQlpmJ637UAbK$ppvU-4uD z-McxASqSa%scCJ)wQgu3^qiP}KzLtbRACjPg({%OWu8i@4Gb!O+lYzLMW-M{xhM|h zOD8BoFcq|}hc(OYg1!buyYQQ3^k1<0Y>2E*j07EZr;J_5{5c}`7nDbe?!q> zIA*nwFMH=AHgZrSKY?bQ>C~fq&m843iHQSL(xjVb^YKBA*LQei7=2#5eg#y^FYc!x zUCGDB_7)AP7rhtz#h4vA$rn(Du;xT}Sg zhO+g#Emvb2s|yk5^F0_0V0)?JvOlG>szq z8oJ%(P6@Y9@&NV&Xk`$Z=~Tb*cYf>%NuV~vi_uG9koXz0ANh$O7L_;Ve^{Dc7Uf%r z)sg*ea@`&0uR)!|*oG1-nwenzPX>yWn6#|Y4ft|KUmZK0wQ3D90#F^yiD+%#k2f0eleoXt)G)#9(F$U`dz>A`ujAq z*)uE_ejP(Y6dhMIb+3>B~A&u`kbiu&JiroGq4Zw48s;^3Me>_P?celZ$5$h5~ z0^NhA-nt{Fpr>xX4B`oOS2fgwD39SyGhif!pyd2kc#BOFSt(CM&z&wK86}_DYI2Ra zT)~C-D`6sU2=E<_(EX&y2b6}RGZUU_AFZCy*E+2Aa-B0HGc~I}U z$_WiXUOgW3i5W2FZb*_e$Yi??mnS-Di$i{TaY^`(omkcjT_yIvxiY_PP{Qa(Ft~i< zCZuZUDyY<*=S-mbD-%!hK}lG#)!%G0b-y$OOBaxzCR|?6e}V1x=5M0`Wa{1UesMa% zTyTv>_=-VhaCvTS+4zsD&&ijRPZ_qAyoRa(C(3?&tTU8 z`DhBT9CxjE(Lw`@0h(_wh}R<`+k_2*S`cXAJ@b*a@C7)I2z3F%?>{}zuZ<_pj4`Ls zOO+PnAs%{Ce=J*@JM}Z&vX;mc--{V3R1!;w8@{A%eQAb< z>$Hf(l&P$O!R_nH+jpL$5NB27C1EfFmXVbVKs&BmO9`P23s}7|%XB6mMv5TTLy=W7 znl0tROzp9V?lJ5r;~X26sv&nK2${?wWX<8&Akt_Mf1Ok*`%duXz+Jk^(3i#A`e}q0 zuB#o$+FPs_K3&>cC}=_XX*>_VMcNa5Ii_Yc68hSy>Ugjbmcq%~%-mji=gnt063C(?kQZ6C4AARc~X0|2xNhzjuL!103UHB_+dj@+O`}?kIqL0`oL#W_VuVz zZ_pl~f8MOA^h$wV(Q^ZNY%>Z1%sd4I=KK^wK6Xq{4(yG6R+(jIS|{Bn#n)XHwXms* zLa+p52scq4#7tRI6{{KrH#Kl_vU+rHO1Cz&XJJ~$MNKkH>VfqSX@XgWBHkBG>uL~L zgX8JP-nclPHl%|MHi)EIcY0OIT~It(lATu9f4(LF#MeV29jO>JAy|x6wd*ZLTWWUS=oK*l^cY#+ ze?&SX>;!fVOxe8d%b1xwUM?Y79pAdI+IZj7@s9|bW!&^BnLgZ3VXk)ZIR=QRZik2$ zKv_5v?)s{=s~f|R(@^@MD$5UH9Hd)`VC#&;6s!^-A`dk&zAd&iXArr;X~W3`Ysto^ zzrnhEWXHhaPOZ&^>`;1qT-#8+JxDYje{auwK+^#R1{)76&zp$rC-d87Q2pDO_%|DN zwWA?i=Av%?sX(Gwn#&Ep7(Nq*8P{-GcaM6L8h5)7LypCfX^@)h+LY*T%Gwxu2~h;^ zd*2>*OiuZOC0g$6EV8~CB(V4=9P^)j1)FYLGgUw;A4G#P31o;8 z4u|F1Z_H>G0?i3gns3A&);zgGvS&I5ornI`5i*rj?^Vcz+-}riryMN7ZBT1~=qvR_ znmzTse7tg6n?*@twh7#vE}lZ zbP7&pjS=t^IdgBfOkmAJSY9K88jSojj!L7CT|)HMc(e^xwzWBIS9fd?YuQDm$xQzD zKt?+10pc{EyGT9QQ@_6*Dgl*Aj!!7yPQ6qFNST^EDgn-i1R0lax_k|MfHh_Yc zcodRdfmc}d*sEZMS(C$df1!Z`p$T{*cRR8?Mm&K)k0i}Sn~O0jC#>HuL+$%HIGW+Z z%>h6^;l-;-nMwblojF(#K8V-?HJEt;`gh-%Q#2}>DrNFYZ)Y$Xhwd%SIpY2K;j;m& zCW!gLj40AyIRRm#aVihTkR)e~mw+GPP!hau`jU4!t!(rN>CxjPe*&dd2ozYv?H6~+ zj-#ORgWrDvJ3wFx1~8{ubDI~DM&8N$;&!`*z*P}LwiKdrLnGFmkG#(_^4}i0n@B_t zMB3tm+b}Y7i?*EH&ZjhqP@$UB`Jbf^E>;-g}zY8oq z{(^6})1%!GEv2psY1c!j4&{?k=&Gu9d~|rjn6^qC)FGmW70k3JTo5-A5dBr$5ONTU zM^x23pyk(Of4##%LmezqbibeWeQw@}(v*(y6;JdU{!D9A9ri&?U^bIq|9@e4Yg{yP zS5qbsJSDO^0Z;c3Hc(vc=feE?j!S>s-=S7FQ}LYyLqyw={kQT`0WY_$9R$`v-)=D? zcfF8p4PFWL5b;V~w^OAAtXS;Rz+F$P!gHFf8P$65f8*V&Z@k%thZ#Uh0x{TFL#RYk zHIG65FB6XwZq^o?lWd3}>7Eu<^zJ%yB6I%%+$cc20M2g#SzJ0Bf=HK%LGF{z)6{)j zo6y}nLz5$v-)jk{zHZlp(=)Le;p{}7lqTPlE_!Xwv3CbRq)N-kJ-cq79%2xTc+?_( zcun7xe{pJQ)LP2nm~W-%Tup!I_$|!8C8U|q_{`9QGD9c^gtPWy?B1ipJs6;+zlmF= zQWCAa?O*Hq;(2mY_=&&>8k}@1XBcZ&VV=S@EN_|xMVc2>-{;+tmdB3536@lt`xs}g zWaz6`koPtXgcK9mUCX`u;OJ}LJnqXxh35+`f88dmQNhrtWXrxon2on8GtvjVtDEGJ zAK^XXJFh%Sw?+H{Cx+w~MF2tV9EFvf_RWc6gn4-#*!b%^&{xqzpNvh|qH2 zf5ua`q}|TAk|V9Fy{Z1?%%>CD5TQCkC}^X?aZRtxs_{@4<*mY%s5+tw=Cz^4f~L45 z^nJpWkj2$wRX>T9djAo01>~pDA+7vi<9nu^N0_+r?UMaYd$Ac=72rI5t~mTfdCm7n zC;C@VY@?U`5AD5zxCJh!=+#$JK*evYe}hE0kbxc&>>tEZMsCCnXz9j9D9t0;K)>#M zb3b*TC2C|?`Z0H%nHlIXf=ixe?N4W;`EAICzNyDxM6lgWTc>)IUZZ`wy?L+rD@LjC zrs!Hb$#RzSwNWm)T~YV&z%?b#n;3T zh>^_9xZ5KoyNE99N)yl!+YP$6iS={(WF3kR+3WL9_G+d#mgL%6{k2;CtN+WRh~N=s z9s#ACoI=leiMz;xWh%F^T647}YUHrJZFe8z%Zd=0ZP)?c{J0o<-4N;w`6h&*EIxu5 zG0Wl@mG`sg3;F6vnv?JVe=OrJIsXK2M}QgCEbnOm?-q+pr+YL-^8R2Q2YP@v8hVcb zbklgG?H_&v#{rot+XQ;C6?gS!U$*~WI7E;B5vUF2cM?-AXrKjk(Is-T%RA^&Y0B-l z%=(*g zLMxP=WXu4Hpl3q2Sr*cZ>riHWqa^2F4{(7^-w-`zLF-DHE<0ZG2s^6h8~eb-q+2HA z(bzbYEt)QIWxHJs8_bwPTt~v@G5exS9w{-^EY=T^lM+|BI^xlhx3E1QwP@}~RNA#J zwmmKv+YD?HcNy zYIJ9QV$*gxM$iydO`_&WO)=PR4ND{5)$6S2g@H%SNJ^eIDWMmCU3;98D-Lez1g)Yi zFkJg4?{Z@i3H5GwMttXC>O2r?Fdf4Snh2TjPO`tv&<$ zqIkNXf5eAeq**cPzSYx=ymY4;+-ZnL{04%M@{^_aC&hqzEC#tO9NxnQt(F({8OjNJ7?9lWd*PsaNU@N#JAV`EYJF%arXm+yO0H@ ze;U1yZu$Nqb=7GnDRMr`J!X7DR(T6e65{&EgYz*+pW?+Lv_!DjA2N&*_hUSK$ZuYt z*9<#L(sxkLq?-WobT$&Y=j4Gn0CV`Q`r1=24cJyLMrvTH{&YNpd225FtW&jgll!wn zie~BtJ8j4RXu0n{cLf4T(b2PYFJXdPe^_+E5a*0Hqfr4f8i9aLG zI9h3!;Ic|lhWI})s`K^n=*k{_CGy^Wg_f$$IQc%SN-BnGAZR;;TUPy%egM!+e>sG0 z1%!k$LRPP*ujwadaLJbt{2!ExR`h%n2%7)t;w#c>CFXjN6*xEncmJ%%kd$Bvb-aDl%B^3U$O%!)4yM(wdq$g*lQSkw)@-^xaZ zMVSsX5;HutxNUvH)#xW&Z5@%XQ%tG#$}Cffj1QRf!%j%$o#9B}7jFree9E(|{Dc{GKOs+z+e_@KD;~@FN za-gX}p9w%)z2Qc_O&0ETmN%d1d9l5r=-z@*N^7aeA@N`NJ)X6kw@TBuhiLF^@Cz^b z*BfL;%Q6yc38aCoUoA|=3+3;os2VhFiZom21YBKbyB-GTpm#aT>1{!AkUP5HxAkFB zx(dE;NtrssN|He{UNZ`Wf3?O@F>8WUpXZ!YzXuXJue4wN(_Rl3J)WLsM(wJU3)n19 zUWYbEG@@#+MbgHDVM9tgd-+0B&#q@g#&Hgn)PJgd$OTn}&!pFM-n~J#chp6G<_|lv zDCc{-nh0PilSiXhuACbZ`QiJoo%`bq!ff8}9tiYIZpkA`5(Re!Lq_N8tJQ+6fiVpt!(wvtY9q$fe zKQ(jh-G^aLSDHqeX?iTz-B4}*Yd`b<*-uGaR%%!D)tBVbe*-?*0z#C|?VBJg3L~=E zcW+??j_fKB>%LzdeL^_vZTlYb7_SRsDW3XONFFe)BZx}&Jae4@= zBj2c-fAXQm#5#Yr&G0x@!{S=cU2(4utNzzGHm0A|p=%d>vm||x9M554-mT3kcYeg= z#>!Koj;h921@l(|B7Uf)a-n}}X?l<)&^leM!ujdTmCf<)$a~wpAoIT_4TJ+nOto-< zHe{LUC2D-ws1Zv-j~kd4?Lhycjwmq%oasA&e_M~uajiMS$k8y~ZcA}F)KPTa+MkHX zxbdx&T6dNmw^XCj3qtbY#9CuJvQr#(ySvZ}Ekl3#epPMt_yR_?efLEHXTw92VTOU^ z2tm3~f7-~foYv}!`?J4fau&all;5vGOeK(iFzdpT(d?MS)ZRcFic6s>qRxt=VKYVC3rlAb{vkN+S95ma(3;X0hTXl01dP8>==2gl?#SZQOkp>L0}IpM*Xpy#!0^X7G$ zQ;1shum53{f4Nm~vootrooK$Dq1WQY`{VxWm0Z3nbqyP; zq!~m$?a$2!X<=Gnidz-s_6zf+A=nyH#Z*x?We7H)h#}VX$frfYoNK3!%eyxwg6FGQmc$W}upvghXYkUV`&K;Q!}-7I_G zAoC6TLy$Y@?}3SNMJ-HGCx(g;LL1}8c(|f!WA8Znb!a=rnJpo3SQsnjBl9UPdkZ5% zGa3lZcF~lDHa*mvW#boyTm^DBYoOtV#=3Ds)>m#5!TARfXcJBTch!j>oR*;)p9T1m^`};!=aRs z5JWStGr@1i64XsZ%guE-fBePS7`>%#LbETp&1t}iTD-GADc{kM14U9v6$kBqOyhU& z4!OukIYwI8Tde206qwH-zbJ^~61Wf`R|AYrku31qZtrln4C5EV7$3_~)XGe4&FE&L zN+D~_tfgtwzk5u4D3Wcv>#M%he|ltsPX8NEN`qOQauzXGEXW$Pe|nU#%>d%5_}AjF zd39UEQrGIq-QhDP)QHwrS#=i$oWx<5hqLP|1{=oTB7Dn+WFcCsk5{phK?&L{qxkdn zG9<~G6OG42Nh-0WoFn+kgKT{=g#wWd=C?%Z`hsd68`E{5j3%dx#eR8-w1IZ zDmvD*+45)Z0E01afA=~_3$07_GE@!yKWV6ck~C^V-zo8%{X769y98#|{KqYgsS3HS z-ETP67z19?<)RA^`dQD}&#_0OMyg{4&s#I{-3^?x4GvjkSK9DR1s>IttGsSRE_%hq zXob6Q(eLD=K{Z~2SqOnj=JK@vitw0o_a6?aQit>l>ou_Se;f9~yP1usU2l!@6c6Cr?22pW*APKpfQ$AU|uQoINoyk_A8`=;f+Fuvn8ZyE79{!@5jJDH)KWo zxp3dQQg;&Me{k2-5=DO`ous@sK4NSR(r$n5S~(g9QxLbyvsWFJA6JY1kz03S2X%LPdqK%VlwGbhY&BXur{G}bL z^>f3zO%#jPOe&!jT4+whHD;^PT&er~EBW7ZkMVWnuB+kKn*TyD?R@7M#vb2JLvu_ zG9B+{e^A$p`7(F*GV*@ShXExrzE8J?b*=Yq#h59!oX`pKY6p7qD_l> z-nq&mO`f-Bo;&Nd@H0@Yf#}SYnq?!G5|OZyjtvLubst_GEzMoI!Uuw{HGwZWG~;gV zGr1FtG~?7W*dA?DYkjV>oxSN*2_#_KB}jgUe@i@tuLo+2boJrx-IeV$FBhmT)|=xU zpK%Wp*c6@0slA%&{0&^D8pGh~*(G>jKCDli*XwjcHwTh-3SZ3_V8D9sBH6@POnQ0j z{5V&9)_xlr#`<0+MVpgWZPr+j1*3V+B}&b$%*{bD2Y%vztbBI9R?yHQi@D6syxyeT=H=ebgQZ8*5YCMpezsI(! zMz`q8YSLD>1QLeqhZdJWB`_4NWgVC;3DE3QGwLsu@Ddf^(1||-lJVM%cWN->f9f?& zw)}DRQ&oy;LpuP?0g3lB^+*X6lJVo`G39;w=FC%}TaSDu^j1$KPI|c%7+KRUkGc)1 zfze95__Hxt#Y^N!pS^^OdM*{8jT`8oNHL}@T02YO0BJYTVL3o7omiC>ikEmE4M*8{ z@Q-p+ei+@AE${6-tzX7znN4oae_Gp3A%u*aI=erdXv!};BLT$gJ8nE}h~_J)ZqqPx zc~G$MW9yQ_DCovRFXR5)>~>&YOpZ+>32m;mw`y_m^vd z5Ki=S&5YA_mmmC6%bX9-g$Pis07*K<78a2ZzMosMm)+;NR!(g_GW_z2qo`X4+tGO= zW7CFoW;Xk2tbge93DS5>e{P&_+!0BhAg3jNLybpNsBNi0iBeA{y+01TCXKF0?oKCX zLifd^PyWbZA`l9WDr;;atzb2Z8X1fj*8CwUin=H!7KSf7$D%Hl?~OHW824W=Xj>3Y zc(Bo-6E$L{A{OJ zK?Q``eTshFklMru>)%{Mh74C7``%)sOu4|#D=w#fNY#c_VOS=OAi%5HjFBs{-*#iU zi^z3uDqw~Ev~i2Ae|k2sL&8>ujthx5T#hhBY(BXjBwedJ1>S_aEYAK&(j$iG- zCQH{&0q2#1f4m|leoYZFAGLt(UhPx#KN=vk^X0L$x4Jxmx2yAS)!ZAppXO0LPhXYESmS?Cnp#;7)7-nIns=7)F7whizhC!ksTpK< zUV5Y1ovR#o?AuMCG&F6`5>brc=bq~K-IY@%-X;D3e~VJ6eIk|fNIMSmj96R6wNLg3 zK}^)4;^+J;ytT~A^ROW@O31#+;;9*=qzadY=v`YELvC_ z?7-D}jYebIGB;3CEc2h`ptVvJ(QhbR8_0CXddBdi6E++*;BPW4zVAKKqD@;{VM6AV<{0JzkLZub0`yU5zd zN(!oTukIP6VI9n{VxU>SnuK*&3!4wfE&}GCz!}57+jG zW#~Gbt*BLkla6IHQ?cwgj{b@}Di_PgXy}Z*f3(t>LC>NDBw27+uIZIT&2kdryvHt9 z*#(+_#xK|bpY3Xf=1A$(=EeaL*8a#8Bo$1LjzjVz44@2O{82bsPXTr3xq{G_>y ze;4-@vGwl(3t!Zvlo>OzeTW%^o`(A-RIO3F-`?Mz2_PW4{+iAwJ;Du7M|Z=%jkgJB zzRva}91HLs(K~ZMDd zEQ4cPSEn_9|Gq@u&@Xh=5&gWAON^)pSbJJDGXHSXdRRIp2Upeiq@-!Nf8e}=wVgql{vnt+ zx;K`RM8`ygQF1^E7-VLposrVhGteAD?^GI|*D+^;ZWV*4I#wKbE>n;&B{z`?9C6!q zSI>uCAC!4KvOs|?h^fl#m1r^|5A7*hFT8+Ln`}(L=DY#Pr~_|0{4;YFhQ}E!Ih+sj z0~sXvbH_Ym@7Ofwe}&vifQY<$*3#`M`*|V+9M&R_rqvpBlle=g7fGLmVS4^S5<(EnBu zMyBu0yg}a=KN6ga;<2-5yNxOiRHZIbQ=JE2_}Ars7jt?$@Xg%j$l?66@owalH48s% zl8Giys|c(%Q%p9N51{Ijr26TJK+Rod3yz%V2o2ajrIXThE!IDwfz+htbQbBoH#<(n z+l2A{vNIEAf3y^;cK0(wn|$Z&wji>F%jl*8qhyUE1e2)>W#y5Y9Du9kBeYT$D0!?+PfBgFxu93&_cokKWCmrvXEr*La zDTm7i1)=JV6T?`&Wm3Wo3h@Z{^lU=mNb{tMh4zqN#)kv3a49w^8CHc@{x-Rz72~vk z&~_kn3Hs+wWzB<-JR8it5nIdJMvN^&iRVLx*jN6(x^JRYB7t# z*x&K~=4z|D+PpWeVM!kyHgY@KuIyJFD)1$=azAEgJUng!Wt~5k)P#YLt8@2vs?sX~ zuS2pJVTxC7uqMee)D6`~!oMUjjeoV;{Ifwjf4Uqvo}q%(CR9G9vRnXcv|KGVB#^5s2TOrZT2x16^&cGDz~qLIZc4pbolQjIeou;qISCCkv0ydlZ;$ z&B${HmUI1Q-9UL2Jxderxc$j&rEy3)>CVS}moT_Em;*)#O%Y{yzWPZT1qhA^4gT%> ze>q9kb~=BH$;>;ssrAAA`h9f%EkwS4po1v(%X1E36mEwUN|}Vm!OR%&Yrb47P`Ys) z-GrUY;<7WS#y+{Z_M!3kF6F`6-*gV9IiFe>g9e(LN}B>2sEPC5`~7$tenLvluiG(Z zuzLox`o<60_zpwTMj^jlm$MUP{y)o?f4qn-4(?)bV^oLxM0@n8@}9SUIrr9y+|4G> zn5}UQokgDtB5DKz+`Ohn3{G=N(r^XJ&GBY><+g&6KE=x}}$!Yo@)2`T_Nr2RV0DH+j31??eMJa6P9x-N-&kRr-u|XBok$ z$b!1nxTn_mad`+$yd_%Yv6$7CmLdW^o|s z%$#)4(6fN!1L|+?L14NHs0H*Xo9bR3(2{$QI5@RA>jw>Ksr>>{Eil(fubbS|9o^6x;9&Xu zB31Kq+t>l@ic}tS*ExD*e{6n&W^E=Wy+#mF5aaC&b+@geyn)#KCo+MGfgINsjk8>M zmWuK?TZKy;WTX>Xg33l@{Hv({#S3&@z*3WK!s9~>;E;Keql6D?S4gLG<$c1HsQSer z=SNs!gqGSL%xr$)GUs9`ci4^G)}+tXR&jmw__U4U-D{?H=JuH;f5QUnpO2?VV9B>{ zPe^8n*6|3{R}j|6(ZJK*O7Ri@c^%$4HjCLv9i_(pDR9o}m04?yEM(4J94V7QFR=`YpPD0*v-@nd#e_esNw>u0?^UAe?(x33>>Z98 z>~}ACG?p@Y5^{&Ee+NBAPuxc{Y8F6H5yEJuLt#lCJ}dPjEM&Q| zRh#3=BML3Z++9{oL<@?x-^d zW*=JYNX;|CWpo&mwWrU`=xzhah*vHzCwqn?1nI>wLTi~|LbV!*t55PvONe3N%G5wR@ge^Q0Wu3#KCjB zp}utH_h}iZ*UF|Xi8`IEwFS!l4rk@E9wHf9{jsv;JrhXGssz3U&?wxpg9)ccK|&!!eImf4D1dSYblCD&$Kp1eP;Z+GN&- z&6m?wlNn4_$2Z}c1j?s2SI-7(cf5IQQhZ|Bz|nYMyMfNOl6{mk^-kC-(Jx!yK-=tL-P!9ZZRn_joU^_T;-B=$_( zF-~pfss8zHMux}JBA*B&K}Vhg_fDip8B0noj@t8&f9i&0*?xQkZnXJ+twG6rxjqa> zVs6+Cj!~C8NYmPju$z;OL-QvFos@WXgw@nFe`=VkA6W!nG<~DikrL3$FqW#NYrc?-hg&j^q3Ls5G}qFhjs^4lyDKtz>pzoPR|x8h*9dF=iP$yeKyt- zfBt@TJ+y17zs)GgWY*}%wjSg7A`*- zd!CbXq5ZG)EV_m?Ow3UxP;%)4?@l8)f5`&0RPTge;!-v;ep^4uRNp zay$BmaHLYMblclPYcAqa?gLZ#kWbZ2t!$3%c_U8q?F2t#>y}b>tBv>r&S0Qs*W!lM0hV+5cVmKpRH= zPuIm!jM_6;!_^Cj0HcbMCrsb*7tVg3uxyE6(ovp@ zk^1$D0@CB6sYomMLRm`)oi{i<(4cCBvBu0|gsZg@a%@BMaoN!Rz0>1Ae?@UzJrsWs zZJ6;)f;?>6A+I3-Z+y4t>5WL>alfgiYlr5UarSm?rc|kfR?q#p)POwHLwZFDwA<(g z2GGCG#A%2%={F3+k<+cVySI$})_rf^8)a=CGz)S1-dZ^9Lfr9Tt+Hzdp-9X#;Ms{r ztPf*YW8;n{Isj;C?=RO(ee_@tC_7`Mh#ET^3 zf}}Eyky03vv(K{|WO=H{EZRsu554_?lkI|LZPbQMopB5 z7VyFVj#_ z?4Rh>6D?~cbYno{pTa4Ytk`G{yfDpH@BG!O&HUNNBE;K8^$OdgW)Us;YkDhWgMq46 zBFq8xdtXGO_r}6Gcpc7a8@IXp)e>f8X&MzuKlqXQ`6j96@aD9jvChKS<@VsWd+L0i z|N6LA%V2YNf5+I8<6Iow=p0MV55$`5=+|#AApFk3vh!8!NdXABiEI2VNqYu;`5aT5 zmf{SvM1>P)F>Q5Nop*%R%1#iYm z&|ZtOL#W5=m_tq2UW?I%w@$+7U>VW>^JVSWf+uyiW*oM`St{s#cQR?k zC|&G~HAfg#XTt-lfl+7$`Lx$rQiWo{?N|~|00D2Vfvu21%_WW`oN~~q5g_j;oQixL ze_G0dyOY9^o5%ZMFJq39>t#eP0GjbYHr!gu9iA_{a8pz@cToKhDn*K6b!#+fvcyda zraYyKlbw{EQ*dBSpsiy}>`ao0ZJQI@wrz6<6Wg|JXJSlj+qQlGdtdI$c{yF()z!Od z_iJmdZ;gI}A1G8@s>648@6CX9>VLWJe6nQme6QY%3PZ&}3~^~#2InCRGuz7fU75UV z*r3gA+dso0M=c9WTkn>>Zl}acV(h{JIr`_#9H7(vdvhDi=NCJ<#pH{^3>;}zVM5BQ zES(n}G!Sj8GNc#yt%FR|LMU0fsXn5!{aWiHq)j7ADWwd>KeLi1AD_nYSj>^u_jcvl ztlG_>uzfTE!QHmgcpv$vh|@Hwuu=XZg2?d$!A3q*8X%J7Jw(nF5*v@>>B$p#Zxnxm z%RGITW-VwW3`94;TKecWdt+ih`ZL&&RJMN@ape_DmTAB-V;fN}kfd<=03&*cXMy!y zh0Nez_)Ku#Vd(t29PedaUId^X)aH~~ybSsT4C05Wu(2w%0J04DukK|)N&YBf-*t01 zS->%ll8aXz9U;Nq;pOUHMF(Kr8h2GRBdMp32zD}sh4Jg19Bs50EU4coCVx5yZS?H` zpwRV)@H6!0D{XUU9t4M@Tx>Vo{WkVJ?~yLW8~QussnyR+JQ5y1e6oU4)Mwt@u$D;$ z@~@|?q2z|I#+(ysS?NV$+!rc3q)51A){FC*@LFv=xxYdfi-dORofE*Ewdppwk zW5HhTFy~lFmz2?;Cu&6^0-`qn?Ps@E@31lLt!H9spwnV7$4clB1)t(H>m2F#0ec}t zs4Jnl0xJ`)vxQh=d;_>jFtiM1pj1wix|v6U3wYtUHS08^$GMfaa=|6tUkm?;MzOG# zlbrsfXP~10eemivK6=J2$FWRKA+N!8+BS?JR_gWRK9$9*JW`0_Yqu8S$3^vW(a*gk zuRSSVo0uaI^4I4YqN`O(G~Az_Jy-c+RpmbdD0&jI+4-6(ug(EMiry^@l>!OhYq7CZ z28;1rbI=3nQpv2tR{LK0W_(9FAiR5wVuJgpBRdN>A=^C zsz16|Q+nCpeIt(JI-Tu>T=<3`2%GoUQkIjI^QDF#9G%VM?x3g3sIKk zL%c=qt%px4+ka#`7OL?FS$wKueBmm1;A9&_|As2s*C7u@nRal_bMd*IevV7$ke^vQ zU<)^tUYAkn#6}<9TCB`G@Lgmr<*XZU#5h}p>?|TQ5dxvvc#&C|kS%1`#Tm~niwg4ko zm*Ro?pWo?LmyzF*hDcOx+s?(I3=Oc)l|Ap1{?GK;>+}1uZ1%_^-Uk$kgRcu^0cN8Eb$1E_ zVU@+mt^rP$dEK@tr?xqWHZJQ=kywLBliidE*w>zIy-9{-YV+y6y()BjRYQwXy)Gkq zd}O@{jEF+&(>tv&z&_6>@14r6%hmZ1VXXcO+cP=LRGV%^yc6DSKsL4S$h-won+TQn zJ&ncGMIyeG0!Fm$bJ@~IESO)i;a&VrN>loFNME?;U#Umc>5)B}+NB~vc0$F7TTr`j zNDEtSnl)TOz0W{bL!#y9i>aPGCOzx6>Y>-Wc42U7=8;VT5#L~yL&RCGCB$H2Q-7sG zM=Q#v!KRiQX$@+`P@gPtv77++hcGg~5NPgk|4rB(MiLY3UW+zu%{xfXn+v+=7}Mm2 zlTQ_@HL!`*3s&6z2qB$O)4p7$mVM&B6{{5*#u2E-Es^z5G}&jnrM*I_mHl1ES7g)f z_meHnY>52Tm^`eLAB5t@9pcU!c3Pi_4^NzwTktS@8%z29&U}GmVi_5L3bJ3gdWOg< zz|3E#8W`m!FknU+*ozgq!g#4d;D3EU=`B(89k8+4Vo~Cn`HIe_MdLL=J>2YMZ5;(Q zZ}f9F9LkZ#gAoqYd>|!vgMQjx;$Jtyne>HR(;4h#w#>tBa7kJ^vcN1 zGw+O@HowdG*o?X9uuVC;ybsdViVe{WoUpd)18QIyJQ!8g@mm!D_weWB>Eb%gt4}wl z$8qbI;G#|`pZ3~bQj0Fktnqc7psk2{4flpFP2eS_;=;;0yB^`FqCuAE&wsn2%7wC# zJ^lioa`Nhe=N+H;{eCZBP*^J;Vzj0P{92S4Hxen=e=4!068hDp8Qw7DuB<-hPhBbB zP2ay@c|h=sqRPquiu~E}b(cHYX_5H*`*IGidG@7cQSGnPD|7D&C?Hp z>C4henl--g%fvc_qOo7mpsh=SP~&SkbzHO=ASf0jt4A)!VrHff9l3`K-!kTd9G293 zthcwK+9kppY5_bd=I4>z&_HiG%!?M$1-!K9_7GqdcC(!OLx?Kq67y^k=^=v@>!)63 zSJ<-G zp8$Lk7(bpFwh&2dC_og@NP;kMcoVqzF zb@+f9b0V!S6j-nyMVJiTwG{C?!_Hfs1S0p78X0N9B1!%hAWca4Ax*8W9UuPbs>N!j zL<0&|-LcJ?lheG&6~7$Bm*mjWK4vQUMDw~4PKHDjE4{U8ckTumJ2U5{qd_)|gdX;| z!!dV{33|LL92T%2vNnmNwR7FwJPvS6BOfkhlDZH<^CZqX zKCD5T9Hc9C9aMdo;T% zAz~2ku7NLap%29F;GE!>Y=(E3Z*W^Ci0Le4p+hyi@0^u)W~eCMV?7xJ6Oqm*dM8>kIUs&CY0$WeT;b3(+t?#W$e;4{7G!ZTHj-cbO0=Rf zS@L>MPet6`Rf5PV&6^HgRYKnN6Sbk{&HFRQ1%G@+Q~9`4oNB&j7TrrYlWzN!LRY`N z9tT7KDTY1-!v`pLbUP>KP#xhD=Cj{s-*a?BR&R1<`0o1V@R~VBo+*zF0O%j&DSUY8 z?*-&PuOa{W!=uwR;f}*Y8EK41VnR{mBq;9E5I4)LecrjuCq~-ao_!%J z%8C+?MI29-zP6Eb`l!yf{idAp(`)KS&bwoB`cALu=QvR`7q)K>VR4U^G=+*bX@e#H z>u{HDo+9^bxP1SETA7*?0(i@sZrpj=yQPj$usyH5{eHZL1eA$Coi>q)ZiZ9Po5+*5 z!v_(ck_=a}&WKOt*E~CEO*?ayN@nX0Gze3z;KEhH;0fQJf5fM4)fQ~^@J0@%;F)E8 z>Dy~~eqAGANl9RcDq|={NKz%+_347DX<0!w97%V}Cl=yhjYiTv16eH9R#jE*eNIAa zbG^61;D1Oytu)^0umkHQ+rk_6Y?K3b>5@Rm1JH}j1mH5oBE~xqGD?{{TFNpf*AWIdP&Gs=~3$W#r$Rd{EMQg?cDj80upKzA=G_0F+F)cy0-0U}?fNc_7q#P5)Hg*O6 zZY~WTyvI}^z&{t}vq4{pNi5k=Eu8Ayp@C3@YIhVrWF-9^U(JinnM>jMQ8*Ts(OtIu z`~J)@+uF7U)ry078C&_kh-`D4F5y1wOsH1TB_@G8!1BK}K;D!iopWWWky*>7W8 z-ozLR)ZWH^;wEZf@k&R@6A6XB50<3o30%=Y(oE_*HWdP8@&i$BV=pU2L$=@Stapjk z1!5u5r<

IR#zYoD+DB0h(3p+VKjqq`B$noG}`@ztn3KYHmterOcXd&XuRKAK=r zTCk4;D!$J`Ge7FuxHB&4zLNj!MxTE&c>~|rvT*Z(Up>sb-{*(8{1!~CVg?@X-{( zstPO!_=&cs<^OH29&jL_5RafBApdXXi%d^mmCN^$*z9_fNF~;2R+T7S6Y?k&Vq+@r zhA1X@5^`xt@+R-3fKfC%9A`-H(TxIBr!=yrcJo z{%#FkPD$7BEH5StyF07N;;w8nOI)Oa$M6?~U;(nSyweqxwR69ibG!t&3E?-!R1>Ni zw{`Tlxn2IrXIn!PPBESBOhXzI9FlD3_+l7B{?R2zYQsHx863br5hZ6S?pC&%3R3}+ zZS8j8eDhl?!pB%Hz`z!Uc?X&O>$FddZ+P78m#*~Xha(Xi3kM0BRKoGyR(p_0kP^jo z{mA@E5xs9B2wlPn(YUCUKc+-h2~i6BBFeVBzw+D%mNM>Ie?v#vb%@(ET$M|RO&$zj z2L5`^m9XY0^@F!gMotU06O28Ba~uGy$DX`Y?-|fZ7LLdFpL!T$LTO_bVtpPYd`&Rf z8^qhm_qeCKk|Amf#wYi}!W&u?N{!n5R6TSfh5k+cF;vz5>-FUbEo=7q*rix&;2GVR zY(-$c1G@?JHNJYwG8rncR2#+jn`yC?mEvH<|MKBn4_e%9`A1Dnj@GixS``(5pVf-v zN2jX;YuD;%Z3z>e;TaU@tAm%Mr#y#ijyRW^UR()sz2;NtM;M!*z9H^ZYqjMmQgkOI z$kic(R#3W~^i-u9QLFkhVa2nEbXN@woDodmh=#q*!wTi@M#Tp7sP8C}cfNV`Tl#{wFikgf*zIwlFtf)Ey7$toLAM_0R!2aYssJIV`T)6|3KfG23?+y4WUE zZdFmwz#Y8m#Y?}baj%GqLzQh`x}ICmh>mqH1EueRx5C|4|p zlGU`>lr*C<)coCT`)p!Hr!>wJ{ z)^n9_A$MeV+7%iMyTbKG$8X3Qu8&}6S&?PW%v*{fKOgt%LjfXG3APgm8c02Owvt%; zBfP@@@*zF!T+UqK6%hl-#K6Zs;Xv4A9IrtDGllYUVh8b`WH6i?4|hlBdFP#ir7PlMSxNu%9y%G z5{%r`e#`&jadI0cHP@}K-n+LES)~^>!gjLAHW{~P%++Hqa?0yd?!qX3^NFy0Oku~sUv{U?=N?e z#NFkwaxJJfmf)$?@UA zIYNpX%|R`#$@T*v=P~JF?nuQ=woZ_2%Am9~9x38SNNJ85r=F)=((vZA=d)^Kxj!=2 zZLpIeab~@XE0>HL)j8^$h>+=#dz-qG5kkXcwl>$;FZ67X9_H3$BvJ6YrhANEV%L%F zw&+&bj`mJ}2BZ7^evuQD)5XkC!G&Yep7Fy5e=FZ8LW|z&M7%h&l{HN=)(V6 zC!Alt!|kq)@k8pvRuc%-xxi zez+x1pcAMqXzdEnsc|XB%|29gR+D5HA6j0V#CHUYp5+Egod&%~7#%g7Z5W6qZ7IGy zZOyJt*6wfM)Kei!+De=->}iInYW79FnBUXhopRdvdoCheE4@>a)gzcZ8tx{bp115O zZ$+Em6!dhZ++Pl6JMLkhp2Z>ZH7x2JJ-xRY9j3z|%WswVh}vh266_h~7jBMAZ3`7; zDP(~BO|{J@F7=2>X-3h0eDABj;mJ{S6tvJu2{Epz99psEwHDL0PsAh4W;iJgsh=fG zxmIc_At?on%^ezsBX>iA?brmxk;yA_rS7@SRRJj5$drFvMa_oRBW5FbaPe@(G81vN zaQided&uk6nOK;2S0ggE{L=W|c=>f~yR3ku2hC4oC*M0Cr_IQEEEHkgvBq1~v6TJD zW%o}<;eC81T|6n5*9Rr7@rg=`XW?0!t&2%bN_kLuG!e)Ilqtm3fC`xOp3RXSh^xd_ zWE^CVUV`>Ya8jsMb!toURQzW={ z+Cj_V8@X(^d2Pw7(?`~~WDRejM4S*$A)JQMaOJag$03lGu(GlzO?yillW#@ff4ge%YY&!alC zuNO9kq3TolG3TMc{b4$uJF;y0_Umt@PY+KwE(#G{WHT0hDH+%c-kuK^--VS0YL4!z z`l{*c$v?r49ei$o96FWyq+bIAo85;D){k2P_}k}hm$T@+hhg=%lTn0uOW*iI9|!&_ zn`1&hjw8xP*Ek@sCRYP8Z%B6hhNEv8?mwGJy!7DFKjt8aokZn4z!6=zt3iU;Xy^rj&45Bc;zFx0(5d55(^JN-5j7&{R9LN}Qm46z{P89?%ZA_AOM+s*8kVBj+cm}rq58ok7 zLI`lB+5Yz}`tA0_T*y4O1U>wZEOkA*Wd0}qm5&D11w^Q(6&7~l&>jy*6bo=q2On3n zyLHZ!!dAqnO4&@3m&P)?NaD{98H*4Eaf)-3csaJWQ8HMJ1Q@qQCi zGAfjWoNmBQM4Ld8Lw@#J@@uRx{I==2w;jz9g5&0?x%$^G7z$#*N7$>M5453=!*|>U zFF+gFAh^;3O-@3FP)JU)hX)X^Kl!8*ea)8JQ7Cg=hTjUJ>gY1ri3|#0s0V*C9QOoU zWFbM>mxLE`@=;m;{zpwU_7(TJE~;wv6{|0Fbpil1ljYQ{%XvnBZb^3-Y?g1DaAU&q zK*|xmwN0O535ZA-kP=^s-mh-~>j|KM>05vm3ZGM|*rvzY??1=uivA_YySjjnJ^Md9Aw(wi;1q{T>Kc=K39BK2Rw^*UlaoblIg5hFwu7vh7Q_;jtM`SR52d4BU z3{aE(%=h`3A6oYQv6m0o*PqgQ>ssRzMH%7~&M3HhrrES}3O*-)kunDG?X- z-^a}Lmqh2XQgLGC(F@d$fj3iq{MgW;XTKPa*FNy`#RhiHZW_{js;Ne!A*@JIQhHkK zwbV3G9q5yjpPKD`R-h{I3E1+}33InFKs6f0S`u}4jhIZ2z|fPx?}d!hpGlDO7(h;t z)Dp3m+kI*TF;5k88@pFvR83&HWuz zWMnUDXJx08m{?mPe+}|^PF-! zJ2~)(k&5>TY|&?yzzS?NYp3RuuPG+HanFuO7S(3`oP3Z#P<{7akKod{W` zK+k2JIp`wXU@6j1N0Uz(kXzg)Evfj0ktk0$bb8R>onEL8`t(z* z;tUa-%_TrUXDTLw1A;hO5IwTiyeT*0@?TxmVzJgZKq)4}#nbGGVXD5n;4QB;q-6=O zY>$Qp_oM5waYB)LE3sg1?8o8Vh?8E{<)y?D>|DtYr2L*Y7i1B-Rtym>v7_HXezt$65e9mM(A{~<&d%oUW|h)zvum=8gytCFg^kmFkKr;ijgt8|S%JopU%N6q}1 z<&g}P3wTsq$npPpuM#?m)?bwdD9rM!uxe7KpK#xi31#i3S+OdLU#)Mqqt&3sQm{au znZAY}z!p-le==!q`V_?JXRYTBQ=k?HxI1*}eO4>!6eSfnQ>TR@OAy9D9LvmRtJr3Vh`FSaLerY>zCX*dN0Hbf%?6Z6a6y;&OS7mxs&4WQI2FK*s zXx^AL4BbS%488+$7OO2ECd?&C?@D`%IwFVR1OJX$Fhdb|vu#Cv$;=M?4=fQ%?gqE1 zT_B2QW?zsWx2Ydc(Ab*P#}*r*dQ>fbjjg%BStS5}+go!TLwom%Egp4Ft{?DTiHeU0 zfw#+s-Yvj_8VswG7NH{b|AESU?+sTJHpi!VjodkXD7$if204(*CfCBG>@~yQS0%( zx9(&9?rDuy^O>GgZAIxBzV4oHPRW!sO>Agp8QCgrP9(jvv(vCOs~b*kw>c2PuxgJ* zfC#ql{p46Eno;?#c%w$Mym6T@hFA5c%%i3IJ!;?hjX^%s^uN1c1=!gvTi4Wzkxp*0 zZIRl|VozrlIV={ejge6W*Zo5k^UKxv>&NxQ+tdN7@WAn!*JF1N-k{|3#c=2Fi6t|e znvvD{`{i}%T*gT!9IKGXjcWiA?<(g%%_LW)Y#3IhjA@pob7+DHr}@?EbuQ*i^X;0u zbi~ifdM>rC#5PfT~pCViOR@Mr<{O0w~N#O0?$ z7nG2dF3+yH`>FB?yB08;XsBb?R65OPjb^EEuF~;gnd_mqhlJVjSR}YH8iv+AhE%H3 zluKXw(#S>Sl&X>InuipPtBEW-25UX?@|A)S4roS|PJL4yGemXsZWJIL?M)@zVp2)- zabMC)*5Zn%Coa{Lsukd+tve0Psp&szOQ-Hxkw*78%~Nx(|Ifw+#|eL%^u8PVd8oy) z?ltiAhlbBp4y%K_6?Z5V5lxzv_~(Mw83ElHVs%}jan%tASJFG&YeZiVssgDcdppc>~7gawNUz73j_=lkp_m=Z!_-9diMefGBigiMTnk+M$C%Mzv9GaGacS?* z+q*H!FB6g@qXgh_Vlud`V}mEfCQPYZu_%Ev7+*$`m)5h)n~wnDWGf~xgrZIOoX~9erEnGZT;Wb zpM!`T{mO~{N4d!)MlS;~gZ!P#+WJqPltH0Sk>)R=ErB9YuHW5|T3uX~s=w*jdMAE6 z%CrO@Ev6@F9wy+rGgNfH!COCH*w8;+>E0~pl~&|?`>yBYgFQAACAhAk%SvYxOONj- z-9F9FHUd`h3xcIvDXYd-ao7>AuW_!?cE_Jm5ecrcZc++Oo&84*flQ4FaAcVwaG}vo z`s1&wj>^fh4L_Wyw~kukZt2Ala7I&jt6f{z6%{sb=Hpx|rXA`z8(eqeT-kk(pC*$N zT;0N5cQYE7d=#@ETsIHWQ-1PQ2P$7hzXdkRZvxoQEnWH+2|J&6h}yEB%6tvGo({+? z)Rk|76yrC0zbGmq*7W!4OQF(H?JY0Qk=#QFKgo{5U0>Ijg9Y=vBctyG&7;tYv)S3K zjC=-)c@C4A+0OVgKfjtPMi(tpUYPXD_@pku#9p1h95)YJI^)hK9lTer6;P!H|tu_G!!*W{SyoduLLSHid>fS8S}67wUx9| z1q5=<8&J*^<=CpM%OR?9vCoZvkm_lt@F2SP=ed28b#_+3#uitMV$SOrTOi-Di{!r7 zW#3_9*3pb7oc6jd9@ru-_3lUAuq}O51OVRTkmJFB!Si?{sU@2%l7!P#ndDd2KP_Yv$$)s+5G1y)(lmE5NuXp zvS<0VOw|IoULv7XJ#{y^?L>fS| zfww8O8d8doY)3oZ%#;2e($anvaF+U?e#K-;Rmw6S#w@=M5uxF5qlV6K&8#7W3fLO< z7u_5-_~rK3nSeUBQ{3n&4ad6$5foY~zKag;fZN5ZGo^{XDLUVY5`}+42cgG*pg94g zf&+@r*HH9veWUgfv?{hLz0e0&N*35R%OhXZ7^v@1eA_wsYBQW%x$+pTch1y>Nvq5W zQgN{6f?DgX(mSI6<U5(HYu`%oM#O9Ov^mAd)1SWTcEcCgmgF0W~1yX%B1J)rd3IyNUp%9ei7x~E% zU5c~9&Zlg3w_ z(9@3_L=kc<6`H={-K$r}i8BDvROyz&GA~XrC7Tyis-EbRfQvx5#cq1N`|9pDAXS&; z;1G7aeM|nOpMc-ig>De&dRp}OUiB(Zu(G>TfVgcI_NhM@6z98JCwH;X_D!Mo2NHDl zKVEFoO=}cs)1yWFWA|_6;GJiI5b&RW5=o+8HVfO4Y?|LUBGkk3?V5qbhupNUdzvq# zX@c)=xSj2GC&pB_zeZ784~y5>?5TJf64a6XCLLKRR)(nzKiQ&|3Cu*G9u* A4Z< zcfK3s+IwA%dX_U^1gaBp;*Qz69A#bvP{Pp(>K*Adi1 zEsHm%&~yq3RKn}2w3Y2pU`tM^Q-VSZ@yen2Y>{)Lije|Z{2y%1W+kpuo)j8Aw%-%N zHzAB1=3S|7d~b>W4&I%xei>O{cYs0NJqoCpyc2QT#qcaR(HiKd{-htAKWlP1#y8k( zyqpzSGF$KGKby!-=}4Uu5&o=W0hh)zl6bXk)k$PHETei^4wC!lcU|hm&iwYzV8SUi z+OwHNEwLEDJbMe~4-o&u7Pa~l7me<8qerL-<)if_t_TdLZr-$~uaYfX?5oMbn9tqu zO7DjT!-lo*_9olY6qjh*riWJFMD8&(;eSZ0R1kvmXTfPG@I^B1!MlEMvf6*ozMzQ3 zbFC_=P$&PHfyVKu?jl{+Ucro0rcFK|TW0Txt{1w1aP(0j5UZZn)2r06-FNCEb>lM+ zIXw;~44&(%C;uEQUOh?Pob@Us0eWus(_|V0U*>;nHsChGaKb{LTNe}}{(8!v@nWy4 zBbf$^c;>IZ`G*8sY&B<{&b0gpofv*QtERV~!I433)+%}4X_v2|G!{=`VtbUvS9!;P zvIYe}mN>}RnTdN8h9=-N1$|R?<%!9`kEknRwUHxlP{t~MGpH}%>W&tDU4zm*TQ@WG z3T5;y=@G1Evia$$)}`J|qM6>L*ymZbPF)yh6eq)KYRy)Pv62L)+joY*Tq|RMhcBQ_ zLU$p5GiHv6`!xp{26_Ne0s5hDCcyjg25|?h&nkqegMljKVX4*5GJuJ3W|5Nit%jve zc)2Ch-EPrEdWoSwIjR_5)Er6;qj8-fVGt^y!)7h`x!omsJT1E!gMY*BH^xP*+i?~< zmfmm1zZdVOYCxFG4*sK@eN;MaeyuiB3d8GqtYBK=b|O}{`4a!mF+*YvEr}Rg*1s`< zBC6DrGqmvYGmplwv9-?+n$~NfGvWlip>2Os%HNgl;6cGp@xx{e)^-fMjQ?PKJBLG! z!O!d%d1BO*j$Zipr6FSr_O?Dy;uEo~xd9Tm_bAc1TeSA*kLI_$L-cfCJo$2W$>*Af z*Oxmu1|BgKT2uWZRYf{uwH&lSr+I5>8ZvzB-bpl$6iW6k)w4|j>bFAW3mXQwBPdYU zj#qsp>93Y6E9LluI_@mHh}Iar8a@t%chbQ9{XHw#$h$&-sx%E&wg#UlsrD;XU7kO% z0HH*`sJElh6E?N1B^q7k%FAhs&@Y{=&^78cDC_`M@Wo7Ga-b(r!u1@p3YW>(J<4z9 z9|ipE`LRyHElY%_V13e2?v5b<^(XYwZ{w6^Hd;bl@@} zujj9(eZ^o1EY&KW?^{(Ua;ja)fJ+gq>O6i6E%(SV6!l9tCRWh^wOX% zFCW9W2-Tzj3P zEA!3~)ZE%oPA`0j}iXqJktdycOd$xk@<;9fa4JXw9)g1%874le~)FWLozVMJrV zsrbiaGkLFzJ?IX;(!y_6r9JnH1Bz`ZHu?ohoo=5@|EU9PL$fhHUyh zD_(=%9F=(uQ~y~r4)Uu`Moo~PHx1w9!RxhOdmxs;~bq)iNe~uU-o6M}x;XXFIdXb>?&^&v4IZO}x>A7MpC!3t$D$|vM zb_gXfG!D-E5b3#AI)+1ZBqrd#tNh!Xow5||lfJpMl2S&~TBwYh$Bm!L=367i?~B^y zwq3z^(Un0_NOB0g`dYz!C?LVknumG}KaDk5Yn%TxY}94LB(CLTus_{Egog;y+?=u= z_O0D?bU_6F7OAsQ^cS=A{bYF+$c=&Vs8Mcr!0*V>jlH(!{XvoEDCgcd&J|0+ud8_2 z)uuzTQlI?|q54BbiC6qWiC5VpOrzoCX!q2PA$x6ru@xG?YNzN?RP$PPXB~)bPrQB` zIc|BIvTl;RGiuA|of+mSjz#h7-infrgj^e+AF3Ydt7w;Yv@yChYyN1c9M%On-fzec zmQ$PHF6;w74&Kqi=xwyNEG4&0pQ7=)<17qFfUx(ERZ@XKW8 zytpuFflXkVZ(HLtxCCTDB|5pBB~-hY>`mkSo@8-Fj&Z+41^X13>VO#3!4r<6Mqjsb zE)mhfr*z+65t8w$BZ_5v5I6W&f=AWS{)+wP<&{r(4wkPOi&}#h`fN?bu1*nN1IH{2 zf1!VxaIup6(mlbgitr_Fw)qCG@@qXCgC(Y%X%hHv7v8H8cmx5<&z|e`(~4_0t$?cV zNK1s|0F(tA?L_2nj=I_PJ<+J7y75(ajrT_4ShfglaG|A1+};uM%6y~K^@(5A$BOLt z@`v7k2i}TOYA0w@Zz?TXI&#!ee6I z@e8MZ(p67hWmS(7Dp{k z)-MaG$jX_PVPcViPBx*Po}%BpIv+O-^iEdwV=Stv;@!Qk0LW1IkFPwW>$fAijcXj5 zlR59YS?Ra0&7dm9%wr5!wIer{;3CSH=tDTl-9z3$x!oef%Ij(EP#zAIEUCpyIS3%8 zlvVNAul&;8R7L|C-(HW6p8U^RBOg?F69p7J1`>WP2XdDip?O-{I7Dvq^=pmP9KPcY z4B;;vj!%*mPp5BT(Qi9W*f7LctT?MNeUi{Y{zwUh^4Eb`q7!&OZ|w={=2mm*33)r4 z>p~E3Ry~GEnTAH5uf?tfzpKKk*&0CEfH~0+kDZ(I6AD}yh$Ltos*Bp7^Al#369<%j9jL$<4)jxMGao;sr| zq!t#_BZL^Hst;F%5QXqq#p{zT45U&vIA%)H)YWIT59sg0MA9{THh!q-ho+GK(iuHO zd8WsSEpIUT2ZW)}^Uq0-d!2wNE*v)em+BVeJcUc52coPOd(3k7bl z{m{_=;3RfYViG@Cu=>VOt}xsrL{{@Z9}wG(P$yS@tPWqPEV+tS8Gv*SwC;*l(InJf z9Gd3X^R`a1z3?s0=lOpQymwiq^j$_C8y)T?cnnkzB3kdg;`lRQPxJ#^t80J2*4ilS z(kdr^)wT)+^F_Wx2Wd{%Qwh%|6&LG^G3&ewXS69ArT%X5g?pVcbq=uY$yV{Ou{Mk* z*DYCr%5{fFo@~~!Pm!e5Za>^5^bp*6^XW-5@OjA=+xGuL)?D?mshiE3Gu(B7w5)I~ zuFify=^BEDOS&+UxS9l##rW?-N9I>?XKrEziLhFF_&<>SI0UKD#f6$Y?WOUrjybGT z9@k;-y1|*Y*^1YR%0fZ7!8~UYO(0%rrQtavcR05E6O`>5x52T&Fhqx>oG+<%5N`ri;Q~gsL3}7_oKOV&3yo>7P9z7X*;=>%^bu#`a8E!6{K<6!jpXNu zXSp^8GA2g14j7YcF@EoF+$JK)2!dUw*7udn?fDhFcRg#u_+MSkHl*I#5abZc23R$IW@uy{yH@ z`e7R`-2w;Zd5<-n+X*{>ws_0e5z{~XH+xopjI9yBX zNhnAmrN_W=CTdZCPZt}`a%l6y(kwJ^$Th2`I-8WqGS-!`^P5C^%^gCE7sQSNDuyZ0i%<)F4ffAit zmB95bAyKn19cqW!BoNI^b$G;PUIaR~gV9_kcqCkiGWZ)m^U`Q>n@csP|yQW7&K=NW;GiC-V zdniL<0`TcrVHGCU58Ji|mSo}OQ>~)B;i7S`)Q#s>Yf#`QP}Rb1LRhScxOFZR)T=L? z3)gW*kKgogaK+Xx2JC-%C;gjzfeT#f-0D%bZc6_(-mg#*=OgE;R6`H_`kWx`+2UpS z9?@(^cu{5bL;D$A2j0fDKGJOc-PVgmO6BHlq1a4i>owilC>$cXwB z!WhkuHwL?*I2&`hZ)|&{r#b`98aM_TC8|QytZ3nZe=QEX>9hm|!wE1V=wa_HvDQ_O zAuKWv@5Vk+B@r{-q;WmVwUp$)>u0nr% zVv%N(P|M}_#|4S>`*z`}6ljg)o?5h`qHA|>u&%$gglklHqn1?yy{tF2d)F#{&;;{8(F>b8Ph{)bZlqTjo~(*Q3Y5;=z6gTAN8}N?a|ns zQ&;enZQ&VqT12syjMxCo)84=j0?1EYL1@e|#`CJ>SiYPU1<9fTB~BeU(i0iR^lHBn z>?|%$P6;VRJG9zc`AsAZ3+Clx1C5riUx9=ZLKBOb^KYUpMQZF~40cdBM)hh~&UkLy zk4^tzVJcaBp$BJoe>hx5+9ydc(<)n{S%=W}E_31LeX|SBq3i+vlzr}4O;mfndFqj{ z5mZu{JXa7^sHAmBCn%z*i2K9P@^Q7Uf5M78q`YxtQ^Il`<4kX)V)7^)G>q%*C+3oJ zwh3miwmi+;mN{t?syiRSD*kOytBX`Ggh z?@cq5$(hOhEp!B&EU0423Zw?-{2&Ntn8Ix$(_=fw7bu$2#!u%-W3i6zaL4RIp%;no zjVmTWZ)Z{a!Yq6LHH-$C?vR)Sd8C_sq#>siT{HH708c=$zqwZH?cKySnJFv3GQ4!MiR%OyIh?H^IDfn2O}Hrj_**C5 zRGzWdM7V7_e0J>c$7tz91X2SMVRzn|YBLK+ zg)Q+#l1Y7fRZ_2k-^XEK#uQU_BG*w`;?3Z$7^vGhp>umSS?PhPRP~;ST4{fE_j9Cr zqf5RGPmIXZUE0 zMc%Lt%f9kO8a2JR5re5zjM0>GKNsy>1?jNhnUi!shu?ctzxWf6&p_}G=HXC?Li+n7 zau31wKyNwL`}M=taTEoto7knOuTFn+-6h~Op;Uc{)!lvxFn_Zeg?Vy8T`_b;9v-Zq z)n?1zthXqtN$2vuVWQEk2sDU~0Ns(nXk-3_`HDH6=9rBd=D6NKI+Zdbo@Eq2uLj5s zE1c^vyVKuXQYyJM(prCp-R`5BI-B2*l>KD9w#;KJww|sUEBd~kp`w6u+$xeVYw(HX>vp(^QzBTLs1X_V>Dfxwb|KE@E``4NFzN31*`KUd#Y)pEH?w=}uciJ@nz!mbBS()X~p& zS&g>$Fzk+JQ4i9$l3hRF%BGu;^tFw*h?p$hQ5`P2k4W<^)LQ1tG|0!8DxLaknx2=s zaxq(;(SK2_k-(IwwS2R`Q51^6-|}Mvfga89qJT-A&=zxbjO+$CNBx zmnkdxNLZ(Wbahx2=eeIWsPY}W<7mqr`|ajxp2G-?f%DJPzMtx=7H#0rKf>K)TdZ}A zWPj6kh`>%QWUI51yevqKh*eM(sS8uSEia|-M;SyL8|J*U+Jm!_xO`k?t^YMWsX->M z($_PAvQ8wL`NYY5y>@4^;y71v{pR}dV<-ok07(QTSBp9`h<;M`4~i(y|Z zV@-rTwT7R-P507d;WT+B0Z}lzc);xEIr*fud$FO*d|G{EdxN^1cQjes&Qui0q;V2; zF{aVbX7vwTh`Gz-3%)@QJeNqzg*2B!!p)qUH*%y3vN6>m))zC7TbDzF`?D4L@_)0y zgiWIT=~JeN`CL}B)e^qc5vu(&jNt9O{n1zptKGQJQwqngXJrdvY5buzyAEN%aGqxf-Lu-rgV*4VN9%w{I(=E0n6&)aUfY zlJVW1Z;tA}Gpzz-PK6CP>d21QL-gj*9106Sp-G6co>2l}P z)}j!g^0d)QixO_4JSA$gKlRPAG>;+*=NkxLmU5QhH#Ckl2^NbJh{aml#3#K~lS7tN z;|kw-Cus)jLa7R)Jgn!Xru&Ddn*ghij-jQKY2JIDQq2F#)5L(Rdydx&GhJ+jAjrTM zW*RDww_{I`nALU41?}#hn12sPcdF~{fwEGy4J6u~i2GcvW8%;+=?)`#Zly?6J#*&6 zr$_LMN7zS$+dg_dzjW$#LaBNiLxYdiM?9b{_whJR!wH(Vtg%8&U5 ze=79HxWBlQhp(RP`{Z!+WAakEoLgyccw1#ZqcCQ;2ZX93j05Ij6IOXhOKd&rec^PG zLSTnZY<3`W|7a}cyz~-V-9SOPFk<*>ZGem+^)rN2XS7j{`;@t~JfDYRh|dCr;)=>? zCScEOd~?f`b!Qk-hJWIoH;sZ@sqV@BGs};SUvuPcJFH>#u9M*`Zlhx+H7bU1PDwZBg>TYA7rliP>;mg7BpEq}k36=2wXf*}gZqijcOlZcma{_fg#0?MIPAXqLrt;9@o}-6&K&8wX#LmZn(rjg*pIomVO4Tl z+uwNd_w=Llx9!!$z$&avKecpM+4kCdxKw^A;#(ot>>7`6c_+P{MN(mKIcJNnWvteS zsufh*$bUJE)PJp|k|A29%3HrY!QgXsHZW^VXW92rH|&>oN+3V_K7&cMNnmY@bIs61 zn!1=sLK8ClQr~hnq-xDaq>(P)U={L2u(R##>f!`I?K7fCLJ^z8&Bzfw!SZUCF{Mxh zFI?XqvDQ1#f5(+ZXeY60`DZ?TqMsSy0x;eRv^L$r`+t#v4=RwiI)n#=7b;ZNg=;mVazQOai#N{spgB29|#2Jj4fy+Ak z@M0)6xIx|Ne#bUaY^iQ45U`WNQ;>rI=_aF07C$O^0-0?D%U{m^1BF~j&2 z$^IC4FB!I__ThAc-?(_N^wm6;tKrpxPnVJHyhJJo9y}w~M!2JU2-e<6zJkI`SJ&94 zyMM|dy!{N5<(gtPtMJLCn9!yYc~PJrK}+pGUlsA1igu6f^ltyL&26ybglWEhN8Tx& z{S!jMuWbs2p@Tij!tB#KH~|Ib?XG(Xiad4qJ!!GnY*8w$eU_-lE*jh{v-xIp)5T^U zyLo8CP?zlh;{w&7HC6P5TI*}9DzocswSUp_t50iqXqVk^1**Gvhmf$ZXPXT*pl~hB%mDG(#jQaDRl3a|ZO|wEY+(dAV zY?0P-o(7-YWEW=r9u~Tf>|>b981=q{)Y`BHK7zwY`BC~I#h#j}&9U~G*|(g!RevXb zX8m{J4hIRM(d{gbawt)JJLhM%z5{<4gKjnahLHxe7r74J$#Ro7N=W|F*)>j{hVGl|`%IG(K z;E?AYa68+wHDN;LbYnA`NP}5yvVWI|CzzDd9x8W?86YL^;-?WQt{usoXzGcD6}e_AK~9t3{|ktw^t8S9RLZUd(3944sDC6KfIRy8 zwNOB`{i>Yl_AHys=Ke|z0afkU;{vJGR~ul4a#Th)h1Up!#Wc|E!bodFlOi@5I;^?c zc1vP+Fw=EVf9USGwFG{REuGN;iL9VC>$q_uHP?OFOCGgk5;c*ojh;PFu1&NHl8PAc z0CpBu>J{(0*>tbmOu6xBqko!`Uc)ww7WA?Q;P$dXsLmhoB(=3rpEkCVDH%X69>{{<&aMlhcundt<7cxDwAoLoc6HlLb z3ytpvF$`AqniW&s8%LL@ws}M5(KRqMpxYWN)L1Px**(@rOUm;CQm4(hyB!nY6u8}( zT-9?S#Zu|Iazx6327eV^+$34bX`4T2e9RK%GDm}}3=FE&$tzkj&tW=r?IRZ*UIxS2#I z&g~feTnqj0S{RzA{&dO6t)csv53|hf<+{M?GMihEm{JZ1*ok@E2EoDbclf7ZXmu#9 zjzF;|v6eJMp8G`ksLbhMsT5Dc?(pAQ2izefxl;WPC!J>&Stf_h;KB(qv;YBEY?+1Y z`74+!OWkgMHh=wF_HiwS7Um9R(26(?YYtb{d<=Sn@22X6TMFGXV$+{!Dss`>+?rX^ zi{iwK6)!^pD!i3?+)9q6EPEalBO=?3^+2mkNm5k_HBvHBhjWi7g2u=Lg0gv+Gh|e{#m}DnSVb;MY|@x_4ykt*4?lyE4A0K zW>rCSzHoHtB(8ivy)i>s?-1AvO@Z$V(QhxHn73Ct>kQpCRdWB8BSb0#zsX=a)e0Ck z*#yRL*3VrseZK7 zGAt8^ZGWruX3IZR6cVSqrx&kOXJXi0AT*LG&4>pyy6NsgG-ykFlCYu545_Nie>JIA zfqf>sUtv;Cmqcp{6!JnwOxU?aYc&o6K6o|t3p+A-SgeaQelD%&wjpci9QB9PkO%vp zZg;l|XlK^u+LoQaMn&<-w0^C1eu$h%jq{1wR)6YfbnT`@_?EzMcY)45S?DTkWu_K9 z;g(NqIyZ5#@{7kV*wv8aX{{4#k97sL5S-@os$8Yum%a#Qyikrnk!>oqoW+LgQB$no zEZ?X{$DZTBG&wPuSlqCKQ*HM`C*76{dTkF*f(Qg>-$HRq`m24!#%=-iEm7iOgv~lX{j!%o>~B& z$Ofs+@lYX-5Umlab}>Ex2M0+$#PfHdSW1dejE2kyCivAFh7K{=j&PN!QXDEQ_${Lz zgX$z^z-g%0j7>)uERBk7-(w8BG0F!g#D5M>(2yO#rGAYDZMex8zEB?TDegip9X?Eo z6((9;5Ta?~mBwgSRlm)e%HmrLlb-9SctHlr&}K3#2ZdG6G{{Fmi>8XPW%KqlAi688FY zDj!~QvLepyV*~AvcbBk*Eyxi9M|hU@;=iT?RmTNx)T8%B(BJtzXx=V`_IE3klJr{d zoVc@W&x4+>pK~(ljHBer!tnd=(SK;a`+NV&qtz?ZkJw_3xW3+sqzT#y6NLVKY2c^w zb;$&@%bpfnBH^%J7Y~h-5IAa)T=R(t@lZ_RZ@$Y_hWm7c&s&GMdYPl7zu%Mk4zl)u zrfeAn=85~^BecTo#RH}F@cp%C;cRk+VWiWqqb@^+Pj`+hl1`6LO*$BE5`T$v!?6fu zIb2CvrK%0$RJZKe5Gw524$_NG^fui{YVDET5`sVdDnTYMiqIwTN@AH#ynSaq zy!yG=tI4;Alz7RW+=xJ84u8E5`u6uDxgD47k_R)@Pgxh;XML{Eq6;UgjKPI_eHn~u z`_B*>rk?KR1)-=$5pFD+)`(pq2Y>RGktIYjpQDTuxM|E@1)=1aH5#$%)LCsR2MNmG z5NTeS-*VEQZ4CCh21&rq<{1&SsF)8-XGj@mZbfenB|DQRW(y5ll`&Opxc33agckywrnmh=F zQgEIBF~|8N(kA{Yow`L9d0Um9!E}kAM$EzU!Eq8_mqexlJ%5)nHlJK4&`DdvXRuYn zE%$T9*8bk74%`I^B?luUr`f&IBni!YvDcM6@lz+$aZN|PRd_CB)>Y*TvG_`q@i2_Hw2y9Nha{c4C2#gZi{`;j1BFb;` zF}6#Q3xzOb`8*1`3{#AB4(o$u>mt^}V~sxB))^0TC^t^r3D{SWMjN(DG!ZjPIHyCy zUrgy=D_QQyX@fs|z z_U>}u+#P?n0_|*(^k~@qnSOcCFShs`ZR$F4ZfZo??s(L1gH;V3cj#$-`fXp@Thn@` zlfa;|nDhheUiZ*oZl%wvWln=mrj`ib?k6DeNFTN9jk575rlXVH15?O2w zG9iqrynn>|+#l-1RVklG(iTe#G0^{x-wB{Mw+CA}?mOhqt_>B&Dh61+yY<>c;=;R$ zl{p{9<|UIhNRla0#eI+ljfq(q1Tk!t@8UreWf6o(Y zZ&v>9z0YHdXOtGvkerFx`NgdDX7UpmLvq33QGZ^*$2JLliFrgjbxd}>2tIt3IxXe` zxdyb(Ak<;$PF)nu5DfZn@B@_V*ST7G5j@%jWA$RR1jlM_3b(q`Sba{Lp`2DK9c0JQ zMB+&${v3R1v$dM0q(i;}a~x*+vExFK`eQ+{r3QNwN^}E53ykEGM>6Ro#fM=_l*PkfNh&DfhP22w3Ie}6`cCq5LS+?@B9#!@+= zZv*#BuaZlMf-FVK9B2338OGjCX<=wrG=HepS>?&?#GEy|GPV@P3dXaM=nlpx)nLD@ z&8Q#O&z*6(iwUf^wA0d-k%fzCKK-dErmXXVIe@ z5J8&MjrH>=N917Q?i066M%D{Pn-~qy(M@F%dm5==<>YO0ZjGXYgx4HOXgFEyx_{Qt z5sX`DMCyXuJ=Fg(42`U9-yqE%?*~p_!@kjo_U5jUz4-sl7Bjb&?|f! zu3db8(cj83kvm-@e;9N886Hm@N;0WlM!)?LXjJHVS>Liv~O%@wI!jNo#)PmPJ2(0@1%$V@A|N7 zP^C(UVY+?YtztF~ZsyBS(SI6TNpO)8&A1T?O#ihXb;nz~p`WL+dEYq}DJ>J}1NDa! zT)N%niz}i943kgxge)$g0=g&_!84TibLeALAW{@|y025~}qKN1Vx`&cxvBVreK?e-OSE#{S$Mb3$))I_Nwjd79 z+pG-n3zA&|5tQtD+kb*c)M4fQep+la{n2Q)A$~z>hk=s%3Oy=Gm_h?(mdXrnpY#1C z`hM0r2Gc|m5ZPXbjUCAPUu%;q9ROW2`M9{Y=*sm?3MR*|*4B+?-xb5t;!4e=grz*TAv z%dF-PFWEW2&wqSU%akX^ZKXdDcM2CU$^Rl=sK)9ZHzu7kt$uy2g)rT(>t39Z?Q+Eb z5xoPp<`l?YMBpm}YUNc*6`xoUQ)NrIHv4xm$p`z`^{ES}wYmV0mqZ}nEE%Hm!E_Q% z$(3k62a&nb5`MtTB+I29_Go5D7LY7HFRXPDRuUL(R(}&gpe`wf8`XaP#$)@VO@s}B zoJkXjNLVf8?o-mbkW##z=3^H{z0pQWjPsCskK^9JwrS4@e2mGOI!PuP`k5fD0Cm|i zY@q+J6vz6Lj!@)C%>jflm5xoy0@WOxl>54`ZIzJxel%$m9&;%a>a*@F(b42uW07rI z)j?$iE`Q}9eSv$|QTzQ)X>?&U2{CV@>9HS|;b16K)t<5VQmWBYm|~xO0K-~G08dN2 z|3Q_b_prH_dj=^zno(qQ3kfd37HQS8h1p2P;;K}&N?t(~v@p{Tw)`%=U=Tk^;EBUK zFgCwPEJut@kW000GAz8L%U^e!m4WaX0tI!m|9|08Hpd2%R1v3iv1PZbz8w1R($BY* zO|>WR)A-1pho1rX`|w{PP$(cJs2q0bK% zTRG%-b5mD|vBYJdqRVYysA|mwEeV3;eM5W;>HA`tLekB_#QsX9F?INgAsu}s$hsI8 z7wNby+bCjMCo`?`O7F%*)9DhQLHl8v{(o%$ApK!A(bAa(Q}lg#qv;Z}bGS3J`NaK7 z)TgIi5{ia#b**+{>MG;e4ESDWLG7NRd?SO-DkHR*{~5uiGWmPLI2HLBdwgw;BiOGc ztMCTV_Q5iY>S;ybHM*u}ZWE(>qehoWZ$GHC`q}h&Wm|l8t2SL0I!C3<_H23hrhiK+ zo1LnJWBAcBYaoRph;o*g36XJ@5PE;-EY8&bYh52vv`qsH22&Bll}$inczp43gZpB9 zwW(@;tu@*;dJQ6haF$4Y`?v!-4;J93RC%Lzd14einVYuFvxa%SmcY!yc|*xWy)L3m zc+o^vQm*ZYFWZdkRenmST}wS>qkjsWVVW7K2z=%dr-fjwUiB#nrvdASbz8^XGjPp{ zOn~0~Ab$&iU|1`;v43_i($ShrQLy;I%${ha;p9Z0nA2?my*yv=v zc~J{_e+-G?o>gJ7@#LO6F|nXKbdC8ucVg#OJuqGSVq$j+i)qh&w7$j$)_osce zeuIDC1+XkI|EOL#?B3s^S$|%{#5|9Fqg8}D|99dz2a1)-F6QTKn`xVEHJZQb)vkuW z|LfV}P$3&7S{aZcJfOzcjLhjJIAsB^qjPGp^edBC>gd6jedf0ZnBRE%3#{x|-kX(G zD;EXo#F8t03)!=%T%Mv%5`%^4;qrQanNxCrkPwT>DW{D^P04D%ihn?{!SGZnz=_6U zgQV8zFsoVR=GL48B2bmqIDdZOLge9caZ;XuK`LVDFQ<){*zNx%Ldn6Y3!++3Ud7zv z^TAxxva}jAxSsZt%ahx~smi8IwTX}jvx0(n3Y!=#R$fvy*c^C`(b$y!H9 zvCiz^mUuOHl759=zki8w4P_G&da6*5XtN*8W9~c=I*rWP)8p|Hz5b?s!cpLlVp4vP zmm|CQ-zl;~qZYX`y+m;MKQ{3>TI@sRpFLx;o5{)50kVt61`uR(rHem}DN%<=2zTS@ z0Ta1wCgY_kf0n_Vd`9C$GDFzgUm4=fZhsVn{k2``M+F3M&+fS7dHbPu~1*l$ZTR50WX z5iU{&XLCN~PBA;k+tamAl8t!XD7&M1rnbA!JL~0pFU~*FYg>ZZq-R*Bxmn&DS{ivU z_aJpk z;%fw*UVr0FRJFRy%i-%r>D*w8MqFvH2qa?evu}SmNFs}aGgOWcq-av$n?Q)GWLjU( z$riIFs+?!xwt0SYig*{ATf;eU%zKT^re(>%-rCmb zWc(1a-}*NHJYY;L^NY0LwVLSs2b3P3dqq`} z&mWe1k4mX7o$%%O65;dGp%4Fk)=+=Ktp9k~ZAfO=^^@Vu+bi;Yz7o}j^f>)^=2k+{ zxYl&3Xs4nzg20glFp+5cxdywV>>ps~`3+IoaV_lbr`$@6iDGiGpI_*TRLUD~fYcKb z>womOOgLfTfZA-dAaFHkc&XqgYd~SA=z0IUXzH9A2+APfqn0?8F;z2Lf{mN*Owme_X;MhmY;YkBa-QA zo*P8qa{@jMFMegWpV~^vncN3@(J+E-)hT$RZwkCjM$;z1G^l^|6}tFPi9 zvXrVs)(ferv(l-EUL;){sJn>ax1azUxFZ`+1N!nyZ5wJ^*`-=ezQBzxe1voh17CR; z%5M148d}ORqeLqJZIL`NS?0sWT+pMNO~MB48NHRvJ`LqyGgT6aapQ3cn=4w;J%44b zmab*?Vyz?kW9Lv0E^m{~Eg*Wmxcqpv%5^*B@-l`t?mF=ljaD6YZ2F{jnLDVwRvumZ z>|z!l+n&3lt4)-Q$$U8n3>uw!s(=Na0G5?qewOIRMM0{o=wzWw22-!H?u$qH*Wm`Z$OzJ6r}>VK*loM+~vA!CdN->4CHzze-QPc1+G={)%eK^Uzw zN<&B)4c>Q#A-w*#UcTY}I&zFk`g{PZpJhL9PS>T%*L+Sd1?aX~lj9lDkOgmnO~tHm74jCEv%*HqxVIcLQcwBa~JaL__I&5^kkl zvI-F*k(h~iErxUdpkf31kTS?+4!7gSAAIF=*Nv=7Y-I3>!6$)86b2Po2QOs@td~tE ztnFD%s=<%eH2Qsl8uo`fWq(D~UJ$k1!o>G0qpXQ`htv^-<>u2R!-ns6IpPeLz%3Kyfbw6!30J88XZwytizg^HD4 ze%k`N-q0%KQ(pmX6?S|lW*nY7%UmrNthD6s)H1urV=-=?=pXE*lKb!uh}~ZDNt^qE zcbHGdpNi+pW+U`e!+)Kf&nA@m1m5G(8PA2yRhwwplVWOP&bF&*wYz=!z7oH?jGWcHJS1bed})oqulkqUEmF%HSiI!umee z%+Dl|6@gX?Md0T&ySC&`7EX^|9X9NYMH>sxr<5Be4d9ceex}`)+7!l?T6CiS@mNAu z`?0cugHT+j7_l4#of}dj#Wl`UI*u7JAkrmnRKDZa#j0&4ZKv3I6sepjd|&hT^P+bL zpyU~*konIz4S#4NNLm8l%~LLs-vX1FCP*$~^rm5V@8c%BEWf!KM)T#oLYUZJ$jK~h zT*NxxYaDZVR2g)=MXj)OWdOWVMv}RNv(^jH!6f5VS`vn}wbfJRV=?OSY&|TC<&r%+ z`-u@xx6ahIyLarxYBi&)@K*=yTxak1x8pTj2M^k>DGO zWxhFQQhy}!p5o)D8}z}%mbk89G~ro*9)sR5hQB=vlc&A{71`QM(0NXzzivDGO}9p$ zRWvaklujfv8RpU8w|9ShQzn&b5KV4Y$}Vp4ona>+LW&nKRNpK7WOBfdzVt;oSU9?G zJOtBuXCj`f-eQ-AeN&0CN~<%u(R`qmByhi|Qhz}YW-yi!6-;MVbnwz0enjYKv z8uQ~gWxz^}(WFGwFc2BL8@=N;cN3o(LYT#`XOl!bTF8_HoiFEX@3s<$%xWm`Klx{t zT13Q4%1M>ZLtCYP#|?Izzcgk}`i%LCXprr?JU*ID$4gs>=&;;UvA7i%E-RK-eo?6&=%WMW zgF;TT(iM^BL=k!H$LpN1;QLq{W2r~doqy6cUXa4<98zua7|U?Jy`0=|>-o?wNTkjF z@~ThCqBF_ngsU za~{2gTFV8awSt!CbfIy*-M*KXY=om-t}TpoEU%B$)YJvgYsAhHlW7Gmy*HjVdVeGL z3lQyIj@M<^0m7LxTgYak8^K{Q)4%b!)u;^l4cEB)+Vq9SMRW#9XMfPmH^>jA;1tg; zQ*F`w+_nEL#MN)OyaMJ|#PlZ;yq!lq6MaF`M78I5TClLsD*B;@3}Jk39`?`mv{fdf zaSZ0O;)yqS{qy!k!=6q_;}5WS!G8%gnH2l6tdBKn6hQ?4MkABR%CGQuHZkBfs`vL) zPUsg$*!;;9=xDcE9RWOsMhs~lF_q{lQW7&3u@b=*8-UHm3G{3)jbpS7qJ->b2PT$Y&Abg?>8=W3luB1O8< zVTn9pv2a5?iTKOr(Bo=7ZGS$9&?}CTBqq8dv-eQP@ze zWZB1cT5Tf`Q3K#+Oz=LOpr?e>__z&0m{XZ6Gi|>cFEg00g_J2)<u5_G-iNL;_cT~p6oVn`?PQ2JJlX}?%f{MOMlm1t@G7$D~hh3T}^93Oa&R}h* z7Jz^#kHpXXD5flPmt#sSPP4y^1fLOZ0Q!gbX}IzbdSck2SbuZ%G!>M(mES9X0mdxg zr+xVkW-C&CLorHa)#B*iYq(A><+mqsx{pRk7 z47ax}D35LQx_>?cq?AwUX%2=IC49!?13e_>ZFLG>h7#84P3m1kImk<*y%(tGuAV8HuKCIah@D zOn|vRe~j5W(gvS&zCKc^HZ5jbk5}uol*7%*&0Xs&*qUtix)9kcBuu1)#pc*MNzTsg z2EQF|zlRmww2 zynpCp6gjJ-?(&k?M1H@0pZ@(-@|;>>m8eoNG_2W*>H`%wyG%XMOsBnAdo>_d* zOe$W~RcrU?ODW`C4s;r0*`FDCK^+-R76qwoH9ijV6eXY}Z6-3BBMGz{0@J)$v*mGD zR@g4S56EdGi2TaI>MXW&3cKzWRUlDeoe%#k_WT%$$v*W6Z zol)(@k_^5NyVdR5nZz0jU5Xtr14A9j4cB?b*XlY`jQV{dDMS*fq}ZlUXp@DKLm0D_ zLh}2-uvjjP7^_+FRirw4x~Q?Z@_)h6nv654>K+w9R_p&U$YXcNQ8jl`Uo6FQqSF-?*Sx5kWZ~x6$`=4KXCm0v>$eBs)DOEY|Madix1UFCpRezE;e--x6oxL)Jqd%I zZOVJ}upU!u6OK}(e4dc+`54Z5iOx->(IgFsB3Jib>kQKGr93VL=sFWIusf~c2h8Uw z`CHibRWj=}*cb*x;AdnBA%Bn=<_@ROPSWmtn3;iN9P8|tA8UT}GdyY}V6X}wY9ukO zr(-yuorjP3c!t`;vpd#k9nRiY!A=?1GI24YrB(fXl}1+s!!DAd*Q~J1!YA9Po0lc=x za*SH(&zF1KQ9B2r@asDgDa}paBCDg=%`J^ab1FI`G)TqbzIv7%F`8fg5L_I(Mz=;K zuX3gw?5m9+6wjJHM}Nu{PMyq2;k+V(R~}Aj7%@YJT zLE}s75|z;(A}HRGzd|Kb?x7j$i{s2NfjzkPl9P#fstR;FpdHan%}#qz+#7d?pwOt} zR|w73|7f(>SWa0ix{I)}pC}z$?R4nGmsTu7XX4LwrEgbH&wu~OLjO}pvso?2Mo2h? zJpZT`uj^;7^)}Vc*Z>1zg`nlH4BZR9?otx{Cj{9{GZ_|_jXWj!M8hR@b$P9EM)Qm5 zN~Kj`<`|DZ57!rAd*=lXE7sQ#gTG8B2Mv?OXug!e$^W=phFeFZ(d?4nuMp}pJ1S+O zl&eRN>DT>$SAV6D`nH~~0-D*c&n$tii&V zjP~pK#O+wT)~>Ccmhe`tm@hllG;6IwAxy$Y!XKNXt$#|4iRn*ayaNyC$5aw|9CSW} zSa{J~S6A?!bl*mA)F3bs8^311A2nfMn`?GOO=h+C6i;E}*?*eljU*aOTXKCrx>-0K z-H7|MNJhat1)o)Rywdp^YsBMi%V;2S;qJq2P~eQTUYA;{@{cbqA9HcsLN^XK9pIdn zt_e+Qj(-=jJ>U6_avC#^++8LLC7C7%pKS`CJ5Sn6lTvTQ(>W3z>Urii2BC!STB4*s zJ>7qrYusEciV55q=QY_oIOPuXBx1d)Hd~w+yV@+A8mldjyiH0XbZjAP+Z1CHF&kEY z-2E^NLMByh0(rKkBskcn6w?`qoY4k$P9nFU8-KgX;rf1DUe7u>^h@Iqz`$|qr{los zqO-oE#JunPXGSDZFyR=NZ49zGZ(~L?64M`vMYFSfXb`*_Ak>-W)1&7-+-XXY@97NZ zOzJJyy@C#|-0h_%6QJ{tLzks?BRY{2T`ghQEZ2}I)<)SyZnaUUwYbnz2(0<};_9rY z)qkOI;msIKra}mV@7rzeY~cd}-$?DV-A7>)c7Bs13lY0_^NvtH?F@=FT=X66t2{DF zup_~pY9>$lI2a~nFbq8u*%FE&fZH_?P1~@W-lG4lSW4E^Mm9QJiWBx$%`s1w3_eH0 z{rcS_OCD9{)~G@kT)07M>U;Fad}fVxcz<>se5FYK%a@nh>vP0YI1;Za_IIW=?9x79 zM~@yxLc_ygbDUvc-rq7VUS{f?w7J`xL2Vw6~rBw>hhOb?eP zQByIORV$Txq(ys}w@6+;TeuY={(trZ^g=uI2!A?{x`X*HNL(Y*-)OSd_8w+YH&&Zo z=ey^_yVhbAZ2phtbB)1U11cutQmRk<*z4}RBfblTdf~WyG-LIHDJ-t?bs?SQ7;pcz z5!3{^Fk2TrbY?v#W^Vpmkb+{ny&AD3HsVqK7SEvS9?3eG;f(^WU9=)*+U&kA0~vip$ob#e=kyw)=oA#~c19A|o{~fxt})W~O&l(^ zJ9cv}ySD7R4qFVq)j?MXC+gYZjSdgT(ws3fqdTaRQ2f{i2G3e4#DBn-D^KJ=;QII{ zY0-V|qH(^tpfQ`kO&TvKPZuJxo$)1!sX=~b{(-~SqjWkx$AUVm)^IHyE*-ry-iJ(r z&55tKVOY*5YZt*MBr_v-_A&H=nm=eO*hUDMleEx3voe5U-Y#959PrRMr!SkY#k?7Z zYoOjLlN)3)K`b~K1AiTGnI#l4qW=(fjT*`!!^7JwXUvd6nktWLNlvL&&lG?oT9Uu| zFjD*6`h~twSE;FwD8^O^-g0Kx)meJmY5R({MgyV z-nCLKDK`_mOpxRlIj_q?VK*aCz2F-J?#1}U+<3VuWL9yx5rKw8L zZjp~bzL}(HciY9*T;ORrcmSi&sPYwWdHzOd8uj*#oDp5xT4$i-taK*{B5joM{?KAW zrNyb$o&F%)hO=nS+=5iOt>Nwk-?oJm75r`M_fQ#w zs&$h9JrQ`OT#-ay_kom}1C-S>qRG9{Tx^B3c7f3HG#Zk}r~NnmZu$xzQ2T~6`sW@V z0fGuy3de8h%ztSW{92}Z`;fT}7PO`twR>?}EnO=m9%pt?oxy9;Bp@+(Uhjn$pDTqikEZK~13+9liTj5)uwMDLTbjri2 zrhI+uOtyTUdY%Sy3lkG&P3t?|98xPo%b<KV*{NK>8}g_$7^3! zKD?D%*(*_GAO~9PV*n}@mhJJ9k56Ht5QeZhpf^w{ODs81IEGv}^~4#MdY&3e=y;?V z`8NA3o!*#n6sS3yS?SxXH2LBL$*Th;5n98p@IDbQ2s2ek|{cAwsMk;=2=A&A#DfPdCXjIFlyb-`dEEkvA{%n#)>he;MqjdU=9 z0A%F{Sdp5j@l>90P#p|>s1Z*`=*ug*Vl--@!<@LjFjGPOqO|;B9S%!Nl?F$$M&s#) zxE}NjB0~*V3+UIzH8Qt%BR@&j9fWyTjltH|5XiW>GCz$JVZK!6-;&$e5V5$824Z@m?tq>RJL0GNH*BuE z%-ES2cYvI}edj(dM*w1TE7)Uy^5#jCa5|!JG zX%hS;b5Uth^6Z?kW!^SDL5|X9%)`5+-6aOVKU{S7;N=SW_wmzuP<2ixLDU=Oi-if! z-CNs7PJpjMDdJ|Iz+1LVPmO4_jnx2mWysQVkM$pa+Nw&uz31`({h2G~YIxJL+(Hf1_rf4nL_$?Rr zI_t$7P9STR(vnY#_tHz{h(c~!?{As<#P3K%5QuP-DAgcRrL14Kb3Se?Ia^dg117L? zE8kId&-lY-*~U^uBQ-!J@}^joxpY?1>;i z41RDe8`N4CfW?a+TbK;<8|ggH4X2V=gp22%Ytrg9G3r>;w&zLGi6xIa8bOxpaG=fo zNuTW7K3dLNZ7#*R(mMgJuBt_7oYK%K!OVHT3 zR=?3RACb&wy*co5J_*xgYsRF)L(i`^h)V6Nm(_M(x{$4z9Bs+3KvND+n)x{U(1*kC*JZYJ&gG zx$}5dW}`)H4z|@ROLW9pIBF$sf^=@LLxu&S^KTB2Wu0dEh1Hs1cmYz|Ry-1axRpi~ z(qB=Y1GfYzRuR6fPXOBX-VaM7(#7e7h>aIu^$&)t1)?hmkYsjY#} z$q2pm&g3Mh5g#|y%zZ!ZMpgG8Bpn&>WQmmkM8j#AH+$m6EVnMV*Nr+0r8Nw?x(4T? z_BJtfnjf6bQ%Be+_ud0d2?JDBA($`3Ej^kN;j1$x#8CSW89#qUM-)q&0g2fd_wJZU zd|ak@{<%-MWAJZB=AxnUKZk}`k}eqdDVLTvb(zwql8ITP#L-R`awPh^Peg1DM8ntU=i&nVCL zQx)Hf%p|K6w|;-12Ucsy8)9#7GC%#IQ(hw*%dr-N{Q(t?!v-fnq#!fQ5RBKPTJuu6 zFCH&(@L=nazD^OE{!@QlTaRS)pqQoeqphTda-9uKFZ-@Z&#l>f{paF)XKIP#l(~E@ z-TtVoVB^?u(PguPL?tk#o9^C)c*MNv=la>EoM79}HG_Y9@?)anp5wa93d+?skaJ~r z`M3;rT}pB#oxs!~XR;qPjP^Quu=+e?k1e^gR*Hwj5s-BSAi1Z>lE`I(K&PEUx4_9H z+T@!hO%Gm}NZjqpx_8&xdFrTZb?C*%`G2y8D0EpQ=c>1aIL70^xwzPKpk#r&ycu z#|aCL`aZftEVgp8`n)r9DkF?75<|XM{gp|WhkN_wI?UDwkI@GZ$p)&n8ZExdwj6RL znt}#3<*uK|;{_!qO2vGU)Q(QV>(t4X!WZlCLbyNnZ`Bm)+pn%@bT2COhkK~xNY6C; zU5bA#NOo+)N)A(~*=*+lt7fcOML*H%-Jd;+NW!0**Co-CkM(C35o{X;BTpcj8STjh*W2xo9z*(yIYWd`Abz0%$k_mhXiNKYQ zxzig|%MxuWrih%QLYBKd(BA9#p$~t6$!dQYSh>ckf%waC3}2|I1T`M3H5%^i^pwC&3hg<1*aU4g04mlRNU~=8?(9bl)@Ql-^uyExNT&Nl)74e*Gh+;U;^NxRV z_mR4DRD-3qY&PgFuzH(Ynq*dcDdsHEUAUE0&p@O;iOwKo&D>EFGiIm2$h|oDmNENf zoAlMpuJQ2akZ{o^x6t8aMu|}zlkqI^YzT?MGu!M(n6gQG%lxQOcOxd}60dXDbS#k= zEB2t7T8-ml!_8%5!|fy2K3Ds*i_d>f=Uc&Nu7swfL(Bcx3Z%pH&2ww=`?qHs(%y@6 znAI38g{PMkT8p16wnhi)pmx_f`{%Txi0oksE*<5JW*fuL*3)@UafK5nL_gEcwNUAh zMt73CEmXS_QEWMD4llj4@yVqRzgOq*xp9(g$4sFaJH0>BC?9D-4t}zxF`a*{PtMh> zD|}!NbLMT9*}%Flvf?uuf~gr2bw&7m#2Bq~q{=+&df7gKU#`_9ZC_W3bXw&ip2Q~B zuTYe;-WYZK*so9$4)Jrdp#|reom44i?Nz04wMFr?QiG$iqBw0JUyi2d>6 z`tj`jMMMh}fex1l!IO$KFn@oM?dshX-)Mxntz&fk#eIJp6f;VQ1xaceeJ|KCZdLN{ z^f^4WRmJUneYqz6%SDm)*9ZP^n1RW3%(@11zBF`(Vt|nU{q{$HCQvMPr=rGd&LNQ5 z-SKS_=yNUW+`UPT0H$DGkr(6p_!br=w<-8*@i!i>*0-L6D$jD>^h19Tt62ron}x)< z9945b+Zyen7nu|zZ})FK-=^AG%kIC}=tUHcr7Xs6`;f`A{uL-G41|B!>4zXZ_Sr*4 z?~rQHYM6pw|LxvQiy*F%^KBc)QIk&QtKQoRvq|_dpEdmf(YM7cQV9vrf(tr-WDHV? zttI+I8Ft{KtS|2f@%n#_rLQw0br62k;?L&ys;AF00)^{CxdgDcm(9t)yr0k(^Kvn3 z3}eU^p=URF*+cp;mvXDP70n}nsh>G;x>RGI7ZAaj968>^T7Z79hKCapqD0fMEqO2( z9)=Fe)hY+F)<=T<-*)o62|Q72%*Sqb5qH{(dN0nN7R%Tw2ep3*$dm&?=4=R4&#)0C zWWLj4+x_u6y}$48%4AbHBbx%thmQ&ZGV~z0@+cIb=%`aY_A(4!~cllvhSyZK2)i6=F?A_};!3L6z#X4e03{5OH zb|{yTO2hl;S5kl8;$M}{{B#H%#La#AJ!IxBy5Y26>?6 z9RP(SarN4YN_S*E6OLxNU%k=K+H8-G78CJMrMUFcUH^s z>Rnf>R-=DeC~j*EBzE^0GviLz{1j*{RR46)34-?x_bh}B%?J{lKkmyW0xl84qQ5k# z7JvRy8m&|WR}zz@!EW2?=i?L41%@s&HR@0FL--{zP%NH zIe5y;24WEv?63_=0r<9Zxj#*}F}+IXcJ?vCRv*{yV1_)*lxMNNGWQd;7LyF9<7>q) z;GA5Hfw3^!VyGnyx6Gz+SZ4adQLe3i<3Q#}|5?l`GlomRU27!^{)eIfoE!v(gPX5< zv#o!)7o6PPPq!E0usmn5s7c%xMAPlxOnd~vSwVRVp}W}T%e^YJGF3cJr@y2%*LbM$ za&c0V!)aRqvrr^{dLOnsuZHc@kFgvf(<*Gp-(1Vgv-0$-$;NKXpuSm;xe5Bihj?Kx zsUZD@+Nd0Po(~3-;cK_+*mz!^2z-Bd*GYdF?q8QfpVKK{ASZ(Y&BYZ2Qrki1ifNbo znTrVo!ox!ZxX`wcs5Am}Am&Sr>FB_8AsC%;ceY|%_#B9R>68WLRUGtDag@1t|Dgz7 zh{ot;t^Mbm)_}~Nqori{3=PYO?6eE9c}ZWwViLOu9rz5fKOZybAhNie#H_!#lka~p zCsuyB519laD>(kggVySD9$sfJH(n+(P>Y$9yZrjUnO>_a$hYB?(QLko-GvS6iw*9y zdt{&Qy`b?xhuP4bldqJj)e(aDgFf>QuXexj+~HceosYHIaqPIMDcNa7cLnv#6UQ-M zcoVn}&st!Q`1sCKPyk0S%a|JN<8ObFkOQi|P%pHt|1{<`gArWz)u(c$w4 zgYkIjm}0?=RZTMVu|M9x+)DWtzLMZtRItViYi`i*uhTs%{vIx|T=@I`2v=TQH*g-3 z3@)|ndA}Qe73e>0lR$`qbIAHpMiLdM+MAlNsI|j^{6CP3K`Ml8bqR~^PeXs>7{Li* zk})qrI`RH(VzoT2UTlKRB5?oXzT@BCk&VL!u`pLH1>dtY2jF0XF}EFvgG7+Ndr#5R z!p9o8zpwvWJpV1YDhQ*-a;|l033v@!6gV*U=YT$F-%9tz1!Ni?kD9{y-@M!cd^!xT z{JTpz(#5$jv`b4AMLiJ(M%#Z81$Z7EhA>)L#aK|ft8*(a6Qx9t9fpv;Vf`^=SOj@M zEU33YXNZ)V*U4eMxULK;&q`_;E_`%X2~H<0A&6EfhKM9$y5HUN#O>8u2!6fWmXvSf zc9(kbz~5mpAedBx^O)Ln?h#rbI{%c+51Z}{3PFDX^Wg`MKMpa(OqqWIC|O?N{Lh9dC~;xrt=^3iVv7fTUu(s{TIJl}j-c4hB{)F-5k6rzcY$DbagV%8sW z+Q{?e+2>Hh73|I71|O zk?Z-vp9dua3=2|!`}R7Ln89L*m#U!a{GxdkxCxFyW6Q$)nCue%|lRq z{3ZaKTG+_uXp07#G2+!P|J}GO;cBC7m0&cO336$MQ3Py@fI_5!BKX4;l3K!#N`J{x zg*0JL6a~=@ed7brJ&vpB#^(b0kER>fx*NLSQVS9h5+$!WPS44byL^6YXoPTbRrdDvFaO=X!+U&ZxZGoAwm$sv{ZUw+FjMz+#hF+H>NhZ}x%nCC)&+7VuT`91{U9@|d&9EbZ%OG{QLR9hO1~7wH%`Eibm4VPD&a?Ptk@IbXs>i3lXw3_QJ;x`iu%m?cEMegIg&D{CCg-fylxZjp%^X z#&^8akBCpfIzjRC_7y%jd`IBcL7Z=?J?&BB2mOEl2G&bc0NRQzpU?bnM8FY*NLnhB zN9YQPSwG(vnyNB>Yc8kQ87S?(qmA%9AR?XrZT+7QBmo@glGpG4cLTiv1<=rL&cIp| zSS}baH(9E^Q4Xu=JFC*lI*l9_X&3Ml;;q_qM6`hV2MhED67&lYV_qj|WkSDkMv<+P z7f^o@51Tqd319igB4I(nynD8a@!f6&0~+q%LIx9~BAII8{gd?(Y*hh4>T?na-2L4% z2~bK*PC_9>&L}S#!k*{~{Rdf*UxCH8{igglDuoKmKXi4B{)yTUAaH@BydVAbUFJ8| zgyLDjS|hYpd%baz^f|)zQn-~BwtJ@D>o|Y*#r}1Irrw&vqW?er*k6s00ElZTNE*)H z1(LAVUR@{x%>x7GDlMV6)^|jXyt4F$H#=VqA2!aVIS8jXqswIf*>v6j7w6rZWVzQt zCyE_pno_@KwX}r}2C#i~^x?ys=sm`W8-4YNJ09%FGU4bS8Djs)o4tX&i2w$q=B$5S z!0*;3pmnoSmic5BboO~yAj0y@dwi+YrD!QXs3Y_fLDfBMMA*nkE8UN(KgPxX;Si|d zfG+0~PQ{1U{s17b()I-Ug0u*I7s_~4s`nUgK+EB}xN)G;$-VxO7w7|dahPpVdf2ZW zW?(Sa-W`1elU*f-#!0Y_T7$gFWqp4kagj7ZKIesHfcq9-zHCc)gU*8ZKSDtEi_Wso z$^bI2pLGi1kkQJ7xN-_0$l@ecJGxrRA6A!qD6h&MMI09T%R*)8Yj$h@*`$PNfs+zB zAJtuc{p>+_#pn)9=xI&nFAIgmyf9W1nc0)p&I zK#;|fw+noYgVYxrB^63_AqLM84J+Ipx~m{9?Jt=uEQRlxs^4TNJ^%XUAE7PE4rrq> zS9k+Y;jfYSvvn3zg+LSN*l6r?bkv&GDy*PX;4rk;eLqIuz-QtE91Gh}+Fkz13gJr{ zK}r&<7{OAo^1phvf(W9dWTk&K%PsV9qsCli6fY4-ZuA9yvfwWVx=xfD3v{S~${6R7 z{|MIkFa98(z4GAG>t|9BstrkB{PbME2(2$gTHo8DW)r7w7B75NM}u{i+Jki!Z6lSx zP^u&M|7UO@dx67#xNghqXCL@h&UTT2#_9kxR>fW>+YCwF8y1;h^ag+72Y)F9=X)+l zNa#!3Q}WP%HWbCnP>1Qd-M<-%dxd+9LkQ!!8RO0qOEvehRCA$U?*(2jRXEq)Ty(@&Goaz#Bn!&NbXRtk(>6-$=0CTtk2%Z) zN$0z@ujNJ9{uA*1uSKvDzd! zIs*U6#Q-o2`p21&=a~3QI`8H^%lT@2V^9EKqa)%r;;UU{R*^auQWKRzu`iv=!DqvZ zjS6WF7DD2=4*q|Ch6<&Zj9!VU?*Q+WyLt;UO+&LWGLRp^^u*4bTEn7{|`A3u(P0-wFwePb|w5z38cAJT-MZu&^UkjBU-6g?J1daWf za5Y@c>mProL17+vK!1nnc!KyEA8_!sPESb~Ni0$k9JgGQSnL=`6hTId>T7F9TiPCwM}7vY5y@Dda3E z8(n`xsfL2epr8n4IkKBVA@Z7^h!CSsK)xp!H*|x7hC<o{mt`FS*$Dmo5B2fao0~ zHXzI)`i-e(AoMia2*C^{;R>yH_uO_g@lgyB-I6hy6&#gQ2{qkgg_nJJfK^ZB(*5$k zNdHY}<-+G;HN@&XM$|AiB1pkkZSJLq2b%YhdbZY2#PibZtKiAfeKk$Y3Lkbxw-o^v#&b3+_( z=$s29we4Pn>_c$=zpNrCB)ck zv(2__EnX%yJ&{Q%U46k;&KGV9s}Ljn8rHlkE^X?-oFJvXM?Bj2^Jj7Apl5AC#oCF_ zUwIRLej<=MWI=X+r})ap{`ZDN>j{f`7y}zabR&whzu*F)=Vu=pE6_l1v(!0^zrqL- zE&Ej^oeeYi@qHXvBp1nx<0fJ9KLCGuH8)z%D0)uew(jG;0D^NqfQycVq&JM8LJ2N3 zmYXmd=i?9AB9&4I0_xXR_}|_CN052gR#Fx?pT3u31o*AuGUgP=R$u^fwpT| zqcShlJtgLZ^kG=({{<-Q&uhoaAqC>p1eS<`>NBjI}xYd1wr zqj+{s=Uvli2hPt6rL4;`jbL3ie>o^FkY6|vRSBZyMSMT{*KoPYO)SIp*zqN4UId~E zK~#B_fPE9fG3|dZgmIsO@pxq1+c)sh0iCBmHrDBy1NU^e5nnEW4$8rK=FQ&%kb2f| z#F4_)J+1lpK~dm%T7Z9LAVWw=!u&V(=khJBY)#SLR)%gr;flxO3f-l#@0rznl?UK@7LhYI zpRfvjVpNu%E+N;_Z0=Pr;jg4~a;c?r((?YE{I~DT_VI9?Tg!i7X;apXMqxMr|F3AV z0-^ZMGD+oYmIiD^zGLM6xUtId*DL#QcF0};tCYJ9R~WKlH!S$IB5y7J3#i%&?fTm#kBFDmYKKkJwf1@+;;)IJ~6RnWg{RPTGB5`p5RN8=O8;v z^q&(rdFMGG61#sfq9F598MgXjn2K#g>=14d?Bh(XslFAlz6X=t&_i@xhLQ_ErcIkY zVbAbi-i+Pz#p2bINN_zYT^jG<$yyg-W5|SiESn2O#9TMVOf1VY9 z9xPKyFD%(XVxu~vqqVe(41W9OAf^0i{Z~T#e@*fdl;L=W2)*Von*Y*Bq#*{W_wLR^ zqs~dL`cQv3Lh@0inkO-*nR4lPbCwUle=t|-9KK8<3FCI~Ln%=A(T!M5J}nZWH>Bx) zuz4n75?|T9G1$1kO+<{9ELS`=d|<$sP^B163IS1VaUM4=1W5k{Ksi(^Re2vp~~0H8S*!M%i6vSeM(*X5W- zA`7!+Ga-*G{s;*bTmSBu0c19>8uL*WDv03UdTfO+3*reZM7Hqg zAATH^QK>!@k2v2R7mHw^Ko`;36i$g~SOEVv)fqHjz|J9yysv}wiD-W zwttlkPLjHdx{Yws}7+NQC-pB1e7%UFTWhynY;zat0 zT9&Gx5M#_@UtN4_jn=Q4tSJbiyOF>vfqDCxwF8+9NoubV-sbu zH`v|Z>lbx=b~vWB4i0s&39~tzsPDKye`lsQJd9NE|m^&7Y48TCsY5dSSpUrcZNkobR=Z#;jTL2~6O zB0EBtqY?X20zrSx^qr7~kFBicW}S{x33jqr9-$Ah|DC=5fK)6o>|6x%R|{ODWEB8| zhXccXzvx!6;3_Z-V*9qO;+i8qzZ7HE4aVKcHSUA1F>>C|h`3LSn4G#LNuq;xjs=S8 zE&(>kl`~XoRYEyK>cnvU%H@9+w@wvv^-4UBr}@t8AK1POpx@uBDz)g=M6=rmP?_9c z9*|#u+#!KqeYzKZw|#n_>Y~frTo=7v#9Ih=HVMx|!1?PSzG7Q8@k+CE@` zwJdhQO2&Cx4JP#qLjv6+t5Thraf*VaOZt4NKI&(dyc!-XbB?USPmg~s>p6)}#f!Bv zcUO(n#6AG<(4??{@n(2(criJQ10iD8-Tq@$ELwfSQ_@<>B1x2t@?EEgn4$$@g$vax zfsTW(yNgYc%!KLb=~rrwQ(s(C*w&qqwNv{ls7mZA+1#Enb6Onj_Ix)vO`|uB8AL>E z7cOcf?bfH*`SDfQ3j%*D$M&b0r%aa!mMPWoqUAuv(~D>^k2ztb-mcAKi$#`hQACHB zCJ5bH5wtH397$_Lc{1FFbI-=$pxVL=QN6nck7^y0L)V=<;LCxAhhLShP?p2X>GsVN zJO&>CtwbzJ0u_e$l#s^-`v2mViV3VnN%OM6@OXpM3+cDxYNlaWY}^8YGb3)6ZspDS$H!NQpG!Iy5K zY-n~KP59PRFIs;}t#?VN#9Zw#PEL`PQTHag`2_uDEDE`CZ#cC}$!bx@?jkF*dyB zkh_1EozW81UeR4UAXR7ni1+reUqbK}q!z$3y?I2t=V*VvCfeImOO8w$MQc`T;mPi@ zk!9K2i8N;3{9M;qaFhVOuBC=z{Y_2m`EeB0>c*iO0>Lb^IDsK5NO{Z7^nMmL7#6QoCxUnTeQTK}lomWxr~_6jDgpZY*UlB5A?J2G^eIub<{I z4g|!uLkfSYsjIm6gKcs&=A#7*yY=UV_kmWP&eR|>&dw~1M#$gyES#J8Cra8D>Cx@t z2IVH>tD{-!W2h9wukbPGjsWMerL^7lH|<^*aWmE$8k`n8aS!$fv0E;!md2_r)pYWw zX2N?)v-M=Z{PoM9x2f33q4y`7Q>fn<_fMNqS1o_#nU#YvcLx^c!)1h~(hSR9Po?q; zJ18nz${pIWFIB?hZfu=i>V^h#p^~c*+3$XjjDY(z<8{*M%;Wahc@NhCkWaqN-hOI@ zkBeNV`ABrDkn-6mCEL>&+L+K2AojxjZne~|n(Fbx;+jvx*rtlZWs4PQ-v5jO%~`50 z7mk1K>tdb~PR;gemzI{cwF{-geB&sGLau1yc7#PE_;lj8?BFt8-A4@B|Fd5qTdU3j z()M~y9f9Q+1CZ4ry_6Ar_`83;qTOeZs?5&(FyfEA5gfOYl$gVIUQjlr8FRJ~FI!#7 zZ+UF`)ShP1cGiF0{VOyF`gst5oNm)~U8R4t=0n4n=1mvOH@AGEe(VgP{g|t3IecEqs%?F4}`AiwK^1~y~#)rkwjb)vc*@~^RR&mK%Ut$|K@PdDt z{;=r5{=vo?V-0;`G=B@*3ht)$@+P@_d~+uN$Mf^FiHwjKAE+-qq z*u*35JHA_kGZQ?tl!r-JkS|i0$iG;+QTMuPEa%l}k%@U@8EuU=!#3tS3w5_@`X`iZ zf<92}3l5i*fYrdnU$ie}@pJ_vErWl7|MG?(8jSSv3+)F*&ctGI8arr6SF`qHHE;98 za~zvvT$xdNJH2*it6dbzp+c!qL85>kq?w!@JCDY*0gI%3VjZU!8xN4{|&?3$! zf}I!R0xEtAq8yF3xSYcKRkvwgwX8QsO;lPFE7Lz283KPS*Y$3lu2h*VPSQj_r5?5=67fJTSZoUCz__em!w{T| zl=tFG>VTwddy-h4lE1Eb<9ve5!e$|!f4)Okn#l}4c<4@jM-?$?eyDAZjk7rUId8j& zPuhOO6{+|AdO$nFzjlCbe7G$A#Z>8CELx805?v>KvkMtEW8#-TnhDJc7q`MK2 zX6R1o?jE{f7)EM{Azpvn&$FNV-uI4s|G&P!&8PXTwXXA8>x|<#&#REkA0Njj46I8E zhP5rnBUvs?F}^u7H{$pD6--i8EkkrZT}uAr$o(RXQ8|k*i)p8(emyzt6V)rJC7(VP zTxxCGz__DRHJWL7WrJ zrwFu|or0-$9%O9{8>+`rU&2TrQ#V*=U3UW@>#GzE0VqP)JzJll^JQmIs5j-tx%o!*Z)XmnG?RRhbJ zww9%Gz1u~Tm;|4ZYfmB{o!w!@LrZ`Jezh^%76dZ2%i%OMOa#Ac%g>m@s~)BOEEsdZ zUxg>@A6O5Y(&;}4ORI-e+7Yv+1Cvp_tXYa|U}f3P;4x_XOh zxe`k?$;Ubv*iTC-KRxm5!LW9_(zmDP{JZCUyNck8fgbsn`6E0)dzt}h3Y#_TYB%TD`?fddDMQjc_nsKu_&4{hz-v^rTD z##n!N2G2a?zZJqOy&QOub(o$Gjy5>=c-#KSQ2j}J!ERvWh!J+6N66^-FE zNM?Q5-EsE3?Vc`Xb=<+JWQujc;w!yi6Aw=-;lp&Okhe(vheil>_`eB)dT`P)+xi^Ymm;^Ooer$1aF^g zA6ll=;*2S4vK~D60_)g8vp)bPhG5bxrc65bXtLcArSX5a1Ztgc zc#HVf2&!jc?~WNxi@sAI@j<@OSYk@2nU7?8++(|(BKSGKtqro#x-E81yQ$SGaGZU2 zZL%3>|-|0CL?YzM_a@(%w~6^fNNJ$*3)ELxERVfZl25AVZ{& z^c3r0BFzF!TC%8IF@xTtP**&q4xVkRBxD!n+3LqsWWYyVuSKRwL+pz6psPu9@ORFI zy!}aR2B9VAv(ohUUb5EeIG_bCN5A<-m!#Kp6p5R3N;0+9s*_FX@>YL~oLIK@1)KSOeHIe0DKLpenvuK zn!KIVXJF)X#`p0)vh#nC_C@l*?_nC<-(#=3eQB+)F32kyO!fo*71 z{>zxegri8X{mFkpxUP5Q<)zQ9BDegSU4pS;XUb&Q6T0JXw9#MtF>=t9Ct74t0L#l( zUq~`130*1DZJS*`F;uc=_AF}%zdIYU-kpz&w&ezV0#DPrFx@I<=)C;S+~ykm=!a?A zA=_AGdsAj#7e(99>)Pbxf)XCciJJdyDWb?=2c>Pda^+(KVRVYi*0F- z%P0;J1&B}@B!A|8YGR7ot+r3hLHa^e?@J-MdQVvgA;PJJ`RBa>Va7sG)jngV|2W1( z?MItx*9+`j^euzp>t7sT2uLvw*ai3dhsX_2Z?BN{Z86r1!N!)b9ElQfhSjZjAU~ade8S#d&Qg)k-&Xn z^LC1L?HAhYdiJ(@;*aafnQRC?Y_*5G4P8CyYoBSc>znPe%u|lWFh2$f`Q3@VJpF&E zwNTGSrA)k7V@2zCHwF7)KD6TjSIoR=RLEkNzrJYKU%BbKA8_zJmDa9stkfHLi68s~ zzwvJU;+`h(F;0KnYMb#v&Hl+xbP9PnxniGc2`Q;Trt+58ctVa@ntt_sW~|-8hW7^m z?FPrt6=K!_;OZ@9SFkX&Dn83shs}S_ZpvV$ygjb@@mt_w5Xy?;$bo~q zdMbC)rKOZys~35!N=9$!zh-gQWIE-!@B3 zBj$uzv}-54KNHy55Mpifoq+qXT+Dnw8jITC5`B?(-WTuIq# zt4lsFBbiQc*=2hnUGxNCpY|3?!e9tW59}Y|^HUY>`aa|=e&{J)_X(81KQ4M0&UT_{ zMIjjKdU)<9WYhMP)|6z4Vkmzh{H=*@h=H$Ob5Fz8?afv262G6Q+112Wd4vcyNbh|5 z`;hh*+ww6FvLk_Iwdgd%dHvnBC7Y;{gTrvgU1Td|l9bKom`}g+C>6t7GG|+z9J{ZO zq(xAdwhxh5Q*im{YMe-I3`YB%zsuqOUNPsBTs^U%P#weAS0M0aEvdho!An9$1Hp zcy2+mkAguav-Ly6Tn2xV`UE6AcJz*!-L3eM-S|a6xm-Hzmi5ff8*nyk8tkbLdOhiO zPPMYrGGrNheKv2$TqFk_QSCt&q7svLGzW_hIhtaIO(vDw1bce=dQ-Eaen#2n4$NE0 zhNdVHl_#Q<$op$=+>gAQ494gp#LwCR*&Qf>_HlRhZtPZv?Y@7DE%Lp}8I~ZbB^V`n zF90Y3Lqp(^CDf7tU-{pQ2O_pCm=s8K_t)}J>@W~>|^rdJ#Na7wJb zws7Fi6u2kj(a(QcM~qltl(B!#O~pO={K+{E%Yn-MZpcq4X0`pq3{>s=5ztzXjAGJ4 zIQWLvauwo~;g!}`OxM0mO^B_e8bgT;6Sq5S3J~$hGVoiw9i3+rB=Sd;oZ|+5X~=im zasg`ol*unlZ?I0aSBNZ&>Ivkdb%q{-j4bFJPq{|EIeiORuBsoeny6hb%ct3gccg zlr`fR>1f%pJnQ`?`c)9TN1a)#PAPvx+ssW=Wl?7=V)5WC!+R^w!!Ky_NVrcShB&JQ zY+MmlQiXrj%%TYSETh4d%pW2Z6*?QWOS57bnW?}|4Q$c9Q!lU0Dk>y_b&MiaLUk9F zT8KrmgDs`S1@R{C?f5CTg(>{Lv+iV?zYA*J$pC*?%Bm+DA(z-H_f*boejTS_=!YA; ze_H9>=(xX{U)M5S5TsT{%G#dQC(dsH}gUEKomQW_w@jFqt>DRp!?km-hNi ze&Uk5IGwD3i_~oSDm}O!7Ef&8t^G!klnLimLnh0klzZ6yO0EP zk?cSvEi;p~9dVoW!=xu{vfcr<}TRqPYCR@FDv}-1 zcx2BGb*o$(TR2ZFPX@l=DYol#)ifZyC9|O0U&ORE`vSl+*_;HwZXTmG#>zPOg>$_uV z7s1dEVIK^m(iUF^cVHiU!SGkT7`4+U3T7*#!jN1l-dI%eCOD$9C~_f@sJaaJl@g3z z%w(FjG=~|_We2C;e1QnX($9U*D=gOiEN6A0?QFUg*F*~gYXz~z#8C^$V{3Pr-~E5C z9khsMOqi~4d^|>^X|*Hc>^UHdaqUu%6=o6v1!YD6nhXkhd*Y zr;&RMZ`EFVyvIn%=?h&zS09eq|KOUQ>hf(3o@S`EA;lDxuyXy)6mKH7ps;@kTH85j z=~I`8aJ}9WWXvHu-lJ?RGKDO8p>_M3s*!%R*)MTO9AMF3l+VLj!+MsS55{WHN6=6* z-^*kmQ;8JoG{*81;Ldu9`&8zS_dee@eKQhI#=m{@oa~7a@gN*BBl)Y^3B01W+pQ`6 zKAI^WU=Mo_i25Q!WG;NW_NagHyS@vdlw`6QKHTS|r@qSEGGOiH)OSFfo^h@^?vf(a zuvl9GdBfDtbz=KGio)C-F#xl}T$L<{$xaj;(&U(_gCuPT(W|OMgFKr`!aG_OPuDks z=4g9?S&oZ%I_Ibk!qRo?uy}k3-G#NwBTvdB-QQj$f42>xT2(u)bHab0HnpyuF<@Id zVTm~;V=QfxMn_#hXEN&OZf4Bu;XtCE&f0H9^S2~Koy9*EUz*F67x1dIpDPo6bBwd; zT<%gIid}&MQ{=V4c=GG zeU6FF)d$H;{O*nA-Yb7W*Mb#ICCG6l`ePjtX20;Zt|sT>uQF~qZ6sp!&0>~ZdtO5k zZ@enVG`+IMm6imK(<_cV9OrsDkS|zyMRJ94(ou)q>#I;9j*`8n6GxHXL~-xQg!M-u z)h4|yKP=u#`v%G!LQmYtOt!#YfJ7nw&pk8D*bTEQ?!c<2ykmdr(l#w~0_dbePN2=! z6Ewnela6RY$JE;>+}MhAi_^^Y1qN)d5#4iC>lF8mra(~pqNDoiLq*`(J2Ui7UEhL+ zDL&3v@YT>~Zh+ZnqHSe+!f4v|-ACdU*_{hc{GWHb85atI8T^1~;k$L7$(sFhm%X*0 zW-qx9J@}kX@>+j{cW68|VSa#xd|LU$l-)tUyIW;+_QRuPuk8d)KUV2T(gfiL%;P(( z#?ona%ea9V-eq0!9`4g^?H5q+nCRMmE`eNP#+eN#M@3w^!qYw8nf=n&&}e7z+QKlu zDAZ-nOoBu1_a;mk1AJye%mjdKqIchwn%HZtx9iY0R$_m0ecuSU9mX5poYbEj`Q0b< zVe;VnwB=t&J^sC?jV62Z=H-8}wAjC+v9;Ok^uC1|+p9LI71m1V7-K8r{-t~Q_VZbK zwVM5%ZC&@$Y!>yxq+?Dc-alIT(mrzPXLG$)4&l5xM>)py)x#7Gn{1RJ6kHzv9>jqB zk+eBHiyMFXJmkl(Z9o(&sHaB50K9be+-FKe~G*(VK})B%p_h z=LOQUHklr^j?oF~KFMc&7IFO7jo2Z&Yu%P;qgfkR7>sRGMPW}H?9~lLAn-PwQ z$;2CnQbBVhY%ksjrSTb3o^?+50r%8qkB|cv_g#{bmoN4>Hxelq^sfVKrJJdUP9Ip| zkSKp|sWYz@q>kbD)`z=0sew+*AA*To5F`w$=Wf(n?6KEC)}t_r_$B4|N}mQiL*(J? zI`yU;kj_+M`U15z6L~j^mZ<)OuYl*G5^+f(vOn zw$4|c{q+^B^&9!FBZ3BH9ct~ZvTW)!^k9FpRVJkXRY0o0EIu~gBy0l5t=sG1?KhB* zcu(Gr;mW`L8o`@CCd@8Dc^N3>{UriJVdjWAI}ps)4G?!V3as|Z7Z)|~ zF0>|4X(o{IQdQ`PU26-uAVM^Um-IFejO$pHU1UYwUop9q_qbtPdz1Mz0S9Tex;al0 zX?XR2Br;IudJidlInxActQf`5=LJ@I*525*%CFsp>?RT_lJup_-10@P&AI{QF`_9w zrPp+h&5hpTtN`%waN9;iUx(Y(UmP-zsH5tCKpD#-&TJ5fk|DaankhBTgzm8&$&Su7 z%=kU2`J`1|ch%swiEC4>{dmf$mgmWsXb1v-dhveAlBK{hZ#m1wci~j@(xKIGJ!6D5 zRy%Rx9)N%8u86 zZRh%P#?9oli=UdB`ew4{lw!tq25`k0e;~(zw?Fd5ag7zSMD5YW$@b7mf$%MN6n>p# zxWx&0^=8scUS7ZywHp3zjT^BDX`lVcuerVA72FRgZ;xeqczrBF#(sG=A-bsdv4WJF zBRs2#;RgXwSfReatXfQwpG(0H^Dz;BqnRu4>aDC&=KCLk8YMeV2dEW;aiQ@&FohHI zA>Vk}IHCHgRP*fXbSpnA;HRg;|5)$m8qYE!o#gYE)&BGZs7hs8)R?ByLj&fKhDT_{ zr|U85Ki0dHvT@Iqs{|m-Xnb=npQc zc2Khv(e!7~R>dW)htpA1>+3E4ppL< zOq$MNw$@q-+|MM_m`{(Rv9t~YRG1=a(~IptHTm5;%&zlCOb|nQIm-ffX;PN+K^>VXT zi%ZPvmhb0TElicrw9*Q-750r|HMH8-)*??bJY z%S3r**T~~-T6MK>;ch>f(m$1n_ln!o!9qlLjy-1gV~mtE-pp$Y)d*FO5B4x1TxrAIu#lgvnGEM z!!hzVtsRJ$K4zeQpBo83{<=;c>_xWXkZuajd|izRGjSJoA>;wt=YqE7N7gv=V4mVO zngcg&EJ|7T7$JkoB?vD&fL)!5ZlnxRmX{mU$j(j6t-Z3$$q+dI386A=v!bC^S)Qo` zyay(7rpuKQf>lnxQLHMD zK|#kIygDV$3k(10{DN`@1{Y!-YTQmdx(4bsJngavw}~Ijmc^yn7h%bcWeS#XM8_ws z-XG01f-^KTeGi;c-fd&^x$Km_esh$>ZKuOWcf7y*ipzQ{F)22dai&C*Co76ffE6-l zi5Gjj&1UF-;ut$JlFei^(QasXWqc==A^;=VjJ1K{sby6=ubh{;Im#lxR^k4=7?wMd z!|N}LVO0h(%=BBSgLj9cB?$X7mum&-JvG@uZDFQ{(U0*28P*d{vobm9Uyc0xE*z2Q zrxg-3qO;89X(*c97+91_x=q%(@A=M@f?MzfMQuNSHz-QhUiE-TrMWRA4|522y`3s2 zEObse_>lP!uFD#WO{tGJR5bnB-j!%Z#D6?^<8`xTr70WMVpof0^fcMGRlQouV)+i+ zB(NM%JP*)y8_2y_I-1xf%vpWR=xGHh~2#ldbrN(RVg*q;)pi+v8l&@`^AM>?+I zB728=l#I2a-yw+u8ZknA;ios-t8A`At*UGl)`Vx0rTG;Pp#7xV0v}R0 z3pva(&401Jw^i;N9ch*$ash=|iGW=i~p7bDk5!*d<#?dn~uH&vCdDqUcI zzSWD&E7c^H1OOQck)?DuxE@Fx*t%GYD%M#|Jh?s87_NTlE=o2IE`gYN#I4M=JC2Ic zd$oN=_B5=|&E9T3|2eHkr|o4Re`(Ev4hDKwM*1Vo3XX+up~;yvdJ0$;r_wkF{5onO zzo2i0={jvppF^xB!f>iw0r1Zh)rOLP!5U@*=XmINmU!QO3fVluI3hq->$dk&q&5v25IuO(=xB6%x7~{2&5$0OP z)NjUTE$0<3KwrDEgRNFp6aVnhi&qP4xbwALg<0L^VA%x<{%VVkj;5T-SL9Lb;Y&$P zJ@`NmuCXF0QX4p@z;mpXJ3U6UNoI4=Grj@*Oi^asOJ2dsgb&CYpo?o=;rcX z-(#C|*uttDYimtL7PFkoN9^?-s<@l=+#qgoPX~`yS~Vq)tUQGW+6>W($L6vYS<{9h zey8v8zjp?nbyVs8HLs8BNT{~lT8qbwvYTw$J>kXgrrg7_>o_55qF|mqFV_uSE53=!fdb1nu7}9kT{t%;L3!H>{ zO<|1J+ie9HBTrckhMHL2PvhGt5tP?DaUA9jmViK4T9MBccw*WO{dS9OnoX{*Fs$bv ztDAIrv`h5y(M9O7-+-up70?LoFhX9gyhH@;OY0MEv|MfplA|@^!*BI6&L}^9`DMIW z*SgbZP4a~dSd!-G-Nj22AEcYb^s_i&g?OlhaoERiLqS%9N4G(+?PmD45t)DY1Uki) z^&uv-xhSB_8{d=i@T$}Yln8dC8|Y?h=W3?R?^S~xAya3LoiB=ipFtLSeuS4dx2)I) zzuhzRd$nzM@;N{7)748r+8tORZn~1`LCaTKuDGB}3w(!K$MDN8-FcKAXl(M4a^gqkUfXGo6wY2^&A!wKOVoW*)^bg@G< zr?u~U;{VN*|N=y?^B#fHJ`QIWZKDbkE$g11BE54 z+$nx?9WWk_5z1HC{}8Ys5eYfwl2 zFI_>Z>6KPN{&HU@IoN0R5yP$*OJX z6nmh&VvL46n`|{eOa|3o7yHL5Qwm;vey#X|X24J>v1A|AgWIz(gnx_7T{o5?MX8%R zG}5EZ1)=`e@ea<&%l6||+Pf@VNACFa#;JCw`R4{DN=YFr1|c()r2cXAIa0NU`v)WF zJnf}_ocqMBF7~b;zZr6OEss05>s)SnUO7RDqRu-+r{h#+k6d>_*#~^~?!+_OH4Pg# zLE$H`sJ6EOJQ#LsmsZ4b%{J2Cl!)*9=CZiO2t$I_tRH`u5(KnDS4*l>J`gM5&*V}s z6K=*q@qa}iyg`R*PbpAG#sr1Bb188n%{20Vwi=SPv-DY5zL%C*#biv;6^&m7hE@-A zu8jbiP6{YqtfhCapQb+)bfEr;nOw7eGC@*!!VDgcdVG`%{@l-}0De}P4j-a9Xwftz zfCqS0PB91BMVZvWt>Z0MDiOiyWCZ#%EU@&6F=aK^+|An)59$Oyzfm2t&zK}bR)b`J zewso0ZbFtJM8Y~fG;7PfVPRt^nrQTiSSjP|!|?VJIWN!s*ih8a5*ibl(g^-ERF3zP zuSLlwL;5g<=9~p@5iwUOqK4;UC}(y%>ZbQro$J~$Itaqm=^S^SFBD?xWN@g_1{Zlb z3O+<&H<4p89f*CQ{lZ4^l&Xt@Y)c(~lQ3f`IBm}(V1J1KkDTkuX-r1QSQ!(wp`yZ|>{MxtrLEy(VwuIGxDw~U^#=ZWY^7o}oR-|p8*$#O<^|pIdr&lghd^LE zX$8C2BfYh!F$qjqcc4PDkK$wqrd?c4a5;zCuztmXXRfVH^3zBDDVXnr3U1pPb_8dj zV?GJ4*?m_-nm!tmqgLzc9^04)PGt^mCzj^1fV~b#a%_=L%kR4)y{8~TsCuk&DDoy82VF3ZE!E|XChYOa2 z{*<#v^j=bmg7~#IeW^bR3<1`LKw04ur8^DO;xV*#T^*L4SX7zp*96%rb+0J|>oWeb zF>acGx78`*btDN(pKmggwi3-S%EZ;%jQp|{pD39d4mR0ssi3q;R(z;`MaI7ftVjBh zVfLmxw5Nm{i}XA5_rqcUcsoo99gAvQ@gOtRu)`b&u8z3bxH&;ND4H5qURb^wd(~zDXl#elYz$I7F)&wSvXie$&lw9f+`*sos zoVzH96?J$O)Sk3TKf!y`8zxj9HdSx(j7NnsvL>xMyGD{ zV%ICKR*$D>#e3p#)ZV_gp7$ZoKx5}RyV-_3dBtmBKVQ{H`$j&m*7~+i59w`w0b75i zv$kWj{THLcwi9qJ^?UitjU_dV7JYp>UMYO#9iX0S=C;tTw;p7F7L=g7%s^=C`D{?f zoH9_*&>qRx;3t6OJI5EJ&jIbE=5cCEXQO?FuPO~Zd^FCEW<^tr<(&k#)p9*&^y>Er zh3yhbHuj$F*G<`S?QMl6$WnOgURZXuxKbo6Q%50kBHG|2@ zC4R~4sxsFMSo0Kr@V<%XH}+#~cVCau3sxy-bdK2zPzs9dV5es=ZtU-@oqJYoVpWz) z;Z|vjO^DD%Qf2PUoLvm29VB0B4}O7E7Uw>j32K;SPY@(sWHW3v$;3=#vJ^cCCgFL| zUA*KPAD*#EbK>!{T;yv9qOwGk;Ha0CJH`NQDNZXr*Qa%V4D`FiORi=6J5H-E#Jg-} z0$I&qieWG|^|PQ7E{T*7p=F0?>7&+V^SK9RhB^yyKkjk@fE$nxKcfhYLWYNJOmJ=- zla~2`rAXRGlY|xmE3{MwQXQ%jl;BBw`wB=Kmung6iJptb4BdvE+9*MaPR3#F%f)AKmg9?YRfJtTTm{PeM+ z53B_LU7^7v`MebVV}*vlg<1vDkNbTjRp2`|3ZO zvPJ2RdG?`C3&Wi1b$|$teX_i!!#kYb=}u7FncGa-s{``iTZk$pXh4%%;K(fu`5TtJ znuAk+H%!YC9B`FnxT75DZ?ZU#sfal0$dsl^F+Xm*bTV253LqXVY@=b%+cGlK^3CO3qv@g9QDM2ZUoIB~5qq&YdJ@Fm2 zHZ###f{5%s(}0_CaepCD{uC)(* zd(zoLQCT^uH-hAr&6xOLi20B{=VXAzrjbqfGVJ0LL_>}uEvg#tuLuOX&Hfwg<3G{T0VNE ziuZ*g#EoSm1BiZh>It9=qQpC%Z@qo35fqh~V;9~l^#zSs2Pme!wOZ7vpn!A4s8lR+ zIA5ezT>_|<0>xh1lH|vKQwW=99;%2^Ezui#UmKM72zn9V4eTkmy|cSI@9$gM2|Ok5 zCP@ycmk&-9vT1J3re)#@OA*rJ=^UwW7$bo;3xet=Odo$gtXmF^^}H^1>UB*pt_{n2 z>KI|Gwpxq-Y&7=%?h@NyrSfaot%28(C}DtxvwC`0ogDuN52*Qn23aXtLu+4tgA?rN zaeJF)s;KFU7$oTR7(Gbx@lcbjvUFK`x~^j#Ro470qT>ZPv2=wVxVrEWjce|H$EvIk zumc<#X*FL?)bC{gXRhi5%jeM^ijsSEd@q-rdi;C#lv%e`yL>-QFU*swL${g%aU*iw z$*ZvlVL9}S`|2crMc>S`uVNhjD;dwkQzCd)Q9sPA5)ahpSbP2kWGQCls1ZFDH&3#r|;LGtUJ0tn~l1AGiwR7tATIo^{!z^lw!S+ms|6- z3e{oR)oT9arYu%99|THNnUxFo)nTpH;Dfi(v?lbm3F1S40H(PvCCin-cFjO~XO6AX zj+dl?Unp0U#!sK2(7l=|F&!-*OkjLfp!r^$S=4tq`BM_I@UAoVHJ{G14cJo; zzds|~D^r~))6$$OvKU1Od9*P3XNS7cqbdKcK7HsraKLs$lbsQgTZc9|Oh?ue&yD8= zw*AGdNLcEBLCqnAH`7Hyg*82KxrduDwK=?jf394!CGg^KkxKc)`+KJ-An`-0xSRlm zL66zh&7l5=F5tAL;&|+&x$FAaLBdY6jS~sJ0;LMU)E*gOK(KzrX(6fLCVf*`PR=3; ziP=3%prwtgnaBPQ>IM6VTw1>KfEFRU+&q2ehBW4X#YOt2DjvmAo{>4jv4njN3%PB% z1oPo!;29ZZ`3vW<%+~yLIFuB<$NT`%7n#vK+|~>u-}>-6JOjdGhF8jg>2p)k$HFu8 zt=wW@y-Rv)y6Aamq?_>C?oy-6vzIBr@ffx=SAG`7yOShaOYmcy1NrjIdfw-LE*JLK zE3fi@=hX5$0PxU03a5F8`wI7gcc+!&!EU%vT(vCmbHR*%@MeQAgvpC^-t@2A$T42{ zeewaNl3U#WT-#;fpLYm#X8%W||OX$Iqsoc_4!wevE!i&TNb z4Kir;+Ev_eF$(Vl?3m%7dyD#Is9aNS%hq7Nrl>dPLS_^V|x84X_l3YOmcQ)1G0@ZN0|d4`nI0hb89x& zhH^5)BY(bRpDxXbsmf0(`BVlnhzTDBH@*7Kyj0n)KG*N>)0}b!H=NpKO?*zMO%pAo z@^qD6$?S@s=3P%M`g%NDE~u$l>?Yais%prD>P1HTAzc!pG(uhNtpnPZaUA`B2G~mJ z=(hX}*((&z4!Hl%dTnjeL|%~mG?cUt-a`kQMQGI-AsX^Y{sM~1D&&1d^gf#MVHK&?vqlCHYRYbyS}r-Yr#Pd@k?E%5#Fu6!)O z2kY83;sb#eHOB1(rzI)I1b8b!H&$swNaF5v^&3QNdKYCiE3D9J>@`7u-X^IPOZ@I~ zbX+nQT{BBEl^`zx3q!5TP0z-nk2b+xyqO&HM-1LhBVLTFS%}4A0$b6IDuYuJctL-( z()m^dI=@{0Nn>gXEz?D2JYJGTxNO?|k zuT;KER?$~TLW?!jC3R+hONi;xxb4`@ZC6^<`Q2<`D&%^~_4dmab+tAy`uL!=YU8fQ zvB}AKftaL{of=d{MMcQ$p!OWiG4c=97C5r9vI$xDH-1~WQBv|(Cv)h|+UKkqst02R zh=dxeWw{mi&2r?YCFQ0MU~;kuY3Ej|Y*@W*Z@(?C`;jirHB+L09tXCE&|(uYCjwVI zlsyX_xrN#)#b0?RzdM)LZuKl%@V)oB;621s%c5~!L6E-c`n{atN9H5ilOK2Z)q#H@ z|DSSeOGji?SDG~D#^vxT*lj%uYb*|C)b9;TuUmImlkBrbW0Q3q3S_vol*#M<{S-L4 zNB$(>TKtgPwI>*V7o6Q?jF~JloFU-}bVo3UDkDY@3^i5=@uh}H|M|q$g7b%<2FHEx z?G!|W(bOWWgq9RU`LCt~$ zbMJggd_!pxzzfz#R4yO&Th9bpEuh0&7rd81AFHA=e*$L@xUPmy1mjO31;^tbGhK*3#O#@V&`VSC7+3pOd4@mtP4{-#T`J zYkbOEINJ%UaT_k+>pdaxp?17%u;Z#RK#uVSEV2Ch(FKvvUm=Q3eHnqOp!iG&2N zMj^_EWy8G*mP?+fj^=~{ngdjqi@n*xr(MsnZR19NYRx-9<|dfbiGGsGCp)J#Y|crD z<}FuCLBAT0rt(}^Egmu_lg>8AK9Aih9kyCDcD)pnx5GlmX!-WOk!*hpp7-)3*dB zl^?Hvzoq=%sVZ7S=uR&3zW$xG!u&u`bv=(=&0saEp`)@Z`K8eh%voBAG5jGht-}%# z?XkO)5xrms!vx)UXGc7bOL8)aQY2Bt>PkvH!oV~kT$Hh4RpI7D;A+Njb*#0w>{g50 zCf>(xlh6ONO&cLprx(Jpk@|GfM7O$^QRhN`Jeyp8>Gnrhzf*Y)av=6f7Vp%1momE_ z9t*E`-Pf*!l~z#>M=KpMv);M*d-OT&a@$m5e%-XL{Xs8}A;+Q#VcTU%{Jkgp;zfL@ z%PF|0#AWqrZE<-Xuk-G)(?LrqiZ1?(tqR3*LdZ%a;v&kXKG}48KiP=#!@aJE>Keo3UlN$V=Jer>C?ewnqj7rV--ecrR%1var^uVW=8p&M9lP5__Mp(smo zC29kAh}~^X10Lkj-~Suu^e)VHzCqT1;MG;}n#-3=oBE~6V=1$C zlOF31hDG^-4^k)Cc&vpo7?*4`ZcLMeGK}o>PT)h)2imC>O~czC+BvrvdQpaBnM{6{ zn});u*Q#1EhaNt7;948yqp>C>J|(@O1HhNk>~D7)h%i`;CEeQUmnscsrOh4ci%^yP_47oOTdN{&c@ z9Y}x(Qn##Wje0S^Zp@6Z*=KJTGmuY9uJJs(3gbJ^ zXY3Uh>Mq-i2_{Qdgm%t<7J^G@Gdp*c+b)T)qmDbWR~k+5qcG+7N5M0M#8WM!D!aT2 zx!)+3uqi1&8xi^2z_bDUM@wx%@>mH>>d}#mNuGLkguh3)}#9*^q`{)CObtDDaa1pCM@x6c-vi%e~QwDVa@z!$G&%k9aQ zxOXRi44usR`E9th*KQ7c2W~0(ENk?r<7=(!_1!|k`NH!eSL&%qg@9DLDgqKP%vCmh z8wo{?mvCB%0l}uaB($MfE0T*LiaWm1!>{u%LE9SHP|MLwo&fm}^J9V{`x<-&Q%AH+ zvXGtBzJ=}j+!e!r<{~DjH2}5V*IJ{5QTv*W3!1JT#8>|7n=`HhIK+EaB$h~Y=j+iD zopP4tm?2fOddOEtx8(Equ|@QsETm?tk%DV-;XXg{9l!aiWQ4YV@aO6VI(*Bt&1?l& z&v1^y={_j!VqGuQ&K*4q_m^-(Kd58bh+wJ%7+imRdw+R<3+&%Jp4x7~ZSa7_r3C!s znQ_HO|AJ`nGm-lIP>eP~q>2_I5feC!uEAZP^Cv%tQSmpr1YaPJupTIdAxUSF$^>pV z5S@3dd0c&2DHk3W72xH0D}ZF-@`#`Ts0%ZnIN-b+UlWon8#VWBxHiegKj>-ZdE6p0 zt|O5Dm2?7sa+DL+2pt>fxbVKEi1%YwaPXKOr*P&ZOweOx>WCU@V^~l; zU3_5Bq8HssbwGN1)V@8Fbt^VGIa)|_P(FN17%(}1#1{#dIW4chV=dF!@e%gkib0PM zAu^dyGW1=>u-@Dx8MwPW2q_m+!6-8H*`~d^@RK6W(?*j67pu!(U4ZYSV{#4R2$?if zC_-d6Hx35ZyMk-U7DjIXkXhBAhYnb3b%MvKwZa1^ctRGd@s_vu^}yW;r`g@s4hq>s z7LV0`ZAF9EZ;g;qpZFsi^B7~|S1b(Mr!AkQr`I*s^T;~fH5Akr>Cc=XA`es!1^;pZ$DWV1T?adxS z$N_6TU#ZFqYapNS`T~}PePQ5s#Z_tavoo20@AN=Vb7|Vn?J6u`0WDL24PRitd#pF< zzAs6rGU`l3bA+$`wD^!W|3WbEk$)x~B|Z2wvhgJWrO4+;NXV!d_>?NB{$#qVQRaAp z_=>>gS4DBQ4By($2}~wucqDEmH6_a1f*_n4ST5=)|9qa*pP&kiRUo2;<(v0b6$`I_ z)-v7e`gnn2&}MUjE^7qfB40l>8uk1Q-^JM`)?yGl8s(JC&pIdV-Ji3O1=Z2&ulA#$ zcKe5(uCp^tXy$VkROOlmy3ZMgkB?)-0vLWJK$KF@hF8Kf{N5`1Z&}fireLf1-e{ar ztV&7WWgv%TuvdLECVDKyPVDUau9@tAJm=7@>FTZ2K?N@#ww%7>ET(ox8H^lwF^k@3 z`Qf%ukhQvit`4@B?vV#I^t`M2G~Dz^=&ZWH87azJhWqv@e5^{TcK^MdhwgeJ*S0`@ zt@rR4g6MKdR=Tf{xe$I?e#enog>Q}Jup@0s3qBd?KRZNbI3a26L0akS6DLXxv+)VB5`riiV#f?jMg?L4s)4-&0#Ud{`X zUqxBEkoinig;yypvIJJ`SGrVx&FK{j10wYk06~hzOWu?`0HDz6v}qM+HrNs=wdM>I zvm#)r@2&A|jDINqv)%!96MwT2O~jerK~r**ae--F!9?n-kJ8I?#Lz>z`sNO3W}QA( zRDsCpR`0c*^n#kQ(0FQF!JOo)tIe3ip{VB}BT>&mBjrhc%=dzJ!V!Xhf?0+jrgmk4 z^};2@fDhgcGd3^VQ0r>mFE8J`U(-kndQ+03%@EurUSC(%_&726N8I#yfus{i{%H-B zmk?S(uo-dClc5zCEUrn!j9-?Q$I@{by36|fxweX+&kt-IrBURs#F;)P!J-Rcm-FGd zDw2d&&HI4NfUJE%>Yy)wr&)B9hzV6`vSf(DZEn_$bNNidpwqHVOqut)R=4)+honkh z*HE|NHMhMl8@EN)(4Gx`a@FRnE(%(+ee&zKn&{# z*{p&oG+GuzU^7^}_&qwOELId<02NvpB6yEK#~^g_6vvb^Jl$sr&sQ#+?{ClDkAJN8U7tf- zs1_Cv@0H1&`+quJZ71~(&}CGB?edNGfbA~IH|K`;W0xZw?IVC3pYxeH=OLy3OvK_` zTWkhTZL@wo2gQ4T_`K=Htj9gi(M|92@y#jQmROUbBDPEOi?>E1=`N$%jUDy}@4?zM zx8x*3-dfWUq%UzI;+zqe!>ZE0TNg8#a4u^#{Uxq@_+_783k$i|O;V-jxj5xdk>lUv)5I68|gI2lwtQW`Y;7^WwbGC5IY;H}d$OP@?Dm99g!WRcVv0no?cJ z=)+!dSnX=$FjJ8Him#;|yq1+%M^SW8R9mZfd=sHP*EQYh*dA)S*}mLz0TKu*r44fV zkbKs`+#fA}2E2-SzcMUlRJ3W0<@f_U>NZ!&g1wdxa$`W9OEpbH4)(j7$*4cM*#Hpt z#oyA!cTazO#0n6X&lvx}Q}R{UL+Fz3~ksMDR7X4KB&M3U#C}H`n@q$G7m(T%&_3cPS!?%uK)qQ+GdE z`C(3QebohMn##G4%UN$~NnRYhW9?D$#_(!*RLWw22Hh%0S+xGJCU7Py@m`m*sN*gTd zf*2ux<-AB38BXm^SLGEne@yDi&i5433=>IW%Eq{_wgbbpu4V-qm8SMvUZl20PFco2 zuQ*hqxdrJY2Mya)RTyDCg-@@l@8^7;mUQ!R^Pg;(pm8CFdA}D~&4{$IoKEG}U>y*O z`f*X-^i@cT_ZIPLd7rD@Vl$CM75?KLTL zn18Wg{5P-tr~A$)b>#P8=k8~MO$Sg`L!lH)2A>Qe&rhzx4x{`51sQ}mE<3&BMM2GI z4ph5F;D0vtpQd6ADw@>Pj7*?#eV_h+&cj6WsBQ)!hg5l{j!T0Usk0L4Y3QO;X>Iw5 z9ZTyA~Wd4l%}rqb?ej$;1sd=F{0hacz4-E}qNdKEbljZ3No>inq<~P%raWdcQT>Z74?^mlgQ5t`^ z-CNR>E;v0F(6Jrnp8@H&v)3M=^bp z;a1FP^VMtKi8VkECUcW%o`nu4d`Ox*o6+y>@JHGFn3X#D-5Hy2SE$tsbHHtHPb)>_ zf(Snzy3(h7{wG_-CwHV0XGY{-otkXZ+`k&p)=V>rDn8UuIiIrZ&C}+8k!;5#Yu^4; zxodL&k5(ePST!xTPP_JvH+6~p>G@9mq7mCXjrKyhKZZ27EgV(c1Xi9Wa$(@_HHb7p z5$ePlPvsNK1_33k5-N3@>TWQWmK$p~C%f4J4nU#cQK5MAh}nnaw-iP`mofUr3pc1b zUoB3fe!#8(IwYQo;L)gmOmd@}hqNXNpj|Kpk9;n7e>f3b3p`iShlQ@*fW*Nh>40^i z8U>+nK2Wn#(5MioDA2Y^{AXs9pD=vOr4hHfNZh_LUAdVPc;|ZV6mcB5OO)iTxiGEh z);pa+tl;g9d#9MKf?RtiZ2J9^U6M#W{x8FbrXJ%Dum1_fzZLa=gLyptwr?&W>Q0v^ z`qr;h;Svaes5;EO-Kme)9ZGDyxXDc#~MG`cf;jx0?fL_3)@ zafQ53w1#&BVRI&bX}etdNDU~1veE#2eQK=+g7wX(#G+_JLtbr5_sXx3Af-xGf> zE(6k_tOgmXySd8mFTP8A)oG?I8yBhtRIOgG(?C-?gCPnFeqsx#BxG}Vcs8zZCm&!} z{$*PESX5Ck*$ZiMPmymR5+aoA^AY{a<1-hwzFch;?g*=YEWvkQaaVwvC;ABwJxbw$ zB1nY5FX{y=v-83;eQd?Qz0mqN$;Ak2LA}nmuUg)7(e&6)!||WQ$B$=;*v!WILzSpa z-#g%nX(kWI5SicX5Tz(jl+gX;v4Rl&D~nZ9)3N5aj3r)45{hB}d>qaajr^M7c`SB_ z_qe2;v*~(&6Qt(J_jCbhgAeh1V3JlkKuw`uSJBT_xgN$XCvT)Z3$0fM07#wswZnWb`CYRb>30yvfa%#PExnp_z-i*eX3=U}GGei| zKwsEr=U3oH>Bp#rbCH*KuD1;6-5<-_$=Fy8HDjvF$#lX8*h{k7%u@%_)qO_Y6(zv%YIiu| zDP6jM&a-p+fPXyxskx)p@BG5fu!vyxa%Zl68dJA26oh~2F^dHGAkkDpz8GpZjvdLK7L~T|do=lV9U4XaJZbt&mm%sPj zPi?pX70@&b<>pCQ?|JN+OZWN$P~?hPd*bUmvfF%eii=%Y$r&8%s%cc3kOQiIc-g8+ z{qgN0p_rxjo3LD~yB00@FT7k2@NAA|5`S;x)eYK2G@2PYxx3>8MD?G~k30EKtJ(p7 z{bM2>uHtlMz7u9F3hi$^{bwEpKyFmJjJo5W)2{>A9QqV_CwWmVvbLkM-nbEEQ8 z6J*vMQe=2&Iso!$@*H1FcHEy~KOTgC+ItG zA!pA68HzK4)dz2J>G_6uYab}R6j2eZ&YiHQ};R=cXn-;7@&-(IUp zXRWmI`*qdHd={yH{s88Q!HN@q0`uoOMt!pvrVD}h>Dc{khyvl~p1`t2-_`4X(22K@ z-oh*=54CeBxER1J@gMx`_p4a@ClcS@Ff^li`YPBjy^MCAP%`KS{+4Nw{k*eEtSU%& z6#h#UiE1Ir^O6_|`Fn0@+sC8CkStI0Wx3D@|X9BXGX`e_0DA=TJh0kzFG&GB6db3iDmrfo2)FQ zH*4!9um)fKwby@d*^1n5*!1S^h_1Xld}yu~NMNS}8=pYgO#DdK_u#pIi#3@+m^GT& zjEH8r`_zNN;z`H6)~~ZPDx(k1yY1~d67i=G@xf|2%^H)Ky74&P8S{HKBH(uDawW#5 zGE?s`J=Vbd%3}Gs4}^}$6IiGGUnv3!Sx{zew9ha?AX87J;;5=9{P56}#AJt^z%#nI zh)6fv0cM~&RnRnVJ}-=agiOm!Nw3?9E^YN^H;~HqKs+R>)xoAe_3L{VR|PebGisbl ze57W3@|en_FgeOY4@3jO4Rt-3|D`%Ui$G8{N|#A5?yN;`)*5>{hN79R4B@RuJU zrrjN>vxFuL$Yjs=;?T@&O1vb^`H^rQS{_P@sJOB5W>Wu2u`Q^7b#{v{OTM%@02W&)`}DsyjLIFd4Yz%0aF*e#s5@ixmIiPA<~QDZ zd;U%vNHm)@47nyoWs#Xsv6>HQsKty-x%o|u#nt75SV#-p>)Od8c)t^#pXjU8e|+ZI zbN`DKy{Fjs`y`crmtGg$v*%Dmluw2o*G6@Lgg8+&nbXZcMynN==D&1OBxJj1GWr%- zHl6|o29>O-8g?Os5bnNIMP1XWS-VE)w8i&sFm^G-WEgBx7J3a@7$t~D(ybh zIqIT*K&_*OJl9PKGG;1a%;m@KhYz+v$iYOCL(eFqiSgrqN8~S+{$zv&%u>JYAD@~K zvkxd3mGc_~$JhYh$h?le>4~^?Yz2lHL?Qpz4*f3$9wE05`ULDq|4A$!!6*Ajb>;U& zil@6h;P$(y*kteXH7a)F9WM6$Fi6f=f75?&k$(XO=7Z-8uG6Q{{NecIlogw|ER|<{ z!XZJx;wAbKcJ7c+RSX?AH@FO$QMz5_Fy$xX|UcOR*V1Mfv{N( z<@t>+O|Z1eZKygedl!|o_nY5j_&dHO&CO+}^DzEfssAv(1d`}Nru+0r${)SG#IV(? zp-eq$V@bUG)m(YodJ!g+8&+KW0&u;XYYbm{;d~i?BUvtu^gpYE690iAnE#9@x|l7e zhf@=reg-X9m&&;weUUkHDt<4J` zb>F=6G6gq7!nFoK`cb^=)1T5gc)n5qCqd}9|B0CX@`IM;0q!P)rJeq`<(b&gq-pT& zcLj6Re#@I$Z^v@I#pHDnTDhyA451V-xrX&OhDAU>RQ`E%Ckj z#N3ZuRpNJQX6ewX4~7W3e^O|$)lg_TS={6BHfydkrv0C7h)>`4IotL)${*`bt|;SA zrd!uNN#Q5_;Co!$Eq4X6Z})hg1{XcYc8(p8>6NYAMNGyEYu zldj_Xeyx`K7_{>mHg9H z9gLP!uwh?7^G;7%)GcRDV`C3T&sC*==Blr%-vlB1ekw^Wtl_waq0`{MVXc4JQXUo6 zy!p$5(x24L?7ssAoAk5`KgZq~3kR^ZsyvG;f64XcjXpqMq2VlbkITsD?*Rm#zVyM8 z#y)l#e*j3-v-!I9Cn_xLg-2)6#9K@IC&KXVoTR_iGkxU4`#LQ_6aMH&8>7_g3iunC znC}MyCt7@TCwA2qnw7sHIwjey;3Mot?I?dwEF#$r4*}D2UJv}^`m%C%Bj3tR4^89e z!vo7q#?Wgoyq81-tZ9S<z^$o4yx+GK6uEVj0m`(J2yA484PGS_ub74y%PJ>#!oE{=^dI&!Aa zldt6GmJ__YiG>2^-$KH|{ucqL;)%>Zn7G2O`t*+&iG5y}B~)D~y6}&GxbT>GyZaCq zdrxVt^63B0UMxst=JF8t!moe4_5c>A!zxi!maDt1L0gTy&EZVB-?#S5!DQES994n3 zHu(!X%fEw(Ry5bY4*fJKWakuXvX)*CuKNCvhieOX+A6dhTqipYs885*bg(WIWc-~% zJuvs^bq)HaZ2y#dsu_WQzXCUR`^ZT5y}tKm2jYc4V2$>QLex7sm~%_Xg4#B zt zp2qu!jo_HJ29?a@!pBp;BLKlYZc;WV7}c}(*^%l$pZ^&pvy zrv~khtAcXbZfEl^T;Rke3*vY+UMsD^xKW`=W09en!=Y!@qJKkhs903lxwcelc7Gtc zNDB7vh&|r&psV*tyd{B0$yVEcyT-PVI(TFAHwt~^FNTFNNc3T@?oY{7Wyci9Sm?~y z`3)Cv)u7+iv78;^k5WX%0y!Mk&SP+MLhMp4_kCV}R&y19ldbLD2d&w^5r%}^ z^H4{yz@Jw5{Rxw_eq++{;W>xt(pd`oPU&gRLp1SZMoj)4IW0Yq(-1Ty74Pr#{2u~C zDLWSB|E0krf9D4ku=M3xEB!w-`fPn;j9^i>W!ZL@*=5`Ave{+Z_AT4!QkQL;UAEa} z+xGN(lS$spH$QfAbCSL8KIbMUxhre0Jq=C(yanlf28>p0rgkU10~TUZQphA9J$9cp znO|rBuHu(38BUX)%LE3-vK?8n;@~*KPZNq1av`H1pkG9aGT;yxARr(xAkiVbIuDu8 z_TASYAP2IbAfzDpAa?dHrvGNPTRSsa+8UahIy2hXo9o+}x;R-H>zgq%IM|u1sltMQ zA8UJB{)-m#WcrnBvi|}yuSfoIkPZUU_5cDx@t;7cRv`>{(ElebI)vdXUHre&py0>) z&00V}N{K=Le`){M`wd-<^;!R;gcTx2|CawQ0Vkk=#cB!!WRmv(iOuy?j}F{S_Cr%P{XNAF^4>tJK}Z^^;Q z)Xmb={r}yorK<-#2q?q@C1|HH_+nYsbOz0)pyZo=8=zB2cL~ zbb*MBIew3yx+0g~eo0w9`}lg%J;uLA&_EMIbHA;%Lvp9}$eAn6IBZ-_3odzk{2;92 zZO4%ED`sMTkKKC?2h92u+_@F{W}taAAXp(sKz2~Fy@kIK6O(!QpY!xw+g1nTsGaUc zqc?Gee06KP-H{MUG?@9!4cqZPXj`Rq$l#(g4}fr&%h|m`H%oi$Gq^+h+Xo(jQuvSk zz#*X@0)TOC1R{SQ?Q4r?iNw!o3sph4hlj}FQ?A;Z3=(%oR*Bivo8QQUvB$TA{Wc<{ zpNF-IqyE?#Y^&7=JUhB*SPY~2LB`Afw2iTmP1znYVmi zpc&ld|3P11M=6RCjc!>|q&!Gs*5(nSbT(Exm=LM>NNSUEuCyBDmO~T~P$@}0J3(bX zILof|{Lq6K`f9mTYOc7y(D6y+q>EJ#21Vw>CP ze>@iOsb;zL zrLgUlQWTQ}x?UUKB3_35TCVQMC~N`2L!J6{oze8#a;)M;o5mmzcjn>snhu#-EM;?` zV?_(7tnt47J3CE+b*fE@^~8ZJe#j({BnN!qex7EUKLtysX{EH2AUAnK&@09jB!gYJlg^?npy32Va5ZBB39ru8kq59&TdD94FX! zZ#@QR57fO=utpLu239Y+4ZaCE!!C*My~$n<>t06#AZ_cl*?Ff4Wg3c+{{CZ#S= zXCz9sqc|ATqogcwf%SnAj`NhRr25#=&Zf+*oV~&=-uo1uQ=Q)GN)m`>w?zre@u6yH znBn$UCwim=5PsXwv1_6D6A^z8$w&a$t}OgkpxhTaT~!?TwNL#uab%`svJ~g?{k^XA zs&7T}<~p!cw%HgK?L9wQwyBj3to3g-fA;04e@f_$z|_xACJC)h6LQJ+wtTMv_m_55 zKjv`r(L?2rP$QCww-A<7Q87?`6Cy~kqMZF(#_a9K+=+as+pPYSikgn$^lK-s1YBw%z!9)?a+m z_8QbB{CV1O`&V_xsHdK0R#xSJ)oG`l-KjHOt6k6LLc=$yA6HczmFmTx-x*mowCH*4 z719|e_1p=_q?>8bA9+ZxtR`HdEyC=e4WWPX3v9*dangk8H=MertgNOIa97_xPm7gv z?C@CfII<#_@6++SZ}NF+G*AVz9R8Gn)CiyHT2%r?kT?uq{N-#&oP}a1>ksd5G)H{1+KB_iJd4 zZ2}fd=3SN@Yv;4y*0yKKnEW*LbUZWa^tW^qv$G5AmAPq#O$;qN*>e@Y8fRM$YYl4+ z)8(CM+yd=;9i+#EM$^1QUZwdfH}eNPY7>hZ3&*@^T`|2yJ#8({Qhg5=m;N&KS70Qr zeeNJNGjs$mdNz9iNvT;1B6N}pX6=-LZi~%=D@;6EjB<rDnl^^tPEvA*)HDiyva7_h~Od3}x?8VcT*qe3fpCvmU)x5f6<>lUR}Hfe8OV z*WhBI6C4TpEl&@^uYl2KGqWCF+ppVapGoKk(97W6U-?-GSw;kFwFhICGv6nhl@->Z z&u-IOz^!3CbB6v7-`^p4_E2geg`>e@%wiYaS7X!w9E`dI3hz!GEUijC%xfGg3^Nz2%!Eab)dl1Q2=#|@|H_n6$9!*!=-u*gDMLXje*C28*0!z?EB! zfD9GzBp(l&bu$#I#U+y4$K2j=T^{JNz0nKC>!btW9&3Daziu8DDp&aVvm3(=^{GR2 zRB<$r5q@IMOEgCGDiu#K%(I}czPzct@t*JBaExMF2qM>KxA+lP7HiRE7bTe`y~A`( z3;SV{Nz)zggx{LFDbEkF16(XTXPj#!tG<4~7AhLW#I_D!evw{>bK{*ge!tbYv|GNC zey{6Sq;G?cVX9lIQ%iSA6{R6*-azfSy!@^1a5`Nh^pC4lg(AO^yC`jYuZp}04kK=F zF_I9x7;i_Y5*Aw9;?B*CKU3wdq~=&9JSQCck}X>&Zi3cWUjDwd?4p`VN((=eKNyOE zCXcSMPO65#t8GWQ%O8h<&OTmq-|boAObIV^9(6pJi$81c?2emZ(m_0-Rl-{1^+kki z_sPt2Ua$(My{46yoFw8`x|^-qcdd>cd*K6SJa$R{*eIOQc4QY!>yEU?xe(Dn(PV#r zxY1kW`Fk~0QB8BA4Y_uCo7np2a94{9aDrNau0U6K&QieZsQ!BmdmW*9idZ?BMK30g z=Yi8z&OhbOkoO3)EU2}yhr)f9_=cJ)0SD>Bji z#R}o*#7dzW33o%VO+`1FXb7gRooWl}J}K)2`^-I|Ppq;lN5`HZ4Zm`imVnY5u))*$ z(MR*N8%wh*<($Zx;geycoUDUor=MNTg+o|`;%UvM=GFFT^fP!$$6c38)G=3#P_Bdo z9+o`5aL`H?7C|^Q?JOwi%z=)Rg|eoFFmTyLq5w0fU~(&itgE8h%q~sdJX7mPYml(# z^v&;Wq?FCp8?V}N2NWNh27guuh=hBXhPQXSVRuIqdG=n|j1qE#E>xZ1J+6pSw-5na z-+b@K6*U$8UKrPS@8AaQKn^C~g!8s%Ut@Yl$OwC|!X#~G5PqzR)U(2*2t%CS04#D8>?S*|`M>ZH)s=;6QWZjTG--~G&$iO?x zGgdM){_>=W3fLeF2PPn&dRy8;b8+#}Tt5X&SdZ!VwiMGXGzwTykKAPn*`#fChTvqf zbgXoKoqYx}y!YnCoP==#`wSdQ%}DPIBGX?yK}1|oUz4}L-)mF^G-3Idl@-5Zhpl8L zID`)^Iitd~D{b!CMc^5ha@GSs5pDp za(;0iUOao8poTS~@`>9rdY-*^3a~(Z51dBhDMDT-K@Gh7RrLVvS;44g8-zIq`iOsu zr!R~AXL7T>ADr7&&Q-2RDQ|6{eC6GX@hYlqwXZzN6KHtM7%Xoo&qt2Wruw=ga^_hg zLi;nTJW)opLp1w&m(zmmkp8Inxi9-ybr!O@vC<(GMh{h%ZSQI#9>{>RV$ zStZ{US%62O#-k=ZG)AE%ls`m?ZSk|x6j`4k#bZmW4jC`Wd+JZuF7MAa-WmT&^RcT&#cUQ!yw|J5 zB-6mg5M#9U{mbvbfvQ9nU8A+~X#vn<#ufroMK~in-#MbJodylM74*Dv`YhBX{ z02m9cgo+PgU&1W(I3;w~{51Ona}rXWv|Fp_#TgFqo0EQHw{xaw zRVsfdCzZO|xglPe8Y6V+5WG#zSX~-)aPQ%^i}zK}wG`$Fdpvtx4H{&UJ~>hpj9x-J z;yId)a!O5BqyIHK3v-fHvtqUD0K_S$p0Gk?+?T^B?$bk>X!Yq%TNQnHo}sodnS^P+AHw4_Qol;7uX;*T1G=8E;_8+vD z^P|gszuzLcD|>e}XHl$E7~@})Zn=KHCUF$CM$g3s?Swav3ZrzU1YYbNVF2>da%?z& zT#(XaE@8nGGaz$5bNb>&9=G*ML;hIPkpjl{z+h1l;;7K$=4{QCud7@m)+=+@Gh2fX zU+56Cg38I;YtOHVumIvXfn%9?T)=8K3^kqj3FOk$!oN@d?if4Z1x=SuO(D)eHE5??498p5yIMw&|3Q>v1pA3r1)Y(01t*Jl=ds7= zGRYo&|QAnJ%NY%IL+gZhEo}Dn+$pQ}M?U&g>;&4JgbJ%2LzGNsRFAUTPX4`8QA|E^qVq zO3dgWQN}2P$1V~zEh5ere%1x?C9GY~i)aG-L@HN-XaiWCNub%QnUFi4VDe||8BEtq zde|ZP-*JwxOItAMOSnPe^p!F>nwJ>c=ot_jV?MFsxo};m4VAo%W_B8B(XdOmX<0F( zt=@ylKpG|on;%XN%}*`ZK>|&WV&Q($U~ehBZ_woODi~oi0vU1j(;WdPD3{CB3Q1Z$ z`%%9xxXSeJsDStjj-Fgib*bmbxtSej90LUiJB`sx=ezW7G{@Psg(?aA`EZ?49Fwi^ zsG)lw(3l3-Xu8P-&)-CQFRpP~6YdYUtGG_CljP1-YD&}stohIIgF zN+%UTCm3aw3xBfADQbk=OmWzz;!kTccYMmOJ8>Dv2Xaf>C4R@Tqm!5r)k-(-o#cji zP0@5on^f}8JFNUB&XT5Oz-l9y(6JXV#-I4dZWFr`Re>)<%*~JvKXt0mQVD{jnB;k7G^o_gD`tVh=&8aXG$@xL%~UbrYXh zk_mU$Xu@bad^Z|GcANfXsq)(qjJ%fgSd)M953irOUTLyXn8&B((oGQ2*quS=OJX{s za9z=ux6qqY@YQBe>Cl&5vh3+2CWMti4Ts-I$oLT`yf|J^FJnksWwpBop0Ojbyii^X zo&b|1)JBGh^T0$x4~f7`C@(?@g06)}!2xFmKKnhd5I?rEJ}M|*+wziz2*Sg2U-p7; zF8I&umW-NcYt&qOY}N%*Miw3)#ecL7A`#T0%>N9MpI3hkymydiPq0cB=;AD}N};Ge zbE~=F-KfF5xN>h?;lm49{AQA!OQ12hSptmqNVMVE=RSiTEv076W%bZ=kY$<{n!``TWc7VlUm?r2w>{Z!~}ez zEH2zPs#-uYN(rV25mSw0MW2H;NyCB|zB^WPgEvxze)DqnK^Oj(vCk%YB-hZ{tie2j zf9{E6c)KgyoH3(r#*-LVY7b)WnpyiTHzVxrTi>{OQmC}QGuw5eAbM}Ir5snV-Zar+ z2UzuekTFRIrS#oX8o zo2oo1VjlW`l^nB2y5nDxUmkQ}D5UF+P4CtRyL0=Sd2@eV72IE6Vh{D>1Y66_Q6L)H zTiEpoblZD63~~ZZlL>mx1kP)Qiw&Bqp%immAE6pwoZ<6dcqsMikQKCQ zxZJS(k`=#fWg=t^0^;48O3lKxnhYI=&|F?JvWDx^TiLxBaK#gTNT$x0_Xs&sTZLZe zFsIKN|A_pXU*lSM%cuRvA&5vMj4#&UvHg^NU9vw_Z?8)$is~9cfsO$1rk7~)pCYyE z)tcp!#S+F}vB zZXT#!Vyxl88;I}7*z$Psn|ooOgqP)6QCxq2>q;W$~c-MIqXdt1G{c&tFEd16Kj=x{Kyjx zbQ$K)v)-z*`b%(YQx`@v{ZV;7>pd#G;z2tu9$==y0)Dn}pH~ie`{k=zy(grO+2@lp z^Axy@e%wB)b`8WnDa`0c56I#`?PM)RXDYr}xkz;j_UHhJw112(4)W~2BzINR&E}dC zei-qQ=iZQDzXa*hBATVFNI)N7cm@bMmQ$bw`!rG-Hko~{XIlMckWgAnkc?)0FRJW< zpxFz~!PchP5f|jN7kX}a`p`*A*EQLU*zloOE_Ejxq943Nv#4iM`%qBkiezaME>)QA z*9se<@yrAk<|+7VaDC0hsElhZmt8_(=NcA}e!W(mY>?{C?j`r9d-=_J1Ud>V{)ao>3S2ZX+F(LuQ6_sZc4dAlL zqx3hHt-ayClWEO`z(hj2Kq_Sa@vBrhre^|Mc*= zVI!9VDKb}m)GVLl3#(kf!h91u?(A7|CbL>~8BF4R+zvNju!-kYmGp2?&vQCUzONM{ zT`{fQxP_&7K=qnpnsytS#U#1Rh*~~7m{xorGd}y!Q>aho8EYfi*9&V9i2Wl15Vpe~ zt(gEkSLU60Dx?~B(fubGkCdnJxzTuC*gCylTUJ(Yv^dAy zo&~ zJJwm@I;$ITDtj-Di<{@=uuz8O)NYFjK>|RrTYO%Sg8-`V=lSf_UUNMDcc_3KGjwf+ zN}PCGRYje_>w8e7$a&R#|=bSwY1@XRtdA?JITfXZf6UH|po$4T!>(ji9a3p8xg^HGd9Wzij zjFVV6M|n6fzF@?q|FVcJvu} z_BOLq30~j->Ch_wDLoiez`HuuEBvaMqY%u5J3Vvg7j`=~p2^+2%B4718%Do|_WRBH zdHmb~`E9KJ;y`uiGQy01Fi)pUl0c7qf+)IBGNNS8ax06D{LIDQMkKxf-x-imi-7vs zk9vw#W5$bx_v4wrh;(|ry2lHzS}#j2Ec=}S&gV-5gKFYO#f#lK*Ub zZ1ofeBMh=fX{m+f1H$U|Kq&6oJKcw2syYL{`U`}vrJB=4(g~|>>N8Tga9E^1(fcW5 zKyHDiyxFoa{!-;urZ^7Ba7Y5Vwcjh34Mw=RUD`B6jxU=5@#{i)AUJu8h&scxN z{Ax4K=rX@DqctRc^eqY>IpOAawK#7eqt;sTM1l5Yj9JItyO?Yg7HQc|4Q_&W4{YXC z{mtnPQx%LLKBg9k3yY;8W?C_7jJ_{|%^X(mL?qIOPKFvo zNMH?eyRBn743A(@-6)WuGc+xdEAkXA>|q)-TaQp&-B%Gsaus;}sIaDkbY4A;w5(gy zFyoL6!{e-~-c*gI>frROTf}9Kz~>5qZDUx&hj+>5XIwLi@yB9R#e&QC7g~%rYVG3j z=@pPWR~o_E;6(Z|(VjjK+HLkK`DM~lIqin3gNGttQq8y;uagGc-Z78+V#o(%! zk%j)|F0!D!vcfph6uv~cA6m(=I{~#67mzUV95~D&zF*R%P{rE3f+pQ$mPq4c!11f1 zkF!wn2rc&ao1V=w4MEx4H_}8%38IQ;f<1l?dK`62l8q%y@SSjNoawgWfU|Z&!bU~D zu7g{P2a{`8MIPYfE#C1o%TAc4?p`UvK(kLBBt2C|b8^-u`=wS~ddRs@PBfo`=zv@w z&LefB(C3_v7x})n;9EuF0~@rl2b<&uiNC2_7$k+_P#Hu2l6PC$Y;qZj`4+PwR#$P$ z?!tjOLPzb7n9Ft^UcL-M4GiXX3jc;8OH{C>{>XFgI?efjJ z!vC=4FXDl8_*RW@yeZpQJ_A2X`?H(tM2n#u)w3xDv^2c%meID*wpabvSKpIq4aP z4=msoZ5ju}D$2Cma+HJngPjn}c26N72Y+FQYtb^W+hpES%)--3z8fn)L6Fp=rqi(g zBT1;QKqh?q14gg+lBB4J*qy3D{9B1>P*dnvxU8hh#(DB!F|LFW{_gP1fNu5f&Dtyx=!^p;h z2769pVEjbrA1R%i716wuf>b`XUAczg{p7l1-tKK$wT&a|wb=6Uzo*(j&5m^#XX^{7 z77+cp&Y%<=*q+Q~Fq05`(~ zWvT;Dx21`aU&WF3yPaC5+lL@q-iLI#-^^Uas1(0tsRK%(QK^o~Gxs-tAfDp#5xNSr z29qZVd$baQwASgyM)K)nx!91J)|(rCu)El1XWXnK;TN@5Ld0yL6e`=+fT0>~MT|FK z{DhcEYQ85|I&0L|ZkVd6uPgGVhtp$tf$RbXmlLM_kT`qCPwYAws}ze~0(sy-})5}xl5yxmh~BqtI*Xvk50h43do3gw=ItWf|6 zhqg*Z`qi-;%HTbk0Bks)cUla^AR_)fsX90}zN*p$m{?uBF#_j1pAZXHVi#r!LH2FX zEs3&I$2R=jsaPiI*80x%Xf-pXOZ=q?!59Seyo_X!*O6ECpU5h^AFmlRrOV-i`wM2RwGI zH)D=Fg(B^RoqzFk#d1aNM_(5@09j zWi(%zI{8avR7yOdhsdukOr++J?3^_z+DIBoFrqI(q4buh-DZys?L?$)ccO*oYnPe7 zl|pPtP8r!l#^*Dr{WL*NH?9THW$B0ZKF-m~Y?IVhD%6yl|fW^yR|989y5z_J`dwv8{FkaL9I{?Y-k7iTG{b`XaW zt(WrwuDb7C^85U^XaUtT@KrVs2rcpk6Kv^hiFZS&c|E#Co?zceidnYo2J>Xm4u=`-*O}OYijw zci+Z1UXSVVgHGqS^csK$-j2~bq19!sGiGa6e}pgTABvKA$RWOP1CG>mN@RPK_MbL5 zlZ}2}2W?rH4R!Y%@+w>kGUwV5x%jDh7)dGY=t*c6;KF4;Fx9P&f)UUvWz+C!W`(K4 zpFQoy;OX_WEkPf*R?^lObfE0g2DLE3nYNDSd!$`9!uxUF-ys2dTAgHQkR-MR1PMfV zh&TcIy*!Lv$^99b0)6=bI-beD(YqteekCVb2&!;a}*wi3JLm_f5XM*2KzGtl-Tc z{5AR=6!wTb3Wfm2;t$io`dBqM4$DCD!|De74Xlm{8@#~lSfqy%5{+aHiok{kmXsft z;@~3gZ-;wkt_Fz=6wGcenPmJ*Ek-lz#9ss@mjjPi>U{$KQ3^y=Q{P=k&Mv6r6Z12- z!3Nvtwhl&f!(r-ErDU??tvLOrD^D-xyrOr6zwMrzjdTNG(|nXpJ>FZBp!IQft^RE6 zS<|%o{X=V?_J!evPKdPnI6pwpfA16MHq5T!Ow3KdtDS^xU(3kMgfGl3U{5uJgk_@2?X8qo#jt}$9K3?MFJ4U9cR7BP$4MpU+4b&t z5|YN2kmd}uRz~UYQzES`__r7PVaSCYz)RckPd7N`IUsFjTruidioxDbAYqSXlD;$) zy5X#TN2X6uLpEv{jO7>gg(zNE*KuPa?zRh8IM8c-omWxv5@+~1Z25~kD&0u^>5R}! zO*~ET2bEB#^Mmza@T@SHu@%oAM!(YE)07H<9dSM&WYT8IXpw_xQqEa4QY~R_R=?4) zk5f06D!o$vfFIj6nqZW|w#&B-XI;+2N}b>S4Q={};kS@cAv|G>(yHY%S0GxBz`pM> zkMM;-w#xW7Uu|#l&^;I|y7kV*vz>O61P0hfCgM~wq&HxHQ>8SJ^?&|0`X}G^6~=c-)66-$`Q0cb(cawYPy9qk zxy1M;I9@UnnWDunc4KxzfR?hQOJ>7wSAJ>=*ir3o`BY`8G-M~=@k){+38B?JNkL3@ z%j_GKL{1Y&?F3e}HzaIOyJZDKzDSBMn){b%vWinEzs|vy$|G34JSm92jOQ!Cd@?tl z>d?O1Xg@~{W)X=~UscdycU9Y}=h~4WyH%KV@|*ujk;(bSq@%X29mkis)N?h^CnBf? z)N{v~64d`-@JXpxt>`zwkvkxbhxVSqykB6Q9k~5RnGgbrpwTPy!!q>m;0bd6%M!mf zp2E}~K1=XFp)vLw_|+yAt9VZDJkxv=QbmpPXvBTbY3{8fWw3A^$W@s%8YbY5e*v8eZjYH`?M}Z}rJwcv++h}2F9U%sm zT^5iaeiv74(b7u8V0tJZD3ukNBldy`?R%2hFM&^GR+m$y(Y-ZT5IQ$U^9v>b^6`l( z`Nww*pW$V_%7Kp|a-F<^j0WA42e++JmEriSt_OK^n};y-wK2xHI4aZR2;4P>in@w*FT7L$IU88o(fuTI( zy|Iym2=49pkKaM#+;sy#Fnb}(z&3KGxpO{Y)rc1fxVo~B!2jmd;qMCrYw|IItgjLi z|AxGizl>FdkQ57&S=0)c5pcvVvT7)QkOd9_A)jC2Gq7Ai*M}( zr{UToqyA|>b`7Q)x)TdHHwmghfhvBJ$G7<`&4Ez9ZrM4s(c?_DC+E2Rl>9|M<^h3! zwXaf}8^tB7Y*=`g>32o(b!;~FV?W%zR*H(9VZJ|=H5({a{3`nSLE!en*m2Qg^!257 zUZzI>kAB}m%BVr+IueXFQnu)#x(y`|L_jn=3^%0p$iP>{CXoJ3>9o#{JNd)hvUjY6sTh3QpEKHnmGA2`J}GQZ0IwFCLZ z&n?!wPjQ9KcYcZ+2r921J5`lLy z$3Ta#njZj${mW4^Q{T z9>aT^S?JuNtdMSf0e|?aEJKq}Ih~K@Il@Milq85TgWQ&;AXVCS)rBHWF@YG6a(nzv zzdJxShU)=`VIMva_he~j;LoDoL#6Eu+5V~by6Um9@Zkq_64Up7^bBG!ik*sk4p+1b(H*h%&m+O8g4#BoPQy2LTH zrEUk~9;+!Q-i^|nK|7Nsj4RFA&7M|piywe#>Ie+e0~+DN1n*??menM&0jwW5w*)7W z1i`a(0|)!7;~94k=eK<>gV3mp?Jd4uy6#D_*pJr@BZa}!g;%WNjJE3xP<1%z)0zRF za{|6pM^}I4!v)&`Y%c!Qo{v-Ged`*m)G-g}ft@YzKns#3Or1~tXg?_F5H2h68a;q0 z#B(|wKpr&ANaq-lv=Yo~t<-i{)3uKq;Ji2R{yjn^TEpSC`e&EjCiYDFzPDfBb~UJ% z+F4j@X~h4>`piLY$Bb9wjn?h9Sn zBkP7JxTP&K%5dtylsO+#%)qszEUZkn;bA`YMLMJz*nM>Qq%5;U;i;Gw+6kL#Kr4vX*8^;rPHIp>zXODkdi4#N2WIheZY?ke8Y{`7%(()VF1<^*F=S<;q7ej&Ft)+Aww_Tl27kMwQ7{Dg&ca757gwx zq`m1)uSFy7Y%Rqmh-Rk6=uayG;;AIT*jjHtu(a+cGe~PB4{Pzala12`)+5LHHsl~ ztd^9t5JV+KE5G@q>tftQ?JA0M=5+Vp=R{0$RA<7foXN3B$%d2ha0!b|Rji8yB|j|_ zrGLt?P5I-{>IN0YVFPi33>$&pIY8p7E3Tr=yx?o?f;z64XYv4a(&bH z`Usui;x(K>B7C1(xTHywsP1%z>4n-p^@%&#W?*z(DMsRnG<0Q#b)2c;xZH9{t87|e zj~1E=uU4LW?Z(AWQb1rpbjP%P(p#H3{9Z(+8IUxX*ej18ECiQXw^_V&*AmwaMHhsg+cp z=TL@k4Foi0kqc@>TH9Y!m{*$T5laO!E-#m>L$MUE{wX`|4h{ytGUDnOzZs99c6R6> zTKUPT%0=+&=fv4*|r0w2Og^3aR@xBur6CDc|KM^n%J8_9<)< zT%WOz6yY7DcQ!h74UQ1buhxcUL=Hm?)t6uny|vuHXNKAI&w3$~-^HGsb}k&Se`&T4M3FvEIGfBonG8&1Q&z~c1jUt5hyUKWLS=0{?qX99RQdw!+N(` z(yt6isH)Hi7^2r4DdfTSi~TI2_eDl;#H+qM{1Rls(vuO5#ENIGcn}74rR3lAG0IQc zCNk7*xQBT3CZwu-190%tUgspmzJ$jqB6Sbebl{SnoK2A1s^5Z^Q+%!#Hw*o?l#3MX zw^FF>BU)%;qdbiX%gV_I4uG!o{D)q5W{5Cqf`TVn5{XA7Et?6JFgA)b*+Zmz=8&F3 zS)9M9YqMziMoQgN%#zx8Xv4}lYG^l&qh}Z-g-`m$91=E3JGO`x5)Z?<*4p-|Lhu?% z@^X-d747^CG#OfMR^O6(`03|8v7@bz*aU`#PSR>6hrD2+NWYKVIRg+lx3(;s@oI0m?SVePHBn5)(P6E!arQ~`OVx2}#&PECmxh3Cct z4HckQbTWmTZe+`WSmk<0PI8vpDz@H3hdElbD?S3Xo7rgGC{!6L(YRZF` z(xDKpE!v&MLmA&9k3EkN;CfsEw5L<}(A>D3q_HSL7jX7n9A<7En&0yN9ddVT<%fn} zU~xB$V72rWTL<;8iwE^}8ryV^!r82f23j{*l!Gf$@~t~^GHvMCIVd3^KJ@71x_(h=MxKk=5=x1GA8|2P~B`z6KrYzk6?@e#XyhRERQ z4{{G(y92FZIn#~H_zzy+80TIpQ7pQg|J+RHBCe8Xh{{E2UD*bwt4Tiho{M(2yQ#@U z`Y~yc^OB6>L`*~8rdc= zY~_t%RW9U3U0L@-6hSPzlxzLrt~yIXE={Rh(9NRfuyKwpv3%FA#mQf8Hp9*CN~H0Q zx(ATbDahTlsiY&Av}_2Q(K)?Jc#B2)(8^>O{(AWLxBE~H!SAIt^ayLYWCOn_NL(el z5QV~8g@T>;vubl>BerxAU&4cIi9ER>$62-N!nF)r4rNnmq)9py1M{8KD`Df~BBa=3 zA`Bw8|7usZ0CG|9mKye~T*FQ&b&U6Qmj<{ko;)r%wcKsj-C-hDI|cI}F_2>XG2NZ% z{6ITAGzA{dT%QZC!#27-AjE9m5ue)?JY@4;D}PEoKVuI>>IQ>izV`O9_ z)LS2*Lf3Yo78f4Tv8{#tMesKTcZNsx+z%LSwE?CBpMrB5$?! z%`FRW1?dXaKpfn-4QaljMuzq(KkBE~=>y`~e;`@-tTR)0)A`Hk*{k;Tl?x5A8G<1i zbJrm;WOZ@1_swdeM$)JrD2$`V=^f958N;2X-UJYTq{dcQc~o7=o`}x#SiEh>B_oCB z+JcR@_s|KKY312I@yVM2lIV*;?v-XYxk}KXD4rUT=*V+U+4nMyg3vQyAg~+r`Vi_J ze-A`jC&zYLkEw|r`q-`rU<{>SuOI*>ZDhHlYdx@h$zoXCa8Q3bWfHbdCGA8(_^v(h zanP>ouq3{^I)kxfu#198RQM}WSIT*p)eu+RCMun=fP=uvv?QyA-fhmLr$u=hV(DW) zzrQ#+u9>r!bNkhbeBulPeRKk}%W=oYe;d!`r~`kLDn>6nM&G)uV+4#8obRrH_|D!K zY0)h9jsGV@p<&}T3?6xOhL@|_lIx$38>cLClI&k9|Hlmh?SP$pF@Zz^+Q{pCfE-?G zesl=mR`g3g^=xB)QUcBYeH{SUkIz@D`f+bw{XT1|jDn5(!fSGpd!^yyc8#Zge_bJu z*UcsA11S4#z$N2@D%%LgWB?}!s)-3AS8GGSy;H>>TTTh)s9Oeqp*2;-#)?@k28E)B1qh<3p-CN4AZ>6H^ej@?Y>PCDIe;af$xLzks zIU5>dCvC68l+*m^nX`cQt19wQ(L7`glOVuRkT$2V$f$Hn@95`QOa#+lf(a7D|KgK1 zLKRC~r(e;eumujf>5&!Fc;-`{vkGY#2JUjoR_0>)h>ak6rhKA=?IM$Tv``m(N&{lM z69Q)c9eTmll_73$lV$Qkf1W~tdo9ag%tEqeqWGRWRd=zj#$g?8Ks6Y3Eeb)MM|?F14|Yag_zn##9S#W!{4na{g&4bSg2B5M5(ZE}LL<|vu}g2&i<)R~EyJ+2Fwy9GkCL=~i--tsIr@d}Ax4epA5ZrvHri+NSTa;o!P zH*)A`i@jhrtS)jDe{w2gi!VNhEnmVD&X8Lv3f$(*y`Y&Cx(gf_3H7qB{p6mt9zEhR zn^)AWw4K0;Q{+N}5iu33LWNE5+0qK+gHw3lA4YtAg3C08JX*WBYZJ^=or}}S-$bTE zrE|J|H8T6JdDeQ$n}`YB4M8LshaW#Zzpi4eA60gAMR|4Xbg=%VLxeTJt@V8Z){gff7qmZ>s}{isJzhjtr_79AfVl_xjUQCEs-rZng>AP#e(~0{=>#5B@-0n za1*T@bcU7>{y}NaMW?x+N}>E@ACtC7qSR4-v$RuCa-j0&Qx6h4iP#8SQS#Rxe6PQM zHlf@9h=3^#_<4s>U=X{MxvV*6-Y0`u9z?ggfvgTMe|@YqVB&TF2+5!)IQk7`k6Vav zUQY``poWsYd8J3CP+QkbfVp$gfcelmAn2Vj$S8C)p-!Ih8~@BEG^@u?XvvY z!8F(&UG2rRNtn>_@j6+w_vLax0U{p!K6=y3-L?3beaB-$ze`=4Ds@G2x~QkMhJ0QJ zxC0fHe+$J*+Wtxhb_2o7?8X_E$m$BYzAphiIYlkrt(r8@RS$e)j+QSx79g?Ho}C2F zHhh>v#in|N1lmB{S?*gL8)@G3ax_#ibf}sRWFct*%eq3?LzQ?SXK(a(s?;Vn zE)z&_X`Guh^iC@2Md#=n+NHob-L!GWc@E1^NaSTxh-IZPOab}5`wI46#IIaWn&+Q_ zd4onwu``+ZEru>KZ;z@3V|@kl6SQ%(U)eJ+3#5J^iC27+R?Zi#)CssM?D7tQ$rV^B zf5Jb#&6)3Xe3u&=YpxQmQ(K(KGMO`6R@o3%PCWZU3>nQIA@3pvuyNp4Fion*fsQ6P ziqr3ItW9xaY&P8uQ&@-Tft)o{0wllAGb1zzZKc&MEpvIcBePyoBf64zNDa`|? z4=1Hq{XR?dA7x+#NfQM6)=KJsrVO}%G6;xF{$@Yc|JLI-lt$?tto2jL`F}LLA3*Tv z>{VR!XPPFRYKX(s8Ac}Z^{3#U@wEtg;7vNeZl5e)=%#l7vC*{$c5&_>s@|s-!+Y4| z+w=}m=?1JnyW2772(IL|e^CEyUW$U_tUPk?`;h8CfBYK($R~b>qBdpaO+xONW!ttQUf)6tyi zcL_}WT$6g+U;!}be|}QxJjnPKhHYgeCMut^0Wd5xFDJQv@tot^GmttnY8s;-?JQ&D zx|!shOxfVQl@>4prUPnjD^hM5_e(PM4Szm-Ne*h2?f4@5`@m_b}I~@`~ z;rLDl*27wHl_?*{RmnKM={`#t6uxFm0=EusgH?0JFYa#GBVibxE4LSmgE@nXC4a+X z+E;pBll412IF$ipC0tfjgELN0T)&flx)RLd1wr!70!vLiFI}{8|4&x?J;z=E;%7h* zHOn`xf@LDOe@q9^`@8mpVNQB@d7k+X<9T3+a>dW~1{`;g2EaAQ(uDRQ+}!TpJSaiP z`~9lPfpKxSMYof>Io-xQr5CKwjd&|=Lzlol*MquE`ET9>-EHMEzuoAXYpt00&*%&yN~3&sdeO; zZ>gyp|A(;@b5~H(ESr-o>U^z8sS$D1^B4>F{VL6{kCt%@2Q=!8wSh%5)w8wC-{xgS%c7eZSt%_A$W~rnu7-^QdZ$COmYI*Dyl({8 z+%6FbQj%y8JLd0BQtstTpm%P?L@y=3rA-)@qjY(#eSdXcI|&vvTHAaf_Q7{M!eMt% zBDgzi@n)6mUV75N-KKf(466B8VXw0Z-4nt5e+^}!uHUX%-h2Zy#izZ`CLIR&%w`-X z^{*4Hj{reu0qHUO)}5f^q=PICoaGcL{wNfl4G2jPx_x!U{)_TlszlO4`sUoR=cw36ys zx822c9kxa>feUoV`lwP}_33A+BK4NWe_{SJ3(x(G%lBUsl_0kdadD_~`Bh+|w*>LG zp-<89bgj!c-A=bigI|~V*>G65dtL8Oni$-)3jK8DcDHRey$TgjS2BK@0AoK}6vQRm z@OQ9wVmY4>*|HK;<&!$shwM}9RE`e9G8Q9udn$@*b`3?s?)G>n8^FqRuJ|B(e^BS` zcgg&DJ0@@k@-3-=lXc2Hx8}2>-!8EXeFkhZS7W$&>zZvi6UZ&HG)M5~qBR|wwk>?w zIz%C`SBelJVEaGC*xV3>FBZM;iQ$-@J(mUtRXvJ$N{KlC(_1Xx6mU&N-NMh|T^!eh zR>rGi1o2eX_J=kFMb)w_&>MRoe{jt?J}jS)?JMmfb)%SYrC>b<|36!-1^cs8k5WO9J{& z%Zw-*BejZ!hD0(*;JR!2h)ka1>(H1tSm3%%xw+eas+@VVfW)X|*Ntw-f14OtBE;z$ zz^Ds06-tQ4H2-)ui-;@tVyS!R5n`q;mv43Bz5?{iQ`Oq&L^;m=Ubb@YVx-CaC{2Uz zfVMhopiQnsC+3Z&1G5|bI_M}}Y2y4s)ZxLhi+P8;VCE}p1|bMRPvGFO?*eCNLbJb3 zq1LG4J+;dKHs>72dVI7^e|eMCQxmP^G?Kq~3f^~VgWvSHEXQ&MF-C)#>uJ49?ki45 z{mE*Wx%Nlxnz76gJ57IXEvzzi~4>4(6kJin1 z|F}Ijh~Cy#k)E`41K8Xc2K!Rxr+&GlxVm|8?X0f^MT|2H!O~Dye|4(GR44d)?^`PY zF?d2~QNtN(FZd4Z4egp4p$GN5=h8TN7a|8j$%K0j>rtAD&ul7HNOf|@Pr_E;c+U;4 z*Y$B0pFCnH{Fb6{lYJI#K5Q#uVmN>LEY#0olg|L>KGy&7V5Y!tb=Aa21u%+kMEm z6N=XHhN}hDaC+pOnVV-^xcOtbmYB0%8lPFAi@Az?f?>wtf0g~3MMbj|Pbckt_msVa zOxy(t8eaEB&kIioA@w)f(mDIyEUofp#OgSumWXGWlsK$R+&Ij<1`91Yb_#XXm9VSL zpvBuZ{=W)np}ZV_3}4Ydlbv82?^oHsF^ZLoh>dW6W(qQw=qWN;x}?Kf0N16vXnjXA zSWyy$mqdy?f0-;ilzY2^50$RyW0&i-dWF+B2$;!y9WjAh0gXfR%EOVAp2h_jY45iW z6>H&g8`^5Vp}|$*Wv_I{nOBSdsZ>@XXP8+pmC-Doj)00) z8A}jNGj?ZTMvF|ZocR6+>UjJ*zdCO7^=lk9JOh_yf4v8}eHmef_T5xM;SYKQg?BF~ z)b*u1g9dYJJ#OO`bnh9!Vjy9F+h9uazPqy16l#vyq{MZXt2D1k`F=A#>D9&t79`+v z^*;>J*-q(u7U(5j-i3s(_YKSRKV?V4+P)0R;;czWR{W;q+(&p35HoXKxuqjj?E2Kr z&@*`-e>-uhLxYS8sa^KBBprW3 zyFG}_uvem;F4;u%5b{G+;bVIs&Q{m*kT_O9!t@r-pBW=Aw2I8a0tO%?@iON$kCL{rIuNeqy4WWFFpGAg;f(eK@QoDZeI&= ze;J~%$2QzlgZykdCdh*DtV^PF^O)^tZEDmTa$drRynPOP+vr5_I_hI4yq+uR;ia*2~7fsS71(nW6^TRqnfAaNKFcIhCsL*q>3Oy`0StqVmszRXe`PG#% z2T_~aveCkeCKK@zOc>Y)4Dq%KTR;$cp2T8Ux1xNGJ%+l3`P66%;fUXzW}`7xyw(w- z2>syuk|S#d87eluOfRKUGWykDUsH+DZ=2NbjYKE~pJL(6Og$V#4TW-s7ZN0Af8C)A zpHG>bJTJ4CDDaw4faYnok(Y%+5iVoe1u=7P^a%sKaB;VYs8)VCuj~|*3@C==C$ACS z{dN}!I#`*GYYXIFmAAm7W$HW6S5<#-7kQ>0MqaA!3*EQL+I)RxL+W`tY(rz+&Xa~3 ziZk{+_mxTR=Wy%C0bD^oQsKDge=}Akr0W;jJk&bawA5+jm~^5>iD^&F5%V>5EV6Er zQiWQFaX6X8W(f+Zx8wogBf^Jkm{Yn$dbWRLL`6oQeyt;qN88%Q&O*)5P7YyICyP*| zmXG}wU`H!9{+jLiLu*u{2-esBIkauhO9$DlA42AE6PcJE;-eO-7JZarf0wP0@)G=?%7P?0a_dbl$mMin$M)Ww#Po7HC^UMC52WG04RW#CR7>FvM4Nz4NF(VlF> zoml5HA~p`WtW~u6BQxR)#Jv=V~o4L$uaEQt%NZ2l|^r zP)*oM$>vXbnC>WAKO-XQ5Ziz9@;>gS+1wU{g5{@f^uUTNC!BXd7jA+NqshRv`}4pH z#NBqQ9}t~_yT^y0FX!IQ1GwkKJP@LJD1g*BQe*0K@$8tpSql-9fBELqgiFRy8Jbh2 zeEVM)pEZ(4Mbx0($VZFbD`&fvSRMP9zvbX=VS%;8zVzcTY3SNiRo=mNC!=i-X&??9 znTV}i3&boTpOE15(U@zE;>ONt#wiB)HWMwcUI3w)dl1%m-%N~4D5U_lJVOW^YpX6% z>(7i(etPdPAE^aPe?yZHuH(~7CHcxgajZzwo*}p!>QqJ68crMpSq7?elX6k{N~1ae zw)1QLTw*fEAw#(Yd<&C?osmvX5FVv{dzv;T0LzGtSZ2f4Z+pKzW;5!Y-2+Kg+JYS| z8{irNgU47{@S<8)ETs9$o*m&fy=KXAnzvOYTq$=<^jDk_e-p$zM&cU*=iw=V2^^os zGrDVPl}z;EA;5<0*8{Fx1?q&@Y!d8QG{>`l3aM_b!};{?o)>?&{MPVxN3+iUp?Fbb zJ4OD4@@97(GTy5ez!&S{p_Aw9e3K091G32*>tfBta0Vpt_0=#el+QJVb&yhZa}z@F+aWNXAR)?L>`wAnR4XIc80 zL~t{;wz8dXDq+Ynlksy&A(?d2t$dQ#Z7Jvsc3?d)bu@iwA!_qaujkwAZ`65@ZD#EMyp73G3*#4=8(6Kd|k4RoV?hK zQP9IVjayhW02fXt6+#qMfCO${an-`1*Bh!=sLa(wDb@LDg*3VwN^IGyjK0dX@VKRwIUNr#&gb6DsM7Mw|;axvUEJJ%NJe+s`G zw@w1HV)_7badpbLn*o~NbRTtFxkKhW@2KfjTrN0BpIndyE)rv@{J?&tcAvV)2^O16 zDGO7)#@KDY(=IVT0-~@?aGOvEkw&FCj~ESPslmH`o5%E-N9S_+`%LKX=M)*ZS25m0 z{^gccI8s8U>;oT%zW}d>)|<~ae@WguR6{1O;ZcbY*HAjEED_sKzC>x1SU!%R)_Mn^ z;KB7fkPvX=&p#?{J_SJ40?GiFyGW=R0hUgI=!@P;*~--CEnw0r-Sy`a91+t6@E&_YCwLf^KhS-F|z@)EP>W5JB2g)^5e~t}hlWEfk}D70*tI zbo)^Aphn|X;JgTANP;_ae_Y+q+zTmnfbfB7yA_$NuXoJv0^GI`C0_yfuq+UAvfv|Z zp!c_73uS@JzuTvyN`oRM-}sWv1UxH8CD6hVx5B_8UhLl+MPI0Vv@$#hd;OVu(E1ma zx;cU3XS^_IYx-h>gel>HyN>+vityV@%5XO6XuhS72*=njDO005e=CYEJKspgdiSw) ztqBj!dg@k>=6koB?HC6$MN5T|>TC3C)Q;A6xo&X(OGWzsa;65Bd0FKQv&IC^anxl$ zCx&p((9(q5DBt><&@?`1Df6&h!*T}HOk+ZeoV)JIJ z%r6DL2>mIxwE$log2S}qvj|q-bypttrQnnOi`*;`%UQPS)?1(0hzlmdZ zPZq1+zq@==e=J}FvRcPYY6E0)11}6^Dg!-Z9Bc2j!#)b$hTt8_cy%qCzo(w|HG3#- z1I*^Y&2Da;vtht`fGKr;06783V!uukyi7sQ6FUJ%e1G~h zUyKV~8I-_oMljZm`pG^j&yfWdPt^=rp6;v&!;ARof7Sr$wF4~4nHl}$Gn(+ctrt|< zS(Mvj*AT+$mc#ti>^o{4t1nJAnbDP^WCM*9ZM!HQ^VBKyk&sKIyg-AHzr}#VKq?|LD zPmx^;e;N?6pD~Gx#@x2oz)XT}rFOAP7Xdxaw|op}f^~*^I>?GAgznmmt84bR8fSMI zIm%QTxqVo1&d^Y*3~c>b?vV#K#5n9*a|Pyyh>2Nt+m(Em)m$#U_vVo!t<{rQmc+wp zpML?GsFwV7C6JY!Ur^-@ao-!Y52CfXUw?gSe-(s{q4GbVidA9T#eMn|yPXl`;S=TV zTtvz5)}rB=US|TU!Uy@fo(=W+ki~=En#p>7_07Ex&7O*_I@gBoG2^eIx!wpc4n21q zkNDv@zwQaIkMoONU)D$w{zG&WH3s08UJzK*Txe(qy%k1a8VTcf$v!P8++iD?qVm?T zf9n3^J|cf{(SdI`hkrAIw{zPqr0kn*Ux=u)c58$!5jvh{kVMh7d)>4)YgtefpChf&W4e^g#a zkBcmS!i|?78K=YjhC20gg6it*&PW#fr$tWE1ith)nV<5)G|CjZ^pU5ZSv=DC4jWHq z!u)XCsRn6%@Veb$+c`X&$?mrJYT+i^izcKeu};P%)=00F$Zg9y@N7z@k2f?|zGG5tTG5M2vL9taV(f+=Ze|aRqfnA=#!jQ#34-Rk#ox*fwMv+KaZ*WsMjd-j41sc+W<+>rcL4h}0L#!-7J63uv8 zrBg23Xvj4-kW+srf04bgW6p|gS5hI*e^vDJn%eWHw_F94 z{8cDJ0cq5oE7jkr6G~srN6}zc9B@nDtm!!}`nUDHU2^|=G}G6m$tZ<`EW4i7<3ABi zsHr%n6T7=`DoYFPs zNICEtoO(K|;tL<`Xp_gVzot#fW$Jqg;7*$w7Q(FC&u(GSyFaT2&q;GNt)z#{`?AyO{lOv0e`281UpifAEYrT;W}Bu;E_J>fQ17 zv<<(eJ{!|bdOKIa6(#X^!ok-14iC?yk=b00icS6=`LQ&%sK!0*>+sN*IITa)#&A&;~O@jydvko0kH7h-*fM8*^s7Av)fK%~rJH3U9Qu+Sb|M8ap zp9fM;u+7oLNQLqVA0CO3r zT>kinKj;lS(;UJ-51-FtBt)IBC{)k>xi_MzpWXktBjl{NW<_kD(%;U; z|2bH~q$D3&!Q`A`j{3(o$I1ih`PNGo_aATXOb%#?b84$o-apTxx4^JcpQy(FOy@dP^6dVs;Ti!UF02wfU zEz$oh6sSlI`Z7RA+dlE+A3IfA3(PjggQHRRe-g<1>!W~S2QLqb3S@>sN ze>(x~2Wc6Xv0jHh>7Sa84r|i&BDz)ZTb1wdQR0RQR9Q(>z z;h*h$FbJGiNJW}x{&7?{K&K&SZ1v^;SyhFIo!@0A*jj=Vl}b-tvn0oO0CqVSdM&(f z!{|SoG+d0B?BjgOdo=;HM&^9gPFR9nXk);+`NSAXdHc5_JTL+~Y1cByx0$#BJ}7R|7#`bc&aUg=e}os( zeAN%(1|PSj%ZZyY3xVENj)Nth`&qoNdJk@x8}>%tzH}@eZCE%feh0v6tAcswMb8AV zttMPY$1%`GklHm{4)NU8P0(yUjvAeafcuQA$onSmq^?vtqcP zK^l(=uM<-vLbzvg8Dc~u*S~M1m-J>BW({1~)9x4Tm4MI7m?cl;n$)Y927;-X-rIeM z_pEjG^gA2T1)&4Gunq9xa}})*pB3&sDGvKv%?ScrPv5AKfkQu6|Lvjof4yy10H^!n z^+FBHuQ!RslrBp>3?+MC44mg3#WlbE%-lvG{q=DpJ(zjtm|FA&lsb90@;N*cbM2Dv zh7Xp<53 z%Tzs;?n!!L1p~vlejv`Bf2_WlUA#Nr_=tdMzUckSkIWiezWctbuR^%_+~rL(JUS?e zQP*bT=uNIZ&3V*V-nDPyGaES4N3cn#S6+6n3r*_E7lz`k0Wz=4<^Bct!+9@7qXmJ5 zqtJsyp1-$DpP2!3lPpCGO&ny|+N~6XXA%Y?W6t|uSog|2qs{jRf6mZZLD+EEfq>U5 zw920yX0LD8B9v@N&r1W&v>79@fNqing?a9o^pm>p4dc0VawHEU>3@AKuAu`yEoP>$ z#B<*&;r?due(wua?TRFeo67Z7;EPam@F|i9b1ywrsVhcB7e+fL@hOet*Wq(OdzuWoBF<@Tq`}H?r13EYj-b8}y*&AhHp4PLk zsl&@ZzZyA4sr;Fbv3S|ZGXvWdxxWm%pK1gG$r7w8-B9_ObjfN0Z6IEZcc0%i@HF-J ztUa61omJ;L@H}+qiTg$3C>h+#0&UcPeJ6ZTH&|+!FpWrpe_m{UG(G554c0Q64Qje1 zbKfh|e}-Dkdp4$;z*I9=E}#L8%QJLbm~KwVJ5VX%gHjIC(`iSgCBIF{6S$UDFdI3nqaM<`a1M}M-yrZuFyE4eYE$epNuPZ*Q%S-1dGST&uO+*KPBK9 zT8Y-RwJ6DwfB&`N>aitGHsiRgngUa+;g>o{0OmmVeBB!e&J=cl;NY3oTMu@Nl&H7h z65<=l29pYs!NQ;4+4|1KWgHtuk#yQQOyZ&(%V8}>ItC9O_GXm$d8YJl*4ay^whX%K zimem&Ij`r^t$Up$EcRKt$AH`M_bst47#6LlOnWm=e<@7r0bPx06F01w88(g4pE!5s z)|H~-eeQ@g8~QF%H29Q{fo>h@iJG0OO7iQ}pQkbH>cac}0#^C=F2)Cmfz={aax~$6 z60axFyZGAB&pKj|*ETz1SZ+gVhnVK~N%lQ-^VbV39yA9|hxo1fo>?IanHq*oO{mpb z$CMC=e`Ij8N7FuTHdrMlZ{d?&jEJX4x10g{c*3Xs-UO^B)SbLd!A(BZrl4zXZ-d42 z>KcW!=#Pgz7t|xL9)0iIGu0B84hER~NT=}i9>;is^+E4-c}!Db!DbP}(`vD<6}ZF| zGo*NbBLQ;IYv0G#b(*9^w!>)&b(%JfDNbH!e>c+aP=mZ~7Qk-{mnsstigE0(i!nHf zpOinNW^8aWPG;l1Jsp*E*tddc-!Ut7mf#DrsxLAqxs60*yFK9RWsKM?XFjRS5wgkW z5feNE?@Di?aZt0?U4@lBt#<(!nu+mA-KZ=_+6(oWY1hr995){l8{E>TXud4Nouf{e~CAlBe>DTFbvuNcU3J*AVO=z1boKZK=-r=R_bwO`up;4Cglha?# z9-aB}W#3&x?;9XD?mp_OeivRWL+Ms8AkF7hB??08bV}PW13~xM;?wnAB#G1YZ2`Vs ztlo+ezp3-+7UP{P7EcRj!OfJj zLNkT+xM%W*UKb5HZUuynuI*$#4*#hI(CaM4vX#cNy9B&`%R9VM@`j!y9eyy~f8m$A z@7cZDOj%rKSkx5QD;0MxY~;JUTy5}Lzxw_g)FNp#LJ$x|J@g!RxXJL!DMHxvhxycQ zDQd>{!^RpaG0n|ZWHP~VU6s`9#GnEhRUtY}7%ZEoDX(NkUN6Ss%!Lb>$D4B_oO?mPO3&1e!4rCq5ROAU<H{_-r(clP0ehc%sk&=us;rb62mdprD(PoAwnzNH)!dIjWlhKnNi`4y5 z;>~TlFyuw=1N+C7tQ9U#K10>ZWH0EU>>)^{#Of(d0uxp zyH>?Ctr{|;^P3hKcLv?wEDnsV`hIyeKco!r-5B1YzdMTido)iDK=d06E#|tl@I0_o z7GxwRIWm6{)C*z@)|1}z*UY%@;%*ELeDLDn7z*Gi;Z~|=ZFgZ>e_oH1g34f%3y!Ih zx4UcdE`C;@+5*TNe`(qaUxJtV*l~)R$>kNk5 z*Hw;QS^6U%#E{oLZuw~k~y)@I!o*i1R6t6JfEi%tF zDfWEZ9p2^G>ej8Cf5Sw*I4_w~;ckws4CIzv!&0u(GF#c}Anb1pYE*8@Uy5s-(x#lm z%WQ?ebD#Ia?YeTjOIW-rD6u_PK^Bk-zd>-y_cA~1pe>vzRSn*PHTh&5fP}h@b z-2=5!dPzcPcyTuR-42LU)QaoY`mHp|W#S7&gRi!%G&1QXOYiTlolRWYAHVw0RRNQ) zE3SM0^9w;%e`9o%@xv}CCU5?%LR{gee<2pLo!ZsV-QqW;ebBpIsN(I3$i0K%Y#u{_ z`CFtkSc5>fDUdCnV zc3zP(ukg~;m&3yKq0ShtkLeHk*}zED^1@D5#-ZoBe@>Btu0=GDcKJCTXu%$({k)8`Dh8(Gk}DFyI}%Jn8?-4m!X z6-Dm#R>ix2!KL~W8XwXsXFHNaP9`5Rx(K+Kk^9DInc;f*;xxn+^oKE#K@(%@C73Eo znG?56f1a{;+nx2vg`tq6f|Hz*Gbb8UPP)nOQQT4rXYHq@`=r?gW~@W$)?we8qxT<= zq=3ce3ZCO`Wrf8lGv}QXnhNf!TfA1%u%+FuD)F?Ll$xRU!8@x)_Y*DwP*6R)c0Y|p zF#x7AO0+F9YbO>7r(wCoEEn5)e3YeGV@cOlf0^Jmi~PcwZzqoA2DjGluA`4{<7`ZJ z;Ia{4N2Eko_ceK>WpJKHqu5yR)n2GQ|VW9(5YL&Cm8+!U5d=wYvB^`!MXwuG|3un zj(a1jUf&lCn*QLis897PPTT3H^}+LV9XYgvsVh<9Z(5_N z6jMO8l!$T;-b@N=4W>FbWl2vbrzdmve}6hK%BOLV!AjqcSUnTc z>_LZpWluiKbZzn#hkaYh9&s{|OO=FA-B~^39qXzXksQ%O(&>9;c+#X6WtB?Vlw$N`(vu16>w#1ht^TyBHplo1_OGll93~qB#|z$8v`w7h@5Ifs+L>H_HIQ={= zGAURz__!}~09;42wTwI56%7{DbnK~$5vIP(%yfHRFPPycOa*Zz;Z8og<*J>RIe`9a ztZZ~r$UF>_uj(FE#)dU_f95YFQ;VUTIa@XPcIokr4T{!5dSk~6+SeMOnf3C%V>@ak zcxuS=ESy@#W+jq1LGGTP%CqH35ic&Ct}DXV;U8Onx(mnA z_A&;4pI<}M5l!vjY*R(n?W6wrd|cmsziK{HEiO@K!YVqFj3)}Ue_B4DY8`(ls|aM9 zx*K{a2m$79mMxMUH9vQB=?h+5djSH*Yvu=(jzp%C!KXQQF@}LLgQg6@Za=O-a5~@q z`q1oArSvbk#9qgD?7SB59eqXuj`<4>=`@uv!GZzvER(xW#X9rcr?{{tIJWdl%)g^< zX1a}hnEHaZ-W=3bfBKY(8nu#O4^qZh(k5>;xDNCTS&Ztv`531rF!zKrzM3_qw>&^5 z9-3T7w2ZpuFQFPr$CUW_1+_z@1XZiMn%dQ{+9~c|4he4OV^m?wx5Mp~_iNu@IY~LJ zD?{wQrxx*>Qu&WKq4M@z@lP$w8Ny!czuoa8yD`$FZk3=Ke>Vjhkb9kQIOuk>YgSsR z;%$a5eBaA4WDHn}1L7U>td-@z&!LuLgSkT!HTE}r2}MY<>JblhZkCWmwx~tql17`N zQBQEpzk8e^>Qdi_)69OM!~o*y9(>bq))LSfFKAG1;*@?<)9CqG#sj_Tr*aqQ)>V2E z{gf#@&Gs6Ie|ooi*7~<|QWlDt-ye6=ID3-deHhZrg_RbH28-M*W2pl6)NqlD)cz-h z{x;xekai7ACm#@3NgnrCA8Z%b73WxEg~K>x4Bflf0ASHi5M}fwu~*{(pT=Dj6W$KH zA8wK3Gmawj@7z0gq3jbt+tJaeWkme55J@7fQu^Bue*hbksZ=#cyQ&S*jM%JOu+22+ zpq^BRXTp(IDXu`odm08Ve|avHr*}!>>yMEp|7#rTLfxv8x`Ynw=Fh-<3YFY5Wp;eb z60fs@2H0+qKzaI4Fg8dkKsediq+Z1BPB;bkHqu{MZ)^Ih%I5b*k_RRx?0>;?^hT~( z`zgyne|UYMNZu?@^D{4X> zA~21lfT=V~(A~{2-u8SlYpVKbeI4}8%IyHG26VOts3#Z?kW#;PHr+o-NxhXv7$wT5tk!PTs^OP4TT-SlI*7J4VcP5 zYiat`Q#(9PtI!pi6jbUa1TP%p83JmfxR2vEw>eU-_!RSoXur$4f>euvIRN2CtqScKHI%% zJ39deiyil&{ClTvjQun5hB7ZM^F5t_1&ypw@NyL<%FnmkfHz&*Szkjwi5hGJyz@&j zp;is0JK#1)1%|izkf<;99t)eAUF$J}7S~`^Pi}7#YZYW2N|YrTCAZgxiw|eJe=X6d z6!1SN4@{O+?HVPWZIU=mrD25U_6}4VX6+fc?~gk(m66E*g~XEclI0Tt4LHlRfH$-B zgLyX1Na(*C5e<%mB?yn12q~s^EJYLMabR*}x%K)N-mzQ8_W~b!JnZ7|EUf9w?J70U zr|trbeh+%Fv}#Q6wcb4j&)l~{e+R&`)kixUj!M1Oct_BfUAV>;yV0{OaB%7+Z|;W zlkaafFN%!6%giA@S<_th+XftMd`~oD7Sy%>3-O`Hra*%Ui|4l7^%qmBKtx=0MgD9< zuvVGjLLs;Ej__8hA8Y`_tM=Cqw1&nkdDjq?=yUJI>jj=D8wvlc>C=M9`yS2ak;x4p zX4X?YYZ}NtFx5XPbz`SXe=z)RmgUQjp&0LV;E=NWY%rF7 z9{|w(Vf|QdEKd-U$HEX28EL~Ot4mnGk8u)bD=1ed?=p!yi`tiee;ri~gs_plp8Sk0 z+l~PARcjg-ug0bG?BC6_a6PX&&?QO4gSH-4g*z7$AK!A<3aK_59y6k6pXS8Algssh zMi15nl+#Nc`0N)MQ(mhI{n7-7OwR z*L=7gh^BPxc|?77f13hBzE<8WS%j{V9B%FOJAeF5c9RhvFad?(X!{C<@{w*(c-v@3 zZFUtB2GwPTlE!jv(TQ9typl|Wm13VN*Y_RYoI{}X*!$ueh3AFI$O6^Ivd9( z)Z=V?%JRbenCrT@PKQ#2ZfGZw73VYS($=Ls3z4oIeEIQH>f;13|1N%*(E6iE^!vtf zFAJGqW8Ywye@pv|cre}H%VFrbPn5Xa&x_fXzgL7fag;SSXR$wLey1nKIVm?)v*&5SpUV2)KxfW-IL&C_0LcR)4v4Z056IN!pK_a@l?^ z3gPnI_=Kx?c_MSj@3Zg)tsh)TIXIz(xXoU2V!5;+NOjkbY!c#`myH;n2I5cYKF-^= zks04OD0)8z;$qJS{4X?K{QVmLzdRVl9uZZk2e|CGcb&K?n8SDc^Tq7fCMKEqg>xbf z{CdB2Y;(5hkAG;Du?s9dil{EKJ2ub81vl8j;)E+Wev@sbCLc2P!3aFSjz#kUvp)U9 zugK;#s`jgAxd+`QDmKz*qVhxKSehzbK~D%iX1zaJ*LHfZZZnx-{H(vDgA#6iOjRjK zT^*VDy{N-v!-AuRsm4Sf#BDSo-kL4^bVm~!-<$Ff27j1m3vf%>xoUMNSU3OfLnqpA z)*`_^6R&2*{PVIMJT|s@8QeoL{6AQXr2rRWzeX7RGw%RR6adZe1c>_ocqSrV0Js{> zrsyW~&kMI8B?rw2`-9IvuW=HCYn)xY+f@I&at((FpwEwDLr8zH$opTx{#UU773_Zn z`(MHSSAVeo-wM`8;6th;v%wz_v`Rq2+D6rh^QCBTj>zry$)jSne-@lAVyG}|ODeTv zPAgPRE?pi`=c@!2?mv#=dvMv`RT{uX9JL$&C~{(uyCmMHl8J51Q zrrl$6w!x+NetrL0=j$T72!q zS@b@58MNJhpN;*`IAnwDH&T2byc6_dzJD43)&NkD*bb5TfZ)?9vHnr;v406(@4tfo z|0BVVkh~Zys#%zs^Em0>?prei1S@QvoRxSt6`e0* zdApp{cwg6;a2M z)$ANT+11ZiGhf)#8t(6|!?^5H$~-m&7N?Q&!Egz8o-D#p+Z$t2&q3kD2g@T>sCHQD z85dZK>sq+=Arv-VGIJ5JFnY_x3{^_Gzu8Pkx!p6fK~Qp^8I8X`gWgSQg><}NmT5R_ zzGhEpEhExW{H&4f&fO}5S}0xG2Y<%D8jKIb0!W;+t{IDfnEo#+*uybK8pZLycNrRw zKz^v;SLaWn!Ajn+(SC3hMR?ltd^$(l`}XWQRN;c6fCDVd(jVq^n!zkb?aa|FVB3@F zfjT2i0&)m*^u<=14m}|!;4|W^cAl2FbQJnY)c@#psdx_m*Tbt!Pjm0b41aGr3^|p& zjw7zI6ffzDj}&CdUmXP7`D(Fi-!xw|l|0MwzUlI|=AJV0Lv`nI?g-;u3ET*mW42(J z0~99ydTvvA|JrTD%k^?#{^?C!_Gbd~quWDoC|Y@U5AAMo!$FFGVAmlakZQqu)LiiK zGW>ZQ1H2hm$1%`3FR%6&q<_QUGo|MFAiq;BkOErSZ7)>*4W5bJr+z(JmpkrPQo)Dw zX)_*>uMm@OJ0=ozw2hwId}Bb-nuW zF^y|f^QM@^*c@cp`l}1SF->WInWoCmPs~-k=5N|*gv(sF#L4)}Bs9MMq(|Uc^l*qU zl&6VOr<6ji{2BcMWPjWQoSPIQFtM}UH*N|+@~V#u4sPlj6nh_4)}GH;&6-}mx{|!z zuUTAhd)phBsiIj~3+`*$7cPfyFXc=q5@&Aw^`p5+Tv7OH4Xh7gS>QRludr>{GdpQd5#b_r-XLSCy{hmP(kLT16YCK&0w#7T^#c0@xA# zBjaHhFow@31@T+)z(VK_6WV4~pTJ3%AirYh(9gBMsRGZ0Q~7@CO+eMviQK(Bdy30S z(Eer}W~Fr2m87%RwEk=A5-h&8Z^vcsByXwt!$qVc=u_q&GoZV5~gM{$uXbe=(m)O!{7;;(0>0B4ZU zCAh)n88KafT1k5rGx;J+uzX)r80#U@npRVmsD{zcx3|vCCrRi0;3HHvK#EB@7syMT z**B0ibbnyPg@TL#_aakx-pxgzDVrY}9Dg=nAIvtBS<~p+yuFUa$&Ik`?omx1v1$GO z^6ahqG(Y8qF7;`|>kYy|7b%CmvM#5KH=&2+z05Yy^7wc7x>p?3uE`HoJ6X@7ZM;`5oV_oc_sbIH zdd`I$rx7*RynZ(6cha}=byon+?ShL9aOkcFBMYT!&20}4v!c$D`!7~yXp+W+-Phe| z2YNsgRP5rRFTLm{SdIf zclLk=vs9n-?uUc4la1QdPzJJwN&cbdP=DgViRao1Mp``@j~fo^t?*IoRzrBW2~XG` z>%`Q}&L2!5!HxOgX{%FUy^EVG?p^cr#ZCrhl~aKXTy+oyjF1-^05di^_CLp~NI-!?{2nYy-UP3^64WR}|Nb>FN zx$pZt&+|UVbH8hS>wC+0{$ZVUWPh`-ea$s9zu(MUGs9@sgn9KNVzR*5sxm(@l;?f0RzpBQaLh*~)aDRknRO*L< zjFdrH&szdSk9i+F;$VOmA3MG=A@GBW~3c`$RG9Wp(?&|>wb(jXP!kC-hb{Z?lbnK_HDlwCrTUpP!1fz=bV>!RN#y$ST$eScI$8w z1Z%M=&95CDOi+1iaZhq76msy=3Gh}`xGkQAr-J+p15IpTB^#VH+m(Ns_Uw0|{l{pu zO}T3MR<~5uPPa~~V$HjP)DHsuYYq`!2xM}PtI*t7`6GoRXMOiOA%FNn=?||1Mk3BY zBV9Z4kAsaOCPPl>4NhHR^mvGli9aIn@p{vR(nlrtTkA7zVxqicu0#Cg_S@Av@?7B& zHd-GY_~lve=y5OAy;qnpReT55gk$5+yb}FXNcHSQQGrHy+(GG&N-jOi3^l!&%lwH~ zf2NN!-l62{G~{C>FMsnt^(OSEtc94y1lYuEfoV;lhbHYGuVK$}o7l!X5JHiAq4;wT zQTLLT?}ife9>UIue*BVd2&>^*KG@l9TC2lN*Z`Q)W$;Qv&#n092{ulVg9;y~qlv;h z*S@EoMDWHG+3ec*q|BXnDxRz999e)U2l%lnakxD5;+j3zQGYJ_X_w|MkF zpaFGMz;OcvP`|D>m_8|k29?RW9Ac<+J7*7SeN(Amk<&)pOyy6T!PH0P z{(%TtNsiF_6wtqXimknu!ElRbx@1&hR=3JCu2)|U+a7B3n%GJfOiiEQyNr%7oeLDv zGo;bK3eBSOFn=ATP;p!}J458axhoN651*xlDklIMD}x`&V7PC|zD82M+W{75`>Nc0 z=r8mrzYRIAnL(w0_4Z}{XW%#%TP?G?0PkFe73HZH{Q6g6GHEG=x6Aqzo@!H%0?2D@ z*X6ZBgU=&>YLN%mpwFZ4hDtun=IpObJ4YTMcYQ-|xBeZwO>%NCbr zV!TRij6VMS)Z;yWt9{~FVYGn5GBJ%aPJHGDWxxa?7nEmy6ZO)p-lCjC&A;N~v9{a% z`Oz_g$A69rj2-FR)1>K%E;WAOn)BC!a?||*W3Wjpof9VOAq6dUDNgIYQjhY#(0&&ON1>UGYNc!BfabFt~>wL!n5 zdSaIh?5++*VXu9E{%Yle=SsMWY+R}zcwc7Jy0~HN;wP%+=;%Z7XMRSS=p|k9yRQt$ z*nb<)U^{STvN#2FkK#8oNjJZyNTQ|zQs+=U^_IOd>nUVewc`;J2+YyD`t$(`Rk zM>#Jq%$)7{xGZq1!}Urb_;;Uh2pbp~>Z7v4U zt0m1RyD~U;aw}JPVSvdhBdK)^`hT%ryO}#6aRZfJlrw(i8tjORNPP0M6U#);5mz3) zq|5MUS7Yvm)~a~DJO}IWfyi-fu|S30g<|Xk=e@*-pU}84ggbCh-9q0DH3`3yjW#xK z<8ZU#1ehqRNbzst2r&T0pfi!35CjN~ga3``o!$svM9slwr&2b5R8a!?YgCK^_Ab9lD9t3lXsD@TQ^YCY{)&ux;{mblZjsnc;tc# zp&%3xa}cxyom+~eSxab{Vsk6I{diI16Ph~HiXf^h?-RH#z|B5 zxAa2wC8XPc6dPpPUD39~(tso5*Xr8Sr+KJy2C89CxmlgA*wCd;%a^Z%IO_@-hVw*& z&f5KBF4B z#FJqi;@j8ITxRT`BDt@eSMy`4WYhdGn=e=37VDZ0#%%!9rhdnvU0tMl6k{F28=Jy`KD4{@|;a1=(Fo z9oz4*H}rNc8U#g)0vqhpV}67J7S>#somnxY)*mn^954t|QGZpf@VjT`|Eub}?vYY* zGnHoa_Q8T2(uaHNPSjeww@@B%c6#n5CsMtQK9u^NR171naAU8P+HbwY{g`QUABH$+Uk-7q#O=@W$c=LLq>LT|p8xtY_Nr0VZ;0)2gh7pkpqs4b zO9pOBr|GNzn;~l4eI`udPPiO-Z*323v0@IWx;0ED zIl$%l*Agt<$DCvYr(y3jKsxQ5Hrsi43^cjlqZ#`T&apxr0LCWf!d2d-;D71k^7EiF zmxtlQe%IU@R*cYDJGV+dW3Sm&PrWs@q{zF_);b6nkcmJ!RP;!*r11HG$e<7EP^R`9~gdjN>w7`02?zK$#cSz#g=hHH=* zV``A2zk`|vuQ?JGbqSJ~6~I4$O_R0&YhGHIZqp@fGk-dD-$T|>glU)djB>|L^z(RcL4eLcl$z@pu1Ml^6|&^AnwBForR{ zV!XR7>-8Igvu<&iU;kkArSXJ7=H1YPGO13&uYVGaxv`aY^uJ?Ib*^WwMm`VzXuOlJ zBT$jwQg%c@7C2;8$OQF$n;qe57x}!Z`q;+kLIuZz2-tziR%KN~p{b?u!8Ds2Yd-+9 zTtW~=+Ifp~lqFSepYqFJevy%OA<7p2#*2L>>BvERs+iB|D&rCrfWBu+9Wp-L+?Ry&URWa+9u+Y6WZf*dmTnioAP$8W z23!Fv*IgB3Utq%=3mO+5lgk2I0dv6LDSsnX?eo}S#G#ENkx5;;lyp1sZfHSl6t3ap z@}Y9aAN;>@q?F@e&|K;T0AX(69t$-wSn|MyOutrUsw|rY6Av!eofH9CQ`ScLecwK9 z1KUCAx$EkdlMCvVE})|u@a^9C4AJKE0a(UI3P2yNVRI=0R{*avp?|z?&a3le~Hm+h1@;}$Z z9O~|0nZ{xcFN^GC?lg1aOSq#(1AoBA<2D?fcy{3l%bcVHfXNis>`jNFS4YP#`gB)- zcRTGPTEN><){{c;B*~xikrCde)}Qa0sciXz?7Qsi_dy*9#(O=0>qi>ix6?=GDTm4j zTh_94g`_jgQSB@5?UzC1Fs7x|(Q(}&45xB^Sf{ms>eG*B+zUDMfVbEQl7HP!Q}w_4 zD8DK@?`WVSdAncTzU^n;A+NSL_nE)G4 zaZ1wb%JOA(l;+nnWl20^8fmgq$qYi$1dtKjk=E|meQ>0ruWIu|?9WIu%;T)mn7L%Y zP#Gk?x3ARbcyy=6V(UaSR)4{~Jpa*J+@{Yn`=if({HA}^@dLfPsVyl2^K%#uK!>?e zR7|EHMfqvoM|DuGJb&JN+a=6k9_;p`t!Ea~c=6-Xa>pZK_ch&)7WuTo$I5ya1*GTGMNC%8z`v zd3xmev6S8Sn&hy)`R4|qJIP%`H)tL4W$D>JE5uifUcjMmb9RO88#k|Vp5r|K{8 zJ1;y1-b<*vQI!E8On>&m5Zd?GJ`6r6;{*83%y-vIw9t0GlRL3OVD(ZRUAsz{p&Q1f zYbuCq!Zt_5swRHfXtX;8Fx_3kZ5M9AJNe++Kk`wJKJ9~%X>bn8T8Co~_XlqNGH~)U zW&^x(fqf)}@w1#T6-y|%{W1D^EEp)hLOrLVnbiE(tBf%Qn|~iLmfj8ZI5vF2?;KT` z+4s^mPPCzOucCsJId!0g>{h&LW!yP2(xQ@_RU=`rxN>v_tM)(uru3Z$^|DMucH(T8cO-O0i{x`uW+=$bs}D(ILzEn>kE2o{=_ z*ACFguKWG!3!oRH7k|rQCxq&F%jSHdy`4_RcbU4S?8Hi(QVrf_>Lb?aW0_1Z*g8og zjR&^oH&pxIA!c`Kuz!5GI7z*r>8fp!$l&8BynCOT z%2?y(->nM}ng$hjn8PXHE6V=%D_(zj1QxDjoqfRe_uZs^fYm7bPlk^D{raDRMK;&p z_A36}Vh6#~nShX^*?{sv_q^QJB`-*#+TIG&cl1aS`;VJ&VO)~*|33qgE_Whm=m7NqOl0(!4ttw z?um66tQ~(ky#iVqSHG12qwK=$f3=>m_nu)-WDp;D4sB#{SO=lguUqb_#dT}7U6LyW z<(F%Pd2Gq^5=cwCI&5WSQQahIt&0y{Afnwszax8o<)vpJSl2V}h9nvpcz1~n+kkOX z2Y-0NmLr7F7ncC`c&YJA2lbHwEKnivzn?Ea{EqHO}U zYO>Me%Tdmon!RSp>KDI)@wGV*Q~>9I^nbb^O55PN$AdGz!)ip5iY}Z#2>9MStGIqghwbWK(0$*7rM%h-UAkymwDv%4wUVq|T z0MJJFv|VsgiE89F`ag89%Je~gpbYr2fK?ym8?Ed;wRQCSvI#pa2YGi}cb3#)o&;)FGza#C5x;B}$-m4F9fF0HCkvR*9NV8|L6j?)u78cuZunDnps>V~YR}>DL z2|{a+*(+S5jk`gA4OevjID`>@9DhQBbWHo~9#iuAn|W&pQ6CbD+-ng^P?f+Sa3juK zU{j&_onn~J7KE|`Q7E4s?kBF*;ZRN)7wUid(-V&Hg9Cu(;Ska?Dm~87QMnwDl*Kzo*I9}G~=e^5ASS%3G5<)|ZN z&=MrtC*=;-@u)Ugd@0yuzgWGvWH(-=JJa!LavfMZWCWQ2{gMdOzg=&eo;oOj|m{StXz3 z4%0K95i)q){EAU9_AQc`E;YH|jSA9WeIriK|^yV!K; zR@QCOMg+`e&3Cu1N>9nCcY3o^$i8m~yug;GI^f55%a-G5F@H1?lFdUYLG-Ihsn)wZ ztc2|E_rNBM?Hd#yTH1S1|4I(1t|1T7W#IV(QTtig_#*<38sm*`jV9XKt+^C?bwyX; zyvHD3*b0jZtp~nYNNgp#a%a|;2sY{4g_ESE%y#^2ygmjmm}S`vM*a*_(H?#P7VJfp zwwITTS-}P1A%EP5ueYaYF)T}F^GD;}yA0{&pf}S5Z()Wmo$Had*hNxEi3PGEpJIo7 zoWa_}QP%6jmCO0?lj2gn3t-`x?afiJlv4Da1+e0SH8-U7^k*$Ea4Klp@J#@#q3#`} zu^?h}LjYU7qpHpF=#*h{L=&P%mMDq?+yc6}hJy1Pu77jYlO-*wyEQ3}t}OwQHvX@K z`pp*4{e?%3i+~E^E;9K{r~RyMX4|FG%+r8LWykIve$qmYIGQ&hvV~cgFdcn3=(8X- zxzw>;xxL801s2;$#ksbsHb9>54o1~_Ms;d7w0~*j^D=mIBc_-Q+8f8i1-4$x%aOra z0{^roNPlrf+e++YOEg&KOI)viM$U+k_2e7YF7S+kYc>>WXL}-9ap60@mnL+IB&Kff zgy_`2%x62%JEvSTaewBSyXy{G&I%{wVxxzRl% z+rIL0gNfU?PBhv#vomTdkDt}0M*3(*CX84GTj^+64-_)Ye<)N9SPJyZOexq}&4O!N ze}4}c!MFEzj7S1bi)KUWr$(aauG`oJl^UTxx`I_g7nfBhCOWt@8=ik?S_OWF=S@5k z_Sxs!t!`6gkd13N^t!Rk<3aLfbdIy_w;Meg_6DJi-u)OQpLp|)v?{IMXFd_p)}17+ zKCl}^5hP&pbO5Q{g?r9@Es@}LWjGoaUVpnb+vWD4MhWaucEL(>yko=jy;B)y1fWY; z7~M7*90aLJ*-xR2x=m(T)wN>WP3I z%;ZgI-}b@ckQ6eo;=)D(WG7$8#lw+n==x-67SaPtT|SWWo4%haSO4HATRpc*n12ij znMr4>mRJVumhA0A-{*my&h3xnC%@01oSYMMTT1*O=yJlVFyM^Hxdc(QR7LJxPyUB0 zJHAtsYWIh1&Av8n3q`$~$!?;>JY4{hKqDVmuIYx2kq(H;nl&-O$CLGtUHVBlqgR-VaOXFCiS(V=)Hj(Etkg!y zR5{Vn?P$;OV=Ky-2Cbuy@Ocx(xH`{l{t{y`?HY{5ugT^UZ!Qz5_+rG5L1wi5xtD*&%)*a)oN?Cfh{Y9-do_G@V#ay!_qd8-WAvcW7;o{p;g8EJyqtF=(74tC zy=AU?wvG*|cLRAR#q%TLc3orw4nMsXq4U?)=k^i)mlLo)accedxB6gwQNg`U5V|8 zFb~0RZddLsa+A*7(6Ct!CIbRLs6B{EIhY+-0j#Satb@aKqm9i;B=^Kn|->3R)Di-YdapzZexs91)`rjm&Is+(4x%0E7e zG-lyE8m8W|j}7uIuYYu9E`sTdJc-7QFdvUwX^EnXddr+VsPw8XmEqQ+y0tM+vm-Uh zKLnWi+qhr{17J1l7B@z#MlY9U?>yra<}uR zuKDGI3@#EB^5H>;IZI$>!KtUqg0>c7?+Zkpa`h0;KqHCioK+J;w>4>3CAU!u#12!l z$jqg+4E*d3@~m+oF}n2e`WLWBoTE6!k07<~ltMHdeo7)m#t40s1#q~tOQzwG>)<%N zMqKYmEZB+#Y=4_HY^TB`x(#j}-{j7mey5vZAi)@|j5~;`ypDQ#VJBZD z7Lsy( z1#H8apnu?;ZCBgjT?gAOfWOp;3daaAKB8<%etJu-f5J}n3B<(z_9LGSK6KxJA8Y)D zBAvQio@EMHAXY-}lTv>9UY=a>HQqF#eo?MtM+G|YytGvL#3!x7gBRS!9gMtoGF^S_ zV_&DwB3H{g32RY~lfKhTDU0x&iSA1V?I-H834f!!Jx|9TdSX_LhhwgzjaA&ZNb148 z6z)Tr+dTZ`(b1&^$OzvZIHV#6yo)Ue^XGpQqLatNzjO5(Z>BW+&NAChz6`AzFWCI+ zOo{fGz+neVaBkiz;ZtMmmrtMJDOh>R=i=gG01=yrcK`y;L+YFv*L=z`9@BX3o%O&o zntvnLa#jEOhS-moFPZX;GutM8;&I8;92sm;t}ownqc8L4W17f<{39>LC@vWxM&uEwim}0Y% z2x7wq!@{dRK3&%h?DIYVTv3NCyQ#Zc|v z_G8TxVk)DZzPpE&ZsE^r)md=$17NmHS%M@7nSY1@ z=(2(RO0 z>(8SnE{>WsSghRF5ciCj8h;q33STJB&^9WWmb+D{E-4NmOBZ;u48h>p2$uKY zPIgMXvlcUb#9mK)cg^VDFx_#!u2v1axgW)*b}748_l3VXHvGI8pHzNh+-idE*_Psq zQ%4_<$@_&l%(R;EPe55N6h1PU%#sYJjJxQdzN$fdv=$XB35_E@yBL$n!GB7|ThX-o zd)BeW7J+s@+FL&o=6WzevCI$32%bNj3^=5Ez(K)lXx^(YciitEI!txy?qB=}p+lHV zAgvOq1VW7GxOdnC1w!83aMqQ}D~2GD014knVLZ9WM#gvUb7Pja%e3RxFVU+STivU- z+K6hEgs@mEMZpP!HL9zY{(seasIQY&@<{XDYox&Hug3tks3yGz0M-S#1==^}G`&6m zWtkkfRy51oXt=y@N`)i275!Ny-Biih6z22mEt5~>nYQ$6zH%$?EW#LY_K~Q9Ntx4# zY6_XSVkg$0i_Mp(4|9_$LMFY{+fR-3jVYcEV5R%Kmu~QGWTTc&e}C`Q2mA)ybJ102U*y!Y?&|Q8o>6_9V~E5z5c9Zr};_s zD(rb@V6R)9$J7YpFg(`lzKe4mD9wsKZRPrC``6ok>Tc@CyMri&_*QtOGAjMOXOxfk zNILZUy`X^*M&C09a4a`wal7m*gRh6L>Tx%v@AnwQ==2S}H(EBNLA+UAsN7&KPmh^s z>fZfwl(9gHhV&FOQ66lGOb_Iq>Yom{&Wd=b!+^3nO`a4RHh*MFLEEYyP`O2>25N77 z*A&gOk4pD^_NBR73)xZi@$KPlxCgb)fCap`?XrBR^W1S*C|e>_=jOM`?&t8r%4L)g z+HASadN4H*VtJ(@c`1Z7!C7aiC^gVuY63Os`}H7E##IM(nIn--Z+}_`QT@}OzO3pX zgmuZnd~>$lWq-j`p5BN@N=G%8xKSMQ)Y*Wk!MutGe4bHt(PQ33~ zrQ_kp1VVfz@}2NW>G+1ciCauEthD-z+S--49$)+|j(-sULr0EZeEk{lM-SdlkSpT8OuCvS_tw^$g}N{y zt$nQMQ3mianz|G>v~AVu6S9olejcqc9RA@A(kB2S35hD?6KdFgT2M?{t6eEtbs;JF zRGEMW&wtZ?{b$pL3^KEEf?dkh#?|Y+*IEsaB;FZy4@(e7NCxc~RV|lH!@7#sLs+;5 zP`*ag0-U~kQ|{h_Fl@!}>JtvYpelLy__IdHk=?yj{pbs;U$hkGqWGX`XqLTco>to| znlMc%dl|kj&s!$7@8k5iYWbm&MdVcBq$=RJ-+#mIms(cQ^sdku?FBo9X|iC7zW#8w zJ-|KyBZGU%%Oe&Oux=abgw^L2l*J*%^G*Z_hv{#tb8B?^V3e>_&R2|1EFBeBvBE;9 z?xJJ~iZz1J448ORmJ)__s_(H-SVQ$zx6};PCuCSF5r>3Gc))!sPBg^UsZ&Pp7o1$? z5wm?NEFNnkN~fP2lA@whM<pUbH4lHQASt~i28XillC z?0*u_-L$+YqsqEcOKgA9FsX%_Zf2Bw-=^VHXyq^+%;UN^#O~@XqmQ~b^A#xw+ke>w zAhHW)MHU*V)@?0z41@osO}k;z?9ciY?{{(M@r$F>DhIRW;bvLDW=SU7H`}Sc#3DRb zyHgdd@;9=3Y9L_`W;$BuoqA+++!#cYT}~iF-Wy2~y>n7-sV2*uvbf9b)c8*m|m zm~H*aGU(34VsFMw>EPS&Vy#=wB!AMvAdQW)G>rb@N}~41b;Si+9JXP1Vsvj{q;z=u z6Av`tM)V{0E&xL` zPs)D9??mW3$io?Rd#OrcllH`R4K4)gYQ;UOM!Q-Ws{vl?bi0aRz!GkMBG;&<>R*xM znS=}u(j97qmJO;W6K3|YMSs9{o1gD4A(76Orv@$)gMHzNvV_z?{6-XBdsv$t%8K|T zV)lrK2CWleUEJV`DLh$w3pz#2q}{D=fwej__>2O4eCGC+-yd9|*O z8PQ7ehooX*Ds55J0tiU$y?2x9Xu4~n%Fy#MqiW;Ww*+4^TeE|`4VNer z-}D=pA%V*UBbnvl`+wX0>gY}eE*_Onr_d;?bK0W97Cv~U;o7h_6@?JL97TReru=c$ z5`LAJFBkyid}&SR=aWRbN6@*W)_^`CitQe!$ME$*lC2e~3(rh-u1_`3IH<|)n^LrM zun(fX4$)9@gYk&WsFpD8Gv_4i3|YlXOqK|X^j|2;kf=qzgn##yo}Sx^{7H*C8)Tor zBZ76SWqS#bb7o@*%cQU8Hl^08XI}agN>7Js8`5+Fl3Pc7B`YzFJcVsezcr`tL0>2}_E|Y>i2Dg&%$TX=S#cZPER*Qb8Y67%lC?Tc zGclE+0+JFm7i&;tkdRN!`!!WQ;Ol3n?ooj@e$NkM_l=)$u^QNvmV7xv?_A?0AbrY; zDM61U$bVl)r4z<*Av8OaAe@EcUL8a z02pD!nNvA^??YTsyG?o5keLrW)jH3Y6>Duf7kjhwM)@1CE!)XKlL(QWnF7|g$2*0P z-VqgX3ttG*03kClnJv>K=~*OItV!2eDGg_{+JDhjv~QJsJM(zfYN1z&t4r3}hyy*5 z;)5S;_FO5nP%l^Vfeaowq?dS<=463a@oLl<_RUv#abrRf1rU@b{^qFfd9R8XzcO6^ z;iJc#J#LKli1w?zm9ZTiZdZLtstUl#+~C9+(xx>0^FX{s z^0^5>5aU58Tji6odq2l_@@|{Zf)RcCTCpg*$6JY$2i!lvxf98;1#}O?BMl;DL6{`NMW&@)%|HU(hsrcxhF~~$O-5nIC z)sNP&;QDW#IhA|kiP-j&?4$pD)IT5f&qw|9QU83@KOgnaNB#3r|5tvLyo;aafAiF0 z#F6H=3@d~E`5>Ph{lzD91=lYrynkgL4_*Ep^Z)mcA0tO#g#MHx@}Wh=r-2_~%T}il zxidd$|L(zyukXs8z6g!P-8+N$zQAmK8u5?z>= z>Pe}JvN^h;sfw98%SZv<)sKd(6k*H0y{i!0K5fiP<#n z&)Y63L2R7I5SCNDN?c(^5Q`+2uV3kfE~Lbbe$r@FjBkt^%_$B;!i#G8R1!q5OAQKC zjf)f@$pkaC#NQf!KVPA{sDI0o*SMjPJ=cepaq^)$5Fat=^)E*^ytnMUiRj6bM`=#} zFleyd;q;h|T=a;Qe)c5ZGCTh3!B!6!tC4f>O(-h$?SfMiHQO7v z2@f7P@+%$xn>2+Wb94hK1#MR+@Sa0TT-cb)Zoq%`{dqAy{cY!xE8+u3*{KBbGw}F5>p_#iX1j)+Azej&?Hd&XPd94B z!Et-*gq328V%xHI?iVv*5=CA}i41lH*Z(21U*D{^$&rPf zKG{}|eg#L(K>LQpkLy1}zP$AmH4Pl-6Tne$kr9?N*Pp6Mu%-uz%m@4r#mJB+6^8#zGgG7C7xB{@X+6N_PtJJx3jofo8$rWB@A{ z?7#u_Ela3&ur*3u_8esFc>u%Lm!SfzDb@;A#-FTT=D zsCGHOfPbr7%6v}73g^P@fr1DgBgXY9ZT&^DuDzndUdBP#ehTIME`C150S^s%`2oRHOcFZRXwsq z)59P3j(_|<^TAGiTz|Hi=Rca~o^GdbDKycp_IhI(utJNbkUf92J8&?+Dlp10bE|zm zYxB}QkVX4==Z$iey|Lj&R`V1ee1U82IMUZr*7W;_+i8ReFC4YcyF#cXsjzB7($IC( zY${Z!@wk2c#@mJ%nXx`1L5w)=)MK=p4EtCF1%HPEw#w!u`*T|)dzg=#k@puEy>Ja~ z-V=jw2ufT&%*j6Z(bVVu+^@}28wKiC%i`lX*;xkrh1POZ;y+@EO*C9o)aQ>eoM+ZH z_LwVaJ#RgL9pZy-%qQBvO|);kGo*9KJSWMOj;wQlLsERx+Kn)-@>P&0y~Vg4esBKe z-G63AIUDtWBEJ)@5nL&OI?R>5cYbvfa8< zdI@&e`CXIN_rW>)U^@fB=oq{ve{W_=Fi*1kf8IH4s9IeaV%&0EL5^0ReO7|2+RJjVGzEu z7Zx4E(}jHus?k%s!w2!=c#chuf+n^B3m={E;5TH1LL6>IEzLiGm3xmR&i zQ`Hk!iMGRQqh>d;Rg;`4!0~LI0*-E2kF5ec75dT$E5T=ATcLGU+Wi&AAb&!7^C38A zdc}B4omI~?GPN{KOWC_-rKs-fQAX*XkPtMxEkXn1Wr<2(yglS+5EeS}swwO;=}E~f zIACSE+7_|=y<*syJsC#cAEBo|cYWZd)j;l^X9h!1g z=UFAuF<>*B7cpAe?a6$fUw?Ip4I>UNEPC>P&$t+Pk2lWPGmj&`KjR1+aByJwT01dQ zdHO)l?K_C4^G@)d9&nQP$fd7f0pyM6tdLji=j{P?sPVdHS|wXXzTk|K-;gRTI-~wu zBPo6yFp@!Pzl@|JD%}C~a(N(6@B3G7M(XTf7uA;pWeOl>VR}KR3x75B8Buv#MP#bZ zlY3DFB!z^-&aJZT0H!e{FGEVoLyI2;#~a%h-63{UKKYh+@seqt%wTlnRzZ%uK# z4ZkI&Eu13}YmEatk?8F;B};#J4bYthb0PRjJEn|R{xPD^7sYWN(z72FUVgeS&yExc z3LpNi2p0tJsD$uyNbin2P1UXxbz=vaf5!x8zKr+fU7c=QGxox=96Nd8 z6NeuB%V+P+_R<(IcfmW7Ioxw+;d8a)8@@%S+%J)Z+%Ken1LW&a@#%lTjQUPKmB9Hx zE9QcSa+-t_i{(Fc^h#v`GbX+<$u0}kz&j-#6Y#DJtAy=$Lf!%pi>L9L6VX_DwWN7^ z<@ZPs%L)QV&}sbwj?fE;hcE$3e+pT2BK9snJv8tzf2Cw#@+1i6lD*bStNf3!2-UCO z!>bD|>ntVQ#PhTAR;PakYw~+EAIVQCf~Ey~37}KdZmhUqmR}~h43(OfwV5+5FI_Fw z{<=rrMW$e4O~1aO9TnQa-X+K3mxK;n48pm{>p`Qoj9?z+)y|+!8?7ES9oQ9R^=24% zEhvap{FR|@K(E+`44qWH5@fQc0#zk2qIDb!Mp?(yO?@q}u3Ue7p@Oy0q{rtGsQV3e zTIoud2i!)OX|M+g8vK)QOFLtmrHs{Xsg|}1D61l~`Sm*K92 z6lo7fR;sebkw<@L2nCkox*B_5{S}u!;rY8o&%-3a1FBt%`lQplETOM#9;;{GZnxvO z(Zbf&-}hQCqvHfGAL z=qy*udt_avJ_wt%4q)`};#Tl9OZSTza~GfMCruT@NN}-~!tvDS4<&mA>>=t^i%&jX z5}=Ddm=k}3wG;A{%~V^7kg=SdEot=tRzM?5wGc1clzEbv8D|u@JvGeB%F|!BKH@KA z2D?(dnhIna>l}6Ot1e4dw!%!egd7nDyUga`5&0o@N$IqSxCRpgk}dh@STKJ_p1!{8s7G}9Pwju+iC0SA8-g&FOptT-Qx=AY2=XDv-T#!hE) zJ=)A{Z++^;m)D0!`}yQU%Tmyy(6kQ`UuCs9eYSgF0ee2uzv^V37Poi#x8Uc?30{zQ zC|>%7elA0M??BTI1^P2LLbZ-S4)Uqt1!a#=Ld%7j$jYi=v8kfqU4*j*>K@Wt=aqca zKtg}&5casUH-7*P<4M3z#K_osh8~CUzc!22q-{%FG&%Z{0WpU`NZ(tyR zpb&y1r`;Zk_XXQgT+f@Y<95Ee0%~A&RNv)G2;=B?1OAh`%Eb-T0)20^xD7d z65>n`SThg1J4lAd++n_0K5b7lDc*SZJJ zZWi+02(5KPeqk%T4^XpFYkv_BP{yH%SUFgrMuc1(7kXQRuam>7Y!SdA$!18q2`Mlim zL0x9(0kg{B-ArvU`k~x#JFyIdZe4%nW_C}Xfn1HLc#GsxzNrriCuzY_IiplFBDRVV zUfk$ekilSN5GKLz>H~j2^I+y{IApu@kzBS~()n!;GsUVUJ@=`==5rRGUkFjd*y*X? zJCmH@p+%3>X#^1YT|2`NuA`R<3H=;^D8OSF8Gb9Ul6ku19_s7W-<8Jtoil#|AX6j% z{eOE_C*TWCFS;`5SA1ntJtS1Sm_GVgU;e9524yOU%e*RhonJXO-40y}(2M3}1gmZ&e2*{vtV?Z&HD@z?R={;XQ9fT4Y1%j{9m$5=5#On>!8JFb`OK1<-M zled4qcIRg)%ud(R^6nkcShN5rWifqSG_C$>S*PHOQM0(PAoTz-s+mLvaR7VPt^*uI%$;X|M3~^qln({KozNoNj7P^}Z$|S>gd8V4n_qu1 z%z4uTaE`Q~8LNy)`G$Y@3$ZS45SOMLZ0T`eY89f$AnU>#Qask1XquDUCeEc;w*>ng zr7>sv?kjecq@ z;C!nDKW~j%q=4ZC{)u_TG?qXeS>RQ#IIQ3O<$Pu+N#K&-hFgfZQ&zz-BpEGD@LOj%&837>wR|(|4s?`05 zM@3l?i+mTO!gql~AZ2cYsGK^=;@c9xeX!+g*T2yo!fKrOyBg!g&R?1@^FL_**ZpQL z@&`<%Jeux27#M$f(YQa~-)|#wkez5p2a+$mVbxnyW?{%z+Jsd^x`7?^;~p4yGVXgs zO?Qpg@VR2tD!}aXWxD)no8L0P;Y^b%;`-DAqRPdCHxM8F8E=3MIbT!^YIlYY*xbXY z+riTz)H4{jNrr_KnD1(^Hp`PQ3UF00`~+Tu-H;*qgS>y^#U}u1H3Rw(kI`1}dAd=> zs?3vIapr;KKHO*aC4cZ@P4@->ncBay*Pcu`J{)pb>k*&_jeFCf-~it?yLKDE0>7`i zs@yXf^a#AheA9LeQC-~aK-wAZsb{%eO~RvRyv9mW11mejXUu2rwcXI{z$@5VuI9^W zn_9qIK*WF9p&eYZTOZRp4Dx|ittZSUp5FcbC^=PetWvxG$-4Yh8tCNNS3D6f(~eUY zGQTc;$=|qGBP*x>qP3A}JAqMu(N%b}Ed653ZsTr)JEv!Z9x8oxL7Y@@8gbKy^Eb6P zsm1gFx^SbW)t_|=sW`4bpsvFHgI=iqR8h_Rd9Qy2O#1UdEuAjTipv2hR|f?Iv;iY} z(}%Eab8lpVT@?BRC2%C4RN(6|cAHZvaifdD z0FOx9LhbLRUku^ilH2AM#yyf zd+Q$??BwY&vWozYs*EdF!bm=Zf0ZV#mb;`JZCC=_U9|#k#V`&G56py1EnD`dDvtID z$cMg16LosKO0#X{enm^fM?2|DBLcYi#5<6d>SFcf5%q8$f?GJCMv(E}d+c>8}8STdYW*(Iq0zvw0#_uJ26Pm&_+ixQynahd2(`fW_+)GIL{%F? zNEVtF193hX=6?;Ov99mT0i)X`l49V`>T32|4@%|+!5ho9U*L_vcTAu)f{T9(oJa#L z=@hJc19(Pp*~-pC5(~Eom_Ajuq>bLcXrVk-*@1Jt=r_w@M%NhJ;E} zz!t7*L^bGNx?6Q?9WKbro2GvyEjn6g@HSJiN~_oQ1R{kpht&S|A#4>e!~C_Rz#}KM zZ$Q%)B!>VMGF-5)dXMTj>^RJKneI9`_`ayF;vGlvcN59HVlAe5z$UrB+v=7Qsz2MyvA&e%F7j$}Sze=iErQ zf8xUJk~LR>uvVuD2OF%5m|%f%Z(BO4WX5_m$*KDeUepV5Q4EZc_91hB3Yl&gWA7H1mAcnLMC0jvsd zPtfW&wLMxceid_1`?q!Y&ma7vNpz=w>5C6b0PDsqy$w{%A7I{6-oQTSt;X`O(8v z3CE#Ykxo4_RqcOug^;0T9^akVp$ZSSzscyYF#$W%^YMTAOQ#UfBh8(^&nw+RfQyI= z`v2CW`p0=~G_rq)bl!NeRY>VI^XqQyz3$gM?EHtgH)($BNlg?KQjIHZtTB)Ja1+G zRVc6h|Ag8v=4WiTssEcdk$G?0jFcG56oQ5FNrhto7587(YN8ica`|uG>8Y5?u{GsW z3apbl$V7ii(6LGIDaU{F-mgcnpBQnWU;)@|#F#@Qy=&xuRsY4z3T^Xthu8qS*_1XI z9%@(;`s}y<``>aXxhpl_OnjM0B|vww(A99(-zZb^@6XnuBZ0t4;hJCrW&R3UF>{%1 z_443vDBOR3@ZLa$@2@wGfE%g+=U}UO&XyPcUz~sDyD$7)e+2EefI_^WkFEYrUH9<6 z5(OPS!bG-S2)sx96J^z=C^({+f91c>g(tfIg9x?%5#c`~{Qs#44<2c8$#^>726%dN ztbyIRUrjVfWj`ofF=0iVz-0Hc&st-*YnYGMG zVg-Mj@U3@3$RglG)r}o@%ej-5H{H<;FJ} zb|>h;F@Me0#SN_BE$0%#GJLnWe3hi>Nm$Bh7H+%11V-t8xcE`CE|Sy`X)TPhdl$Wd zf79#S>XCzir7(|nre`ALPICEdYgFQ-w|;*x&3|_4e%iKFnDi?!Z9f&rb~03`(cNC` zvD9B~=QXN7&*!UO=kw*%GI){uleCB9I5>XJ_mgFI{InBtql3Trjl)c9OPu%^G^4g1 zxA#EE23ls-tH{QXQ2i7Pt$*W3m84egW)A=tAD17BKk}Yvekrrj~#4 zAdC*6bnS%Hb_$qkMics%?>6+V0V_S5?75s*zz$wk)%9G;GH}T=nDAODtonL6^v0Qp z!QS3T+m(coiexa%{Pk?DU}4Cl1z2Um#H|9=YFPJ@k>M;e4YkY}KUf_^W1y(MZz+dkTNv9`kB|BKOAsI9NI%R8ZEz_X1$H;XVBfK9ewz z7i(4^Z8O;Un^`_BEa`K87p%QJ^ zc~2g!UBVN_oq9Ni^emE4iGHj;kd-ko@3ev2t^x#KyYkT`oU;Ay3pf%s^?83XS=<+F z$`&bpoA)meiuy0LZWp(vD7nvVf>T*Hg*z^!a$qYnxOuGgz2W<|XT*QkH0kvR8iu3PJu@(4Z}gO7ovtf}v8J1`yJ zf|+Jk9t%`F3u?WmGY3Y>h|qCox22^GC<(i?RpRP?5cFj(4#L?6PD0gabDR$5X~mU{ zg|NBcc6NERdVGo-@F7KD!_6uu8DLc#dxW(SuqkBa1{jYS#JJCdB@TZT5?UF3*8@hh z1)v&)R~0OuryZ>1L)|}b<7!9^Wo`Noi)O&8j2$n>3|8Y{9|RukADL|P(M0O-tzk-R z+Fu=i#eW_D14?}7*wlPRJbd;*ZeC6tKJ`cPW81IAD8dmiyY-BK6!fTD>j$! zMwoQ~^n3W%^rr#e8bg2c-B_YC3!;BqZdedc zq!t*MPI8-wHzHajdvNQ}cwR|GMa}f8l}U6rk4=|mWVjk6p9~ZLTiAtL_Gb@9?&02z zJ(By^HW|rF#dqLwFp5`|Q^2g#F7g*=%l=<3qqED7$0Ub?r$`WyfpJB|zj7n?i$c>V zCXKFA;c8ezkGU9^D2@E#8_*3vqOQCTH@c;8&XVB`Pq?8P24f8~S#ic4+;Z64Mlx3E z&!RR834M!aeJ_7Ie0z7A!@hpw!4MuCSvNI__SwbEta%(|l#Cd&Rd!Ur>FRRQ10BX8 zP@48CP_=I`_CUztO?TZu>@F@>z-dz0WqF|6e#qF81p%#A*i7ff+nlL-cvpT#CoOoa z@bPkZ?l6yPLUdR_t;8c~57N$XdC$8dTb-;`4^gQi?$3YLRfnzcu9y}5JcX2$-ANd7 zZKo(%B}G=It`anzMp{Y5HvOo3Mp?7%b-2dFI^SOMve%1)y$udaxhO$V?`0%SEuB8J zh2E$w`36SJZfm>Zd(Fycv8JM!D=in2S8mN0|^@#63ir;7AsI{Eh>{ee=h^s1^5ie zL88N2&4T!tKx6jLu%h@PrHPLgDlTmn>2RcG7r|gw$~c9dBoJ+q{Bhy@X!Q^u_g_WC zQs4c@--W!-*T?+wcsC3ifW^M>M?jZc5}{RXfOvn`-}uw`p0Qj1*~CFGgQwi>y&ubQ z#RtJc2hNAlUHJG|aA^c=6tFcWvsS*g?>bN^Gz@|H1ldhaMHB}IX{yT%>R8ueBa+C3 z-PtJh@HWiz0ETJdS~_WPbT3E3!KTlc=Srbeg21Je-3tj7a9UHj8rS)Fi`GUyl?>Mu z@6CVqJ$*PsAb-K6{u$%Bes#I9;-v=U!JLpw(CuxM!VHmINnh&==Gnc~i5_Q@$!~0`pRN7Xiw$5dor!g8wfwTX$n}QxukXx zp&qRNOYme*YX)~u0oNGBk5Sce%-VCn>XLuubx+PE@{-ezNt}^uo|MFn&cZl!)4;@d zVXZBqwbpJcqum^GV4=m&yhp2@lwx-}Gre6KHj!{>paRkN{gLgM?3kVS3xiB#LaGcF z(ZAgKZVdHF#hzw+l0up+C4oj>PbUIwW|FYe#YvhxUUQO5w%)G5`UF^Z5f}h5H;sS5 z+_7`{QSZ~`v}QT!Y)#Lul(4R|0dwm!2x8`dj~O4agmaY1D*)St` zMH&ZKe0Ye+_}Vo(X#}1AY{M~0-{fE8V>9wgp*H9{%e^#rLl9UwTo_rLh*WBnv8BxN z$OV&nCnK6qN);$f9@SQ#FTVXX}62=fchBcXNVq+q#}>JoT$*>5lHDLs-yWjVIGmSRcGrJSpcY_mE%k9pDuABRnmXjwBS#1p3gh7i8|M|^h%38; zv$?9w#Ni6kh)Kj~+VZ^<;+!m|r9RFll7t2`Cwnc82 z0t%14eseC=F2;;h0%*L3v$rJ+t&}!lQO5gA(RZ;YlujxE(Ft#ujR+EoCU33vD76hu z&RZANWEXKvW`8y2v+92jQuF4pU8&m)=1Km(ye#L@8eOcpQ&<0eGOvEpfGlOaFtA*H z^(X~N8)AHg19F(2!vZiKgPL@If#W|I{D(Bu1P#dXRnA<_F&Zd-)~_7=phpMdvr`Kx z$ZjiLkcFnvK9hs5`fP{*fODOulet2MI@M_Y=!!i!-AQ3%-f)N zE|^VH1h+AnG)a+3yjU?~EusYH(6wZCM|mZn%6ztYEimC%Wt0*TLY^lr>-2rcA&eC> zI+svZtc`L6En9z`xv3zoaH8(n2Sn%~gH)_^!6GX=zLvA%pq4wdeg*8H9geoEE4)IA zsaGYiG3f6N(hOvuKqUAx#w_lb%hNYhwXhnP!jm(Ji!wXA->*azNC{Cx3-+AKHx_xO zoKFn)T8yZRuDBqlm4H(uE9$vDMPth^o$5;@kMW7;)kuGfPVp%1^pp2*-jfWk*bHcm zmTScK@1d-2F4f2Z22`&~lOLP$GfexzM%tx@qO2{g5WG&6=tf$6Z-$aZmBt;gxmb;* zx9*2b;Hbbqe(&T`GEpdAs^sNU;_`9EY=|B%WPuzkJmrT)3-~_O#HuUbe=eWg}7E{ zk54LIAg7RIo*5_>`7OuM>Sw~;Mkc?c+vcRWB^6W91qR>1l1TffU+xv_KWZ{^h>%@3of*T)c;wiF@5I%pr>&Hc3n@pIzE?b?8lO}TVd4`ET z*pAm^XFb(#{Z}6HV#I$u+Ysgl9%P4nnvogC~(W|HUYTCu1hN3paUi$@Al+{;$ z0jqz`@E1&IS?5&JflQBBuf>GX5Aj zme==8XT&-7h^4EayC*}m6-vbqX6=CEWs)7S29o!o8k+JO1a_l2o>5R61u2q{MrZ^{+?2P$Uu@vVHTJvAD-V4a{gQqPiRjA=ZDJ z=8%QPE~dmhkf-YdRv0&h=p=(Bw&yP3jGwmkuAxC0u-3J(_|vgeWaP4bMm5WJoY zAyfj)j{O!G7|k+4@A_CJrCx%FE--%~8XQk@XgxP%h+jl+rwNd7tdu)dJ!;9qb1TpG zRhUUqF~0V5`{BF8J$A}h&ysKg_8G4;LSS7g+n#PdDT}&)b$Y)1i#n`I*V+10bD|xl zmyM;yY?Ttb1Qp9m6eB+bhGPR*6Amhg#n$Qfm`>_jVYoAx6V}rXC*D7Q%Upk@AfuK* zT}s+QS`ErH7c$f8Uw|EKR2pD1QdY}@ZRO4t&~B8qa2YIQ2AioquzuMty`8HiSg`jc zfKh+%W?-z=Zi|pc(Z$e5y-w>`LZc-S<2Jd|HE3)7vGwb57kg&!>43v@y19Fhd8_d~ z5!*3@&9wTk))bqn=)SMP>nwjL>Z|&LtZqU@NyQXTHEH%UBW*6y_p5r;y0)mWOuB~m)bY{jj@EFz`f}K1P^`%Uw`5qb7d#v) zSo__04b{ODwhZ?_!iP`B*fkq0$ry}&I|-ds_3Y7palIb`OLD$3YW;t7Z_K(F>ns%^ zA|JZV%78b5Mz~IR+Av=%-?H5qF`B8gfi%20FS6mI`7=+$62Ll=+Uy*|#*YJ7Wj~uW zog|_&wV1tiQ8WS};X;)7p~(=$VqIGil3Jj{@Ti;ytWTMz#r$B>$vUSuL;ID+_5Ob! z2MK!pOxs2>VI`k46q$eZJ&?`#Q$w@#EYM9o zmc_kpup>&uUBsE4* zT+t*xMOF2Oi`IX^$QJ)2DAKv*ur-t_qW6ANJIYI?3f3nvKU|;k)HHxfN@8qh8Vj zfF+p>h7T?ZGwA1r7cQp$Ow@i@YMtJGU=H%9mCsJy!EJvPS}(mRi0Ka%*7Bhu!>Nx@ zU+=Ll>%>c-lvCo}VLwkk=oHSZXz)x2t6b!#V*<`9FgSf;^`;*U7ovn#c63o^ zs}TyhRjEu&U-*fxFd5D0p21d4;#FOVm96hpO_wyHWQH2K(FJ>um=&Kwt5jvRFh+Q> z5<8EUHsWE@RsmGQ?5qC5V%vMw`Iz`OTm1U7GK_z>)VN3%{Xuv7Jn@G7h>;2u*37&2 zO>M(Mz^N{7V>p$-fLd*OV9IAXunkmE%u=!Iz$XqFPv;Xp>{j{!o%3i_g3oMl{#6bw=0Bdh_PlaC z5fGR%+j-fBrR>Boc1O{FQO4#&iM7V#hya8N~>V&S5>KDtGUqZM=aWxCc;Q-OE zYo<8fV$jd5s{Vc}eujr=vuQezt5GVXUu(3RQ|0(WxXm-Fvo-yW8ukB5rU!1;M7diW}hb-0Nru|S(a|FdJq^`XIVHi&nroeDaiNs$Q0 z@lkhd+xKNEZ|sr9c_b+yIPmIG@8*9$I~z&~N6qF4$==AQ^adlfc%NLHVpx|*RvmZD zplf_%rR1)*Pr1`h+PS(IrQTAjHqXVhpk;F9;tsHVT-hbUc9U&n8Q97PKsB@Cj6+hm zDyDRpH{ptaaqqoH67Xa=*tIQB4J1#@T~|cg;XTyRR;-98GLzXD&ilpu&QpI1rwIkQ zv*$Dmm-kG@DB36oc4`4o=1hp&i>U$Aw@Y!xJ4#iteH7Oc+}y@Y+sd3|aq^yQ`}rxk zzldFbq0l;u1&uL1`9o`!?)}O=uog}5LLwQxJ7YBcE0_9sAN>d?V zC&>wsFJsBQfc@%LXzVJ=gZL<-6Gnz?&m(>`#E%FZ6797tUecJ+mA}kYhFYl%2_3V= z>lL0yJdq;2=gplZ<`3_j)3`o1xm_touvxBqC{N2VO!nEE3L4!opcbfyYt!E$u9llz zRD~ZODHWYShy-(oMaX{=7WZmpx%zX|d&aI4Qq~flY1f$%#+JtocDlOzVY_;$d$|1> zS21lZ2*22;Ez6vw?6@*;dnLT<#?+^1Vt!eK$4ql%`ylQ;mfI*fxzXZEA5{9H*kX68 zVsB-l&%~{VyD8f8q2RQXjoS8mFCqb;h)|#yG#BO6uWm*DGT(n#EQ~)FPOcl$j^-n{ z<*NYhs%vPU&pj>1j=8pEYuGLM>nA?NVvR~B%3me3AJP4p$2w#U=CNA74g!bG`4kdS zapV-p`HUjM|Yd-0SV_a`i^?QR8(EjTjyk;8uQZnM*Xx*+ykoQgaDc4;W=0M z2_vDv%+-bm5p?XMRH@4nfSS|lFA~#7EuKu#Kti^4n5KVZnTHk+MCEO=v;g?#igsAl zrb)Os;fl63khGdhDEp<&zW>(dtsOra3P#7dfDr6a_|hZ`?2Vm#f2LEL8Ei1f!m$*+H*Fu>T!Qz6yPDg?j-^({~re6lE)G66N-}0L1HlAqn14q|D1t zj>|P^eP0?^uB2+Wq?5KpmB($~_aV>~Wqab_P2yO&LGk*bd%= zO!DCFh+6AJEyzCq8TW12_S8TlcdF>62@bG}Y3`OC;%RHO!;Iwi3atDY@w|1y)r#RY zz@-&PEfrD~M4Xq<)){VR!}x<&n)BEL&YPTvS4P-fH<>%w>WL`k|1=%Sbu{5nW%`O) zTbX|Wtx2pC$tZ`54F0<=;)}jX#LcAeXo4p*QZN9-2J~kwjYrAu7ve9=b zef)%wh_;yP4?JK25zTIkTUNy@xCoZQ5|iP_Ro~0fdlCmHo%>a>b9=g` z-j+N*ZbTOIf$YW{pHn|*1Q6 zwuk7SugJq)JvUnDWs@RplCQ;4H@~c&jN7~n%jxiKTk(2u2{?o~`ocM2gdSY3K|6EKo) zqfgY66l}j(@e7^fJt{Dis?Xh2Jy@A;@mg6(|KT#ae%9MDH1> zUwvhO`*EGLcQ1L!j>u%?m%Qp_)wn;uvbopAH=w=uy>8(js|B-2u*_bSYJ{6; z`ydzh@~NMGJnz#XIHnsM_y5|Z3+Bb zLD5i!1zEwZna6@Uep}GM`JaM@*>kEzYEpt_*2$`1eq$26=IUk%URZxfXZ>Bl0NbAh z1Ev2%!2luib!@X77SUhU2sV11LMbp}qZTwW`_*OlO;P!*iX9&=DSWM6DKao+TmJ3s z%$on$xJc{UBv=;k=@k1Uf&Z$A>A|1GqYd)^tn!Kfa}_WLkVA(KQ7F4RO@x0>bzQ~K zmWXNjTwH(PFnK1ux0-({Z2GoZ>pIiqO4SW~fG9W5&;zmUgwjd5Wr1F}1~g z?`$XV$S8+_sG}K+*$Uw=au@q`M|lXP+?BlRuO9b zXRd{;VTO8&qqKj2GbZ(J#of@2b>Z0SsPr=H3g~PHU&r5mO+K~$`=E$Y6+hM(Be>!f z9c>NQzZ|-Y+s0yRkt96tC?>|TnSA%nB~FfAPIwlpQmxPD@7}7-V0K05#F;`>L}YORh z7T4cxZFPS;?Vo$(8~UJ|ul4zER-AEy&_V>GCn}(}Zj!XNHOwwkLnf@$9XDOfk-?8- z5PK;5a=Vjun}jJwn7iAA$3o$Cspb0J39n)gdy9H^@>a?uI1ArpcB+Z$>V z$8KNmV7wr+J6?{qE^ZjQ;h2aX&n5;!_5w#I3+;b>R>V@U`9j^2D+O!8>fSHiy2OSH z;L{pP9D^K=-!H99>>m5=w$3pBXHW4s*09f7-us=92@Z#Er}rJn?kIU0mo^T!@q?WQ z>fkVU@WzTk9QJgC`+YaC-;Niuh24Y>p*tYhIez)wj7N``G21PR)%>LfoR57Ku8}BM zkbHkPG*DFbcgN@EqdyPnu3hhUy%xoz1h?A$q*{CL8FmcT9!C68yDey7Z;o<6}Kw~wQckZ#}v_PUZ|U>Ta*dJ~P%!8pxt z&({0Lf1coC15rn9htbz$P>O3Xat|e5k4=Bpca9&j#JT>O=7g^xu!0+C7U=0~$*yCT z7!xkyND6Bsy>y}qH^ykqG0vsx#GUmoM=cKXI@BegF>9`vLOU0@QEFxWZt*#W+*|9z^T{~+3TUEa4`KK&&>_J(W`LRYWlywTO z;K!YMel)Iaw8@b-Y}OB~A;huTp6RfJyP;sgCvnXUi6)Krkj6WzZk*%WZe_P$we#pS zaIFJQK|%!B+0Shu$!#KGS9T*<-PwN&*RY$g`_|ad3tNG4YurkTS-~GH$RVZhBPPQI zR!T<&z=_0Q_9U@*i$-ow-fb$7t=r%5tzFy3K#pt$aRo_U@gOUzHdrior-Bj|SOea1Oxu3YzqP8qs$|>G#X)Db&{+Rv=!CM^b$uoFXteQhCH|Eui(EZSYx8l1$2_q6l55Z1IT1=$1o&6ks z@wL`bfy*~dG=AQLkxEIYt5NFcbI{Xf;Jr3Z(dzBK$HUJAjy1uS;(RS{+QZP{cjiBd zb+(2c&s|&y9Tt8lx9xjkHLgiW!J}h$VyDQh@cPzKuDyRRwRhu0VEYH7J-+xG?iO8i z*+Q33W&U96kvo`TWISe9PZ%st79i)q2staU!Mo&XA@WNMajy2(r}_0}Mo5AXVfqO& zzFSn7p6~i_&8o*FKOfg&PLkQzpXDi})!JR+zS^C+=utc)iRmym=^)ft>GeQ5v)K*( zB5yf7y1ai!)~na5Ui~*V?987wEa;a%2|FV2@~0Ct)=$^hlRk)nMkVNofb4Efk8QQ# zouB7>Oi*Ix7Z-DM0Iiz=bk@JJX4KnA(D#B;$XZY?p^iy*YLkt6H9}eR99st>_u<`214NQAX)kS!SJ%R}N*WDEP7V=3Ig18`>>;49Ho$zCL`e zT=&Er1aJJ~QLnl~fkEXDltoUGt3FK3_&3nx{lh=~Sl%x`#*3Q1uI|y1(1arGPRf-R zdmMjsA`d#n_NRwVl19%E{8%08KBT=4N;^D6_HsZ6R!__IgPs1f4}8g5#=A!aVi}x; zYIhW)w|HQiJbHzrGp;oH-lP=>7vkN|kcSTRk&0`=)s2hgBb6d3bB)xN%PV7dL+fSQ z;u^QJny`A$)b@SH6EalbR1ar2JC6v6^mKoF9}HDBiWM(QAosfh(e=Fr7fWrj)9!{Q zNbZtk$r}+u1;Y(yr_7@&OWE&H7f*1Q`Jiwz{yrt-OVF!! zk5QDJ$UNM~uTPL0bdjnzuCm4>BS#}@;5nO*;pW41z z&0F?1x{o`wb=fU#f#B}U93%B*v!9O&Zs0cx>}4MuQXaUbIsO#N`*XUa`DEt}&fk!x z2i$)WnGu2Wci^{~p{D}|Nb>!@9F(OLvLz|Z`xd00Q(xx#&y;w|3=eS2}ek!^Vr;tKTiBDeelF%?naR_N&_ra6)QgZiy*gYq@+4G^W=gq^7N#acty4^{^> z)w^1TrG0-M6XU$4sA|&QGK30vjX1D4(gt73Jo%Lx;KuXc28t1KgK5wKNy7GU1xaj|!H7gCB9=*arhZy0# zY08VbYz{m7NWqwNPXA?18W!Bps@!0M#rk=%;`na@j+axp&CA}oa`4#$O7bbol-CGG z+6h=8ZbZPOJADr}+KP$N=1O_03TRGrq)!p$HX2H{+!MWyndmtPQu+{fPDO7fVGx0Vg{E?rG$4j7!P z-SnXHzW-dQvAbsRX0JLG7n2RRuMt3%ig zIM;u1Hx#^KZXFxya-Xbt{TQy(adUqGfEVpcIVp{64-Osj_)R0ZC@< z0Xwz6ji}VU(7U1PKFgX>$61%{iG0OXe5ikyxdO?=@dbRMYxA(eik?|rhcr);bKl#! z@4q2?B`g0fN>J=qr2!PM0c&f&0(REr&~GzmwbQT*73rU!6M}iTeLJl*+c`S3U3m#( zg!3G>JHLp+tQhYugTqGl7eJd@3(#)L?tzQ+{$?C}whO2H#;NpDBT$JVWyKZRoggB&|OQ>$OYz=VJlLs!c`>hg?<&tAg5+ug9393^1a5>0a%Rjmzd6+=UUC{@0myyBIe}lmwxV>!69Nz{6fLL(y%! zSE*;hYxWA)DA6)CC?H#`Mn~GQFj5v$WZ1SKN7>hPHqsYmQuyI4%rJk}P^bGzFJzRO&9LMJ z`BP+ifsS;_vBTRs6@4p(pL#D_#OUT7QOv%y`)T7vj2>^u;E!axw)d>UPVz{O*g?14 zh*qqo2m?maR41`_8U*dNwkHFB7OP02qUcubE~xMTYp31rEr$FAm80(5{X#{flVC7t z)b`=qA*S-;hrcl$5*>#Xw4H)F zEa&AEq37FV)*QBMc~ee*VA;&%!Y+PXMed(H7W#DlI#gI{ytwv7W`aFc;y zbv|`fY}hago*;v#*(TmH28epGuEfvtE0Xh{Ot``MPc|IG#$H%|Q_s-1P2ToCD+UX5 zGNdJSt{+?9cw`;U=nMg@elkU-b)~?bQVRG3GeSz+%JFR8{_V>1Q5MoO=TUNW^1=BAoL8hm^#ON5`IY*cN z&inMmXiMInBH;9Y+6+ILU|P+nn^_H}T0E^F+49FONy0!>tiB!L+1H3LqoO|3P_DveP-ZXLB zURn4pfI3{RtxUq+V?E|?rPMelRsnzfNEpGTW4Ki42HHYyE52t|P*`su`}DKZO;37h z0Qc2)GS=mPryouC@C@zW#>0LkMo6`aS1ed`Xc2pWb&pR_^vquEdlOtdUGw>>Z3F1%fWBT_o$#_JL*08`4H-hCY z+w<~b=H-izJyrkGpPr5Q6Evt-fuMoUk2SK23w7!+i}2!vI-E-Xn+V`M>$Ewm!wv;k zpZDc$v8CbMt!}feE~=jUj>7T+bX?jP6CAiOXbnBDL>~yG*@i40iE#lpw$m*EZGTZkget$lKN2Mfcx^|_Az1~O-$+<6S zml0K-8qILV_WUg!)5b7PaPW`1AmDHvS$;<`{tt|L1B~BDEZ%Y6*#k)`_aq!bS~hG+ zXd?@7y!_y$0ob=t^ILK!x92r&aQ4rv7eIEISH&5|~;!Q`N)U$)BQ>A_ZEA)6Tbl zqTe2In6A9l*#`!#UFI496n)OnG?((=uDo1}uknynGIBdh$S4X6IQ}H*g*!d*+ec$B z<3f$M)7!x)MLWvVcqd<{Yj_I)*xp8W93F=&=0S4YWtgv??o9Dq&ZD=-oj&=(2VmR+ zd{$&~yhZYzI-~DTMmLOYC}1$)oa5(z;%`j9af%wp|Bb|~)o42Zw@6IFq+F{L>_Uav z&fi&skmn$^E5*o0FyA$T*-GABT?06&YgO%Fp|+r_hY31B{>> zh+MRdeR*e`zj$a@>EcA!Q|q>paMA`iz{p3eu~t>Lpe|>42o>~`e#P8}>n2=(Ds8=1 zOWOo%7y$q0RtFBz!+?r4u14FgM;#_ zh--DYMLxH&&HlTYYnz=yDOhU|nbkGm$KYrcDS$yu##dOb607Rx1T3(37|Isy0H}pS z)asx9O}f?Kf1q!+|F27G{*!Ni{zo+bh~^*B{3DuwMDvem{t?YTqWQm{Xmn2PH8J|4 zl3a%LFvSNQ#m>9+3o<&o_hL`&;j54sFp#FO`&~)4tge}3twd9!2mrim0Qt}dqn(w% zN^&%Zjf!lb>_td20Yt`=OJ;52VBiJ=a!@;5!|s}$3)~<^tEOE#!Zp-?wxP#ot-PPu zlK(KOJrfM6n)^U|WOm1&afP^eFFl4VPjAQ_1W{1fO0-@1bVd(2pbFDN8t-v&t3zY7 z3;Wc1+?-_W7_jA19Bx$n^?TEy7HxS-($l9g0`?xArw4Rdtf^cd$VK25_UytZ#ruZn zPxwgK4B~6gRSVgzcnrv;v~{ld zE3P@`NOi@AHkBPG1aE_dUrQkEb18Oj;jg{zig+Y3JTrunMdvf<#}Jcn-Z^NK2pRJh z5Rxk$aW zJvu&D3Z{u9aC~0fl?Z9aMqMg3nMHZV<#ajB4Cn4ABoZz?+Ux5(A8+9b)-=hWKZB!9 zS#MOrZ48i{y$@gj^%ADHT7lsoBex$LZAXHs=B?*dzR5;^b2})vX6=LIci4vc1Z%K^ zP!SPE!BJd5>hIPrTMtKcx(=H3ZU`O`$zY^x3Q;!S9bb2e8V=Ir--|sX(?(cykQ*|t*LgMZNvNS&p9F3{tx@}P~rV4 zM%eA14UnjB0!nGWIWK|qAWx`b+ePnMREG6GPg+)gMO&o!nA`Q=4P7>q(0?xZ>ZZZ2 zDC%WKfzn79GRVBEJ4n zNprLm144Iq)U|#CjwHa#H%(x^Q}mGA!T52Ngp(J4ps{b+?LHMxn+bW1u97TMTMimn z)mb@zs`LlUq&hE*S!5h=+=;cdAe9#^J@1{R7IJSgJ%9g$ec{3F21nlsi5x#xexWMN zxg5VoTb_c1q|J1;2B>NUIEboU^rmf>qqWGh#=Wm>jV0zr3Onx8OeweHk7;TZd03YV zEib@hJzLLN{HXFe4=r7Osk&L%_khvky8^d=jbqN1~IUXOk6I{ak2W*RlMjU|$))NN9{EVi`|I zpCT@*=5fuv{z}beh~i1&iqogkzkld|adIZ*hlOl;k)4?R>arheXt;qG`gUu*0@22% zE&M{K{%vpaR!Yk5U{|zULm}{&`(c1G+nR6K>TyRffkD%&zCkBQNp2G!TeaWBW02O4 zkYtOIZ&8k?V3OrsS1HC@3BxMMgn^uH^u=SBa;j!V@6#HTe^2){)*J)7y{f-|4eu3O z840iN9OFHzmM)}XQT6Hm>%O9d%3G*ujR?^B;(oRnHE*|#WVWDuBtgiPia>tKVVuAtHf#fz*ias4k!nfIrM-hBf`=s&9k z$jCW}Cl!fLCd)mz*y@zL^;s|Z_CA}=V`Y~ki5$Ac{8_eZMlP=$HA{UD|E# zqAIs(!`o%QzwqSj`EQ+meXM@0ko^vw`L-6&Rws@k_C@MEQ(NlOlYdZu0Y+*xeSWm} zZm8l4CT5~%=39>JP96hXENR}i5!bp(jq~X7nlF|xbXR^0Y&9?X&cf$gLOCZ0XFFCJ z3Qv#ZhD~p+J_e(%mDy*cUKFBPpbq2&+_z7XX>926nNy|;+Cu{ zwegK<*fbRWe8%;(Fv({;`BKsRd;Z+F$J9(}=SjPhuroSYt&&%NzR#3K_^jFtz5g^V zm6IN3X*@&q^0UhOo}699ly6woj_t?@Z_~=6r3M%CCNAMQhuJVVYla{=zcSvZYwMH9 zE?7=&v(GzI0{{w}{P7zw)%m5+v%F!pbQ#r2ccojk3P@=2`|+hd{a$Z+v`-(E-H=od zdHbS19C^D`H~Z{=p?6WS4v*jMg>Fv2najL(UD?$X@^(&-HxW;o+jKd%dP?UUa!J)bp=<1ajO`hnWyQ&Q#^8#7_W}rm zAL?W#33Y^?yZZL&c-a-%2zd4_23rL^*ID=yW6gy7`|IHu_0|42@q@Wu^(SX+hQr>Q z$Q0rSdFxWrVQu2Gw3KQWje5a=hmOc#StghRG09yBkNjT6ySM_zgAcHiWbABez{!NU zh(j?^>arexnxc z1>q!a|A5UsHXU3L?n#JAI{+V~g5{HMv$hG}3AdhqN|-BhG?ZG-%`kVlK;Ya5Yb{4D z98Pp2;tBZOQLS(@J!4AgIA!`a+vIhJ`Z2djpM|Q=!Qve=>fWqG97L87?!d^s>vTx_`JPBNoRuqEFI!DVKKvkbt>hSd5disYu3#W*B_I_USEfm zJVj)cP8Zky=7`ufS`c=$MOadEG-bL<_Xqiu&voxk=5n@g_T3}JoEz_yRKWB^Pa{`< zr}W&GmakJQb~3%uLHfcY;ZPNCPNdG(ZK-JcUN3eI+-4=)Mq-J|KD#jC|HIy2M@6~r z{o}X_8+41Hpa_!Ejeydrq?CZPgfvpp4I*0shmdXtK|or%8H28&8%BYlVI&7;n3?Z= zd!BR7^EvC;+h?ucZ>?{v{m;N3cU{+i^^VuO4blW2Rt6z?*Sb60LJl$yoZZWMtdYaM zZgvCba9nQ266vA&-5DW``C-)M_3o@h(EroI3d>CVm#ODb=`hlx(5Z2{#mu|!_Jc9p*2qLHNyijHl6Ds43;1v~ za_8bA0Bb2KMMBLN^&Pl^^Ju)Bia*@UULoAGdt&zrw z#(m3BmC$=Y*lb)Egv|1g@Tg;d`AP|CpUFp5zSw=Jz`b}Q9>#Q<7ya&fu#l`Esaiz+ z?f%$v?^R2WBSO+8&Ivz8)0fTnh)>Quv)FOdZjDuFd$+^w21&pk7+;cMV6=|#X!4R zjbJQxo2R>)uac04wIDM|oy&%^8aeAg<>D6w07(68C+I3yb>cKQP>qOk}UTB$PW{6x|vLX|2cyD#9U$ih>a;wbW_}XDI z6plDUb3$SVxmX*CdU;k(b6&m)su+wxly21aaMiQMNu-N00-S-+*$}?|BMYl9q5j~w=%k`Y z9$}<%<{Ei@ZVIth9qkv*+#B-7TE#guT@acIlK{t^_~%?~+%BRYhzn^?Px3bMYUQ=x z=tbPJJ@=)WIqHLd_69?2`o2`2bO_QOX-)i9@yD~XE5CIm`HKaZ&r$U6%sVX1_}Jdj zgSgCp$dUaIvtCT>AO3}dBGxjg;RhzpevcbK;~?54ra19HaXgQzG~P{zhje$jWQwK{ zQu>~-9*1{mU-Y;mNgA9tARbR+Xxn)s6RGy~gip(%OXl-`at#>fJMc~p7xquMx5Z>= zq;%?j&z;O`6$@s&txd$01|Iz3OomZfusV*QaTuT)av!a)?FsvXhASG=J+V*~LmPB$ zE~8k^l6EV~C6ZSxUsh@?tuNUNQRfx*tQhNV-QZcVE=!k>_tVcJpOe zu3RDR)c27%M`L)q>yz>P&}}VCd(%#NtTJ~v!zlCS&Gx6a7EeR6?hMuHhf$XvMwC4` zrQxjpdRP0ZO6Z%Zy;RN_F=5xH%9ObTIBPEVCY0!Z_YtiKdQdMnS5URmt|Y`rfHBH%VtTg(7ZLlCtz3k3Q#9y7Izd0wjTjs%+>E?`5Ya(1zIEDKR~O`%ZC`=jlq{Z^I#fB#f!$tfX(P1AW+#?KRFawBH^NIp`e6 zyhFkE8?V(^C}wt+-4UYGNV0ZuvPbb(KpP{ZFSoh8wL#u1GlIas}@epd@&nE@NhYQ z>|7f|7_|`ZgX1soopUFnz zICY6PI_>vv61v|Y;;IHE^3_@`b`9%b64v#3Mk=%~(hW`n{kYOt5BLHHWK9B6PCcpf zPeN4-ePP&zgB||xZ~F&$Ym(|N!aZSsr0?f=6~=Toms4vOxp@F;Q_nRbBY*!7pARe5 zKkPQ&ZRcZ{?;C15Mx_R>2Bq|>a(Bu*TGvyEWcKhp+=Y-B+hxI*w=vv>wy|)Q=ePAF z)~Ujwwmd`^ta58K{P$s5{<+eQOq{*_*TeONE;1pzdxtfe+T3;JNHq^?Xt)M{ceqiN zIh;t0hE>sb{blmy5kiOD<>s+V&#{nJV-L`M!UrpDG zr(0+p(1joRD+T%gwq>_@-SCqVE%JriK0jWuu3F2L_=w{ot(IlcGC>v`*FXAyQ zEX!-J%1!FCTvM&Rs{c82J*^I(Rl`_W^4E!T_eNEEmmOke1C1D~-6oFM5!zi&x%Yg7z=3@`QJIMlU8R zg;7@x(fGkX?>Q8m4b8-gMp`?kr5H}CO!T)Klh8P;w6*ju4 zGLj0XHJ$|}ulZ|m5%K3il^@xY>?qPAQ9EUX5(2c}TK@%@iIDLo+iNn z*u`J(&O2Y0_~k&^EveM zf{6aGf+rkW+ntv_zWcmWW-sv~m2>B2G7}L)Rr`M5PUEQfBAuiT{ZY^Pc=4y{+?5*r z3h&;z47NTXO4&0e*GLY3*QAm2BZg`hmo)dWFKX;?ono_pThbG=Dy}f6?e85hE;DAg znZ!BUPX%jg_cFUnBMQ*7o@}a1{1&8}CMj};!o%+Ee$Bo=iIieX_l+&wW(LpP z&bRJc!F6ut?L+5Z(I4{2^~*!1LowWL2Ulx4Z2E$kF-wo$EI&C6*C)A_@|(#D{#y9b z18zCEOUb2w+7B|yN>_EA^~6`*k0V)Q)7nxMXpsD(9x$FOuVYP-Av#0C33Hhu7@J(< z?Fhi0*`na+5_X>-s9)Tahv;w8dEb)U?ZEcO+U}h`>a(67o*0xcGAG~clS(W~M#XW(dmT-pxND=T`~=n#ywJkFCYZM7g@ zg8D5*C7wy)PnM?!@@{m$-w))*b`4iaplhbv-(5FM>8?*XZ(HXgG7k5Ym+6a{dz)0_ zceEvR-JvQSm9#zg0A_C`8s(9avoU@UXAHa8+E+UfMvWX` zgr0PNa8^&xkhvK{JqSa`v2d$TFlZh@5#M@b?t+QuKrJA(Q z^+=-De3wfA8Jv00_;~2*^{^0~DRB;0QDi_3|7%9pH0f}O{H(K3-LpaqcBOCbWnu&i z%AE{kQkYd~nNyAbGwjo(8rGl0SRBF-OAlM zh?YkRr!|X1KhpDuD2bFv z@|agQ#_BC=#W_f+_VP2N@Rch@U2BjF{C(%6jogS|C50Lj{I*F8laO5p^c`H4TK&?0 z{-v)xBoEH`?D0MFkikjEnQu1acau5mYR>%VZ(W@IP~L}3m4VAcj1Z4_D#xSe=oN4M z7XfJ4+YY!Y}gam;1`xxPpf&qyrIu>Dg zsn%$P@~Brf6*&25b_m8VRO(qfC$_(0UowtLFnB|K?L0Z{`qemhx@x83?M%pje%U^r zC;X%eb`*M(Z;P%8r6kw0{P;8t$4bN{HYWN2CtXnspYyh%)j?mp)r8qBoqC=GyR;uZ z7*w?4mZa3+)+ZWk4YwQPaY^d#<+GNWW$x>nxd;%xY+UCC48Z@if84ZHX|u9RKkLVI z@S6!Eqgf##DEs!HvOx=c=sVGWVXJj9w@Fr4=*{yNM;chgKf0OY&HI+WTJ5 zu5q;XA+FPUrND~&&q`(C!MEq~2x*8GBxVV%ykhE8CU~$mATETpEaYqO>V?-YRn;>o zGx)F#v!oz)U805mGlAQ({GSG>l?!a{=x%9ha z!vTJ}*8Z@uS+E>~=Alqe8Ykznjnn%wc-}_eathglx$8%9w!B`ua&^Mr_s?U);OGUn*!sliDLDu;N+d)*(vdN_k3vDWWb60_>1D|KUL0`uCh z&A;tNggl=ecIQ1%4t}jG_xv4w^bR70J2g(Uw|ONH!svM%e-gWYS#+^A>L^LXdwf!C zQo?IK3SEPHZMPN$j~K1Ce&UMe38OalaYmE```MG4BsJWPIzOyGR)R<@t*9JG#VSc8 zhn${PuOWSBtApr*4ak>L8w8>dB-gg9)5{i15xGq+dzrZ!b0Z|Q%&Fdyh}gXr zfpGW|Zifu8_9Dvq^SV8JnrQsug06l1_^@OUb|Z{>PbNN?XOj`mI?duHMU6^6|66*) zs+c`}oRy3}_N~Y3V*T*(*U~it$UT7&6`NdDi5eAwP!Ps{&eeuuKMm-~t~U3XLXscp`zkbj=4Q|-R7N5yk`fJSWy3CWS*#1Yjxa!~SGy;79(&Fh#wszO$$M;cf zJ(!52rMJ4ZC7ynhyW%ATxn#5O=*_2?C%!q)pNP_=5Ea#hQM(_lJ?bmXS^C3vahFw# zQYG8!8RA`k68d-9vmak0WOLeX8rF8pyCl(V@n|bk021y#DB(7a#R3J_%GRqDqBB5A z#!Y%Lf`l{m%3UQ~1j)<&l22JTm>X)20ZElt{#vWg+O@9%G$yGM54avpyc*HW>Fn|Z?m2po%ywh+({)@!XOp6s~Z zShsD&(0>ewtv33`-kn&qu#Jhs7TjsaaP~D^oDmeQ*Tk%u?5&^g!IUNqENoVX=+WDQm`ihE}yIfL$A>tqO_kK zq~aur^83Bpzx^hdcA9*igzl*5BqY)$^7C0^oeEx#c~S~cu7mj10Th;t)wSS;ZAy(| z(p=T%!Te23cZ)FB`hLHlMNM<#y~jxmpOd;&p;WvXRvMCfP@&?L>JBa9y)*FqTl`(0*+Tuk@$9#*Do|%BYkr3{PYCn;#=o_qc9=5Luqt@vHL(6Ut&Uz*l(GTjE5Jcr#TX!)0>ukX-f-RI@npjPKNkBi7YY=oM(n*+Mp--8h8 zc#P#HqRV3(F=Uw_HW_Ekx5BxX#y*dRfz=cu;T~5xFg|2&ks;F*&uArG`K1h0(jSCT z`;$fP?;Ehki;+k$9;9^-Y~B!m08`?3GFjZ5wdFc@SeI(b$!lpOQ;xP0D#z+u?@P*+ zrjJ;O1UHh^_Mr0k0`q;MgK+W?Kfy}OE-A9ZUMQEwCEG<0D?_{WLDN5LokF^zI>$_6 zcnNgdqi#MR8mMCG6e>N-DRs>JL5m5A3Aq*u){ETwvTu~!z3>iTm!KklK5PtSa*t?i zfQu;q<=UR$+p!F`=hgrteRoX9mG;}UIT{x_R5!xsH@_xmHxR0x4Mcy`Asz zyoEjuj$D#ri@gMT{R{WI@6&#``T2KT66`mFE45tMHBPzH-g~7G(Vlh##!4}4=0RO$ z@XmmUD^r33&x4wq<;5L;B!bl;xMj1T)YZM_*3PFD0;N?cKB$gLEQ`bIsAW>Tv zpY-5QmTtoQ9QI}cX_MMF%WYm8)TW1|ZS~StE)hf_x|3CMm-~HOm-Cwaq5r21Ge7_D zl%K$dU|DBMc%P8e0}W#)BH#2SZxZH?sG0qekJ807oI6XuPijgUXjxaBwU%Lmo> zMN|93HhV0{(*Ub0X%u;Bo90K)=PrGcmDDR%<0K}|g<9^vSvXseg}k;;Pq}hmx&2*O z?!Jw~;k#BAeytOK`%emo{rhb0#&)uT*!#6EWF0ndsD8|}N6>qyw##f0PVwsI`w6Ev z0)?2V43}5!Gq2PdO){+})A%LKnMvgH0=Konl&yMCl!C2)X?LTTl?^# z8lzyJd`a3nPEsc;L{xhM_XUaS1+cqFP~i7v05^#HtTgw3or_QFci zY^|=AR-R)i_=qhH~_htYgVMX$8{VNjH zZA8dRa)CC~-|D*=2HQwN^h;b;oKF7SU8SwOcmhEsXw;q7$SFNC$7{tB+*W)Q%0dC_ z?!$u+6ur;%Q{w{Tb9VTO*rVj`EkL426R<1C-PTE_cscIh;}Ah3$^~<+x9jc-s1if|;K9MW7t=elT_55U|09X1?B;r0T?M@A073hY zvS$T|vOsAXr~X05Wjd2psQ!JL~3+eMNV&>q-3`6Z-xMqJ5^K>xlsR zXdv%ZfPS!+h9{QpVZfTvW`$1a8764A&#I+sYb-2{vPDBEEeYC*v9`>6QrL=db@(Ja zsZUbV(foz4$R?Jm8X1N75K3&Y5;fU_@D3@i#a;Bm-hqHPTbI3mClJz;J~4!97Wj9 z{j?_06Si|n!K~3=>ssFkqigg5NRu^fSRLA7UF+f>9U1JBoD^<7g2Heg$2mi|SdVA+ z#c6Q7Bo{>{z=|bgdu^W(PYQS3ccJBf2kCvhCDt$AK>_2xKX#((Y&e>FFN9brY4JPN z+|{`xw6z8M@Df;+)$< z$gKhv&iVtL-QiEx>2B+%01BaDi}vnn#)@1aE4K$;xb*vh2Q&W#F)9lB?{*u}b~ z4_IZ0$|iU>CZB)V_W=oAN1%J@I}Lz~p274yV7D+al(x2Xfc+=c64Pfo3-ay)iEUt4 zND|C;zd~$At3@X3mL$-B`s0f~jK^MPQ_yg6cV)W`iEU_ieg=?5lE|)F_4Z z!nLv~LX7O=?r;iOKeJ(VJ{ZFs4e~#$Q$KzDm+jU_;0tO{Vny>(h09B@d*?jq@21I$ z=5>kC&H5vb>eQd*Zc0>Qm3y5}=&b`9X$XAz4+}YD?`K;8V?$RpkZRbtki&c@63Ki4 zcgl)ZE1^_K_r+MQqUtz*Ovi+;|NW)`I?lsbsolm?=HQGD(c^>eb5=o&4b~e);$fLY zuJN3O(L>k6&xDjb{VAoUsK&ljChQyVKM1D*a6nejey`J;;&Qz#%eXLzbQr@CJjpQb zNc-8^nuIDEtOG`5$h|fNZk(j!OgC%`H~q|E9^0o3Y; za775ewpMp9hNgCiY6!FwMOV(DhkJk6;@(-L24y$A^6NKUOaRc1m~d=#`SFB5J82F$ zrM#weu11qZoJWsbtCNoecinKNTeyk|J?6ibEtc8U>fGY6m9|hwe-SwP_95N}5wZI5 zf_6qxb3rRFRL0qVF)B}rnfgca=qvsmu%df`ybIlj%cy@j-$tG1mMNld|Ui1=T()4dMSR+Q_hk21wLCwa_8 zgd`rO4q68t_G|3PP{AO5A-9g=j1&7%^@FQ%2vKCS^_-P5Ow=os<5$G9Yq(e|5v|E()BcG))>P)ky>b8Tk~8C!3x z{Ygaa+JmAIl}|tNb-KMikyD#wO^xY3Y44fi6)^3thWpeqtu`;$^V2#!o)lVygw*i4 z9El3>&e{K;-1Kc#;2-JGyVMM}0i)Itqh8EyQr}(>`8Y6S8aIMb$R{=;*{@qO$*=p2 z&B=Lxfw_}nG8;1sOVkZUNx_5IvBZxHF0P)Lu5mKQMhAF<))N>3O&G#--~E{Tv#p)+ zwXww2g}aS;V99xdZv2?#OIOLrVI0^;%NL+qkFU0chR=O;x`KB-p))?Nji2OSZ{#{P z%(=FB`!;@MAd%f^c9d5ali^1EtP6Av-ZG_sSe@AwbK1Pikh@XHd8cK;7tz+T-)w9&Rc&y>z}-LIHQbv>X4qj zx9GE5S~~74X*sS*YS{u-&)2lq^a;};quTvuY*W3#^xu}2#iI&3Tg2ho2sCfE3ckC4 z{Wvgsb_O|B=5~vISd;Ws3#s(C^9x?*d*M_L-AG?b}N9V?|*X z7ut5cD|dYwS{ACyq>gIJB~vCISJzXh$YqhQdV`6byepy@nU8&F8+yZ{ z&kr;4jO9X4l6-G9PHKi)`hqp#EgPSIw(^|k@F8a$Q9g0Yy~>k1O&RsHvuPXWPPbdU zB`sy849Jnj$hPtmEyoFE(+QG28L=ztX*gy43=}t`Zc{_#Y^QF9@ z7A8LqzdE4SpRGr zk^2~JL^J3cTfi=&+g$^c_V}J}*xk1h@qHSGIEg{57uFPd2^qCLd%h~rDk0We;+T&d z#Jr;j{*XDeIj$t%=Or46Ob>2mzi_QXC|ltv^@>b3cV6bur~1OIjcvAn1uwf?feQj) z$#wbZcHs|}3&Be?CW$OG&y&9$XL$Fi{|9u7R{3{mrs(IC6W^xo3;aG!96G>4T)Qr~ z>pKz~YPe0@`Ed0-&6ug8H|)q1cJHJ~?2>bl{F8UV08N!cc&iJ{^yKOI-BOq1S?1nWQtE#s%l~I-cpuNM$Becd?$9*=vAXLUiuP^Uz=ilW zVT!MU*@CVYgcdM|i|b}j9C2WJPyMMoU3?pQu>~xAw_kmcX&Nzqt<-=9Ij>B+D+#|R zb$|AVxOSw=g;IpQgTQ2EmvZq|LcR94WhR1O16&q6;97aIt|HeAq0L@O(Zn4F!q9;= za}u4tdH)HCuQCr04%vg_dFLLESdoRO{M=zBpX;sF;=im&`BZ=L-$fkO|4%RC_`AO5 z@1yzqX#PH$zmMjB@1yzu_|beL)c6WoHfSQ3K1IyMbM^99EpVHO-%fq8uAA#;_Ga|9orAZ?lmRaaP%s{I|$|%S9!do>JZT9yKzvT96$+K zy8ym!|Lv>Ss~ZNY2ZLf32x0QS=2^?f#Q%C{>^k_)Dydw;kFOL(U}J73#sduSSn2Y% z%En-}?_xTOJDXc+WqW|h^UuE>bW0p##-RASE+0RWESnQ(;eN`q z4*4;oV0~BZlzCu*ZK|2i{66_#bUp?@#{7l@E2Hci?}xPx;xFgcSLQoH#}y%9|Kg1< zI=2C`4LlYF<-K$zQLXj^-GqP5W=UzGCtOCXyngn78PXN?*YfEOU2yCkpiVz30f#>d9n@t=M~ z$erEhuNe?_1XGO<;BuiA%@b1!cC~&h@hYRi=!I1LaRW?WjY6HTN^v=GJ!ga^g;auY zc2K&1CB4@{wb3nycbD9FpEwFO7!sNS6DWT+eHX&6c0{aK8p+xrlMQot)r$wB0-tOC zrC?oU2oqiUo^Mp72zhN?AEPy-nMD;vJyxWxO~4@t*;+D3<4-h?Zi$F&*U!0bSTkXb zlD)1;F$Euv)b{a|zVBuXEaR!^y5;m99E=cuz!b`IYB_jsN`pf7#Inw9uk)rh>)~T{QRlUi zD~;!N?U+d2av512l^n2D=@jT4gQDvyiP7Pf4sOkZ^1lvyQ!O+o~AJqNjDn+U zgF}u=EzH=xR>INqp4xAm7=h(kr|9`(=MXvV!1ou~%VB4l9B{XKIfVMVE}Q>&IWF^h z^Zw#r64DO}!C__TpzB*n4kJHGN0m8$*EoDgd#y#o|CD8xHGNq=r-g?SuC6J4lY?Iz zk!ew6)90%j*VW(Hs)Wgw*9=Ko3s8BN!SI-gb_X}Dgh+;XjYk~Mq9v*hzNf>3R-M(T zZPU~Im%~UIKlmznd}R#W)(PD`vYRsV)l^~hUwLo0wSK&j>fpYbZeO@lKUYzIA&{Wy z+JsEmJODewABIr`<;wbiWxtbMpv%dclE0#LzpwWDYdV89uRs-D4K~=}Y~G^9+0sqN zcdw{RD8LrM6Wg~$?1#z{Ec41iRu7)1`)kl(^DmN70u&C0pDmt&bj zcqM@K^EcqA)FJ%!MmJXEzP%uS+JEkb#u?G3WjZdNI;Ku^K!O?4N= zY=3MAr))gf>K++a(|q+~TT z^VUD|FQMY!e28g{nh{`0XLB=kYxAB^IXwu$?idbQCddV{|23GQ{--Q|>S47@sC4o2 za5eGGQyN&mvq)V-vr-K4$GevXx6IMaUYLy{Lo8UlTDz7LIN1#r;}-BY%va;* zajWkl7$1V8m}kAlG)oJA1o!uh1Z)7IQejtMJYLznG_^nMSkbqZnnX#~S)0}Quw*J-lsdM~-{ z7lkbPC~_Ef0X2HzGT&RDMgO(K7`LGOcDpBApz8?Ngvi`QitNG&_}cC2{>jdkRI;eR z7+Q8}bHDG+8uRLZ^LmTmXorIQI9bWi?bF$8hig&%{*DQ%=bCl%>(Z#vs9MT+msT!0Mk6P z73vPQ~Y^IuJ)_-nJE z=U+|DK9u|t}tKB+%PedVo zC-~$m(-2d4a5fj9DJ3M9qeV8If>42oOqY#BS`oYVU<(5Qsv4=Yy|~aaIds6--4T`q zZFBs7YFyJeaM)*fG)x;J*zRZ64bVmgwVBrl#vhto{FBhFvJ}R*9szh^ zZY0}(0c;)_cd2^KWABy-lcI;xVxJ-eOsxQdKz+ZpPN26Ud7T#ZE)dv_lTN`9oxz+R ziO@4Pm)a3!?>(?##fB8(R10kPS`r+e`(qA)7ws&&QMChTLTR8KF|;CLCr8rqwS93u<$Ef2#;FlC?zyqsuGY0h||?uEgro z<;+l8R{hXSr8V9`==E39zeWd7|BVj98-JpM5=97=wzN1lXjKHNro!O;xEYXPx@>n86O;yO-`bl; z1{1&`e^vn!aVPZs#(Wi8$%9Gu5RXFG18>ZKLU&%EKMx1><5o^SSaTo>pD*zzdjgQn5W9HNC0015B~>24)UQ7uuvI>R1$!239FA#MF1j^u z6mEdsBFi-ZaJk>hX$VNkBB_~gU`taG({4a>Wyzj=rkG5fwMo>}E9r}SvA)p~N0_ZjZNZQ|$X4)ztrc}#+90c2 zy!bWsjFtDv({dA0)L>99`SE^2YOBp?FMm_8`ClUAlkRK(g54+C4*7|mR7E@^R+DAM zjz<%Ql_||VQy~+j=E-0bxOVg*uYMRFe;hk!{gMDmQOwd!SQ^C87Cmhphw%*Y1b|v0 zIOh#RBVq<7lMY2}PJxZX9Ra9PE~kBH7k))lHU98zLz=`Az#ujz&x#91MWCkxUVx@D z+!^X#_J-k(OYwn#I5J@O2E@Gp@;eVLTf%)npFOONOqa<)x30yw^HFIo%>w8Re{N^W zf~10;C>dQC*2F&>jK|sFFdQxR}2-!V2W+ z_I~J;>E$U&Jmc7F2Uue;17MBDMZknf0knBd(1yE(b=RA;bQVgweIG!Iz6bkypIYMr zcIu`Iy(f)}FhSp6-7EwZxoiqzf8J-#ML$mkFFoSKrbZvQT~%`fVl2w&_bu#MDBsO} zu+?k5aNBTJwA4R)0r+${Q-kyB{-}}CNOHI~wQK&G{rJ^Hzxi>sU#694RFNERV*=ZY zPX}ZVfBke`#scdUgEA?jpCx`h`L7pO_pimt*l?f=fyMD>0#9jN1@k8Be`oA6gHCfQ zL1t?{998e;<4w2>HrRDs9Y{a5t|RURC(uZSeIHR_(gd6M1WPY=l>IJZH$YS4mPZlq z_i?X_4sKAUQ5zW!#ckVg(ag*cc!( z44HPda(hTw+^vhUefpS6r&MdX?8#111U$-XSb;P~O(RNH0<)e!2FQyiZL61)0LdrS zb2McEr6>R!G&usMe^{j)0Vx!5V3Q zBktM16hc6c#;zt^wscd<6~j&7wHl!9R5#v9sm>Q)sF|5UZR%Y0T|hEoDAt(GZZKoQ z*WisfHF|%`g3$a-fzQ8s=?(8Ozg$Gd%t5s;BG|VX)sE#se;`v#1-v=ohtiHc*>TeQ zs?C7q0h(x;)K8v&zFoQM^_L%)Q}S=y9d;hT;$Jyd;Y%X?|c5f z=kI&|zUS|I{=Vn`W%u0OD=Dpb2&Rw+{P;{tS8wEgfmm6|5{xory0!byT83XyF6f@YH8?wU3@~=H0aQ2nvGWZT^BtZG0Gw9^EP2;i=bAw_)98mq zj+}q`so%KS{^bdOKHMZ8f+f@;a^71*Yk_Yoa(g> z(`s+1Xxz_7QL`?FW&HWjRT`atJ4J|EW3*BnMm{E9=e)CYfs$XjmFOPyPdE2Y-oNoF zsUKZ7-$>q5PHrvBj^Ef!qWwsp$^PlOEfxQp7vhJgq%rW-TDn=EA}_1f((I9s2mSem zMgIQpe+~NvY}jLVle1MZn3~vvmtKaDwX%tAvi>MIFs!zNQvL9HnuMoaz{&TBv6`Ht zvcqkAxfc6$f5$bz5=sU;an>&Q%oQk68?(v$_(Y_CU|{Aw;hyQgmhEfMooW$sTKnV6 zWkb|Kn_&CNl4CD>Dn0s4{A4{pt`LiY$?-RSe|zcJdBZU4W6;YG1jp=Ro#pmD~Z}#V5_4sJQgLlI^BBmbk3zg)9{JAQuFC* zf99BWC}!;NxB=Go%#JM;^hgC~W>IOz%3{_JUyHTA<3Z!mFgdVutRpn|SPkrCht0{P z0H={V0`|DZ)6*I#jt`YnJ*C3v&sP5N%kL9MT2ec;wSaA-HB;B!@`z0U9EQf7AXPv6 ztaR_qP@OeywstkcsgRT@=+2-fRKkCZf0BRoeTCCsknbVYKiPC$>?LN~Ek7|rtv2%0 z0hmXPab z>jWpA)P5zGdW;qvSt7jt3??-nE}%lhyyId5-q!vow8&&LsU4;*Vf)jMy6V?=f9c=E zh5?A}P84;fYqR4|Gj280&HT3k=t0)nxbGQY21~r?Gu@ z@*-xkp)eS|Acg}ws|`m{f4`xwQ&!b=Ya;ZX5mZcjf|lQk8lu{jo=VA&soPkB2x(il%kkshg_U{gL5Rp;lVa zyy@&4`1!Y+<02AC{Q`#%Ya?yxx=df^W0&IO@P1%CyTw=gbPepbypd)dt7f0$Zx=Ar zZ!YSU*J*RhXmQSlioeg3Lr+Cu%x2^5dj9DeRBliV5p;UkUP)C~WTE>2fBhS0vzG5^ zlx&=DTW@`_A%mv#{;B2=ZRNsgKw@yRgOD*d@-Kvcq3*wfEFYZm6G=5|8vz;{Ri<-! zyzsraC?~`Gw9jzJr=iAf@ZnD%9I)vb<^;^tlx?3PWK9O;&z&;a2-xAHRuF!JVO0YUCg%XeVdle3hxox z9=#1lb|q1*2PFlM!B|dnj;f#dUI*3Kj6F0v1PpRb7+T1B2{t>6w1G*0a3L0FRui#lovgcZvp(Nn-82XWgO@)eUXI~c zL}~PtlB591H72o^rI$^GI|VyZZ``Z&*PQj9GAmWgu)y!9a;IoV;bZxCMP_l}kkL*u zZh%eH1HOfkj)GlMf8T2dEm7$=slJGm#e>-tV1(F^!@YGLY!LmpFcet4Zcto`na>R7 zd33U#mIS-f5?niEKXH@MYCB|wm8p35X`X1?K{o1Vm>dJhSgEd~?-5K|5sU#2$$wCZ zB(Go>+`Ab{k+yw*BuOBR5jNpsfQYl?e)#8#`;P5@%UgItf0r%okf+3^qnMyboXdKN z`2r%|IX;z?7B%Uar$k)HYP8@4u?(i|5w!1)+`pC;*L*1c-FYl$!v%Nei zf`$ZmKV*|i>b)30e)Tiq;);r;OueaF+dU=S&pvZ5sqLqOZU)%bnzfN^;Z$1U?;gsZ zu5ij3n~Crte~v1ACtS9WquCDE)NT@v-3SJBUV(~tWU@GCG3+KQvL^P~-kV5^=E|cs zRB)wKmGsw}JD z*m$YA%0`PZd|@gk5FE{|AIh|vWbdtag`isNaIsMQFMvnVMNqW$N=ltydDk#hedU3L z_1nv}I&O5{J_%HI-Y-{(aK77B-9wu}`CVrXN0X*-3`ktY*&UI_Fkr6&%o`G!A3?{| zY|^}Ee{53+H;F6kyWf7_GE& zLT%{7v9BHc!!bDOuC#LGQsmTBq-f}-9<(G5B$cCkpmGi%fJrdVLhY&F1Xyr>&l8wa z>q(nXTU^Il)Nc$->p%%94Xp)|Fp)8pHq;mtfA3W%xVs0=QaqYN;C&`Rkz1OWq%$?4 zR{Np9yUi=%Nr`o~O}ugT%BxKy8Ze@9Bmu_EgSB9?!+;@I2{W% z^R-?++&S4agf$C<0{GK#Dhj_fa7$<50jE&jt;?qvuot&kyNI5OK0cB)_5)gp&Qq>i ze)Zw^v%=HVu!0y(sxhoRo3dv}z?d9Scs0f?2s> z%%QQnwvNiUnCI%ivrw2QQYbyx>WQx%;=@p!b+)Z-2{tAK?EAdwlVt6z6=)aH+uG&3 zsQedj+S2vEaOK!zFAX{|#vRxEMbS-qe~uBBm(2jRPL`=&==prOfA8fm;mCrZ|J%F% zbI-{oyY~P>u)eS((Zn2!^pxy>HLkFy`lbha;6xtK?X)em-@{tEpKZ&%Ssl;(sJ>i; zP9b<3KtmO=O1Ux?BIZ&xl#j8jccAFhC~D+USD>6up3=Ni79#z|`wGsfn^D)@f6Fv# zd(bzrV3cTPsP_`l_pHh>6d3h$FtzVZ5t{0~L}AP1opStPmewwwD`9Jk&rP^g#I(Ci zJOlv!lUPqWEW63!!@{1UW64S0ktHmG5HeBt90P_Prx;uyY)R8hgwE;6VBBmHmXp)u zNzh*BQByXpZr2mBiZNr?0$Ijhf5xgx`UB4|BAaR3p_KfEOBNhce~wG8I`I4h>(_dz za=w|fak8FGP)H@`Ms4+SLN&7Gecm8te#-VCvKz?nwYcNCq6ZW*wI=nq$yaQzONy|( zr_z>o*O>8$daL!VQIcBS-|OXSu}GlO=76wJVNlS?cZ$$7NfD*TSSbbke|3U)0$&N- zpEhc!?D_@(yPP)%JmT9i6O{(tRuWMlOa&(`Y$WMvsX~%SB=Sm(=TgMfEGmDE8 zzrD3ek8Mp>kLVa>0<*AW@6I_r5v=AkQ&Dv)t5WvnOX1ldJ}O0@ zYS;448rCm^u_Vduk6%k{e>Ej3UO%Y~k=gu^U&8+A%eNvqoMRhoDl;mDd}Dw-p?OQ3 zaCR^jE46EMv=&V(QqrH}-bK5?fbvs0l1(vg*Ss!GZ|9vrRAS4z z?{TAQMXB9&!2GIhA5vw>NFrYeR-m(gZ|UsBt(BHP?@H4X2`a_*E&wE$F;IlJ3Z~fi zu4x?J>Nj3F2x!x~o8P$@h?1E*tx{C45VH_T;rCyrLl~jpf8F^DV0hg=(?1`veXU6_ zRt|qMv*{w4-w9^UmB)R^G?}Q%_e?5ZoHVc<|2zfSuUnZ zs<6vnXES@Rf0gh+UjXhfa_PiCggl9}JNkl9jl6UAIrpPkwG36U`R5NdiD$<^aR1EW zbgh2QRh`cHqe2~TU#G@(f6?$Xg>U^9cn=fDg5(lfnq9!Ix-OzaA9C{jXU5(p-)-GN zzehtf*KWQM&Ss18+AJlw1z)h4f;QBP~0;C|j$=4DOh6FbgY%{Ix5@~Ke z^#YU2BA7xND@5%!BaOKx%>#FyS-@sk_oJ!ZV>Cr)-pk4@uB5Yp2`-0V>z9El54*S0 z5H-#be->{_eO=BG)?h%f-O_8HiU@Jp%o$;civg1yyeZuR%y^L4bw2gP#UHQ5#zjm9 zxZyc5>osH1)WMe99Dgp9((-?Tqw1mJ9zRFcuWYjm%55&=v@ll`z7uOB3sSSik?+gF z$AbAy6YWDD+(+#}WgA9;@Y$oHj2NIwU#k)MMoDJ1qdDGnQshKwCNS-94h_NEVCxs)uJSgktGP7M) zBObF=y8I8S&M{`~Tf_8ErH!y}#t^jRaEB!We%2v8()}3D0H)w)tbe|N%y%08%hYz{ zkt~d%i}Ps!rNQy<@3pUgh*<_KT^k0Cf7q4CkToz8uis>}%aTefxIWK2!jizu;;@ij zY-PZw6JMEV!TOy!Ru1J4r}O0FR9`t4U~dQ6^kcibBJW4Ot)eU6Zce

}S9FM}UQ_ z=tLB?`vQHB+~M#8`%*zcJsl2{pyL{7`Rq*>)WERNoJ9Ov$|=4r(EWgu@4``mU>4PkkDO8aNG#*NS=tgnzxpbmtYO0ddG_FU=+0lW;FF#5n#ml zYRELqd#Ibq3hsYi@9$F#&YqD06k%+)46d ze0+xF%9k(YGiFTP)! z5nXD1KrZoKQgF9e%2I}tk{@{3NGvFoy|ZoQVH&lK02(cQ|Jnh2Jm)6oe*_r1M-s?w zMsR9r8w@MEzgGOPw&&8lc&>Q+$Vf>mp9s~Dk@f9>eONq4RioR{4o)y#you~quN4O~ ztMJk%-@>;JB-Ka9*E;V3_5|Db%9c{KN_77RRy&!0SdqwdMND0>)CfyOBFgdW6|t=3 zMDC^8n-)Rt+2VRb$D?<_e{CF>nIW*5+zqt}*?GT|?|G2v}s#Qlc)p z#-R~ljjm)5!&=NuaPO|I1?_r5OSdA^`Rf;a=8XLd-+X8>)KwVJji0;md0+S{t|~6L zOqN>VWPcj8DmJH)sTQ(!p-JgfQPd*#Ls@mFtC>?fQR!TLaCZvZe>|mHRy!CMihnlC z{zoKkPd4FcA9nL7q=b^6C5oN%o+56mtY)YzMr^Us)p{F(l<}b3=VLo%WNBiLMLHb< z!Xx47m)lS()Pc;}!IF_yZRA#Rjk@BgY6w^VYCyg-Ow^mxL~(`*|0BX%YAL|frB44I z(L-0$`M_!AnG=xWf1H#dg*i<*y{%y0jio`>UtrjW|1IOZ0U)?EMaad?uI=Bi$i&&ct>`pF2H!A`8n2*etG@pgNC(2d#?K@UvfAQ?!&N zoyZ>w%J!2v7!cVC*iX1xC1=+kiD7MAY?D@^7o_3`XNe0vq3b?HhQi#)` zXM_;!j@asan*P(ZY~PAGx;16W&w!0$m>h>*n34mN9n)`knC;y{2aIWo4H~p#&raw? zv{pR9B4JF{9_E9gmHZQ4O4GiT)Sv|#QlvAB&heTY!9Sbr8j^*U^LP!v37m7cTjTfF zGYOB(PrlqVe-H^hw5*mHh^N?xJ} zzdGnMgbF2K$J3Q=gHzeHLEvcO@a~lzugDFlyc{j7fBQe3t)OSMxX~s!rkg2L+WW#f zn4F=VeJ-Re)V0ljDT;G+Pz;fkV)CpQ%wp9lnFVyIIB4W;IBEx)WQe+$Amxgw)*ihB ztS!)kI{M|ZV3y=1mI)_V37GqO7ZG%zk8b&N;qoD`=p0@S_>u{uEw)GR*;9X+Mm zyb`XKf4GmSRjYzof*GOs)4&_*nsbK$F4o5j*?kEuY2e67EXE!{2|<$IUJ!BUJ*NGu zD1;uzzrFiCbtAoc=?)`5d~R^c=l*s+ikd&SG~v%lmbvbq?!vR`+L2y>aD5V%h?Sqn z=XFN<=C1y^1629(znx#_-0~BFCv3BIUW#sff1wDuhLRur^D3JEcRT)1KM_%A0GLb@ zq=T=MJX$Y4eYiA`Cd8Wk&-h7Qp_AZWxG^ICTl{3*#cLmQ>$MZXAU3(tYvzZL&?lX-LKFN-jK{_~oeMs(xp zTM{x;0DvisXFvGW4HwDy?>LWsWxhgF4*;+54^k1oy5TB~iRJYB2!&gBoPkpz(eWk! z)eX#dYCnBcqX@riISoK6vb;Y>F{%V9Y1~aeppMRxFGSX8(w@GzVMYpe}-)I z#oq_=_ksLRbzkh4M09b0AFzZy)lFpnZ~N3wa$b2#^R2%q!2*fVEX>Sj z#;D>wE(fBq>{{&qN~txd^a*yW*X7{x~wMV*+Z8~MwP@dOP0 zRgYBF+O+;}CBJ-)9vAb7)$c-_p4`CWb#!0<^6{BJ9}oNc@qa)5|2vPzbn}WhO_u%i zE{e$#&JK!-WOf{hSfw%R`4dfG_ZbtLcqNi{EmKY{B`je-752VgpoLFnf4_&-F;P=g zOw9Ub@aYcNStD&`P%{wzwR4&_SjBxTxzuI483TcUUB+ z3BycO{wy~G^ID%-(1qr?e~ffZQ=2%v+k}bE3OQtlOZNv%OjHsAlC%4Q!=vj$;7cLa zJ3~?^4X|YjZ+We5z=pboWbz<%ypA#SZ0~RihOgr>#nTBDC*ENBGeyK^LZ&&reSvT> zikd?1>^Yce*7Zq|nT7OCH~sp-?UyC)F()Xn#13qTVG^)WFy_(Ef1jFzKP*B^IC-?# z4`ro;Rn`-)3{73?Fu~|duA9h%X?JE6yV0rtjb)1Yxa*R2v@$%aT zZP_6G|YFI6Hc%SZ6y)N|CP&henna1-ZPl%3)V!yRwp3`|{; zCoMv8n>an3_83Z4q8>~ZewUD%vsv%Cg~!Bf>}{gN7Ge0K%A$))-N(lbFqzJpCVJLu zev3|(ZMtmIbyZ-!;>JzbiITpzbm)@A^*J0GVZr{Q6MMt(e~EjJ<|mpJV`}tx+tC(m z*Fz8I|F&upWjyax)(Oi5)9SLfDF*b~J4rkrWtv%PLZ z>~d&#>t2jo7VA+9*Z)l6htzK!M&@i|cW^4Iwc@8MiDPEI3-`KyfQ6m`;0PN#3mAnE z>nDnkAK_lze+<3@25c|w=Uf`h7-}=hKe72_c~2T|fT0Yt{c^WpCzGeIZfwGlnVv%u zoMwSLbw?FqZ5x^sm4;gkdWVH%SBC}_As4SBY1uFfePGXw!%70C5=XARfiHeo+xY!8 zU5U{3n7}=VMHU$9;&!ozL9S3{ZBI>xgng-j%Sr(sf9I`Fm1mI`Y6U@Qo@L0@Ulf3C z+X|L0%IV|p;$oAmv4d_Qo;SuB$5}Sp!64z7wNKK`)^ERsn?HL$l({K1;ieH6G3wTe z{yl{7og!2f%sVORS9)o5)Ze+OF%B)MoyUeZ zqKx*ke(zA%`YornE|?}NLabvY3Nt!ce=JxNc=r#d{VnjndE)0+>ZC@jvb`tO zehvt4QPhyZ7YGJB4unKeTZS+vDbiJ`taB-#*_+s$$=bX-TJ0D;@}onW^z z$n?0BsLB$Ja&txnBCxt_e+)q8nBmpt=#l@BPy85he0G%LU|Ketn;x$9WXCPkG^V%_ z55}O2THw4oQR#E#&?I>&G1f`p1f3?3S zX^TeE*MmjDNBuKr>O0jh3!{Fwy#vM&cbXJc2NV;Ecm%g@9R_QJGk)$d@1;m|3FXtK zqTai_bh-u&9WH#lru3Zuf}IdM7StZQb$L4@Zcj&L%$GVsKKC|OI;<&93G%w#;>z00 z^_xjCM}6Ii3q7;GXD!brtKkD7e=Zak97=bPc&E_L+QfjI&T&wNM0Lq0esXL#h#urP z5BCe{@E{f<$GR*B`vzoFA!_bC0Tf$+fX+2*j@To*RL!n&o2~~iAqK?SiCBKGZ9>FO0ywIdn9PtCY?cWOjz#?d+b(QA;=6rAT`k#kU@5_V@}waY z+fcjT$D10__CLF&!X(%Ltxf*RCf3fPyLFs;V_UTd;y&8Ieu`NY@L#x(rDT=cM+X@KXEn{(( zMrxSlqC20ru4yM)9twt@_Ivel7p>@x%`G0+*8*DJ5L(%~f8Xb|*Ll+r#X7=P z*b!&(UHQ)4`oquibL~nHnq@+cT6debV$&DfEKIS#{ocQqnM!^-+0dT6Ki%Wf zTRC#GD(Q36T8`jF)Lr3|6T%n)BIiF6vLs!kF1_6R@?``@bv7K*E# zFfi5cX7Oh^?iS(Y{G!C7KXC3EysTC8ZsC5yP0g+c=(O07xHVC|4ROpg(ws@6?TSfzOq-gjz(#<+f_OtgP#V@ z*rw_;_1kO9GQvNWRy3Dj8d2@op_FKmbo=15Im|K>p|M%{%=9WcJ<)cFHXCN`e;j_| z+qmJk8IsaDK=1Fg&$l}jiuARthQH=GA{U;-$Up*Pf9gJRp>C&;M|Je>8h@-d4lxJ& z|EP;0f;{V50%_l>GD>WJ{JPJ5zg9aH8e`2M{Y&h|DID1$d$Vlg8HNy_d*o)wIs;5Czbp!#dK71lvpq5fb*fuyz-E&eeycA(G`NjM;s}c zy*qJikPlLuWTo_cgUVpW`RIrWlbP9VK9&y|e?%=v+x|}>uw#`^Qk(1EA?=d@W_?2H z!0NvgA$RuFo9kt9{M)Ohv@5^YWS!GtIKx6O#&%b}JTO@FTr_O&surm_({Q$ZF8s1u zBE68035D2CX?X8AR;b^Ct*euIL<6&lE!Vi-gP1w_Em+I4`1c#3i-49gtf2*y5RVow ze>!(*2vvE<9bG)^ZrrR(hE1ZprMq@3T6Rs~InRd`NGF!_py;0kLQ~5D;UOvkbp?oN zS78z>jc%F4S>RVgylOB_C|@gSQ~&sio8Nh zw5Dh zNJ3_s4Fi@T2R!7fuZ~ZAVU${cE}F*As;;nQ&Kw6GJuW41``= z@rDP2i@)gnT8FP+m*aK^+2uW_Zy`H^M3&cHfQ6~L(JepV%L1fy`4!bcFhvO0O@)b* zqaeQ&wVVEY0DH!LvOUukf1jnDiu-2ZEb*OfG4*rr3B74!7bT^Rik32Ww-T&ACG3lE z<0S9%bo<9m9}fCgTa_-TJI9|~hxx=Qeo&2kQ?Butr`wkHQTgvaIPNWTr}DZIlyY>` zWr4=r_6CnljBwhm)Hm|pd=z^%P+XaS1qn%iz2a$akV@W@%5wTSe_KkObcbteF=8ZE zQE3;WyS5odw5vK#An6ZT+_~E67NV5hNv{aZ@%Y#UKS`$sWOW76Iihqft&tgt{wUFh-jj-nFK8vE%&r!lX;^|Jx=^iQ3L0W9e9oJ8g zxek;Q!jD+#kb*^67jhjwOpZ1t$>|ms7|a%$LPNjzd6+< z@gc|Fge#Pmp_k>_+t^%BHnNWD+x{Cb08n>ft!ek-es}G+DkVBrhYX?!A@3k!ek?oB zW6q-!J(4#8^c6nIVF%OkM6dQ_XEQ<>n4&Jm+tl>1e?U6k&Abe>t(GT#z0~qHW@&2l z+Q8Nn{-U(Xr@Qs4B)$E5NuNJ?7$k(8uMzHPpM6%Ue@V)Ha5unHt~+VxT5AkHB_DwM zAr=V{W{Vz{_0HP#y!OKDjPPq|5VrIX7k;8w1!k~Mw!RyiyWu5LAH!KItnQz?A` zxyKxcLpl@^CMa<}&oEEPn>AYw=}FM94A8IpLtB;_K2{loeH{2)8FiIKK07rg($gyaWfrjx2`7nK zEm((ruOM(-XGrww=Z%X=<<)&WRQD>^fA)mGQLA%xcPhh4|`;hxh$Yw}oBgf+2(HLv~c$o43UlQsSePyPGP z<97cD+qDc7fLip+>nu*;ezwHC-Ob?p1lPauq@9edyM6bYIbY4y&l9c5=Nr$qe-`#; znZiC?3{t-Tkj+eyQc*%(QKCi>_QE**@i2okKbAA4Ju01fpSqV#NPy)B{n2Xk!Cn^E zBNEPjOLqao$#`s;g=xS1Q`$(rg+7ZD?d#W3^)dKJ6xcxO31{ne8+y^3Qr*=Py|o-C zmjha*&$?c#^1L}0l5>NjFGU2BfAo3WqbWgnooQ_)Kpol}qOx>mZYoiaWjE3AA&b5# z&Y$`+(nqmcX|WIG(6p_%kr0kK&GKVdY0={VC~bB*Lx@*Jiw6R?mCBa`k$%E&R58U| z2pj1n>*g#RM+Ft4me@S59$hFm@ zpctDM9$nDjMfHqw0TALwfF=86~fi}rtNo7LayRfyl^`rF<72<8y zrrQLTz*f(eIB!mqj%{L2Xpo0K->zJO0LwQ@#y!Z&hX$#Uyt4w&tm*9GN6jneoM??U z@pb+K8he9oAvMn)q&w*3=rL>`aFQ;ygl{bDY|6HgUGHqu&J-oG}dN!JBFxmHyDmI4NW&4j*T6e$Dc^fA*OyY1xt|NI?$Q)SWs!e)pF` zCh6&(Rm;_9Ib2g%fe18z$mnEE-8!1(9mzBND8<<5-y5HnZi_pbLd{KjkRkCg&Dh@r zIH|738M7_jT(0Wme`666QDyX9s1bVnsu|xQ9uvrw8oa?XDmKC*$eq-m8umHMK_?as zP|yBRvqej{&!LL5X19Eni21kLR`aBthU6OiB=pP<BQ0yh;uSiy2#oh-?eZ^~o0 zX)UD%(+-EnME#da%^G+ln&;JvL;Q4;QAp-1)!t4C_=h>qe@WV`4FJk#jz!by_8cX@ zAD-$>GAZqAhy}lin=xC&rV*Jj_1P=X^Jp%6p3$}9j_11P0_bd{li?@ABY7J&jTVGt z^c;?1{m9wD#BD=(rfL#}h%vY3f8KAek*EkUVp#_rxm5E@h@lgF7wldYGL)`_BJoTw zV6`J(EvJO$e}5K3WT1~;zEMFt6EI~+)T%FrU(`@~@vPdE{M!*P6D%q2n~(FgZ5}Fa z7DrY)c#mmQz5k+E`qZu>M3oOTq)Hr@t*OK5BV&;iRv zJ{LT`B9gB}TL$6Yw{Cg`j)I*iyCuY`=IqHbl1J96RIN(X~f3}$zJ2t%Giw7`E z+LT>wWPOTAkz#@h-+{+QL8?K+hzzS*UCoxZgt^seyeea4!L=o9Jd_A$SDU7B$J-U| zAe3&Cf5}WEIFhG9^X7$D<0}7!lB#h z1a-|(l4(eeqN$;YW5MRIB7o<+AJCrVfL}lt8*1U8^kfANzPLHlpCD5+t9x?6{A@KI^`ZXo4t(0}Ww4evv^*p9 z5+Dlgi?%9S1l~OE&*_N1aP5y6X0>@=ttVXJ0k#VZ{q%pBKs=I8~v^D2L;MOO+;6 zRwf%jwi8&1cn6&!c^!^~j%v|8vmw(;MTmF`igz)!CZ%JIo^0#pJTK8JWaV`K3v1fO zZSptWmG4<^zn{lrreFpN`01O$H>0RWDy9RHE6LAr+7b|TX%Q>sbQ)MyvKFvGe`Na% zr-27{f!vAb`q=V3NZu$=^6xv7HhuA+eX%=3Y$e@|X0A^g;cTC)N?vXXiq$8T3MDz+ z)kt#C;W)2IP->IV;muJHBDYeDJYih>y5`Y|0iX!JI91gm@*e|6K`=$ayy z#hFj}@5@c!vPY&=2SHE`Q zzts;ODN?w-wU6l8T~o@sX?hk2P2G>`_)O>4S~Y8dmonH%>wxZtig5;{e`qfyy!n%; zJ{vKciN*y-9>{=G6!o>}H_SGP$omiD(r~x&+IVwQ;;?;cr#Y2#vLWoZOlfT*joGY< zUr)1qyb+e(s$?nqQaS$7)!wVuxMKM!FE1IkR2{V#{51T;lc$6QiM-q&j?<30bsLa; z0=qp=z>eEbmLOxvhUUh}e^GMR$SB}xd2*zctZ9wl?1mRb#Ei4U=Bc)d00hH4B>P%xwEI$~2-9hZG#N zo{T=Z7epMudI#DdEI%P=IeE8-QJr5gw2PWt=&M@vXV4zhx)WuMqvT&QV!N!wqQe)M z*zFdbnDt?}vfMWJxo{Nqf=|_T)u0BJ6AO+fNnvmY3+IJP7YGf$+laJgIlGWvQ#QQf zRiO*jNOUgQ!_?0sf4#(d{pW1QMbc5KuG(lPiTVglkg0UwBy`S?7O*Bx<@)RfN@o5L zdxIbfma5xby$>j49y9Pzd3Sk6eF+cp1smmVr9`<4VU_2*UtSh4VwV8>6YIQx7KkU0 z6DzbLi7xy}v{NEEd(qE)GE^A--v`KcwNJDBILnx7Eex=Tf1MJp`T&q_P^>SfNs@QB zr$hq(k_0q0DXOpUptO&G=Ojq!>~5jk>3#h#<%QVx<20d1NHX^CGz3QFeUp%ilx3H< zdyivJJvQDJ!k0pDVvD}H0J0wo=*N?<_5L$HRAGGV4FHZ=rKWxVYfx**`z#RXPaG^0 zj?WN?-9P&Fe~n;humPzg{$zRUD;2KA;^yIh2DVQKI}4Bo7kx9*AFm#oD-QZ`7hb&i zApXA2wL3hDnpN_OdP!^*%GVt;B(~j#pB8?gVSaZz;bEGz;$FPf%S#S68Icc@Xmn-2 zXz+jC_3eFPozuC9-R_pa=G36=V?Qqh?i^)oiL9AKe+aF*;W+-w22Fe+4@p1aK=`x% zW@2+rf_q71rz0RK;A$T9cd4-(PlAso799JZ9MiV;*V3_g>5yddEh z_^Ye67D50~SUaY;f_@rColgA?j5Q{T&5|aQt+svpO>ljY9RBHHJ)Bts#@3kkzO0gn zzk1=`f7Fw)C*BkKIJ0IX*^&*jY=PXJ=GhnO3(5=U11&bMZyx2&STkX#Il}K3`;vTq zs8$$DHZTz$Sk?2w^};<&K|P}ZA%qSc7mqx;Slf8InwbfX{q&VGt0*GYv_a;|L&y%v zn-`r!N7Xp%!st$Aou5}Sa?x`Rvvik`64zwuf5LFtNJ*jxDSBX;wdnK`r}|fd48V3R z%y~~u2Pqwdc4UM<2zVoM?hUhqPXRXL>YK6m6gHAsNs`k}oyxxa@7*|5dC~HH9~4&0 zpsS0tz@w+)Mft+_Bn-_wG%xY=)JhjbZ|Wr;M>K0s!CO5lf>ztr&Kb(;ZQ2j|(#g(ZH4o_k4DrFr45jon_%b5?&8F={_gtIr zAN1U&(s?(}&(VtznO}O@^UYi4mGT86a=Eh~uJVOh1kkUWxV+%E9J5fz-v1Raf3W5Y zt})&H=nLTH24m)jh0*kaMDe}X6Sm21e~)+?_B-+n2XBtSZ1s7>LeEzR+>r?PE6VmU z5u-7oLk@no5$u`!PD`fI>vk!&Y3U7b#+++QIgWpfiY;<^FD^8wB`fsxgdEWi2^Eau zrRUS31l*hx`~e?(Fzb0TZg0~Rf6N+MU7+cGucghy(Ah{E@?KLUw>PtPbs3w8!Ukj` z@DjP2r4~0p^13kMe6}rR+G#BY&$eD>oxYFFt66}6cWIev45Qu%@4cg2DBUw;`+Js0 zJMqPA_b%FO>yXe+Uh@%6R-Jmf!*sqYU{oVfL$qKeK=?N`lKXnG()&F=e{sg1QBY3+`MeA5@^r1zPA%3J*= z+M~B>Gi=Wtl`T2ULhG=fMgMqIC4qr!N)O{zuTC8+t-Z+%Ig>5X)|_R+1WTg5Nix+X zzP%utZBN|DdPpn!2)ccxfA?9U$9^woLwbnrR1I9V1y_!cJr#n-)#~&L>&ns1X4exE z0|k!w*o~cYPC7VKI`!oaoCV?UB~P3t7*Mm8S@yv)ood0ceydH7@mfjDO^z5n28Vci zo5YJ>i zn(%F$@{_T>aQdNm{hx#bdx?onG*<5&O!i-QsJdnF^1ASFs#s=3^7|zCH)9eHsuyp$ zX-JaYWA8b@>dyG55)_um9;bYY(CLPM3#=$@Man*#s}`6_C!_cZ&{Xa3+cHbbqj}$IW9eFFE?W#j85*Hnw8$Wzb2#8KSFTaKY zcYOwq&P3Ao^=UOjYXVH^EXv*Nw80}1BUswMoc1PtAtfr%G&7+4_oK;m_*~$KZ6&yh z%u=}_FC7aPB@$90%b+T34wzog=bJhLDbCu^!mVw;G+NSe-m8h1(T?ns<5YOx_O`(f zXgzKA#a9W`r*xl!@A9VD(5T3|-eO$HtZuV9lNk~mgJ`2#{m)LmdMGC`xN+)}k_R`{ zHh2qjbkOvOp8*QQsJtEU@P%B82FFa+AsT_nuyomNS{on+)gnzkU}t@dNgRAnKk9B8 zy}Rr(3vqg9&YV5SY?!ZpxBZh%!^Q5Zf1#Eyy`|5${hVwY{N=)fGp|SYC$i&lH8Gc4faSF++Bb8)fTGL@6wF}qx@r)fi~jA8!r>~C}x|9mc75>K_$w^Wczo_93FztJ-)t7u*YSy0DfJXHu;=v!adBkdt=s5 zk!p)BM$CMped?kw<=qpU^uG$t}3da}i6N9eHCse7}4( zbNe^AfWvU1sDC@^?O2i|QJ`_7k=mX#-)%g@r{hl?KTy+kUW-O|yM9}S`lD(l>4>LY3n6hKRrWe8{3_Cly`29$d9l1_V1}{@_Q6ccZm_wH@EpW z#$whPf^YdEP0yTwv?LDPY=ErJhQ)j9wEffPMZqb<(tVe%H+`45y(N!)z}F|36Y_lZ zqhIa}=#lybL5jdr`PzD+RFA?XDF2rj3h2^(o2yq-iYb*=eo6D=fDqe`+PdI*Kl#xR zI4$iv|Gq#XsruP4o4Lj#`5lLE0#(;qfam5Uf&BnND8{1EfaQ7bDisVLNFvx1dbtpa z_!L!$#=gM#!U#kC@2I2!?wvES)aKNz`t^j4>ue(a^1r9zV&GMmL!frlxvALFJLk=D~A_9>5tAfq_377QTk25VK4w7-FW?JrRY2^+%%moG< z5G`;TmGKX<=x@T9`WRnu{dS?v!kVs$zOv%Mv~Ws1m3eKB|K+2BkzZhYV8*q*q-b29 zWt?=zDW3>)OwB~=IY$o6lGj)@1zM~q7F}j$2^gf(vD@WBQyRT!Bjl!MoK5*2#k6Y% z2z$Q%<1sqv?woIBQ^wssTY-)6B1A(JhDj{(G&5~nxDRkR{LFMS4D34A5@5Oi42 z=9v(+##YDSm-;+%(Y#%riT&_DrRy!_EKf!?)Y+%Tb_LYEpmy83CTA+;8%YKHU5c%u zZ|_$E=Vtg@u=^8&JII%_OQ9(PaL{5pmu%D8AvnSW!(x%V8SCh0E0+AW9gAO=yap6x-+BM>?UZUi5j zoIm3DpS6|G@PGpd=Ux`$5=7S;Y<%!(&(XyBp6UfeWLkfodOx(4yeB3+Xgm^LM6m2~ z@x@}EH0Y|OA$H8n1xWs@(92ado8{^_9OKXMs&b9ssJ2COF-J@8!kRRatT7xLo|$l8 zrn-{B+c930_V|@JqgQlgEbzm=Zg~k-{3eH{XEz|%Pn(yxcnt^l0Z{tvr-Z!DZm!P*H}qQ=*8BkhY4z{%N8=`bV$uCOGO(61bU^c?^O!R;5B@9a z1jO?(wzkNZUz)25r=hRJnT4W?Ga5Z!?k!IR57Nz|FD(n3VImvcnCLU%e24Kgya;|7 zD?^*Sf-FE>ko%q1MOJITM^n4bXi~R5zrpCyRH=mX!Jm=r-OdACOjTrd>SW#S_IK1O za;GRIQ8v_&d_c9%KC1K-28vHfFwgX3r~Xprxb5Sj^g9T7&}PjeVCxXyk0dQ$6)RI( zogQN5&Mby#VXE~AOoc4{ZM86fnK-xT)PX$MC&aXYOK#rf?Lrw^Hw4!5x#1iXqIjh* z0=_K`+~p0{bFYS4V@<!(41m4xpY7QHo30K>@Ss^ zdg$MAU8tHuMwWl}8ftx?thf? zk8y{Y)&e9&i$Be&KR91Xb?WjY*mFzMolv_DwSmYW4vl=-+RQG@&eV2Nk{BX+yQ;M; zLXyA``m|OKECQ7Fk~z%Ad=GH*04Ts-)t1)Pf+oKmAqpeQf7A zZg?5a>?9{{{zytld|2M%%w}$pOHCshC^{%iJ5zVwS`dS47Os0SDt^`H#gw0K4?)+J z?m)MtaW+JJhDB0czjfI+!oF+81CJN+=24uA3O)#Wo}~TTBN_O{>t*OSe6(dHYe zcy4R#O}>F77u?BmC`nANu8y^IeD&%&;1rP4x!HX@*kEJiI!;(bz1@b#U(Ojb{|N~6 z-5rn6o?y?D&6{8!+hsY|j?_ECl}BG>jOjStQI`(;OBz=o{3eGje`&Y69Ot=zu zJ`;N;l*MP~KR$uE#`Z?+j{owLc97JfqMXEWXN`k>YevoG@R_!JTu%5}wK-m*IOP3( zcjW#$$W8KUSk+(&j2k~Ue%cQ-bT89vbIuUEdmAp_oOSw!OxWz`*|S3DItFm&AIenF zxER=rFnz;7ARbE)WQ;pYyw>NrNhcghr9%@o#3|~wS{pCXbXa%ft4ys>`|$~#4TiqW z8dRiwysH_0EQGEKv z_Nb6?isu<=yyddNxPI9sIvI#1;vk&hF$*lHZv3%m_iMFISKy;we`z0tf3~C`wWBWB z?1id7+su3-z%u z>btx3eu0zA z)rA)^ehdb0OCzmEi}rjYpCqHxG&#}+Q$il4M>J~e(`44w5$qnNsB7Y~&+z??q~$-= z`X|{pUjh9Uj>7^{#a=)rw~?01XVK(ZlZ+1H$owt=b#*#!+cO z#E=LSI}x&ahbJ7-CGe=A*sL+`kcgFHc@kmJ`!V*arXw_O^n9t{c~y{*h4TsL@EO%% za=LqOj>g_4yZ6p`)jCop{DInTwP_gt*y!%+S+G;3S=rf(Kpyz|^o6~sZrAGPv5Mw0 zR{YcooPp9x6b0|_4@URWh4CMFA0KR4@op;&_3?wgFEZzt3XF7lO7g$&$95mA++Op*8xYd4P z;htd>R0F-{x-kJW@%iAyb<^6tNNZLRd3YGx*+bp>S(8>z>a-F0#r1^jf2k|a89s=? zqg&Mzb)0xQJh#8u8PROmcuYMK$VEA3>7VR5k*<*$y+mq03J{iGZH3vI10Twg)%H5= zM5H42%C!0Ih2QmW_l*xa$*gu}2PKo`h$0F{n;bbCFcyG1T&M1YmAcp}7d6FJ+lP@N|4%2!FYalk`M-}wU{$c2@w^;||a_39Pe`iIArztmu+AN7e7ojt(} zJMi>s(>_qWqLiYH>yJDraw9ODu64lCd5@U3zX=#_&-E4>trWiX3ioI&7pZM4?r;Ur zkE&_^X=Vl#1DT(v;BF~ppTb(!w;udSqNgU04TaPQ&CD=c8Hjt0W9pvThrFD|jq&q~ zT&S7mNfqr+)xNqJpM_MjVSO45bnk8cyh_AUKFiqq9@qZ#GNt3oHyE)J6xy~?kX-ei z`djK#3K8Vqv~C|=>hy{tbw+*uJr6cz*ZOK}Of5Y?J|pwPADlj|e*jiQmO-$|$C+V; zjKj&nkq!Vzly=l(x_8XXBLlWG-_JD&Pk%i{%CDr4|&^zmY4Ti zaq`q(s=~<1=4RM51$Nutde?8oZd);qnd;eG;}Rax&7eI*2CE*6%65P~AQshjJqo4W zv7fYFiPZlnjLih;xd%&@&KBbWFNjg1-$ME2gt?^Hz2rUp`{YTZrH&(N1QYdo1t|vh=y02o+k&JS8M@pjfn$_eK%YUV< zl#BOWJXQ}%bFKml<8;4n ziMlJ%aItv?Ncf}tta5GrC#s!o!ZC|BU&nmGSBcR=&cca{UWwG&3-edaqa@_b747qIS2G^I@?Xu5SWuOKv!po zWPW~L&Xo=g-!5sryBGYNxxyzH_HN3HKbzfK)zXc=DlfMjvFowJkXzsuwsCyRU}&Om z1;1o%IiT$BNs@@-`xJj%zwrkMo(#XHgI!iN$+ubzBO26;hatTzI4~h2Zv7_ z$@4twS}1gIkEUE($eb#{_xF>sAN9BPyr|2zo0^}8%#u2*`;lSlrsI(?nSpnc)^$tm z`C1&>^DqdsqzyY%Zat0KtaT-mM?0jmQ~WVL(MYdokwHB^mdZ|d6=RUVxodW`)d?34 zx>!P9;$itgU6ebnKQaj<%SBI*9alOeGCxv(uR%7h71FWpCZj7Se|=Zix0KHx1&aG> zuJ&XV_D699A$Z+&-#f;Mvaa!Ehl-5eH!VgBYv0etE5>~o(Kso!?jAq0k%Aeg8?DO- zyDU3Q-yG||y6b8Tp*ghxrBc&9Lk=Ul!g4I|_)>TiWYoRYyAT1op|E0)MPjA2v8i`j-~2 zhinAHs#i=C%pBc63AU@*?2gk_vK`lJd-Hh~rr?z#dXD#h|*cXx$ zPV0~VvV(U@d(h@S?0Qp)P1Pci&<>bm%0#`a8cO-4*~C+F5Js(fUm`k6J=peS!quoo z5;%yW8do@ql$ObSynQ`)*B3?*GT~@xJ%i={HF9;N3WlTU(Ww6Jnz-lDx|*6%K7~{E z20f9U$l{hpSaB>-7v|E7r)WVcZ8TH zFG&6o>tbJWTb6j+n{TG$y5hTBSs;^v%SmbBwy=<6<%`wTo)fNfqKwayrMoL4YJNMV-jsto#-M3D7+o68=d|g(2Fz+6ngFZ7JK1-i zH?n!&=elwBRku23UXW0kw^QYv-!G%Nc*&W?N|7fv*4O>0xPWulqC8~J(EV#KM+e^h%RNOY1y;RU9s;5j8$JyIuXGF?-$n6wdv9eGaqF$<)tBn@oUN4+H zt84iC^n9>dZ>TIu>nhGKAgT6kFg#!t5r^tU0Da64kb$AZIL}x%D1CJle@%3Ed5T|G% zw+np-qFT?}xV-1YLaeWwYpKF1`HioOUlo^KL#fxAkD0dFFfRC9lW;{02%TH7Xpl)` zb(<)UxZ5StK_}~II|-Czvgub_9Q&nYRc@SEG+368!6yPUkQBNN?>!%avO7_38KZP8 zn{e5>Cxw1{tSX@2wA`YBOxFvb|`&$)0K9fe)NF0Xiqx970ui$>geah-r@2=K@Hl06}u3> z&cbh)?0}dUlczkz!tZRH1lzIg2xzPY2t|&Oo`mWdDeImIy<#UKNl@$PCVan!>Ay_^ zezlmpxuNyMNhS**yShhpkGeIvMY^Hqgzmo00%Wg$?XVD+?@on~n#1{ILr$D@8Pi5+ zp@_lKbdr%bOWtsiLp5V1g^abwujBJo=yGxTt=I<;`wFcjhq8A2>F`{bbYL0TQ3i_d zX&xwec+&|#1tBWjG{H~n1<89mabPA>8H1zx-SU-ruS2r0}qQ2Sm8*-^s-iT zG~E+rawoL~Fvf)=ghfRjRnB;1OyT229SG4ne#IWaoWqir$nvyf|KKze2<>WfT08p3 ze>l;aVs0HcU0(+#$Wjc@?xnaPNZ0XSlZ>w!Ff1N3oK8()TcPJ8d z3y@YJxCRcm89OD1AX?{th(%R&9Yid?IJP|HRas{&?bxx_#yGzkrQl|DPW-{Ev(dbO zlmm`;3wB)`dcvXYM7PVhwXGcl15z{pOq@!W@`YcY8OMb zq#`4|)y_!vWkFW!95=VIk>A)AFSsUdJsj#@eag)aOHA3K(QE>M6e$u{KV&^F)TGx?@Vy0pmw2uQe0C-+_a*M zbTs4iDTvebYD<{{srojbc`b`GpFR8ATcHL>81$JYyh zKh34uJxRyMxHl63q^*XqFfzf%ZebP-6zZ{CkT={bPjOATtbh&-S_XG;pslTMM)xA~qw(@5Ql1SoT9;hRM35;6}2pX{)M8TLZkL z4qN5yH~m!SuPexUUHzl6u`Wh#&LwshU$ZcqT2AYf@yYEz zJWm-n@6m!;x#$;%8QE-XxWBCh_F6ToioLsked38T2r|RvkCGd)tXFR`>Tha_fFE&t z8*T3C?}C{`fP+lo7R4aNPHlS0%!af-Px+h;Y2PCN{S=H#G5ZPkEM@1xd|DcIpKScw z2-qXygp=?^9Uw?ih{+s%^o;AW?;E0L0Vpe!*S#EMul%xQ4;0U!31L{4E;W@_K4x$C zJt$8~Uw;@O>z8hyURqy!wQZ>GPtd%$IS+9UB?t<}tZYE*vD~<)W?Qs}3U?2sDPkwfs5~Vt3xloG#B1~U)?7Yj+b&v2Q?TsZl{5!a7wBJzlw`%lx6(~ z+Bdt4_SE@ox+V#6#dmE_<2re}%cqWw+I0ZDKfh6z<4t^cKGVJzRn|BHOx$;+62UVx zm4SV;if}GxuY3wNz*lBi4m{JD<72-16Q<}d;YG&>M-QO##+6fGE3J2C2|`=dJk0|< zz668Tcs(E}vba?B)0*pwy{^LN^=|oK<4+IJX+?vRa9*!LLsq;OEX0r^}H>$?|N| z+ZRdq?I<`q1N~>Y%om4*zTNKhz^dB3@df#jyHs9jP83(2sP}V^d&c*4>ikAUx9j-Q z7K#QQPLuZDeHa13M-P_1U$l$?z}C}xu-DWAv{@5f5NEqrjn(?bO~ul*!9m8dMF%w) z%VmA%UOhKn<^v&&{UIZ~i3NYQK1H|<8A#b^2C+CWy|xm;8G zy~JI+RdpjY+QH%u(Jqzc$gB+=EaoG6BgtvEbj4HHTtj2?Nm2Ky5`nJ@I8WBSo72C0 z5WDHL?V-ib3Kz6ARj5$o%v73upTsAt(k%U#Fmk_OCgzCl1#ZRT=Z?z>KBv5DWp55m9 z?`g6fI*#w{mQ8`SVR0KW^R9|72CWXYYgcCNI(%>055$gpS535btRLDe`-DSQA)oYfVRa9J4~`zQWfTJ!sdyU5|r}84qu(oS= zzSj-4vsnV>llLQls8nFBKJLoL`Pgf|G3OuLlIqKzCEJ}772io>srjj-emOYbu88w9 zs#+)v?}hWyhfXO#TWX=cr-R|nI>Lr#XphCRwqGHmWR+Ge6tZ>t2tuf8P{I20Z}D*Y zIVsFci&OZ-Evw)F<4%{-a?Ql>>il!B%-C1_evbEc?sO$Xmd{~IP^CAW&Y})9!Ue4U zYZ-amZW|Z64BRk5H`y;UEgm;0H+Igr{@pWk6K0vJA{Y=#vd}P?5D*aX5K+Orx(7Fg ztAA#oA;jWfA($WtAnYAn&5Ru!Y+M+v?2IhTk}rL9DBuotysSPVl*m`-BXmDPuZhT4 zzjcQ6{~)}Ca=x;|z(7cOz(TP6F9=j+1FHXv;8!*X)g&teAdJxPLZ+thLgq`M|1cnShH~r?pdg~- zKN=kVn~|i-iJRPcz>M&Jmot9A96El$(lma+8iw#6ME-M6iWW8`1hgF_1jTEsvK^ZIe;h)n4kzJ95Pbx3 zlMd%MCX4?pfhLZc>>C9Xgj)#|1oeM&vhge0p$YycCY_EwG@Xv4DV>g!hvGko{c)8o zHa{qcv3V#6#{UH&)YuM9@;?#hKTJZ;f0#C%|1eYV`ws&7ANiDTGbF@G7!(BU|AP4U zfgt*yh_qR=rnFh})jGaP4(El&#a!u{f-~MNMe6&SVe6*#l!GF>v zK$jt%KcOHBpkN@l{yROH^~{pt|J#P-sTC)TP|%s>M_4t1&aC@?{%1edn)z1lY(7D> z8^b~H{WrrUTdANOngRI#Pbn)mPb3IPm|I8)i2qz7r>|~mIx8!TG1%IO!BBQJ+w>|C zI389`Nayq0=cvNVD%z`S^v!KBRIm|+%@_oM1~?QU+w_K0uSId(8Jf0@5ji`4j#3jd!nv}q&TJc z#5m8K?QKgVI}1N>)@c)rogCq<-?rTAqNP(g<@WZ~ij1OhR``fX&8Mu%QNtnF(ss~` zzf^IdrIum=&*z0WDK|#P{Su`V1cFcm&>!)lh$UrtAH3!*5VGsqakG5a&-`X>qWX<4 zph>Z4jD=*(1$;1r0-G4inQv25lmCbsU|GI>zm^2z>~#Npb#;&ul}pIU9O^R|ev0mm5N7Gvohgta%^Wm za-#h?$da2(kioaDn~wPMwhnci=8g9P>bJDdsqCu5GOL%t@=vc&DFAlDCuq~y7fAD} zeG-`yt)*7i3g67Rj$|A~W6?*kd-c=_@p?F~n0l^dj^yGIu==Xd;;j3-l21i5eLv&z zL5=)3d~V|m^VMJKH=3n`pnAwLwBD-n#cX1JUlp2~y6`bWG9`ufU`wX%0sMu}3?PPq}B{|m^b%M(oK z7c&^=08xo@Zqh!cBg3`Po(z#*78txhuK&J=?~iX&lu3tyZs{NldLaw%sLs5sKA(8w zFwQNcA$*WWu>Om+0S{cT0E$YHSqDeIH>E(AR8Wk_o)BO#e~WpCy-2Y2ZhJNU9vn=5m*yyXfU|s5^kBY(kD5S{ zi_nyXZ!t*H4r9?540;){?YeLp4`M|zK()TDMK!KsKJ`Fm4E1+DwhT5_i(Q{pny|O; zKlMCSlVY0tx=Nuc`x{GES9>#CA8fWJm7%A;MPiVGL(mkqve~yPfDHy#^TN-|Hp;(y_8F6FC z+%RHX=aojh=l8Y!m_2l!K$e-VYb#UpSi`Gd+w5xaip2JOqo<+@!(M)g++5GH`S)_> zFcHrf`A|l5*%^0QhdcY{zd8 zDvv05aIoDn^m9I324wt*9hvmtEhuBUJ8y7*d~d~hRQ63RI1(2B_Q>)(AZ&|e;Lg{x zXfLR#R#S|-Uc-Pm)h9iNu!n@pCz5DK>|W_SQ@cKW*J3HjFGs9)G0m#W;x zhXc>?IXeo|H%tRWdxV@@_ZsySg6zh)C(AJg139|zYR=humJ>+(#D+naTKhCKifR27#yIrr zFWki=!lOsy#T{+HoVoDJa>&(`_G-$1BqYpZUbk%231WrqK2L?2QpI zluZ%!vW0+bG&}UJ<)Bu%*8bs*eiCp4e~n}}f_s4$K5D@&9;3PmyW)f2=qe6B zZtQNPuKz{IMIl4`5NwUdO)z2Q+e%kT876Mc+|vO97Y1@RLr)?RHW9KHn%QlmM&?)0 zFE=CS=d)>?Nr)fKnW-*Z@VTI0c?S2i{sBvq^M#R$$%dpq#@ z!XNixVvL!>9{}!a=U9JV^hkW%aktHcn~8;y;%Twj{A3dj*<_-9rWGWSO}br>&%2Sz zeBReWocT1%LM6*Lvp(}lc65Dx^|$g2%}~ifJwRij0l8?kkwBoz7;Is{mB;KbDJ3{) zWnja}@1RCs<4`@(GFU&sk?pbXc!%ob-3TZ~iqq1){?H~ZwIT)8&FnR(#y>Z%{CqZ1 z#Kz&J6bxX-sLRX7ua65m8N1DVcx{zloIY`$%V<}!%7V}KZL9dcfXG>GS^DSCob;2W zSeDbZ9BmS$!%U#0uuXp3*466z-0ns0bm{#3Y+9epe(6*Qpn2)ojF2wA|A0C+mA+S2AZekHC>)#T#DD{2{uMH!ZivLcu`Lm4pdE3KxBGFPN`quWV`5VQts~+HNDP`FG zxoVM&>9|h&K6)3cWLu~(ov?_%=s@}N7Oz)gK)mN2tQK6BM9ojGj-M?8GbR-#c^jNK zqsb(PF%Id;ZmRsZ?YDe7d{rWfEX1OHqJ7c&$DEK1Db0tv@|@^qf+Kfs;0eFB=OLTGc@r@ih**n-lSZ(xK8T zb9C(cno}zZ)%GP+?Ix>lCIhMq(Dz)6Lo!1gcly7BGzP_tl@rR_p-%vZ4w{2x!{3E8 z&r^2PI$yqBOC+F2(yAWWh{-$e^jn){b!}bRTi6nIbo3hM2F29((DI`#m_K}s%pXf! zNc?hd&asxXb&FmZXgAMjG>TWflvA~i)|ZVenHINp;IFY$d6Y*I}YwOSu71RNaWAf;>uH9@wU&0T?z;BQGg3ECNNz>Ij;K`Gl!i$c^t|k?iInsgMqUBu9o-e| zWthszn7K{^>1k#>C6KBBS64WlNJzI{lKO0=yMKJ?hpNv!ND_`ovd{fzgA}3&RF@AC=9M9N)VZU+!};F{6hxa^*QyQ%Jj;k z9|uvb=>7{xGVQISL;RG=N7XmYG7O!_Z0R%lWdFrkWO=QadvF zsWdmL6(ZmWuafb%Szl<|5#`YfzeS*rXhUHu>+|=uga~5~DKP z)w>&A^xf0?Ygfz^cnzth^L{h?+8+eGi@#sLf+wD949<;mW3qDe^_LFc-EK+WPM_c1 zUM?>6>Y9$fI6D_Bx~by0xWH-9N;YZE->(*9!~z+nZhGJ~(!wrK&QQtJt3*Ru&HSX( zjIs(Qn_Iobv*Es_@%X+O_-5guS3j^=6y(x`#@K#(MxYDZA!hm3D)QOQ$0IFszuzFa zc`0J@kgB9T&e4?_=oz^SIYWu2x{JhN{HZre2Qx=j@NO0JL?L3bx4%#JZ~Xk|XlQ?^ zAJ{)M9vc7t?OXR-NMdGUB9AJ>w6`%W~}b4y&S1XH{5+B@EJ7xMby%eC;jrX`GJ!|1}RCv5}5 zoKP|ebG~rAMyh)!2A%{t<{me?T%0dEfk&{s+gooL zQ=FXTD$EZX+R+zxFPhV?Fy%vY<8>;s8+5422-W}USUHgc87fro`iJBWq2${ct(2y2 zv6B*+DG4H86T4~Yov|i_7hOd&a6{l9efN#f*G9}yTVorRn@ zg(ICtWd2(d8R_=bN9C%VPS(trTNTlZxqaJ=XvBB4nHVAajp%xa(}~4ZXw-pP(!{mf z(oUc+h;sXWlBl?0T@+gW+okD@2X4}bXmLb(!xUtHG8{7LEsVky=CXhaa383@!{>y! z=hLC-gBERs;PUICpCzZ`__uc)4iNwe$75IrJ$v(4{%wU}gSBW*F$X{i34i;xX8Jm( zS~U6fis7Wo;&b#HL}b@DrnAQ)vP*@5(3qyfT zcD5i{tq|PZ&Gn^RWv+@d!@=f3x9=%^6g-_ydV|I% z=W%?)zeEGwPi)@-m?lLgEZJZjF_W;xoem)n39_sJqI#{810+Gydbd10nj5UYM224} zOvJ}PY#iBm;*Fn)`oF~5LcikCqWyd81FCIL+ z^WKH8!&!>6+Pm_(AQGl1!oAELu2C%d#wzK?CBL?I1*)Lg zrow*%c#e79{XBv&WNv!AI&!HbKXkzaIrLIKs)R~wg;q-AoKUyEGR0Yew+eEg`WM)R zipzq3_5(U;`yL)qh}I%nG~ew^ZZgZdp;q4`LBc;7m00>TA1>R^dJM?Q^#}7TJO849 z@hqD2m^EM5_}NS_Q8qO&66lI;X_O=U~(g>W{X3SR2?c;czR5VZ;Tk-3sy&5LGP@6LEs-VsWn6JH*V`5`i+ zZe^1NwF#A2>f~f^qs4I5)oRwDCq*8deM&V8118s)d;sc>-irzmwCd1 z`&Pc7=s?y0ni+oEA=+IB$LP7od1HTLaw<;8cDSwkH6%kK zMP|EEg<-3ZV6JCAv{#8q=H_X_+NmD56|jKa-;7?ftl~cc(7q+-(+F(>aK6VtFp(r` z3MFo58VY!krC#sQMWwjy%M~|ut3W2l`pxgI0tn5S1(SRpj|`|Sk2RKhZBWa_xVOB_Y+Z%enfwh78GX6mBEh)tYm<2+ljv`ORW?mDHHMg1 z((=bSVuo2hf6^wF|GE^lXpa;FIhYKoQiuU_`|c+-FJE(lC=k+*5Y-pC4t=2N?uG=e z^8CfM@>ZGfis=eSl+z^jja-^YuuZB-@6IEFFDbq zZbag(&G2+l-|Z^c!PlutPmd$oIc2=uJ9}bO^tQcmH9TEK7Z|~Qvvgdt1V8=z^DC)}zo$BE z`5@`kU@SItE>!z_b^{i`-`$p;M>kSU2JNJy9m{B;t%W?L?Mhhk3kLtKxK`zRSjja5m$M}2!#%)bRqW8bIkT2&j|i3l2r(2z!7mc_qS*aU z((-FU3&{-7u+_`~pNg}2{s_ubk;PVYvJe;4Imj!D{J75V!u3*5zbVHl(I&GGmRSR*WN@Ty<&FEJDhz*p>QY#&WYW>*keB z(xGGfel6uMq|dHLss3RDxQ*WVLZ(g#2pQF?;!FSXW8)NnicG4yD|p39wQl$n8Q%M> zWzp0!?eZ{zsxpa}{4+Ik2!B^{jV$;WmjW!lPsCK zOZ7fQrpaWW^Xj5~j6^2o6vjJ)7s0M_jb%la$W80~hL~_{V0*AxIr{Xy*3Bl)I);e_ zr!tBxG)dw24`|vNv<|Q4gSCnwLu!}rj zwxFyVm2)+iy5dH07bUm*eP5I(!li`GA%Q*n)4#6vh5)%{1m+tV zAF6%+Zn#!)N=O1Y4-Oc5G+4GvYZ(cazA$oPJU!3dKTnwJQ8dx## zZT5Y#lR|^}gh@(A<^mJV|XWLIj5q>|!sHgLi!{lQYsI zFx4l&Qb{5a^2X_C9S`@S_I|@L2(2D*t=-TAl65e+K963zo@z4>i%JB`C+RcCX176Y zKG{kYlv+8^EKAx6D~`v4NFfsnZ;k#~{qh_dM>BJ_YN#J(+x>*?i?+4S;D51)L>F*R z)#h&5rDBs1J>))F-Zl-lQc!)^Bt>*Vk{OCb`j7b11XHO<9NA9(X}uaj!(4)qt;xd@ z*x`7!zxEojTJN35x?}I!`KnE`{MOb%hHdtMux>|+tNC)Hsb%2wIwLNg{Up(#(R`(a z{>*X{*ZHEe=&^+nTB(IZeoD60*@cc|9yjCLO>dP&RUYO=u78K@F;YezAwy_Nn$F=y zOY4;0qugWJWlLN2^V4u*I9_-D9VoUQAS7eH!@n()0pVyJi=8|u_NkfnKv+ftgC1#rDM%huj{dNQz2q5B~wu@JOBx{yrE`xN=qcQ-w&pleHXsY>(=fsE!k%6Yl3C(D+-s^y^qoF-^p zj_mOqTxwABHHJe4(%LAKq*sQGL73JF{XYO%K&HQ}y0xhpXlJtI((b$htFeQi>Ifd&sQmo z)N#$4=LqORUlXX8mQP}%pDfinojovox{rS>LyA!dYT!`{vXp1~$U8K>M9Mq>SfhdN zV8U|H)@|l=;bNG}vzu`b5U9tM%FC*5sN4wDiQge$LAXc3$!#OS^a5ee;<8GZK*g2CR)s@%n!% z%8rb;4y=9J_PaeaP~8Q5Cwh=*-(P@NyweaxT&qItG%9m(al!lg7s8T3&v}pzK%O?1 zLEDr(4rbYJuWu^NhLYd9EJZ6h^d~7Ct+y;~l_dE$tK!}I7*G`%>*7(0zS`vtV8+Cw ziWvtX;|9l#UkTJ$#&Di5SYPk1mQsHrkjHSw6hYf+Gj8el664a3rfKyxw@Jq4{R0uP zeA;-7+9g$*tbP6n!%bkd`~kQ$n~=#X&act5(r&KC(4PyLmzcPTKNVYAIk0RUusv2C@TyiX^!3DUtY=57Q4bR*ho`kjoXUryATRE%`%om)A6jtDTS?`ak5Tt!x) z0~$9VjJUS(vF`f*(ZS3OO<;md-?}dbK=+5Rh>cL#i2ZFaW+{GZpXe62H ziy!Ymn%Fd-S-Mss%JnRGADe$IwzwbtlECjT$H2K-V7xi5Fup~`PGb?*d0jha?Jp6< zoHS1H<7YhN)t@S}Xms?1vn3h%Fse^@)Ak+u+(CxH$iE;%SPLPQQ>qX=pvK|G$GZsj z37Mr1$B%CFK6(yego+29i=d!|*6mG1OnxM_7~B3_+ly%oj@68M<1&A6VO`+3XYV z<#iv=)x4wYEQPB!8=KGdH^53C=!H=Mq_mHqtw{Dtzty2#V0EDft|`Lihk|zTFZ^hD zXZZG4u>jWLYw5L=%&32q)!`==ph16#=C1@U96SGKU>x$^Q^$?-xZ6~a-)X`~K*Tf0 z>8?wfzsniV(M?~CoGPj&t%RQV<2zrS1aP{anFx67dGM)o}fj3njwGRZ{&X`ohbA_pm?m5CZ$hZ+A|48|Ft zw`PmMH3nLS3%fO&aePz&h1*(bY>7ewvPFFMKKt?de)AS3`!%*k$^B8MgG#{SK^0rl zD|cqnEx!VgqoIGS5*owJbbJlLReEm7o!>1?}iW1a0ZprZXuuXOH| z=O6zFCFSJ;K!gQBhn$? z`yuulBLY38M`Lgam`3+15y64#AV9`5L6-S7$rq_`{PuNJtY!!YgHpcV2M{m-1aH%Q z3KPr*GZ#0mApP?qJ}7)sZ@b&NSo&)2#`+vHK;X4R2tFSq=Vp-Io03CR8}jv5HE``P zbgE2`9*2K~E!t)>qGKz;?bdgRVNEnq1f`BfR%yTQG%zc2s&4kV+~JBvlq~rA(mdjj z(sAV@MBT2F(LVJ85kWt3e)Z!K0Yt1)uaP%Xn+;m}7#-Is>(*$Jt?M9f*7*Ej>pmbu zZH~|##-{X(Ba!TyANWz{ZQ;c4)Y7i*>K~Bizw>{&pumyW&m8F;?gI{=n_!=8qS9B{ z5c|I?UWPA{1LOSpe!Uru5~GeEAUcTAckEyLAasbx;`HvN;VPS$!FpgF{gX+nQ*K&; zsH6KarLOS~k_Q>Q89Hg)e@1=QKirF@GRb84m{E#$*{z(zev&!>NpG&OMDI!5GAl?A z*@b`JUNftQA5W>K?M%OB*9~7z(DXRC)D<wRpB-@ufnG|umB`;1O)lLYml)q3LrN9#fu};cA93TF-Wp(}tLYW>`&T}? zt0(*uB^jS6c3Nj~%#Ec(wB?b00tG$=Y9n~z_2p3H`_BmtvWK~ia{VSGxA%*QoCXqk z;<$^o`z<>YJ>FoqnAzrS<@OFlCaI;i9&x4$2a(J z!QsWscFcUEVeH;`_bpJ!=*?YzIhHSXbaM_>2qR{ti{cm#*4M3iA5_MP#Bgf&InB2k z&2>-`c1IzSac68Dl=Av_7CR4*sVx_hugAd#H~Q{(w1CM=;@cJE37e%h*_K}GgXeeW zUKMB-CI-fx8h9^w0bp*z4)lNNTN-s&u0qs1CfeDa@Fka>2}bYXUto&&_I1+FlZ+#xULZi0Oc`@zQ8`242_$0L_p8jTgGkVGr} z98zpN1Y3EA3WYP!j~F-`jZjn^|InZ1a`F3vso4V8=7e{Fd_uHAvkQL%V*aHbL-ZIr zK>ZatZ4hYy#1=WT7R%shq+J+{atu>tM)dAGQ*ktpu0_~mkpQH(+7~=h7HzUTTG7ji z@yq}N$d@FgHB_Uvrldh-}$XT$pER}dC6v*bnx%uSC+O&(uk zsv$Vsy?fg>YHm9}SYKKh$^~4SFQu8UCGOW)+)S&r{K#-QH^hGsnZFbIM3L&zxU!g> zt*Pwgo`d@P=S<96WzkY$#Bv)02^&kk>`T7*$@VC|&$dJ&q|gB?nCVou2C5E2nR3!f zMsy*O6rMo#cvk;Xjo03-Ec%FYzvdO+X$-Dw7Yrti<$a9{)^|s( zU7 z;uV(R-Z3Y9{mE1wkj%xgqka&@INLKPmv&ZXb?>wA!}fvXsq^dek`aH1??PYiITO$* z1zPHIWlyq?($&iXz!#XLV)b{<2=w=JYHzrfw z<#@7pN@st*ik?kusU~v#(3gK(-Ylug_F-gc;mFeE89n(rcw?ng8DZ zJV4iPH7d?!Ims9(U5?6{7TRbJS9`*IxrC`Hf%R zSvYLP=Ujd%eHI-croDv4m%#N7aGYeHs07`MJoZ#@0%hr;1EyRHnmLp#M8(FS6WT5x88pOQ(w9e}vK7G`1D3Sh9%IyI|fY?A3DVOxw{yYOj>^$A9 z-^YLXvkcrn?@sJ|D>LO`=jZx+0}q`ooGF|HErkO-_GVUGW3&yNgV(`tH+waa0h5&O^L)B$DQAIH}(4V)fp z<)eh}MO)VDd(P&)KDz0mE(#qDKP6`2LMeZM-z8_UZfOJIodT?!H9B`cOyTp4T!3@p zQlf#LU%!|Ez!r4Zd?)de;F{Xcp`}MU(9#y;3!_Qa&SUuwZBhvHv9Bia{8K*?sK1a8 zDqJ0%Dn+Iqqk)cZe;r;Q@+_NSr0rEs&GwvopqK_+C;Ds2GrE6T zFsfu{uX7?Lj1pON=_I}|i-<8j{3K8Q_tLh#q0 z=XAcj7yW+urqw@7X#f890~BZhBxZlRFlcbN9N0pC#yj& zS(NhkyVDzs@?}Y3Qjv2kWXNv-8f*S{!Q~g>#fSA z$3?f+NAn{CvD{C_QFK7pXH3gji#3% z$w&McjV(CA|m@A$TSzrN^85_6Y+V%vq(3m=wbRhTs~_AUPjAjw;*om|ABwjKJ{u;u8hXy z$EIH=tPfP@UwLKzYMD^Tk$2?5A9tdH!M&VMAOB13^?u$nCv3ijdPI29o`*mU0Q>rT1KH!43F`_TaoE3e9TXD$k-7~Y?I6+$dRr!lUEkc5xz=m^obQOjoduP}UHz=3~zyHhS8Hp5T*ksg1}Ev zkhwGda;YM*I3=3E0-TAfRbl>K>H|Bl?%Hl?w_wcE@Ba^CXzv4`^Sy&dyt+AbB2tK& zd7n2)IY56pxz2uUZe0fRI|KdyK$Z4hG9X;4wBdEU)LM#0;Lz$wj@zTEe*jGwq z#{n&sZ2CQ5ilULZ&9l@(nJ6lMbw#b`{q{1Af&u&GK8@O_@Etqxr#AE>FHk>PuF|x# z^JV6sHR7GE_3?ME%u_GpDs+GkrB-%7ViG}CzQ1(pu=^l=L8D0Gk6EdR29J(wPm6yi zb{MmzT{i3JX8CJv`ZL|{Zv0lW%K=Rn;Mty9vl@n67V;?i48-vV;oJ+|;Uy^kxq})d zI$)(;#>Awu))F#MIIM5hKekcxq1!^hZM%v>z-L0ZktKyo;cRa<-|H|}t=R;FQEi>A zc|BGk_OuP^SK7FTIAnkbtPiK@8|{Bi_1BJ|qm*3veb@IM&a=G3NIb-ekGjjo0IVp? zY@<7ecB9?@zKEgOaH^CAAk9kDS2%%1lCH|;RF9O~t0^za!B9v(XC7N{helVfAF`>G z295g{PFys+(Ple|(kWsLrlP1f^>9bPLYs6QeDwA*^nW>^@&+w~72%WKN7LLX}H znIG&aZ%lo}*wg&LPAznn7XR8%btFS*YkMuHFromVUn#hw^T_!9)Smaz>YyP7rQ&lw zdHF<^0~Fcx^kQipI&~U)^$vf#%cMLe(F@uuMlzJl|E5l9f%;ATEnnjKLkVd4bv^k2 zMv!y?X7+kGjqBzJ{%$z6f6=$L%PZ!7&Qx`I1sZm zutsR4!eJ~CMdNz&lyZN+0=p=@NP&^rv29kNHzI4jDV-8(&xsDOrgT39Kl_^)vW7BU zP|V&!^SpBy;&h+!bZ5$CqWT&59vbNPzMC>SAm*}~@BeNSdykre)8^WKhkNG+HS{bc zeD;tILI=#n&AYidxFaJ|x78Lt7OZbx z^YybfF0g=Y__Y#kpI2p5f&w0u;5P+0HDr{k+ zNMVQw9q@mxIPC;T;6$?=e*$lr@RUR;$fE0?Xo(%xX968;%dD9uz4O`^%j*ZvTNJi_%a!$8P zH}15;!9w$TVVifpM!zqeX)p7lhx>%oGp8#AJiveKG}Zm+=8SSv;QFwSIu_-!Vwtr8 z{YmVV@hr|!`AWXu*1WnZbf(Yu8=)z{TbJM2W+UmkE}Ly{4@#XX2^sxB&TvZPSD$CT|Y)a!$hSNgM-;Rp4W5YfxUs{TZt z()E9zF@^(4%(PNr#AkHGv;vU_LKb@r7Pl>JZZPaw8;WV8%zy)HR3HdDU)j9A%-0&W zIeS$xcmo!{Uc z1AXOExz(~ehkM;3FY_h8t;UMzYfZD6tbl)8GLuz{yN11Q^{u>s^fHkY32S}I$+Qd` z{c#HNNlY)VGRJ+Fc|VS~38)vm57MVFtkh|q+Z!xtqXGJOUlkK??(`}d_s7x?4xdk< z0O?T6Kmt52k(G)kDXCJ2KCNTUWZwoXI(@|m;ST;368<}d=J?W5ALB+h<7)QtY`}jq ziOelNx5N9k#!AnLS#7*qAIYf-Lw2#}a7IeqouffEo+Sludd~{-g7;ne>mx!AjeMSl zOYN4GLz%~7eZ0G_r4XEqE7$Qt)?VM8=<8#6M=Y(cM}L5;Xt93l7gPl7sW#4M_MsBQ z&T9WBTCkBBP5NeMWn}qZ>7FzfWo3WtRzlb`nY14Zc9Qki>O$UT4MTyA-EW}Q>D^hq z8pW3dfgFZ5a%)_LThHW15zW2Z##v8&#K!ZRe^9XQoy|LPy#t>QCTKrI)>@9Pnjf8T z*;9bt_^iJ#RspByjev8{FY=-oyfWRVcqtE6pok=6|7Oh|=6^vC-`JkWsW5-W0ts%u zzY|7z*C$=ErQRLqn1jpsAwqZ^cKx?>TM>NxJJ;pgQP7N*wu704y%kLKFUpZw{4c2k zgjEX_GuS$=U4B(p0;NEGwn!^=>*>weK=q~IW7?~VMPIW$nKkpxGH;iobveWb%E2qz z7OD=pwX4W0ab_Rj73~>F-SU4yn<%bY`)@zyV$;tgZ6WLkyE#SJ%w@fJ$l>fJG3TK* zp=9uvkt=d|Sp6>V13Sa*hL1<`NoN)p^zXZ0@FrI37@omB9|F+#waRVGFP&7xreBfK zBp<=JQ-sY^l~o47pT4-;XjIZHws>%U)b9$GIk~wy@TE?E8-I&LK$d?pECuZ6>h2(y z_03S6%XG%bF9nw+ZOEe*7Qbejh?9eHh$T0#uR1Oz7yOx@kkxj*S&@Col-6 z_*~tt)zW?6PaL#j@kf6fgL9<@NvFp~ZR>`Hwab6V4K~*QJLf+7!@RwH-G8ZiydXm##d9b$h4{YKW-kz}OiVOXZsW!ZJxt@y82sI$yVQW29XfsbG30XfL8$e!GS z-k-;{E~i+u$@=A572~H(n&l#KNo)pe@K%FMADCT0;=fM3wO=d zf4^0`s1QOw-~{&C!LD}Zk19E80B@uup3#Qj$Gey-_@*-S0Z%CwAm4vKHO+^#!V$AO zW#HA&6KSvNmTfHAvsL1XV?0LKHn+A>%G*dv#4xzjvKjU%qM&bn>po!77i=53voUMM z;@55-vBiJ2%IgC_;lTBFPv1FDXTwZhV4~Z$8Xk zitFBv)p~vIN&}T+$0pws7;&g~$%ZMX&t6mD3ghk@ve1dAXD1@$*3RTza`3MFe)Uf~ zdQhDEm?sIL=Zm|Iz#_<|`Jnayj=j*s#~D7>w#U-l~+LN$JCrJir!~J9;pQVOX8A@KKl5TRBh!O zc81l()mO{;Vm>WIrAt1Oym#e4*v@s(LV1)|(Y%==(PTc%G*W#y;aEC_1DI>zJIoV# zC98k!ww@G1?X8Bq+O3Rk9kc%QO6mF9BmNm7tPL5rn?oZBrWHb5fz!Peit>S_C?hEC zaMqE~&e!(x^O`T|qGVheCz$T6a+IG(&+8(86!hVOio&iw>*qdo8e=RY6T2u9*w852 z%29s_NE0+V+G_20*QVT;1|8RAqE$o~F!6s(9M6K*)*JM3QuX7mz%A*qc{>^4`(`u0 zMO6`e>hR;_;n zEK=Z*ky^nj;mvFNTT!MQ=aj1ulP$oJl+-?ac|+t`=1sm^9n%d(h#vQZpxsYQ(YF;U z=zz-h(QXQiA%8pE?g;J!(>HmMv*)Iy|M%22NEYLWFDDHPx;A8&6<= zu9`mGA=W<236&Pt1o>yx87tluKbn6S;=;9ZuElpM%QYR_ce={Wfl7>MpbjKb|A0wt z&efi{1~5BWZrc~AKCEm1$y)??RX0(5n|=Men=Nb8`*@SL(4C1zH^TH_sa9L7(r8QY zV~EJT4{*8}>QVQdyq$Wo0kPYQ$WHY-s}HoF)jA)`_}$*%={EX*+WpFe{D^-eXsBD> zqGZs`Bj94dPWFBvhPL(k`WQ{O!#|Hg@VfNo*^oIUhMbJVSdHIh66eXg-|KawjSoln zHolO%d_{d0C|B~)YgeS7m1HM@`~(-hlsi_&jl`nq6^N7CypMiPY8%tr6rIa}R%%Ib zlgbS1QLL$T8>pt0iA*&-SZsfV^vT@p;fMGhH0vALA<}OMhn7?4x4EVbJKKCfAVPJWljo+0r6n*2%-6Fl-i?L!I`Fzr~s^t6g zzBcT*Z?=o9^kT2Tf`A!Ysr%li*G&OWS8TDlmNHpMGSfucJIUm?Q*&Zc8l-~{wy1$f=%2P5|N-@X)xMn+#FIm2@ zL!nzGumQN!8F~{O;3injX)Z4xly;o?fIO-8bG7>}XIJ7d8sM0A1 zDR3+HrPFV)t8s)z)!2WW>WwaEdZM)OLMz8l0KOAFwOCkG>zO^B4gZ9W;gBd6{$low z+GAjI7;G$g(wseAlEnpowpXhw7Nu!B;Yq2foF#T;G?dJdx3`GMJ2WzYV>RgZO~BC7 z#Ezotcf3wL%dZp}4NE>>?%keGFwWN7MwJ=Wj#O7xY=6}aBOZU^g_chBzP}G}fScp; z!3fhdnV^>Leg6iA}jJi3s#CAq&s#Y^8Ultvm#ms*oiRF z0V{Jb_Tml}TGPQY$mMtR)6Em{vthxWm@h?CRtw_Xd+&*GU$i@>L%ym~>uPS6Mmp4) z!2D3?wKh-mPJw^j=;73rJSVy8>ehk1M*ckCS=j6L-#^saI~Bx82ViU&etAqz4}>B* zRm;(m@%HEC!unEEf-r6K9$GxaBYV{n)qZ#5R+&-r&qLZ4hiS(414j3(_F~e{p>BHo zr4p1FpmzmiqB&@k?%VBPk)bD{lUl}6)M5^b_s?sTx21nTnB&&pX;>>eX2&g4+xF*q z1s}IZRgYo?iY)UyOKGs4T~)Q7KGT)?Eq_1MbqmPG?zzoGrdXTwff~%NS)x&=osxI4 zZ6$iU5Wve#;di%F15f`aE%5Ty>+w-S4B#c_K5bNNWch8CxAm zz4sAp3;52Pl`258rG3>8+N4wt0>U)RD?8?NsosA`zpYjNRNR%N1!>@tusAd2Ei#t5 z0*@~m5gpZZ{EG6(r=@|_+whMYO_8n)HAp`n724Z!bZdQ|HSCaQ1FLO-`icNE#je_X z&fC#AKja<@l&Sl%UHcUp1OPn{jA@X1RKk6=pp09;Qw!Z{M=pAJgOz-_-dn(n+)ES3 zkQ;xznI-ShQHp8$U5V-)@xQI&Ji5EX(wY3D*f+kVJ^Hm?W$dPIo@oR72SK3l;Lj0zTs|q@nHsl|kE`1X9^mBv;FW)E zcP#s+EGmgE)9l+t<<$jx!w)kBDXkSg`7Pzu)13iN(*p;;&()K!ugK^NHf($Ea03PP zBJ@Wgj%%1y35Kob7j}5O%lYF*HsZg8KCdq3%v*K~E$1UNN=L;}DeV5Np$z3*+lOG< zUz;d!C4BK~t>(j;g>M3jp)iDlEcJgOv+p@<%Sye{ft5?e4Bupq{QBhBzZkmwa`--X z)qgz;pzo&zS(9#~MX^=%I1>^j$}C^oEL4^^dr>w?LWu!kA{#Pjv8t#in=0)Hi~}o+ zQA|Le*40ppXAwlC3a>tYq-;1!^g)CI ziUm$E9tMC(5sf}V{u8)$o|(eU8!4cy zXcW4b6w!tNKc?pBtxZGz61MH78&!RWP9)L*P_|%))ceaD`5byHC(PHW4^snwQpEI( z|1^IK;$$|g2z~lYaBa4P7O#D*pxx4tB$f#)@9oo!RE1CYcj)w8BawewddPWb%M3h1 zpZ}+C*83?5Tz=SBp&t2#T`c{{5?xJEuiM23_|srhz7$c@22h5+crT(oFT$OU10Lfi zoben-@Vnv^i|!C3*rA!e0~lv1P<&9j71)5$D83fmbt5Gsj;*n*)-f=rg&l9UAV`I7eHQ>$@Rzd`_VX0#5a zPk-fpf*BSL$TnDRP=UBqH!oRBby8IC1;1b!K9 zwU>>TF6rmA2;`>MD}e+Ik2h#+C;cW>!F#kNFj;TGMT2)QTN{1bXide);4lSHEy89k zDf5?QcHnq&b=FsK$i9@!=Gbzw=daPFFXsQYcp>Y9@+1*P>?OL!cc!uwjgxFFwHwemiu*;_u4?9dZaETOqS`ac;^K3qB0!FW%lzu&!Zig z6BnYAZx1QTa%sK7sRU;9a4+;aR=OjL+uTYxGF*n3*ue`@7Q>mA<*Aj%Y{t>c_}g>8 zOIeKvJ9^{&w3-iEY^7upYm(LEYjPAY&OFHO7Xz%e)JD z*4Q~x^z_p@82jAk?B?&P0uQ8x&C04k8y^ZApon;;^Joj)z)^^aQ*ZFWZ|Um9XS_BI zpsZbDXD;BfUIYTjIF7}He`KT|)Nt7=MKocgFT9fvk_OjM&)XX}K25*7CkmA3z>gw=O(;qcVyoD%94P>o=KzZ<<(ZC<~x7Vi~&|F^DI0U_!9Ln>|r_hv|euQ zYOHmu2l5G+2GkX!ZuJn5VQxJhd6y~X3)ZQdpgG!zYtp%DXpNj)7HFW2Tn;Pag-R%q7iI)e9D)MF=J`J zr(u8OX@;Lii%PZGEcWR8kL^R59J<&BW5MGD!wm@-d#O)ZOGdHOsid4+ZpD5mA=gwsQn8pwbI-P15Ol@wqxz z(QcnpFNvdzl*nJTYr(-=^9_752DLlxb(tR|*dpI-5IbR8)a5FJMxsY5`DH@JEW<@@ zzq6tg2G`X3HFNUV$qI3efA=e8^$dRi090I?^f_~wjApDez5Vl%(L$Sd7`J~_Gtl!G z^5Ujeb}&4}bz=jslD{E}LR6M%$&;gVd*=D~Q2kE1TSb(ulSCSw+{iZdzx|H=FHo9; zUc8gB&_TwGA@A*D+zXa#lW#1B4R?7udk3XQz^&ut&1M zY0{+3JF(n=4q(VciK}xwv2BN}QNV?Nt))-Vch97!l-4yUr;W;}lYU5wpGv{#N|^IQLv=@vc9eVK`w%3Y;>Y3M<&WOcBo z3@r40OrO6FtWr745H*dggU?Kk_K;V4#>k}coK9HZQ~mVyEH|ev=zjY6SK!Iv1MgB_ zYTw$giS3|?$K2(*+4q$$-CEpWS6}k_EwlV-q_Cl|A{=w`&9derHD7k!Z*- z*cRBJ`WF{dC}3J+xuJg*I47lc*oA=0mZy?{V&u1F+$a~NWW2vw=(gHaoAt-)A`x&g zMxCPMJlg-M-s%JJ&UVW~hRO-02NV{8N5|EEGm565UCvx)1hMCqlk88ipH2*&A zHo%$DGGye&REJ6JW8i<~-O)Y=uX(6{0(@d6U!nrHtwRSeYgKrM`iBU#F#w@^1_>S`*dE4MKjm&U^-+`=Y|MK{SZ#gvNUz06;=v-|D*C*8>2BC8SA z3}MO^mhB>oGhcsblPVArAC5}M#8HXa#>{SMgmGx34Y9Q|dBkA3GUcJ;NSX2{Y>}3I zWy*hBtf#9$L9QRC5#N^FLu){K?JvaOV$1y0h;7EZySQ6J6P@j%-2Q8AQDl+sgx05u z*4|6-vB`(a*h`nS2!Wx|vI7>ix#~UwR;q*w!#SZH8r^?KJXi!H)oyMyxIr1BK*99I zVPGutvPR{KMD9wNRl+}2!Xp>}Iw-gr0Z<~c8!fI{@@)h`+^kyy_|H$jF?>p!8};|s zHi$8DBjYY@0#dzu#IKajt=ZCk)-aiIxx*`E_&V>6fFPIFyTwLep93fGi>uF%ghA0H z3V{l~*0X<1X*CIcTd;v0fXhnEu8C>_Us<HFnHg1=Vb8D7h zd(oFj8dis`gGxx6;|K#6NGbLN#(qBi+2>#5)T@8HIQ1ZFB4n}MvQndRR(vE~U{}EH z)JBscC{*8({QzskY%o>IVIXOGTy!GXGSh)^1K$2T6{YTXNfR?JnaCc0cw||ufWj*o z!`IBzCjosFU2^B6Pt)D#7)mF~)zz;$wkTG5=ZHsa`QIk1MCPgEEonC=Qi%8r!5y@X zybFH_S-ipw5;*b**^M-q7WB1z&b*bU%5>qg4B8dxHEyE@5T7QTC3RzKvH3>(csJNs zv&ZaF@fKy?H|ww2P7l;QVfFPsSC6JmtvBu#Rth+S`C;<@K?mxKX@@yV1 znf3H+)iJ*QpHyQ$O@1^tc%x>J&3yu+HI_$#nDp1}YP@GPp_5Bm?ufkxbHH?o`Xwyh z|7`U}KndlpP_TEb%9+cTWb8cSU@Y68QYND#!F6k-e;6Z!79cWp{lU9jpTU)+5z&8a z)n(V8#&^WHvv5$*aXfKlqscrN^63i0<~#4dO7h{IBm*-ILdsOCznBt&gy2nv7K*JG z-RHmr8^OY}BRp{_pQT4PUy+0Cm<0yGHj&*kGAS;0Sf*0lZ^dp-YAqpmj$cuZyQh{4 z_Ra+JL$7i&&0SG+i@Q{#__e!@W|Dsp{574RV4SD*)OCL{7QegyhWn-4&SMD7&~vfA z=EFWG#K(lqs>*(?w^9=H)}YDi8_H8Sn9qd3!+hoxOrY9`>b=1$qsjBKcHv7&SJ>F4 zxo-m0oU%v0o8v5Y0TrBE;#rL~-I(5%61J3Q4`m8S`JBTTXr+N>nrW4KmGgh?gD0e} z`M|!4$el}9P(N$qy5~f2>nd?5seoVfwybf`K)funaNv}k`fi~O_vxW8kGRm5l;!w+T(nJYc^Uq^{;K6bn0&PD&X zeNM3&fr09=E+L*(&Bw&DX&#gmLSmCg9Iyo#y^P^I<_%`8q#H!QAvd$%G$p6~ju*ry zt}v>84``u7WqP~AFDPlKiuzuRW zv=ZA_p!(CB@=SXMvbL%uBax-(zr!}mbn$Jr-TYZQNVHkn$)^%3eg`HY-rFC+9VfuV z3G?pW;iJv@c1kuMJV6?bzO#`xk|?nSOjpnX7ImBAOwc5nsb8OA` z0X}$eFrWArJo56e%o=~M6+3H-2hQBsFmV zwGx=R4>;E@rX#y2Uz9HBE{_d0)H<1D5@d>vj5K3k5l-W&D_ALiKhU@KDm9DGS090; zBILQBXv32ZsosndrK{|in*}TA3@)A5ybsbhTaf9dcilXuy$^rBa3M+I@y*PG*IBd2 z^b?3-`;dwEjgXQ>fRZF{tM!lTrhOI(?A6D!Gg5M+MT_6=s7IrQpWY?M)?}dKH&cL?YU}7`|m&LQFF1V-GeD;@|29xu6 zOM>T)>bKqq&i#LEnN*MjaVZpq-&nO>-$2AR>+HxL9I{=}HXxJj*}KA@6m*rY(lh_O z+vGG`-@suK3zTPG{5|j3H|&uq-TU3OJS#TI$ZyGr=*;Ej3;y6V)NgmzSjKkjy=?Xo zDk#WTKik+qeI9b-ScoaF-NX5>%|uq!YpqJLy!LuB8TEfPc9-(rtNmMLn#Y><6R&e} zU*;M0Vu>X%B_^Tr@Ig&&^OM<)z1i9t6wZ>CbquqcyH04cQ4An@-B$P^)fZmMCRh@! zSr?Fj95i}{0MA8jv>ViVm>#F@oeJ)xMuLfp{LXJYiVe@vvR>V$9j6e!cB;(-`$9a; zhjc7eHgiwClrL*|AA^&%UyNBM-j7kN(Z$63o#F&^iII`ZX`}PR=^Lkk#Dl9TI zPwLat0)?b#F#IICv=Th9BidyhU6=FFf}7uzU5RWuX-NcHhq8Rz<+t5TC^W_0XpX|B zJCntXmCs*3D!p5HlfGBnmM-8rRl*5zJp6wykUL&*$;@Rjl8_|SwmNgYxm)>@c9_a- zf>O+X0gC{IG_A?)$s>-oO_c{rCT&FwF}R+^&DVLLl7#k4Eyv4?xNn)qqrs=PiK$)p zWHy3{+}RB)YVP8F?PVzQ*kF1}R%j#>LD7kCK-FWUQ`4-Zj5ed%1siSW#3;duZmoYs zyRf~Yg!y8CeGx4O?5iR&^x;$}G5Z!4zqPpI&`*5IyIAf{F=X4r+NQd9K*Q;LZPoFX zb3gZTt$YY}(bf{@?OGTb`QDU$<#T-5P4{Bs+q@9--D#3;AFY6?(xo*`X7H063IsEx z(VG1>t>agzBCX6ZIuSF=K8&xP9d+0>}3RZsR=I zFC)$3f4%l68!FXpjiErEe!bxwdXoI5YmYuTa6G?Xyy@GZKTIp!Z=8oe0oO54O_T$`=@^*S!lid zn-BH(G5uo%PpRV97yvr0DV2>-|1?D|>0@%4t@mJ=h&p7hk{2%fBPPpsx7k@d3lM(Jnc< zpzCJ%i8rVNfzIK!r!QgNxW*d{_(Qn7uA((soK9 zavFtnXw4=URY7mj9C0XdEx$9+0$JSpD1|%*KH-b8kww-d6@+buJ`CTJl3{AzkMxz4 z@v8}qRLRZzX5a%W(9)c#G;9IC6r>4j-?owv+=c9Jjmh;h10}4#)js=I>O5EV+nYIyrl}p_8Q*4y8)@sa?Ksgs$Ftw}kEX~C%Mbqb z#)}}cXD^oH6@TYYX@Ry&-aiPqD^z3|@>r>tm^n9ZB&N{^fZJTIFW@!v$q#MrHdXJp z?u+i%=fdts8J#h(d}ly^q&p1|z6z17?~BA2V&o zt1lI(hj<*FsIyW`c3pq15K>GNRSLlX(z_HRH=|y07pQ$zwz7JR`Z8^r`&yiYkGO@lSgqnTbEr3I9ZlE zbUCtq(Pw{R!#(9{_Ob2caYfzY^f#>b*0){JGz<4%?`9H@y?g#C=gv?3 zlb+myv8!g1n=j}eMLvG<7bUd}*D?E?xG?ZF#Mwer+=qLGBtkKSIgQ8nXliKbxbIoa#r-+o*FUEc=#aX@d3Lg6Lib!nf&Ws zuqjG#dwl>aT^Yr(n79S`*O1LbWf69Tq)YsN(dJXfNki|Agc>L0 zQb^O?x}7TZ-oRQEPn>qEhZ{HksgdaA$+76mx0D$+y>*X7$Atv(e*9ax6N|3o9vVn; zBf@`6@ZInrWlLz}X957DQCZPrazXKm_D!;mR@-Qc^`fYJ$Bv*##*XWQ_w(8691)~{ zJdzdQD<=K@%N`YfNipxE{5GTo``zl5ve+X{t0vg#v|t)->V>V4T~lP}@mtxK z=BvBnX2_NQL`>Xuf{7#`McTIR;MCLE%c9 z=M^57zyq&A)=Tq&A5jL-MxX2-;K~XT(^6rt-_fKF3sP=VnE97wV{J^h4MV?eV_MBAxHR<^WcONTr?Y;MDUe3TkSz$A#9 z`Ne;#9J)PM7vUU<%H#K2Sd-;(m0^VbP)F#~s^&=E)4O%b6poE7K%*s5`xg}6xGRwF z)ESHTc5Dd(bHB*C{a}KRww+`PDy0B0d2B~+4bJPclD`mhcmCQ>!1I?0sGegtu9jaq zSd#I8&)RBU0w?|5ub3%+&6!*`75(R7Ur>@q!*{ToP+9AI+r1w5C~o7&H|P!TG^8cc zD3M2jnCrEP<2LU{^(Z5t`AM?#nZ?m?V!cZb5$tZqe9T@ctvZa`eQ7#3h#$8%Sj}y& zIj8r;WR1a}DFlxa;G~T^+jeGS*$lc^tgv^lYtt&97i@?iPBy9vt3QHGkd9p@ zOjFi7sGAFt(N*QXtZ$GG95hN~z0C4E>Zc(wq+8a0 zQRaD2y3#66Z$DRe_pe84TEZRSj1tr`0B z6vCUaswRU|n>0#)i>VNNw6UidX`bOx@Bb-;qD#eIrNYjdwiA)F>r6rU2YPe&L}~Z2 z{9Tn#fUrcCgoIuzekfTENh5VCyeTjGu5!3DU7qkhup62}&5JHnYXRYskiP1MJFEo= zK9j9T=S!u9p-j_&NxS5Q;EhT)=UtWCwITH2_WN;p*<@gU<-F(h`*7mAPKrs>^9qrg z+fMdeF*{r8;0V>cfq15u<`L9cff)Jq`ZGXi}xJ3<7OLheF@r@ zeJf}Za;9Ht#M++)FZ|qcG+UuwIGoO-%nY47;%Kc@t|KA(O0IkeUi zk_uGjD4aik##4iNPrvytUY_ayy=_Y*;@%+NJln$`6_8BBtJ8hb@1xHgN?f0Quznn7T4!M(?)9w+&O)N!$$CYWvTXE0m zL*)|W&%ae*>Ysr9tF8Jk5jHcsmv;#o!NmBo%rM!1WWIBr&nCymeeHSrO(5>3Fd#$F zJaVYc=2UR@up4l>5@NV%`BFL4U=vZ2^K_;Kd9_#LeN5~c&*_^5^yx77x7c&f3eVc( zBKtKnDk1qv?NNg4*(6wuLU7E_U$?J$GqrS>ajooB#rV`U2hwfM4CV%%%s1OMdc$4U zf3kUhyKE(z^o0Iw-x&C5=($8ywX2i{=Ij+b?*4GW_w4hvIxxakV{X2Ghb`w5OTn^(E;!c@5rZd*MOfc| zxRB>@SvLj#;X@zt2~nvW;|9Qm@=vb*yNUQ+li!6da{v3}ttDKzv*2}2hauE>K~sBi zk_YpnuBzlyI@vcg4}aXFVJiBXO-J-25IdPmUgd2Mi?rlZDaAm#r#~p(m`u4(FWxm- ztn_y;uP$_iC_*y8L5Cs`SNM>dwzhVE$7F(X#(gv_@q6e`0`EO|aqs_p807LY@S}S3 zIvoVFuNClAIGtyzh(Jd6?@;_;41)N4)%J96n%~_Kbkvs z*n_W+Di2ik>+L>h>A`O0ef9<`DNL!net#HuXq?*2#gR$&VJ_ED^V@Jc^t`=F(rr00 zKbOVbaWvswbc>6ng7+jDaojzBF6r_H9$=XYuW^-gp8U}xc6)7H`+kMSv-ETdQ9Elh z$ZEUVuy7x}`02&`v~k+Ball|t<0Nrf%ITCbdwrAkkFR46VkL21p<2dT3}N}sw~0&V zTO=Lt{jU$94qeT#NFh3@x_2RYN)^aGUqr_FP<9t61vULR_gZWGMcamdd5_&P)9b)M ztO>Jbzl)f@q3zPXw5v1+J2w7XgId@t^0-HnvhOzul+$WQL?M3mh*%@!2}mqe3df@P z@SG3oXFI}g``P=jM_UoT7RlR3>05{2Yl-~;e7&krJ2cfFyWcQBSJpnI$M~aVRq9ZG z4_@ZgEE(#pTni047t?Wn5WQUI@^nhF5K7V1x&(Rz()f>_k)V}||0-Yy+l>g_K06*) z7E5vclI5URXQj1_4UqWn65n~C;#4BoO4fSxCCz~+$-qam|J>JeR)N0nQ2*=upVr;O z6CqU2QC9+# z%?pBuO&4sKfwl8>k<`ekvkgl5-a(QyY1BA3alCDPb_nA30lR*KR32 zTevkUGW69IaNfVnWscss-~(!2&K(@SMV@z*CF2X#g*Bh7svabNPY+lzKk#k$IUQ;f z;2IX(y%t~=7ApZPgApJ>49*^>wI9wF=Xmy|RI}KA8;;d~9ku9|(~i8bHqfkY<)!m* zoPvc2Px}Bly)~k?Iq|A=ivu{^c8O!iEyQDQa?o*prk1zv^WSY}{mz~N-0oYF0?hn3 zH}Kn^iUtj~^K%pv-nV*~S)1sm)5f=pdsF_@$KCj1EnBIg`~J1VQ#!E2l<)jc;mKFYgCwZvRg$_ia_DN6HMOUwZWKXHaLJQsIq3{-iYh`C6jsTfT@{%`jq`{^Dw zz*?n$Y?I$i^m@&B2rlh-dYNRL9P|m`8j{B6Xz>M?lma*Qvi*2O$O1f_z+Quw-PMw> z4WZgxsI?6M+0EfhRJL7x@!qTM8%DXTl&Nyha^~$_NkJ8-cU{GRwy$)#0}pJ}RwNQn zHN+Gu$i?RNYR{=O0iN!L$sPz*LzqU}YktjtU#x}HYH@I1_f24I)3~0joJ(cPv%*(I zj5`Bd(;9+E1oSqQxZI<0${Fc^yy)W@KE|<3+Cy(C92)`JB#A2}4njIEzAw1KKWTMY z5Y*oKQ>qpg7nb3(Qox&ieYu>!C(j7=w~WUrLpa&nPf7%u27@e)hsk=sti7Ma%3jug z=-O|bffmn^tAn-OXIul0M`fs6Bfic>>yXfwpVzU8*>afPa1*6lA;x{z%nMGtFsXTO zFCof>rS(6X3ipqv-5D*|vMa=JV4waMYdd(*U-OqQ3seL(JuH64dS)}~DHYqopR5x@ zT%?`f7OB9D4S3Dp(@sL6?~FB(|6PWEBQCRE%1O#b4CQ9L6hdJk6YB^geRx%eqaF^U zD41xa7R3*(2UOODs$nwra|dg=!ktW!2CnYG^Tpiow+0(>fiAlKmsh{K$A^n0ATIbG z$f8=MC|ZB_EhB$rea|c`3{Ys>CF_~z&VjzJ5M$|}yjZ5J9^Bf{f6Aoy%9@FP-@Dfj z$@_?qi{$q_-4FHu<@|zL6z*YwTGIHwseSpVPNOJ=p@*U&%%@85`Ev{g!t|YGTe;ZK z7k9SQqv%6?h*kX1FlG)PrK~mJWVzu~Kd3s8qx6X`iVV#r_j1FJG*YKI?1kx?#!O$m zrRc<)4W7EP2Bd0#Mj9X)UPsk`qV!>X4do)#sT`erlvR@w{IAsuPcx@BELjWf#k=}D zNc|G!13Bc2A2d__FMlvWg|ZHtC?Jz3n`W0>1r1f&;pfZummwqY>%ufUyM`GT*YOr} zoZ~IA$_)O9xs;oM(n?f>^Brfo{Y_iD)UuoW5SNz+bFZ;d*L0YNspt5AXI@kqL=>@b zv*rDl9QM$;f*$SfpV(rqHO}N7`BRVwpB_E$SB8DB!rRYY`X0=kcLV;dHRW&s^)a z+h+6YLC1j#(J@1&UvIE~b=yAgw}ok?G|_H9VYBtH}ldsIQTrs`zmF2b1h1Y z0}4E6ukG247)sQ)|1hV==iPx=gfE;r-=2sLCtt=3ZCu`-j4N**GGy|ygD+|>ca3i? z8Wv2rY87fThVX>!9JO>e&Yef;L=#ay13sp}-fm(9+;=}lS=`cpL?gTlj5sjoU9!q+ zb0%qhf4krsHhipL^d`&isQB`7>ozrlZ|NuZqOa57iO%p zlU!Ldz;w_)_W{vs;W}sp*G(s^wc_#}F+gOt@8(+1RDoxGA&YCd+&zgsGqTSpu1{(q zL$#0*&y8(VB<3oAtwdrXMjXw~cI%XvTuRSjoBD7!&zVuM1gjo5*KLEj#g4@I@G8B+ zurc^*fZMU5Rsi;6%~U-`D}#u{+hC6HVBd3zOCbqtX5_!YL(3<0%C|VPS1lt{JDI58 zD;0N~JSMv4JlRN1+DvtGEEvFhT23AhKvzM z$x73{G%PJbr&@z|;0C8^1~f;(tkNrMtpPWK!lx4~$4E@-7MEO?>aMHZ6ukZU)C>=s zXP&m6dnm-~Xl>l`Ewt@e$)_IizO&{Osbutswx?>>iiMF8@WVVd`#Ep8(oL^UY++^^V%JihdC zi9<4vZ^h-%6wj_>71(;Jtaug$<1EVSk%SBNO_iE|u#g|^aHbGt5w5|us&%R9snYf_ zMDuv0l1sy=Gh_Kk3+sVDWW;cJryk$FwIh<1K!|&GiM!F0qB06+-2rV%PM0`Ptw_Fp05C$Xva>eZvwH7_VXn zp{*;WqYEnso-pjcf_PN zeq7SIZ?(B>%y3%Auln07|6gbsoEzF55|@KX0UYWeksZ8p$3ZVka^9rR`-7??czmT5 zE%J81)UJLxm$_~@b8K*08w;ET?R7bYPq)wKR2UPHXaT#c}ee=ohnKnD2kSJjuX4yQM92jYEK@|c-j}V(Q9%68u^^Ikq#mVJ!1i@5*B215=X@N0wC@>kJlx2Zl{QnYtu6$w9+SVd9Mf`AYgQ5P zx1AO8SUCE2nY4di>Db37v|pDuYkKbCFha+cAxMnEx!1?!T$3YtZN~?`ycO`_$cC)3 zp^>+C_FLR85#?*E%B>v)o>AM?Rk+BN5Nf5h-udUlI7MaF4z$Jjh7f8A}8E|Voh~A`k`0K`PcuDx_8euy4=0NE9E9E)3e$Vh25^&vZXF90QB^0^nk8l~+Bmi=^f9Kuo>Bf%0^lnohjo2I3lBu_kzBj4Wp zKg6FYaoJWUx}K(@@$8Kn<_Zo^CGfP>7(G*JwAjWva)Nw!Ks3ZttQW8vpu*U6bapoHP!;DsuPK;xBCgQ| z`vLAt;ewFCUUB6)TUBS*8g7&Etd_}puXb;m>_ZUBtcg+2y+26ney2XqiXh|V?uL-1 zZr_Z5-2HjXHo)@clK&dC^l7)UT~G_g0LzUa{&-?!v#R6Ge;Jv7wnr1tLsQqGtT~w5 zAqoR|@JDoGEBx}8R-dIH<>YBCGu6T=NNUjJu=XSi!Z7aA~ z-KnYDDYNqFhK|-{i0%=(ZfrYmw(L4S{$vg=^=~;@J;r7tVBAkK;GJ6lLT;*0?NRBB zV$vtPH(QymEtlPYHl^mP_#Kjih4GIuGyPK4IdQsSt*=I+_C|+a-!v1?#wHT=B(d8i zS17QNnYU0)NhO(mYg+O^7;g5;HO_zAtE?Tlq}I0sr;rM5*38IN4L8n>Vh&!CYHKrB z5pPtsZ0hCpEEZWgOFF22$>glcrjgg%m*K`8g;Ji)K>W;qh}5U`Y*}lqy2h(<_)9SL z>gH&>kLq52QtSY?=czVBZPFmE+3%h%hYqs6!Sob(L%^0=iLZyjSs#ah$iM2%u3UGaAT(@pd?f(_#fl?;`o zk?+m0yywJ!>WZeekqJ?|kd`0`RnfN=);4IEmbcFSUQHwP5q>LN&8+F@C~N^8sD=T0 zXa9b8s&&r8w`sOa^^^Sh?!tysAE0a#;{CnMlt}NWe?^93jqpRld2MEv6*-_3ee_ z(i;goN(v?(M2pIR{D!UIaq*JYtqSM&vzIi~o zRpP59Peyy(M*kKRh&={wEdkd<^98z5IVmSRR{)chFp64i)B(GACVY(s7b$mmOfm1Vu2%7QF!M-yz@|Na z5|%c3cClS2x56ew%$C2N=@HNe6@yyl@LnhfT;(-il_)69FY0ZC^`5%RXOw=OswE@1 zs|ZxPN~*n@;(rNnX1Nm8F1VJV<{bt*YNi--o%wKC?75HTo8M?HK4lWVjI%E6c(pXH zmQ_FD{TRYJZK>IN^&unY>eYY0R z%jrS*Tm-LEpZ|IgM50vKZc?7(XHot%7>C-3QFcXIPb$$s=oTm}CszO7&#jUmorusd zucuClZ+4=rc4W>&`;ufcPMMkIgX~iUf@wLTbO$g@&-GnL7pW9J>&T5&xzmEkW#UI{ zfPCM#8)Xiza(BA^yr_k&ak8L)5!+SdM2;n_lvMb)B$w5DvGc{-xP%nbD4|i#n+0ml zRLS|D2Pc#HHN>oKk87s!BMmBDS2}$bW|)Wa(;WHc4hycQ5r-(h-lvzPo2h$G`#KKZ z#;C>E1!aa>^3RrA`m7B%Qs6H-jw|tU>{cKwTgj0MNn%DH6k7Qln_+@~ja7sZ^UCYi zGRF%7Q8LP{Q2cpA!)|~WL!71B>lms0gfjF=1T~ANkO7g3HUHM$hr%sYnIFA*pXN6F zccN?4F=%eI3?j&QoI))f8p~L1pRrCD04M1a2nS;o3vIt&OuV$@-%Pb@MpVc-j1Ea% z4|b>xm^@d`LCqK;HiX1~^l;b8CDaP%g_gz4zVALA7TXi(=+@e*({VYDnAVyrAQP<} znD>RZXQ>Ltn~YHJ3)P1$TJw1fZG`s9>A5FHu}@nj^G<5t))SoZDYXn^B%M~_?z{S* zPD0Dj2V_z>vxS>)roNp47x{Vh_C*iI9izSd4{;9Kh&MHrnJUpguiFje9IEmJ?NS2Ynq?UgK4c|{Xn)$X+n7Gl+nhC`b#)_GvmQXZa`tHbS}Q@fX>%tX5VMQ904u!i50Pd^r^^JC5f7 z0XwTcFQ4ZUaph-EgDj*)L%m-;}$k?FJEr=`fi=+-nn6-_$79(es#A;Tr; z)zbSl-df>*X%wq0Ah9*sXg%+5JY6}_ih=}oJ#lr=4ueoa_j&+H9{2Ol@QVVHsG6_< z0#4vMPq~y9<2qhlXn!-Ib$p~se*9VaKF-E2{CvkdA2xNmJY)`b7;ByG*9(qU`zW~q zTs)nsh6i%sWCCdjL?b+;V}j2`4rcO(86`&C&Q*Ya7agb^KQU$l;4F@5jtl?t$aE;X zt9vL;8RCNds^q@LSv|R-{MNHmoUG(S75V}<&oLx({o~@>M%&#*!z@XAWIlN>qNr9h zO6-GpVPyeUn75yqPjHrU{l zGZxr?ed}_Kg}`;j2HX|gC=qy^#xvF5Bu@*j9CkWnM<zq~i+=w5ENdyu-^2ecP>fxzA>LeNl4P(P!mtCE+6vA!?6-i%HL1agN zXxWxgInuN5BO^0Qqb29xi#SS@ayM;KPdg>50&lV#_B2{!RMK`=^96Ti;{ZFRSNP9Y zd()YA)hskpCT5nxSR#y_jDo+Hi4{V)2LhfOrvjmB$xxn4(b5NAQ3`rG48d!I5_ks& z)_E4Gu*%}|x&cV)n6Py6tuwTga|(ZdH{6#2Y6B2GeN3zHV+T|+a-4Zyat(nUnU#(6ZVHI1*i7D-K(prAi%|!seQag@y5G*9 z>#LTD9@m*3w8eDSK5SG-0^0#<<=NGiM(OH3;$?GNh1^*v?%DR0Jsg$y+)Zq?Ab#Vf ztmLC`P3NCJHax_T#mkWC7K#IZlB>u8;kiNAkilUD5aCN%t*Zn_Ot%9I63r#gB%7R~ z-5gAP97Um~Z%YAli^sL)yWy@Q0#GY8X#xjzs;0r_Su8jaURnM2Enn##&uTi?&|A!d zq-v*J!PkMn|co7C-C&z@m;7J-XEyzR)}INE!cMY*VG&3x>=82u$3yl z+DI1w9#xW>uS zC3sSUokO|ZcOp2Dg0F4(G?CgcQ7eO}koLegnfDMm)52yqlO5zc{f@*&Pf=i6?_M0E z_qQedb_VcJh_eEt>cF>WJZa6HZ9L?0464R!9W9^b_trlyYy8#1WDRd0aO@R9>#Oo3)5I{o-EF z{o%WkLtpdvp9MpI^=tm_%Gsu|b@Nf7IGo;9ZM=0pOLN<_)Mvhq2qppMC<=6TdlMvW-$qdvVTyFLGkFX_#QL zW*Q|LZH}RTXl*lTp*-DN?B8~qiEYkXz{gb6l7md5)0@`lLnJB04a}Jhf!~9Xm zcz4I7s-V_Zu&Ua;} zSuq(f#3`J_aDBqpbNTJ!o<+DhVD{iK0wCY4b%V{m14?&4^B%DoNz}JH;H{KR$g{K+<7n))vOyZw*jjH5D`Sy{)8eRD;!$tl**a^pO-PVTtIbgO z?*9nq?8uZN4$ez-{_gWK;_$;5(jcczJJU@lHz}?4m&Cxwy&_l05E(6deMNeGwbazt zNxVxXrBM$59N%C&oM2pU`eK>!5}f$nQSB&yzBRi{d2XYoA7CM}TJO)K#K-N`I%N`9LkDGkaK^y<`14#S?0tJhvSs~c+q8;VTSbk0s9Z|wNqTDktJ z(8|%hG5EHb_~rk_-g|~M*>!D$w*?ERs5Au;8@-8i5Ks}2UP28`KstmPdO*N{bm`K6 zrG^@6fY72KAiZ}&?}3CKAe0H-&-=aa%zX3Se#bF0KW3gk9CEO|viI6+o$EZ;Ui-=o zF<##3$5Ogo04X9|EAgFHPv@H>uD`=2yfvBN?RgM|U|r2KH%sE}eh@h}dI!kvL^I|y zq}p>NV4GA_RW`gRJt_bW>HNB9gsD+~HvuA_$5?P=@u;%Q5|X74l?!l5)J|dr9CJRq zeXXFUp_J>}vcT2iFnqkM#9gv+EO6j4{Bi_uKriC;>OHKa9^ z&l=?ae0lX8IC!o0Oi#WhE$c%@Z;-7f!^ z18eN+Ms&Vde3){qp4fC@GwrN@u2Mt#>(|oZtR2;@=mR2M0p{8wCJaC_*&IJCUePm; zJB+D?e014F2-)LI`(f2e2YEpl451-Cph9me=kq!I(48;*BOnp!Y?8C9?GN{vdcy zD{z@y^w{F5seo)U6^Co%=>S)ileFH9_u5ISut`3Ea%XvOOkd8ahub0WDAhjszJRm6 zo5+rSbb#ZXM|~HiEV3U>Uq&+dU%mBUjoO?4vBV)fsf<$RggKBg^%W|K4;PrNq@+WVS|4bof6H(s@DyCj?a38lx5Hj#UdJiHum6F`Gyjsnv7^| zwXa_~I#&nGt8=lX!wHT0rQ+U@{jy>IrEcAoTV0+mX2X6T$>W{-0kx6%$>mqDU+Mdt z6}ijv{V>X_gcKnmlc6UdHHnS=r*%G)uGx8q@Qn@lfSuk6HEeSh?c4_ZtYuYiG^b60 zzH_!-zM~H*CcG_wVp;pzqXp4wuwbs!>hI&>E|W_yX;((_N0QX%W({4D>iD+S5yExv zj;=&9h$d)(4d6I%b6z2EU&F~#S3%;L*EHBr$6?Le4GQHsxy4w~>vmROSzmVYO~tAV zGyqwY!%q<&I?Ik$%^R2UpO$_qqPeK4$xgB5+ z;EhICy%SO7D#j*uQa;wWT}N9mZ}J!I0A_z^G1T4(Oa@pZzJ@yuI=d*J$bwK zh1dyY#F%P-@z$2D=e7YY%IvE^+5yH7XASH{Mj}Q4eVC_eA1&AIwY92*=BpPVU<= zS|ts_BBJSvUM!6|P$a*=|8nZ?wDAriqjOa-5#KX^`Y@|)suI^YH!c{~tj=x4k$tq2 z$DE@hIE(^1^w+|6^HKRbYTD7wtwQ&`wo5`At6~5jz$`H(1}b{l;pEZ#%iD|f{WNTT zmYrLhlCI{6o2U1)`1%)npOzJ>%d{Rqv5WKl=IJGmi=XXT0KvXX(EX_@`JA)jhSvSC zs+-Y&*aquK2bK7zg(PX@FP8K3XPhVKqc;M4lTt`@>IFS~<`#gp3A62eCg{9s9)R}# zhr@&J3A^_3+jH~k!t0}}@A09EE*3VemQ+a_$L^3E^z`-l?!6OQ%tgn}Y7E-amaV#J zKQDWeDt0xZt0L}9$tkc_o1~3>6U0_3^-iJXzIzj%bsC>#>pyo5bHn!8U~=Kwr(BEqHUomQ3pl3b zHg|o=xBfbVds?=L6C>dP-You+%xcoT^85nS?G!pSU_Wg!3!HDsgi4vmgc2(AUs}0; z)^1&(OkPtmP-kUqW)n&|?-5DVw;2XCCPXNQ>A&5--e+7+ox<4E^T28W{y4^#7B|Nf z&u6sV>mSS92GQ1u=efbQ3+pz4-I|($5(z9`$0%JVhEfrF`j#uDT3Y9EX ze5S>!nw4@7$vzxPQonjeu3M#`=DdBE$i{cvS~3C0d2(sdxCU;MVS~8WbA=ByYFHc} z<v1A z278AAuOtW0%lF%cX@UwDUvWxvvlsNMZXHoy=g`$!852e?M zL>ncannqE3Z+2d(ynEYW>FCCPK+bcux+rL|UWU|n>^QrHI%rVXy$G@nzu!n-Eg0?mpPM-#TS;CoWTc>5oKxY-QQ?d{mu!q4jwUPevM z&5FjXT!;&5U!vIy^)g6aBTXWI?ZxP+NHMNvnQIHS4QW~wJ)(te>uH96XW7(@Jf!dZ zgeh(#jdCGGQn`mSz_cj{j>?bO`ro{mdRE*TZChuzbPMcnpQsR)`~XId7);yP*rm0=RDQ$^LcFg3r;CX zSv`HVXmFBChrl$#|8Tz+m&{lw14GkbmWCu+|0>X9FJOrhB7uE5DQ$gV+AUsST-t5O za1sN(iR2x=_%lR)$uvciKCedYiImY^LBzAN7L-oiv_<{nk zN8&k_LKyet2fqBqQK~hXql)al?oasop_*mQGcKjNvXt&T&Tw6o0pa30Si_Qav3T0h zvO@U{=%S7u~s}Jh-9Sn-=@|wH1 z8uwW@PLETjcioA4V>{6?jIJ%@lK+4doP3*Jn%(j>YQAVD&>|?=iU&X znc;^CgDxt!>li3sOQ0;l3AsH5!~9BZO@flyj+LDlv+m@7*0(dNE}~cNp?d2Z-Ol+d zRldwXA=w+o458-z&HGt(T&^*+)^bFE%(6C>wH}+m{RCa>a}O%y1`B5yEAU&)1i56Y zqJk+<$K)NU0;o`02hHKb%|M()+nXhJZmwT!)|PnJM#9n&T%a0U3z@v~TIZtATTt<( z4p1Ma;&8ry$h0Gw#FBN>L%x4Vr>A|7;5~fTgh;J`2=qj@p(Y{g8kO9Ecs%eFG8HRA zlR)ismMlw^+%M;2TL)_vw@!o@_~728$xd8prWn4DoMb;DK$AE>WUIgfC9lut7gb7d zm|J%5tq7C-2Oi2f)kO_^6&_IjU-QrHKAAR7p*H=0FzV_GYwmbf!t*7Jb?0pWbK?d3L5Sp1<4j1}&z|g=XY5*}Pa{LhodUF2HZQB* zlkhEnkX?rEq;^-wsD5phj}sZ#JmM;^a}O84#V@7rRDAOb%$an5tDIFaqJ0y;%L#GYglh4VuJth|eaaqP z)6czZ3J0AIpz=mSj=nk6p%p?tkh8(>H}EjOWNuMu>0mkbUB#x5Pq)HT*Cz&zWeFYIe&hwpXjGe0(PW0?k;EpTRBBIQ%C%kU+e?H-hX1Xq_qt&|1$3&0#{wEOpPx{Zg@N%T!x=^IH8 zz^zS9-@;!wvd$fNX^bMB9`;m!93aYnn0B?E?$k)rKAtWC77l-8Vd_DJXfJP*As5oTdqzlYcFeZ zYt?}DsAIE|9|r;KqX=}!ACW(i+78TqK4mjm%_MDaFp1J+9}OygJtd1R(=6M65YSDK zDWq!6{bUOoe4_}i_faK*)#$-sZI8TK7aLVbsV(2}EOw440(WCO_=grP-=({>Oly(K za&sgkWzXE1l2HE`Pck4RUrovYdjC~oIfz4E+%oPlx?==i*88j=GyPXe+0ld}iTJm( z>t-XGou~5@+9=g}aEPS}t5O;LqLuE!*Htq!Bk6gMwchf7!Pa&^NvWnm zllk>netA`U()rFkeG|s8glFR~pEToC8q1g&mvS{GpWBQw_a3lKZbrKJFxT0^akI(q zCa)CgjYvgkX^9mO7Rtgg12Y(U9bB~TqDmtk8MB>cHZ91xbSfOpB9xtQI_{<%;;uR5 zVA*YQ;7N0&FfPE7TGIW01);h}?if@kX0`L}S+|qhh=+Jk)uB>%*iudu_!Q5)UaMC$5BX;fT@VIau zeGGB|jTe7TSl69@jgpqrP=m8+z9E6>+t%R9mR87Q8!ots52x3DTlpz9DVBYW#Z+*^ zYSMj~9{mJPPSdjYbDmt6#LuRKMO-hvXXP?dFD1FxPNhXLLod-H67_`$D`e=WBkqd( zkliiV5&-kY{%|`Yc`m6F=NZJ+{gz0V)R50MMA%0l$6L>TZ4cWE$cuY=72`O@+_i4c z)^5-&G-Bo&0>Scw9&fb;qFY9cB2JRY#E_0rWtuF9?Q-{&{u4dseG0gu8`ofIUQ3hT zxFB0pyeED4yE&5@7ZS7NL{`T-fkOR_@mjZc@m+iFVc;!R{EhDR7WR-AqCW~UPkzM) zY`W!YQ-Otl+2p;;R+c$0cDlQy3Rk*9FZxY-y35-+W@a77|3oJBI;0h!w_HVLXL;SN z`~iR|beBbaF*~mfZt#tS?|7A2W5hbO3{Mbh^eMQ+6(kWo^l2J!(D8TBX^&bnr3>ui zf_bYo8#%I72_am0_OdjBrNtEiclSBxDU+3G?epD#z09C~Tk{c${HQZ@`|M(DLr*Gi zPv|`jBk*eX1gLOg#HmG6CwMK`h5_R!gS|I0{a%dMRnvtbuVxBmA*LIT+2yVs~e9y9^6IwpY_6J5@-xPnvvfcj{A4e!Y6bh>+8Xl zz2eHbB}Gm~sH#CC;n^vSUT_%Y%GJ|E6c-Yxqe>X4@yQUvO#x+bu@0%HAptVZ5`*ie zR5n$1b^46)f>3NMrmj#`Ok(M6U;=*_95pRqn4WMg%e3*-2x(FTX0Yw3U@>FlW z#C17TV6wm@3->cib|J`Kd~v5tn%CM9p;*^imDlX0Wt`7JdaYLST@~$pvzJHT#vDU` zy7We?ebU(vSKL@PPL(E4GFvP9UhUN{`0aF8?!6hY`0^H2P>`zF-$Ra8m!3duLb$Sz zVPkm35MYkJSQren?NzQ*~yMW>4FUv#C9=N%WQL6IE{ zhn<7WK6lraoVH=y3kLSe=1^6~MUvKk$N#C|tOEZcRYEk0A!-9cR0!+!fk@EtO0q+s z_n_02c84e3HQ{=!DIyCuo>m#9TV(hI(pQ#qC5?Z*R0RI!)y^P$m11_z%B^?>zw#u* znW{)yH_M^=@0byQT)bN!~BexaoCQ`CpMK8-(Rm&(}L^qxx5oPm&gN*M2dnHMyR zS#+&Q`x|T=Zw@^59EP-=(0hVMX;KiZ9aVpe-VRX=3zdZ5dY5^JIpKS_0mTWuQYz&O z8C>zulT~`xsd;Z{G}ofZMp4DDyB}Q;OOC9|xXd0GdXcf9F=ci|NuC3i%=VFXXS)ogml6|RSbXSXk8?kA^nC-|0g$g^)l!3nK9-`ke5CNll5b` zzV#7iN^(!CoZ1#`UY&clrW$F!&Dp(Sv^#aU%}*;}})b zJKhhVIAU8C&V=EQQn@pK0OLJkW&BB>Sm}(@NcI~r3e+tymp2SIh+DzCKMhQJ>gn7; z%5F-=s{iha6lL+Wn^S`DH>YM}qvXB3M31vI$COki|17YF<`t%c$O=?(Zg5^wqi+0o z_p^FC4a(#U&Fy|J$yxY83Fqbs=^l4HhsW^Vr55e%DeD%4Wsbamwv5EL?Z;(VOi)$i zhdApc6)%)v7%|2b+kr++F0WAq<66TAXjbn%WooErTJLnO0jX3-NSDi4+j?Jm0Bt~$ zzskh=Q>b75yU{?)X>UQ^)qpLHpcdPh(QcqnOojDG%deZ7j}H%`l@$nRcFN4mB){N1 znkXWR;4%($XV{?lC6%sce=IMZ>{PSr?i}$ZVx=IlXVlTQsD9c?I)34i%&uVmA;Wi0b#s^wwPlm$hOa!q zjZgo}%{uUTAcL8?^JAinDU&~FnD)*LhMx<>(>ytLVDNvF^$!0<(>7|5Mk=(FbVjVAcH=SMiL=N-( zn{VV$iMKs#xYAV%URgPiLSHya83wMT3VNDl(>JCsUP2e&3j!*VULDR>c+aRPnL9(N zKiEknW6=Ovi>WPQjI`x_y;HicS+;;xq=3n0S>sbRAI~%He=uW??AtPeuRG{3&qpCB zbrkdRKT#|=IOMKR3K`iLljN(l8~=t%r>*L#ldfe$uRCIkh0}Z%D$_b@hsQhIp-dMT ziSYwntm?l>@!*$+`C<)UCpl93OUS1;PdX&ZJRU`seT%umXf#fipy*07>f^M`$yXn~ zf$^56v2HALf8T;WRj7go9&|W*2MlK~oT>)r8*Il?F|_zBkJYly;vp>uWgj_zlyDG(s5ApYJKPvm0-fT@JX)Qe*mdS%|xxE<>Dasl^DK>-Cq9+ z-cT+x>4B_i_Iy9U0jOC_E8V1d*w*Nwl>UvS8ZC^AC*7Xn9A&Z>-B--GSL;=s#kwDw z!rW|A&6k%V3FLx{1 z)F15T^aW}r%j&S(Xmjm}vNi9=W{w1+qC>Z`K0WKm1280ls4Ay#a=T8O-Ykm;3ybd# z`bxu%f+I`FDG)Q-LvExU2DQRi^kZq3~{|uhkh)&7)uxmgoX=kku?r6l{E6`@mQzC(th$&>E ze<&cS)zh9_Xr% z^Ib?){t)c-Bcpi<%E+B~NS5lvHR5%!H*DDR^7PlG*up+_{QDzWdAo{V|?Eur`=D#tsIny;`5y_2<`t%D_Zggo~l>|#wns0p*BH}*7ZMi$E6f2qB4 z_MY>$<=rlMp%O0Biyf_6E4=FVBJWY*N~P(&{A&FJm!qGFJdTOj`}+_AI7j)@0J4gF z8*;05eSC_G`UG9mVfn$V`)i_XVuU;?t>ejF;E-lVGW(Le&}%s6=ghIHc@ju0ytLMk zom%lqva_G?;sPoA2ibYJoDK(Ne-#r)EwF&zG3oJGp{it+JOq}KC?rk?sz?Vo_9LC7 z_1;(1YN`eM0kpjfGP@nG-#CRH-l@uExZLG+(=`J+Jd)}ODqVGOhQ?-Rxw}o!M|b7! za5*^0#CkQBB@}S|dbSqN?X6WEqdVnQG>66;{T=Nt!dbQMi$j9R9fCUPf7cilv?{uL z-_AS^q)#Yy-juNpQy^16Z~E6)LATccCRpaX)#Lu+M{)&;*P;xgWR-R9O$>AOT9#^o z0zhS&xZud6O7TIZSF#7E*+bsX6y+^sf8geJhmSMl(nI$JOA7_6IG zmb`_vc|-Gd_9Vpp>NcxTfA-iFsOj^XulIA%K8C7P$&IcBOly_TBR~nx^@x(O_NSpI zH}or7hM?&Mqn6}ox`J0Ubfsd|;Tjcs<|z_dBJ>R=Q|&R-N~WrB;~vlTr@w17XGf67 z;7$rU&P$csL#webiauXz-o=OHCH32wM@li5)k6;lyqdQ}~Vq1*H86Mj$K+DXtJrPn&d=p_;u9-1o}pWi8)C+pj2-$2$A_ z6FCE?fyno&Br(Qoe^4*dTjDomkDN&9mLL_%w}G6Z`(HCV$IPjwej;$Efgf75`gpt> zJP(}-%7{qCM9qrKK4(@1p1S^gvc{JqqwA-hAbeZb;r#}jhK4<9t5-h5{$WG zx7xvE;Qc@tuaiXCEtq)Hvm@_S?XPia9750yBia-C?4mDje+6!&u797|*^=qs^=uHp zj3uhr%axxWbEn;rrA+Qu_+3bRX+8d*g+zctN@Rc^;Ol zdBprwe$l>{Ov&;rLfjEG=26ftXO^C@FEql9o%?jw=nGcYc}y#HAig-BPFK%#B&u#d zZq>orX)?g@`AXid=k&??!Y;RH&a~U!A%|49SD~`_f1>kKGMT$}ce2$Me)!rysWk0a zl5S@=XsqV*Eeu-7e?j?ykB;HAAZuwPRfADz9}0__*KoJhmRR)kS_+a zWW~4Qe;g^9obC3f2pwJLCjjzfLtWyh_K`DKQRalDE_ROVt6oG!U&Cox<3zX9RW&Wr zp;$IU8C-RPS?bcFDpzEqO1b`cq8*9=w;+L?U+AyF3k^#ZnctUK3g%!}n-yb7-Om*^ z{gr@KVI@@x{?12^#xP-lP8S^HOEQ*rqco;S0& zPKZK%uQ}C|Va0n|^iXp-QJM?;tnB+E`5O*V#SNhWhsx0m)#`j=S zT&MabUGcVk13}(21LW5`jM4?v1bY@U;5nGeUQ?sHuJzpY$N2!ILEfLN)KD@bqk7n; ze@3jb1fdi8UR|P7dA;LiDkVRoYJtnCIRD=1(4vT#;+5uxd2t3(2sNNP_ZoVT)M zKCB(GRWhw>#Q?N?_nV}rixAptv`S50f9lUmwY#W2uXWrN+NnZ8Hg0t&er-u>{sniW zvY^Ud2*xdDld8TCu8KgVT4 zF6y{qOJtz(^_oS%ucDy(AIyUL0P6h8>C)`ORan~O;mSP;=K+Alfa0jM^g#Bne~{T& zGyO+UX`Oj(lf9v?5bn;9#oG^ePvq!#oXBSn-%K}GSe6TPr)-E|;PZNqLFPsL&2909 zUY6%)1S$E%++RTub?CV(#w5RFp62N*%F$svPN_>tW0_YDWyqMMe+F`lGm3TwD))MoY+)LPip{NB>;Ll1=I{8Z&Y%DuaIzo8)XrZ z%c+*(;`yH0n5IkTjqG}fe`h^d8~urz(CsL@IRiXy!mflDUmUc?>OzAz3!^=d4>{#f zrzVpc-TB&JmWoa>C^is--F_N>3Aymyzs8i3PM*lKRz=u5$;BoHl*Nr{2dDW)OlY%< zPp~PIS)VzI#RSh16i%LyT`!v0Fge{`H0+e716icHdw8@oYQV3Ne~VPtI#HD8>&NI- zyw_}r`qCI^3Wy8(Aj_Jzt&zKX7xsRUQ;Y?q^EkRWQ0O}I<@v&1=xXrr{rTehA9`Rg zbkC2DDV*Qa*@DKE%yr!_giBCx+A)w|KxE}-zLK4xt!d5hY)+{89LD4&3 zMuXDZC;gDhiIu})f2-sY$18fw?NDQu+6fno8;5dtzxsU1+PKePy2dX67o8nD?z}K+zeXSE8#l}wsjPIP+x0U zyIh!7Ap*s2eo&n`>F>sf*l)zeuxNQ9C8TWSjR zUp{z!b-p8NIK^%|g~&_wTrwf2)d}rpOsBJe2GPw|VR~0-LVPD^k*|9gjnY`)#cDT- z`7nOo%&9i1Zu!QNlIbr;YRh$N!|zwkYpB`-jXC4Nf1(-jP0n=)zM?K}CxBdA3BxpT#o*P>?9dIU=(R*XmD< z(0EAAe=B``CmHu&6gxusFA~^zkAIY!N~m8ni1MiZp~Svixm|0$aeUS-7cE=ZRo8n%PgqjWFOrc zO7G{YIK$N=ITE`hn3YCPjvA@M7 ze7Gp;p}6asJCo&9cH@~KE%KRqd+)_iYU4L;t-#tm-O8~h*6d$p9hn2z2yk@f0pFI} zf2(DXlF=(vCN*2d*WP|?$?!r1$BaUGuqKje2z78QUrw01DpbJMC)@Le0-t_=nNwi8 z^QHY6z0P5FMYaYtm6rGxS&v6INso9WZ*FRpA_0T%A;Tdce|p@Z7T8plXFbof?3E0xSh3VM=uSIJ7NvxJ ze3X45TiG#IBPW!tvzlT*O>fC^OG>TK5Ach5`b|iCBu9BYVTyniM^YxgVOP$SPBsk} z*|w(_SKP{H1(B)Mul@aaDuT!<^UFQ5agvL>aRi0*c4$PMGH>}BHCkri=Ona`vlz1frQ#LXphsu1J)6707 z@+isj)@do$`+$*3$|0{yac*D94;yI8-Bc9NF=nFCyA8A2kF8dpe6XJ)iKj*0j(x|f zW1Abl#wI!SB!q*f$PCTQ4cRgs*vWm``Pr#wv&toHjiO^bX1_)#~@O5W4w}i;2~i17Vao zSr6_}>0nld>*K-b;~G(_5iQmLtwiwa(EzJZpNTFCM^M0*Cv{V<+(e8ah_DFFdzs^>8OIl4~6){05vg->CCGg(EwL!oWz=U%dN^#UxwWniw?C*xy-m;&EEMhMTHv4^EaAA(`d-a zmXa#rd`c$Qeco%qe{9nAYXub_EXU|&2##<~UAPqR!R3PJ0U7!~-UVP_KQ*@W8LQ9x zqw&J1fUeyt$Ldc-mH-M{WZ{(xP*qi=+NZ+gmwP{(bhuhy2y)DJSQn!#i{8=CQtGZ1 zRYW0|vASqKJ-&APAJb_eNyi*!&x_oG@9S{^D}Me;tNu_fe|@vd6X!F-RKKK)p;v2f zV|=_ECOH19W(~>^onmOHOie|*O;y}eS~VuN4~CwJfjz>0ONez>d(agst3KauJp3_&E9KMk(m(;{NL=f{z@p&Sm%htf!73?9zQO>&mz1 ze5?gfQ@Fv2|KzS|rL(&lOh|t25Cs-?4-?({4srb%_gInd!BBZb(#)PiO#V!aX3F9~ zG>*5EoIxqX5BMOoJ%qd7_*NC!gM4N(ROnxdQg9R55FtQW$K*TH{wXwM$^ElxSO%bD9BdNYbcXFrs4RoVuGB(E=qaP zH|AZ%^X2TeTVWfunx*ft5vxwa>*k6$@i9%oZU%7W1S3NL8Hy&KyQu_9RK`~Sk~lfC z(?qtR_EMJc*hrXTP?~MZi&Tk-U#y&J6<->me{W3>c`*AfVf-WZs3xCu50PjoL3Z9n{Wn_j#ME~nh zg4*ABdaL^Qv#6&2c^Q>g7W0z`pFeX&+ErFgo0cWv-UyEM_8J%sXQjNa*t*G))6oq= ze|`3TdUV+^gbYC@d(6ZNI5cqEODb^#5elRey?9;-h`3N2u zxoEbU${sR8Xj=&OfQw{+#3}8Cb>UV43C!YUlalDz?gn61Kqyop?r|)`)CqpfJjE0=#6U?t~Bh1M=ms&;4`9X*;47sc_MU$iMd%&wtI?#JRje=9#u z5P$?k?(qi|z7Tr}E=!sSHlWGj>^3&A9V+L~o{9P0@z9jZQvot{&H8y%BB2qz(l3rr+R?P+!1@kD=BvgX^0B&h(HiRY@ z$yF}!-x_J3rk}R>v#PeCCRki^f9qXTH;F~+YGUa~64GfoN%9tUU0_A?=Xam~0%qfn zKKcP{Rg||sX?ZaTY-A_~d$7hf?=;Q<5Dg?|h5mFNH#@3y}5` zSJRD!!4GA6$Tr%KZ)*oX+>~plv)1Xj%K+I8Do=b?@Qn7kY=SgbWd`YAe>MA|<&;I81|{acP?H*=_Dc>Lu#EnG)U&N|-~!Ka<@`(|^Wc#aaaj*_4Ez7(?# zCws~WQ%c!)p{8{i1?lcUe-V|Njb!p&_mYYYeNJ3yma0u61e$stBvP3Rm2mxrKT-DW zuMFvafaOaf3HMz_)^UY%(zq$H4dbO$v--VZp0x2}_d(zX zpiLzLy_NXMIT#y7VSFi3M%tD?YKEHKfLypfAUpEhAcLmz_1>( zJOlNnnk%l6H$@mU4x*d;(9;@tWd2h%W?Rk|Ed33^oNzaOJRBj zZnEVZM?&-Bl7O|u>J1ldnt_%$0@H+GCAhiPY%7YUP&YFVgwLdIt5cKCBUO@3CQ zRIsa_JzCN}4;{ZKT3Zj4NjdEd8KPc<(4Z``s#)UIXGNps zf5?*0=wDl@jWKgYm}&<@%f3+sQ!B4{ zS_#E?diEOzf5XI$zzE(|^Xhs1ksGwVJvSaY8)tWzGM4IuxG1^nFl+R5CW-OxyB{}0 z*W?+L(uiP7$rN_pV&)JMP@~LZ$tuLS@h`UMyhH@k!^HdvXYN=Tos~^ZW&!KL-Ww}E zewO1EL5EHG1>?;dcs&AhymFW8*qE0Q6=v}Ka+n&Lf2G4Z&1Vam;R=C??OWiEqHh)r zh2U+)wg&;_dda}cu?zVTlWm5lZGO>AsQTR0=n!ee-R0jXY?Z<#VELS7pS~xac^0{* zW-nKsA`4_DBAzV3I&Mf=nIp1~FuvGw=m6QCjmc5C`opR<#+4fc9WOJcqxqa)8#OzfH= z75iF~8>5GQ!KqRNHZVatjCTBVV7gPDyWU3suLxIath0ExTzt5wR2*E!EM{w{2V*vX z{`MoM8(7rBG6K}9Hb{_zOt-lzwbSmfry8$8f1k}aRZu3!2(Yf>u>;dyI~!F;t8eG4 zifnxLHLOi#aS?A%-ge2ZOhr}mG;FH`AC@h!t3GrQFNq$r$h?Vm!;ZO=O&&~3) z2*MOlHpx3=>CjksJ)&8@+l6w*xO~R=j8ui3`5;Vga<&_Io`iwKXJ)vrCbE)?uQzH+ zfBlf92>@*kw&E-Z6SW9S)3qYKo|h=MyaYkJ^dt&W!r@TSxsq+YAh*eFx@WuaP(8z;)g(gveQO+@X|tX#QPi>utr`uSQ%*fgQp~e363>TY zXdS2VJ|Jrt5UfQ@HgSV~l78cal_jRW2qyaBipN{u+7eyC(yqjZ^2)VDo9&UU!Udzy_D6-=w zH&NglEsPg_vGDc$Woa%va0G2^KWc;0gy$|8Jbs`}1FTbg;j7O=pY9i37&cf?6?Nac z`w;Y4U<7fAfCs}f=DL}Uiu4;=fAROfj^#a|i3qnWfz01D>D1H(C9r?iSRjl(n5ad$ zpZSwJ2J~HL)eNm&BF(&xSBc^nXQ>migcWUTi$*b{-HI%&X*BAxG^in7NOAr)@N9Fu zd5rX2qveHJSCsD9ui%f*ZjvxJM@y$AdL&RkcQHP}0LS1KoR1x1ky={LI8i%l0buWO{;JxNfgT5t z`thE}Fb%bYb=E8D_Vf4De{sS24r7%ErCG@rqn<1E+epmK^|&g?6(nBXbwnY?8BN=T zgqNzlrMccQswt$3EMlwQCz!|*K}smd5Agd<3kL3oQ1R=fX@6tX+YTeGf|UeiQld|e zC+1Fg(Jo>)s^^s*NkdMxcL5=e_(@2)aNb#-(tWmgzi2psZWM<5f9AK57ZW|`B7<(1 z#uZJgxy8tHx?okrL30`cwIz;|gV+Y(&=ojJPBw^03Co;fz$(mXy zTXS*49A`9qWJK->#ww8BU{+3%c`-^*OLT=>so)%gut^o?7_{H>1y|mCPnl~r>)K4L zfKzjJ*Ngd9{Z+~)eoI*AkJMKR$f__89c| zxz+;_cxgM>q`19Id!dQuH}MM~TR-I3W$Yrju3w3bR=#gG<>r*2>^KZNYJA!)+@vGd z4x7meS?V%e;V^F3H%$JOG5o5)ra^LmG^Eaw7(C*Z!z8r+e_{8)s#9CL-+Y-x}*(!SsdFfBChWCGo$I47?GH zSji~q^946FPduX-ojsXvkfBQEo6X}3eE@cdqhi-DkQzQ69{=RRw-}Xm$z5?>q2};b z!6zp>e=~e}gp(qs3|5P7_O65AcaS8zvj}Uieooh~qJ@aE;v%wF+#bar9-G+daxFe# zhd*FUsn$v+SN0rCP#$q*l~YW4G4=}`5ns`s5y3qI+cqCOjIRzlcpe%fzYA&`??XQ2 zR?z<;iQ09mCW`8971;=;=ea%-=07;KyiThXHX=&R zD$_i@9a?A;){>fTiTz6U*#66$k0hJ$EP3>!JIUlr)8^qXTnfh?tnG`UhWW(gJcZzWV+r>FqTf%_)}5@XYqW zKDS-bu>Gf8Q(mnfEal@8`0nKLiKFZ_yQMmlY{eEhNq@CRkN0{-aT{iiK%?FBJ=Gan zW%F!R^zMh&Mo7fP`9;sk_mj(RHxMdTe|dxaq8l|_(K()Ok>HPXHmHs8T5Oxz%R;T< zR3{y;=4_hk${rNMA(1M<#sRD-rBpAUmc(dOJY}ovl)DV?+&XXB73MY?Xr2-=0MevI=cf;aObyUY@`OvIbuiw|6=YDNzT{5(Wb$24({9RS%AyzNzU=t}GKhe;; zN5M9yih2@rG(0y|TJNC^Jph%0f264@zj{1wJ9utO!hH79aZlX3TW==_@m#;>MZYws z4(nvwscgV#gT-SKW@z-p^hA*^f^qm_wpC|nt5u)Qig@3BW_f!f3Xz^RW*ue zQ4nEQdHjUFS?3)e7xK*aYEcMoB8BkldEHVIP{7Wt`=NN@%mYcqk#hc`nP*&jcAa zt9Pa)>+j;%OJ>Nhx=;Gxf0o?WCAB4!Cby_gl6S1C?O%!;_5&hy>&~liItyg;v!efn zlXTt5bs>S3_s}0=g6;40WZ$ym3Q8sOA>dA(26I)B8M6o6mlw(u;9=rt$7F(IWkb|S zN&H2|iaZ%@h2#=PzmOEsPg*No`ZY9pf~ht>L-1()1Y~mkmnd>9(892NN9D!Vtfa;Pkr{W7Zu=Z z*h`xdHL4HpvS5LHmg#P#GhDC9NP+1*ZoN9&I4xE%Me+Dd+uAC~9Z{B=1kGAVp zG)Ikna4Apf?C$Pvz`I9TFHd=Pdt!n6@lH#L>97 z>+_-zZ~57^VeR(4Q;qHAUB~+s(q{s@ooYTQmMg-X4t4$F0Xf4hjW|yj-g6~Stl(}B ztY;(*aw_GOK&R?7Jc)M-;V!iZIf-J2(u=p9)NMD@e;7)q>Af=!YNoN>_VH$|ke@Dx z;gb3xHOo0nDdwqZiw)$q329YiX}YWv1gy_H^t)C&)D|v}HYcKu{A|al^Lf*av5~g8 z1J@v9f$U?wFW%Az2bpIr%1>oJ_!jeR7|3(xP2qetjp_(k&#@I#%f#9CvfI&Dt1YtIOg0o9>nB(}cM8mwVPn2@dODOUFeO|pgDNG&Z0pHGHn>|+dwi`%K$M|HA^lA;l)6MQ zrFTPes!xP%JBq|BDXCrTBoIV);o{|Mw;#RwfA|VRdQAucN5Lr0e}P`(qqeu&U{mUl zphmc)h^75S92ZP>kZds7tV;YJd|Hp7JxO=Pf#&sHWtPq6sqRZ|v&hyY!o9`ozn>Oz zO}(Rs#8SYwDO^Z^w`QfW+_>rHu=O2oLB>}f|N8qwzgz~soZ)lZoVgNc+#XRj7z4M+ zf3-Pv^G?CWPAg$D!|xBqOt4FykKWLG^*t{I^;e?u$G?1ETvzYle$oDOk%#yY7Pc=r z{1KQhhF$T$A|tO+Z#FNp^KL=^5eYi~7sAjP!g`P{0rIqhAnY zGNFp#1soe>bmP6XO4C{`vV|V${C@y(3lCC(r2 zkL%IQV0A?K`Cn9MeJf8xLi<^&&-X4z|B<3%70x^GXVaM*N3CpP4yaU$CcK$imv1S_ zF7*Igi<>X1nhJ_HbT7g_uEoB19xisY$)sqh^Ot`=(&DCX<+0O`z_UZ4 z;feRt48-+pxd;o}J(KLSKC_FdMi%AH_VtRzG_r;>?{|~- z^PHvYW%8}09o#hap7?w|e}lcp|KZEIYDC{*+tBWM@;|)$zxjvC&5IY6hI&KB|77s1 zZ*nA;lk+P~_(T3+@W)sD#9h*m2ZR6eWuv-npWh!o#PJ7%FI*gY^k_!?W_|9> zKN);EU!54Wa-y2$e=_)*x<7H7`FAq}{$%jK4gPO~|J&gIHu%2{fBtWS|J&gIO9m&! zjlxpB1zaXhT)W_l9vC?Nr7MLrD2WB~F$B71=w~K5ng(*z-8WM&ezIGx?dsoUGFk1Q zlfcf664C@3A+mT#*Xa?RU2`5RO|02u+G8i>!1Cwk^@A7k+;`BPf4jL_OEf2%!T#jqf~Pxu z%slTiIZciaw#E)cLA8(LP7eF`+H+ET#CM8EO{?n5=PfH>G7oE*m)0i)1I~o%K$l_N zDWa{U>>RqTwxkmEJ6HmkU^)No(tgk-@`JNGFoE$8n3@+7hGuyy!}YrL+fY$VF#$7b zJ3>!JKF0J1f7@%ZOpFlQU6N5p4E_Aj$kZ<@pB{amubELIU`uQtZYh>hfZ(Q%T(&$vrn|G)R{+`KkPCTSUuiMn_JQa zMGW)^RbdY#(mYs>_E)1fXu@Z9d;MKo-;i}1YIR@vf1}9ByM9ebQ8D*Tk0A35pU3W3 zZS#7?`Htd-b|CM;!t2r#7u^xW`ZjxU5O4QO?g>MC-@|!df?Wh35CF`fwr?k0JUv2- zUu`|tIul~6pG=HfeJ9K&E-JY{w*Hfkx4d)PK9L$hTAd?&0j7?o6qkFPRnaLle`9N5 zo?N;(e^t!n*B1ukJ%qq#)l6ygWl;TL4Qo!iZo{EZqZY-?;DCuf7o)&I4|qtTa08LB*k`2i*8&KGJSUX zC-C`Ini%?)pR6}Jau{b~a&mP&QxuKYM@#$i0IfnNgB0Z-0ao&~U8|)sxCkAyRHVT` zqKFjw+q`lRom~xo?f%dCi0~AbNv;WM`}|hRz&}CY1fn-~T`q{38@OvP2R@rmgNK&) zf7Dw9nmNJ8DC9o0q z=38}pM?0zj0AM}F9j~KRb1;v|~5}Q-?%7^@p>UCR7e-k_ob3bi-W^PUVY-3hR%;%!NZQrhaEcFKn z_B@QZ(w20m_+*#dckHQjObJuR&ULhFqj;A7f?O4viA(* z*9?>AddXbliqq7}c|KFIqcow+Lt>!+W3r`zfW4n03TAuxA?Z)-okB*G=Y2Hqe?2ZL zArKVCeT;EEz^Ni##k#}sottcw;WK>h>2Uqpmh++owZ3!uC{c_XM~7Wi;2|9hptGaG zHQ-;ntV)tk(UWlsoh?+AD*sWUeNA+%E0%18x5$f1l7Y8Es0hWWMx6o^gm}W{K`C+BSVsv@da&|Kx#$^HNAAWQzKd z5+KvSscUGiC*JTS!npfSR$cLwXg`7RFkJyfdziCLy=)dwVi;GpTiFIMAl^(d6 z5%9qEk9P3nyzAClr}fj+t9K$}=e?-=op@($Q=Hz?4d-8$1Qv zHdgvf=C;}4-QBLZKC%5tf0L0b9n_)@O<`3(@zdn2=QG&p5>&RST$TP`n!$9#h~pm- zDk^Rwq1w{V^7|t~HC{}lNQ4bfPUb&inOnpVrka*wd-v~&|9j&9p7_5f{(tAhZ)%me zH2%EDYo-@UFClJa&D`?yI)$-uQ*+AWKX^>;Mc-RU0q2!b9ue@`e{FqH*YLzduD(1I z5nxKlM*!XW2&E0Ju%klgLs9t6VT0J~f3OTW&fX&H_?5oQ95#QYi%Qwvy1)bPOsA&; zCw#`v=y!_xwa(#0)v{>N9o|9eEwMQm69!}vUqFW+GM?YO9wr0 zPpRI+B?N@LlagG6aP%IJm4fCxrA>YtBkoh;xXRUhwm;ZelFMqNmI@%I z60hDob9~}TU%fTi%RDfMb!Jcm+vROVi?nnDE#y=<7q8Ol*SdiA#q|DYo9Zu`w1!8h znzWeRT-TR+DtuiC;p0ay+Ae?7TTSerL+2Ia^#vJ+wXu384`;gx-hk^u8vRKcSzMBV5upzA zR!-bNMd#JYym!|F|0J%i4bt2bY!HxMx-zudpAG07td%*5wzX0Ip4TY4PDp=h-1zBw z4y@qMrv}TzjFy^oiU4}#@`2@;Co_XKW!nau=r6N8e?0hK_b1$<;B{81e^Q@#&LQZe z9r&QhrhFGPaE}7xX z03gQh@1swyPCMA1YU_=}IRo2OBE;yfae^8D*=p+H#Idv}tF@4i1A}3-oR4P)St6qU z1j)|Ne>txU@o!qi?3a85Sc;*;{=mBph!mPgqPO8>Lb1>>$v^lQ4dzEPME{D52Y%1v z=zRF*5718J!Ug6Fu(GNIcRv!2V?v}6g>bop)M%;*x z0G+R5pZ)+Nz(gn~&d5EFB)PqD{qY~5VLnkGf84ca>yYQ1QKvQ|{{yI3Px<&Unm8i^ zE206OdtUp4U|##zmZ<%|vL&&U_D6Fe-1GAh5(!wB;<9#b3|3WF^l_>u_Rz*>f6P2Y zidv@{tLHg>FdLY@wLgLTplIMW3pEZGHY@(T6klV&-EiO-dY_lcV{eU0;_R@$3uN*_ ze-50OY8r6g{Bw3Z(9r`l3;ah%Wct^tGc?t{ujUsNhm?1V`_xRF>?d&#p-l#JRZAP? zQ0@DIXy;6!`HE;vHv$&Moa)BrU4?tC6Rj+Kcra?!(Z!Z=m<=BRq8{EhgzSN-pV#cg zVUo}-A7a+{#N^Ms*_#m~dg5YL_MHD@e;^Bre>sQhxpPdek+Xuoe?e!`b}q0OpU<+o&8ad3`1*)e^FC# z<=R$jT5v~kHJfNv2iGMN|Gg1TAlwz?QO9@leKse4(#hw6`{1`iAdSAgQek;FFjiA) z?;ob=E5P)(H)&NyT~rF=KN#o{kJuZYe#Rczc5(@7Bi0_5 zl){D%_eHVcCRbCwQVAdBHkmA^f4Eo_x;3Xz*ymSssXqRPl<591XZ(4N51_rk7?F4C z`B!3V3@$3&#D{(aGzoP7LlAtg_zf6}6G;P|1>@G=SN8;`LS+~rFB^{5l2*{QS^`Dr z?GA1_2w~q+#s2tU%K>c&>g?t&-&N}?o(|@5Ss$&&PobJaxa;puL-6{Af7Nr_rFa8R z%6gf^({-YeLwm%}o=As3xKGAe{YoHqtXT8V)OlkZvBDqCko?@}jo0oL^VU+5U8}BL zJ-x&YdRmDj>u&82ky6I>;E`~p6cI&l*-Ar>ck0B@(n@2V%Z@h;vqPP5Eu|97BX7j% zx+Ema;xh28F7_+EMqBv6e}_TJ;)ixwmRHPEY~#MVrn)=s9fCW>2{z4Kd7_$y24Ln8 zo8z4X^ZCVi!;bsj56in>$aAD%0;#mLxt6|~>=!ySUQL$kJ6naEe~F|cNmnOLPc*)a)pa6{mNJ-h1E`+2-f}-M@{g$j5<^u%=nXI+G&!+ zxuM=lLJsyXPk#c=i*z+*hhgr^Da@WrRt4}V33ca_lP`}Qem(a2ynJ@@rf)}rFsA3Q z9prDk80*8+eAwpyf0nKgrLA1&pEteDiy?!~ER36W3>5 zP`#LR#6y4^+k}#Df}-Hq^Kjv#&qS8oO1V7X+lE05#U)X3;>WG+ewMbTP@aGBIpuOK z7(;EI>L6EG+dZxtbOENecSdK|Ky~$@v$ogaq88v@)+Kfhf0SM_cxmPMnv4jofAL@2-=TwEfT3N`taP%hB+mV6fB}B9WQP`x4~|nv^9Al8a*1Q* zW0&F+$EQ*de|=YuSot0Zw_u_YZ6qsd9W_05Uh-LX#S{&ERwU4G5)O_qc4hPQBF>+G zcNqOb%ogdrTdl=4^n|-&G%2(Uye$Qy)Msy!`oCpPwYNE*?`xfJV@~T1=BWODJa$dZ z@>4qB8x>&xI~5uOdzZrcVcN0N1)rM?;xG5da5HXOf6ejbfIRU$MW-d{?{b;1vsLrA zZFOI2^>lF8R^H?F5-k>qDaPhUxQFj=9jt7dkPx`LYbggO6 ztKmL1f6l8>_0|XqtN9eysjxI}+Q=c(DfgvtI=gBSvlQ@fO^t0sv&)!^_ioN{P&2ju zGt_;-X=nXsT4evQbrs>?o8$GUSp}bScI@<^Zl4f}`8ea1@cifbV&LR1|KgPm)|r?b zKDzBD^3uKlx}UyRv2;5NIUhQ>t0p~UlD%7meD-#`gE{(D5Km@&O ze`6buS?8#>MLNlRYAq;8T>wRSCCDtvcKG0w@YrGH@zs_%T^nlNo|gbk4wBE^!uZDO z(}1M%6F%E1ryg-tz5~afk*zKD_!Ca{s!2zQi1`Rn5w$O@sI1jBrTbpeFHd^R%Ox$< z^9ta{Ca2Emni7NNi%OCK$G7P~b*4Aye*`#cnMgYJ(dNZAz#q`ACHAT~$b|;e=)>39nWkf9H8%3W^WqvC+ff<_qip@h*VHtq#Nep0y!Z&jau> zng83>_q5HH6Se?TK&-!&g`$Sl4&Pq~wmgk^8Ao4p*kWxmW*X1>lK#wVY3yJlm0;Sc zX0eW~wF|09^^(0%Lb-W_gQc0mlzNbMm18Y?!BoQq+TU3f~kO1(*I9f6KKW$7`U}HOFl*0_JKXggv9oKHbqK5XRFML zv_CpA-m{^E?TzR%cT3F=FS|Dc{<>!>+PlQDwt+s;0bj_=aS_WahqDrK?*>^(2U(S8WsFK}N&J?5aR5JQv z;Y+do` z?5fAas(~o>I4QR%O@leb|p|7uUo8)rx z_TAxsFz5qczWPZxYOKHjntyoLG4bVO49I?CC_WyNFLIuwJ-F@UH!5Hs0f|xAbky_& ziz%(cN{Z`r2-3-vA+^xP=(}YzReS`dIEnHdn@baop$DVOxI2D*@W<5+8c!TTrqmeR z+OH1s#PVCJXO;Je@Fz)g`)mg?1T6W9TbtL`d*Xvp686gI3dh420Dod61XYeZmfIwj z2f*)(d%5kNEz*vya#oGJ&)T)T~Yq!fFgQ(Hq$d$uO_Ko zwrO^(=4Weo&C8^@=YOKTp{UEriN4t^|2>$uv-D1W0c3wlx*PdhrHQML1HM}lS#;P= zw^*j?h}iz5V5<9)VIOgUZmUqAVRqen_T$b2VRNAaOp0`x#l2@PaMbM5J+HN_JKxmo zVLoTE!BhpJ`{xIP7+crQeLNB5gx;S`52eV{5jM^UZo8$-Wq&5;$O`g678!3KMm}f9 z>6OVE#Wz0cb3m#l^zV9vadvP_NSq&dS}#bO9|1ryImg>rPh9Jgu1)-gQr9D&E*+I^0jf8J+i8X!E+x4-23*0DB-3U&`2F`%riiO|@C$$u znshhuZ_Bw6EtkjYo2*8X+%0-GSA86OG*HTqki2SB0!sE&I(ei?T6N4Nn5astpKmV< zSjf(s?&ovBEML?s?|k7jyIQl+;`bH|B^9706`~5~1Am_rULCiAr;Py=KC$o+nPvachqJV7iZwb29kEJ2s4K=zO90vz;^V$8^oC zVM`j-S$!o(MHR+{Ra3<#VMAIWiHUG13}Z_sG&El#`QOW-V2ZE6UwQqFiq}FH9gO+ML^6$Q5PFMPxMU(ycoVUkm1xkcqc^! z9Gv1h{@B}R^Ry3PeezA-7dRfcfMTVgM2pM|}TePaU9?fLfdnW~B40?|CdC{{H|YEWQD6l=7nD^*nUyYYaU1 z76B=$`BNmEB~OWpaPVUBP)rr2eZHMq{>K3+nlo0elliPK{jALNH1gc#>L*wE%SMPi z?tiN}HjQ8H$z0Q!7bSPyrdr+ZHv3qN2T@j?`Nj`v8S--x=AnCVJE9mWo;h|$_>%v=7w45J86}ypqQSh}HNk9=>*xCN6jb@E;UfStOmuV5DaO`eP8Mb+ zC;{qKEp6JL81zxOj_?1aeS<<*l-!;D$KH6f9 z?Co)xz(u84ZF~d>6D0Ny4dEyrLB_eE$edOAfx+`{G6pDUvCOEi7deFf_W>Ns+N=F5 zPpwf=R_SlIe*G`E_BcBsHy-G!>VEf&%RNQ5_U93TE>?3zM$QU#aw%G z@;fAQN*}2c$fS4)zfm=93TP2K#(O?h3>z6O8xpwm`^o|cUDM2~eD?HQ z8)?;*derZ9*OoSGiheK^oU(XzuA^!&!M2kw6Dd3^2hqP_-74&K+B5CaZ-2=#ZfNt9 z{Q}I`F6fVGclBF-jLAUBuQ3_`~ge7KQe9ar0Icz;KNVw6SXt# zM@i7k=Ql^+qlNIvvVfsQdBq|F3>+ovwLiApS8CE)eUvXDb>_{!$X!%*y4A{HQr=Y+ zQrvJVB05EcG4&v|U7th{F4E^E^=gPFX*|`wrP#;GZYnJU`sGwtRezU*mw&n&pPD2C$m7?fUO>>(h7N^TeH4FDbb$=Y$cO z`dUrNzgokE=UXtv;o^{`+_UeKQ{)mi!iuWUJPikP9X5Pfwb*K5Y6G|Q8b|XKh_@bi z4?O0bxW=XEu4N;#2`r^u-J#}iC4c2@`PZdw@sUquTE5f! zP~aNy-Y#oDP-v>1ECyL1UF9a{-IL3y5Tbj!{8R1Y1&P`(R?ZUw4FhEs zzcuIYdKq5%U&dSQCaim^oYywSB#Fbq@}tFPIS>nX$N22H>yPF199KF|w0}r*Y+~qc<`cDzi1W&AU0Z~EVc9nKd^~D}hJs1yGr9P|b9nuJfZO^O zbGiJ{{wf9e!bL80&Cr6KYX3(-jiu3Mu2yksDZ$Pqh<`+z^Jsq~34uv4KZ^}x;UmJ3 z4n0%a3vR>GJd))8T&W3Ll)V0T^K;S*@y1VR?=N?}jemG3Hs?=%MlqrBF^keakN53*;p?zc-AMk@xz<)~WQ$j-tf$%UUMnSr()74t^$ z(>@HISjelolYP->LxU=b4!QXKY!+2lb-GJb_ZD#ML`}XYp(O;x6I6( z9P7Re{Mp72p|zFy^A=n1^1!Z7j~L-?*0;oeUZ%|NWeVdu_~NPT^Od!T%dGJv{R<80 zfjFGp6l^7Cc72?$WpEVp@{JnUvKuA^pEkCC>LdrG$$&zBEK47Vxb z*MI844rI&6ALrG80q2y;iB9plQ1*7u839^77hEu()^p5mZ9Xg3|5?s+y}??uxN#mj zp;$EX`Oee19xZ)3!6|2JqDHr@C~!{WJ5r(TTB#UN3f3`CVB8)+YRG+O9V%5n@tHm$ z|4HPU63wD9N1l;V)WP(X6HJxeVOTx{V}BT}%Qma|PXZ2=M%Jn55dlXYl^qZ<4p2Mj z{*R|J%D>GqMb@V#YRfn?sJ6k4hoFwP3l7dtDF}}xjZ=|6)MQ-M@BK43_|&WTQqbmU zJ`W8chv3c^n9kAz;}Ej}1iyO2gVq5Sc}uHmJ4e9J>~DQ9cRJ}IN1dwMG7AsM?TTb~t|su6d418J3~vwFve^WM7&lr@uqN zr0{u4tX8HqTv0w2k-eYC{?9Y`_-d7eN#X0g^Nx*0MUj?5-aU?!OYLql7k^+cjFmew z1D2wk=UaxPBe48^*-kJ^B-Hv3{NeI=!uRSe1drq3*OlKL7cJ(uDD3k$<6ReWOL?B) z%fK7`!ZofK;4DJp51JhQ@|C-NkKe;4lwMHW2fmLY?xKnEd+A5G2q4t zE4xu*jr6zVKj~cmb$NSWAb&eW$83!%Ma0Eo;Kg|(4dHhqO$5C;B}|}KEB_Dr!7OZj zk5?4YH7-F@^)(0=a1P<{qSAok=(gayOreu6bj_B=SXH!jjr0gSMfWs4^$%wDsCVgd zOE^74W=m)opV>gC#&ZwA+Fuu70^79V4|%x)BJQOOY5vL!b569n`+v{d(9ZIOTrgW^ zA3MfbGvv%VKcI_My{N=#uDdN7@qk}II+QjVU1c5Ltr|@yU1&m=_Ox{29q3OMZhn!% zvD`oGz9?5qI9(c|?q3a4LoC;X5T$U1r@|*ObZxAPTJdp|&TErMfb-g@>Zs-2Ak-f; zkZEv#d83G+VETEmg?|A5$l>GEFJ_}fvQ7tGRB{gzsDO9Xe|stu9@K|iNcl%f^V`F# zCKcX`c?>RGX_MuFtpP75kK6m?d>i;W`B}-NH6mhsJcx%UzP%%2eJ~&?d+ndS@xsNh zYgyL-fs;0dDI(y{@Dx!QMaB618v={h|DZzpKY%&;<3~UD*d2iqdBiq|Q7)-Q~!K?4A+(wFjHn^FH!y0<=xwrpdF!3Mm7cRi+KJowA zU@BrFQFqYKc=KmAT=*&5!L2MAmlM>-7-ItMAy5A?0+5wp*GRL=W>Uy&xpyQ2=a zCHwgbe`0?GLc=wZ ze_9)9R{cR6c?GNQb>sy{E2nZu7=t05No!coT55~^+J8{Fu5(A&in`)?6Zl$JWM)gt zXH}CjTG)GlB9}Y!;9!7=ipC#Ug7eXrMkNzRUf)H3C9&F(w-H!<-5J>;C;1DoVF$Z? zz`!rXCa0%@+1Z|oBP-(wr*hov#Xs0p-!-axqAp&cw7de=?{RFjZA6xTv#uP0j6su2 z`n2a{Q-6T0F45-%|G|?q@BRu*?so?B_6)L%fE9FYJ@43CT{;mwhV}7?2!?7W#=$+S z*6xcJ-s~-S^3NXkt?POyt(ZU(9IiXU)mN)kC7^ZI#?CRrhkOU!fb}H>EJ)kQ$>UPd znI=?OYiyy~;6Lz64aVn>;|1@F8>HJte<7Hw z6wj4noHHf1mS$t(dEUcp^e;8?TulFeVQPx{t>N@0!wDXmYD0Tp zAb+Gx_h?JB^PBvs5`I(0UCQbqAx>TWBzD53Xe#V`(M<4 z1Z0X07;Vn{u>30{qvL1L&D6vUFG;3}`fZ^rj9MGV}^$Rd*_vAn5*1v7>>VLPG_0`I(2qa0C!>oe-z!LjU{ubu0`u{%` z=Kp`LmB3>@4xHbu%#Jtc9-qvViO>ci(*$czhzct7{Aj5bj=xcFl8_y5wJ%0nGcz9{ z5yaha@Ith{k+egm3dFw`oa#9xs~uuRa-x8imyMjBX5Dy7|=TP@VX|u9rN(uCneYqd2<#6vhCTt%!7y0<>R^f zU_B3z5|E6X9P8328{69HC&^CxVI9bANF?xSV#eNh2C zH6yp-nm}Bg-+)6?W5n2)c^W6R&WBku^oC*2i73k#m(P!Q&zp13zp-I!%zxbQ=L%1z ziSOZ32nltv1htM^`U<+%RaC{oo5JV7$l|<;KL#aRPCoC4U7F7Nsii(Y;wN|Jp8iw+ zD;kSjKBGs>ThrBnds4+rD#w^0SnN0g;$19)D7ITQ?`VJ!J{V5tESpzc5&d!ff41pw z{ysRp5_{J^@ww5PV~>=uxqpIUR5&G@`&v+-am#&ko46l}g449jDPBOAE<5}v=Pj{? z?`Hbb+Oy@|;u7Z#xP`wYeNci+tk(|fcCenM&2bYW`rfl7=B+iDLj9`g$*>7rwJ6#9 zEX=9%DoJSzZ2>>gwCsjSWETr7TWTV`9pd9sTqecOyIP~rJ{}--Qh%81w5No?v~ewy z##0h^foYrhQOA%jHl~H4tl+9w>68a_C4JOmDe(qYZPn-9@h36o-3JfdHfPssI=E|v zGSSX_M?4WNlDrfyNZYyVe9~OE^89 z>lD~d&lP=hEA0%b%xtGYdW03SCyZM6O!E;uE81`xwglyuTYu`uplSayvlPeetz^ zv!mg3<&@*jsN4KaC*LlEp`YLWXC;52#=lVYefH&KHIso~_qFug!NXVzWg&q|v{{GQ z{q>3+?ADxo@qdSrN!K3R(%v^@0+^!~QbB5a+Vd;mdzhO~PKviHCc(>wCZ$p|mg_EI zsq9kru)eE&I}4Q)8xM+6TV^gAq=MOB0E`^OaK0hWhft&!asGAU=URKv6d9;A@=jk7 zV!$S7Gg>0F=aZg%-W*|5(-zQ$tdnX7N!X9r#6&pnTz|!)c$d%FXuxD|)c_CmCB>l8tOR3hgN`9sKpolb69TWnh4d*Mq;=VzD4dvO!x zwCiO9-?gps7?}#Xve3j`UXcW@zMpNi$-AUP+)i=+jFzc^Jbh8l<)kCBnMO|{v>8Xf zsfDIbHGl3@7D&#VAL5){x|K&q;$&4Mm;R9=T$D**(*7im%2sf>Ay1MuqIQB(?5YNi z6qsADQ-*)7oNn-JJtb_Q7p$`GM5sMy;Y5wuF0aSiD>hVLCr&OrtLILH`2Je~DYO~# z4w72${VCIA>Fkrm8$why=S{Mu-6GahMSt3P6WY6vi=0a~O)3X3zy`C_yt&e7RZVcK zgYcBN9VqIrDeE4zNf$pR;+BxrLarOTVDm7h76wg`m{n(+^R&XiDY(s!E z0;~unJMU{KoxpmwE~oiK4)KNcSF|zFHGkY233FdcVXj#|Izgp*)7sT-Y3e1>p?_H= z^zCRb;I<|7DcOsULrCq6-1D`%KLQfX_#xymam%TYx*x2J1$)-78P7x4PFH+}55rOZ zQ;H^qxA|#*D^?U&M`PEp=W>iO>={Xs*!&1FNt-^Y1fdcIbN2T(T8$)@8q&^~`fHu0 zPsj%$5P7CcuA>5UDU}b5O3r*Wx6*kZZdok>rCThFQLfiFr+CB9cuOzc6=Q;DY`;C36E`+by5y9Pccof%QSI>!O5bSquwiM+Ir1wx+ zFD1Y^yPr)`j8$|Wo-OohYonS)n=bvQ-H#bO@$`@Q5mgSyS2HN3vVZkN*<70aZf3iF zc_O89$MdyRv){ERQj{vMo5LTmG+$JE8_e?gBdK;tm6lX-JwCwiqFrivDl%cLbA27; zf^NngnncOIkX_tlQY?M4dmnyE};{XJ&jF=q>b*}Z$ zNekHs9Z(J$-_RQnaeoo-o`T@T03jEZxpdxZ6lk^~6pWZ|+p~ol^wsUBT@zYOIcH)n zW)V9!L;3hgIy#S`M;M}(?yGANqy_;)iu&Q zhv?x-5LyI)*D5ztl115HSUw2uv0=MJRvq-SWJRXH;?^bO=S^b>w zYLK?uKFTrqBVfyWeb%Jv*-5<$CF$XHAxlEllnV?k%mkyF4B-};?iMC=W{eiyMRXre zO;_x2qqg(+gntMZ0aa~Dwv`E8$=*JZf&ABn?%1=bGT##Q|LfwlM<4%pKYnt&`X2NV z&>5({yCw(xzDmIPl(Z=7CIPSi5;*?$O*Zh`UCnIEXzACOOM)^DV|$p9iL!4)Uo{G* zs+7e~1~spuw{rU{_g`PFfqPJf%V}+s3UwujPB)zF>wiuV?Ywa?Uz4R9mIVLUY7!v^ zVu0+Ho@{p%SK8g)9zFYbvi|K7xihM=o4vZ+aWQ27h|aDCoZ{%{O0x#6_dr-41)Hap zcL|_7*G5G9ZwY5r5&|J^=BI=oftBxn%;#`f3XEkXZy=YkKRoRyoh7PF+)A^yHuu9f zJVG_#b$?RE4dSpiy!cEiLx~S3V8oA&yJ0t2Y;W)dVY`F-(!A|t!vj6SqX`d8(d4ns zLsCpkmjEekoEss7swXVV zcf@rKczcU*4bmrt^MlK^DsAphxJ=jH8r{7yp<$1yibFMl) zIa)x%2@Gz%0XG?D<1iytRVglOvXRBCoek!sLePjYF_-BsL?adQl8^G>p6P0|@&t35 z*V3_bGJ|vPg&?_-tm z5Pze9JT=A%hgNz8w7{B3a-P>+J^6<|$JeVqOHdDOW)(2t6989SQp7$>w}S<@~LQtN34HMH)n|R*aj)Yb17OgE7OpnM|B$;eY&| z<@853`8^y-Pl`2V!1m~IjY93dIEi`|ljm$88i<6-2(pIejMgh+ZR*6$x?c#^(VV-( zHL#tB!y}iq#^t%|Y3r(bCh_d_O-HFYCd_~~*eNO1vJhb)qr-yyOBibk9w8q9@H5&@crqn z%4u%`aK}uM6tdB6_kA^1%b=V>>7`;Vs4G#RQWS)03zHF}T+miu+4E%`n15WGL?AUn zay4o#MN+-+@iHVMoAiDdEA_su9QeGtbHmWh&>aA&qtCz*>@seC&1eLv$Zt0_2tx6@ z(tMmfVfh?ykT#8Tn6Z0Ae%Nl?w9%ZO=k z9O^J9a#&=MVPh@l_3Ow!la?_8?o2CWktZsG99vHIx_T(t5|hUBvr_^t=ljQC;6zOY z9_nJ5cDg_6*NU##^sl*tbljwJzw&*~k(5 z;ZpGn4WJAacq}{a4j0O}1Ot-nSn*Of^|4}IcPOzdl2p@5z<;>AO^h|j6nz-q4fMn2 zU~e=sOfLdqOoPJ5&uguI{WWsaXz=2!9E3h3|5$8mfp^{i(Rvh+Qa#~z#qDTCdBeQo zy&f{BZYz3slH?ckqTUS7VYKc|iUQ!Z8-Q#VE7j)pt{=!fBC3xgm-QFSWKv*x6NzG; zsWJvYHY(g;&3-2mp_Wj3JMyl?=oJB$x{uh$z}7H9z2!H`v+lIy z4EP#`y_Z#i1q(j7dOSr48)Yt+L!;EP2W~1JRzIA7(SMM0Z&72`V7xn8`iuz z=rrpAs4z96EUqZR1*iM?%v6~(_!tX&rRXosd4G3W($lcN!y_5&0f+Q`+B{jUv4&TM z(@)57#e<9w_Ex)5xg%B#qOPk;zzdqaPKN`^S@kH&B~P$7s`U`Uxen}la$B1XxqnYEyS&{zh=&<*?x%qIntGZ~)w>ugDufFE z$q#}Jw(bzVEr)X$(RVHF^52a1j)a5md-ZYG?aAW??@k#Q5J4xgTPYPT0uRvi^&08lw%V3xwZO@GkY z)lpT;p#_HgSKA3hFVvfUQeMQ)yO5`G~5k2OtD`7;6swu&&iO> z60`dqLg9#bJ7XB%rBcL`r@<6^-8F(o<06bb|MK8ret)hJgids+GD~WH9Ivo!QhkwE zA6mS>6Uy6j6X-$c+(>KG7Tu_b)_*A09D-1%`J6(N;KQ(nv)RCxmz1Dlw?x3=Q?|;j z*43XP{PO@z%s`p;B{3_p5j{_cTE}}*kNms)RnwYRM_M8%UUw2#-j4Xo!J?N<&+59C z75jS;%l;Gi1WgLYa|0PzyAPo(>;@8TLcL#c~4SzZFK&kZF+LkHMVh3{-DU~=D49b^cqEBN3Y7(;Jep0UQ zE_Ml?^>pLFL6E(n5l~bksVA;J-j+g-Zu$sSMe##|#J6>tUt|@C8A&&?d>Mb`R>L^L zFY5$lw~e-(PFkX=__^`IYxFh%?{-RCD7UL9$hq8T!<;hOwPALU7=P2X?GXt~Se>wY zul8Pb+Nk|XtACBvuIYfHAVO4T`oVpRTXuqM?^Q<$L_cXTz2x6pBzIs?qHiB=L*;d* zp=u2f)bk0W8<9Lne&>oJ5LIL0@*xc~&}u;V+m6<9DqO-UvMa5XaCtePyFPa~xenJm?xbbU?Fe2NRf`3;{mwuZft z9>|%ZyMM9YW1H<9bv!&i)aQT@Ru?vX3Cz!IGfG-BG+j2pg3o~V5c{0XD)s@d_{y+dg(of5 zaA*@dYI?qc*I|+dky?#)nnhhj7`UQtoIX5U!xU3I7El}^r!nhFPQ+ZjLBi8}#PWZ{ znMWVLqf}@1MO$jlB&ZkaCC-k?G-2IFB<_~!hLYd&6f^@&1IQUtYmpyOhWS44%yj!X zF)%{<@Nz5vx_`ubmC7WR$d1^bAEV!J-K48>^t}arLH$yIxhTckq1bS{AlJJ3vDU43 zzV0r)nF=lThjAkE&qUl{xjlVn%9beb#aUyed-d zrS_tj9c#2VF(WExLL(Br8oQ|hx9Hc|EYY84UQ;ny-G6zfZ^W9rz^mbF9qn&5#wZ@XGZsP&=@g?d8O^O9kS!Y0PL z`a%3d_>F#viXqa_no`kXb8%Y3Y$|#@AZ5q|`_4o4M9V38f*cw};z{wte(~i_q_F~$lH-FVN`^r9>1`~qUNVDSX7-$ZT)XT>=V z-bH_V^d%F-qmj|rGdpjxy9B)LJU)OWanbD)*T>9`J0;w8x++*|JZ z;ZR_Lrb%(~54-_3LZMx!yOv(QaI!1&^GIVChiTno}o8h=*- z8fpZib9;D?kpacBV;pC%?T=kZIhtK{S@+9^@(FB`g_*Znr46T&*U zD#sdyITg0S?R}*zGkbQLf`8$c2HJ?aAd^AxjoQl9cTsP(0$}As>@CP;PW1uclgq@} zqms_gyY79&{X64Y)n@ThF1JKJ0Q?$F*z%H$vjJ#w%1K`R(l?ove5bSNxMDxZ4BK?7Vs>@ z98F5`0l*wBSLf1Ddr0DTcH(l%rg@gAq2SxD{Z8l2#8 zQl+orL{3f8O*&3fQ-3PI3aA9Yd1H7$m8^=-2U*N`M7lH$jWS48^nqx0tG8S2MbN~% zPYZYxC*O0!8jn;hph$ICZNWqf0AW2d zChXl7p!bJDWYI}@hwFG3(^zxkh%%-21mkJxiwJSXfb80BvD2FI^mcr}UuHQB4Jjp( z6ucDckP@Zu27jv#a;uma{_%2Lx|0soZ;_hjxP}iBYJrqaXMRdF)M#E;f_-}38Jayu z)LWYH2qmVhBV`yOAMHEz&|6qqM91%BliW1<% z(DPaoIVsj!wr26%$8ApAw<|THO|ebECLLlq^vXUrZ!HHVhm!&k%mHJWA&`p(!($!h zHw7nj{yWlUg7Kv6mW!I%!-&YlQObQAx;TA>x{YbpCCHEQCLO>6m$^y%M7Cv|T8+)t zjV!eihJW}=D>QPIOcA`r-Q0&MOAXCtmv(3=Q2bPpZ}gbWgJ}XxY#t?hAj~fwSCRk= zRRx`|f0a730CZft>&A?e99_{7G^urg!#tgFgtt)Mkf#Eo8+v%iP^yU_kaLbcx#os( z(3}AC2agDBMJK{HlmXo z?qDui089qJ(J3HY+Y(O{ytOtGxSQ4XI?6#`qJ07R)LMSfi39R~_o4s4%NZBOnQgRp z>3{hno`7x3;|p)96|t85Z?_w4IFGoER(jlxC9%2l17ZjO{QZqSc_FX+hlCW#IIR<_ zD@Gf+U4~es)r#Hb!K6!|9Go!^?}=rAV{{f|xQFm?IMhQNsGLDXp4Pifi75CfSTaMG z`}T{Ja*m6O-T=>~41z(k#kVjhgD@Ml^>ywN>N-OX+5)KuHeG*gV*+AbOPen|6muG@ zpLHVzz-KbskqY`P{Qbfg>=~c5AemB^krVUC;@%U!sY}9fU1Eorl;xU-8|M@bVSn6; z2r7x|{nepR`wZL|8h}zr)}AfIqdY~R$8vAB#k%!B5Uy1$J~_iDvK?ut4fnHT2sm_$ zz|_eX$B7-4A2cX{}Mwa zqwV4|cnr{6R(2A{j#@F>Um%1@M1L>%&GR}TXbuic(VLUcJlAzf#`qdHLT=&{HI<~* zREymP(NtBYitvFUYF0lxbEJfc40%bs)6H4uvW`l5zOdc0{5BN)i3 za#5hd1J-qFxdCgFd^EqN(Rt(yL=1gHr_!=hM97hi+x#3WmUT%L7_9#ZMt{v+2q~!} zMHc;VFw0w=twDOao-10wLMYeyhR1CYqHo#2zK+%Inf?*=Pf&Z!puIArw@mV zyn1=aTArs34ckL9hskX8J_lXtYWuu2fG4N^WN(6Y88T*&r-`GIfH*fvm(t{HfI+`R z>%9%XOdcmvYmwA3ElB2y&3{`6PCh{RX};z?SQ*uE&Qf3!3j7^%S52Mg2$3xgG2vNE zBKk8~>Of5I=U!r>2yH|=Nt!pW)%U|PUThubp@4G}k?Q-jo@&Y#!A)2y0prBX3DJ&& zr#4jKhUJ4sYLl;yC>@wiH|u>M_fN!qppcVfNb^_;CC@CFwBY-GFn=tf#z$BlNt!U2 zz{O8SCAH~urt`XH@#4z`(d{~TV^+8o6J=2Rc=RrUo2gK7^3s4DQMd2R1zeqL&>+~= z{5;H)y{t0tnZ#kW1K|SbylsA%U%Jlow|2sdGFGfS_c|pxIcp|?Tz{sI~b#2>$Dc8y@nH~WkwF#nd`dDIA zk@|h2$%L1(y?@(s%cKOe&J8=M&hWV9fp5rhej>w&_vJ2^(rOONCJMmID_5VhTevz_ zByle5R9I)falcL4+F4n$wD&@Wde&aj9FJBzKb5XF_rxN3x;n%c6_aqK#(!|?s)@6~ zLl^~+I+i*$vsWwMkF*ml_&u?14JBfNT_6Gyce$*5G=FxO4yW9dszq@*$ULvUKAp+Y zV&6S(p#O4vwv4MZi5Q#EtwF;gNrb2=E99K)XyDR-wX;=u0KDgqx%70#-dHB*`)h5~ zE?!s^>C9xYF{L@B!OtOx!Z#Z^E%7Xi%Yw<);ZXdT0cWDr2UyQjqpJj)=VaPP+?eyR zdU+){oqueK`xTh1pkX{d-7#ugj3_Szd=3_VdoYt_)g%@7-G`G4XOx~jyIyj2l2-7L*iX+#oqQ*a@B z^I|Sa*>_Jz&+8n#IhPmb?LRyuaU78p* zt6rN*ZYer9Yo9_hGm_zUx$fWFq%m0oLcOjs%Jln_>Zh+^+X z<@lkvJ|37-M)cm_(w$V{doNJ{HTCRko4*=*4GcZ^LADmpmb+p4C)?EMt{10drVi`< z&aFQJBKlc@w6Rv_r!#vym4kvcgF?4$bAOncvWh)i^&)0>l?tI$24i$`m{R4sc`;Qs*N96@GvvPJO> zoNz9M(NAs%d%JDxIFr5FHY4r1x7eEjwWf$tJw!x54AY!;p7&gb;42#$)OP%p?|*F7 zsW^6`;2evapme)%fy-Gbr3t^%#zSov)#-z)=kA&c`bcSPCEk6puII5B?ykcB%iHnC zwx-y&Tyerk)m7XmK&U&--79CiO$s&`Ao>ndki%klyb(7QomWpqI7yc7=l>UbpMIbO z(ksBq#dk|J%ZJyqqSC)83I9%gY=4asyP#6WFo>?1m-!<#ZBR8SNg5Y7Mqq(fnWH%D zX#t;TEc7JXU3TkupQo_5@>RrFxF*X9@aFvZS*3O;KH4Mhb4)Anx6{e1plkY3oOYP~ z2c5oaJJPXEqb;d+rdio^r}SMti)N_!p~v%*djk%O%_Mo+brmpt)E7z&J%4-k-UZmr z#@Sj@`r*7loW9!LbngqEVJQQMb-JQjnQ_17<#FZUEObyH~JlKFB)u+Lh`#K_M;}OZ9Y55(jNL2Dc{MZ zoB4owr~PKn|JgP9s~JD80^pY!dfyG`{FI;D07eo0i>-6%CxK z+Qdz?5#QZ3V%XVnV78snsj^MmtzE{kz8cLS+lwo9fPe*#zA5R1R@ON&hdx@o)Y4t| z+}d&!;9tn#h1dJssE(AUs?mLoA5Yvx)H==#?s+O~)9rY|f4a%)xB$AxRu<4PU1Rw^ zL|{YeQ`UXuOB;@5N`J<2u`Iq%oF+#6yEyr3teYDj z3UZ0L7tu9}mJ-rN2b@tD#8PY}(H$J7hyWwz+~>i$nW`$8Qu+}B#g#Aa(}GDkxwzNH zQU@;gZYRJacwp}&Vu=>B@pgxuko%&uG=aEeODH06$$dQQI@(fVj^m^@cs*{o7@7+k zMdhOEB2*_9Qh(fAct*@rrJegliTvLgiN4#MGK#>-axxrJe}JuFY!%yMBVNqswrkJbcdQ*uMZ#{QnUN%- z9StR_fN(Q>e8!*GoI12S*&$sCv)13Unf`>lr){Bm{KNHg4U@E76%)qA%Z8@t?fN>F zsU~kG%76bijr?@ndu$`Oe4`s>)j|gopanK647MW;ZJw-o+|np+rH@-8VPabpw?dmy3_o=EGA08emLPbKHuiMWkk^=oac*; zN2ty@KDx2W{r4K^{XHM=$oIN0K~cokBNC!_XGzLaGLdL~{UsW}Gpg%xV5DS}1>w$< z|Iy&^$T`&ZMCHNmWC>5PZ+7OUTHV&PM6s+XKR9nY)e|w*2aM0b@{vlp587+Gor3^X zK&rp)`bdA7zipBI=92KF>Mh$w?+IPi7^c}0On=w0T=;*Oizx%-%CkmSQDm^mI#nC>khUj{ zz~t_9CI75N7qcl4I0$C$c4QBTI;8i$IR{zniudINx`)&&*57S-%&SR(3M~Q7JN^vs zE3s8&J#7PZfF~T&mY*#=Hmg^4zaQc`w@5BFm+0_PTjf|z$ribt6jb+n-xxx2H*XS3 z@p?-uNqNbNOHBp8PtbF1h@% zf62UYCbXI9IWWkPe?6?TbA9V2pHbu1dcJ>NyC3Lnr#ln2eUrP}`NS4;e(!^r|IGCz zCazkM%lN>KU0O2xLP(R-NH7e)^2qHr7|k!T_5ncCS^rjdh?fn<$04MB)tJrw0HE;0 zYBw51cWX@+VIH6~hBk}CXGu8+%Rb$;p|ke(1j$`2#~o53I_!f!Le^BMf|@V?D_>J{Z~IktJkO%$MnX|Xce z=1<-BvK|@&uwuMxhV^nTWMh6?u1}3J4HN5j?#?hOo<^rIb>C|QZR2N)1Ymz^<_M5F zDooT5u3Gble~DIIX&2<3h-22p^QSJTWB%@Foj+FPhr`>FEBLm@>fOo90nb4~@*ZWM zeCAW=SeleFwc3{8cHMkj7gBNAJx8BX#WZq@SagtOqR#8|Z|E&h+0zID@aPQY=u(FM zh2l9-Y0Q94i=;5ed_ce8<&J+ zW}Mbu8LQ!*)82Yqa%qh!Q!?$G7UwHWw#FkTK%W|i@4ch}x>)Z+(yZO}^(4o`d=u)n z7G6-_$}4S!M$NHoX8d0R1 zv^>M$;yQ5`%|)a`Ahv(^ax-ADxh0il-Ylk`iSX3?nIlA06rSx!(`Gk$^qyzS?%MF~ z%=FJE!?5j6PsW0JoB1}wJ!vn%!zn<6pOMABAl| zYHrCh2Gybjb{;LiY?1@i|Inr)!4eBdQRI;lsg=AGyJP6Ba*fx6(rzu1#VhQVvu)5- z$)h#d-BT?aN8v|kG&dL9r7y~$BOhQ1N>fpo*?MdfJQb1KIKslAcEV~h1qtEwdfwMz z9M;p*{Zhp(6CHoH8*d7xw<#n|b}TvV7~ay19A;g<>ip>tGfy zqnvy++FgfPz}(BxvJCx40Tck9Vu?(H<49NBDQgaASxfo>Lpg?iWC)B8Ktp_b^@Au3 zxaQf}O!M;#+|w@J?nO%MRx~}BU4e%5LX9dtn>;$-IjMg}9Kn{ZS@sM;DYLL^Ha#D) zSSYjSY)3QY((K(PIFi9tF7m{2V=W;uP&-@GA*;mk{3ue}kW22L)rxOM-JxEbma`*h zT!dfT;fXm(G3Ea4wvEi(69e0cN^~V8M9w)*!h;&O4?S}ny&eGEJS&Nw#F_8{lqR%Bkg|0=pDlGFfA7`Sd8%R8RB_E})md&_6!W%2 zxfNtHS0QwIfgxz>_!(}Ut2mNYZIp>-a0k;&Fi|{juW_y8Ac_SSL3n>Sb|On%z(CFQsxL@pChX`Y7Z&_# z;aC`tQMpiv{^kd(N$DL z0hk+=8Et3D7Mm4-2cC_>Ymjw2Z9|_mKip7(VGJlX5#5d}1SP~f^zN>*ndmB^@7PAeWPm!+FQ4#-5XxvA=5l2ZBxhp}lBmC~p z0vqr!-Juwx2}O!1t)1joHcx-bkn5F;Jl12{G{-=9w+k~`t@%FYgGzhs zboB-(^TybWYy{sz_KEp5eIjIi!~=~ciNnUCDLE|DKiE-t+Pd@(m#vX9Llsd5>p+L2s0>zh{T<&A`>FcN&K&kuQlEZ&P^ZMpOUlNNuKxuW0?M~s*X?AC5ESW1@lyPVZ7!&lkzEkdiA6oZ2tbhA_n)v$l4-1eR- z7&lS@--#3O&ynmYh0}hn`RP+(2D1p3gk%%b%D!^gZgI_=6YN=u`l*;=A7kR2cY%fi zYD!g@?@i+Zf~C|Qqz4f7W}&0IPrEyc3@LQ|;bBssk0=_hlJsnM=J=kL-zdjb4>eYi zYhN|IoBCA=nK4DUb#fI}am0VtCra-_vmtZ!h=pzj3mKuVAw;`L91ABHt~R7N^d;K& z2J@;g#dR%z9E1>;tjGH|=h8GLK9*yHtusT;*tE{h2;gIo@ohJSgs5U+{pwbWCx6$% zVG1wkSCpgLWZ7sJ#9H#z2O3B4Y~8lLP06914F_3W={?U6)r8 zu`TMsbUnx3^Hp06u#9DWYA+T+X3->ys(W_lM1C;iv9?x$rFacbV?T9c0H7%D%yUE))+6(>rT zaNaO4_^cevu#gR;!^nRK!H*3x=Kloq`@Tr&eYNP7SgN-HVz7hX$zENega;3C%7cvi z?do0iOVdj5nPGzI?sMWLY_LH&Z0=4^@;v;38VHBW!m3l-Bc9^Yv)xO5-MxbY@zCA2 z(2Ls6h&+c{m;)kJauK!pIW`du5A(*Ue6fN!*jj!zb{dVxqSnke1#Hn?x%IJ!LkxMeeWnh!6Y{Vibz(SW&%6qst$f;MN0kZ?UMSd z#+o{GbC}g)8*P!k--T+8HO1Yja{{bXuP&Ye^JJ@=W%_P{cZwgHF19QzrW?YrP%KpS z9{@@l4wEdH4pM)e=YTMt00dpbrkb873<4Aylihlr;^61&FnCA4n*w>sY{0&JTX2)o z=4Fqr`|jer?utB<*TytRK1oXX%|g2U@XgA4QZ_Tjm>V)U(Vy2gZ$=2uhZT;;YSPFw zOc3hc)vpxzjt9@IJDEnEmK&PQ(%a@W*83Cn>-U>#i`YK)rV(@Otk8t<$Tc z0ajp5a*0eLFAQoPFQvSlj|_z^-cb4l<)LPhtDDu;)%ek&d1)!d7Rk$bdJMet5{^Ag z&62TlQf52KP)&qIY4ajPV!ZZdd?ex?G$aNBD%1pr(9niJxRuMwREs@fhz-R2Z7GMj z`g~V-9aDd~jPQp;GTRZ)i!OzlgaI=mV9x1&5qh+Ud(!iN2SS%!y3z~XA1blYfF0up zB^%paS9tMb|`;O;QoD~z3_-ECcyS+vku)7m7MC8U>i%m`E0(i<87 zN{Cfo&1qz*svO{FI+%^O< z9E5)YnaE9A=w^3^k=2UL-GRB8^af4zV2Am4XcC77EmzIA?la*6H?Nge=zc`xm$GW+~inh>{<{mOZ+wXU4d@@xldVE`5xvd)v1+i)t{7I}YH zuJj|)c7v6JQ737T`j)u#xl`(<`}YtLT{Y}ypCrK6L>{U}*HpnjG~FR<5VD;qlXT}% zWmnJJMgnb;hTs6-Nbg?p%kSwHQE|B5O_-Xqc9*gx9IX9VBxq<*w4#!2zTK*<{^WO~ zz45p6p4|VAkm8TJJZG4QMvjJ3FaCd0cOT7>&8JmMId~Olc>L3U#kr-Tj%~ zzCISvkLLK8>@U%$46~kj1A`jxKs0Cx%YwvU+3V}w;651=~E3JDs7GK%G zh4?z|(56%mPKdQogcw~r3PG{1YXh&*ANPVwM%aI_qo%C5 z!oHGOjjUWon~w2*qfdAjYg=Q2y)^+QW=hg%bd#9Q73RWZX*#Yb#<=Mz9@m7j1>bgi z%p&vZ{1Ua?^f}3jV#Q}7#e2LfynY7!c&L9|)ZN|XR2xy9 zN#EIiTcrEckuEdjA`kzD_6WB8p&mg2p3M}OE}9UJhiJSip)jZ5t&-Mc96dc$dX^o_ zX$l=t;m;|=3mK)`l+AX<$C2J?oJfP8VX$0`$_drU4@*#bTVle`G6&0S(xvSK1g<== zmw8XGc-wVG@=z+QlE;5lnI|JocV{$Ld$npimF<*Dg&jt{3fpp(Y`<;Rs>K@=?7JW2 zs9Co9cI#NsEgA|Ae2I1vTYQEV;E}Irw4LD&35bmC@=D}J2q@XUPB!H9XjJ&Px>E_L z2^8*hKG_=EL2COd9$^N1^L`OHI(eCQCJ_kwC3;S4wp+MxTw;HQHr;D&a=OOCFbI8- z6o{2&-R9A=CO|Kr|2Bo*A6T>j8O@QVoiSP`6@KFgZanDH;#@1KC-8|9^&6K`zOPVdH<#vD^@r!_ef&SO4L=t+*A&HD3Wr*G)UaDSA5trHhtfR#;899rn7* znycRDt7_Xr2a{qGEn802lF=XF+~X(m-N8-C8+>`;L_`4@cl*Iz*P^+*U9R4vr){wD z0%}P#ISYSyMW_{l2hAd}dqo|Izu(KdqBGU*g094UX197~Ahn$#&2_vhEd3MYG(X#E zbc5z0ZsJ_CUKj4Ime_sTVj+*))0w`tJv64|uy{RCbA0xt;CB2r^xS3ClL02s%@S{x zM``hfv5?Z0(Ul6s2X^e+emVw}?Jn-!*B6ayqU?Y0WQ2sF;x4LovhOyenP`<_4V+XG zhz4wi7frn0NV6_Z&b26=Q&Q&UJji7H(AHf((7!()8Inum8nX5P4b_p zC#!0DXhdZs^_Wn{zRmQqf+9udd-TX+bj0nXaO;P#^rH(CS4bvY__)voc$sO_L$r3cG(_NE1L{Ahkm~BVHCQMQc=38KQJ$&G_`}Gzm;cS3dwp#d-0%vNw zjO2Yj(PV-d{#nTTzsOt3f|@Ojp6uh2!nr4 zyq%W`A+H-ek(Qu1@AJ{~6HPRJ7C3A|W&aW=j!kqXkMNCUP#fT={~dZeLQib*zo0zf zb%sVo>tUDJs@WOPOZ%lDSQ~%j?k|OumYUsd-RT&RxjOzQ+QU_OnM0w1N@$NW zrOf!_+xA4e5AM#J7TbKj2w&36f;)|=%v*CfpF~pyb?#=S{v71!*{c%C!rVR|-?}=6 z{%8}!v`%p!NX)N1oVC_Qa5XR5BJ91AnuGx9PU+rBz)vPuFap>*B~U=Ri1B}^!kJ{T zXzoKDu#DGiDBEwnc>KJg(b;?|iqv+biAlZoCv($l5LtQNCu3yERGptuIPg;M?gUi@ z$J>6P!tQfn2C2bX@sm`yuBsKvvf5|o9vJ8;6|VWdFj-Qr3ZDm;IoM_wCmp-7{RnO= zzVOS;ae@g?>#0K%Z8hKpPE3D(Uj3-PD0+RyXUJ*=$47))?tgqEMCM;_g#2V=x=*X` zxZ|OnNt7xW|4@fM)y`b?L!nw}wXhlvxHB}}5oYVJr(6Sn;&k%eN44AlZ36OrjYy?M za@)&m6^qmDnA;tG9_yC$#{Ah@RH%EMo<~ zzs}rgrh+KFVkUuL&WwL?*OF&~9gU#<^QT0%BUsCVI8ge5A^lgr4-CA-9K5IVef7|L?3IT(Aqqe>Nl`!D5Y|w z8YA)8w>g~<6v}^@m~Nry;(g+EG6#}PPK36F#wlc|Fr7TE&9qTOge*>&RKmF7Hj7sD zJ0L?u<)EN*H1J!~{?%wOv*nQk%MhZ(ZkeVQUa#l}z;xW^s~HGzq)tZc>5+9ewv;lM zt0OUfxU3TO0KmorKa1WyS>VeEDbX``Wzb#{BtE$oW_o`|cMKU*)tU-1k`sHlScSWY z*P!V-f|zWJYd`^*Q`pW}RhL`&ec#e=J(xu*n+Z3Y>Y1<^hSEJduVzelz2oiRT|q<- zi1Q`J$zUSiyyxD*_JsV)vpA+C&G|lNyYBZcx=@$uhYF=er2j$7z3(A>@w*Wa{%eHL zmY(rz(wKku-ZjGSJCq(_p#w(A@ou{%5HE&oq_1vanm1pc_8d*L-*yPb`CC~t0_-T= zF>xA^!k}bCMF&G1oSQKyM5rqh{!i1eY3RRqj=aI7&Zjv+WFd#4@+msd^b5j{}Z zwaB_MQn(iweJkCrR%}LStRqa!s8UaJt!97T3v2u&*z@L`M73NIZN{vdLI2@|aO z<7-aXkU2&9g1pe`%T1H?sdzmPBb7pxa?jJbK4_bvS6`PcGu=WaoOrdIW~0q|LdV%o z2b+KHSY=32Q`<51W+ghs;s*)~X4QC}!mDw^xA4nkkxtC_g~>nLyh_I)f;ezzO+9dd z3d}R33>HLmoUjNKXTd|1s~@{x4O+jnF12j#?gZ+6+ZLn^itv431V>z0H>^Le2knM4 zm&!3}`8ABUd;F)o=&k$Zi;@%Y&D|>@O)r08-LV%~D~^3;ZodWfW0pd*J3U=BjW4t} zkM!CBzQL~a`@8r5%jGlP9TMo; z7oW^Cq5uvwDro-d7-=o6s_BC7KPateeZW%l%CBT&Bo?W-hk2bhjrBZFz!t3=Q$&B# ze&eI`*;*-^|BHvjX#R^^7pNDM8Gl*+-VVr!!OEM-b=r+6Z@#(c3y8fq*|ma&?0vv%FA_dy2w#;<<=dorJv4oSq+ z*T^jVz0aUjrz+3WslL-$L)^=2#QcA*AFfb^QJ?B01h_rRCIRJ{z1vVLP@dvFApP)n z2mRMAx&%~=yDu2TEhi$S%-hN*f7wlJA@V$83YktT2Ze5_cKPU)a=)d*mLG2eh2>^G zM^~7|8$SG>tpa(;`RkJs20Vd}xQVddkqQ5P+*j1`5*-VNHGTt|=A#nNv&Dbz(;nJ% z_1FRx%TfVERpaS&J5Gt`_pt)--iPCGSBC$EOT@75j78}Axr0CUi_))*BAox(V*iJ2 zoA`fOIf(avu=4-%((-}!Uz!h!#6dt}*dL8Tx-a$KeOeSir@cIZEj_U@r|%z)!mxBi zK~2HnCh8NGMY3GN6viKodQpEi{OGU6p>zD-#r+>!ocHB5OZCgi5!F}*qQ29~Kb-ED zmadl>U!ti5#|l-M$80vH7e$L@5>v{eD7UcB?Sy1Jqc?NcHLcI~7kk&m;;_j^2719B ze{Vf3)>C>{o@`HOt@bl$PE}0h^h%9h8tVPMN&cTNOC;*eUVhhyGkbr+bJ;~IGj2y` zT=L&6UN0zrn?##}MhHj22Zbgyaza+EM(vXcKNB@P{IfWEzo196zpl4!d9r6=IV#Il zUM%f8sE?W=b=F#cfdHRs*_K&VoADH?>sR78)$Ba~L!-Ty#5=o=09T(5Hd`KLD5Ode zNjtCke^HX;&ScwZhh%@$mF;<7jWG#x0iR_g1pFD^UM!$`AROHkf!j=$(G^Bg;K%nt ziu(uGhuf5=-P1y5o%w_wsykGFcCURoth4cCddup3mAT9|p+>%dIl=oN5dz|COlZg` zt}p|D%vV;0u{qJITB9y={*BfjItuR%5XEBR#;Cbe!@|5WzxRK`kXy9iJrqFB%G)c$ z{`Av68}Hgi>J?=vL3SsCKfL3I`;^urGz>Pnt@JVTRZnYl9g~sq<5*o@>WcK*Zq58e z?TIzgNeqt>Z=4qvJQ{@$T!%YK%6q^w!nGZ}4pC zjVN<8?Jqb?dnA8hWBG43luKr5swJl?{8Q~ey8Ph-INk0CB#jsA*?3=rnV(oH);MZe ziZOAW<3@goj(;QG4s5*-aXTTseD!fo;g0}&cge#$MpeyMJkN+E?@Zw=5oyUM^iIR% zEq z{S;06Vqt~)Db=6d`kp@oqb z7o?Wc^?iT&=L2H6kt3+#uk?fd7-#QWkn-dZ|M8#FG*V4=AhR329pxiF|35{`Jrtzg zj*kEI_0I=T1CR0)X+(+n$;CP-k^U$z;_Y@SIy}im+ZSfJYrG3EPE(9 zdKG=R+BXxbjv37`ZK3GJE_j|(n)dfo*r6YqCw#cB9RgaQ8x`ntU=AwUU%#598oiI<4{P>b1UXIpDXfE7*W~#C2Ls+WiCPBpOaiZtXfJeJd-5Yk+io1R|Wr?GESc8CXxa9o5 za%EECX}HfP2kU5!+5WH;dVyDnw3<3ksId_iB`ykJT>?F&O>5S^uh*%&m`k(KP2x3; z2_`o+@r8S~cqNTfU;nFOk$}TsjdN@Qx&?p7iL+o%gB4k83YZ-Cc%UxF+w!Ko$VUWw zmHHFfd&Xe%sbBDhWtz)#Q2OpdI~l}m>{bburgbB>l!G=!i2tvC#+8i0_-v2{Dym*czf&*^iP0fEc zCT^8t&CXK{G;_hTeZ$MOZ5#X<7P2Z2_C+{fvGDaPSe0#yWbrEs`=7$WMnqI@m#Hej z>!duZD{!aP46k6b-Eer4`dZ5abB%xyBK#vo3C$WyYf!3NCKf)0LTRGaWEmgF3BkX= z1<+bOZ7%IDXc{Rw15o32-n1@lyvct}F1C#p0VJXT(5665@wnBAxcyEuwO$?ae7O!m z*|2k~6LY?e1vl&T>u>d4bLRcYoTqZI{TZ7k@T)B@=jE{i72qC|+5P|c8Mgm)>7ak# zLZ>>AsiDEWkZrvJ?NzMxINUVhG>2Mon#^Tc9}e(Mlu?JS)g5GK$e);>&W3+P3p_qT$QqierXZQ_5mr{`t2?PVm$Tw!C9N3MU2-clrbT?9fuTe+g8R1I@j zjH|cPNKixcwe~mKtGnSGOtZd9Cttrr&o2tA;Z&mM$~eo7<)lF3l;})5s+o%#MN8E(+gKc*>7Hb&ncK`wnY`w%cl&vd>$!8wW;dmGh3J1ayZyGRi}cG9 z+qyLf*)H6jCWv;b>2k;DXl*6o>>09VE;k!BsRMC^K?ZoMY?e(}|1?3qK|$hG{0k|9 zorQ(x8`02kXg0&PVLTX5(?2sso9?ld)^ym$R!CmOD=z$} zE8)Fci3!@Wa?fb6`S5?QC|{9hnzo9CV&Mq$(*{%k5t(^vmP%c_NldQ<8^UV6F`TiS^-I&76iOYgi?W9lP7+W6 zgM#%V0RC0RnlJ#x%uLt>uJdvPlltd%1OytE?xDz4sJ4HHvF)l02-VmskkdKM=V!MF z*BQ6@P9}6&G+LxLpJFtN`utCGWQZ72Q$&_^)sI-B0HSGxKB?i7EplpSbpfi6gHuYj z+KLIRh#bLA>$_akyR%#K-ZkiyDP|EugB%0E?Wm;HI_(1G`DH4q~GM&Olvr~ zILd`@cxZprST~MY4b3(oGaG|a?l*xN17zA$9qXSf)Azfh$z;$Y^G3;H?sNRPJ?IxD zVZ>%BaC>J#bbDrNtXaz?)i3C=T-9WD$@Lb6tG^bYZ(f@tf;RX2+X8ju!P z&+DesoO1jHD@$qQqA0pX&2l{rF@tHs&(asOQlnpPJa@j0$~jfG71L3;nOY@t8bx^r zlM#!zNHPc>?O<7Mj(}jCag0eOe0$ojh7VQhWI4;@^|Y;H=~b$%VRALGEDbAhN^c9@ zh8lmms>=&y!3&x{TBx=li4*Tt}uws>tWcWWv|aV$)mi*>D5dt)m}(sayYYl>2; zn)bbd>Z3N#|{5&WAxZ=nyZM93!dzoq5@z*2k ztI3koogq=2>b}z25rO6gy}gzLWv+jLKPHi3OeBeD&a00%eOe@NnnP=ZL8fc0tZHUc zMo|e;j-Ds-wVzJ8%l+@HN>}+l?e1(C z7tAoL-YOr&Oc_9x%(I~gFB@h_iCAS0P?waUTrJR}9a z-y6qdu{}{*^ZP510Z>^TbGqIyqpu>EtFTxgjAqdke(s@OZ4d(wXk*4r4rK)>*FYWV0_|Hum023jTr$LiSf%0%;qLYL(7zu>cE}N%3znmt$ z+_|2glmTOlE%zET>#dVd_O-W1oj;BA0R;1tv7gNfnI(Bg(Jd%XdSaH&*vRb_JAksT zmAI~uqIc)#L#C))n9F~)?KZUBbAwinx_%|!~8oVg_{%FGQ5 z+VvZ4Y1aJYH1-5+{Oo2UL-3~U)1TB{T8+~?hQ|hAtB!z!$ zjqTJCg}!pOUlXk?-dpv9HG~i`)T*Wwp!WsU=Y@Lcn;I;4Veh2bIs50njyQ!o1oR^% zR&wpp0EcByrfUzDe;QK9O1031dy@XU+zP#fSH|}zzZ0lBxee1Lsj}o?HeHl5+G{*5 zJ3lhMv7s^i)4G3CocVoA{?A~nW*@}dZOhuA-hyh?x^BHz@;FnaZoZ}rsrv+KjYnt3 zOx<&nG}80wyRWk+lp_)}9Te}rE_GL^INd%N-KpPeitv!#>dx9gG#r5(0iyw;>vgCKf0$41q2#>3 zvub+~?)>+GR`ke`eq1@Znxb@gkXsn##P=^s>XCMT-nfcK#=*#mYkAEd+v=(i6@&Y_ zP|$yRj`7c@EPXd+s|Z2k_9sCUpOJc~k9xU2s*mH9fyiH2kVay`QGvv#=k=e@J@|bj zdWzF@@p#}hVy2IDT!4`z7?)}_=?UvgScMmaax2G6TNDNGBthnnd&?Oy7=+R-`Ids1C#U~deQMY5%07bk<-KyUm!l#?!bc7Ler$hsdHfY8g4 zK2GIA2rxSlmr8saEa^5Ba~mRZmSFSi*<^ef0v{4hUf7H4aS#y|KNwp=wotg^l%yH zgL_0u3i76tN^qTS-DX#5+~ic{D2M~hNPGkQ0V+T@3o8wKius#W`}Z_v7Z*d8e`bk& zLA>u~TWx{Zg_{78qXd5>AojDI5jTcz3>QziRIBdo(LvDvq2usAqTUcgqTgnU+KJS` zQBgqec?%gY+Gt79Q31rZ@~Hptb_?HIE;IR%u3vnK^Nnb)FPQi#`!Lo&i>NbOj0P46 z0n*1;DV}UQ1(La>~T+sV?LgM zsf9Jm{U17O?um<@BP|{`+FAa2Q;O;R@Mo|G|HKaWd=TM(S(C0su781h_%84V)0?wN zN_B^~Cu}x%levyLTk10Y;l%K!G5nuU;1F;!24Il{QU&75MN)}p*y`6og`SW@2yTyO zC+SfEAD-W@Ss8yM2n&#vOBVzoiQ#drU6$xSG{H-}vo{8)A^6l$dj5bLXe#P@IdV1J zX71%I-l?XynBBSFX;&5w<`*B|puGCe3;#8r!~ zu{hB<> ze=qOhKZ}3X*7xs2Qn+<=em$jBsx9MR*UE#Yo{baCR|mHAaCv#nYpw>Lo~3(l&Xcf! zJ;I7kMgP&%@A4PU{ej-jsXD=sr#_jUJN4C~Iq=u7>4X;uqmE&dH3bA=uw*zEf+l>D zHB)i2OdmGMSMZOYLE^(@GdI*2Bv%FNR9slfmU4f_*Ist6qQpock9_pqUWImg)P!ZD zr*hra2J`!Ji1j~vz!#$bAxo$b)$}QE0%qraq|=Kmtk9MfB&h&11w5`|belY6hqJ_U zKR#(I+s%pl*4zBgG*l>&?1Q@E>!(vORV1gmbH=S z`WqQT5~X`+uJAufclS^|s9W43-)2T~3vTfv?c+faQ92{Ku)-iLWK+5&rZg6w9u&Iv z&u+TlN1GwF{-YrEyWk5i67Cbqx8KA37?*!{wSo$Fwzs!PtMOMGifp6@j;iG5y-uwH zGVxVtlM0+fp71&n{jVT2r@nK{Bp>nEe!SunN=`im<1NN6M^r=C}6TTsseq}Jhhy*e@eiuYxKK;wZ zp>P9^KU`NjhX13Zl(rE8OwoupG6~+D$2pUuT0*Yht0=U z{|0Xvx!nD_z28ngaD6YxdCw<0;@)** z-S2R}dCoG8{zs||H}U;<_t8z$E^8d^90>=D0v@#`wa&Xp8GFVJfqwHPu*NX$Px9gi8qj->NfOjj&lFzyXDbOgqDweqEWuy zSc=?5EQNPC(Vu1EmLR}A6cFz9{04sxmA2PNI-q<(iuz{^ zR`%peOHUxmyWiYZ@aX+Q_UXRxLUtVJa9sIOv@$r$PIS$z4|87D*<#)#7}7 zntXE!oc&h6AZUMutVCU&tiyV*^y{$MmMG+TpAbgW2&O(`B=g>B@#Hb3}y2ipx3nx6$Wbh^@Q1;tpk$ThPu?vWs?arJvWhyTsTgZC8nt_YN z{}Dl;VDR%)gO0Dlc1se}D7~^7;!B77wYWC7#>kKv zLrd!W3qk6SPt);fKEBt>mXjvxjmkkF|gvj z7DmjV&-b;%tuY4k*(weJl1#m4i){f}Jnv|g{4t(bNIxf&?Cm=hyqo&*RC& zFeruV_q$@@O<)!UUz1$K*54^cl807dQ_1uC*$ z6swo=g6Jqd(#h32rbpwufh97D?155oZ_M8ueg2^sTe=yqH%Q8LAK%gCebIX;wBR5V(*6SLUo=(K-& zL+^5>j*(LF3=JSlItGHnzJ;saOgGoTmZ%wqV++;#JD)SH|4fVRig-p`B*GfO`?dE+ zQp@!vZk}0awd3K+pl;YJ*{v9M&uN#%{ewic8N*lZC_XQ2m)z;nNiOmf^HIj~bj9^P zC*Z~m-WLUlnEFuIpXD?OHe8dhah8ARH4BS}#7}TZq|;t}bK0A>e1jE+wasOji<9bh z1_cspobJLnI%Y{dw>;k3%oYy5e`&0T{6KKqmkQ+)kh#{Z zOrGq3mM`C!$GoX#b2jM;OBY$D1#l?uL4RO_Er%!NiAhO3>}=U0oHrJ{MEuppw7(fq z_{@2tn}vHqs7g&7R(^hn?uy8sEH`w81(uSzIg}m()AyE7bNW=My=Z^FM>1&{Yc_pO z(>^xCL>jV*X~s!@n==)gVQnz=!};MFXN>;+zc&_bAE9Qpn2Ai>I3BF)VGpTq5ROoR zut459T%~$s7dd)gU$c>jMLhG%><$rLd?hwidhuf2Mi(N&Kr@iJKA5JkN^|nYs#=Ef zHTvUeZ`3paS1IW(nIC@^&3f5XDt$_e1*=yzmGVPur~q?ZeB{NI-bOa|t)vQ{{g;<7 zZI?W^^;#;D`+jW_yiqA2tk7y@!*=u`XkHjd8oSY70lU;*9z`^7oa%WGCv({_YJjaq zeW$@@)^#`mFBa&0x8wTkD8g>JJ6XaO4I zCQs%rz|wcz9Jj5oA?Wi3kNc5HDR~_h8E?MCG)ibypHU=ov&r1#A=DV%4f#1 zZUr)vxo$JO)A)bOwYt$LJqabh`l!9|OrlQB5Me;41vs23tVBH4YIHjZxw*M~4r_K! zPviGAR|n~ROlQU})~$QffjYS(veDltCG_wYK`b)^Du87EQOcPH0AfI$zpzu$ut3>{ zCyex+Pk1`_rsL7}gthr#%Z1ckLQ)V4FJP%iwizyZLd5vV(HdNTNcaUUr7cD&am#Qn z#VkgM5*6S^lknfQsLy*EJW~`}2D|7LH3$Gm|LP_{`S@&0%k3iK1mSuiD2~UxtG71r z;k-9CTRG8}Jnwj0DCRuz!H1zlzwW9<8kwX7i0lbJ-(XN1=JIiPOBy)Y>;zgj-L=>j zly!rZt5}tg?lLid>0U;3=Y6XG=6!10rR$^6Tp^oVakJ5EoX9F_9~xVFg#vt?CVQ3V zq#*>foD@8@ebts}izFc=0le-vzSLEg=k)Dc7$S9Ew(~e4L=3q<$gy55VEa68Y|2_? z7@xFCTI@T9Ht9&yYv+A|@iN!ipO6Cgh6fLgjIQ0l;z@ab)^RgGhCESzXe@MOJ>DF= zkC42qSsOI*%Ap@KCv$3JDLtpIrqv$VCn;C7Z{{t~-c+Z)d(Yd}=D)90TM9_7FzQmI(Bx{^4%Q)Gg#vK>~`1e>Sk z)E1S0M4Nqo<*#^Obm23<>KyaA(!vTLJrIXX_MOQyiY)a{MkiVKoTssldGWZ>E}n?o z{`XLA;qZBn&s}aofR&OW{xKwFmWaFO<|u*>ONFkmj^yAfh%&Z zkR6ZPhQv%oW#WL2`)CrF?6HbEe;XA(D#n?}LQ8AAa;fmUdK#qK`S9QZVV|fPrPFn0 zw6ep0gZ0?TAB!Z;X?OPH&O+dR+Iv$HZdaj0giT+Yc;84^a*gGKWs8-e8%lnuO%Bta z7)kG(e-vY#qr@>D6VK&EkKmshq?hK;V<#IdFFe$Q`@A{a;DdfgE_KafIgs`u?t z_!lk|+PV`zbliOLLLA4PGr5YNoH}ovc~|>?ll!X4wcjCA#Y=3TXwqSS=dfj-k} z?j0|c%rRG~Wb%IVj^)M^AmI!jR)nrHtHfk3xdTdZufN#L3^fYfX8QL;$=A@mN~v~# z*a>Q^vh5Hn#z_Y~J6dzp6v9)3@CDXTS@f|4MOe^n*%PvSJP^Dso8LrA^i#{g|K=1_ z0E=!zM*gUAAG7Ibk;c3FSo1}_1N}yiRddMel+ejwAEh6!zfwj-(aL=Yz*?)xshg>} z{k}H&!nrNqG>PL}mDyYgpcPFgtt{vWh%S3dUg zJdw&5^H$hHD&@>Uibn9K!|E@wYV%blf*)o(im?Ky0nYmj*RB#9t1uwt?H676WRtHh}r~x56l-90p}Sn!J2Qi4XcR zE1T141#jE^u|>@kL76LbQ3`PKDY1fAdn4QjuAUqF3cT*848#AK(cg} zgXPXq=k~>`CQtOumHc$8?N{P|As_pbIH7Up9(qk4V5b!L*MmSTvT*#@-{kbXBEx%Q znPekn&HOCG&fY2&$Pg_OI!z6Uz$6GUPMs?(+rH+veaSkM)5U_t)M<&bwdystYXO zsW8cDaQ%i7I1Mha=niAJIOJ)WnJ36E!=d2$Kr*10eMG{nD7L8CJ)t#Sl}{%JPa))x zA}m>BL)mmo<~O3TbEENep;BHoOk%K62n)0H8-4$ruVN-VC~NzF%gkiW?F$NC#~5Ue zm6Yo&m&FrmsvQ`S&yiYX6vE=Bm@$TGeD*m1RIgi7d0wt?MhjsY|y~Hpi=G zG`+oaRjb|#izT;dMnQ&i{^e~t)n5rohN`{FbZ>0n+Jw}9QlB-(vgkkw0*=SrI{1hS zhjnK8j;~KT-svh{XnVYj{J@=G@hU58fgS9y##B+P`+^*z&Pe08_k^QBEi!bv>Qk#n z$Ka0}j4>R*jR(%!pvFx~9Uc4`f>JPh5T7PHX36!P&NTbd6g5Ktw#D^Mg${qjmDg!Q zU_?6Id7TP>Wh_H1==BnhPT-A(LYzM4-*@`BQBlMfs*Dyfr|EC%2wtu(Arrw)=BRfo zAxKA{%nZFafjj!Fp7e&Ses4CIP{~6t=-z!BWCcWcA*MM;ZlYJ?)6%2n+qL`BmhBLu zmio6e|M(uM1?{OBqdecqo#((N52CSIZ2EziD}$nc^H?<@8kHnZ$qm&%+b(VBiUNs4 zKmSVUmhTdY$_*ov*Ng1!x)K5C9UKi34F6p8d*fcscwvyfhGeCK@IsYRQ4pY6sp#kd ztZp8_?csVTug)|;u(_-H6>7-%-JXzDIeFx~p#oz#$COa&a+EmRcrxgIN4sPhNG8(i|*(AHCEN zKeJuTGaVLzX1W^oCM;qV(mNS(bx8kAtX#+XF-GI}?rRfO=;E1?UaF0`;&2<Ecm(Vg*B>%D(gSd(Vu);niP2ofl$cRA|%WjlTVvJou^Ta$h&kHuq46~+JyG|OovL7EO zZaynjs;2>Nj6|QImGN|iKl?z`8#~?+vv!FVTA3!~Eo|5Y7}0G@t`V?bKo239I726Y zYW}Fh+*K)qF~*&mkhQQ~^KC;bT~{aID^*FB`AU9EaAFBzein^v;!A&7E{f_thTb@4 zNgduAFd=WTW@9+SX>Z%n{iwH}OwHUtgc)(_@MVwbXpJd%m(Lr|w~K2{$Fz^pv{=6B zF~?PPINnoUVn5r}_mH_W(jK+E>{IlAI`V{i5tR~~)oN7*+o$g?+^15k%z+eD=2J16 z)$on|h6RfzwTOXgM_b0+&5=gjU6^~zc!_J^=nnR?SvmK78{PH(3KkqdgAg$pv}mlQ zY}?Bdq995r0dl#f4mdo5$k7 z(`adBGf>BCCIjaLKGi9otGA$>nWCsnnTFAFJaga?p?eL!jBWDR&K=`qt!tVcz8VD$ z=Bek0I7)m49+x{FkQ=FF$(%~o%taQ5o?y!%5)VPAeMb}7woIT)i^cvb22ENZn zPOLQgk9hd?uP|y-Sc|GB;;|`t!gFQQi&-+K-q9+&T#Q?j-M3p=&c~Ioe&ONiv@rxL zY4XxK;>Yqa4qiSLzo=SMsa};*X&O(7`+PsN2!7)kukP}hh!ICW@@wpi?kU%wRups1 zqb~qUQ1xO>=Mn#ZH>#Hv^r6V2$BDT@DMrPG%GRJeY^`{jk0>+|(b`rwoE4}TY#{!) zu;B@|<#gG|n&9&0`n^4mwrn$Yr`gYZxEuXycO zYJW7NpISNh-7nFK@DgdL9xvNX(TL9fmQ6{bRFZ$5xx`buc@TPwfa4z_JLXTxAQfw z1afGA8#f%|n!>LBX;T9!uE0!{MH5-CHR$cv7A2p5QQ9SqTiSzP5r6NWW60b!(CoQ5 zQqw+ttmSs4w}sG?fA4nOCYdv+4!P+HPpl_YkQs9OqRGnJFYDN z*Tmv~=A>|<%4FBrDiM(`7-aYgO%2GCGrjph0$e>}G$`{Zs>#7)t!=sfbU5o0m(H!| zX-uQ({o=W)Wl&X&A+A_%=*P%cO4rYKtL%@wDv{-D=|K@yM?7^dLS3HWc&wR! z)}?#K0u+9Q4*NQN!VaYYtsTh!(9DB7=kx1G*T4u-h-G4xMZdOQbCLPC3r-azO=ZWu z`LF#4A7Yi=o2VKNoLY@vjuql&*3J=t_%xQv)VjyGEVric!!GN5~KGXX*QYjD_+Q zUh^?*Pc;&7n5z$J5Y>zmy-}Wi@j8_Gkc*PQTJZ`Wnmk!yh2Bq6qJLq^q-}VAxXOXq zb2eS1Q2G3td1S+dbZ73J!?ajX$tPp|pu*Es9`3qn)xCud-?S|QnG}P;(N7cMDhwo3 z!8ElL&d5+50ms$G+naxo^V@Ss<|@I5n@Un4OLK z)Uq5P?7*ujkHa&szQOxVx|J($KAoRec<*sa>tkVkduP&TftPQ`%STLq>_$dUp5ih; zTVu(CWty*Lq&Eo*7tdH56dv30Y*e%JR+$%~evo#$BhX@XztXGzbuhtb;7Q-kh+HqP zYNm7Dz?iZZyvJ?t-o3fWe}4-gxur*Mu=UFC8)mG9xmgG`pEGEOuxHlj{laU>Pk)oc zd+0vjiMd^u=U(@q*Zk~%LDEDY2}{03tlRO22I(t)75gpz){RW&ZoBgdhbsF+MhI}T zfRRl`$z{D@4QUlgej$S?tt5Xdkj;W7-ISA^_nm*0gUT0BQ zg3{%Ux*WW!e?k!4$-|b3;WIoG>#&l^Y2?#;F%`Pe(nWE6*ooE7tNUqDe(FL`$pddS8&S&id^ zmY-LGPm3gzGDZsYP` zLfvgp0XsL5TsF*7@iZJTf&tp*Yo|DFqj(Mn-ze@p{O2o~u+i8Z8m>t}T?)$i>NI z(}hT>mFj@nCBikDUaub;pKL+dY^>wi`m|rlZpHY2ngEK=w!juuHuJ*u@e5A^*pfRr zRN*>nr{qjpFtP%bVzKsS$^}ybqt$D490HNNPN&^+EJ>WjfzvwMGmPpVqj`xMy^$*cB?fDM`Wg908bNg3uR>KNx{Wlm<>8ZccI7Wj zU`aV~I3gvefL35*PkDJQi$qD=!x|nZwF|p-J_!*GH6gG;2bBadpJqVi_LsfRPO*Z4 zNjrnkDv;16c76_Stj@w<*vM*u1}~4jX4#Ggn-bL`^jPSn$i+3nArM+t5ms%bDPPaoznaOp zN|<%mnuy`CPc7d9&Dw4(?D;>f4biTWEgsnen@{=kRy$vz1zQ=Szbilm%r?YgxFbsX zMHmr@R(m&}_O=^C_&^sobm$)ND>Qn4ts90q)8s>rpM!DvN;y!mQ17pLJ-aaPq3=sb zl;Yd)@_Mg0FxB)S7UsP!`@dJHIDOgPovv$*(~Q!=nQ`2pSF;$Ly;!FtUd$zERvwk` zbDPIN^cY-F86Bx2tRyC)TtRoe1ZdBLhm8buJ)TGJB&&-Ot4v6odh6YG1C8i^MTo~7 zG!?`1myUoDXDO>LFT9M*h$Sjr6mGwfFl}|;?KTz*-u(TjG6p}&OD0?UwYd%y0vuvD zb@2?Wl(djNZZ~C{JYZ_7<8;JxQ?AEOFE!tKpftLjk5n^PpQS}KTbM|^e74x(OeK5i zoK@9k)qhupMG25BW@ohbslNW-1v3{KrBo1GH-hrN6O}*Q7seBsgci9QB=xP2~$h3W{#PPv1xnTXJt)evH zf7yF}&+y~{s5XgPKf7q@GzN`f%AS}V%A#2NCC*7$Zv!5>dql!4Ux&%DqpPY`pdQYB zg4V1&h7Ky*hM9cHN$8*#PgUuqo6?Q#CCWpV;vyb+qQqvD)$fdd`?K`hyC-{BzP*w_ ztD8&BN0#Bg*riXes?kyA(pBimyy|*cG)`Knz6&-&LP!Qx9-D7_cLv=|+C_iuF2W`L zvNO{dyYw}Fo^2AA@oAXNX=m!%6#Da}6OJ0Lzgvw_a=7F1#-k~JP70>2!-KIPl@MNC zmP?YKgr`Z|Gsbb=j&j0p0n$`SwacI^Cim+NbET(RjUVkef%NuWPW;&}Ky)KEWC z@XUw+bZFocDwxo}udr--fWJ(M!T5fr+$vRPi;kfcv!+zI*rhQ{+}U~HYBqX$xWp@+}$W>o|fu0vx*ti+v+(qGqqj;qUgL5D71iN z&d>6CAv7I%Y-fVgKMc9}hRI6Tp*f>85wyfNKNk#t$HDHD(@lo)j3c>q4(l1>M<%4b z{5oa$){oChNIaG51GQwwEQWqDi@#1M;m3KVnvZvmyNt7|sSw?DvI`eh^)w&>!GtWM zWX+Wd-L;Zc`|~{O;{x3p7JfD6cuXm>13Ur~+oj%vHMyI4q-@OQ=FBy6TFfqf)j18y zxe23x@6GG&tWGx6shuC)OqJf^)k7Vzn~MvBM}DE`5Km4N&v(4QSvQ&Stgm?nM|fNl zF=({H-<7|=;b#}aaMJK7h|mk5<^w6e0~t6TKkJ=|W7ZmbQQJe`!N>R(vt(aG@moaje5qXR*$5AdQjr>TJcQ2tk!bnP6N-VcCG?vji1rq zkB5DKg%^cFs|J@{u4bNS;dTH%0z^iGZ~!URR_1#zr=M03_DW;>OGgM!!UEAK7_FDs z?N&a7+W ztRo67;P0)HpPsgh6jB@JjwFh9R}o8(70}Fr%r&b87iKthzTGJp;Nl2IqhnKNY^6f=?4=+Uw(aEs2&}&;n&t$_ERlTH zsXyV&CNx}+Hj-X{M4J`Clx+M`Bv_FMo26evozwp8g{Y?l3h1sK^~3jn92?k=wM+}h zv0DZ`4ViirK%r3N8$glM*W}Rk$s~BLv0m{YGr(%>oZ2uzsw;joSn%?jYN@sgeN-y@EOpEw`Q)p3e<9i8Ji0vZ2N}+D|_q5Q_4GO-7nFZUHlVViBRE1G) zXIntO#+Cqb%Eq!}=FFrOEUKejvRb1N^@wMx*v}_}no7*}W?wR}-WvDMJQEAuc~kQV z?x-Q2teMJU>_aeL8u*yZ3h$9hiL!3}__24x2AOWTJri)xOeS`J?IZhom}!fS&uB^= zme~;AG9`gH7D+aePN#xnKi^X1aHp_(q8wCTlr53qBgs$B{S*!g4Aa^S%?qwoE4w#H z@6wQx24MMW*j;L+HLk8Q(OZNHhz+Q9T>07>*IfII#|=d`(VURByzf_au?6F3PI*~n z9-hX2-OE=RBiWpP&H@^1n@}m4m6Dd8t)njimW=Y_=-z8gI@KfJxB6o*V^lE3EBT&y zKWZD)+6jE+2h&tvF~lk99yC#H9en3u^A3xqK)Fy>2#ZVlnO$c*YtZR#8Gl+ucwxSj zPxRP2A>RNYgW}I|Z13`TzWp`b)xpbkt`pMH#rE#ch$@qRuUSFg(!Hx=4Fc1H&!-an z_F&JNWaO5`G~Zpn!wwy3CPBD+mf5DH7ynxSR^+g}#w0sbhAqNmG zZ0SiKY)uJdg~d-j9;~|J;ggzGq6iQSdjaz8k7ud*SkBi9PYP)1c_JM1-;c1q-*|b) zb0%^@OV*lI^R(;b;f4no&52l(DJv-J-(P7zblAa?n$ z@#;Gr;Ub>KTJ@hUr5Ccd$f2R$XZ5Yq6k5}%d%wOZJ@A@uh>nAVZ)0l-3(WfPBP<)c_vr3P}~x%hgv}fz3m87cvw$&UF#4nTJv0Ea4Q~`j=|879{vpNp{g$$ zV8T(XLHmNc-~^q2ZyDjR!)QcEQ;YxUXA@6+M zCd{FW-bt;pq&T-RKKokEZq-FDde9`GwM+i^nCQbdQaQT-J|e!0Y^hk0W~pRKL3?t4 z*<@=ONp7>YH}pEaZwT3K&XhfBNf;HAjsenkV!ZDG)F51j)&LFp+)|cxhgAk+7rQiC zIe;d&O`!Jff#cMn3bYb&uiZ(Sq<>uT@);%(^FBu+lZ3FJykCHQGFORx9SMWl!$M_% zDiU=7F8gj(1QxO1^T~SHz_7S=yAtt#;paY$mi>`vSWs64*^g3Vo=lxEd2t(`VBZzB zi_`g|CYIL=RgUi7vj(=`D~qB1>UFDSUxFxBkM%nThq+H?@|)(4t&TzuXU{`_NIQhf z4ZAa6p*w5bysR}p;ayCCs?Gb;1R5yBWR*d1PT_M)Ljk|Gj{KCkDrzSbs!YdP7nVBG zOPg;zpbEg6UM|aNsa-zDcTl5lvPO)tGDSw)X>sFhp@`2E5^bzbZ)fW)xUBFmLtm8e zsZA8dlJ!+kJJ7jTGuo?7S4otA?tLAt>s=5!13aI%aVc4JtY2kNjaOj=S*SAQetR}| z9cc>0wkPFlLchSLM+02qlILE`bG~yl$+IOaP-rap@k?*Vl~y%HyRQGh=?3Cmq4jdU z@N?N$*TIP7)l68-)ILwg;5gdhI!Eod-rhA%A?XBpX`2G)lW5?H67i>hN<1yfaBW`^ zB&BQ#=5FyMq#v+-9w_bx)tOejom41$Vb`C8t(+$>*vq7ddn7PI|H8H+)0gdUwWAKI zh}1%rcy_8oj2i&oQEh=xn;8JdOyQ;`=Nr57TB_nC1Q1L})ZQy!`0$W#{joLq%7${S z6i>m(uLDF*Lw_y6vns}ae8=+mx}dJ&nH6u{`U}~Q1gckMNdZEP#P)=joR(R!(Z*A% z=pik^qm?;*X;b4YN52|=EN_4W(Afh8Ouqei3SnekgD}z?(2%Pnjf2`$`koayGU{nz zSm`Qc#`NU%iFMDvbm--Cia(^D?D;xRS6LXDl`vZNKj|Zwgf-#lrBz2BT zEc@gFqDRZtQo%949<%5y$!4|2DiOogh5_&Jxas`qlK$%f`yphxrVTRKLh^_A@IbTW zgv4kUcAnM*&IMI}pR3bD1L1JNSt9OOdW~;1a*kI&TC6Jgl&4uLwv%ZqwW9QMlCg|^ zlpOW@xqu_;^~FA$MHzm{#zdcMA<6B3c_+(C5YG8P{!dA38xAJr zY(`ka31G)4Hsq%{A2S)Bbw!-7MDtt<%Ak~OxzVVURBG?RgLgG*|JuJQHj}$Bn)%4p zPQ^^@u-Q)W@_SPg=t!07eqmLB>h57*XAuwP)7cqJ!k^L(CeJ7uCOt!SYCqqd>*hX5 zDlEfaQGaBA{A*Xek==>Q3R(hl&zBtK45TsXPxu&sMN}56k;;PPl9`9r17!^Ixs0yZ zIEp7*t7y*6T29N)>{buXr0L0x@W;;b+P@SW^P_Ia5hn7 zVFia8`rnVr`?S#D;iJ$tl104ZMMiNgw#I~ihI>;-90_-N zFLg%0W?U=QY_6Q1C}Co=Q7dWH;WgL3s`JwC(x zS5wU~G!`Gs>0EfR09xb4=AOGjKCXSlpd_ng(gIwo@nCDc`X8}i^675)cP z(ajn|TCf1NhF~8v2WjbkP>bXk&(G`8P;dKqdMPk3_l@<>lk`x;GmXh^6d{K^ zeC`7;7jJUL$De&NDu?QG#`cE8D?%uLZ+qkgK1`ML(xC$A%|sTm<8moA|Mf~nrb-BJ z9iS7cnIhn`XJNj^!e^E4R5oHuOIXD;GlzCQo65#gPG2>R{pUd`ifM3v1ruNL zRHI2cEOhI{B7T@;bCh>RMyq=Oedoqe9HYDhLt#9&Q1}*(8E}1$MS?v;E18nuwCl01 z-p)`=p7M(dSM(JG?FPsX+GUlfwbiFfwjF)11&Sw(F}cu>6+W`RS%W$ zun&8{UiC_3$!!!7xn8WOZ!aYZu*$=|A@7ZDPH25ubk}TfK2`SQsCShsePGi;m(?Yc z(WCUuY*+mgF+;B5VyupTge;mW-Ik2Xtmd~jUKDMJ$|8+-bt-zQ22Xy9M^ft6>z2%C zS2m|--3qvgHDe@O5Q8_`rBZrOHXkP*J@rpmgBAhF*mL~QBPNrU1{1h!bVu97h4s4j zL&$k5p73q;EL?V6U?Vd&%@Uu()_37wCR&@V-=xq%qCw$`sTX^HVj8p94Cu5Hu%lRk znr$QxD0tV-PF#Xsp#yHNn3Z1$5z#4rk;0554}AvP-d55AhC_lYtHQK!w@sugfk#V3q{F-$LCvaW;2kGM%!@HA;T^Wl zCm1NSYf$f2!eSxSTRU?Br}4#4`WlJ60~Bj@bILDC4T3U%37?H9%slx(vJETb8cy0g zswEA;q6ApHbMWEUgngxw_7n2!)I`|K|G0F%$*uRkq3+;n9`0$}nlZC>m6Nx!J{!f4$Q3QZpw{I+iWqRJ@^R8GbzPeG8UGUn_1g zyYkOxxAr|~AJ@1bcc46{CT*69)?Zy%8NhO?Yjiay=H_vG^m%Sz%Dx<*Xmbl}F;Gye z9nUU_pTy;iSxy$jGYR{UemK}LH@bT605UV1Evhzud^tG2%h$=uEkeSqJFbhOIrc7I z<55Zxd5SS0Juk5-O4WSR39OrFl=&5Hf>^XJdhMZ-0HM`^m~%~x)5>V!EFC8;wCmATj_;(@Nj@#Q~d zD4Q@-l)1UfU)1T%!#fN|3} zDfaHwj-<*#d+#!Y%0l+VjVofWURjf0iQi?vWl?|8F!i09lk9gLNp4SZ62yH8eV~x7 z@IC!}frc+%(B+SrtW!GOTgscQdQOzDshjM4Iv=C!XSI1G;@1*?taNRc z-uMWB%WY~V&&ysX(43qsWXYd8>mW~TrbNu=nMRM6(-#+9K|COlP9tTizSrMIw)oNO z^o%Z2ckcxZPY7^*Ii8+W8Ix7*t8`bpuas|e3W0g6GG=o^xJjwd;#-0{Kc_NyWu5qq zQreS>soC9@Hs^a^zc|9K{3L&W@>J)fGx-FM+f}sxfb*lmdV=ixB8A_H+20^GvqUBz zpWUwQqekcMrfB3kyA~kJD7yxQ)<%!FZ{Uz+@;rWr4Y`UQHFS_K+R^aYaQmC2k5(o) z0G=os9aV>NxxVEt9MGXr80O-;=9C;9ih8X0+(KP&vF1WY>ViXq@$SQaq|g|?)Fl$P zoxF|-AD&5yuR6ZRtVV}VcRRD1eoc8?Jm%RvKU%M**HbB!X%B2ts&RLR7ZP))p09Eh zXoKU5)A>h1YHH^4LL2&>Wg^2U^;+fJfQMLv+)QcptiWkTQdaYZ9~a099(xl*fXL}* zR|66ESe{Bc;H#cjUdS|mxL~y9rqQb3mCAN0CM@ORm)4K0tYh1`uMcMz&8Yt$_Wm-e zu4ic4D`$87(7Tn!ExVr^dK+uJ|ySux)ySp5c?7yA;zUTh-y&vxx zqX!=tYs@*Ty6UN?y1Kg}*+|DtRF^i)c1H|K&ARHjath z=8fABLiE7+@&*Tb3Bcz+MRL z&K;xe=_*R4-kg}vLbSdHY(g87V%>kTK<)@&2V0G!qI@2;=NmQGZUX zLJ-{CUI>1R8^Wu9Ltq}CU*$yDo5qhPXk|@c8L!iMmfdV?l5*^nJdlI+EXErV%4&xc ztXzg<`G3vGW|2$8u*Yd!jp>q)XZ3Z{#~e~QOGh#Y+00k#hb>{I%e>~CBW;X^Lz(GB zMHL)vp2?>>DBxtfZ8K=E7!D?xf7@*dk<4oa4_dUhJoi_B?GM$GPF}mFds(j3uL?tF zWwY_ABVWGxV6mu~72-;Ymq;XCqM}PV4Yn>y%J?kFF{(9VsZwmK(2FQ6``C5NV;c}U z#+)}%hp|PW**mpe`TZ7>z1cKPk)8<#bg65{#u@@Q4m;Sm|C+4sw&tWHg+xz9w2$xf zW@M|z?7bO(T4+!;*vT_)3J9L`yE)AF$Kb&WP5#m>D*01Y+`N~FE_@aJPrO>m@8Y4t z0};S$@Sc<;Tir0VVZfKDLuobO56n#G>-Q&w;9T$NEeKwHG)JHYd9?yQON^oa-}YKj zS?NSUgkW4+-*58Wvzo}eM|C&@O*{9(^nBlY$py)O#_g7iZfpas_KSWJ4M`o&exi+p)MMx$dq)zVImO(3;PwKp=w&?-&^M-g1B>dvXkc~ zlo>}C(xFU3Ml4;-GDr9n+x77fGxw!K$`e6v339U5?ELE52hQhgOX#2Z+C1M}edGT(dgzKz_2mUv)59=*%I2qg~+9 z%K)rc$QP0`S;%3trc%PDDb)4~%m28i)9g4!oh$FbANsKx0`c+u)JQ)&Ut-#lUB&Hp0y5@)jCZ{Dn;CN`n|hh zRFlzLTA`FFDX z?xv60wH_R7Jo)BG^&ZdB)w?0Bdv(2V<5gl@MfgB`C2+`BT`y|Bz{3FIwx5T8B_5%J zmC9F+C(nBLh7-Ph(X48c5d@qh+${7jF5$x=gQzZRbL8t>hK_5U7+bnG%P;--W#VJX zYvWg!Y2>OX4Sf12#;4J6d~%#$2`?sCJBi4{x6lk=(x(v3xmlrr`NMq2wFSCG2*)ii z-UDqxDfR5}1=&pD4N5Bp`Y7m+bmgmiyTXj0Wq{N_93_o6KII%**1ZM+dqBDelLWJik3_d%6 zaM;bz%Ur!a0J*v-b1&D#gW26VI6e2y73N!Be%(3-A`)_KkAH0gxt#v3eTO0n8;E?o z;JzAoJZIKV8oVdU*D7JYI$9c1#H8?sKQ?V^*WZocd8xb#^CKk>Keko-uCs93l329h zFwPm}AVFxZ%ADEgY-=okgpgdB!R30rplDmq+kCu)Kz$P&({qh2Xp#!YvDw^&!TQZ) zKFQ%^nLK%HR=+>Ct@?0JA|2xHgK=A-An!K1z1wnm3<{4Kd$81oyBHwGt+-Bfvo9qk z7?U;^-;*0j&x%SI*k^=jsV=j)#%A7)hr_K?sO@MVLf0WS0TIG~r;MHvK^fz_kz~ai zg+eitd>G-Zs+Gkt@41?i>z9%UBv=8nF{!|+VGbwrIiGJbvFO$u?%5+#K!ELu5DZ14 z!sQ58u8#QZ4?a%p^j#hD_gL87f9I1wN!7S~<$+3wG};~h*_~;gz_axpj2Q{kGq#@r z>z%C_Z^Mq?vCd6~Ek2i-en1+Krhpfb+LJY&_WvhEgq&%}B+QVjV2B zM5bDbMR=ajFAs=|cP!>5iw~&eVL`Jg6t2bm?{D~ZA;E?p5(%x zt`1l(D{7TaFh(e#yP6I*^Csvf0+KW^ihPw-%JrDCjN=}EYz^;?L9Fo^+oKd#>u2{D zY_o$ga;SvdHWG_1NI@IUzQbCzM#VbH+YkLo3L!^$L$#=o!1waX|5oCb27G!o`y;WL z)4(V@N*flT6tl%B-3axQqe*@RoS8k zZ0;Yr#2*uXHSitbltVj!oOr)aBTF{B-f;JrxV~b(5P<&Ky|{=O9&~Bk^ru+vQdwgn zOx@scOcZ8;b?KhV0|j*9lZf|r&gC2Vb;f@9GlF_n()j66$nIcVp@zgKqU~DG?{I+)}h8+X1B=1UbzKru3`0m%GXnT7Okh&2{90XizVDuNhT1? zn7IfVYylyl_h4E*Iz0XT>0je3jRS(eTta~y>sw+lphn1@?~E`YydSW!I)BeO(Y;2( z`BG!*s*~?-qCYvUD?7AU2F^}n2?=aA>G+X+#l1sES64SJz2truSd8bz5~zSNWTDZ4 zZb@{1s*Rv~50=?t1cxaKrDmqz_g7dM-Y;4_v0ULm7qkLayp-DCN67`?v)cW1IJ_NO z8vy^_Ewae>+~}TsjzyDgDa=Yy`OBeQs4b<|Z3`!kLSMew@n}3X>D%3Oweu%fI)4O0 zfyZ`vLILQTOq{8L{gA~}!vMI7km+h#Q%e1R;9or4^*4ti)QZJbRXsZ4a^%+yPd4!< z*ls70ZQe8vk7EJgcEq|O4sSdaC7y2)fGVpo{8t`lE9P|eOY7laU;fv3&HiZ0%kj~N2Aa`gMS~z_sI-VH%xVL!S_}4nm9uv&od}`@4t3Y_tFu5)p^Xcy~?Br z1(%S+;kop+P^X6+IFiWd%H02ndxS{jxwkYG{{t5at0agsr8_i5&(sIsx!OY0^yaSA z_Q6Fgo@sYR>UY^oR-bC2dM%MQSy6U>?tBVi#`wX3Hcyf|vvj013L4(+>C`?8se(Gy zBPG+BdQFz&Y_1Pckipxhi3Nh(bA@N?c1X?vkNs0;7&H=ox;cl#*oxybv(jgvN{#0T zb1)itb@lLmYMuRriC7$!-%@om`l~20JZU}}XPd5Y;9rX^FJGQ&4e6gARvZiqHgtMl%rJ~gkH7In}f3hNnmsd*4`%^>_e2fO|Dhspg1 zZU^%2<|{Sht<5>rx`w>1LmU*S#|~{^M^`(Q1OGmYCF)s39A4KFkHti*zqxtA5CpmA zSc@igJ$>`Z9jAhgj#qrTuM;916^>h+XjVy&Athm92EG^^5}*v?XwK2 z)%s-^XgxCXrqCMwrUG)lG4o{_m?p(jyqsIKZI=MLmi6EN+V7wM{A?wE3qYgXpB%pw zFwOS`6js4T7cJ*yqwh$N)^(5x%H?`byUh&m+I^LFllcV~jNlx$1yP-L2{ofW})iREoub zmgnn+qTm*MkQqn$&o!*Ickewce|M3TOs6Y9+aE0g7rcdJjK^`iFcST6NAfY?X;mxf zljs*i?va>soESNHp`usY#U_!7yU8mx0;6?$!u5H0$w_GtspYdm+oFhoi>!8=ypblU zEck!65U@{-_WwS4^UI3X27{(^?b6dtW1!UL$GQCM=Y{iYkFO?!kV%|ff2?Fu>Vccj z_>xN~tonl>5nGiQbHO0g#CY07k-drRC?N+qkGp#z;Y>7XD-PBpG@vO0PBR*>75UXn zrUqCN>d&gl!b*x(fwe-Ukau7@P{A+gh~!M30W1K?4&{=K|sBaIJF#Gv{_wxnbT z)*u6GTi?A+f4d~6)qY2!&&0{j@2!L<9*OM>Wd&=?P0(h4Zq`9OT(D5v#s=c{?+NIa zhO9TcLvLj;FV%WLS*I!Z!a;-8JRkW;#g(ej?0#yn`TP&}T4Dnpe;!v_PhjG3+GVZH z#neoJ`}AvA9NZ@YI7W2t?~exq{<_oB8reaqQ&@@#TUC~(;NZa!VyOCiFElyc|N9kH zyt`z`>JVU9jR#ZXR&O6gQFzcSzVT{}In&NZ0?P|NE=rB-LH~4HRVK`%)0Gv#cuvN2 z$GxoJS*D4|*+xFXe`N&nS8I;G32<39%f@|(h*Uc9g3q#7T#-_Y*i`X$E_TwgQllaJ zeT;ciT&wYQtBl=!E6R{{?oQ}ebaFmUHFvg){Vj2vnYoNag{@t8F#syEpa_iHuB)&*A=p5w{WK3h9}a4WLUQ&+-` zc=h0&uW2KAe|Zer?G(K6EeY!B-NI z^eulZlF$80$l*dHBK4Q5edsy{Ce8VPjqM_qe}yxf4F)s%#jO)4yd6I^qKhyFG##9x zp=V_>#6=WF+Th9tkLiU@shFzS=%iiH9nSTdIxuj4fu6rII4HEp5wGE{bQy9{gKc<{ zFHlX4FHtt9-oS3cP15%3@&?cF8DX-faCq|7;$e3QhUzUPWjpNsn$^w*OfbLK(G^C> ze-6e7^)?lND*NKGE|0_F5uC2fZjKFS5AO^YTWT9TUX;VWGMV-K}7@53BNLq0BFNMc#DBB}uopp8jO%2HfpSo zrOG2J%CL|1jt8QWI8!ssu9qz$vD8{$e>>diwU!r|Jc~Ezoh}~^W;|iCC1b?U7Tn{% zqj%`r^ad`?0rUL6r~fSTwD&c2{DotC&KYH;@ChQ1j7`4SWZuG+ZE zyZvU^R9|&;|BJp96}ecWNl=_IucStQ0nubBr=nSuBgf4Q<=^i`O!winf(2F6f2i}N z<|JLje`+W$eF)ySnrREVas57lsLHMFKkerWK$DxHB>u4SbPm__h0;BoX0ldGY3NL3 zLG)&2s$$1jeaGWc*bQq4QJ70mtN{jvc-CYQ{dhlf{&IBr!WRK6d_Dr_S5S@_#&hyq zg9Ut15<`!)8L8FplRCe*h883+e+amYmC&K9WZg!e9rnp3&3XFrl^T%JS(CzO2aq_3 z>Fne^!-uYD^}F}&;eS5HUd~{&hHE-)XKS2mJ?=bK-u^CJ0Ub6iDA8Cfyg(LPVg{@B z%{>j3Ogw{4zHA_xE&OOv0AA#M-J&6NblSeX?MY8Vc}mK_%-H*=?eNTvf7!P99v}QX zWtgj-DaS-)JXnx&X=er@B4$e0!y@;)AJwvtJ^U1yg4VTSjJf-GleR@u=}=pA3W5tQ zQ=a#+KBUdX{9Cw_@jt*m!PW%A`#G(+WnUmmg~ z{`hrOZ*b7tPlKjCk#T0%rfcEo!8)0*wAcOui%!T*YaF@-WPeQ*5rGTOY_Zswy&_?& z(HBSEkUl7<9k4iyf6o_QSddnJA)NZy^`RISv>dFV7)=H!EHPmjp)6#L!+^f%2eA=f z?M)4xo=?QJ290c0w%be4quOkCW}WaBFsB$DRHxh%(SG|3reNxMcP9Q3lW4C)43w^Q zLN|hl-!O3>rgoWZvi`+zowC?$%F|YirWng;2KS~9&&_hnf3>ip_f+3y9c$y*|1Wra zjbS!#FruSy;imEbF5}BxC_B$)tFvydlBJd}fuRE?a(~6SvSh-Bdp0=C>GD=ZLPem= z{gwTJ+mS3uX32@kve=I`i~b5zB6ppk`IY%>#f_b?qGUxG9l81W)$b>JP5o&%QXY!< z#CUGA*s}g?e||YM2@^*gjxr(f_?sEb-*{$9k^P~*-hbpU__?i`L(e#pe8Na6M}mcD zzJdzxHzbpz{!Ku2MV?YF5Isb=E!!D4ZeSS>^mvCi*e;5>*SFfjfq1yIc6x$Q%wlzF zHPi>N=#X=O)HjrNKQ;X7nX{+cev2YsZ3l2i`7iFMe}?$k`aogD0FIaSTS4o#^e)e7Z+Sym%P7^>+|U^{)jx~Yi?*cQCeD%u34Jon zfqEJpSb-oXz-mw~N-6tcNEz)E5taR2b107dF2Q(j>jLYtC+?uq{7fwQr=Q%RJr2+0 zCz_3u$ytqqWzQ4mSB{cf$Cup~En_QEPmY&5e~V!Hl5^DFw^OwpJ7xTcEEY?YSgqyw zqRa9L$kl_Q&{#02k)LZ7+bl}29ob3)57pJ$-wBDN3pYjX0^?Kd!gH#kEAwj7jcjC9~! ze-Ac2ysI#@)A-0!nkxl26zv%mKOcT*$!b}P?-?4kpn#3cnwJ%XZvVg6Sl^eo25J3B zK=%f5tScSei^i~%UhU5bp>du(C4(nY+P)K6$v!f44X$7r8`euE4THeXG__EpuUe@Vp*=#~r*L zvO&*<3_&tE%75Z5Z#-TgkMdX$=$(=A1Imm-2pO^=Ff1b@mkC2U3UOR83l_rrav2tX z{N+S)R=0#?X4?k#d0U{op@PxUpqZngM$h9WRfe}yr0IcEe@mZtcn_U52<(54f1svH zOAyRzdhaNgE4CFZtW;P%f`DVj65w*<66t^H#yFlmkf4xT1^+1NZd>0VF0zF|&crkc zD!Seo(h?luzKKSC_$d|suu8jcrG#s%7>}<00h>8EM>0)+N{TI4vxo{|cO@aox|e2w zq&>Bz#vJ^W#)6^KdYgeL5{LUxf8REK{^d3XaJdXksIs*a9Qcb;%R~5^@~Uof=YUii z`TNYFY^!DFE$~SH?#seBWP9aOY75odp3gzRk&I<-TB*F6Jwm% z9`^@e(Fq6)h7})(EO2fgf4{3%IwdZ#8A6c5(#{-=kss{?gkd}dOeXFX#H7vWK6x+U zRN>6I$A8q%aeJl&%(83j#eF6=?j9s89HhOqXO5#*o2MFaEA{9gyMGR-djv8^prWA- zfhU_eZ?^6%lNo%P$VaHOxOniL!898ABu~8%SOT0D26k8b-|8Pyf1!;=mfg3_?hhxB zauq_7J!@lb3J8#{%BYAh0Q{Ve$9hj;6q%7u@#rT(>(&FHtZ?6eYRT#h2euZoKKe&Q zrN!!8scs@@ z{dUwMZ@Sqos+H$tFHZ+XD844JX^Rexf+T&Nv{Mo<>9ijKgK7lr6}EU z?p>U(8e??fe_H61Hk{vuY8KnxPoLr9syZ$uPF8@kQuk_bE5Ph%zEWW+gotX8BCe&0 z3IHQm(Zi70NZ)NVJ6+}13c%`P;y7K+N&q{U*82kL4?n&e?1v!V({)+a4SjB?_k#vA z$vz*xYH-!$ADf9jv0P8{+Nh&EwqImM?s71nh5 z5|=<^jGMO-)Ua1tiqIypq~H2FkqRbrJs6_4MbOMIkflJ7m(fB*X?bWEO@-22R+&yj z&sUoZA>dAD!ei0pesM@EEj75%QY-53Z@Xd4kws99Pmam_^3_IDkov~@wq!h8T$Y6T z77344f3V$PsUBf_-(+Q)F;=E^jA(Af^GUmA)I64@+9%c1Z~N4{JgQT?JG|wE39)Hj z>}M*o`AN0fBU5PbPf_0hp%Bb;cpBFEI$NxxTf>BoF7rIO96QQ;Ha~6+si|z9 zSg>ok{*Su=M)x!u{EW}~?RxO-b5Q~kDLcHqdIQz%n2+&>5})8{H7s7zvT|rh9zamo z+nd3yeNhQ8D5jMgT{8|xs(Ga`;9FcWwVG$im9j`lL7{M27YK67g>rktdC%1!3f~gZ zf6mqCs@)r%pKy<(dX@>83m2lwZ{{}fs`kjxHCm|6v0-?UW`kB@racb^&pl?B^s0VXHm0) zPHFiSkob-W?r^$_vC5=x1!5gp)Aa`be^-@fq3+tw4dPDnr1dLs?sz+WUSSznn_j&I) zI!S4)xi@&%?O{;rc%!maumgS^%jVmdb|5Gw2`z@nQM*tt=q6gv>ZS>W^=_X_ITf~O zVZLEG{$~DoiCr!PQ!*DO;EAJ?f37xxMx~hlu&*JGEr?Y{Ft~c}GH*h@L`HnU%w(>S zGEQHrQ0-+|m^SHKx+c<1DYegse?P-l$oGj63do>?!UmYHuy2^(zb9?aDW9T}EgtHInI9w*u}_-MZvar+k9&>+x%_l z-l$y8D`pa{EZS*szH;5MzTNX8R(`u$?DBHK_$Idh%M^FE|!{Q!`T}dMDpYyf7`Q3f2b!2=f@ON(Fz+| zM6$g3Y;_+0%ibLZROqZ%N379KqT%p?S4Cr$ZPZU`sKa-p8yn0uKkot?7#y5|aNu##t0){t22AC* zY`*dk7GmHe*L2Nse}C{(Q0Az61(5$cZreuiZxKvlVO(4 zNLA6sjFoXx^E_?3lm4=nKRz+%$Q7S?z)vaPbpLv%e`_*c8?~jp6VCD)!2;-)eY%d_ z@uFooe>z9|*0}-ucH%N816J6CLgO`~9c^K)1OL5;T;w-cz3Ns%bJZ)7aCI?u*xb)b>~HluMneambJpFUghT}cIRD+h2B2zG@}55 z^sw^8e>#k)h)@~+$DmU&$nU7BBPp@1T0q4?l~Z(dbU~pz_<{90fMaws!M$S%To5Lj znl~JJi1kB#(gTT33dwV{33n=CwPxGkmYx)%H3)Nica?Xml~#PZ``)kgef| z;X`rd`Uw%KQMTuGO$rX(UZvV!aIlmIf^%Gaf2&zHaQy#gU`Kpa%R#%4_@1$*l>|fG zJ!ALx<$m!%Upm!`2g1tlYp4U1PFCf=q#@u{R$UVWq}eu{D-`FKkESf;R_Gkjxwl%| z1gPBlyKwUj7V5*p_0uQtQUz*S0rRnbg*7YKJJrfLc%IZA)0HLTSHVP@ESk<&-|qJk zf9}Hl<1I9;uenJ=M&qaztdz;9;psXX#!ZQP=KW}FSdpO<*X#B|MtE{`ge?7)ZKWmj z0k=6jte0}?i+7j7tb?BGXCg+Uw(nwzO-lA3l-aJJZ*0BE18 zOJ=t>5;LNRa_-WGi`Q4iXX`ne<3*98NJA>IAd?{ygA|#)o9>a9o5aAMcn2=e6HGa zj2^Kd`7c6}$QuZ7p`)bR60su`u=yD2(B}=z?&mjNF$CfM7%-p@s48{^;y<$n!oBt; z`{d+bh84Th=ll8dXnxiw52v;8R_!xa=TC(@6K2o%_$f3l16!|2C` zr?K+MwIrr@VvNVL0X~Mt(e~e}$6LtgUc*Y1l3v19Z&PFW7E&Wwu(8S}PgrzRsyJdwP zxYI6CzM0#A6*PsRiQ=)@_9f?QUU-GOCfQK>08T#>SU;rExN(S=^F9qXY!q?&v*if0b!g1o&3u4wD%ZjM^f5~L7$c79Uq=%WavC%12e?Y)L&k*9> z$PL|`3l{8DYd<7x7B8tKU76U4?`jQ8v9-1~X?|>PXI(gDs@B)1tKPEW7tnNWORu4) zJjrhb^otqwq7e76e_ri>=I=mSCL9RK<0^y%E;ZPvmsy4)%OOrAP8jC$z{|{(tn8jX zGQLaA0z+Ra>~OA8dJkG}@MP12-8Q}yCe+C|dt7_&Tf5w*Ll4e5Jj|_mFtXEbVDiPY z^)*##FTtV8cyshEJ+!8`_L0fwnLnJA23}8$kg4gHXx-one^og0OU1y52_fqMiM3qg zTsW;t*r=DDLa^v$H@1%s1tEw!k@@_iHW|-+ka%(cF)d;&lQ}Zi!@^g2D=`sc?py?4 z%O|gwYDDA`hVcu}&5#6JDG5F!+;rJ!HZB&aRP|QpPL=r*+MQ(KAc|8RZGSrGd5{``(DQD`R(cndOHF%PRAQagKd z#~J{0{Lq=X2j?7ij_PpCSWTTSvIUPY91T+vfZ(3EKUut2nCoJR$Gwh_G_0+I@6J6_ zq!yCne&Hv_k&6E6etMOAAvfRe)Cn1{xstA)%o3Ufe?NJjBxrer3y`b5XbtZEuJ-Ff z+py}O$g+v8NIro&mZu%;;GF-x3gAKZ>UC-KKM0coTKkYj$4FrGAl|}&F8Tdi0rXNw z({`{ic5`3fQ-qjGA(SF+$)AMRUk-)|8D)Z~7Y3AyM|Gt>lv1fybRS)|{{k1CZ{i+A za9+nMf26v}sbjXQ1Ii<9j@9u}49+^fIoK*?S!q)#=l+4^7tFG^D7gma@kri7e}+YYoN~yO0f&=;?iT7dy1;bY~1>&&AVr zBU~EeSD+|O*YZHtm>%j4?mo_u6fTjo!y}3$UFe;N@Emq~6B#W6TQ6^@aYpUFm+b{? ze;Pp*BewXIlIGh`egT1^-?k)LYIAiGfB41WfpzObingbD7Ntak=aDtx`h=qw-a=%z z)vF^vvE2_|CQr6ZzF_;PaaJ)Q-|Eps<;M+E5P5I}Pom(0b1`Vm^@a0gGqe|QmvJF8 zmf^Rz?0TY UOE`gX616#n`(qb>!}f2rj#HF38u`vhmVhT&VLVpUgfDj(7+*ZzD~ zZwcCLUT88bL1pVay^y#k3F)9!Vevpsp&{bo%(59JYlwDzGlK8|3R)D zF=@hFfm#L}ezcdKdPChd`079gq>acj>47pSq_7B-Bj_rjTn>+{eePOWqS{pyfAks9 zfaTNJuM*!!%nFoBw9}12^$9-S5YF6!7hLga^OYuiNY}Lyf}1-u`rrpi?wflqu3i&S z1l&nEe!2n)EX2hoq)kn%5%z78jG6GGg`|qYYtqI;vzT33An@^0097U)4A`E&l_FDC z@XT_QLX9dbFvbbtGUGZC65h;^f8zES0a7>wrg3RWrU>(tf1Ei?`9ebCyFXPRaC24x zHcyTho8+*WR4$<7oVxkSu-z~Po9xq85?+eR!ctd}j#@H{Ucvsj8xzzommQ^&uLp3^ zo&2&Py`bpJBA!rUWkT;*EZW?oOW}WFJEv&4mJ1W#1W0UKjx&sN{?P2^e-Uh$mssKP z%ejSaeP^C6n1c|!4Qgx~pcw9YeTghE@b737sxS`bDL1EDJeSnI6TT zs`)lwu!e5ax6u4L;&2>yuGUD^SjECfNB5NJTzD8p^T>Vo@15*GZy_s)C zUfc0$RHmI@mnErF*~F3%o0==@`?tuWBD{hgJ*?rColW7iZDkd6a=554KYnn%MM(s) zJ12#waRTaWceY0=G8}X$9#nhB_&ybs_gu*JO!*=F7wy7sEs=k;?zLE_3}yPZ_DQ^Z z_+)kP`X9hamGA$Ce`#afyoA{3CDO#g;R-8=SV&Tt!t6JdL@O4>(sl6)5lJlO5;0`q zHT@lL|51m>+u&F2G!rG;AMvlo+Ls`>yoXqReQ6)A%eDPyt8&BWY@0u??US_{-y5vA z$2IyCV9wD-fELQMTle;IGR8erprOcZ;>9n)fBcY*M=b^Sf9oR#qRJ&t0Vm!+YcW~! zyNwfdxysCO4u9Kph55CKi6kl7LN=?Ns_UkVOem6M7juVIaY-Ix*^8&YrO14+N z!Gl^{!?}LNvUiZTGN)j)So`_PqZT@#ut3Vo%L^ULG4+*C0g{=YgtUNn@rZbwMhfJP z$D;S{Q>gt>f5eMagwX%25h67NGV5J1&EfBUm%Q7kuMCoLosz5o*NkKSE+ zwcXx2nXH^?{$7Amnoc8R{gHZ@+&n{G5}NsM`^o;x!TRYnoV}yy{oNbW%S`kQ1bGt{ z!s{Q-DzynxiE(|Ahd1ter;By0`&Xh`^Ub-x&e-wA3ZjeZ-mn>*W5wqF2>iDfvh z;7Vyp-1+OR#G&Khh`FaZ#lF9ERrf2i%|`71XR}NP@!880g+Anf0h;4qA^cOMSbFSF zn+7G?e}n(-_3ru)gL{ljy3=N=PY)&W3=E+nRpv|&sjGAyp^*91?yvty z1A1^?t3OiEKU#g2o98F~`xVWfBcqd&qh+aG^wkrF1rqQ}-e;yWKmu_--H-bg2jvWe z)AuA{SIf!D?oRGa*qv`0*Lxh*n=o*Sur4W9e=%?}>vu!(;{Fp;!bXLGV-@CmE5bnI z6Z7|9D#^SyCSS#XwCf7w_IsSKBAa}{p%PfldYeHnxtnBEe|T$CSo+WUMZVZWynTCy zj#|3a68?4r`i~(=e)?L)o}J_G{O!F4JCcS&BTlG)KQgQFk6m%Jda@)sj%GY+vwzs0 ze;64G`f%%uy`dMx-`mY`QAGhmJ!^aYPD{3bA5IXN@cxt4IqI*Hf`dQb?$TI_^iRLt zSQAvu;6M4o4)?cNWI49i8!?HZ?W!9x>`J#0|kBPD-c074(8^YX6%*6 zVCxG1qj`NDoe*H%M_WjWf7M;ydTmbX&{$Aq6oKpQ2 z_o^Z2-D9S_-x}HguE=2i!xFE!;7#h>)gTmbK&o%N8j9!_MiSRzJV}W1pOO z{?#V0#xJmkEAZF&u}KTP|C#as&+!q`?*CVn|Ifim@>j@!j#0%|us5f0*A@L^*({m9 z3jfeL{^#)d@yAJoTOO2kZt*4 z?~g+p#FqYh=aT>D@f~?I`(jkTXE&jx`qh0y8bglue?O~s)*Suc?=Ag*9;cCig(N@} zQ2Pc0s!YEO`>cT`nSj3%imLdZw#NPd@y9nmKAq2CLXdk#IhJYsc@|bfn9hGD;u{D8 z)W1S@Xd4y0_ljYTZT_0|f5$_TW!kC!Q|+RtDE%abkzlao~2VG^KZwx z{y&UXfxkjJv3@Y0dZoMm<1GGml1v&7Zj55eQ2(L!zrg&3fy)6cH5kyN&~lf9{qJX~ zA2J7f{fBPkB>GoKq=57y97y22^nG{apI~b#3|sJ@YXALT{|@!Ie=~*mifQR>m@WSE ztdJ$)um7nVA;JEoLTg_t6?|F&xMC|G$NzAVAqwsPuvtBC>!<%0#Qt9p`~Q0otEaXS z0|3(H42KJjw}V1|f!I1IKYtAx?$XF&+!>ZLp^DnXq69^F`}k0P@9dpn8?Y9O8!sl4 zQz=jd(f1grbvK&@e<0!>o4~7-gMRxpFy7Y}_@4nzE3M$)@A1sES6U^}$$=7|k>!0O zGbN_4JQy1?ZLug!*+;1;(1#Hfvp-$0}1gDQgVKT$p6n|e*;tdFX>t8k7F{H zVL-uHv=Tj1sp-zeeQO zzsSIFTy!$xLTnYij6wG^%|xMumyeGxzk-r7uX3%#bGI|EV*lgkef0Muo`2lHi5qqD zANDc-pTo-Uujv+F&L!wKqOd|iLCX}7*;6u+bmoU((P`#jv{))SBeSZcTfgxBaoi`A zU+@1@)8Y8*f4J7*lV*y8u+}`$H3<2KYcYorF;d^$ip&iI|34_7yd+}% zpW+P5zfj)brV8sucvZey9yVV+J*$IzNb4G!l#JwcMk2#;6aRpsg&n@of9jT;|3ZFf zfl<{iTrat}VEtT)z6`{N3adNz;0UQ7t$m;ISs4C=f6v8T_Tc{%z@2#i6+ZocRRyy^ zfb-+}rT3Iy!9){TP4~}NS3h<9vz+*UQ@rM1FxkaVd6@oWWs|0oGyNwI@0)DP{0{@^ zEz-ZhWF_IvD=ulhB|rL;60+{m0smpX`M-n({1uiyV8h-X2Gl6LtM=}1=)pxQfvcm%Zr&HYcueg`caaPX`3DhLMEPDGF#=|kE|I>&4jZ`-n zUafsAO09s0KjH!FLni(?4$Q^!?pIyJx;fBiYqCq&-I0u4T?hs#QG8?bLXDQ)^-Fxn z^QcfnM~O;4Hk*vJIg9t^4}$P6-wPZ&Dg=Jwe`WuO{ON__%a=%RC-N`S;Zuo)%kCxj z;6EZN`HV?pLy#jNyBqWW@#+L!Kl6WV3|@nzku_|-7(U~GrR>c{$ogZu^hA`Zw6U&Q>%L@7S2@*nO6&$%C+(!>l)l7O<#*lLm{sc^4mj+V zu3cA`%Y^?4AI$s##sA#lvIKY=zmXojgzfec#`X_e&60WKtU~^F30vKf@FPcWlcmO7rd3op&(n_ zfS*`d1NklZ(qsN0DtAjH5a$&YU_W`LP5+4mEJ&6AJQA$pd-*PfP<(%3AT>iZjzNKt zC=7PYd4=SOQ~|K@#j#xFPN#l-G`eV1RFr&{(Pt_dCjay80TI%Eg&fX1IGJ#De;IT< zRDMbTEM9-O5U?N5bqY2j;|HG1vORhgUn>H{XFeJ`xH@}Z`#*ZGwWf5Fz3cI(1=;eo$9VX1l^4ykg z%_gR?K;`=(ORUlf(bosCyX<6Bf35OBdU_s#Ww#gNWP!3R`xHEk#s`&gF5TwloNFZE z5lpEbX|olSvrANlp3?F~qru95onC7-(jhO_jhJ69wVjmGF=!th886GHWr=T>XziG(haw0$vSjc zQ|hyhwru$6m;N|nXlwK@!Gm!j8Rx>X%Zn(njGVFj0zP@X>HPBP{G?*d4ne4NS{a?b zPW28~a9tNtUh8G@#hRvYe<5^$x9N3m-3;$lf3y9!Cmc9*dTQ-aDMqf2)s2<*g&Kk*33B zg|gl)3B*fB$HeJx&-nrei)g?n&j3~aBg<&bO}@xRsGksDh|NY&+D3;+qsh*T4#d{A zoT8Jf)6qb|$^`A|4F|@P5gMQGv_$TOgh(&!hsV^{WS0D6!MC^@KgUYc8ix{v#Jiiz zs*Vg@?i{v+L)}scfBnMc_7Q6A!bj#elloP$H{IBn7Z|SfIqYVwa0TWLq<d3?fYZH2(!NybeUA-}g@P8!i%x?F$ypHOGsc-5b+tg%JGsJ{_Qn zE;fbAI+WE?M*0z~j|p=!m-PmDQqGX0<{$TOnO0hLDc)P0e<`wieDzXZ5u`j^qm(Mf zLnlDZ5R?>x6PA1CtjmRgPiz+uP+mlrn2AQEwibCBz2#e+UT*Il``ONe}#w@>7bC@bX9shsgge2gZd+4Ne+2MbiOa_I@ISEQj{$ z^zoG4pFWA5A=|10_)4OthIdJdZFVjv%SOu?E$~3^b{H4zr4d`(kp?N(#`Y5=w^MM^ z`0~%G$vyRN<=XUR$X9-UOZAh7$8QS;jEwjeb7VPbf4x}!m;5!1J?&?&cI8X-{^iv5 z5)JUc&WP-oObft`ncRVZ@BpUXs$OI2{n*jhd9N+e+mBnp9??sIaf4emc`qa%l(aH^@hXg^ngD5*L_RJIFWR9wh1|K1xOBT`_#Kl&2^9RE((_VgA zCOiuypvkGM#`06*(Na9jVV+9R| zQ!`Q?ZYH3v=g~1SCBEKeQbsnhYUo5r3JoY_f6IPLF8@XK1BkyUDRO=I>-JzK*X4mb z7W}>J@GWXnBPk3;i7w6V0PtT%QQ`B?sL5#>(HpM`&wU%+(*Kd4WW#<;B@F}OWROV- z%@VmKwU`yf&B79q`_NT?isXXnq<0U2pu;yf5u-7y|7?XSCDtE<1|QA}fgcbfo}JPV ze=%A*f;paEp|mHa1E*nlBB_lRPsczMnn6UIs}N=)Ut~r}70DBPzKLlqX8{*7DDW9C znl2Cu7<8s&bhE5+x!_NWj!Gb=AfE6m@Dv$Q#xXp&6q=mWDnlNV^(b5S!C~ymbcTXe zNi%=tY42aML!N`MBJ}1c{{@u}|FG_0f8h*KK1;rgKmssg1C-Ay6~`K`a5K9CzF-n{ zHF(boLi`IZ8NKuirYFj$aPu^a@+Z^ni7zM*UwUAFc1fNij=e&I`3Y(@KY1YNc_Rj) zV}u#9*fK@D()2Bql@lAPlz4v#T1ay2Y|sUStio)qZ^4YwTE6+^bmhvPtQ&)df36qW zOJ-vfoAL)#u{hljIUh0oJhL(#bbNfFI-8yJQM@_|8b?ix96E|mi4>$>bXu&E6&@O5 zBEYm9b_FpxFV4EX`#AB#wmoJlmCUaL$+CO1sYiC-AB5{&wiq;82&A@f?0)4-vK!*& z>>6E@Be!4hkocbXm_`5Puq@2~f6#|&Xx=V#H4T2uWl29_$n-RN_ZK}cVALC(5`~*y zNE_xwv(tSDvPG|Ry|NNoi=VZx-e+7}~X&`5l zk;^RJi8HyhL#SSjz}81ciUAJCDE;1z29vS9@Gi_6HcKmyrRR!Imz zHog^gybh3Y5`BOfMe*qs<-@eVjJM04ZnGC-F@-O8vq``xPBM62F>%!&o-EbOy zj#Ha%;;MZ$eC4pG1p*popG`6U7Lbx+N2&jtP)z$4g(b)k(lhI=`|(xpI{`R)3A6cg zDb<8`?{XEZ#2!V|Z-gU%#tN!EKPxG~!69bg)F@bu@G{diy{vv^e|J4@#Ni7w!I)~7 zWF|n=XmCim#%C9PK*HPTqRn6^>&n$LL%Aw)Q|ig?@GmgWR-!v@7n*lnrNHO31HBTI za+q{h3~JfYq66F0Q`l54#0q-`7bzqUCJD*w^TBf z(k>q#Q(sO}1?fi5SO924sTLmnme&R~G zf>HdNhtEsRri2;|E?iu;@sELlAo)Buz(Cm{h{MzC`iQ4Lq(adCr%%ZQOU3rY{SZg4 zeFaCZJGJSFB!edq#3}w#WmsjgeSuk$BQ!80^Ta;2+uWeGf5&{n=xT5xxv~6lv%;aX z6eSf)6XX^=H(blPkcxZW+;2Hk{28AkFUT97Al~Usf4sYO|S~GDoK=)IalS@p#GAXlK}Siepp5v1-bK zrvO*I$$+*iC8XjYC(z@(JhI3t-5VVZP2l|5mOAJAe{+jD#fx4*)}O4eC6RA+^Mrae z%l%JUOTN@){9hYfEh+3zEW|EG?-GIDL@pAi`X3bQSvvt?73vWl5q98neR1V*jk&p2 z_&qsiJ`GtQrgtdz{79V1Tyv-!{l%3d*~DBeya`Rj)giwEbzEQk25S1kT{Sej+3dXC&F%SC0n^#RoZV?JY(u_wgL_a{rTY%IZT0*}R|pIK0#k-dm1PBo;qWO&i0RC_@g=m@I4S0qsL6OKnZ5bS z;G$jpY>V3#l{=3krTKC^nOiKdiD#b;?t|VGe~1zRpRtYT2ZO$0Er;GHvHT}3d4GL% zwmQA8w9AWjh4>8JfNLfA%J>h-DW7$@E39hR^zA>Y5&6Hrc5kJ!d>0}L9==t8ZRGxV zq$bk`mD*C$^esvcv6;P;1!IuVdD7>DM^t(3h*HK}=SSI-4+qr%HKkOXV7zIO>gMj?E&6rs={3{)Rhv zB(}MVg{BlWgS$0Strk(NmIIN`uOTyVB?dyE%Lgq4Xk%crHd29RBj>sxFI5Z3Of2$vbl&^0AympC%jC_-uGnHa9QCA8Zp9m6R z4sW#vMGl3~)`%3=NLu#V2Dw~{*jGEXAbvBRxc*5q-mZdg8Cji9?3G6&OkNXl)Umxg zoj!6^{zG{>70~0H68fl1uxMlxB7{=kdlcI$o#j+#&m>IP#hE5DMJ(sm2$e-N1^ zF*n83+czZRFxOrQ#?5tw{LPoKfk_gquwHUxyw^Fff9%t;+q?e1Ss>Mb{Rkh>{NX`o zpqnQDT<`6e;;MWK*Cgk7W{f!_HbKkFoTJ?jt}W4fdfKEDT_nB@GGI;VnQuyVIiX{8 zbo5>UcFXAWk!$RmOldDS`FcGWe=!LO*_o@$xk^j(nG##vrxI~$I^C0X@c;x%l)#xB zDef09U<#W^+bir98u10wVhZf6(~UoO^oVv!5lz>h98cSc6y+l(H?$MDbYFbq^4KqA zr(cgeZ&bz|A%J!`T$U=2qz14ut@^2p^d-<3l%DS2@YJX#4FyAhhdfU*e@;AAg^6q! z$VDK)5!`9|+9q_S4W3*3-%!`WN$`K4KCw4HtSyWDO!Ov^2}02>=%dhjo5%VKr;Vka3Ye=Vx`VXa2qh8x1H zVoBU_d|K79v!@2*j&4cxj|6-e{Rx?~(Pf7iRB`W{pAi-4@rAyLW9idN_Q^LZ>IUb@ zCZLvRW%?}4Of;J(&o=BTkXpgPz-8|vB+2a6n^Uk7Y2}ppwNgs1#!@T!o=@uikkYrb zHXKRaDD^W=H4A$je@vJH%KUQdS$FY6|R3VMa_T86xkqNo$OWp3^fOU zpX_R0F;2?ZXY0DL$yf&J0Fj8ajGAt#D)c9MSWxH2*!-1e0gqEiuF&%-mx-0uGp1{; zpNPY3VppnzV#$MekSD>R2i8=+=2-qhh*DvubBH4K2oNQg0-BHYQOSRXWw zGM!}6Ml_@GoCjpd%V+5;TKf`CsQhmSJ~o$h$qRGLf1i)k`!HKBXAG}+xPH4o4`C`%a=<9p z7XCEic_tP=^%oSMx1s*os`hF;@v5j1iF30Ae`@@&Ov7=dF|}26{gEp3uj(NUc=|88 z!>HF3G1on2c@J7BtgtodmTtl`#XS>Cj^DEH20XwAGidUx(f*F|Sm<*^8ipf+4B@_= zg3a4kxAQD^vnQ6UkZUf8Ldr#ADlU~Gh6N1lX3E%?tE*a=`0kHgGzt|jnM?Toy#W2I ze;yc2&(4$9#2i#9oCtujMM|ztIQ8)zhm})g9l})H6ydl z1%D@J;lqG@`L{`6Knwx4E=79eUUK7IjCL+muH(919u%&YaN1FF(iE-~o91rZ$B3 z-vhhdG+YR9fzn(9T3+*_Lr)62Trr$br~I0gV>nO6L<4i`w!ms^S!>@a)yW89_=_)@ zRBA@~{q>Q+Q)>Uc=3wYE$AUq?m%r=9th)OL7q7^eRGRSDG#l{{^Z)D04R1B^cYjPA z?DovTkG^%ZmRzh`BUnO)tavKRoKZ}Vemx);0Ls^%(lRvp+v3MDXFxl59nmWSvqB>h zf^B!`tVBb<_xIR~_h2ft2VVY6VYLsZQ?C__iej5%lz6Kk(RhBy;}p6Ombs1>Tu8Aa z4pTlTnmn64sC1|x+=*9GF++x8{eM_z%n?U}A~160{;7kG6(_{^g5BJfV{*va&7K#&~V!CvRXn^uln`bkt^6t22CD&$z(<2Q|awm7Zo0enm0Q(~L z&1TqVDV4ct?GBmb<4sz%s?_aat-S4tlWU4dbB7c4&_*?QM z(Fz4(N}T3L29&eFpCXN{VIEh2;NHXzogO6`Od2)ui-gDF3m$Jx_i7e*_N@mRN-NY7 z(Hloo@h+Y5<6Z8(*78IQX)YVX4Z-YZO(+COd@BffJEO{=TR zlMf=I0i9cvDRTA)8L~Meit*^T%`@+zv-a)yafS?cYe<`8^b!i`vo>aDeu*b)u6J5v zFrO<+o-Pze>*Bta#tNe7I{F`X0l;aG5Jm@MP>Q0Uu$++EwvaMydVdicr(qM(g;2&y zIUCt-VUs%u{&c$CrDSh|;JXW@lK+$xj76(w!GGjB-cxs>G+M5JJK1!1Kj7|s=wMJG z?_bmkU?)F8IZ~daQL~JONp?8l$?%=2cCdp13D#Da?{Mp@Htqz8u+&9VpDc>IBH_ts z(~#0AQ+lM;7Rh1gu>0uSd=ep9!|ze{>bI-Ob75$bW5q36_E3K}vhg8yVuFzH9gn z5g}rFEjuH=O+ro+g@PB9`QW+|^FVBQEE=$cqP=(g!~*rGbbtG#e}CAOZA&2CIfN_Brp8|}Ut?q)eIjCnW zw%Hs4cNAHdW`C-}NLOZqw!eV84YQ(lN{+gY2KL6w&%{>I$zIc2X|ztBx37c^7G&X@ zw_zIHpS}+~RX^ai!|WmlOY6XS=6Ug;$=#}lpIYT>Wuq$4sd%UD@x}Iv%D1%Je{Fs0^ODX+!G>QO0Va%1Ex zCqn_x?DTdF*lRcFJxaD*7IU1U^VJIox?=Mrw8zTOQ+uA-j^X2r;Xs0`7b83zd#7ue zcIT6n{mCpy3dQnK{qd433zN%D4}0hdLkSTAL<#iZdU}GW=X+CXe0IB^`HDpn7*w&2 z#?N9~zJI=uTS9xMQJMXUa}3#Y7Annz3?4kPeiL_7nE|E{bbEY!Nt?Q6Y-deSsQz1h zNQQ?5J~u-E=V$)x2OHM_{ntenFSYO=Dyy3%ZGXww!4A|Jyn1Q%oZD6Na4ELVs*r}GSJo5uL(!aVsb;d zQ?kVr8O&xe6%)+Be)O#B4~FV&uS%H=2fHM4*#)m z@Iprqa13yL8RLV6ULa4AQj9IoNdn&{#eZ|nQ($8m_L9U!oTB*op5Q*M%8<;K?`)aL z?NBgI-!<*l+s{`7js7cYefG=~e(rlM@D2A9^S~>4E|6dIzAM7?Y{X#gYj04;Lix(+ zWXVzBs)orK(0E2kA9xM`R?qP3&p+N@KR}5EQmkEQ?ykr!)fB^kwrMpbL*h#^e}8)4 z?dl^&?O2p!um-j6U?DzGXgh|IwEv{S=$Af$X3udY#L=Hvz)2O*bED^aD(%o=hXfLB zCb#H?xxAleCW25o)6>JwiJnLx?SevZ|#-z!i z&{pVi`}BFUIrXXl)%xd!K2LmCE`L&)$d`+X`55k`CRhE}km}#shs-$VlEl;rZ11HY zeh81AT(T9|W$@fnZ6K0eDJ>CLMIXsX#}WxuI;yms3QSs2N$!?PZ6j*ia4=s8R@NUD zgONpP%SzA*!O9UK*v%4L7REk9W^vwEx4-STZBjms9$}4Jb6DVUH6EX2%YPsIj+xC4 zXPA`!mYiHjn#*S;WblMnR_OTY_v(()-5doa*DDQLQ6FJ8Tj`P~eI`Bmq%qHu&T%mH zrJQELM+7+Pn&xK8=Adu5IUp4oI5G$j+_!l`EaP(S<`5&=KXTWCCSA6%(>xcvB|8y3 zV6m)mBbbfH(B244Aata-kAE#6Z-&YC)M)Gu&jVq#CAd8Y|6F0lKJlr(+r$5zelPW$ zn5%qF(Y(jjSgHOYFagDw{QasKn`WI1R?e}*2hB5A?hX95+&c*4mtvpGR$QJ7>o@Qq zSxV>x$sAsfn-M-*%a*Wx`NCfQBs{z8c4yUQW5$rqVY#aE>|BwW;(u81_s?Q$H$P)1 z-&~_&2&b!)Q?Nz~hvQ0(vB~9mTv4GXF-D&!(1=7q1NdF#5ouxz2N-k%J`wbm$2szZJ@EgQ|nqET<-7hc*b zgxS$$3$WDuRez|+rfVHPd&$(7aaYPhp6gc`;M`U~1`pJ;)}2j8i~CNIDQ!8P<4nUg zryI03l=aayChGK4xia-NNneI44`Q7GF8e^KOp_Q}@Q$VF08v9=L@;`PNi*Av+Y7V? zDr=sw`mKH-QK##)xD?MOKf(hMyVTL+~9vd46jKVJ3134N0?@)-H!s@W$epKW@qs!l^LCCpM6lip+2~dc{8P;~~te zc}?uB4C{a1AP87a&T7)aELI#TL_It!q<7u4A$iYqukZU6{p8cTG`AmF;hYERD6Enx zXVgKWeSgT3o#Z6tJGr)Yaa{01LPGh^Z0D+amFAeNT-`HP4Df{Bv$(>rnDT*G9M129 zu~+~^Np4vpAu2H~ zr5=W`{9Xr3rPHi01O|N)~`+g3MkHQgb(ah<$Mm5R^?<*}D92J=iJ!Xq!+$v12yJdYY{4CEK1?EePZhZ1F0D+dh7zY{ zdgw=Z3rD*GQYWGXNfcYXS{raZm*@SjV@H+}L_D(g0r)6}eGS@1qR6g(4WgbWvzhI0 zLT35yBbS><37EE)<48^@&Ow+X!3no2pD*|zcZtH$N*vA`%O!12f<*z#nIwG`{D0f7 zsl6aO?FP9!^%XWy`D$elD{4@?yNPm_Lw}D#@)S~ZRH~(KmU;+4aGl-bRkandt?kR|P$AMz zf^={nL^Ec(-6h7Hz4O#=9w-?+WiZ*cwq3t^p%#O{2TF7oBc!CkaKk9+sSlKW0QB}F zzn!r3#jhBQP??Ng6G~d{0w;|e!{Z6woZG73Oh@nFv5Yplhx!Ymos9193xE3LGX$;0 z+2+2&gEq0rbilj2M)?kE7}s{VJ=#wW=a1sDX>un`4`fwFW0U#K)?W;VN0{tpMeQk0 zh~s8$V*`h(&F&V6;!Wl^fjB(RT$#ZGYK~Aqo!dl1tzY^B9+_gCk)Rm)IvdiY^DRgZ2iEK3-GQ$Q_CFanm&iaT5rv?C zRrRR@`zE;w@~@w+Jp+Q4{GJoZpMz)QtWJb-?ml^5r+36%jPK zdqARHmkl5KxdjPMHh(nB?3rpr+`T!6Z=U2KD%?#d;e|XQ-TiTOdv`*e$htf~GbH9x zZS_1aUA3Z}6zfyCf`fYS9iusDo@Y;4BpDx?mXScD1X%{f=Xt)Bz%S*tz#^E zd^bIG{0Ze9tC)W)V7QO5PA7QzK_%Z< zYiv5^@eW0h|L8 zN7BEL))DD2>|?7rpP;PyfA)=byLbq)@QMl47Tle#)PFYLAI*u{(u{T>olNvqFdxvq z2_^a5QGGNo;Ar9ti*PaR9KI=kSsOuqd)Du_~K zOcp3=FS(z!=duvw&mR>}Dov*!1xkP#;(dordoDT*TdU_@G+#nt2KcH5I%R$OF;rUR zyd4O+BvyvhB&G%uvc$1UM&BJRNHYaw!EZWF*2?v$9Lfium-kY#v8DQlJ@~?H5AZb+ zoPUVdWx!pyeaPADJiyDK;|m&&VKYk!d?rQLr1Fg7YoE0)WrzC^l9h`(#R#P$3;{k` zcu9wL<<1Zk{bnK$))s(P^UI3)V%y}(!v_YYvjL&{8QM;N$X7%&;&$n)Vn#NTm z%F$?Wn3x#(kfDorF2%oGq*~XR-;DQow}12ACFA7HRKRkRi)@J|5>-?z2#sD$%kA!) z`@Qiz7rV=)>13Yny{1@pxcRx@DwFLt(&c?_nH!gn(M3)br{^sX4a@l126=9ePt)Gz z9;1ANgQQCgml6g|B<8$wa3Dp2){7w<)18pXVl7J8%c({P@?7cK!B^a)pEV5rcz<{% zU#ZFewU-j^pX~r|r4?z_f*`kT_?Df*SQIXO_#|-r0bqnMI-zjaKz62yPG&Z4{KxnL5g>RGGcm*O1=9vZ`fLj2&zoSMt|M<%TJ10jXSpSEB8(`ChN@6sQR>*&0+$9-Qf>n`Vlyb zO0Z~y`syf1eNVS*S$krOWFyMX)AM9!I2U7CiT%29MLm2dT|o(4u-KHTT6nM{f~Q2O zDkL<&%ztty-5VW`G(0XQARGAy&|lOQR$Ib=vb6S^7(YUgcgSs8Tz`Ku%(BTb^oB_y zNfeS4X1myF-rU1v4H}A<+?VZP4quVkY-}(-e#F4bG4pYm4y+;EsIa&{6ryQ;PHtky z$ih-S+@J~ixUn_Eqod8|y`Od|r^Em`ts~XG5xhIj97G}7_zVz}dPU%#>)FGpQEWpg zC}6cYJ3DjB%$0sgMt_|07hT}?4IK(@$nIz`o+3*CUM%vD zEk>BI#1u$i!}1Zo7YzgIT7qwG%!n|`3iK9??%OCQtMwbJyEQDCa2gT#1hbr*ZL{=( zAWE9|5;=jV$YKZ|azs5WOZo{E4l2GEe{!|DH@-TtJ70tfrhkk~OPVPGWQ)gUxP28; zjz{Med0A>pO%4o>rxU;`(a0=|0n4sIpiDZBe(%S{7Z8)^Y7sIbzlr7?k(-RakX1SC zyhBZ7P88S{Lbp0`;b^R-(=#BkVz=3(7)pGAH(!3lYPuNyc%5TF*ljgmnRwa#vnEfk zPx<~1d{U&!1b+#PW;2bvHkxaHU=md|yy#+ljqV`*h69BM`}geQml*f~$^%_mqThV|^8qfQpoav{u! zt>g>g-uO^R;JTM^GZ%d-lOU&#(?jW&_I1X^u1`u4Clj_e0W8uN2P4m_H@a$Cm;$Lg zquNCA=YM+X1JkLQx_4wPzS}V*pC9$Kmp~szxWn7)KW>GV^`5noyXHpHb+cV4Vv6`x zI60%HCk7IP>di*_<$e{o&mNehIN4Kd$HM;XPhu9IX+=G3c2K%j7#2D_&=~ne4>d=a ze^4}#OYJP#`AoLa1~8i%Iv&av5|==WlG`-8a)0MtqdkC+sU?{vl3aZ(zvh#=;Y?(^ zBKn$A>y4exDKR3-5v9v&5dj)M3~jzLvm&gcpl`M5%4gMj9WGaFHd@nGn%puZ zU4Q-86!Q6Wi6)=Ye9y<(aC$Mj(q&1@FYdYb ziP>ltx{)s5?h?5qzps3E(Yl4M6(RCL_D!PBO%jV+I$j@c9IPa4V8Y_c1$!(tJU%o zmM3R=y!Yi2=rc@gOzXQ)xz|e$eHLp5Ly+fD>o}r7fp@S0{i5gDt$k(pI>DcU6?fELb?J+VSjhDZ=%wC1Vt=ND#x4cLb`}SKUduh(BGHp#!u9~ z`$5B?!b?89i3|VxNg|rTv*_15;@v*J4oXi6YYTE|u7S8(K;yTiTa;=RKVV*LC(=FM z*fj(XC745>#L6a{_Kf+`D2?1%{q5yL($-^m;&owU{VORx-l}S$-zl}ZK7UE;qxn&+ z0ggH!y*zHv&9g7Sj5pY!R~DpizNLC;p$IjnCq9GRq6mV{Wj5!?lbZm$ODg2oK5K3T zwHhx}v#X84VydM^lk0)7oFzzZGWXS)gm1Kef14_$Oiw-e|>O@m*cIapKz)}?}7t9?5j_B z+^V6)eQK34%FU*pnUI24lpV%fqk@cbyI4?`j%oZV@SuWV*r^tyAetE}$Rq6cS(lsG z_yvflGG3%>OPq}L3$p;(0X2R*8$9v+-Ga?{{_DKSN`4d^wmy5dzJKHKW8C4Ic9Uu^ z-d!{e-7RGl{q?-9+{)9!J<5ZxibdUDMn}ppo!v;B9qh zBTMPZDRx5iG!N~oeu$yHWE=;-UmdRj7q&n470|K!Z^f2o`;OiaH4pbN7;WF!n zU?~!P6VW%jSkx}loqzF2I-QCz@j%c)GVTgSFJ}M?tdiWQ3RAL#G&0oBFyxzpF638# zZ;7LN{8_V{tmk#c6X8}z z@0zjUw}SiXZZ1AhLEE;%*mHCmcjP>3N|cty>s`Z3B<(~fsecZu$l^sw%WI|eC9nL< zh#l8a)9-zZ;V%-Awft9(5Vk4pYB(1YSFscBn}jBlN3y2xdl###iwL;ozvHZ{(7dt} zC>C?JwaJdyKJUIz7HljjPiQCvnLz=^l|U}9#R0PoO+Wa)evLDOn6|*SddX2w! zYKq0yd7>}I%V zR&yfrkiu4Ic)@Xy>%NiD*nwz>6n^=kc_c%xx0(>2~MJ^ZwX zg@41K=Ud*acg9I2pZhD1^k7M)!XqoHBtj;UT8XaOsi!~mtJ=-xHVPz=MMW=w3kFnv ztMH$JM;5(Yj&&^*qJSRtuPFOgnG=uVc8Wg!@g>J^3K}D#t4u}|^z|1yappg16~?w_ zdJa=owXT8RafJt!c^$ULHtaPYIc%B3=YOlrc+Hb%YRZn6(rLB>PsQ5_03WGfMf8&g z;nR^;W>T@{spc`u0;|+jKCm=uVeI%#?>M(ji58!K>2GMjSnJ8ktBC6@x;}~%^|d%A z!)IPJpQwH;5m%gO*EfbAU7zFQorF3R_g zvG<(1YG&}3TRekwIT{+b{d!(~kbkR|aGv&AspsLr|LF3rhC0@Y^=zzxf`lY@FvkEE zIGz_v5hJTV?vH>i>h9toGjX1P2_jiDo`#oF-d2+`(u|WF8TZa@dWbn3`OpUNMFvF8 zdG1nl*{b&BP4wmare(2UC)3WgVK|G1sE#WM);`H*p1Z$#Vva)|8qC%6$bacg>!Dv7 zmA2w`(BUXk{@Vl%GvLuer7WdN^|+;;hEg}h!5*1xQm?T+pHV3(>)uABtnCcwX}MCt z0a#-AL?s&&TNblI+)d`b>udlIJim(`Q4yM-5944>w;>+j#0Iid05!+cXfySQ`>cq> z`nc3&w@J)_Nx<8R=ci3OFn>@n^5BW`HQ270hs(xovh*|3Jc(A2vjP|<#tD@QOf}#& zuN~<7QqS+oXYz?y@i$GkIpRaZ&)rv4D#(j1=gOo=EwN8{-hy~18&z~oI?`@J*lV(T z18uj;LLZ!6`()?qYUF=03_CI%70PRCpn|GLYUB4YHmP7oMtVXZlz+{lA<4C#H~Pxu zxG+ItS?Wyi>s@zUm**K%m5(DEu^n4`g5wESO{{FaA*2+ZnwGs7O?6DVbVce7J~nls zMk@m@vMnA7Wo|gaU9f$d9qm$9b|lW$fCd*2^>CMKPvt@m$+97$y>8FYU)M{;tL|m3 zokTj;GHoh-fd0v|u745TNF(yNb&`@zKy&(z;AfGY!u$bs6p&5r;&L!XfziS>zTNJs zHfF&F1shwo-J$>6Su~K#_!=L*>R<#(qYNv zs@ZsgEdH*VY#0Q{dVF*v6%C@;GARgMy}TG{8AX!7MUTO4^jfWbrHK2A@;xV&#E9F5 zTtzhiK0`q#@`~zSxiDhPh0D3+Frf+rSH;H#iqZSLWJfscOW))nq)B?T9p=alcJpAx z=%7S7|6Y*+i*9f!6ZLgL^`_=(VRzs}F<A?WHBbjw}r$~%- z!{-lZR(s#6YzUaU9w^0jjq4Y%kscx`2wS%6X3}g$hCZc@c*l?^0lXD9{893=vZ*QY zgqdB~!wAo%(yj=2%!bI-vcvSa7@^3W-*9NpbN2TQZB%{&{%SZAbYjejqXSd_0VLAm*>Br6EA++R~BEJ7vq(%;jds3R|UPn$o>kpdKitr$EEKR!4z zKaHqC%SD86#H-LUQYW9JAH)?V$EC;_B*xy>KksuhXZHld0qe6;)ho~l@HGs0s*b`i zL3Sn+3_$!z)aWbauxL5lXSz{IJu>6#>X>F|q<`AjP-Eb$6VD(X3iuQsW(RCuD!d3s zVZ3|yv=?2BUkj;GONEgg^Km0QtrDhe+Iymb&_kTVfiSE1aV{FQHhPoMxZBOc(@V@k ze55yLr>-$iG?d=9uM$n2{V3mqYww;ffek;ofcm3n2^&?D(c_0XRFK}C#4}=ytjqX( zXMg1{L{K0pL@vqg?)BA8LF}NWhIdRXQVQD(F==3*WoXDxeR!ZytgPqqqoyh!Upnfs z{(R~x$pA^!J{BED{mm+h{YZjyecxTvbw2mL+phDy0$A_f$yE*VThrij0SR88CB3WU zOL!4{`^I%*M;^sd>^kl18**cOab6*ytAB^mJBCWG2jFz~sBu5Jm}V@ZPBkFj1-k8& zjZvM6IEN4EOCSYX9O=qOE4uNcl9J`fntg|~H^m;JjFp+BsK%JGTnLRxSC&8Dv@u(7 z98VVK&hGufW3s>*eUFGIxOBfVNDz}CWq_%+($WZ87F zXkzcPKrwg+%(Y|3)x@t431(ToMWR){1+qo$Rm~VqWC#&TCo`8_P9+c?Xw15CIWzFZ zRmAw6G&TIFcciCnQbX%8Gvz&Zoqt@Jx%PaRQ(eKo7(C}3-0#8v5c=S|s+T1L4<4!? z9oz?F617ME*Rl}o13S0d1 z=K^Mc`Y}l!&Ps)hV*mheH9(#P@mT!by=DeW*&Ta8c7Dg>sXUM1By}CtrGGqlwQ_TU zFfR93&*q^5K|TPR!&%wT>|R;j`L^n3Ol7rAwBi7!%;l|njavPf`%hX{Kv2;fJh#l2 zCQ0^cn`?kVtyu!LyE0;BG@9DWqR5Ho{W~5Q;{CZIG;wjCNgqSkCBS*dgxrcWk}9;u zTud>((Mx(IiL+!9cN+E$gMA|$jjg!KNBrU93h*%pU4Au7wo|#OCp5wp(EaVuX1P&;f%j+NEe9IkKA+??2 zavRJ|&f9=0IzlLry?8mfbKZZLW8WH|P0UIH)M*-5LxKt8Uw?|D3pe9gO>VEK5c+GI zQaz5l{HTud>mk7^>fD#TV@?X?7h?~?%D&D`p3dKLbn0z}w<vZhvHeM_OHsPHRQ*ZL=oT5#o!pzD%_9OY?egWh=N`Q0R4J#3XJHUPg~UGT}`qfjb$^rf6? zMsew zul+$0i(CRchg`vK{NyJF*)SzYk|>+3S1{OPB1ZAobhY=N52$GI_V&)@^AR;HP>EL( zw!(24fMIt!MMl73iqv%9TC81mKXFfDv!%Oxc$nx$D8L{b2rndKHoy0IwjXJkS}bXK zj@8i>JAVOxvmR6wkg?oqImn9DHMcNuCNQ|#uuwz>ExNx49j+9@4;6ZVK469XH_MHU z#L^05x#hgCfFR->18hCp@Jy%800K1$c$i$pxoRtbx;7-QBrlv~Q{6@ex_*OjhLuzP z>2z568OUs~Sa>H=2uN##} z&;9UgDd60Mjl5j6pFbhoT=eHb&chcuD>2D|RVKfPsdti;X*az;7g(2wcu&T&l^E}> zj71O?qhrhbPCT>w4`6G?h%LUKWi&48EOjNT6Ug*i*qhBHb%!?2xs+`+eeD~#|6Hq2 zntz)6A!KObHvLY;xolbf821pb=8pn<;gsrChOSYOMEKpT5IG0d( zg!_)Z>4Bld1z(N9BoV{BaV&dwbM0m#B7ao2JmAXYJn|uLXFIn%D-Hv6>8`%nxnPN{ z*Azn4lHj_Jius9DUnE-Bfb)VkF{|=5^jntp60gTrxQfISL7t^o@fqY`?Ko1nC{kW0@`z7th+t;{%z(6@8v;8I5z$E=BubC;;hH;BmZMqWW4X z(Ya($BLkAabZ>4pK?c3*HGeo^7vWD`yh(Mlh(~=(O~yJ$4`qs%xk%nEnOcxlQSfjh zpOJ1|m%2Q%0&RJI<6RjHvTbULEq_rvpNpWn>QUv4vyU%Q7vZt)Pp-^!(-h(?F)0xD zo@mQ1nMjmeDP7&<^H@%$YNCwnBB|>M3Ef=l2-1ws>DtW(HDBvOI zhI)U0miDIqH3g|0REOK{%vgP+6$}HLqlrHs1x?OnZ-FZirY0z7B#&+??tkZ|#U@DT zeuHK{R~G>Bx0VE*SMtKt_M@#X3!PwlAHZY04kBLbU5~9hzzlWAp z?w~k_hXWHydBVY%t4~dAjbE2EGbdxxKmQ;i+MmeRdQme-Ua%yT+Y)XhYBcK1>evxZC?Kx{k6m*nVbMk}M> zed+_(<;^ z5Hh4GJ6na_F{JJiY$r;D=JJ~5#{LrAnnNOIBU1DAjUBOzxr19C^?u;EcG9^|vNclN zbAfqNF*l~OC2jDY{eR}SsO-``8dd;NTj}R_c~T>`0J3_d3P^GQaZ=N&TpTM2EJnmK z3#eT{Ahg4p-CURu=(w)u#|r$3GRHfvJAZa%OCKU8yw^3hO+90llRe-e$1qtTwVa7+ zY*s`M5gjN>6e#h~@?dtYrGnXTjh{p4p=oMbqrVLpkFltC%YTYpjVTYT%2N=0rQ**b zVYR7r$8KI70#TbeK@%&LBxlOt>|Vn!v|Iz)gu>xAO9O00BLKmh@W#IKrM*r)m$j># z$p`Y|7xJoA$d78YYjzAntIJs`f%BY z@kP~PF=v?8{Hc}JNK1V;`~LA$q24h~UsYf8{XY68Axh&tJ=$%YnLOGF;WmH=SiNH< zX6nqfOF0!7zLeq}bN28s2R=>@Je5ucZ%jE8j+UW;Cx01C+xmW~R4X<;=UQ~855x&IaR%a6 zF2|%m1D*7_C_B0pAFq2RT7BX+luwMNYnfAxC9R1ek?F1TF!XC;3W51NC!P} zOz~m6XMc)hg?OBbn6=NLKC3XxvG0z78E_W^$!uKRuZ}ZgZ?77mBHON+s!ruxTW>B| zg7wnFz<$|vAvPb0YA>4#L7LGc9AG-k~WnzER7iB{I%r{pSFJzu~QmYky+&xj}(t#r9teQRz{b^Z3N75U-A4ZaD2{ zpvd#hZgv*ozg`fS#DP%22%~cSq_9*%+y>~*62bv|i-H7)hSJm<@SZT}8|iV0cu$<& z%3aPU6Xu;`$U2A<5UH>@^Io5WZWPvovmzv19*-IFI9%|CeN*>QL6<&Hn3A}841Y}1 z8#Nv!12eskXGjqwwCTOB?lcUvTL-BIZ~NIP0R%Z>>z!k|uvFT5=W;js_WHQS^+<-C zVlxhpHNCtS%Xbn-)PXOg&&0Zx6)IHSGV}(QYq_3E@p17!5At#0rd^~@BzYZ**hH@G zTQ(4Fu}aGsX21jEnQDwB789d7XnzeN?#7$3ntiWouApKv z$VKiJBJT-K;3#`5tSZs_&>g!^cUTX@Ngh_T#%SkkF$~5S#zdckVMvE$lIF#2G3DI& ziR?|V1a21DvKD7IpRdRg0%A}txR;li?AvD<`pmwOvg#WcOk`nS#9iv2Xn(saK!QgS zihr%JSdhDJl?h%dU!FNu>fZ2mO~$WY&a99dUAIe3o|6;S`XpU&P1VEpsHo>l-^W05 z0@^F4vh-%>B{_06%&6mqNb`mSimF{-w8TW?Hff{AL(ij+MZnq47w-o4VvgyFRqGgp zmm+W^yvU(8eQ zj}MotxR%%sRTH;LgFv^Fn{)e&%OxXS@v;T~7kmF5)zr7Oi{thJ3Q83bP^4F-ca&;C zdXuj77CM9)1Vp4uXwqxwy@V1VBE3s5Aq43?(py5l;OBGBdGEdN{r%26#{Kh*vH!?m zF!r8%?Kz+IJaf*qR)0r?%n!&7)ss}V{`fft+?&GCCMGBgqo?MLwrxWXyV^LKWb8Nu zoyIIYrqnDc^PbYPS%`+2>k(Dcf>oBmo6#!t&oY_APPM)lCXOtf}zP)S&J!*Zst zl7v}^>SF3~+FihCsjzmo2WYq}UyU&S&AufRpKHdXqC&j8Xn)q%RsQN6_c&jnlSVRC zDsie_;hOa$+P6b$5o-Y!YNHiEj-Wls@cfk(rCve1yd`Q-d3kIPeH>*wpTk?KK0Y@3 zuU>8;(9>TJM@uK3Ql#ir%*D#>Fh{Dg5x&%&nu_&e^AC9UeSI{WE=Y!t9`kc~ammo= zcM*%GX#?a9AAkMGqvS6lwKF0CJi@65zk)G&-k1xC`0euyU-yEXQP zMg#ccm_h5){e&^tNUnsUAj&lmnw}(N(K~RD^jy7iCFG5K!?9l5wM*WkN-<5qT#Pp#PNmlundUl0+coG_?vs|`n^-31v%97&_>>1p7Lm= z69x1D0DshC8I<=|k*&M0goXJd7F+7Y>SAt?1BfAuM5Y(~#EC%&h;yMU(SysnoneM* z>kLGg82JrDxOX$&nvp~+r0}ysd96?VSP&T>NA5L@e6<}A&zR_>SJ&W@oV0Es_PAv@ z6lP+Yw0HeEG2})pKlZ_!4)*@|H`!;G89riZN`I~sJ@Mi=UjRf(gX?H1KPc$;bxw** zSKe&mwW7)ud6<&YV219Vq#E(d%M_eFm*UDw4xegvm*YW(I&1>Clq4zCZe8QWzLd)l zk@XC3d(5|OOu{LX|MhkuMDB1uJGotySwHXsMmIp#X((Hyz{B!^DJi74;t5c8BD#747Y)=fTE0%FYF> z7qFej7;bMkJ=BFPjnOqvqDaZvUzI+J_j{yoVGbsxecaYfac%=s7J%ILp{H?!m#4}$ zlC`J~cb?C<{Dc&mD>_03zr@m$v%Xay|9@0nMYvGCw^@tRsQ`b>LY&~`!87Jcd1W@T&!)`F!diH)> zwTMkJ_XY^XD7^S%WWq>9y2OQXwd6z%^oF~*IxXiFY(tn24DdOYRDN9VV&Xy z`*tSevsg#syC|5&9x(Fb2%&uz=YKu-bEE8iNb&uf)CykHkZb!#-*PyJOsO`dxLDaM z3;9lHs#jw$aplE>A)GA$$RzmO?fSuo%(vT~9gvUCDtb-!)FtMba?_5@#QV)T#f|iP z#I%vwg6~x0D8^UZGWW`kf|R8 zO)WuBSfH(f>KtfxI_vGw1!>H7<2XKfSGq3tn=kmBMA4nNR6;26zX<)^p9ld4ed|J$ zEtmHu#f9oC_f>8XDz|WOdsdQBqH7FbFW*#8XokxowhQTnxPr}vCV^P0xHcP!#Zn=2liKBG?J6au7sA8jv?c$g+?2prT`3r0|R3%YHOszvCQqA|_*Xd_Q^CTPVfGjyMV z8>}ecQg7W-U>T*j+RaQ?W5?X-vD`<+$cMx35llA#;d;yM6Rm9_3v1*01JDn80`nn# zeQz0kFV{|*m}HR)Qh$~4QDKfGIZU^C-^k>CXtmF@WD+llr@|L*n+j|z=87M82GbEL z*EIbt);#}xvGz(!HoNCq!8$ng!soeg4-|5N;`icD25vCxT26)56t<4 z%Kp=64TWRY>^|?08lXl0uWoh)j#?W&7M0J$bKQ4dLwb0%4ZM) z@gHUwlaqN3;HkP`7$U~_x=?ML5nFPPw>2@!o4Q>t!GHHXkTdp6A=@5 z6OZIGIpUf-RXi$5U3bzs4ef=E)tv!7MNE$BVfK*XBW`5?VnCh0aAjQi{Q9E=C)V=x zI)-=Gm8}!otT}6%tXlvB8g{NOkij&!kxGr^suk(+5hWQW6j9}k?3+yDDhLJqaP|J; z1&Qj|p-vFPLKJ`L)|q(Rq`9Kv9N7RNuOozomACWKeGJ)5kR&OGE+nJLmN6MJyS{0P zu{KV5)n>Uld2zGEY1AO%1bDR}fl5?nWtm@RYR+Z8>xTH6wAtE=^VDaoF3u_q#X35t zsf^Z_(DpFXD5dF}%%-7qk+wMs)L@NXN z^>@?XNYh{I{ss+M9c|QouJBe_gFcI{mmgG7EQYu1NfU1YBqYU3EFNi_;Z!9e`6yy| z&Esw8IndrK{Y%@yUiCTm!T3Fl!uxp7UGL(D%n8dpF)|kP!`b#49aUYcuN)x&KgsPZ@=r5=G8SHv3mHzHqPUqV2Mm zM3q(bjGpoCtL`RqGmFZ8y2VCu@XkXdI<}?p7`q2DY5nq`;nT=w11v%t;-T&(P%`8G zF_O_-{YOZ4YqAgBg^5mHB^6a6^i>h>p3n-Q>UD%1j7z=aO-8d@bCb zAaChr0lt|kBQ(*Er7ks+r+qx&vFz{Mdk{WmTRU8q^4Y|voT*pFrosI?bo8?ZxgEbw z5Uu>5LSm5ubNBI*kO7dp@tihQ`K|lfv-AoT)|hI?oe7(9-;_sV%|Ld3pf3k??Ph-# zl*8Ei^($@2up6%4E-Z(A^>)6^LY1HH(FX(V=(U4n&;?L6G^EzdnaR;rVtUf(rrf(N zC5P#dM?E>c{b_RygL$>s`h<8+<4$4$4NhFhcjs2egG)qwpZ z?kab9zf(#$F8p!d3{bSLfxmAiKjRYVL%q7n)>$9 zk{LpAw0ftEpCh+;o-DpChit(vF4k2GPn$9`JXTu$w7`}=U!g$7l+gTc*7JV^>c*1> z@KPJgf>&v{Q%A9^LK4K3Zn7-cX-`iWNS1PI4^+x`a*s>DezvrDaoyjl0B}^)E;}`g z3W%ARm~fcQWS{_S4=d#)S)JovCM~sPGm~(jr!G*Sel|Vr@bl@teauu=c zMjIGwq|@qKBu+Uj3U+(b`3`@oYZN?ritBAXk`Rag{Mv$B-LbSNG@MrW7<-fcZs>3G zx89Hw9(RBYyx3BQOXJADHy~y`2Ei?R7_+Ft%AL@EH?HBb{+npvo$eHnm&3WnG@rx3 ztdN!Srfea%PbnJMzTcQ5cZ9U9Vz43)4q2YpCDQjmKy4h4fXW#9a&6_^$ZK}|7oz||1-lO6HvjSM|+^BfySKiwT1`W%t`?z1TKNsWKml;!RsfNjY7_)D}+ z@;XUMj>|(Z1&2=p%pRp};nUL6;@WzDf^CTH>72ne+r+phjTOun>X{xQSrqYDAZI+K zjp1z72>?pQqfCmt0k9^vPXO)YBLAx1!hW%+N48%_c8FJNG5+jpLy%5Jesb_7H!iEc)KqNmp12N3Eb-w4zbcUi0G-O=Y(N;DmpP zsS6X0^%Q>q7H2K51YTK(;9J)w%_4(T5RI>uY&Rd8c3oHYZ+RB$!<0r6N%w3(GUpr9fvy>0b}_xm?(aj( zsRDlq*-tmVf&dcj)tI-x9;#J4Y>0%OZM;Zm?uC-fg5I7*j+*`ilFhtm&>Nx^bvKAp zvqi%0Cseu`=8)WVMmOUWzJ^9bWaj60jEN|Fr!wzaK9i70gCgvdV?uM~A|6lc*J$@q zW^!EPrDY@K4Ipb3(5yQ!_4|8l|I-jfSQ~$X z{_QMtW}T(2p6$-@=9k1bUS)XQU1VVXIlU(65+b4!pW?&%g^^j3ONTPyn-Z_zq_m%l zR}8~TPe&VduCOME71iTac{PJjE*7DWBe7y%iYC-d^ti%u8dFT`j%~kG?(MR$HE_DY za|F$ydFpy5VvK1baxLexWDhsxl^%bOS?;jLb|s(v-hJlBO>8q>e77bvl8Fv|ML}6e znK_EaTzBP00_VbL9#QJ*u09>TN?GV#$1z3XrgY4Wbp@}Ol-pHe#1~TIsOD-*H_GyR z_xYdXOJ}!Z#iRdzhrR*DYuj(wUr5OCW#8!Y^fWonSTtTxED%1SJRV_2m63l4&bSpn zNa=|wQ(DQxuD-Z^W1aRYYcI@q;%c1x=)!!Y8|I45$roPUD(h)mO#XfWAPTMYA!SiZ zC+TH7;1krgbQ7qzo}>Jchd`}{v8mf|Wo0r%!(hF{F0m~%66$g`DH8oaduZXuX-)9= zmh{=Im?8?+3IHTIGH-FDR)2rq3bVu4v!YoPkVOK)?+}hH;Dc_RsxREiBG{=G9Z7kR6PXHEw|5swm7vOF$W!$r-ZcRfnrmL)U8v^t zGp18^-epXmJ0+u-{n%chNWu>==0p6-2W`pi=2uUJ>d)>8O(AqClXP)HR9qDQm&JabfbYjYRUr>^+T{ex zV5WvJ=O*ZG1v?L9PEn^)I)OYvqo_0 zyWsvxwWk_Zeumic)+Qk z?*T21Xh4>}KtW&b=oYh|Ef$gXnw&M=)4!!H4?~DmhtXDqb(#K z@{_DDKx5=Hhmg~7OjNC4bRKtms&ckLq;pR<`<$dFhb+1JtM`8_hJvn-$al|1b2MU8 zeFA*M!{VMiMll2^yvv37dYpMUNpdKj-1zFJ`oi= z#&6_*9ltjKrImkJh>5a2!7~-lQ{&soHH$gKuiJ72hWdN%Okc=BWefEY`N#rKxdK4{ z9O9{=Er|>}82Ofc+)hF;n?%D@o(LM>PpxC3Xm{;eDXEbAduEjQrkvJ*u&AI8^d;c# zhe5BLPdPII^xJ*XcU9E0O4V@h3FdF)y^2o_RS-I0WlevsbRl42XAW))-rYN5Mt^*@ zK2^jW_wj{;r_+h#YO0P;it!+Zfm@Mb;X?AOWjMkg3V=RYpa!2Ig^ZpMB)4# z%<%>gs&NJ7gWMuBLNssO4&;BG5qbmb)2Jof!RD@!ot%7`bx?_*ZH!2L?O1sgD$vh! zsRPod+iriEUj6T9`Xr|J9Mn3fGdU?KH&OuvUdy)jXWDi7d|gfH1}8`3J`g z-@G=?xMVQd_KQuV$@%xkh6J0X-`~q0+&GuNgFWu2y&x2HfIg1 zcXv>Z8~ z5)}<2dy13OW7W=Yxtec{T-lnwLJRGBJ_swQr%1K_5WdRx@$<|s@w(*ib^QI`LOpt~ zzjyY=G#;xM+Wp-P^x2*?#eGZsjqP}xoS?#6RBCImkXa^_aU*Wu2glZXP~9ly$i{!@ z-fU1lzCwrjcg}Y*c>X@NnmCnHo>{*`E+M`s&}{3meDTd|ZqeWM_bf&fHJBj8gi4!& zP9Y7DVLm3Xao^w=V+-I$-cvt1n+Fs!cAtJWFRab;sJ#Yr8M?K;uot8{-U$@;;<+8Lt^HQPC@XA{l?%iM^lb0IrGIKMJ% z3w}AFcFj%Ko@DW_H{R~O@^8Y*f455`C&TYCOS+LZ`Od=q-0&2!_>^_0dSR26f^Ui= zCu%pC1#h07I2JTq+1)*Wa^6;n_Iq{bVp<>i0^imFyp;_~>5GkiD`Bv~#rS_bDzY;@ zC!mpot!FSdd|z{SjxT?lGr5w3{Ebsy6$`J*tJt5WD&n$DX9n7Y1@GUiu=_XY#SeH$ z{MFbgBmjCV?dW{_fM9dvPahZC3)kX*|8z&yusfU6&2biDqG{u$`n>_9(Y3)0LgYpJ zOsq+9j@AucjIjx+9om{#87(i1oY%<7 z7u7!)-7>ex(U3?eB)oGkE51LAg_$=nsiywx=cvZuEK?Jkhd1vn`7q6?mC>Gfct_th zM!NkoB`Wprr9}U`D}U*RMKktVzn7;7p<+r|8{>M8SSo|y&0`OBs|SA3gctG{~lnzHk+1~!nf=>`vme=Qia;ChEO@-u&IzC$kxHa1=#<){6T0jo$?m< zm&}|+%~K%C!VT+h=Qn>V?DIvM!datc>@t!zR;@i#U+hrnCyyn1S{Pbc9B5d6|B&(T zpL6DZH$})joT4uoK-Y9eL8#nJMzN@P)gX&2#XRpR_%lM3D?~YGv2fVcblWdhOuR|= zeb2&;JdU1T0JN>zKqMwss=b{tHYVmZt?Y@Wf#xeglry2BFHV0o^o!kl5c=JfYZfHG zcTFm~M*`QKyhct&|MD}GU6G-q&3J>UFmD0vSoZQ003}Gwh(_`T^I3xbAF4A+#~T0d zA450Ff33SaWPc$cefaa3q#8E5i9_Th8j;T3&S)_-(0Z3_s+=|AvH0gNiePdIGSv$R zfI=D6um!xasl9&~LPA!N2J(aTvB1+qx>pwh+3o(-4*lDA->`~s$0vymN>>2ViT?22 zp#|rOBK;f88-e!;?q!JDO9#a8Q+hi5!E#u~7teox*xk~LL$=R z%ci7$J&efr!@t)2@k?k^C^{WM?CIqZU0Wt7%Wo^YHC=z3-PP6A;P+c7Ko9WBoTs9X z;n!PQS`NOQPq&S~tmx~-a324`B2(LNe&Mp|n|Y-P21di#`j$4sbkBp3r`$H>e zr!0-!r0Xr+=ZU<3(vWA*+e!ZIQ}q6Ea?G7?fAE5;Q!`FthF+-qbY$DC58H`F~v&4_@@{y(psNjVxgEty6M5#c1X=Vf0q=jW%2Zr6;eL08oSH(0W3qj=;@P_g0>WLm^YsZ`NpjX(WwHcAG+0Q)?p?j}>dLhTAFteadgcG$&#!=|v1#%w!if({D}zVmQ%%4} zaCE(DmT{^~=a7Tgu6;dGPHOi+Gf( z^EKu&6DNK(M?H_sn@Y*?$%(gfl|g?GjXW}g`evId=3-;L|J)tmwD|bnxa&WDz7o2Y zxi2QweK(=915nrypb=%{jXgdP^y5YQ;feRMakayty@~h74M%*=Q0RUn20UjXj>qmL z-cK$|O}gesD1tV16+azoT78GGQ`3>(m{5O@Yg~EsZ@%2cPweBnpN9ADFy?=ahZmL3 z@!5>#SL%<-hQ9H98oNh*@x2(?MaH=al-*qYIByE5=(Z65`)~Ix`5(9DP;)GFBQpC$ zSC|n5iFs&fFa7Th|HV%-h6f6R&tp>px>>P81j<1VwRK+Q2T&*|yJ>{E{`(*DiodEC zZRX;rh`j25jQNi-|1su2#{7TBnEx2_A7lPw%zupek1_u-=0C>#$C&>Z^B-gWW6Xby z`HwOGG3GzU{KuI8AI9u=ka*w&iTgLo5Y`mBW*@4?9~UJ{a^mtjLg$4qZ-OAWX6$=l zxi{Qf)=*vCxT|1sFqVX48Uv}b zfl(n$ZvVY`AvdJfJU>BV9(6M{ehJVZ{bB!B^-uOaF89|N0*~rT!)1BftwHft<9aUr z3bLbn_>0H9PxG~h!C^lLdK0yhOVlyvWxp2Vbc_XC; z%KWX$LQ_7tg9%1SzvF+_`Q=2akv_3KZ2F%nM@8%B0Z( z9(P}|L(_s;S?%n!MK@zS3)#RE_KmNZ@`-{gjan)L;vkgCxITYq%WFc&{}tnn<#7SG zePPr7Ezm4*8aOtcB4AUO2uky00~|VZV3J*yzi~GoS1CFa#p^eKM;(}U7|5=WtLF(^UKg{6%|}U|TMdZfrCd(PN&AQ;v8rAk@JRh+MJ`$m zk(Ddgp54JxbZCFKSua;rZJzCO*FLgah!D8rPek>=isV{m=lx}vLqKsw*^048x6cvk z$in~TvY@_Qvvrf0bQ=xn_S{Uq$B=qwWQlAavj@8jGAd zhT20t+i!rQ*|bNPqO$tg6t8KkY=1}e)_68tcyB*?8-h*YS?2(WVH)?6m}$MgbJUE| z`C?qgcF_29;;NPoPQ#S2_`Jpx=_>?O82Mar`$ME(dDu8rUId>wXoHqMw988B+OuA1 zQaQNIYm|Q%QYP<9!4)oqYsyrEG*T{@@#ZA%7LfoNd}Jt2m2@Ljz)ii$Hsw`@`ZX8R z zLE!m0^+6n*>#LWCU>=8bj`Asw-I>8l>@21Gju(GeBy1bv6rV_Eyzoc=m<)Xi94oCl z*{>SxyR}hf$u#R^`oaFr$5$E;?+Dw!5CoH&Mfkd*Yoe+q8)qtelh&e?3KBTM-bn(b zdYDY~>fK^+HkC@{v%drE+s2*0>or&Y*Ln@=R;`-jDOERT?W}bgcv%j-rHdRP; zuXum251L@Sg|#M8kBJ79E@>l^xDm(0eS&D8+|E+e+X*l%5f0tKn9^XPkzwe15>TmbW=k(x6Iux-}@ zTowdhk5B3(1!!nQ9`$Plu|2EweD_j2`bwnYKXEQ2SXX&WC^Dd`3ivpLV2@VqT(s@l90&G1Z1Kh||J_@vVO7 zLZ=rASR&RfFF043m%H(IrTzi@)TSTYEKEteL^ao)Tr8Hd*VD60$up}PD5mT53%qsr z69=(w!H|u4iC+B11c=yW_%1@s_I4&&%J~rjpU9~X1eS>Um{QapZ1m;PrWSv}zV#sG z%$9_Yisawh12v4jPWWB$IE_ASpkCs5iM}TBPXn%cR`L`xR)t^PhCaVMb4T2feM|dn zQ4;sqEDvpcvLN6d%XB|``{pHO_*YZess*Cdm33KafajY2-6!sw58b_YngO%m=n1y& zPGhrc;OuCrmNmho9R}~xXJ3CUXEGq?qcev}U-=K?a_2^47kpim{Mkr{#EKskv0RVp z{nxnO{d-*R#-u$WBTx=`2NtIM13_$zfdM0fZBpO|vu;Pe%fFuo3;(LTRU@CCUYmX* z+-8hcOS_t8Y*HxQA!_?|CE*)*UNikw*RHCub8A)ohaar50WnMx$m)Ns#iyHQ%o2li z8nxn<(rk!6_Gid1#ufx2Ta>k~FuL{&fT)ftsi`P0&y8RN&H2#}q@QO-no{b-) zS!2~LPn&c-ldrN*%{;pZZ0)}}5VHo4zfunc;PD zp@TZHF0=k~7XRJ*{O>7%RH~(5T=?rl@E=wo=Sn?Os@bv@Ob31ulP6|2>v6#gHut#S zR#~toM+i=Tw`*AIu!s(#6o@1GfjdLv1vF?=Y-W92B2SL9Rr(-)2E)ZQ%a4_f#tSDc z(>wyaM*nS1y zKb-YOlYwyIyn~!5$>R)MpGnvlP_1@9J)Q^UK3n0cVx5Bh-n!lB3m+F~wryBIV3%xN z&Qd|}g%Ec_(TX^v>qx}rI1>%3;@lNk5T@&u2t}Rvo$r6qxs35oz!&%@-u1g$p~qT*W5h~U=|EYY{OKDU}$a;wbQz2HtsI>w%;Mtn=KoV9^@ zTwI^BsGS3aq1{lU$>U>o!S9!#^evbc+~#w-wqh|POX+k1Ph6_JfO~O7re$v}8MQIFP z>M__i5r4<;(`rWT&pG(&alf=(F&$xWe&UaT3-x%W!evA_qCDXuNxy~d|qSm3l zbP|6q__9PtX~6y^gAkc*){w6nR$z}GShx)agYh_Y|Kp-EoAGKlLw$Jv7y5d-w^fU--!0EvYO3({TTWcT&y1%&eh~x54m(+F5Um|(ie=oLKv{v0 ze;F58qfGXc4E0I;v{xefJL`%u_;b)+Ij4UmX4-X~Mm4#-xFk#qV_(_**$J~PW=0)N z_DWkkLDQj0*fO$yZ@}A7^mm{yoigpqzT&zc1ozu0t&OSgDAs)?69l3}Y{&POgD8y! zFwJMp3-H5f+oUZ~2Yoe~>;}wEXS(Y_eGw&v?qsf$XO(nO;ifXWni_=kf8jDxgcR!4mEH9 zXNyi)((!!hW^>}+NOCi;qXSpTUh01|^me0D)^{LhpAWX#VIH)a#}6Z7>3NzKpR?DI zGw4T&JqJOs3fT1%dfzUOV*pxV=;%vx9M+3%9oc4nyz*@NRTD90QsjhO->-mnL@;c9 zw8joMGj+(noN7rc`35xe4IFCRdAL2+dT{c*WofwNe5P7HFIfdh=a^`oP=S92QYR}3 z&USHBw9#eB^ZFeq@pT`X&iZ1c_fVkwXD`8?U)G1-W=eWKD5FvlUl{1|Wgjp-BfWk& z1uyQ(OG~m&OXWu%h%cvtB#`n3{l)NY5jo#<1p|!(!I9mUv}isAz|#mIDWNGTNNbsj zSD%m9G}h#S%_nSS9b}~^sEL2-tW7GA*cy8KoZdSt*$>#7@HLCCG^L=Pe3`_#d923A zl(p3`wle!3chJ`+YR|#kVd!8dnjE`A@+gZaw6?aFtG0OPUE61B0dCRRm!=)O%S$@O zF4og9)7w=xRa!Gn^^&K{&8JBleRwc-2+S?G$sO-qI^$Nu8IxyauqJ=K7eq60wxI6E z$Bx{Rg}Y9VW4jLzh7HtpyOHT!%z}X2W|S)~ggM^L7)Q96rM-tg{R__izvoXL2JE2u7?-?4dHAJ+TRGrIDtz#-$5HG#73fsT6O z9WCY{8{@|#Tw!79hxvcTSq=^Q_D!vg=cmcb4lwSjBcrB%0h;M(0J`#P)9VK z&+0?ws`{D0KNPO0>in4fG1gA6AB)x5VQ(%U5%xKVxE5>NdR*rqxu7!MnLgBstY3dIHfzSt_a1W!?i2%2 zkz#glrifYY^T-F@;ugopCa0`U;ICx+f31rd9=$(dIP|{UfUe*7@bQBqBaE2DPwnmJ zvzEFluQnIQlbM8|%!bO;4|eEA=7J{CjtiX1FWV7$3$m2DZBN`+pH?MkB}Gd~3FrAY zjSL8N9utr^8+U(G@f=^N!hCE#Uz+IztWmlD=w4pdw^xZB0+mTv90NL;X4yN1w{>G4 zcfMN+vmT+1LOZXp1!6l)b`mvt`_^1qtQQV9hHAbx0PWvgZo7smi%%h&3f8? zj162gTYu|gaMU9$_1PlONLul9aiTDIGP;1^#n4loeWMf7aY*g)d|09`@5b)%Am!3R z#_O}>H24v}%W|V3K|_odSKaZCJ%tq!mec0rU+x%KjIiZ?zSnMc%%w^Ur}B>0X!}xn zf>3|FVS9f&7qZ?WemFb98hUcR(76;FQn~ALfky(*KhtfH4b1c4_VCyv#ByKN0Ph6a zF-+^#X$xX05(xtWxgPOOv(`|kIZnHV!IsPXaQfFz&V!!Y8Pe%!vL{7F`z)<|vUPZI zfq5k~-%B)Jgm23tY>S(d>fdkbG6s#UdUXlVL)L%r7Go|u?%PJ9-}5=0eL!gO+C;Za zh8j#LtgN}%xpSw^*N!>wHtmbH``anK}2}I@?;lV%=RoCp{4CnmL{3{rMrfd%Z2pQeS2-k*IS=rsks_|tMjguq;KdO zw?VX{_u5D!UyABOx3ZM4pLRY0LDckOW?tXkwu}_AD;M5fB`prvXc`DR^#zO4itF%g z7nRkhF%)h(s87oF+LOJ@2kS+PWK`VC;>&-*Q2^5MTq#b|Z$?~dcMkwYh-ar?!CvM& zj{CyhXa#i{8o%xn3-h!7>Ie1dB?NN+*Ag;`*z7!kHQR9mEFr4i}^kTo#-T2L0|j5x#^$YD+bytFWh@ zQFQ1j@3=DUUQ;pA0LS92sa$Z9^Hb zUZobO-8i`WbJTkIB?^cI9vN->hOFBaD8YLHD9h|97Q}`Sx3C6vD}x+ zuUcg}Q09#(kHoJ$UeJXmu$SADLSNQlE_Pb*^cc79IGYD%7H-^Bu^F=o(nd6UcU=N4 zUEOd+C}OhWt2PkkhKZvHP;P(55HTZW=s1mRhHPwgbgZ1Vp2HjxYBDNPI7+4eg}2xK z#@i&}|08Qt7E6kMdL$l20}2uL+iZs_s!LhHdV027S5Zf^piq$p_o16B7X)owR|7y@ zeI%*Ps7VJ%%D`Pv`wgX6+~%BwD8U_zY;@h8xm5rN1HO(5HDe_&h7W(}ExWKQUjGn1 z^7ISSIToN+QYSpFT-Ys3k>FYk=xC=bt1@F)%HJChmrh9Cla#|nDLOq8TfDQ@xp4J zjgNuB4K0UV_}yO4mEwPz#)Cun_-Ws5#wAKtvsmhTp=nB|5qa_y<^vK(*H*OHuj-T+ zQK7*rhL(&>UgsbE&xCH~tcsm-UXboGG;#4kL9=$jmb7hztTJgodr2Kc*QS86q~_aT ze`mVSd?}`0p1bO20O7K>W1)5$|eovsA=|Yc`*n z)~~qv4(MaP`QCb>=|x$sfJiia*6@uJUFm)>mS9}pY2Lr=Sj3PQvd92BYC1~JjuNiO zam{Fv)pZB0b}@FHa!Nt=nsnK0W5xyc=b#6aGk|xMls^1FVon zHX2`xiC?GDwf-{mOqz&CI{3+>)bX=1nK}PJBHk#yqPu_5`tFJ5eWC1J7;UF!2c0A9 zdDWsW(8f?dYge~1lr+%G;z$CQDjQT?eYgS7TN`ts^aOFqNGN=br8%rp5Ny7-NYKme zplxI>iv;5)8i6IaZb(qddcM9xLx*mM3*~0|SD*~T2)@5?1?#?D$j-%ch79@oW=DQ& zo)~|eSHFMOQM6I>@|>5T=;c-sU8i#=SSW@WWxZFQ&fZ&CR!v|?%Am)%o`A^=equ&M zm14kzI9V>U=qwcnN@myx<*$=8X-^#GM8#HOEQS_;WX5leW(2OeFA|gp}NIH+=*1o4l9CCKP>NDcvykU1B^qeG;*j zcWlnB*s_LzAxKzpw!@pme((AYbh9y!$DDsG)7@;u*&7(^4qt^%-G0J1s{1*ll=;Ba z4C?x`lCLVg$|uQY$` zk=ef=b;@j9Z1*GO%BW693=j8K&6 zDPiqUZO$~oK}}4t@QFF2il2xInQvm{d<_WgzYPapXdMAX?wZ&$qAxuV4nLF|^1vdi zB)#UHJZ?x!2e=&DH49tNQ}6kbLKaqw%9VE2J5}27P7uSneqEm6*4c3i?AU+L8~zbG zuaz?+%Oe#RRP4pbDXu$sYECbcN_H{DOmndlW2@6WPv7t!2JAp96n$;iMZLZl&P;$X zt`}-AT)SwB7Pdj9!c@5_=B^&UEu*}m;K1IB}0 zW&HYL4{N|NU1QQJt}L~f)|5P2ZxUF{_l_gaANM>`Y=Cua)qw6x>z02_HA9}iPf25Z zGLW;2!-5UvXc*aLA#=+0`J z3aCc@&x}Ngauq5!3&3l|wJ?cKvnJbMxcLPH{J)>RZJ&qF^{h)lAHb#20r#dm)< znJX14%0c!7{h&Rt9x&QX%w}Q`tE;k&Tv_<*~Fk#M2rk%9LCG=%9p^izRWud`leUYyrb%B~ADb!qI4rJM(&Rh?Yp zS-rPPCA1r9Q_pjRciJ>&K9=vUMr$sp~dYn_g&L{c_p61S2v)qp2*}+0CPtIL^-CycRl9JUeqJ%v`2?JV$nl^i6Mb+ttw*Ut{V-~ z-9MnkKc5KNeeyOYPa{ibLM7V?zhsxti{^mJG*PoTJP)Q@;c!?JawZnz8#^rdTrZ;c zBt7L^bqdDe#k8hlUOSj}U4aysG$-lVwsU~onc{zLkjRd$5OG>#|KvZPz7(3vUDUkaI=EX@Nm)o-;8+8-!21$AtEu9Z$AitLI4cVa z&%9Xr&5!PUwfUtjxq=cEIV5o6DTCan1aZ*9sa3xQ&w=6wD zwo`jla@zjPUM>{itnNt7v-)uu7N?#SQhbg*KSIK-i;ZTErO98*RqRln#FbyG(sdimLguA|rsJ%6Qd5k4($Xtm>9xzWbC4 ze7X~XU!YU=%+T|7g=QQ(WgPV?E<;l)>@^U}fNR$+I9b=@CwqpxYj5&`#Wk=wZN zTYU%qVkhUKGd7K@&0zgnav&6kZ-m*FS&cM6`Pis1B z&~NhEwz=ILEcjcyyX}A1JJGegtNQiTC3o9ila@U0igAvKP2mGPs=utzD~q^-+_GY+ z9NDILn}(?~{L>arQA3cj^>_NGIAU?HVlBLLUp2c%)XnNv67JzOy|U=h+zDOgF7?^UQyd4aGO=yR5nlTd9~ zKn7{poNQ&SF6@8z1&@>7XX@Q+Q}pBqW3}vu->`@F0@+!hhb{v}RX;sSfxJ0FY6W{b zH*K%2r@AZ1uIz6?v3x!52&G>{8Z#HkXJ-o?DSL zS8a8!daLyu)}p-(tx3dovyE%+!uKazb(H?Y>D4E?S~TT z1iN&ugrS=ONnOJR!yZ9@0MYkkDbhDl-Kb$}{OJ|98L1Ff>NBN_ zcDMF*Fpi8I!JaO$S z%h!K9SjQzWtekNd&M^da#{)$gUV3s$Y58$>7c_RK&x%Gc&YE$4o&PV*OT&Maoted9 zV~DT-jRPu5f@hkq2^sT-+ss&nPDuBs0~RY4H<+p*KcZQ!?h9G%8gNV(cd{G_C8OGI z*vs`g!jb~-Kid$AZ~}YnV@FiPQtFyAwyo7oV*01+x`Vk=Yk8-=IzhTAHRM^h9_4?O zKfihZ{+D={otBZ~HY<$|{T?geG{_lww|hO7%x7LX5WSntDQhNRRKyn(+crOVvXbOz ziz>37czVkTn3P^{s!6@-*V^sX zQ*IjdCOrOhVf^i@DGAwGnEnr(h=5{?^|HpujJq-*8get|vTMfPn%8z$$m^~Ab ze%@$R%aBk56xN-MGl|2%&>wX2%Bxnh!oFErh^e36I>J41Kdz8Nw)ZN3)o5$_FtVbJ zrz>D-fbVGu}G+?-@|Xk zuP_K>{IEFwd+mJkJ}_bY=x}LOsK3m(v?qdE^S=WldN+0c>4dDU%5?XL3V@nu|H*k6 zn-J?0U<(*^AP9d-4x4{}0IF>8Ypl7)l^5itR~{VruRh8*%zrgs>`!Xrxd@cUEDcyb z{pl1@AdP{}<%n-_H(_m<4|{sK`L>Y2xW}I)=f3h;AwYOsL|-L$-5jRG&7WC=oRRFO zeN*w$Bhl}bHdPh_>tE^W@GZ&a;@|1+AXZBCNpwz@sj9V)Jteq*_J~Y!?tig@-?>j@ zLv9EGxTFU~GTJWm>%sl=XWh@qU|8wEQ#|hm#KL=D)rfo7Vd^ z*I)x*mp)r-_`%MhXVxj^(D%LCGzNP7&)9C}@#T2{>j3PQ%ev?Dq}fX37?#MhO;EY# zW(BjsN&3!W>t#=Wn`XLS-!vz2>1KD<^{T^Shs|e#?4Hw4Ss$cekGBls8Ee08*^(iu zjV0~RwvdBQKz+ZgXZy3GlmxFI79rJ@V;&~OT7^@;aQy1IKeVs)r;=LNzE zS?*|Fna-7e4&`?aB%-yd)u-8vl*Wgww{u(!8#_m6j4{Sm-7scj%$d9KxKaZg>jNBN z-SCPN(p{Ktl=aRNtB*zT^ zcGqjfAuwig+)8yb$iV1Qp2blM=I2PpQD}d0 zrflFa8)LU#>-HMwCIxt|u4qhqIB3*zhw5lwP!#bfEdjLOotw|==1Pofz|s__0U0gQ2LI}+MmYVo7fFGed&ZpR#*?l)l9%*^`FX>Ad&`R% zhYWTrb&=N3)j1G2IDgHd379oBS7zYz(`Z5fS(%!=Il(X+F=3pkSg`E_XN-Y4e{KJ! zW8{@PyKaIH1nuUC_9xjEYiq|N+{XETbs|6n(oTb(w@MD){zXu%uhUOM%!5hmvpDv$hVgKbrU3{lS2h6GUL=a?=$ZbprKCV+Fp-2qLEEH#PIfZz@lRl={A}cOSCD@wYtDDtr;gqKP=}Js2PKKc1PW1$1$yk zW1%;ngl-&z3V~6vI6)6pLTPUV?@qL)&IQXIGLTP(6#HpuhmkMWqEbvc}$ZVirW-J-Ml z>~#>lE8lYs5E=xY$y#_GDKe|O`_nTzq(#j7eqxjwb^)Q+ezL=yo}?miJ__6rK08V1 z^NnFjM&P#eR}Qc*EfZ7SSD*Ox)c|c47A^Fb8rwI=^8s8X+o5NJyBoECF1i@wu0wv> z!5kM!`=wiYxD|<$eT@+q?r>YZvjmx}xaCf(s~p%$2i^}kX=~QNAn=(Qb3^e-(P1iV ze+n2?9&Gn&_$%M%@;sk#FVeQxAA;%EaM^mcc^3ax(Q8^wc&J-AO0ev_y#<=(O@);q z2?{EDDwfRr4wiJ5Tcad@oiGDMZ;y4&fvvu}n}g)&v-5*xv7v^pa^e zfkOGk`K@CWBND&Ie=w)*%2vC=Uo0i4^m-`ImnzmI3A|xfq02D#^S1P9JiDdX`THio zoY3{u4Y-{CSa&-jPO8@b_7kB|F+R5>LfgAV_onke#8o9)kMt0KCtmkq58Ts8&KIrW z*`34d?-1aDy@;P#2Lj7QW#=iA2{Z2(;_WMqk)-0-Ih5LOb3Ys>CUoC)*N2@}?AVca z$v@c8Aa_P`h~ov`$9w@w9W{NExH9J{2=}R1sW)D~ZlPLo?1HPzRQrT?viFyzujt~A zzv5;*p2Q5;lGpx!G-bDTGZaJVf^m%;Jg|BF0l@Su2W=VgR;@;Gq#JTZJ8MqU`Y+^p zquo&>p2`u~W8(IZnZ0VAuBfa4L+l%Tygduh==&+X5ldYCQHgMh%}v)74A{l)KA@A_ z!z`_-Fn328@0lxH)mE?q4Cpw9cJ>~X?GQ$ob}_Pc<{UJC`3j3D?`jO??QD^3tJyW-ar^Fq4Q1vZHs?fD3z=3)8%c!GS7u+~AZL0bp+b(6g->>4} zBBNm`Va{)@um81L!;JI}bmgIfvU_;Z#`T+d>${vB;(v*r8|J=qHYO_UsG6-l!m9K> zGK{`M2=R1(ufa*fh9&WIJfTcrn{9Ek%^K{2R`8ee2(k%@OZa|Xlbrjm?!_*Ayg--W z$~1tpG&8~eejdsHm|-t9HdSEqy|}0ovct88t}H?=H)L^%YAyu3URyz+to~$3K-Rfc zMXX~V1;>t)4aDF`fwnn&my)GjG&BsxYh%r9-;CRThTrP3(yN)Rbd%q!G4xFD6wzYf zJ>Mw1lVE;Nx63L#tT?}W7*dQ|oMJ084CYeqW(N2(0l%`-A{2HR7tbJq2KUWSk|$U5 zt{k3jxG0cfc3f8*^8-evR*c?rH81Wm3?)vuSBHk14wh*MXinKS|BByEY*b)M2uJ+L z7waZ}PkhwJ3%H;LhM)XSpNDBZHb~>Qg`sZE)6@n`Aux1<5d_MaNS$=~dQ-&xc=D#| zt#*2tvB|r3#KWizgN3{wdn{Aia*U-SwM_wJF0tB_GMK!08<+86jz+u zj`|UT7m#b| zu3d$eR8FsYV}rV!TAJ8~M!rH2m|HOvRk7!R+08suyvG?MaKv;WHCBzB9=)>|W!`VO+se-zHK|O8@Q4^vQHyUCo4k94AbDGtuB@Bb)EzWLV z(X%iIL|pxDCE*4%?^6%G_6K#gtDN2OypexH@ z@Nt;4?)IVRHjg}=e(e{I(tK@zZZ`pQL%NQIDOt(@m3YsalP-|8uLC+gas<17%&HT& zpVt#aUN{rPMW}nAP{04hcD%~7_mdfP3@~P89182zwbn`UN$B)TlL+PB`w_;)yAnyY z!qrY*T-Lno1+TG4tu#-P=;VZon0uo!sOHeMwZ@HN59=jV zHWqu?viJmqeM9;7a`KIT$$7S?_V9dM&-}5+PgTnk<^cd^mPe5rC@@oApE_eWJmb0t z9O#|NpIcn96-)ITSlF^nv&0I#)2Ay@w}94gvsdQi9hvsvK&Q3g90~U(L0^Xy4(npb zJGHcRr*wCZH>(CCVy`SFNwoley%{>|VP_UnWJ&rM1|QP~TF9k;yA@7(WZ$2!bxzG2 z3-dgEqW1I|S&Esj%Jpf|(x!^)5?cmig!=ru33~_njVM#@x=EBxct8xkw3B%N`y0E^ z8ui3ty8&KG6I#2lOZg`oNgyaQ6_?V*pPQ~7A(BCIOKoNkE5PIn(eH+TT^4ku3&7OKOOx znW0!6GQdbk&SngQ`MmKf! zDMk2lRJP22xm0zp?o3vB7Fz6lZ(ohx^K>lW>!Chot~uR|N||DwK5uKIIwBR8n<4>J zurEp<+QUlytCfoFAFZo&|3~X;6(14#)5^uBZZk|u=v89JfZv8H|Qm;4i*B~mvze|}tDo0Qpk2H5XY5(?x_>GOOm!7k z9qL)MlgD!d2iF|%KVu1$n;D~X4%;m0(m4+II9$)xd$~ZHot!H5j_i9|1imGsQ4ZOE zg?I63)-NTv=#Qvad=UTVPT zSewNh2FHa19brZeP-cLn^?^y6ggjYnnb6QA=ekdG;Sp%&gZIb}wgf?k*_qHW(oX%T)BbHq0^1iY-aCpK`PF$U2>akEl(?(sH^uW zEv9iLE=Prj-IH>7ef+0ySJN5BA7*@`Uk{#i-IXK@}v;$q_QhBROuYHq=Z8u@p;xd>ZNx8SBByWHLw=Ixce-ZTPwS1@SiC3U;pZ}$893uKoqoBHLoUsrmQxFh%!gD$C_ejf zm-S7v2#x95yj3?F&?seFQ<3#WdA!T)FGr^SBjzy*4+wTnCv}X&P0zjd7rEz=Lu9FXbnI2ao*XUH;43?azNz zMn_(JR`R}iGcR|^(8=nr12_?8SdM8PktfTcOp&%E@sxT^D;j)1unVLUSGo0+1cgEc$BIKr( z#Od+Eu+119__(FoN~_k*Fkt|h=tP;aC8<7CQS6%PI9#wzbs>3wZ#bLIWCqiingL&L z&%zz}H%hCt3m^Fn>xWA6fVSY(uemOG&tEgNcAfVDwUtGmM~5t^&(kugRMeg>%4@B0 z!Ce#%goWxk?%v+&0e(;?;9Gdb?q5UTIwVW+a&XX#zP0w{^;<+dQ4%7k6Ly}L-z*_~ zW)cI{p)5^D`5O0s;-vb6Oa}#Qydf#tCbSlDmA`XQPH_iGF3l;eq)G#ANlsUquL-ZU8~0`h6^!#6_%G&a^|v}A4h_H9Y*BV2qYP&R ztC#Ui6FCuoQy@j{q@nN~G~-z-ywd2j%Z|1rzRB#J- z;JG?!Nm4t1DpoMXdlYtRLE2$u~AkuACx7(QzQ+<6L!9Of0|y_i($3v(pn4wW8c z{bAr}*2^@TfJ36HM9vuCyeVseC3AYIxSCz5x%V$UTwrP(!n00i=+nILxMIJ2F*q&?kBKKsBF)?iQ}&dKdsTc)s#zDdI(HghQYJZ38m@C$hviL_KobLh zZ_{lLaKrbnRVf9#Hy?#Y;#!Ktderq^o)W{oH0?te<5})!oCc11u73=s^Bmpu`~FOM zKovgL@vNdm;DUC~NYAGKC5b6YV;6I-{EqU&ju8DQvt^o9GfMoG>~L-avzltrEoLp8 z+h0x*&h3qa3yXNx#_y?)i;DfuknRtEB$k{n701qr=O&^~i&A3LTeP=C5@7ekZZ-Fu zsKC?yM7fd2kooe=OSf^B%30UxgYn2@rf^!{#dCG|!l#b&W90s-b|Vcj&1FVQ)m$)srZUeOSN+rNmC-Hp z=g2SA)@y{iwVyYo7491J%sK0SVj`wBS!a7fV-2wStY*}!*DeImaBrD_Wtx`YRH~U0 z=fzxnbD@?{Hr9}VTdDpP0akr6q1Ogo14RkYfOyd0qtrd)9J-hl{h?Yf2@qYqg;nJ^Qtso5$5pR^}foGiujM!jv~)_AEIMp7;G zTE9$^*HkZ?v*2v!jp%@X{<##Z5RF@3h;p}%lh=pCn9GI0%Hgxcy4>F`dsVz^v>gf5 zwxtq+iZQYDXrT+i%)?xwOrRAUY?n|~*G_0ki}yTbajI9Ji8&C80CqdC3LkodgHydG zw`$QnPc}k(=b!v*myz$^{^sZ^{~brK<_5uu3};;QP_76R%&wb%F~RDoJf(#sK5Gq? zhd}bYTG@H+ZMb*jgZXP02RG7d4V0kMD0mlY}%mh)Un8GlPz^1xm=Evoa-A<=+-8SNLeUSndiROOV z$0u`v#229=KHo=wOkAPBtlsy=`SB~H)rj`?bscs$zV|vIy$UJ@+I;n8CAax?*Werj z)1`_`zn7{T&N?=HFq;vc?&PX}@QRTE@rd>7z%l{c5G%C2 zrVwGC44fIa2%X6BwlAvn{miYZpf0*K)0^Me%~)cfLcO?azE(HvbN8S*Cx|`lcC!7@ zc++#;z{6&(j0-W>Odd-E!qvLjn z)>3=+*U*ZT0F7AzDW1z3Msz^(uJ3SDE#tUSy6<_tH@?O#*+A`P-q)?7Fx3t^#QpI@ zJ3*93&E@K8(sXt4;dY|#vuiR3yv^Gar%o)%W=0P&78IR z6eDd)mO|zLglEOIItLSpPu`<){V1w@T7^CjL}^nDHF=2Aifn^%;?Bn}swUFl@~wKi$YH*x8Rmg*;_o4uhd|r*Ke++U^(Np zo4#!-uU9NGcLu>tVM3oL`&h3yda+u6DV2cAVCcuwvEBQOOqHK?0SLN4(R7iYU zvE^R;-6bXMRI)50;wdgmsG;?)AuayAR(IM1#;wyiL~qV|VhGw=lJjKBtMH1LTKF$5 zM7w`;i1Tkcoy-KvW4|81wfWOC3BB8^?g$aPf6waUtI`<(dOdRF7@L6AQ+8*6K;Ypi z%jPr?BnaU4Q(u4N54plkd|p*l^_19~Ru!8lRN<ImndQ1Rn~rivBdtgjidgZx2gWnwudW|XA)oloM5mQ? z(41=DG>co?4iRKhzGBGx>z2!Ifg5HbBk*=oD>01$_bWfB&xKqy6fRzW`ISdnayVY- zP`@Q&59F$^vxoPj+r!iUmL2^Gs+Y3kA3XypW96*cYc4PKVe`tK4TIp@i4KU8sz<8I zEr7uq``T2E8C5Ee;}$q7p54h)F59t6rOUqB+)FDL!=f%WG#=1w z?F20wIYgnZG^DZC`nD;5aw9C5#gEv4TP%pnmZA$f(j(t=G!1^Wz4|;j>G~nvTqI61 z2`jd{x)5?@kUH(m zy(hqOY|rqVI0y}?<1y4mnW2jKO6Ugh&ccH)=ZuaVblc^ql)u+B?&DT?h zEnMr92$YTea1qu(=7M%Dg}Rtccqv~8Dy4R|LceJ9*Lo4h&64A)AuJ?T3}L?=D_A`hHJQnSN2UZ_2fj^}=LeDpSzMx;^I1Ebkq6G1nR2 zn~SOLtD9tW+2uHYw5s{U;7Z5gUnq4f)f~mkZ~zfx3{CGOKsW6)znm+lj04D;1NR{fD2m zKW3fm;x~4naruOb^z7vn;ASOVqdw{>Ds{uD zWs}}f4ehUOjC0?GRTFZQf0Pl@DLT(xAGt4Rztk)Vt^&*A!A73A%_noZvh&pSNLSV@o3@#+NiXtq@jn~mnWJx=0v zBS?TozB_DxSWKlTkZ&O4B_}5X1nGx3Uw(v~h!Q<8Jlhoh&YAT!Yax7%I9Zlb2{xBw zS;H9%3r@03!}GG2mS=h51BnfA?niCO3t8+E30XTth9H+PR|fntb?M8>NgB|);`U5a z0(Hz+1k#~tb}0-~2C1BvOmI)p$(59QC)3hhi!I=Pyi!)%lYd!TV+~zjDC}>Z@3{-L zjnH!ZTlN&r1aT08@my>A0+-8vmN`j#GKxi5HEC!I4F?pw3x48UzjpdDnvysBaE6b# ze{(J>neNl=o_Mzt&P=OdSnbkV0@B;At=nP`rgc`;mGxr&CizcZH76+b_ck>hk^e^_ zLcRrmKtU?vwzhqkns7zef64)UZc)OoEr>rIL`1-1>VE&&83-?dDTit=FyEiv@Kh2x zT1QPNS*;nwq+Vt&^z7>EsegN(m4H&mvaAT9L6mx3VnQdSw@6PeP>!6~Z`Zf^WSRGR zaphZH+h^vwws#8*kpjE(1{ZQll|Y$5mED_vzmA@nLUem_s@kq1Xv$|#Y4iICqIfJ- zu)oe}LGij><$!A^P;fN)?L_EzAPiNI^D$Q}mOo!$e0qdroPHvh4!pUmRCK`Sm=)3? zYL95SpI`sc<-EXkuKY_tOjeG6f$wc1HlS#x?ew%&NzM2RZUnY4*>RjdP7$hjLQk20 z{s_440e^$t@1zS4DMh<&qz1?so0Wx0NI0w{Hvko3BllmLFuYR~J>Vq!AK(Ok1{n@*LBIMOc&pxCBVgq(G z41bm<_RXZk&$|4@_GOg>%-&BaPB-s=+f6WmHj)aJ=2=1ID%loc%NtA3M%DU*Vh&92 zpcQ!BvE&(7|3;5xQb3I4hg|>VQc-s&gBl2*#GUPxyxjuDgsqvv)%)Z66F&AOxB8O! zk>NE5{;|A^J*>Urd~Qj?bKf%uqGm71g**RcdS2VeT6WLU7HE08&)hHT{?X8X*?n!c z#_R{8t1ZL0^Kb(mEWGwe<%^ovaTUCt9Za&c02; zeyrMj+G}2U$aI06-BEse)X%8SpB4^4J?am}w+v_`T$VZZ7Q^%0rtUqNM28FG=|L%c zECv8k=4YAHI+U5=lHi%qNX3VLUO@dLqDpO#*x8en)WOD+apYlYs&L6Y-?{UU z`_nFdqNJ>*htnl+-#opf`x6fKNLcNbD1Yya^M;J=!M3lZIag1S4Vw^a^Z6mU?X<)o z7YbCV(80NNqswM_kGuH=v2O2|1#WmxmsFgAsZK-gA05f%c$SwR5VLrHtp9G^cg=3k zy-1*>uW_&6)XuzxBD6TGZhr$Bttpr@`!%gu<`R;^|&yj{gfE==VBT6HBQXO65ewS;f5 zYW-4+lzSdlI7O!?zP*PR%6zW}@x{1-3Q#iQ*b@eNr zXByV%8xYb9Txa_)>8u9{#UcQ()ab9jvRTc|y$c9P1%WY)!Us5(Sry$r&=sd&{0KGi z!y#TkMpmvkFUFEueFs*yF=hlHBX?9V_#4l#r4Q&lP33-ZH z5<)gP5=#Da<3giU)x~`KCTD={lsvaSd19LL6SADckqy)`6Q=L>-QVf7K=?~51bzg7 zU6q`Fxe#{6y0-mh!q$NXPKj~xd8;Q`mmIPZ;g-#Q?7p6JnX_R9SC^jk7o5 zRcZm~`|qQV=DY$GAaQ(m^RQy``5zdIZi2t;oTFGFe{E3Z(QT&|ddqNSm!B6gv8U5(g8nmyk=b3dgzFmehaNxG~ZOH4whZ zjOrbc>q<{S$Qq&I((yLEM7ly}2sF+(1bF8rbI^k|PjA2z?7s zv8K-K<*I+`R8xnw&F~#v%U0|e1n}g%R@m#V>nC)Uqx-!cG4w)&fE19R(&o(|b z-c$IfP9z28mO(w1<&~l1fiN;M179}LS<~==H0B7EGwPfyznKY ztm)v(C@~S#$w(7D?pxyp8IbRZzFu6p+d6umAs1kjDRayaA z*0GQaEW*@niMXKL1QXVmvOG&)NFBN?I2{gFot`$MSB19c;0b5UzdWuZt?4w5!l`|M1v+3`RzJ!$u({G>`S505Bd(y@A%F#(*<<+SziLybZVl$3 z{8s6jUR~>`U6IrNjSEx=eFFmzifxFwtS#vzaCJ{tk|p;F>-J!*DLLhTOOLkUiE4Q% zk5YI41rFe-614-H|J@09ODJQ zb#Hg-n(aL;w$%LM!kMhv*-DJ8c$BE)s#HsN8}MXTtMdIXlM|T|rfEPEFN^nwI2LUR zX?t}f%VJgl>AXeq3h^}8y=`^!;vO&X55k_d(~-OE*E+c0){W)s>3X>$7}w@lVOc&V zi@eihHM$Wt*cL&5@9uV`IJa7MY201vMILaX=?vB^t@ULphS8?(+uKJ6``_BwbV$Kd zJmxo$jiajCB<18Yx9PwP9|)8^h}f6yw`TR@xs3~IZ+x$b+(O>z^Yy+N~z10{R4kzT|1l0qP1PsXcY(_&MNfGC-%0AB(I#WgDL5_(58o$ z*y)@4rCjRkAs*Jqoew8PuYK5~|0CynDA;N+Za->`u> z*>h&t&o>`bt4K_jw-^j-TfTp0HfQKtlLbv(!IGta{La{@Ywb~SfHvN~NjwUDU}qYz zi;8#ME8XeU674sqdJyLike(dUIvwWJZ*qE@cYp1GFrL^rMZ{1y{Y;gcO=z>+*}1Pm zUa{oy*p1M%RTWt^#{dM^`YRUo_avn8>YL}+$trr|IN2>z)N~s1Pib+fFG7ZuD-c7j zafVrcER^4lP4p23BNl}X7*j0^JO?>DZ9%POp;lFc^2G59v$;XbT_^S)@3$w3*TPxz)SV3g<8s7H>T<^Vm4GamJDMn%D>pL%VfU|l3k`q;Ey33a0q9Rfv6SsTXPJg zW)`!5xC*g-jdz$deO}$pROZaWVbk)1;|CE$$dj>)7;}&hZqRg;dZOz0O88`qoc$nw zey0xB%LUJu`0Y7-(s9@@D{#U~bCda)8Y)gsL4kcS&xVF3Q$h|jehkNKPUU-Edz%#x zC|yr*-B>cS7Q{?FS*VNR)&79>bHI67Pp(SgNP%Neg$f>Bc)7+3(rkTSw-TZLecyh^ z3cD128vFc;)sgm$n}6pHb;k=62&K1 z_mq+adnsyggk)JS9$x8HLRM7+(ZU}NmbV188OX++pYJzLeNF0}LAFW4#@JnhqkhjeOQJ8YJD#PcJ6ISG!`d#Vb) zY}83R*|dIEDn!+!e~ODL-Q{XFq}<|5Fn^_aOKmoxci%kA&TZ=Rr^=bBl|?1`8f)K( znc`T%n2<1$2VkG#p0W~sNiFT1!-Q-$?Vre8CPJjQq`heSnU4Ti`q>lh`%hIOALZ5- z^_z*Q(*E+6^WIu929IxlMezhbA%Pq9T6mM{Yk%-Y4?`96Vja|%whB{|ZH_+-)`#kU z2rn_qGPq~D&~&@VUG;|?Etgtjoi^5=1i&_5n5M+IS#Fu}vret%1I%u!aQck#hIXO` z;5ivo1hkeL_?YKd!deyISN?vaA-MS-nY85DybEjY`psCy`WHlh3EEl19*=r{`4z1w z#)+{MILkk7Sq00)3Yf>fFj#`)O^h7XKEsO>vKui%_VeLd@1N;yHQcWbaEjgHQxf|r zFnJ-LO=>Os_QW7p?tcC2`hF<)JuWSn-LJAgfk1f*d>!xu6gOmW62s8YsB{fUbM@^P zCXH*avo1LSkjg}V5^7MsJ9bmJ^(xYKRNH%Y!6QlU;eA43GLolPNxIywy(Unur*-Iv zx{RKCR}+A5Sfb_CJlX{K-yIOGp{j}uX6H#rX=qJJzMP&npl5YDBE5~yBwVK3awuX< zNBhN{m`U8`HzDxInPGz*Tx;|!7dXio8 z-hSk+$s*t%@%rYe;@<(Q_=u!My@l1EUQorroKNR`k%_S*^aerj6<9S@>0v0HxQ)ju z%xm&_v+iEe{`iv@aEZgI=pfQeG9Ma`!NA?s-;Kxjs0$ApBzYZr)5#j=pEJC=#_;v! zHBHB!M>e;AWUeu&-nn4_dnS3d@3Q8sPh6mW!cdTaQVk$a`%FOse0#Jtan z{YhijPkfgoC?aj*L)enFiitRf_EvyZZUfwW((>taml**Os?Meh)(xx_yuK zV+ys{DZljfXxZlA?hE!(+Q46TNfQ&rx(t*We>cqKdY>GMLEw!Jr)^a{KMZTGO~Db^ z^D~Tp^N*d*o=-c2GJ3f5f;yfjmmYA3{?*OAr#y|y7ca4KIY1TW`hlM*t=Kc)fWmr{ z02BHSY~?rEVA1uM^OoIxUQ>QryBHDE=^L;!^)8M{xqPEQrrl0DKPz*neXU#ca)OcO z_q=`^{h?%sCPVGwcTh}=5?Vr>pl;#jABUrVxXSWjy9nkHFqUgyi_V0FBa3VOQ;vp5 zrI0WQeQWc&i{Y$)Jr|7e`x@eccl;E>pgO)puxmZadz8 zoc4xSTk3Gt-6e-}i-cHMqGhf{ixau(yPmmuW6G(MW^u|gO_d$bE=&|L8-4lBem%$t z(D>yDExRx`Naed5HOcrVF8I7&NqOX?spz63s9J`#4xd4K zyB+(C7+nUtQt<9L1f?wfXs3;(ey{X@jHEzxac)?>Vm`W%Th$1BzC?NQyEV#eC`;!C zfz{W?{DnPp!Q0WeF?i-5m*IMFyta}KbsiJx>Df;=>A3lzj4M^N>i}5VAH>2#bYT?o znLlB&A(!n$=?H6}o7$VL>6*pvhn~pMHA1ZPwM(D<>w1vup@mHrzhkNdC!1NpGfao^032XS(nof4_!L4p1?`_uWxAx;dA8gzd?w&ko#@+6T zQHfCgq*FY3m1ou8U%kmIu8lwKkD;Zc7voflcy@^|Ps2ZM2rO=F%*KL2M6p^$B1-y&1-nYA%^6rNbFI6t2L?iWNW)FkDxw)Zr)$KT0bySdU$ z(*&n1m(Y9kYeePJ4gb0()9%&{?>R21z;rcQSdk7-0mr30N$h?QnS@$@z_hP0r73!mnRCs0`B!hZZT_(jMM78yk9TVT@;T;CP}-1bnJ&e)sD~(QEn4%Z+@-;FA+dR^wle1+Mcu>bn_%GdEwQUxMv-=_|+) zr6e9D{YJ@mx(KoDI#W&Wg2qmAzkk8emAg4o{~vpQ9TZm=t^2}%3Bd{O5G=S`aCdii z5AN;(g1fuByKAGt-QC??8ZJrRedO+QzuNDv^XFI9Rb2&Lte#_z@f*)G*Ibjo{Y75P zZ`k-fC1So4JnT~BHeVGY#E8=VPjvs+UQ{3zH=LS}213-7h3_nezB}+~`3<*VT>BQH zkN+gAZZ!i@9}LF6UCGQa-Ip8A2p~$na#W<(l=%d$g)V-lg6RUz|_q zr>b5tpO8?~EEves{%8j=0w|B4|293q|0Z!T;w``JQQ^kTqzJar)(*Ooi_nd-eR*ZO zxYLU5fXAi)!1RESaAzRXD+VVXX~uQ{ml`#)1fS!yE%u~ALvGPI)3eT+~NhX{hRkPX-((>8a!Zyoo}xi+=}AT8azPB2(~9OHr+QMG$^nyQD-nV zH)D1_lJ$kIIcty&Y-IxkHO-@1u=C*mN||gIZh*e&We;*6;mu_>u6d~xj3$*=&Tw3R zD|YC%om=&mDc-oD4UHT7lX-b@klcd*vW^gclK)aoFv5j^%^`bZCeN0zSoYOs9YY>4 zh%Rvx`X{Aw2GVH$Eqmn&HpIt!;(3bm0K1In%$5}HCmpSdWiE!@X45*}K;qK*whBLT zr>s&oAg{Rz_b28*2NInBZDgj1|3>DpfOK`A z64$c~g=gafn+7>OD!II%J9^&6&1mC)`p^z~p*f3=$E8w7Rwz1c!h`47;%&Z5^`{$a z3RdE(Jg2Qc8CI`&LH_Sb9p9(uAKaO>!3?kdc=bdMks3rvJ7N8OtTBUSzlCS9*tYPy z=;}FL4z_>4nC$kRoD)tpDBL`T+{^4>EYaF*+qf zOJfS8tMmTf!Lb~mxat@C)%@@F7MW`}iT!V<;KN^{AR~;1X%+Y?^vf2aMg2kJj69(5 z-;Of}>91idx}>3UR+(0nw0$1_M`^dem$rQN%jSEh z;*Wj>KEIv1i)g((pLc11;X>J&SELf+e%ohtI`wk@*x8T|^CyEFh@HYH|DC3?e+c~# zWBfTZk&GN7o{S)osWa&Jk`boY<{y!88A@0g{V!E>6XD-UIg?LFk&(4ovJJ$^7id+k znU>}-l`#R3GbQ+6#r;*jVnkfT|8mMj|5*zW?5%N_I_$?b%xu4p$q@=Ap<*SV_}q&X(J>;)?J|s5v+=7yw-Es& zrHygv{FAr6p zU-U8o^?G_VcD)X7++9BJi1mYGmyKX+$vJ7-hMUAKp!{Dynnv8^q&?P=BG}7pwM2E^ zcAat3ypJot8Pt>S61)}cf6h6G`#08uxWR2v#sBa7p?6&Ouragdb7)OgG$`+q<8Ikm zhg`cYDNY5$NA6XUM#=Jv)<2Hov(|yp`Izf8#dd=4e{S>*>Z7M3ZoI7%#1 zU+Azj$0^D|1WSW8NGI?vl}e{to%{8{VDaS=jlAW!LjG)`#_mBc@W}dmqL$NQ$H%;+ zmPlrcI2rE6lFSPEf94nz>K5}5?q+wYSfi9y)oIfA5T5$;GB2@LED)Ph=IXF)`jgA= z$En*HziHP@M{u-2r@R_QbuG=$4ZaIbqSmvPP{ndPd2j(ZIJ2P{%&RUxfxHod{Cu7D z47*=|d_Np(5ogVYT)S`r66(1!SFKKJ>rM|2H-AZS0)&Une?Yw5Jq@>CE046pI&(>~ zkyb3jvnusuKM@X%Dv*BF+fl2q{r)eMHFmRDIPFvU5<}U$Lwjb5n3Z2hH?C5LZ@J|v z251Lr#lV|jRh!klo==_2N9?zt1~#Xhh&Q}68ntVj zPT89ac6X<)e~yMfO|!UoyC98PW}i2TxWzKaf}v509(JSLS1e|J+pJ8*7{_`cFq>Ap>EY6r-F zQp>{}U$*mXmG%vQ+4%)6^6RD;`D9so{8IYTia=_NL*%Truv6g;$jMP`X8Cy5dJeSO zZEYAhYn1>#84vAJ(?oAaa*(WZy)Jk`dECyb~RBM?xKWxFU zCY$@of9*T3dUk%syNv5-(LJm45YNXg#C%R3WS}+(*601bWck_0qRvqh0A)_uS+0DR zKK+m*C@X;1p>ve!D7?iY(|`kVfS%rPH&3(Woof3XjqwGt^h)OcQX*}NcVG)V;S=7z#1 z9lb#Yq*=9Mm*Q|Vl_x=A+$1k|{wRfVw?Ttr|4#~#Y&`u(3Ix8VfZU%HFy)Ff|0e|; zUHzNVX0X`x6Z_^c|L1>Byk4}hwh{(y9VHe6sO0b(I$~|oVVG3Y-EyVQ|Nejf$^;Ty~x}g zjwL#|8Zz|G0!pLdFY;ZU?C3fDI4lc0H_2xSu(6ge+f&q`onFR zo$GOKmPDk1X_<1u9z5){htb$KxQ|Jbf@>GCd;+-9LZRA|jb}IuGD;OjFKJ?f@(@Sy zgL~f>^1^LnO9#3YZTJhX{l#d9B6)y$yvKM!@49^I5sGh)wae}xS?5o9&LSBIaZ^rOS$W%8c%xB&n3~%HL;?6a_`!10u6espfguV<7lrHtfaFZzL;U=Gg*^hW|VQ(j@?ySTj@Zzp^DbCqtuD>S+qcwb+^_t$gYnl7X6X#>^4e2 zD@m6Rg%XkNe+LnSaEiqD#v8o6HTJ5g2Rt>#S;D-ZKy&ui-w{0}qm==_Y^Sk^>i6c$ zlZ1Ht%OI>CPX~s%mSn^-8I5usZOn`#$$0SG4I_Q-JPa!7y2ql zcV?Wj+r7u&Un6L)+$Y++?&@@52*e01HXf0s6V+IFh-J}&nZPNNiuvneSK zf2N|nL~WE^`j$buf25Vo6%6RRl>-sAJNEq($tH&!3RX<-Y+LA21w!<+b=9-x*2>4^ zXhIT_HI289mx2b}1?NuB7%LG_|1zmm?gTuq2R?CJRsN7fvWzL-k<20eN#jg!l1$G# z{|r{1e=5P?xOmFvBP-@Ea_98(pmy;AnqeL;a zy>Mo_U?B606wf+Zd0v}1o-z~Tu(>FriXsA9S}JEN#?iWs*kp#Erll&li$Sh|2s_rx z7A37K|M}F@=%k7Bm=1V$K#-{2-LKhjC86iPf85|{oXq52arYX6mMC7C(MSw?Y^a(~1i&NQsSK_%vm@d3r zZI8lwnMG_qfE6}aRzyprYGDkvahC;ivU}u(dEK%3=*Lj+MzU%yHCPhwlPLgy%PG&& zE;f>lfrwOmTIo~*^SL$}Z{Y1Ynzc~q8HI}0BL7ldi~8M?dnL&D5_;A6QBO~RfBAKQ zlgHhP0X*B;gF&9cfK#@(MC1Jt>39TYWqU8PG|c{ZSh=Rrz<~RiJOJv^^6Uky#%)&p zQIq_8Jll>g+hHDZ-W1dT>8{}bOoHzBku*YVW-Fw&ae)~$6_Gd|O|#h|l7%MV& zDy8RUh3QUH`wz%70MGYke>Ju8SmStXTr;ojUgB(%qo{oTQ8{rcz6n}dB9SH$O{HN9 zkg>W!g~@r4F+oQ4F8-A(v55e^$%@r?^CtG&Sh9l}f4+~ZFBZiS zcfKB(FIv9=<8%TgaszG!u5*B|SZR&{UvjQ4RE{3^`(@&s*IZ>f-Y!L&+xm!osd?Ic z?JmMu))O49JBZIaTD9-i%xTHBTJmNT_HMRMN?$fbu#GAh?4JfnHFh|RPKy+aW8`0h zdh%Y?jEAh|67tqAE|+OLfAO_rXG(nTc^+iEQ!fDf1|Ci!;-Avta7Lf zdN*=nJv(zy{dtiU*5;SiAzm(n zTuA4-LXva)i^Ar{f9<*w7Z~FNS>>ABHI9}UXobe zjI{3FZl=fWdN%{bU3XmAX%BPfbJ(6u`!*95q|6ykTuf2KQMpRj?JO0si%N2P8A zq5!V3#+Z@v^~NfjNhGC4+F#FWO4wHbh_gFktvBhy&ox|OsVt|)eq#!Qo&KwHwzFp! z67~#Wq}LhuD+c!<&>K#;^rz80uMYzAwC3|270Vh+}3oSUkak8C4bFjqUj z3Wd#SPw#q$l*|p^6`xV5S>_j)37?iV?!#VW!*}0%e+$#mRlly(`k4f{u4qIzq*sSJ z*7dO1KwX&dIep2%L&UZ1%dzvFJi=g=OE)>?0~w8KJjU zK0ii6f7${Dy@D&)Uh3(|4430mXUC}h^l4e~qN{EhIxmL9zQd&b~0gRq@3SM`hi;0w5A&0&b5@S-iY8FS=QN*x_u4 zrz&~x8dISudNzp1OHZ8krkB~ma=`V7aOQhUf2Z8!;I9kN6*wrjGv2Vhx(oooKf^RK$MHx~en4N`3W_B%GkUqFMgrX>R;YaVtZ7)haOk=-%kr$6m!iwRi{U ze}L&SjdXi|jZ8AgvhG#*?4Y7E`}Lx_zQjta+OEDMl3u;8+0Sq%cKl$K@lFBlnAYOy z9AqiWn#KwkYy;COV^~N%zm+JxYFuS`#AVWW65>^f5(B%d{4R!h*7EYkUqwmd#c0Uwte$>Lf12< zaOl@;vj5mO_zIVEgNIJ7UGmJ8s+g`O8Sai(xQA6za;ZMu5m~wsXSO_Dd5Y$|0vghw zP&hiJdeUIOd|Vmdem}OHGiHSi$OeY%X@VQzBZT2eyXb{*!Ov{WiRr8tTeT9vz#L}k;$bnZBqjnYHe2ybT5OJAzgZ~a(GA30<3 zN{YBy_G$>cEG3aEE%5$vA)NndfBqfQe0?Gq3>tC8cnZCk3{LKbU{$KlIQj&rdje+>&Z#bPsUzKH#ahrlgxtc?axa@P8eihRjr1})ia zP$(3h47aW5u6CKrFIeE^>dpw?RJ(WLV$FXM3gC&`T(kgBrpNOajZ}`o6Yn#C*wBQ? z13c+PZKv+6I&Fr@b*AoqFIU5rYZ{MU2Hl?vpn-ADzc*O%5nCG$<}8|Be{JwyGoI#B zf)xh99*fz>&#!qLr-Rt?Z{NN?pG}1rfmx}ePncr5oC}C&A2)l5TlN3hxRh>XK}t>2 zD##KvdfW8dEQvEG8*PR3zm^ZRqyV{6-C6bVcN0>rXT*KKM_I2f89g>!UL9{Z9oq(m zxm)?3Ds6F=IHvc0;wF7Jf8$kJ--+{Ig8Jcx{P#26mj+Z2zm2)A+ZyXnEFTvsmmf_P zsyMTqOw{}t1Hu-7PoREFqyziyrK{@=Rh*C9Bx=aZTehE5RBGs3>&}pek@z@n<50ZH zxAA?T;H-;3!&c>xe7Z{1TWXqfMVGJAf*{vW`We^x1Tz1|E~>@3e{;HgiADJGE{Hl}I-U-UF9QuOT-KdRRj4_cF3pnZ%nB_|=jr3c?Lx+y z_x*%q-K%dCvVh(&e-*k*82dUccE<}PO&@oT6)^k4M~a}Vj4uo~wt%iL3oc+($1`kd zxbj2rWQ>;8jp22b%lm75n$M{}`q)(~u949n=Rae7aXy)3Fx|r7Q(J9(G=sV z)Bfo>KQAv48Unea)D0SEoa4Mjrf~AX67TYiz0R6?zOJpie}n3!$t+n1VY-;50uEa-;zWXuFw_Y$#8nEN% zwc~3wZhf7=e~+4ec^cq=NX2WC3@MW0o%L?B<0dobpZRP@?sv;~^ImAD(a3Q%Vzm+# z@_wz@(4fIrTZeKz$m>QQA>K4?l)N_6!`z2cBvq0OuY{3VnU=jYf75+@jJB;lt4`8F@TNk2sE7VGx>*p>UP9)^?XHs)$u z;|LcPe`olnp2kY+6(SBd)U*ogRrwaD<8pL*mrVb!RtB@uN7JPm9UZyzI9`Rst9Y9%vueG*UWX8OKMu)5>FkKcNFbOT%-$}! zQ1UZt3)S&-xq+0L+`{R~@v>vgqFJ?2k}l^-e{pmqn$4Ck`ct*6CI-U}z1rI=1Xy<$ zn1^>`DR+J5)zT)a^o3`C#{mvHy+zNE-nB=PhV?48BUP7@J9GNgcpmKuhH}LwTTemN zbrH^h!;#2xX3_qEfWZi2h`BZ&8_&xmK;B_q>nk~HKJvZdky9_r>iMvAZ76-2!zXT1 z$hb?E@L}Go&NANQWtZ!kAv!IN)>}qD(-zLj6$}Vyo)x?tGt+f6>n8;eNUZekkuH8t zek40_1KTEp=W&QscX>?d3i{N$99f-de@j0G`39ZZh_g?&e|Q6!GTry@8-0cAeu^x5 zo`3*YacKki{hrU76=*+)m2AXWEl)2yNf^EN9iGGnlS3)#j`addLz<=aB=T6>@e1TH zjn3_d>fAaIo--z^Dar9d{nvwdfBT!cDX;M)^HG|1@%!z`PHj4Cb^VVvvS}mse|H>C zOvZJ>)Ezx9Lr=G0%P$z{9nTdVZrlsY{oibd{B!@@fH6#kLQSgyg2((k|8vJm%mHS_ zDSM;N={BY}&p2r+hUJ2;apJCyn$?I8iR@cJ z&adhL#Tmbp*7FVx?tRtzr5#@he-KpaD>h{wghKMzezYnY{|C+WJ7FJhxt*lk$e4Y_ z-90TI;8k~@ox%R{j(gui1zqj~W?wj$>35k6QujG4-kWpx){*2^{OmIO2faWFp@{Tb zXy)mxKHlRfg8Sj*!px)NTX^x@&hRZGN_##`&;)!sy6=?CGi7R0!o?q(e+`wJJU+>z z*OqVmRI*Va%SSe05}5eGwzDCSr&%P(`i7FBM-#~Ea?#+URe+4zz?U6Rhttmik6>Jg#nN8 z?ZEU(dOCWOi(LJ+fwduZ>^M{Qx>g!`{K$_5+MZ`K=`7orD}bSqeeuCvuGy}h-j6$B z(?>&uJ-9ccjV2!&EoPM`T>bFBsCsrcWK>NUP>y`#zPD%=A$NLEe?zRsE`7+LRhg@; za{CaXg&J9s)p$~eDt0MNM}DOAb22*UV&LRH+|Hq|CJ`o12eA#((YGaEDK#QE-rl55 z0II}r@Kb$0p6XNg4mju`_Tz)Bj=3~*Vl-jEkfWuQXdAjJ^NSRDb_C9K-j{r;I^M08 zdv3q=Aj+3#Max*Te=@Qy_U=?>(=_YAKDhZxfyZ)n(}jbT(74P@0A+e&)BwCZXNSUl zSUk;2sdjtFo{zIN=-dYp9uDa@!3;+IsqpoMp*iJ}>ds(%FjAGr`Lm0zCH>Yrmcx6P z-421_c%{-#W@royKtiJImoUm=8(xfxBXU(Slk&LC$!Hc|e~)k_%X9CG=$f56^6ZmO zMv|XJF~yk9_8zlx)9C>p@9UW@vS4ot<4X1Rh$1Q0E>}Y8jz4 zODKUpO22lP3xu;Jh-V;6Z%d|S_{bwa*ty1sg>H$XavJf&kc39WbKJ{s>aG?ZzC$Gg zWYBV|V8PNUtp>XEbZGpD z;&{^1+9+>WJxtWfd0{StudNWLl&fLY&NYe4|AwVZr5g-?TYkucKfb!QjaJXQAj_2A{EdE&8OkD?s=~aY=TDyvOa+A8XDx&XBrUPtun=K$=6b?ceX@UeE;^%g!g z8@K0l=XxZVBwV`-fPCA-$E)Hak*sy!a}RRJV zqScB*@iJnjLnrOSrrBr8)wWEEPk975)y{EU3q|W7{9)wE#mc0UX#8*Rp%N$6sK^k{ z>AKD#wPT#CE6^4az2!Wc*{gS5D?dWH<-X=*#A*g-YJDdNDV}N1$a{RP>ly4-e+@eR z?bnR=(a|j(bZjz0G2W~_fv|3frOp0ePgPjK%p1b>5)d8KVea+j?&)>MHnC1G#cr>{>lhy@pR1RaX{vb_% zKiPW!fol%^8#?HlW*ygQgGR|Ve~BzzXSunIAi2&m=K37>^UmptLiP*dRTt3Nl;y#| z+u+WFdI|aTZE#u7a?w5hE5=hHg6hW;=Y6gxP76fHFS^o&POIQ8H1}r#h$y6|)lUYU z2I%&}y6YX=Qo$V^I$pdB@La62xZW>f?e2W4UUZiuHb|6dk;zQ zAf2X2E?fd_gbHO-+)gfb@YH^bxkr>z3x{^qiD$V z9+=b&(q~F(gkJ2@+$v~f5dVB8O7wl~uql?AObx4Ja%2+eIUagmfA3XC6^YTK*}8b@ zHpb8RH>cjtIoZ)=-7K$9IJLzaKLc+2619t`{h9z%;+Yv**RW**&zQ(?s`-(8cDcIS zTE!+Tnz}$3Wbu6pB&qRj)L7;7cwL@44MINkyo{LdmN8ZKT(AW=rVre{x|tE_|IvBRfy`9P?K% z{x~m3!}mCpQ3p`=PP*{f`jmGaeyS8$70=!Hwh*r5k?37A*bkxfsm-l^2#Q!za%+7i z-`w|;vdp%72&GfAcfwMD^+FWHmRt0U=RVD<1I~xQg|hAPSuvZX_M3nlI{(B(09exhT^6 z<}5`rvXu{FGwn;K*4;9IL8DhDzax{+ZC-<=Il;eB2_fpdqQB&Q8*+#C_cMvH?9imxwZ^&Pf{?m1WV% z_l2}2e>`N@<8nsKaW(|#Iu0G0GdCVa`FpHdUOAH9u9uW{trYikA4A2ef{`;&$^6{= z`kZ~vZ&t2p=GG}lLai0%Sm{1mC-%gW$jp`+(%S@5M^iEUu#mh$bA)o3cuCcIm&XISy&@f_{} zr`C3>PN^{k@v-mj7OJ04OO$20L9jr96^9DIa+em0&aB>#$My)l?i*UpJLVGE%rw;sHpWS{s3<-vsKj&&O%ltZeaIQ~G1)3l@83BdVDtT`% zf1~Ta+iaFk7Iz`nkpmytP+R>iWH2zeM8`>_v(?cX&gcPyEZL01|cZ?SPkZe!AbvQv?UbcGtBo)4rgR@T*8 ztW5Y^WETYq75=h*A@=3t>AcRaw$bKBf1ElKGEjz$b8|J<61G^qzJ}MQpRz$mJHV+$ zZrro1A-;d1hs*hXw2_Mt%`9sa#5^D2;**GV`T^|nIScr&@GXeA$Z9f>WvA7}%Eq;! zB~CwTV9WhMZsT1K6IlhJw%67X_K$VB;Twq*o$>)AUy{&C?xHf1c)$W=LtvAbe;V&? zh^5|Ea$x`r8Xu;u+FWbNyCaPbJ=Iict17`O9g;)O>YOc9QCV&5YeozMTg!WLaTz#B z%;x-nT$#1%Urhn|lzd2gKk|Kaj=8! z0k!_MZW%Wj>uvP=i{@2;;?8;F^m&7A#Cpcu zMq7W99IfBUWS+eml^u;NWfds|jk;uy!+X{~gAbazVkyIEd9<$yD$aU5rpqVZ?*G!WvV^?af6sMSbGB9=dj26J z(-#>Wf|Mfk1D%Q`3_BWi94eASR+yxhURf7NY;*+6N*RpQHFgM!z)W7D90T1v{h&uN zmLg**kL_yf*9m5U(C4)(8(35sVD0!k*1-ep=`}hjPoM(zM2r7jIih4t_7p%cpL&5E zfzmHb36o5gT8&lWe?{4UVZ52sIeX*hLs!T&@#vLuyxX>M>Ub}j(L%`xngw-`Gyy2( zwXRs7brg)U1&xJMK9=CCAZ61?8vUfR1m&`I=d}d|D_<~--yp)bheAnQO1k!Q)NgL7 z-`c-rf@miBB4=3VTwzi@FCbU+>u`+9W&nP=Q6Q-ROfRBDe|GumjnMNX_TlV2Q`_Qg zNGy^*_>;>*k_`IZVfE`Gvr8cG_Zb9xwuvH{V}cUC}D95GG2t2n;^G8{1k56U!lk?2) zxoZpu<&Bb0=v?15Z^teXinp#B45hJb|A_mc5C>N9brFuj=w~FQLw>P~+Bmz1qT#NVa1JF64DOv=iB~LZ6 z@^e4=e^XU*`d6RR)~z|t`CJblFLR;ejFZZ`8^1ST^Su~Lx^vSLSUtP%&X(UYw>RxI z)}P@sEy4p-|LUp< zOY+(4L9Tpa^7*a;3QVj?;;wwbi`d@W8^(k&fBW?8p1MmW(I~34&9so|fYa=?9>R1( zrR9%(%yL+q&+Rboh?MG^rWxnkYjVU}Bn)t2PLDZ}b=#t$7;j~_yi2&$ia ze;GP6b-{QVbOW+RGP&&KJ6`MWp9LRXjXSkFZd2NcM}juthl?G}_hFDUqO#IIHNUNK zd@ARxZ>jKWyS(EVV%ODT;#=j?QFUseZwk4?U24__?Cd3%SVpjeqUq>pw13lS*;u<9 z9-c_Gx>?mpqFZuK#SY_0D?O4ns3)Pae_V4}>a+wS)f91edWyU3K;x@s1xMovXJ7>% zq*t{H9k%}PDoj0n%IB&oaHGEWE(LY-Q9>qEO_)u$R5_lWUb;h-xRqi2s)DTgN@@4a z@tP=(wdSe|%6NC|%hQiT=8KH`i-d3pZyR^h9jbNk<3y`w*@zBOgV@PIc|L1&f7)T( zTmJ9e!U@Z7tY?Uc(_mqvt?IgH_zCe67MW~5!*$Me>56#e_Q*B0Qi*@%*g-)B;EP50 zpZf$G+^1{pDXx@e)UZ-LY(?wkM<0m5Up)eN)1hgIX#L6+O0kVv@s36sy3I{?v+U}{ z(oRwv0Lz)D>E%1oMj68af5sC|>`QyK`+)q>mD+U$y-0r)Veg^Jv+KP?hoyr6 zg$2Y=nPV1{9Mo@By87J@#z>*NC;JGrVdgzgZHPQzq$=BC$y=6FF((Hl@ZYuKT0${iYp+-~%*l z4g2YOvaf0iW*iGq!r zWEXlr5R^sXXtdZ7sp@WzXU9{^{P+zctyoCU9@BNfP*HF%!z;CJD~Z{cspG7M64;u4 zFl3*CYwTsO{OIng7614P?HQO}L=~K+fJX8i1peGUc`Fiif8Wb@mQ=O)1huNzbqG`4 z6v^+p3etDXG>Wr8(oJ82EGW|vx3wBtsD%67D&W(v+4Av4Uqmi)e5(4;p=>(z1a$0R8pj<)>Czwy`NWhaXFdfTjJ01e_T%soDFwsiM2E^r_D%Wv&@N8Nw~78 z23|4T6f9SIn?L(I9T8WVs!# zf8OCa$>}l0ieo(vHOQN$-d^j9e`!M6+Zy=X0qxxc?^8cTg1!?D1!gIbW(ihkXz-C{ zxt3i`!~0f!#R&~ygth~t|?c4kH+ptp=&@Ol2rxxMvSKN%t^nXq>a?lG&pWy z{Kr`UTkrH1X`<<d~J`! z>C7#`_6w71ASlY;^(>i1C(un-cBXng@>KwXx2%L)atm#KQ)W8Oix7;j!a(luf9I-P zAdbs3HyyiN{Zp|oJJO@&Q|R_5YHWhAJDiq-(|wz>!ilY#(?h!z(96*T-|3s=+S>}k zpig^Lbk3Wot$1JtSgJ(2p}sWtOhx`;iiA}=KjQBi z9JAg9{B;^zK(l{0gk&3**qrGN1uIx`0Lv?%&k@ls54$~rMW_D|nm)Y8M&o)3v?dD( ziBiezY&=7S9!%-31Uiwj<&sE$Ze@e3X9b$qT z*0&aXTc~Aez2{=dU>Gn`9w9uA*9L#u^M)UEFk!+-DAKxT=-<72K;v(NsO!im#tJ?<3C=cE0(&JL~u|s@B+vsqvp0LXC-%@>A*;zQz5Ue)B4d5 zPn~tCEdg!bD*0f|_pKR{e*mb`SUq<4(Uf_^k%fKag!z{^@v5)Tl)3oZ0})fpR1f~9 z6^naTiR21*eG(`vjbHJi*UiZLAlqOh4R!f@-oXl(p&-%i23^kOGewFSYBZ;W62km3 zpo??`s+{pmj<$polQmmK!sP^kf_9)AOcSx-$F^q4zWuPhkLHROe{S?Gr5)|+Wn~+y ztS#%MkLVXAjIl)UCkdjoCCxSsRiZCyq_!(PGhE842DiOSvz0%kk-kFKF1<|H?bhAm zQZC4ve{5(q+^5#4k4@CkE=wPiS%lqZD<2D>YIpK`wK1nJ4QI#I8SSPsK@`2qF_$VmCc z5Vxp9l`GV98)*St7Hocey>!*fN(@=%H52$MuD;7^o*v6xS8|EQT4|6@f1gGaoiAF;E_pGJlwc%v ze?1{KlY;>r+A@*Bu+{Ry^p3^B!{L?(U%Sq70?&6Nn_osEja(l0PZznJW58G6YUg8oTaB^N7RztzA6Q9lpvYTNLQCfgb z{Aj7cWia=Je+wgGpPu;#wL#lllHUvS{Ha3^G$pu`U$4~wN$u;Qkx3!;YiGk=yho`q zu)A%edw8dQetiPEhBCgMV1~)ISJ}q*wqRFcpVYHW{I*3M5mNjz36g2+3HwLqI~SYE zRo*x)9pTVYbWz{EWlvz8zJ8f!>v5jj=yW|&Y~$;|e-;^;DIc!Kh80_*oZn9tnqjv@ z$1-wCgh@@a{g$107br353+j#3AG(sr;=bC|1CKS;LRvM|9}-X1(04NvGuN$K!)_AH zZ8G0!h%WQ29``iqbna}0LTY;+JP;_l%3k=uCZSo2?(mbGl^7S2H5loh)zO`o4vdEM zxnl~af5$HQ9`UhTSN225K8O=UmoNDUy-WO$N!o9Y@I$ykYFX_LlW88?l2sPaMqk4C zHZz>V7#3L|I&d=DHek=hg(?0*-!5k7YqpGwhUGTPk*!U@5)Cn>50pk zEBdg8b!REMivC%*wNE9mP3~(iQ>Ab0v8^!2e@fN0gG=C}`C0RcA0|((S7yKG!sIJl zo8?b1TG|+M_1#gW1}n$cKL;7tC|;fBIeTjbab3v$-%QQ7Ss#v5j< z-?vZW0{TFw0fQXUsgOIF$Q)GXX_INq*Ryb_EM|!d1mGNfKZDb~ ze_yTy7HX55_W$g;P8PLS#Za_?&W-(a|L1wmUZE`|n~*2Yvd}JhvE3okJ$IttqE+CQ zaHL!XP%PI|Yd9fr@6s;%(g1;CTW#>nZ05gq8~p4=+7Fu9e7|=*?C4kg11v8pnC-eO zkx=ptbC|)&c%o_D^L}*)w7&Zmg$c@af63I5UnO@j{#!<%*zKxk13sfy&sT)9DU8eM zE~b0v?E`)Ql(K%?vX$$$+s^O{?HDTe1vmP{|Oeqv}$Kk>-UIpTpS?3(TF@nb|)3D~7Z@cFgqj>H{Ngih|t zI3dUB+Reg!8&199yPbyL)t>ReP{d$#`X7qf z%P8#CR2vw(i3O&|A+gl1f22|L>4HH-#J54*yK)nma-PAJ6l;KW)zp=6W($>{=B||W z7oeBOXy+ZiREuT0WF5yzUBPw6sc~$5Ezg6EU)t;jCr+L7Z1wTguyiJy2=+lNwT96! zsm3kyl=)nR0*vMNBLG{IPqSSTbtADfZeu14N90kVIf-0P{Wfl+e`kpCUvf6$7+B++ zwHpsirWqg(LB*RLigm?Gv-8o4)7@`*WzL$X)983T#%s^~=!0$06}nI}JKi4W25RaA z?AClo*U5Tb)0^wI;5l(1HHTPEc(zi|7}G^sMq8$(CFNL^8!~+`x1NMJ4zJq65IJe6 z&O}w`>CAe71QUfUe_O}OrE`xw(nv^qF%0LX-Kbh^3==CQ_THf31^#Fl12qS5Bg_cySDI zyYv+h_w9wq67Ugyn~XL84>dbP)p@qObPgl*nn#{R17N_>B7PML13>IQCLz-?7#)2+ zoOx{E^xxTaeFP89qjx%AMa>;6hP_oKEJNy~+f>^Y+DbY}_3J6REnNNW{^Nplf5uw$ zjvK9F69hk^e_^5Bsfx;aXxcVUp181oV$;oI#vG> zlDRAvFHWP&w4cwJ57=lWhHXfg?wOlMbElZAG|X11D9-VDoJ%PppCpU-vrc{-NUSmvDWJ(}ef zH`pyn!P;C=sB71HE!9YjQa94-Q!{NlfK!e_B$HsTguPuQFr1>%F;H|qaj!StN9GD2 zxq!1+1TC#iiFLomqZ+_<)kF$D=><+c(fU9`SAXm^>Udh0FTlaHSxdIUbdV4KjdlW1zaK$yQWi-3%+o<_TyR z7=O1NJTJ`1F4~k;ww>dKv80t9E%Hl_Cpl$sp0Tu8RYPh5YK|p3Z<1`MGfVm!c?~!L z`O8j$_A1cQa?fXXy}T$eU)LU%-hFEW3f;KA-Y>@7fx}GN(TuNw3B%?pO|RexyfX|q zR+h(jz_WvFln~l(BYI2MvL~H-k7(@?W`9xY&n_kxbe5=}&^!pLntV{lWD`o2N?UY~ z1vNC|Z!3-$$c+5OL`asKD_oMBwuAbFVo9W8g(MOvO7qW;`|$5PpFK2`Z0uGuq#9pp zlflzTDG6(%MYLuj`3{-ptMOrB)Ful8``QilHE2F7*w`VyW2QB@G4z}=ot|pmm462n zJV1zD#NGBPO-%vG)0%Njxm?`{z&so!p-~sNJFH@A?o)*Y2|)9|jn`++n+`wC^ISf@ zX$F$MYqUnW3*Htr-u z6Zqn-FXagXlNQMB(V1E6F^}8Wz<+rEo=&*N-#(&~SR)HBOd9>ZHFf*^?W`w4Z~Y?` zV^23K(GWi@Srp$PY%9b6gpI|C>mJik%G28OiKW#phwx+n7TIx+y*okkJjSo(^R~i< ziP2+5Ff}Q0w8ARqG+Nsuk%hF$AFb!;9z9Wq3YqXspgB0jJgRhH{$8Cgx_>K%#|MAp z+2WRNcfM~dL3FOjpAGl6^g?nr-3jJ>p{TI}v%MH`He;OaRvY-b4Jc_R@jzXh`ybrC z<-^^*6}lT^Ndc!z?Twr)ul~KcgCm|jn_cdBF(P=w$3+{B^E0?4Lr+EJg|bP3Pz`W= zBH+slnAP=P#)+;#5TBodI|)-gxD%*r&TB~-OeR& zw885mK|qU;W@;*s$Yd0d25XYD5>S0k;=1f5vjl#Eb*a?lXm$s`qkqimnyGd6#sr?& zFFkJg*VduppVzLf-~EVbBseUoK7yS-XPI(6fttM~5-6FQd3ot$YSp};5CP??bs1y; z;KTKn@?r$SWJDglmR5EsFH2_xh{z*-PSV>L5v{SQ-s;$8I^CHb=Vpu~il~BAQucH% z65s~jFamZ^^XO31G=KNkEk2isU?Nmq8-Tck#b&udFPC=Z0Ygwrk(D^ zTYnV0Z# ze~^OSkZ$D&goP@lbK=bU<*6HtWcPA8Kqk}$L@tH5mjVnj#Tu@chTf604kUm0OgTO3 zj9}!BeCBXv zNr#%y>C7$52btW2Lz_^sNHeozHnN&@rE>>*jStCQ*6Hk|!@mXsyNesvKq;jWYROK+ zS=P~1?*nd5Y)jY1q~9L)BL}zd>WlWB@TMJcQp?J!wtr1jH`MhsC>;A>Z2hm))?Ysv z%}g8IuvIMjGUVkfq2VY?K#=z_{c0%L*lFQ{K|74|cHvGYq9kNY7txxH9GtSeYEltn zIcN?~Wgb2>&d9ZfAZKk3aZhVm-M#nkZN}(<*2i#It)xW{FzBStryzjIemVHJ8|C9GJnC@F)nAntuxZ*@ki_Vdazh^?4*Hn zrWu}^M0fhQFG8b zaawK_OI^Y8n#|Bo%Y(tP$lmhz#7fpAo8k>a(Vc_IKgVxsO3rJ}Yok%A#<@@x@k+3Y z+JF5%2m>tK4YHvMAZQY`FeIzpc&Ki3R5H@7b4(9&9v;~42zi$5lktq_p&IZRC&Px@ zXo{6>?;sbc6DX~_`-u`XysgEs0)XGy669 z8-|t2?N23WZZ_O%Y1rRdc+1^JQVmCm!3xRHlw2l@P8qL zBEiKA#M=6p{Q&!De3;k&N|+O}IltHZ zfoWOePQsV+$Z>S*r3=kJvP}CO=_2|I=*=r&@M^Vq-bNQ-)4 z?*WCvTTi)9IH}GT$D1)`YDqOA$iOPLmDq5U?qt_tNOr@UUTMNQS@o69LVr*xbCPfDuu~w5X6O3gmEeNA^j#T__V1rv!9f40^8GRE7v(u)Cj41nrNqIv=BEaq z<-Ha~^(aXLLLVqsuLkM1x_|1DEe~q&QK1A$kg;l0#Zn<8>0iZEhVc7jrK^7PN;KIy zK~?Qy`SuEV;Ya00mha5*}tv^dRZvnie-#d^HPF_7anc%9~azrf`VkkApGdaPFFC?z&ZX5X;bd?galxrK+0l2%a&< z;*FgU^exMKKH2_n?5#eEDxquq;C@mJ%=fq(1ghDL@Yzwg*i(w5&b#SYMa_n~J}4TeGHzoe+^sS zW4P>bLB8@Vi8*{|u5i!I<9HWLJntA|n9U4H)!C|>Y?q|p^M9O~j0ZbEGs*7_y965Q zQ;NF0_s9sw?;*}LH(MRU+(}ThNB5>HfjB)KpITSEuYyJ5W*<}Zw1eCGnssyrXNE6` zRv+$Fbk6hJR%#A@^ri|Jjwjf|2~xHCw31T0xjL4YPX;B*OeJ+ zQnCW-(lPVatUEo8c1vj?grPo3PviyWm2T|RZ-5?4Rc)b=6l{NFand{mcG}X0FcsO4 zx2mqWJ!`Sfi^CtUtHsKCO`vB`688C~vd$ za!G1;bZC6MFjuA~mEn4%)5qCK^m6X~rhsN=cF?|N>+P;!Qv+c&tu$w0YecEm;?Ud= z89TYpY=5>9sHnq{t8LI;uA)cA>iWswZpqItfCMEbf~z2hdEaaW;3g570lblMpG|5$ zcm4*gE}P``JR1Lq(xb_w4D7yb{hrvlYSk)MVe0CiS=(j)eHI)XQ%8$kCpA298=hy0 z?DmLdZhVol^%Ivo%gM*i>fB5O)p!P^cUT2)uYY|8&1*3#2sWs(fq-{a#PuMvW_pfS z`G!sge%^LHcDEgTZN+sKWjii4(Fvmf7<7Cc-N|b4K{evVpT0$DF!|zg?icI?-9hQ0 zqaFNrMJadB72VIxNrcU$bu!Dvt+Mr%rtLL#9Rnlvp&NxM`>DzU=(si~7RkjfI>uUF zVSk&`7}55W((vTG&dqi6Q){bWA9Sxh8v*_j`)BYqbrY2z=OonyquCnOMb@}~(Id+u ze)#-x6~H6$6wbQIt1}k1!M}aqaC~WV(Ma+4^Rub!yOx2US?<~$mKQ1GK<48ARzcsIDr8)wK}Mqqs}?%OdCtZdnT< zP%8W%b`jA{>xJZ%Ib^KKbR&pH=droMrPR_QCVh4m3XKoDYL5-Duo8`z17EbrS@pIM zG=RzeDE;2&mo0NRMrSW=)d`AETz5NW{#PQu!c1LHIjvdm&cTgK{Tbd=2|b>}QGcO~ znCI6&bcd-Iq&|$MaH%XcoR@iXKi@!-mz16I`yeqmkA;GmcL~cTWJv;&p!=0mwA|v? z_RfAmixgQRLfhG>&82c4=cMB)D<{>Z&t;a8K*M#rJ&sc8Av?+9X5BeAx&{-4{{?J_Zjxomn4P&3L6_*Zl6I*QETtqQ?p z>DKGdzQZ7yw1(7Gh^vUyE`M6l2Glg>(q`i}l^|2UGPUa#T@4=5^9-%sD>T;xi(9Z` znLkQQW9>kpwc=jg#?7mn2mmJ>g>7v@iLTQL{d%dxsSAFMTV9R8nEsJl90}` zNmPp7RbbT(&-L$TP{XgV;FBFaJ>3tv$;4wh-j+yPSO!p!S%28hs@4NPbG&U`)9%D6 zV4Dzg&fqW_ONP~r;=a^jH-*nSmB6oL8hb+BUwE03f4}Hd9E@S`UaQN)F|<(n2&QNu z(&TaI`zs+DAwy-$rjFWdv2?rLOZX87?F?0=K7n{BWJH?dG_RamKQs`(I^N}f(oGNj zoo?8-Gr#{g#ea1w^Zew`_=^1jF3X%WR^edxZrhmHsy1VuBYoNP_S@rpye(7TR2u z27~#!pvwlFVR!m=m44rPjf?rX!}z3L@va&SeL=z7MG{XHej*!#(3MFxLKK~!%qyg5 zt)Q$yK*!l-rG0AVvqW5-T>8p>cAtE<1GS|a8R(*d5SWiWzO*F{fbtr1H!W$vFn_kr zBpjdRDu2)R3REq+J)*n@o3q>sD&GJFyPsF4oiu>w^LlxhmXU8Z!&w!+r>T0^Q}}Rl z!w$$gWhWK#`Nso9L>!ugDlls z5+|A5g$&oM8R1|nD!Xh)uHW^_=uxsZZ-JF|RDYeph|R&JFgkbCHNOg<&iVVsRzrfr z`}Kdb9e+KWR$4~TXOOv>Kt)NKxYXYu*b`bg`=(q5< z1CA@XbEg%Zw6Pgk1`B^QRXm@p=5%w!CF%_+0inoeRI*0E&}JWcI&IZA3yJx&_TA=7 z+J7QyW*MXzjrkHBW_-QSlvq!D7j82kX5%R894UX39PyuhZ^B~6?>5m}GtA$xga?(S zGjCaVpj(TK#8g`BDY$CS?g?}M7d+2v(W?7D0;^G4G=_)fS(l{ff0#pEB9QHoC^CEW z@$gJCt=9X=Ga!n(W<3|;k5XG?9JO20<$w1A3RD~MX`tA{JUa+H&CHy=-do@zG~vcy zbjG`jka`-Wmjpn;ph_zP=|@)6T)J(qRu<&4JvCI<{K2$?biC;FzW;wNK`r8_bEb|# zQ*v5Y8UoH=J*{bV2_XGZu@Z#`!OwmxFBywrC!tUYTXo*57WmSau+MGBaHehm*MB`@ zeQ$ZqA4;*NX2k1=Z^q{o8|7pj7kr45on&OtWLn~W4!KD3zM{co(7sayUva*oW&afQ z!zlXc;0H-24nraun4`J?eU)jATRXDa>Uf@hf95`J|8RNMe=>5?7;D_YMyi8*+~IYx zqcDyLNh5yaPQ-6r@??WQiYPl;wEk$yQ`0=Isvg4kQIW)qeum`C`>z zd(1}mSw6EbvgW$(egPIjJBicJ+>WNfy2HweNZ+iV=U(QAF=(4$tp!!?(4eH9J`GlK z_{zS>FD{iRr&ED=mbrFD*0-AYZdF<$zPm8L8X}n|>XRAND%C0hTsONNl{;hBO~rnW zv@Da$7NomJtha6h-qkn}7XzNBxd6x;#9}6#+SBUNGp#MBsC@0KKMbSAeaB7r2KB!9!!^5nC(_b?%T z#a_@IGxKGhDo8IaA+NZ3lB9qwgJ2EHA*XU)ACEe_Mc*o2zH8jOtRAq+nnDPc>=^-% zAvbI^xBU*L)T^Dbd{Sh)Hq&b~Q!sc(a4Sy7uN}p1J~(nVA}`IdD`B`m2T?4y?+!>D zPEvaJaWwhEN`wfelYgC5I_%@s-fw#|h`aWuT-AmQjN>#RgkWU+P|-$W2c! z-N~<|!Z}V%ThgEFGqo)<)@$8Tv_VOKGR}G(qD`NmDc`(gRr>8FeS&}=93BkcFhoM` z-Mzz+ijawpu)54!ORqP5Y5LixzET^ik_sn*k4mvlv)@uBUVrN`$jR2&IN)O+C5yE( zn>NACd(sjpdR(}?1w};S04t`t(u=H_7OhYwJYhWJ{N z^)M7;S(Yu(k$>wArySGn=ay*On=kqX-SI!U(Vp-Rj2-xr{h4gzkem8`zJk{1t{p3O z14hdS8`c!@LSI_yQJaht^N3iJyCoW9HblNomXEd|rro$cR}lbgfHe*)bA z?u5uBP#v1mw(1V78aYjJpUY(3s>CaKN-vr08Tu^?qv(Rfn@`mbTie3U%I;pP&hQZ0 zx1Nf${boqXe0K?Q?#t1dT4(~5h8QM;CNY;D>!CxmOecFF9pAHz=ZQ~;&CM@RbBdrI zv9>)y>3_}DiatVbl4tG4`>9HZM`=iFe)OD3C}ewXR8v=5?A&8kR~+lr7J(LB0gM9| zG+GY@E9H8gPh2Xg4Uww27&hzDHr-4LZvA=P*sgcTvG#QF%JPV5QMjBjU5rq3B%9HB z9*+zZ*aV#mNm`A33chPFTY@I^Om~7WB)23%GJklv6`HSbsJ}AN6MKk6ai-@zbi5Az zqi)x#AnRe+`92>9Xt{GrSCwcM-r;<+?Z%)sJjm9`@NIAmnho9f`CNWxZS2ZY6B%!# z0~OQ@AIwBQ;~AvjDyz4BYaX(%N0H&OnKfnP8_WMS`89>Yq+lx;g=89(G@U-QP1AmK z_J5nxh&|}rG=k)}04RP0@T-8MCX z5n5S?-b)QPy0fZjrto|t%EsNvy1M%S6n{(e=M`uOnyTg`Sh@|%q*(SRc@Rq2s<|xC zR)YX&p^*+xh~BSnxfS0bl^#UBV?O~7CpHNcJM|b%4@tCI9KLxDfi?%3dZUwJwm7(k zRc*sx&BT5>TuT@|0eFnxo^RJx9iRf%WVmC8#D1F2)m7KEWd+`2tLm{>0>MX#0Dqsx z0r5+I@C^yv4)mhEpxDs&`@$Xgqh-0$uT zL8{Bz0pvk7iSVtRe85^rq%;QI&VK}@ee^*K%n0cG+dVEhMUV)WaHy1dCkndMV>%`x zk4Yd(gJ^@YM))yHR?VLpZC(Pgna%iga9GJ=Pk$D3866L2 z_;8os4xkZdJew3|`Y1DF3r7&FhWm#$uJ=3kf@oKMHF#xloi4w{qn2X>iub{s@gSSr zcE8Hu*=dX)KzLcE3|^%wiaP>{(lbfj5nyk;2c7GOvF{DDK=Qz23UVv?#*KFOfC92wg*3s|W7}K$rEg8T; zjL;F6RD%BWygpyT=EHkL2u6vDf`kY~t#;Ft56ByKrQ_Q;%vJ&b>u?pS>X3%ed3{_T~jrx_@Aa)&kTA8(%bY zZFTp|kqeJCV5`JA^W#){Tub5?DCq(^0EgSGio&*nf4*HL$Z^8)994s6MDg>HiuE*e zP_tFUm&IA?$I|nf3c1{pG8ahFm>VcrUA$Nhj~u1wR!vRB?mYVKn%W}TGE;kJrOTIsRX zIcRA@6@Jmq6Lv-7Pj8m_9|cQp3Z7I~WH26Dj!}2BU30ZNt?G%x^qdUQGk3^jdS)GtpTVTt2dg2Q@HW7((j5DwWrj17$ zaGjsvqHo%Re$BWj$HVqwT0RFF1A8mO6(4Q7`*EGijAk6(OvW>TnvG%b=+ib#mguU; zrevb+s$EK3b$=$t?B{$>P71ffad5+miz=)T0A7DEu?>xCH6*(p*kl1__!7G4mm?$0 zalNr-&tqXR?WygLP$gjtBRRjUoA&3KA7|Nm5$oQ<97c9oNcoreh;Sbn7w^Gc2k_US zE|VtwiDo@0rmnV_gyDS&x)1uH=;l1jR8*=lo5Jd&F@FsfF&TB@?O{C2dxAF%#mAEr zT_E~?&3uu}PmBX4^`wT~{)wtDpc7EaSX8FkNf(n{N!W_0E*ZuZZ9-(g!qX15<$iqn z7_;+pd$R`|slQknDKA1u!iL}5j~&W9#Pxb7AwxGRa4@!CestMTZwbqAPZ~NK}l}{Hry5yB$}jLZBg#>+;3X3eWolp@(EwnQ*!n zNZV!4)~7!|K57aPc-Uc(sEn^v=&- zSTIeFY6GQYgFdFDL@F|;O0Ta&bl^k?Cl`IZe5dkK;h?0SU8;t4=%3}@tUKFVUhbOx zLwyDKtDCJ>!4LNY$&u2)9apBW3K>Vk^Yzab*JzJQmywRsIx{hB%3q<=vh zgxJ)}F0<={)|WdaVGH82+e!MeqK{Ap`cqle+U~0hewemOHScuA>Zk5>W5=NcY8X$_ zIQ3M2qeL;^hun+$)-d@)L583K#^`{7yk^&7h%y|gs%{7g&E4Vn=#c79Iyf=P1t@ID zmMd6%6d>7B=#>tRv0eC2%;=!Q?|(y=z0463Y<1y{yG-}(YK?-!WBZC;tqTUXC2TWA zs{U)>(|vs$hnfWBbxn&=F@CI&!Y&JW{1Nnh7At;Ruorai=`vWfI{8>>kTqnfe-TV_ zuk@C|fVAioX<1A*FwoZ#@WS#1U4U?~)lH{S=Qr*R_=^j-%-6F^lgKRN3V-q0X3I4s z&Oj)Yn-J2GZr%6Cw7yB-Jj?b9EO9&Vz#hzN;Fj_f9D$)t9<7i+Y4YU>V=tE-t}5n> z3zev5Q>DM%bNVEK8hz1Z;;Z>S_!WI_y0BBUuFp|xG4gt=FJ!{9Q7Lvi#d6tch#6si zxzRgH`wItRAU$##LSXRk&VL}7@j_(sc&jkAFny`8mx|@QFzv{GZ_d-c#QD1~_>vt- z*6Y|@hdvSpVO14`O3|YZ#ph42XBM0;J0l07YBX5D9tOq`0-zp}yFf2pJVoC7L*A!J zsXYS$$(|?2zQ!g+t0N!OkHlVwi9<0(Ta$cM*JD zhbZ*K5PCmqef`rQ@a27E!T8*4(p8G}UB#?;?DsyOBG_L9BNvL67D8^<`OTiUU#1}` zfRbALWG7pR%(eW;*I=6);4mGocGx;yXzEub4f?fWscdT|viRizcmzt?@_P$8oK;>}7tpbhu|RheT}IKi;V9hWIKVp!Vv~y> zLM1;nGBsVk)4F+m#v8JwfiyF#c+V>Sj+t(p2)+ke+9iq+?0;<1NwA@NVdr&;=T3Hf zS_2O!=k3-!N4l`fD)%o2YzSxfq2c(MUU&Y8Fw&wUy)JWt&xPBYPABhP!RWVFXlbu) ziMT6pTJ29b!?v#qT;>u5^M~6%uYD5**_EJ`O7Q$Vl(_WMc>eY)_p~egQe$d~PJx3e z$xz1=Hy=H(i+`%^PbQIHFu9r0mVTVUSCLlKaB^|f8YycYo&g3HPBL&_htIQYL`eV9 zS?oOc0z)r{GK)LrH{-i$z#gSte*L5YwLi;Ck`YqZZtFkL`_ZiVLmRE4o{mP@w;IRc z>%?h~89ZpqwQ_k1Vi^U08l-fS760!RL)gjJ`@6^F_kW8ma5hV-?uHA!9DS^^n7ZLM zKSk|LrAgZGhdFsSmY@%}mX`M9X?r!*FJ;wmI{L4QWUElJC=ViTA(Wf;;9 z8rps>bh~SPNj!Cs-wb^_Fs#=X8EJAKgf(z?khN}%H z?=`{k>3@Q1yqaq8TJ1fudAQ-swanxJJ#rp8eB)KtRQNz8c}8Hd%a)H%M!~zWSz>zP z7A=k;K7h+!cOqnH)|%h$e=nw!AVC+alC0zq((aR3(|0v7k4@G1!mAe|tVR7bm4wwN z$U(Tt{R)}vS{oyl?n6yrA5HJvePd$~(%tjjb$O-LZ(^J`8w?x@_%wW2OPb zX^y{rw^KMi?#_VpL=8{nnBK?l$yJgc>;&HX2LbD5x(?c0i*P3$L7ss_caBLkf6f<9 zQq_^>G^=OrX))wWd{zR#3-Z-mW4z&`aIR_D8(7R4m9v80ACNIvP|_}y2Hke*&mt{q z(0^i@_DFSnS{yt`d-@;ss(T8p-?Dun>qZXcTw!sjMZ;rB#-pqhBoWf8V%t`G(*3)g z3e8)Zd_Erhcx~5OwRwHq*XdR=dW+J6lSb6HW0n8>Io+wAu# zk*qJCt8umCgOdJane?(;Zt%lBU4K!CNxR`Z7yyIWN&G2DrOX#Y34n5C^a-AK{>H>D zq)ni6woJ>s?GU4!fLuZJQcwFC)O4Cp+Ple9W&M%`vS_*>BG(rT`ao=GTlBbaeG6(O z%1`o$6yunat@$OE*C|af}i0gNtJ5IC0>8mWPi|mR$Fb= znbM+i=m$#N4_IC5t^HiB)7H;{!+*`PY~i8}PpzwC_E$pf-BM$uiAlPts_I>?>jTaa zH-S1}NoaiR^%-nf6T~~7RX%v^ukI*{Mgb+wGX|hYG5H}@6-%l$;5t_O$c}c<= z?_YW?esKXl)klk|2}|8aYM+;%W#gE>nW)HU(FVyN_ebQ!7?u1QII%a-{FZ*7<=I;N zWF#JFRC4lZItv5eLS__+Nz?Z@wbq75-e;M+saV0jYpwg}GxHc5)PK{6oB@*Nl(>Wp zUi-cYK#=QLWgmgYQeRk7)=*PF%xF$>N6MKdO)-*ytx=5gc}a}`6xqeZZ1_`b zV&{}yCO*imNL5Kc&3~mb7Ax#)+qLZqnb6bUg&CfpDOxjFJlKQY540?-6N=)5`fq@q zqn|F39(<05=egg3+)gaavc%b%dIXV1OG5w$IR^R#a1sv-cgKf%S4RZ0xnF;mDiW_} zm$Yetl&ay@_f@EQXgwW++dt1Rpfb#^l$Wsk z%DTBqf04%Q*?&-X^uei#rix*z;I>`A&@8nmz9)#54IypD9Z<6n5T1JYvuY|<+2IA> z+6M?1s)qo4&Qtb9UPF%k?vco5zvmx~!gu7ultu9AIiQK8+ujM#eRZO;uWBgi$fGir zW|wFUrWc;uP&H+{v5}pmHB;O~|HI?9p6%-@OV1Ttsegb!^tz-c?(ybwj^yj! zs16ajlLNXzQ=&3L-qEUBD6VT>x-myIhM!U%UT_xNc+xFaxz1ZgK9T%*&z0EL~R)wq^)jAVQKHDOgptc1) z`1|52$$t$ZzNtP8-GvL|-LF4!A%@h?2U^6AKl%kmToAKvs!^fwCU-b2ic3st%rzL$ z9~e#wK1zw_w?t?me7U(Gc_NR>ER;+q9<|Sfodj!L8is_i@`JdGwqm(;zuOkQw({0a zhah){G#4)}oDypGnI$BHKT)RDBOs}?aSgURHGc}gv`sESSAF(+ z!*Nz5E$e2Gb4IM1rBiJNd%AdCY*h33T`#}dc$@~5MaJuvBBmv4pKy*D(l(hQ)vaSC z!Y@vThd7+;ENjR^VH?i<>B*Jq59ZRdLfAPCa)7=}VQZ1_kc;twP1Of0ElMy)tF5V2 z+kdKAfkei>$XDo;Bz#BaW!nln3Q(HAc<6P>ELU4rYo?~lhjVAzHAGN}E?B{Db^Ax= zdy_U#>GFyCsi{-f38^&e2o64_5luZ|8eW1XpnM-m!19Q&7=QIUGjnib;Q&D8A=Dw;Jcs!sX=VL5r*cxv z?S%STJ@+wQIEq=zU#YWd2=rJo$_I#77}!f=N;>Jh(bJ7g#ibW4K0J)d1B-2Z2SWi&3uC$=8CVa{g0Cl?sP|%33R$yWOGP$6OnfbYHC>WNRDM546C2UEdzH-~J=WCl0&sgS($*=c5`jPQKm z^!4^{<=3AZmU4o(x~91dV-ka-7%V%mbiV-upWznB_5ov#c+2cTS`# z0dtraRhvc{=OdFAZ5cN@wST;-`oXv(APNFG?T6wJ0tqYe^|@h4k|9Ng=w>(3$;D>w zoI$xpBV_Ft!_}~h4|2jd+} z9D*JjdT7IkBFGG*u<{Guso(p4)j4i2<{K!w@~%G&a}$Fv(}x6EP+h*`B3W*a>rrwx zI^V;CQvcS9x0D0!GJl>o!&Igy@p4vPQ`egXdUo3_^Cw{Bdj8>38k?P9S>2BuJQ!FN zXvSg|m@RxK6Osnq3s+5qS?M%Za#Wc{6e|3cTBAra`cEM&O(pEt!?X_D;fs)GPx9)I z5V$(tKv*2~b04rEPFFjzd^@W3kDdr5zY05^6Bu;dsk1E>R)5?2kUEQvt5Xq`U!>DG z#SW*mNy>0{STbblqTP8O6{TtPls*b-An@m)?w}88#;LU~my2#o4(_rbeUZFTAfq(7 z-cSZ_<;{Nh7|e+kOmsfko4GLlL)EY0?fKk0YM0KtgVP6%_SWN_a6$V`Cc~2fTBJ)U z8qlb5R(m&fzJFh~pZSv$h=*_c`;Ymi!|UIbBeUNQn_NO@+<%+}puIak(?s}-gQSqI zV5xGP`hDglp>PrB)>Sa)_ara$}#NshLW}JviUM<1L@Tuwq^^K*TvFPEsiI7`j-YuJc#FY zkX;Din!=v1hCS_Qv}F-cR+(Kv0Uvwe4)`luX<}!PpSb*lF0a2%gfG@SAx#6fLUuZW zFD>k>-haLVzQ$CMJP+U>@~?YrOz1r+AB0{*tx}csh%85@jJQv9tSSN-ihCI3?k-PX z=zh4KqbHd2VesR*^xV!5JAZ1F;fYX^^k+;KoZf!f%Td+n5Gxz) zWr(;H?>SvNWx@_&8<1-*sM25bXw)U&ayzt}B_e(W_bV{*Hj(cd_8%mHQz7?6P-NA=uY=!h|TEeI$#4ef{eCW88dhE&UonJMj_o{bEfW;+(E|c=jimYvW zr@@NC7g;mDLi?Ep2L<%kwOwm4svs^MfE*<v7M(0_jxZ5lop|2OiLQC~9>svelg~*!fgz&AJ#grS2Dm_u>3As{i@MI|KRtXUz^C z`2Hgl_-3FtVC(p2ZaLCHc!uv$2}10`>PH5GPgfBD^Gg{TA-x_hMa(9OnjMH?%-?J{ zKHv9Hz5ltWA0IRsyK#=KHm_intACl>TYVW|8m$M#bOFbYat8TAho%ZTF}P4GS1hJi z4p?k%X>G9j43I;EVv?7fl6}b57Q8m(Nacxqu+lSMT}&Fw;G#sY zdA>iDDnGMCONch;{`zZM=6`dJ<+9T+cSuH$g|0u>d#EJ%@km^@n)bX;6`slJ|9wQ# zW_xaP6XGzKn=2OQi{~?ZcJnpbkO#o%@0 zOv_v;SLX;7xz&q1?4qiu@hk>MxxGf%nPTn}kE(e3c zv34S4G`TLR)lsnP=}{K+Jk4f11dr$Lx4U-{Qiig`IS~SW=gk=VfBxkNsYLn1o-(J5|`3294g~Z-0e&oLY5df3-7Tr-XpX z=cRZr@)=Ju9GfX>fABE!@6u7YR2^txcxvN>h(_b^$^J;RpZYiR6VU*D7-@Phy+8LO z=sToCJky}0$?v9Z`(g1vFh(f+3~eb^FrKvTd^^K32q>+{G0wmgzmOU3{?E1efe=9l zG-(4x=Y`GjAAcgfk${YIy@B9x+r_avp78dCpgmdpc3F=?5*dJ^(Q5JK3Ww(+lghO0 zAtms!b1y}>IUGwSYP$uSvvvR=UlI{aY*ZMZUP3+RDA8$^e`=@Dv5!#1pyFeG@o+!V zz#T*gM zS&^^bzvMPL^PeQyKF(VhI0Yq5%5`ASO4iqjVjP16<=v>ekZ9rk3<8J4PW6hHZQa2p z^ltRBe5($KoLCi06eBkJBmDo5_4W$+ay`!TO`Xi1Kqv&QAmH{q%)5XU;{T1UG{lGJ z^G@Q;&426h``>6k{584luv(wot1j|UZ*D4LH6z zxMMr`r?M`EiLrdAs|`pGMw?*_=OotE6K|cDqBHDe2$1+_;^*mdDy} zdLntI&pY4vRJ0t4S8affigY?wL(yER$A5lJwy!U)Pe&7SPbb$vhrJQnoBKuYgBHp8 zkp0Sh4_CV(7X)GdcnI_ih508~j69mcUpjvPK#n9h_U*qHCvwQYU%eN*#RO6s@BHzn za;2xrciiCjIWRswT5kwCGi!xS=d8j`FSHJGxe!K=}KE4ujg|D>nO_7_;d zVIZDY-eIkFtQxK60h*a?Wy7&~-e5*!P`@NJeBxzF>0ySjBaxWcDsVfJ$Mkh;_U72S zvshwFB~+MWBxTpl%x2#*jzkKJl79+0jf$+yZ$@6X9DVE4e`nHlLFLK+$!3-jcEBME zfHZ~EGGSzD6bxFn2i!T}BH*TNEQKqugTpgrD#&{@m8n3JEq!%H%vYg-ibeKNUWnX0 zVdUm|-1uO-U1y@>Q(*an6}8;Yp?Dgr^h(!zPj-CGX}&1lP5#GXS%8OwV}C&D!NFzh z&E#F=VWN&#{5`5jXnwJL?${%TxiYMQ zxKHoREP#-~xAfJSxZY|~6o1$I9ti)zgLZ-`Ttb@8r?FKbm?4Kp$s=r^80bW#bVI} zHH(Cm{3bsO42S|dWh=e_wPst`-6;=#rO`-IG=&nG1j|}dc!g+YhJV-7qw&*8EiL(I zmyh3=j;)Z}F`uK&(=*;7Fw&_+0R)O-)hKZiVrwZwrZaB-dUuSk-m3CzB&w+x*9Qq{ zO65*cDCxw!=AXD`t3PSlZ>FNETng5I@Et{?JevPun~?9X#AJ-eGxCinm8TP*J8CMH;itkd=s|=NzWhJz z{bf*H+tx0OCb+x11b26L_uvk}-9314clQvS;O_43?k*FX*hkjd>%HgQ?>TjE?fQP* zI=`rz6q7M}Z~c+h+8C{0UQ0L^JVJf8v7|F)TLzs*maEnb)_)6Q^7VC6pw9AKT*_Yf zI{;BYuD?0FmCC^>^Kdn_9yzCO5Pdz`QiNTf#D7LIh4NtkV)w)k(%w64-1IF-vf3!x z9gLY>{(WeIL8C?~kwlZ!SPKg)k7WPj-uH&fX0s+k-p_BPB_>dYzB!SEoO9gUhyV3% zag!LCEYqBqDzvwy{uh6=x(l=kB{^qsZ;BR-^<&00`hR$qG5U{I-`;PlI*-!JSbmr>i;LR}2dI2&FB&Yx=c!Kb z({;IvK0&9_&(37#Ma^VgqQ@&v2El#%=Ka5w9{*o|_|FA@|CfIhti04mE6eoP)CV8WISgW7{5{}O zgiR%;rHq)#=M~Lvw=D1BXoi`bQ9DJJqLXjS)IBV`vm<0lRgJm%Bj!KEm#lyAg_451 zyL71{X@cXIAEa&^zt2!>rCw(m0v>@EDMVXF|y5LO;DZtFi>qIeFF zziV0?T7Q-(uJLCsu0>=atX<}ZG@w@g`ALmZMY(?|xs??zwDS#^v0%3%ED=qvVM)Rl z4LglKloR?(l~OL{L*?`Z<7?Z)1SaUt1&j2k|F8c9N@Qzj+8EG}`F!-N>`emv_ORg8 zu8E%(PXFI8W+%Aocy@F4Phaiu;aFPW&HgxkRk?i&)u+$&3{+dB4j9<~b`OYPvd9Ut z{kwm0^Iw{iJ`RWPD@5j|+ZNL6%Pj-onb2=0NnKAUt!kW-e6#r|9PokudkO+Mc~uB_ zaSNHg&##C;^{JxDVSDs-Wkv?mX5!ys*2;zCqCcaT17^kCchw9K@%e=*JK3z_P^(Z@S`RzFlkfvltE4HvNCav1X5tPOC`t1yaw+l%OC;aNi?7y#{#IYvX*s z#%1gTnAmA6#-BByW<*mZsm z3=qw+?oIu>iT$Tp4z7qwNhb;6YPBu){!kp>45f)Xe*pizFV*-==@#(it3;;dIQP3KTAyY_a<$`2DT8?9;vYEknHg&j3ct zg*u1Y2|MYbwQ_a6l5NuL6ivAdcDnXX2R73QZ{2)TzUFzQS1}&f3n`*{L0kINoMeWq zs?F%t>l)+1(gFNCQ{ee(eO`wG0A^T5*{?UYPvsbz(dXniZUIZV{o8-67tXBHB%Dgg zpA8ntm9FOo+;MCw*5n6UYzJ@+?8wO@^W8sw(rPj<9h=!$Cme~QAExrl#1!v>{j9Q^ zG8rt^1>wf_J3gJ$5N^KIC)6by578GBcQibyoapyLlM*NnpQETGd;7r%uQ;^-$n6(4 zV;8ICUq%VL*vxDa3g>@5-==E6!y_R!UJ7@<0IejKdk#9`Pg6#%sA|TcJ+NaE=3ZBZvpNDcmsXFlI|iV@ z#E42yRi_+mre1i0S=iLw!ZWIvK!&mZv1>Hsoc+(RkA-%Bh<1OFqSdfb4&m*xqXhkY z`MFdUbdNkX6GcY$fT!kW=$f1;HW4)449pO*xZSReiH2|CciAb0J z#6%6{7N>LX7$twnbG+Wb)0u6O;llgHX}QTEuSq${?l@TehokXZkE1vaM}87m&uYbT z!=(LK(7;3gHCK4q@hZO%?3K=%A98nD2v9F9kxgS(rr*9N<%$^Qj?3I+so$q7O_IIt zih-8D7R;Ju4gS9QNOq)gHAv`}Pd#siTFxSzOqV9nxQ2gci9D>DE!V%>IJCp2E+1x$ zrqU#aqkvQ%|NJ<7o|U$h_+57D60?w&Xn=q(CzovW>&TE@nNm_ZYqN}o=eXR5(Bnv= z)Kk?{2(yv5v{l%b2%F9M*e!WY8H{1iJJYZ@p#@ZOh+p4!0kZ<+E@@29u2d8a?>)TB zZ)`6E8~1-#-V;!w0fNelrMbH+k{jIw3wD*Zh?9qrf|f%6qPc;jkDGV{;Xls!bKb6p zZI^5k0^jZAv+^4^d?jdEzm)**PQIK3v<-WQ`uAW)l6>qJj)v;z>rW9V8kzP9g|ub1 z9+DG_hBY98R+1TzTGLzp7Yj=-XmHn_lad?6dLDnyGMEeSi+4N<_ShM# zbU))xi%-p`Jh8sqDw4GQwugQNuG-CTFV&|pgL}|U_dX(L#shKF-3>JyJ}-a1xshQ| zJlAT>_^}{~2eiH7u`QEVA&+4}u~QZ^s+Iqumd%A9|4O87zZ2gD^+g}-Wq=_r68zyq zt5<($)P(<1t&~inPEzP#@F6zyjiA+Y+E3&Z4_kzU4p4I>5b)R@04o7=4)E*___s^O#+Bx|q0dPxyr<5Y4xFtPWiG|1N5aYL|Pe(&DvEGSiQXPlE&&F zGxpyyqARpH7zL=bb$(X>{#CZ|uEllbDMENtvFzyO@ABIcRz@&)M_!io4;S@6E-_$C zC>Syh+ruY`;d9|LY>9ruJjo;MrXov@&4uN?c5f7mNKvZVcr~5r4TUfK`k?1bh8BPG zBAR~SmnSb8E+q|>Gbdf_ZqS2~x#Q1WVgd8H^IFD|02WABd!@6iN_87U!4;F!<~aFh zHDrASVo9mhE9gjViP&GFe3-iTn-<+9sUDP_yl|@QV$(+>sAQWNSb|MX!Uh%IhfK*W zI_~@ov#lj&?9Ocdim@+Hg#+CC9O{2DV1x*xP=#Oq(duWZTWFUITL~S1@`D69XCdqI zrGf8>>6g`rSAkI%KS%tB_h*@E+_r1Xfa50ZNXU z6VHAKUHXc7BFbwocgCtx5I-D^{ikQ~%0pVX%kCvUeK6o-w|Uj&TjDd72PuDL?WHM6 z$vgEeZJm^FTgh~-kke6&al%7qu^&w3f4($UJslG^kA0DQdSn<&IF|%QwV!W)@YWA7 z|2d74mxGhItzJWH|HPnwJP$r<=FTg)_YB9Myzg$pdiAP&pkEV+R*N&ZPxIxctfwcl zmZGU~)fQDLYOOk?BAFC=S_FR=!_x(P!1ad0p2g_FgJKArda1Tyr{C)xW`$0Da>t9K zfREFO_DB{F#mB)*On(|Vy=g}24_9t>EE3saOuLNz0h--)_86XI zb#xV10C+vVr1E=fQbO`g z7SdEoUSh-3YI#4eTeNHg$L5YIfy`$Ea+k9!F1SZ~!&PL_3>{aQ=n{q76_EYi0`D&( ze=-(6&-b%oyA;POriJ{F`zSTdiSqauW1y$}+rvLAf_H*QY=7M*5Gd~^;cu`X9K z24^z8ZhY;gb3&3?6=*7pME0LovH|yIC;z0`&?Yw|!Gor(Vi?wU^9i$>L_ zKNKG6dnfZVJWXcweiq3a9#^gQLna#B;l!)V>BhPskK0+YDQkZTl;3Fu*-zE5oyyJW z@T;EB8_4@lqdxl3je!(L51W1^v>ZALMbb&Rt9E7L7PH0m0wj|K9*~1StTD4rCM}fS zH#TpOeZH*lHyxDIuBD?oDH7Fb4EH6{@8ESFCQDiREjT^bTP=x4f83SOY(<9tO>eMao^qsvjx-}8TZC5|7jQEzxm#~lM{1&XB6 z5xeCC(zsvIJz5l-Kwg9a!$@nwn^WJ}SSY3Lm+D@;ZgI#{%>KTLa*|ziM`wYjul8-O|ccjrrrd{&iW7OhhHrQ=8 zNCxl-BQ8BNnIj&k$p`XG#r(=g5>n!+;<+y{zrH8`iW*?Say#E>5ADF*B?f?-_F-Wg zj{}Yw5MQA0&o>oL7wr;FF>Tzgjl@j6^aQ;XB)@-vm6)$$748DY4zkc?utD=C3%R%Ej_c6Py#MJ- zkoDPruS`6bn*Wt#`fh_*mBW*So*2`3NT18lu_R&Ehqd*#KzDqPGY=T#jmg^K^}4`2 zN49@;qwRE~n^gu$!**{~(y)RKEVlk^nKJe&K8f%^~n%q23 z)#Nf>GajBl8M*gN>`+z9_*~CrbaB;qh=+3-+#D1yxq|ns!DH1>URy<@*9K)eny!B) zcKnPRRY^v*G8@CcBERZ-mdMR?glvoA85K_!uFN*guyQyLlsVZHoi|&P(A~J&_y*KE z>Z>&wC%L0-Or{k|6J2Manr_&hPE3)!eH~5HFU&Ttn!*%1^!za5AdA+jGEGF2Ni3mO z&T)yOl0D+uAz8p$^Sk*S*)um}V97z-C&XM(Z(O6n>zXb`?J9Z8O#-zW~#q+;o4kQ)f*h zzH+38dSy?`G!yV^xtDIr6ZZ%eV8F#{vG6q5c2y>GfcuCr4|j=lVV(c^MW*>6nMWPr zm;U9%4;R*HqpifMT&HRCRbGb7%nPyBWHG}vztGj>f>WzblS@T}F0%1LE8J<1M1)@~?_7t4pB9hQ1T0EJZgfzKDnSP0lDm1ilw-!K{t|jb?)a{Xr%u zf&_{vG9;2eHpQ;dZZX6;0GxhhHk@peIqth3Ba@%+oQ9|I#46x0$1i`F&*j4cPZ_tj zOp~4y!nBs?bKZ?NvKm|^C|?JJmFrf2pAzcZMvuXvL_;n76%2b2M5|2PVFfmAdDtNL zbuKPM9nomoWNax6u->>?Dzm@<;bf0K*J`G|*iVo|pJ?~t|7x@65gD(&9M(jeznAN) z52on*a`6#iSfA~GdKiC<&Jo=m=qwN5ep?xhZGUM>n(up*CqjTHylzu9SYmeG%99mE}SeDk22CHf#SdV_RlvFG3Q%Lxq;C{Exglq)kN&CRx!#*W8t$j zEEKcvm2J#>daQqUj9)-fP`#ck7g$x73vMbqWXkVgm%sPGY+|~M)0@*2%M&OZGbl(@ zkAT)RGBjZTaMO=eVi(TTH*+mm^oj`kjeaNO>O>Phcw-5{8Q!APZUZET>f2id3Tdj^ zNcM2#jm`;sNapCjTdQ>%s+@=4Zm)5eh0+y^__Nwysnmb@_KFWSu&YmGvN}uZH{0*v zS@o!$mHS%kuxgTDRh$KgrC82W&uR#fH=tq^l%o3X>ndzAl(*X+iDL8_6OwMF>wEO^ z+@DRSX>BetlLw2m-G(qpbiRpLY`#@ptq_AaJ{`p=$kbM2c2k+-OYHd?-r0q*w*+a- zSxw&v37LQJgIj9YG>^0Me*(>>jXkWElM}_`t;8DHfGDyi_g_VXRjs7(| z*9GMA<+{r#FQ!ffBtF81w-rG<m(JrpgLC==?KYLeTQ$Kv_OJUl*vL5knCO#SPx> zm2UGHTlz~C&*t!WmW-r>r41-aibjr`6( z|9u)7(BVXTK8gAoBKi91=C}7TuQ~|xz;GG8HEYm@-GZ}Odwu1@r{?2kvk>AFq_85r zP|-c7_)MVp>~68fX!9bcU?{rrhFO1(pZ}BsX9aE%dWH*AF&^&`^k6f}&dey+fj7~h zA^xX~i?e{c_-8G)5(*#yuzrR|199&So)s}LbFjR*c+R^}57t-KNH`Vc-E1dN?{O6u z+uR^cnDMshZqvfxvErPie7x#+cY}S3ClSMw+(0qJ`59}|&CjcJG=$!EWe>1t}5RQVX{6w7HNRSF)LEoD0sBj&Eo zXuL5AJ10=TukuKO=Ps6~Urm28R;u<}yT3g24PrSXRp!Y;so10d$C}b~(6G$veHPRn zHdOTh{jbQh$~c}g7m{gR_|s`gA$Nz`p#lgPjlEwq<&wMtH)7a?7(|tw#J@G>$X04C zqS`_P;z9;lxF6(9yPAb{N5K*8{_z%u=H90);f@9?-<{n<1HWKeDW;`-y3mGEFto2s!CoAWP}FL{CKnzP zL-iVewy0J10!#%AFL6Ib4@ZQ}t|~7d@hOYfrC2T5fmEefHjzR>9wg%Z$I_FdtVXdv zQT7MdF^x$Jv0f690-g#+m`=1;LoZ6QW9||Z3>*M2orlNOkI{cbctV1->D(T8E~guz z3zeHhtDg-%!&R$5@X!zuL04!tOP6oB^pf}kZ9nv(zS9SZKLixwtnC*Yu8^>BOG zL2%RVuqpcf-F*$r-K|;Cw*7`(I#y{4fqMCUUdZ#Ojl_R+v`+!OR2eBp`1-{UiM!dR z0n=JFG}%snxK_?#w<{wN65&g=b$@p zDt88|^flUT!LUg=lO!~ndnQ<8P-eV8?kc3Y08)PL zszZTn8XA9sBYqJAR3wPLGdmx#)Z$Tr6ungnoixE*{_!`c12Or3-d|u$x&{04{`O2* zb9?)P4WsALaN=|pA38!3^>y#nhFem`iF44%Gy05+(&@k|*d7<=XRZM^t34i39sA)gAZ9R zHyVSw8=jioJA+Nv2^E(f_0`VdTw>}%OYU25UkSLN+lvOdeAgtI8q&$7j4k0lMW|z` z3o?J+r&D2=(YMml)gzzN4zO{28zo=q$KAt%$Jk6XC}+*pw{x+4J@E-$#nJ)4hy5Tp zpJ?Q9jVU_%B+QoqtqWBh@{s`j)}7$2&EKABjML<4C;CR`DgGZWC>7TzJM1Ai7qL@( zu9LaPi0TAH6w!__L4w;iz$K(4TmFaJ4~>7hn_neZy{Z+yC0{Iz*Z20l(C!f5|LVmEK?NptOl`3iHDR)Kn3?YXWhGU9?_7UV zjkS3gCgeB~<|ZP5PCBZOAo$0d93vqQg>y2Nh>&6> z^S3swNJB&|V~xJxag!0CholK|X5UJPgo<8(38HZWac_2VMvkg`B)kFLi*{bs8pV&y zsaD~OrivE4)0M#+p^>6+9?2z*(dvH>E)3pKSD3VjG!iBZ?cezv7@}>gQRha)YUs3# z*+U(H)0@w4e37lG;WORO5EB=|#^$y#>%tRV&nMjGNw`%=2os=Cem;3pe_Co>l6}J_ zRPMZz3b$^M<~U#80OurcEx2$84nv^}3B2jU{OLNrI#0)i@S|Q&JMIbZOCx{3-vUk5 z=y>;#YxyoY>b}mU)0wd_I34YJonKvsX)~;o%2-k@H2zGZQ6MaB1&t4yv&IxLm!{JP zPmc8M4YPdF#CEFD5qA3=5i}U=247Sr44f}h&yxzsz5G_X1_t5T5ilWUY;%j&;M?iR zC%h!Nh;kLV;Y+K{tZ<~dBIAEUeL~ly1g_z~Cvded+k}*#PA3H{5|`36 zij>8ym{41#fdG72?FQ?l$zGj{jDS58Og{ROVWQLn#D-H!!rrYv}eY zfcJR9r-uhe7H}pNn+YvVQ&e6n{J zVG;^O?zJ^tn+0#(QGdkUY4#SD>GvBAk0e00}N&gVMRO?y|J3MppU^??SkMf~5HKdGQgCN9^M5<9i6MpD*kyjkaS} z-kl1}k&yd)b126WyG&D%+oB%E6%0;)QFL9nJk(TCa79eb5+lC^7tmq*cDCmGDy!aj;I8$mTF*Wc#|-jj@cZ z-A?jj&+zBdw|}`QD_=7_T-0{@Hhgt2U@l;=4+$~~UTv^NUJQh6d^}$aJb;JggBQ)o z2g1^RVQYUqv^7L7Nqvbva@s(>Ab||y#~3lD{DCp%C(@E!SMoGeuudj~yeMSwA*Z{E z7r+d_i~ZgWF{JWkN)TpR;`j(MAuImavzS$e@?W?@gHTw9jY#fG$@pWJS>`GogP8=<)~{HLHfyIvm#~&DnJ+PGb~f z8aD{+H_=f+%O6Q>=VK;2;y}l@I!?mwKmidn$O6AV)_EU$^VB(>xm8u?|6C(BaCjWf z0~N;Lb30!l9dsj-6UqzE%0!}ol`mewx*Jy`#F5dCdc%O7wKT-sJ%?OfGqa)Y+F7SH zq`QCRs&T?Frict{wARDi0b;k73ONUMrK?@)>F(Fe5^4yF`t+s(*Xw2QIqW(4O7sKo z0(?PHT#b!?LjhB?+}6H<`!jpL^ihGe5o4`(*CwuKMeAMGV7)vO`bqYvt@BvPz_tz* z_`KMW?KtKRhB^D8Vw4R)m3_8%3cbd}So?oA;h9t|2_5GICv-MJvH`jsjzRUMic~k! ztfmZq^V9k1iCV-SIRf@p4`C;h`xisCX2*16qHM{50AGhRO1Ghe(%LPKiL0LF%jH;R zWaEq=H}?6?)T5L;HwpZJx6c^%1lF_Z;TUvZbVE;FBoQy!$q~85{ck#y(GB_JX+3{? zNKV=sQGQ@`4bj`@gxEoKgT!aF;ygRWWj=q1hx@dl6ur=?1^~)eFsm>z9n(_i%2iA{ zEqP=OjAFG*(Na7o5aY>k7Q&V4`PW^7=3 zR7-V)w3hp8P!d`&5-uSkEZ~F1SmuA#1>HXJ#cCyr;cF6Y5Q>rFU}&*-w>*1RB~$%%>1BhRc+n6V-=zekm8@foj?iI~0LzzGh_xP7}9BG`JMcOWiLf!|ou zsOMm5`8Ro}4@!Gx;29JO=fO9T(0kaLGl)q9pWqAOxW08~F#=hZb4Ktcz(B50iwp-xixjDwBJE?hEdZArD^ng#1}I8h(S-Sngi{k-8c z9|f9HKsS-)e7IW=22`T(=gICIj?{NcV%ZoOdx%RsMocH;T)&0OJ$`aB zefCq;i0dZ8+z}o7pfWms@5gQS8cqReEkZ{U&!5iK)hp%upf`VY9>2+jyE#PHVV>Yy zew3#WdTELUh23RULj~Dg0J0Phl;WQQwZ4q7`ve5K8}viJ#b*^7A7E%3l5B|fv;#k3>D10usB-3BEO z1zN_pwb^UBuWBY&VRd~uBN#}OyEnvN^-pCGi3NC}u zdHzTzM9}%1{Y+aXBsnT?&%8Wvw8@*?fb(Px4YJsP&siwJU01~4-a2ahOo%kN-Gd&$ zlwivR3Y(1{)Pr}9O*>&kd-SDRC&<2w0Jh8y2KILx4A3)jQ}9>cLqF%#_?hPeflql9 zs62l?HNZ25`IDPw^*v6+kwZRyCj9}bd;|m0RDrD^QaL(ggq`?0!tie}B&q1B*OLIL^08`-m z0p;mDWq>wcxq6Aetjvvzj+GBw^2R-K z8I4ncHmK!+fugO#V@a7qb@4h3?E6&8D73J`Zsier6At+F+BFElaviQud$WI&r53l( z6?X+}`BimZs!YND*;PbNN~kv3uRd85ausAp=a95Qoo^=qq=0m3+*hHj7B`u(nh;?0 z5~~H~tjh>N=c^7wP7~%S-k|8}F1vhJI>f=k6oZkRad@>}6dMGFI8i;(KhLdM{dvyl(S+lyVZ+0#d7ytMu%#BdHGI28^24P(n;^j=AGHxx!+btPvJH@LHY~tc z5dJ;8Q>mtWnL+gL5|Xo5C7mA=MV!2t=rtd4CaM>=eF8#(W9YHWRVtQH*5P4WvL> z3g~~qw%phym>c9>AH5gF=k9!?c`e89kPiYHpJv4e?#3vLOztq((Cm>0%OpLy8MLGbR+-$&+S_7MR&=eCtI3R z!}svdJnXsE_pu(fV zWk*zv~ z(Ef%%L0BuwyWoO`60?VW8O-tf0o0X@<^kB-KU}Sx+6~JP!K9!iZ=3}%@s$bkMk--SU&LA?$kEqOt6V$4ii-qq-(gm}kWB{fbO^z8 z^_@I1C0Sq}c78{(>MZZ%DTgvtnu*{o40XdvC>VplF8Ep7?!9kmk%o0hWEN0pf!TD} zGr$bR?qGn-%hVUKWzsU)M}My6g0gX(2dO>V^%j5Q)qzs3*98QJf^T(eBScv+c@pgt z>eP>|HTX6DnG5rau4K$bR$;xA$Ig{mj6DM)YzC4B-TZJHbrbYokh@lA`x-Q^cr;io zB(iCedsFU<9`@T+BIisNC6!tJdd`JM{wjk^7uuKcxz`%qg zm-bBKHkftCZDEd9F^QIa1r73$U?Vberhl}s%H{Thl@Xngoq!+`4trszRjp+-Xu!nA zy0)-$Oz~-=6?q-5=MyW_Ak*|cXB29#gx&J5iaEDWtx`SUq`W58j3A}Z|8%u#|BrvJ z*1HMZ*6&P`PN+Ux=Y5Qee*|xir)QzPU2JYp4jBgNOKP({&F28bzJHQ9qP%Ygj+za^ zE>he>DI76tTnET&cNnhaYY&1wL@Ymy5Cf-nim0Wn6y^JnKTz@er;fIo-dPzEXhaS< z`^6wmx679-ptIW0LdK38|ZEf|)VW#E{?$CmbY~1yn#3Ur)V-pPs#8IvSpyhuT zM+KMF8CI6|Iw}n_LxZYQjDl4nuGo6F9im#!0Qf!7zfi>%ouLf#EakZT!99PNr7Rxy zP+HEDZ5J}yaw|?vAt85KS79lbCxhK|BpL0$kPmM5D97imNN27;pe`doS)Bfjavx^m z6-wT=l}}BlSAy3rO?DHoAuVxACv=(@K6|Rppf3+Yd#8&eL3<~2>GVz3YuGISywQFn zDOQDW{|;X6j!Lg-`SSSfh$4TIP|W3Yt`-uyt;@L94{9`MG?*wtVb$#{o}L~#I3YsA zYp!K=4Krx@xD{uv9unGlqru0_^RhkY-ZO*fm8*VK=rl;*W07SK^Xoi(Cld%z9YD7v z1=Rxum8lm{t6S?#RvH;t&rfMYn&iG`)9aP-Q}>ykxh$s6PqIb%Zlr%v{ylWnIRy8)u0oA3fx(`U+ijwo zibfRpr*{Mw=rxkKABca&50vByIJKtmhqd*VfcgUW zKVlP&JghNAL$9(wVWN43WA;gc;p`=QepHm!Mwu6g+19H50a#v0W!9eVB}`m1u*~8@ zoIjk+NXvV&kex`H+Vor8ZF}I)tMoS7*wxHqL|u-=?3vfidcuFf%QXcB^TE<)WSrI4 z2w~ie>`Q+2sXnP*gTML;jRcgkOWg4d?MmcdPV3H;d1t)9=UXw|*%}YF%Tn$lbw)Wu z71+hrqUr$bc1+N*CuzFLt4aDthM7>qxs|)K-u&O!~gOz_e^$g{b4BHuGtlJqjWE3VVDOmGjv}-Kk4N<`ht(RwJ>eC#FDH$Lh zcatQZcl8W++1wfqr+qIL366cfC-@GXX2~@ORPzSF{~}Ub(uK}1Ppoopt@EyAck4~( zGZ`&r19O{~Hi*N;)iB6|dH}p37|PCBB_*%gIbPffq4|Ha>1wdTG25oI8KZQIwW1{a zD}}w*+k))uLqtNSnP?!UvEO@)iP4{z(9`&CuzM2gn3r;E)DwKu0}gG|s>0h80;ur~ zx-oC0G(iF7(-%6(Ctu-hbc;S{+0+Wy@HGhZPiTZns^r%jd{}@s|9~v%KgT17{D6~u zh2WXzkd{zr$Z=V80!6OSs9-$pg#NkE5j zQ59oZxR|akpy@0@$Y*+!KX-9gaj>P8zNFunuR`OtV#fJ7c=!?skLsG%T?>tYn;)nC zty4&WKk2G`R&Oyz-U`?-AUzSbvJ+F3-3!Yv>u_*>Zha1TcigJ0mG+KrBKbl=%)6Anul^Tvw@y=|kGq@T{{G7ID zDi42B&JE|EjxXhBYfve~JRZ~epU+^mnjPuzxja=j8fZ}GSUw!K=fJ6Y(r&qC?NGmF zW3{LDApm~f$sAU3f#|+p&&S9phm+L%&&$TZFp7bT6`GMU+4&iJ47hg4hSV4W{H zdYXsPp>U22+qhD;Cv_r(&Q@K*TOG+U$sjo!u=&4$cKpHO4k{9~bvNczGG?Is0Wleb%i z;^VjW7GnrF-BT?$mn9F%R0c!I(iF0P2(;{H=6>0so@jB7AKu5TQ1^MzLcKQcqSQiz4yX|Uc zjqhZsf^+!fWjaC0dQ#3?xx9Vtp$sA zFLiEQItzyIyh`1Jb=}PMWG3Qkg&xK47R!F&7T$H!>}6lO-DfZwFC>l0g>}z#>Ktm& zf25YuCIp7407|mEtPE=lhRS7eV~HzY5O;PUG|SaU$7K67>+&TE zeuNk?i!4CCJKZ%P9v^M=UG0C#oSk?;DW7eL6JT};EW`m8EaJu9>rbskee~X>aDyR6 zsF%EdO>;70)uHe_POmC+phxd9)!Ofh-;&huKEpQtVs?&)Uauv<^ZtK?=B|0Qv4esk zg({^Qy8j}-UMs3a`O=1>=x5Uhq@kz$1!-|+>*r02O;j=qxaPMZQ%{-0e=MvkUq$8a z?@gy(U4}AO;a0tgzybB$drx6?7hx^(SNBE8ueE;^*S_j|xhMl(s`>4igSaEXu~xvF z+zSwIJ*x8cIjvGfXk&l5%$u9<5_3azwIGIS!`CYL3y1$6m_x1iVWL9E092vqKdmTm z|C?46yAtTkH$hr3il&4B?b1l>syuvcj+1F_f>ZjRnHVv02dhk=D`M*-A~}OzNUm=Z zq(FI54t1=1cS1iSB)X=pi;0e-N<=gRmwttSkU@knVX;x5*r9(Xq4em>(~ZVMU}1#R zIgDYg-(C9MpC=dQ!jBn77slsr)bola9EdwXZ_J*=Ax(Q31lGHT`vTKY6n=GoVTHgX)r zn;u=#LiuwZ^PPVnCfg~EMVAK?Jur$+H<6&>q%E{BmwI=wBL-&&{uwB`*arUQJX3C{ zJa4qDj_L0?c%-z#thPBFE z=W-tN{Gh33JA!2Cyxn2;xz71m8YoJOgU9bEszvV@K}mmxlF4qfA&l7IGk2cETcz)| zUSH=j+L~>cX$ygkxy6;>;pMv*0Y)PdD|XpN^`kp@uV$5y7QtUOTsKup%KkMo0#UBeGEL{Q$)^*2q)POiw=1KtTMDy`i*D{Q zM(jjkoiK-zmrv#+P3wIMzN1l(i!}Sd1*Hqwln-2h=&1c;!64{RB?rson3qFH=_Jcv zYd2dd3LUTQVy}b-C^3z{G+RK@FA=R)|C-Y@fk)_bdRjR%PPi%QsOxLzU zt(FRgRNj-7bpAjGu}?#BujDJ$zRlo5v35q_7?3qcnX--BL+W2rPcFAz;L4uN->1Kx z#0!7u>-|!#;his2*KQFB=tuh7Wv%z5)As85r11&(0xiAVOPo3ZvPZl8Kyx0Ki-ind zw6DI-{%SceMz@k#+-yZ+5u*%luB1mP5`6Mi)nQC-3$w0&>hm|Jplut=CeMKnvlXZ( zxC{iHxq}Sqm8|0)bcQTb?G>vt3l%()o2-8ot%4VSkB8|-ZqPWbH!fW1-&;E*&1jj| zt#U|6Ncxtrzg9L0J$>EWM6^JZ5_*eCo6qk9Icoe2LEE0$+d@AOcJjB+<;|i;eCXQ2q-K zB0OPGDI>p)d8@@@Cqt9;es);~2BTN7_sqk**!Hq1v7&_4(#=3eV|GLHu+*e-KF@PG zrWEsC5=mm99QckVQ_1ZPbkf?rFG9(;SO9W)JOE*ssF^53xvr;LE+-pu`uKmGlCJ&D z?Qp+tNhM;Y&)IN?R7&_qtQ;+VDG2S@KQDC{yIDv{?#sOjJ7n#L;Qn;KCJ6wzSD4Zf zVFV-`zX|QnUvdoXBy6}o_nUR<>uGp-?#>?*rx<$V+q6#1aCi>@NkF#0!8-)Fva45S zIzm2jY-Yxb;sr^Cfl;*cu57XlJ^A~85JEy@pbXgQ$#GZzOK)CRBgOkOz)W9%!O^4{EC7Gj>jQa)x=tso%0mUKZ^ok$tZ_JvHX^F z>3e+e(M~k%du+EPb$<{scxD`n-aCw3{&}XT5lyOE(2Rmvg8RTReAAcCJoDXuG$0|& z(X55mJ&NTjs{LT(?_({I^Lv#p@yG_HN1XcptzWeB?BU{`7Xq5lB+@A6ft@LYmwF)H z_2FVN21h2L0RAYK>)$wtG2%xd7`|JuGFL4*CuF;-zkfj zO5GD#+C;PcFR);u5lBX;bb{9QoA6+CcLqVuO0M6%=Qv~3iE2xlRYNrTO&90$UIt^S z;1`E9i3~EqFb5^pn!kr0cP-u092av|Amqcm75*TM!Yd|!kMI4ZfFPvkV@Fo&!-!N} zi@s1FQ#K3BFpYl;27n{7mhKE}aJ1_SGJCSZw&Us%= zeT2HJo);}K*J->98ntm1=|I7|M98@HwC4A6*Rb{0+}_*5j3m^CVD*sSb^if=LhW>z zpJyval#*V5f{eB?Ok&v;ZxL^cE?Ea6Ry#Vqcc^0YesP9fOPs+M7_COb->f)Y4rrhh ztGl2hs6hd={BoJBS>$paTD2mnMFB>5(9$VkOt7$Y;nfDKitf0NF;MkjJ;I;bcG2l| z6hoxiw+Cz`guKElssoNCYiaM z&K0^6x2t+m8QpmUc108A>c_xv@D8fy#AoT89_ zi1=~Y#t3-4c(bjBzX#&QZuh>yXQeD^{4D9MxtY04Ja;b7Z!oN?8Pm-E>8XDOn?N`t zF+cdZj3Y>Br2`c5ICkpzeC2nFFYwr`I@u(g2&p(y&oOahQX-3|?8ftT3|LH63x)W)xfDl#h}LhUYu=CZ7Qk#j!_;^!$92npp5;}q zfP1Vm^FWyf@^m=jLPLk-*+da^=!0GgAsz(V2c$GXAs64e4Ej#TgQ)F|5=Dp0Nd|;} z+_NjoTi)4KSu<(uh1>8&6=%@%0rJUCR(J>OI@|O(70hH-gocq1 z4#4eXJWjfyDB3myFuU(R3r4usnN9Miy_7;57q;rXYQ8o(CUGUfbd9o>a6w)$E^tCPJ9**3pR6cAc9XEkqdxLyn`x-7<2b>>!crJB*|CTB29 z)`jnylZcBJQI#6RB2q@zBPg9>uWXh|AwY#=r3`uni%y8WTjYqrXS<{ z)yYuOh{iYedBAt*lU`6iARt&~7+ko}35Rp}%s8OKDX~`u<7E1`Kzp?>YW(zw>7;al zhzH8{=7^dV6px;eWu$16jJp!JjaH`?_1>No%v-7M8lx~@onI&e_zCS#s9o`0iu6CiSDFV<+Wzne2*3+-$ z;t4(RpT9rURCSm7nf`tiQ80o5iCI4KYHG)hAEr0moXKBI#+?j zea_EvrfvyBI>>6EO`T%D9Ys7ib+HMVy&aA2YjVf?He^8~MX=f&`x#p@*C>YCC9JRE zjnjdh2q7eY-v$dTMp*T!Q=tQI&)oT11J^`vFdf&(!P;_I@bTJLzda5J_jZwt*O*_s zS8>Kckih;RH*hY$4$hG5qn1;a*^6_AUnV!|9*`IF1vK@xcSMGG;;sIN1nyT-(BMy% zuaCutO!`UqK!qTh{+IbL99(=vrH9sk>6FE5wF>2buwsR*-0&@68SXx~2x0SU-Xca{U68)OtRrwZ6CX9707Pw|IQe73jzeKOCUGLeU-pq6THtNSUY z4Ql;=Rq{8#E4($}q|@>9_shBRjBTBB7y@>J`=MmX1#MRAVk3fwVQvq>p~^Mt`VmND zQS{#1==|DlW@fjSs@5g|6HbYc18Dl*Z^3JPO=gEO4k6+b0Hn&q#n4fPPTm7(QGNGc zDyl0q@j8-WR%JZ45I8Ap!hr~1XfsUthrBSOG(Z<%&dQbnu}19JYQMuaxGMz-hGPsNXagy23A z2?%ezgkGUJL_t9tlF&{hq3ID5$MlSjHBk^2;MAYYdR_W$J=@B?WG*^`uB7OD``1QX zCF=o}c6XtOX#}CN!m^^mUj{^z=*R37vm4+tzgFnUyh6ami2-n0uQp&XTWP+1Pmu(r z&_18|rhOI9@~LuBXZdnD*9PQ&=}25wzCb2pH=@E}x$5F)g1pcSWSo+7_DPe&Y_jsE z{~ONjIdTk%?1PboBD|fWUTx?O^8VuAu;N7pg(Ey)VcW1$vO@@X1-F1TJebnk&F_G^ zG&70t0PiXBr$>QMQipm)9crlT|GEwk>m+ijOwnH#Kj(2yaq*N4qB*O745op*k>3mS zdyBB#GWm@t5!YkzYmsz>2?Jv^v5eYOEa7mcbn>s_2 zKWGA5oMQ2PDKmh}Qv)U;2iFc4*u;L`I@wRMMJ z^PxX%r?3UZKaXdy$k_^i2dJ-$!=EDaLL9T&UomQR>~J-2w~9P31dYe#f=`a+G8f65 zNQvM-`z-p7&D^F|rRZW0bM2EKS>K;|4s$O4UY=+lz!UAFqyv5}$6a^tg$JC_oceI& zZ)jJAvEF=Nyz}vD7!d23TIC|KI$P^VQ+0;LM|JVOnUhd$mbUSKIAN`>4fEHdPe71P zqByHk&p8|%JvMoiXbf4kfb{Q0^~fpO8)K2T69}x>aX^E4VCk4lW+#oL(1>_AzN_gc z_S5N~srnnC!%F)`M72C1Gk{X;?;odIRTjT`HMR(Cmr+W8&3WBu&GlAp*sQwE@EFCp zt~CwnB*}%w{8;adig$fCI%L53)Ej?6Tspv~W^q&ar56!3ry3QH1C(N2B`Bv>pKHxro3|r@SK869{#^6Y<<+eJF)V3vsZP$ka-o|o$`Dm_{0UyY+xi>SAn(xJO zrNt^i=%Wzz+()tpt*Q5M1?@Tb{LV;x_I}1`8&8u5k6U}^t_054rCN8MB&uBwPrBtA z4c{G2CJzSUUC-7?-58D6goiYA{i_nq(yB&fyo;s z+b}2gzu^u=T!n;wKx4Y48qI7S)o#{sMb3wR z%OYmUFh2NzQ!0)~A#yrriO*V;ZC5`TM~cIVS@#!(F#Q9U4i{DfA7 zu>#Sn_|0YyQlM~Q1dnJi#xjmAtIJE~;r(n*wXE+-t`-z)WyzYu5I1@_+Nzo9!Sc!h zIq$hV>5Ni?f0wq@o*CzMEkj)9A{@Ug=S-=?1iD<(Clf5vF}rAJDI7mH|qa?uXtF= zxFL)+e%bMyeNm`%F6@tsTHcP&{Q^MBD6J(a~?L{>6X=cz7{Vs z=`*n82M}>XNv33ELhu1uI2-0CUI*~@$6-CV;T;|1DAt!I_V!in4&5h4%~5@1oRBZ< zA1)S6Fvvw5wg1*r0PdBOTj%V5V&e=eXua1^7Q2saEUbU}%aA?SD;%=u9oyXT>DAKr zI5pU=N{s}jqL%A)#M+Lcc4i;ZvgELUV1s|wb|q+d{DQ;5ZZ8eMy&B&9D)PI3u%v0& zHWa(lLKnoj-Bi^^l!g)WdNV_RbE0m`SFy^p zHngd7%a6?uJ0SJ@n{077`UtrU!oFYc!1#Vp@lVA8I(=pV`lyRx7)hjkgnE+2|usTgSU4@EH=Y;+~`V%`_WK24kVz3=(ce5D5>@m!|7 z-cy6+YFdNcCK+Z4J8WQo0FBrq0){Hqd$yTtRp-bXqs7psb$@6#W_8>69fa>|Ur+^{tz)!@0pkB^`7_F}D)$oH z!`(-}X|q&Dq13srM%$gD$-9yc?VK@Q(FCX9Wg{<&X@BTc0@@Hv@D<1_x&(go1*`e) zn1geHVW%U*KcVA)bd+W0zf;KJK)EtlK2f17zmK#PQm z^H04B|2CE>WB-n*Rd1B16Zs(#88F%i%{QVq1n&J?zO?goAN5EXc0lhJf$*->*~)_g z?;=_3ztbEgSMc;l+Qm*3>cHIEWO90hTQ1mA!VA&2T?wauvp9vb0zvGurWD-HS2DK5 z5CKO&LQ5?YLOkOI2g9gwfjWS8B0w{4@aJW8;SFaL4D`J3U-!O!$Y#!flJT@z@`3BX zwVAwI5kR`ekYeHrguGq*EB$=grYd##R6|Ob-KU6Q`59A3m&t!1^c85i9zkqkX$xy{=T;PA1{X?YmRBcie!hK@RtbXzDLp0Sx!`@lp(R zt%&D?4;0&SJ_|(LZx4mMEBZ>80CZ1UvXfOk?sNt)=ytpH-HOpiR5s9aMs*hOh-Zy3PT-mTSN zSE9%YC`6a}3#+cVhY--gt< zdT0KBkn=hG-bP#5j~I~~y4HO5S$k9tP6Kb{_pND20@G#angxmbbFU7D zYLTOur>Cc~yL}S-5Ra0@HST=Z@RMGCU{t}4{*1*sc5n7LU3Cr!_Y2j9j_tw z=$rYmX1hJHEop#g^Od1Iub!blc@GrE!wN2cBkO7dc%OEtl2+w9&Zpyu)geZQ6Lo@Dy zheF|Eab20KQvtxspRMv!@!Tjj?z}&L_8y>Vujr6y_e;q9esgroV~wHHYFXo;JEiH$ zZw5BQ^G}{#@Bffze|0xxUaRoZUuk+`u!B;xKAd#RKvMLJ_DK2e`0b{T!uUcPAAQ9E!xi1VFM1)yLHsCx( zLd}jF(&OoDU(Ra5ih})BrY;pRh7NN+>@-j>Hfr%9`1$HysoqmpxQS#+*Hk*3(U6X+ zvFmDa^G(JM&nKcd9GonFh)x)ab)T%CRop`kgUsog;UqOc_X=tq+Dfj@}BbW#6PteQShIy z^x^O{J#`&M5SO1XQ;*B`gjvf@;YZ!<&l3r?m`TKDavLg6y=;+xIW7~^C=-W239l*r zfGSe%lmVQldaxxJB6_%NMCpm%(q@5OhX_2*27@Ul zVDau`0r^#$ZzP#6!digxloi=k)knWS9HW>l;(+|Q<7J?)dS4ir%WLzYi51qTJ|CtU z=S~Q`T#=rCd)VZhojM-01;a$xq2}=l7K{BNI;Z|gdo(!uwe|Nr#>*aDRde*_Z=qGN zjffn7)S0EDUYn>7RW0><6Pz_nHaCd}8(}4eBQA8K5+_ zI*KRyEtnj@53}7w^e&vuN6zjcc;KkrR5u;cKEO;tzz*opds?0vJl@`)s>J2^rBEc3 z6kB`NxalhciP%oR*PW)k`5oBo0r*WOL5zrh>tuFyO{)Dbm98YN{TidEzm7^&x#?l1 zq(!~wj>?8CUdpKYmDO;~ywpwUa6%tP3j(b|0=`>tps3kE}z`0;;;GIRFFO15qiG; z(GQ(ee0H-AvG=6{hbbH4o4XXJ`MkgbVe9W*-`qIt7CoPyQjsMm|H*-B)co&rpm05~ z(p*8g8BA8$YZXVdNUq3onDooUtn0s#Fsn+sH;a?a6;<{4@ zB_!>{i=CJ9)>|)gIWIpIJ8#|gElEVRddBkq$j!Zl&mfrWaba8A7a=|tyV76+))<<= z@Ej#ExTb!Z+w_D!WeHo$Mk|F5fRlzSX8a4D zO~Oj>3fvj_4E~G(xztDzt>2!1Qngv)O#}cQQ7wTRD}BR}iwB=P+9kG)O5?`V>xfcs zM(=czNPAfHR+$agdxoYX4oy|>f@mtmXcx5X5`(=x@n)y}FId;P6@Q2%8w!q z4U-Ps{PViIplf60Gd7YqQ<2ATfcOE$kFRqlfaSAUafy2V z=G?JVvr%CsL9O_+uq_jR142F)8RCbcOUsQq18Ov}F+ZP8fU_y0Lgoj2qE;@S827Z@ z68W%UJZRS+3scBvi!$j#hddos)%@6#FdmT4P5Ye_Ms|B@WdsdM&;{?cc+u`-YMJN( z+`D$z=u*Y-HO`mDJm8>I$oSa*v#AD=MvWhMW**25=655(b@tnT%*y91MG*2rrrkt; z@)5tM{~Hu)BYoJ|-%K4#Sbwq+wI^t-K3Gpuhn+5~2wWjQt?1G@3WyOO)JukanFPe# z8Q3@`m>&S+C$Y6FJzh-Dlsnf)4*x2L6+C=zOt0+66>^K}~QU9j2N^W3fw)oB>_% zR^to0Mi_9stq4^H>g%Y1#K^EkPjGl%dCYR&ZQX_Z~j)7A6 zPHv1~#v|~5JX~T#XVdcLzQ*JZN60OCGPrANPQlJ-pInx=1a&0A4&?3RwXProI~GvY z{8-=EiF=Ns8}?TSo17EPP5?VFo2ZsPQvQ^~od`n+V%iuiN7guqH=?Cu3~1)~fg*0{G`{^_`X;|Td<69MBc zbCH1;8$ie#d4JABw0=lp<5r{itoR^bGuR_<|A7=;Xu^Q4?%sU_blEOvYL%8RQ7Grz z@evHk^al=;AExsmaPIT*3#B9vL`)s4y|Sr)_#CVZ zRzYW^iHr^|Q?Zh0GmpCSuL{qHB$bH0ggs__>*_^x-4XKmd2};MUjV$jED{2Jo~r76 z!C}$(VYCIdp)`ngMlME~!x7zqZ59biHV8V7qmskFUKENjNBDuPLC=gx`0P7OHBJD3 zjmXV>@ZY8ak_=L`SdV_`^w*l$H5y^uT0?Sk367Z0yZ97anXjz5C=nipi_c%0Y zoQ;A;h~;HiF5wLl5pAy9{Zg^Y|RXl=USM_!;usO447KA*uvoE|Mb0hwLiQwlh z9apUi`ZhY$NKLNI-K+87PtS3PS}Y+q(9u5Pw=5K9Hn@AkB}4{j?(QMu&X*J6C2dFA%LT`882B_~-AW*F^lVzUIvZem0H!viy5H7|c~>tV1d&rYxl@6wg&&}-{$z*)b4ots;9INIh! z%lf{=9M`#?rn z28QTY!2ATD_?GC8-V>UCWOUmIFu7IrUQE3-j9&h^?NdxLP`RxukyL=jqU~h%2k|Wb zz$BPQD5d_~+o0y)b4FuPyVt|$C_}j}{j-@FPUa~ARo-Dpc7Keu&!X?smaix7ZI)?> zZ@y0ON$q=I`=O{Lje8q{u+pkWDib8}Ny%g;DIYtre%2Nh&PEe|4eX@6t%>E~714VT zW3 zV2d1geo5d@9|bI~?GE6$Nhh4{L|y4_>B;(qA%)!Cp;e@yr*KztRwx~R*gyz;Cdw;> z_f6ycl=byBm2F6Wb5!s=PJKKEHaCn&PKOaOQ#*3dh)PfH)@W5nzV?Tt#@aA>fb-k_ z>PuUk!|xN)@p{37y#PkPQiVn=HiO%}1QwpjVm3Jwe5g<4I=wH85lwHIR+9i>$W4!) zXLPeSj6MvCJQ8rzgG4NooS}mEOKzU3y`2IJ+SmJl)`!G@)$!#t-Rh{Jt=SZcR823V zQ?bL35+f(qd4jr{^_;OLjnJJ~2TrG;lcI;jP>8ir5de1Gwe>YxbgsLC*1`v0oS;`C zI&PVt#jJ&C&(tz`_`oeSec$pE$+U5MG`M$g&u`Xy*g=pRPgG0`hHFjo?tGJ8+O|vw!3mnyCtAv70X$SVMw9&@Z)B3irX10uu;0Tr!(x@MN;%Vu}vp` z>`5s+!By>jl8dY~pAq43z0>(S(GS2~~mn!+`M45HD5Q2U+#F)psod0ATXM($L- z_2P%TgxYTYe@!~G2=5a8{5rw?&$|GBwo3(@-`rrk(~H4(#;`t%Z`s9kb@)J*9)nhx z@MaY4#T+Tc%si7x_JGW`E7aGbS<^o#@5L5iJ;ikPq_(<$qeWG;er( zMB~V76nF+Wov>^fQh%+3;+v}|u+lUDn{Uy@TstI5i6`sCcx-|TY-_t>X1%^J zh26bh6gOv^&nUl>bGn~Zl zzyEDo;3U*XXnFeb=&=9wVcXw512JwED)02ToJITtagNo0Ks!t|P1P5RK;pt~BiZ-$ zPCadQUR_`A>2_Vbaer8U4pYbda~6g}SmXJU*k(95gj?~XQtCwMM@2`i5uA^z*gR1#hNsY6^Vqf239lH`A5e5(~3&@ zZ>^}#PS5-5jnmzrp<+WVD$YB7R5cOT@02lDdx;co^2fn{g*wa3c**qZZFKZd{|ZxN zyYUqlz{k~(BIT9@QMCE#In}#9vfUs;zx6@gI&m{})nsBL7=|ncO?$Kf1^d{3Q9i8Wg6HH5BuK zmnKjuDgJ-(QsHDFxtZlZy2!47D9|qGziUYP&HrpX@pt4OKmOhzlCKmuyVhH1*gbQm z27eApLQB2r#eX`d{_i>%0N3#UhUWhb&Ho#k|Njx1OA=duhJm|n@`#rRyrnYhI>E1J zGcf3Xx*iwZvlvDH`yG&hSucAm?HHG8QGM%$kf#cvw;`ze6BAMyT2A!;HD7o@d~4`b z;`sT}FE@g|CH8{$>z7{DgWE~|k_Q1!`cbdvbC09+vMTup(@9E=&b#|oR_8}b;Px}2 z#Z(u+>V2hhTs5D=4O$l0ZP=btxw&D><4y#B$@Ok13j)Yw)v6?FcDl6Y?={hwXgpy} z+m2z+ZNK+d#FgJ}IeQsgx1E%3l{$I1)$0c$nBNIMtOAsOBr`%0G9ETLZpN;;7GN~& z(*n787^R=q!Bbz8RNYDQVdCn~vo(Xtc96hr3=idTp9emDRC zC^0(p?t;pcfm)jSeBE}bW64yizxX@HTX--gCmtJp%#TOcV?K#$0?>NdQSv!3WG0SO zqEzk_vecyva@YO!`Lyczym2#O(ebc5&`xP+(*2i6MUzUU__ z*aFc+KDt})!}W*KuUR2Av?X%b3iH3dMrYj$4G3-lj$=xN_tR^u&sXYlKfihvX?4GL zRPZ)plDT7_td-K;T`behk2q9+GA)>@v|lJh`y`&sS5)Zl7>jRZ3U!4bNWM2%A8}7$ zUp428gB>tv$dVG^lU~wka;Ai1GFZxD?(MMLVonb!8Jqr90C2V06i#D)rMW*U({@u_ zz1Q7;+KquP34uQuxTx{Q3=Ba01F~8FxWC?>|L*=$>{vh!X+ag&jn>zn z(1&t3m@x|ackWX3eZ@F7-1%tgNeV1NyE!&rDA<{()f-sXwJD@NTZZyDJ@F97Dq-^2 zP&Up0`M#l*4@PDQvX}{wZg=NnVsvJreWShB#&WoMYt1nOFtLn1Dp&ckcXHb<&VwJ_*k*4BRRqgf`r+@;KJOD18Ak!I^gr@WOt;WPJ75b2!%RmY|F)1w ze@AvNoF(`$VVle{gAE?ZkD1!Lw1v0A7s*zkcI4VTT~_W$*@6s=i#i4Tel zH?cr5o|%lmUxELBHSvO4T@IsA?DiY$p5e0|3yQuBy+1q{^NCRj{C+b0OV8C1rADv8 zUNalR)Pukaky~(D&wDVojQccGN0P~3YkYHTmNQ+uKN%KF#-TEfGrxo8eEqZVW$dq| z&bdLkNstHY0T51@J@ImZK$(7LLUwk4N0^*~!ZyW|)*E$ydJZ3VV5R$r^)|W;EVBF; zi?qXOj?KI%T#i{?Q)#8+8L}B>tBFyR52t1VJ>)zh{;!$}d9TjE-vZOYc z_0jhk&yAGn!#9iPkb?c>A6&(9pcFymKF3KF0N|;ew1Q9LMm!$^5(6m37eK;g&ym9p z^8xDLljy&H|IIkcir{52)&2=;oaZdR|A2LkzR1B4)C(VyB)oqPh4pcj%~c=SYkLAf zD?hSj$wH(#Y2orqfssuQ)tE+4MD#V(w~zH`k78hl%iP&S$|+(8F$BiU(T{HW3*d6T z;&0@;;k)DA4M4hKllnz+_HROdX151Ex5=o;K_C!+h+j=Wt<+l3abY&gOyMp$VoevULcp z4YZQyb&&GA{L-FGS%9-{BD()oynkiRwKtNi?FcKzK%9_ecJ=Y&tX`kfkY)xS3or}& zpy-N!|0TL4Utc`%?gmqtZU%p-y!ap)fR=PlwzAHD8Edlj%&uYr@lyw5&PSjun0#eX z)|T$W`?wHhW!UZ7KZyvfe+h)T@>VJac6;W$_~^>zKTr@9{B5%%5CEL!*f*}wT2na8 z@#BBJ+KmZ1ZFRjYf@Nrny`a{O4@ak=jd6K@BP0-z!VQYJj1l0MIm~h+*#E#@YBgF3 z&Dync3MW2seumDrB>Bu2$tzBg_s2njBj72sO7M+osr`09<>dCj%HiVdp}x0Nnwa&5 z*XdlMg#hq~wY}OF)yMlF5)NFsPLlY~P}X(tS-I>sK!%yXQ;_S~k{n?czlhnSukZ(d zETy@>iT_e+x7_mdCtS0-SZfYN!1v9Pk7zLkF0!fS8u(IwZ9>v@^ZAxmREprK3D|vq zd+4lt>Q_H{eTbRH_J3dt(7({(D*f@ZV6&23g-6cXY_3k#c>>!RlUhIJ8lU7>FEIhznSgF)Hm7V zu$u7p@QH%r0bF=|uO*T63MsW&B;$nm@9!bdtkhTjf)~4vJ*Y|x*5{JV@!$n<-w7t^pc$_J#HGdM-NB&Ese8kqQhtHRZsH-;h18;#!wt+o9YLiCyPyHg9X=SCnF^*gJAfZ1s4)I{J4cEc3WSi5&EWg!QTTl zm)h@@BPld=17$i)7*yDQGO`iTP=1ZJ>*Ai5uZYL1?=PXYsF#gDW8rTOCo^Ox!c7UT zi0V7X>2_M(FWVoA@Eng#Q${*kt(Tc=qgv3J;*7~nLOgCUhx3njZ~Oy8ElwM<0AaDW zm91~wR5Tb=&XBr3cLGuKlEaVr4-#@X^CdKu_9+?4)Ltu$_BWP)x1W0Fij`%)+I#5_ z@^i|E2J^lSOgBO?i!RV|C>2TX?{p~aM$FRhG@tXhhFxBO66uY*Tj!h%tFpTzZmb&JNn96c% z8BITVV(k36N||qycW$xdf-DvjD=IF-Koj{Yjit{r@n}$s!dsM@i3Gcah5a4?lKsYeZ*t) zvh!U$xSvy76foGhkxr`m#7N=0KX2n$dE1L9pe%pwO|LY6SZm}P-0X}t)2T!@`1TP_ z%C>9GUWh)|{VB@wRBHL0ueUks{ib)9Rh`LqDvt#^WZ|1s2QhAaMYr5sz=PGDXCC1a zeqibw-Cjc3zW3GFUz+V&l$wK7o4S6zcft+B65y~01-O-+8}(4^gG%jD+|QDPK>({aLQY(GS*vEaYa)3NIQd2{>GW5$7GhBC5LyQ z5-(kbVB8GdIO`jp1+=FYaG83k;2VR(y62u0Klqih>1BiiR;(b+-@8Q|30^-u$CR8= zCbhGKFEz!UXi4-ipsnS8l%ey7!x0&=9YKLYKW8F;pV!y!lJ9d~y1vV%$`V0hPKvGU zbdouzGz1fyi47i|b~KacV$kOhy9$@t>zT^~6UYRpo1?3C?KHrdS>@0)I&F+!YP;Er zujY~4IgqwC3n6Pkw9XzB4FW*AI^l=6?btGm09HU zE8+-$Yq1}djUPE2v0H^^YxdN;-}Qcqv)MGs4L+&9il@%|RxfIucI0sC@BDV*P8j?g z`K^cscn=I8Jq+T>gH}EUwdPhMgS6gnoe?d9gLvYNzZNC%PtU*n)G5!N_jBPlT+NuE z7;=z4)U&m)TMU^(|w9IJQzLsM5z1iA=iz3_6D zHk>A2_MA9;6?owFAzPpnMBLcybIFoQg7+B)!fWr9mab9bzcc?SbOisCwdMaGFA%ts zdY&frf3mpjc%3VCQs8o!Mb-bXo^|Hic+^CR54qP0a(Sh*-;SmBm?C98dFRBaapTut zZb8u2V6$~)e;@MGkdI3R1hUvnXxrrtCk(XE>rs-ijr{(0hZ11G<~XNO3UIp#_&k?j zOn8^cDy+%xvSO{(YzZ$^?TnJhyg!nZ7iNTVe_^)fq=B{BRew`sOvt?~xotkSOB&2v zfWvhfOM9JzX;97RyTjVEQ;|^3IaO{UazctmZ3}xc;XT05758l{^VEbaXhPe(f9A>2 z0YccK6=A(7E`w$JZc$evMu0tXB$@8K4c*Q6Y6}^MeWzxr$9YZ80i?ixTLf~Jj;of- ze{R@G_j^Cd8@{m6LX(t?A)XVnre|A%8Rl4cy#il%Me}RYqxo}+o@dxw>gjdgu)=KN z@T~6TZb(KgybbIlgI;Z>@7q&~39n*IHebsWsBmluaUgEd07oLmIY`32EgujCO_4l+ zX~(!AL=oFi6+1bdZ)qT@YG?l&+|6=Hf6nX{G3j0a2L?h`5O~4;xt0>-HD++;=FT_6 zuEo|%BH(dJPC`xa=UwybEFqz9B!>?U?HYx4YlXp?xxWA+1Q!Z)=^QF0H;dxBHVYyQ zI&&QpT-g<<;u1#@E?VzB(k*gWv;L}l)zuA7UNc5imBvEVs%fqJoT)Nc1h!h8e>f?+ z&=53a(rSsDtY~Rf9c}!HB>GGP0sE;nsOHl<`$7CmQb$e{0(U{Gkq}Jw(lM`5s8(k9 za)*=9yL0@n1j&sd{4+V}0$7{H^DMD6Au+Pbi6S13_V2C*(;#s8;VHkq>?5{SRaAY! z7W0pJ)8%_xvbS8ytu#{jU?2c{f3sR$M2_!USURf}9XzFRm?W zcogX>*uA#6UPubjkVVec=4^6Z*zVKfljC0KAsaCawoaCdiiw;P8L z+}+*X9fG?%Hw1Tg`yhGVz1Ld%oT~j)eZNkfs;l^c+_PrSzDAGH-LpwV=-rtqV%%SV z9Ne~}>ub3Efl1pIN)$C#e<2;PGS{N)oKqSZ{qoUOkFdl(K?`iVD7Z38zq@hLVLFU8 z*$T2fu$h~t58vaE&wUzp!+Lkoyj7h6pidSU!N+~PMno+ z07cq9MhK)NzKLZsT1);z=K{EG=2G?NAiDZ1d^Z0|q|+}JwMo3ze}U$jif))b++16- z?;Y)%I85qurA!ukADTKV(xD$qymBJ-?RVSbPR#Fgs$VeeyR#zTknDf-F~D-oH<{CR zcZbIu0G(;!aTz*m03uqvV$zbyB&;uN8U6P{7@C5k0fe4No1nD!2(5Iqc+`Cw(;5=i z%Ch~kq_l0?PB*;df4Ht#R)GGGfhpu(dsPOmTQ4HeMvld93Zj4 zMZ`jm4OZr*4+ie7_yafj*2wm2u^~WtkzZ*xDud9>86;~Br6OuRU(s|Ku zf)KFLmcQo%du+px7ERAGdD^iVBGH@KVzD_sl5V)CG3zsEe|k&?e46PbH%qyFtbc+_ zt<-s3*xbx5pHjt76}X))Glw*rZf@GIKV>#fsUu`1l;?FN{V_JQi06KbOBEV8*L&bF ztEfDc^~|}|)cze$e4Wi3=VEOc5?5LDGog^FmlRx!gQ+w;HfzN6w$k5u4Fs1?`KPNT z*xY0WY0r-_e+lq*+b)DTZ{$LJj*jKn(wf|YCa;B4b~F6q z2&_cd`+efs$6zALe24JhcB+Hs)HP>XRknm`qn4zle-OvqhBbh)BK1wks8$+BJ@Z&h z)b~?$Nw=^}OXbl^g+ASDwj5zzQTttPO8g9ld(mR8vY&c}Pibc}3dszX`K`CV5~JSx zl=>VzqTb7k)PEC;F(y}w|2;WM(ozPiL-)z8(v|Z;>WIQGA!{UwVGy^$>>O7Mb*^0F z3-=cte`NG@CA+Nwnz$(ZpSYANTZ&AZ`@s&*c`ye|Hz5Im^|;NGN?8tyV0m0%SwO&J z8qX(AkBv!i|FmwpM%xpQE^f`UH|2xtxWA-+au7%eioPI~$|VH!xWxYw=a1_a5&m6` zKyiMCc51eCRptokQ9{eN`g~5^ZsM9%L*y0te|Dx^dF2<;TI=F;uADM=#Vh~b!BU&E zh3y9j?Zq0yJaPZ?5z#7rX)Fgx4>PLK^yg5GMj0`(vF=!*OPSY;U`@=aNlUlxlqxW=Pbb06s7uh3pl#W ze}JwtLo5>kl~rq_spSZETx;R8pL3{I8)Pr}#TnkKv`z>v!B3jTbQxIt|}ntsol)sMtToovTdMe+%2vEQ>+aybmjuCx%u4uBQv`=qP9AAz(U5 z6W)H0;JCMvrla$bUwzD7-ReF~A2b(z?0l4q@zzuuQCmtw=gd~&DDY9x1>5iTtI&fx z7K?)nL%T~**|KAJ9PAJyS;C9!wW6I!hn0#-ffKwNKc;y8%y8@rwG-OVNO+pte*kz@ zNkzNb6{@O#*0l80bfg*V*=J_UA(zeW#23#{EN%_V)zM5wi09`LC$B&{hJ+5km#>z?Mg-<*;R zKjpS(y{gm;l{5G*rUX_wWN)o-f3KH)!YBKkj;3PjEtj#x5%s51tlMR;u7MNe8;UT` zOmJ(TXF8N=3W>6oiN3oXTc^SVsTgu2$h@NDQ4Y1Yd)&8}oRk*z&8T2@J7<~fXEzC- zl2mk@ldR6K1CmM&Lmu~|x}Il<(;=k(hAWRjkAq2bZzhG#+u)beW$O+Mf0)yVyeD7& zZ)0jCLus=DUAUB`8wAYlqsS5A!Rm=V^iWP|uV>E@b{NeJeLa|5`VWHpin6q7MEw$) zvJhXdPP-c=$x8`hp56%m&+}Usi>Ka1SU;C-EQML(ubUw+Cskw>w7Gs2@Qd;l8KTIA zL#Njz5nlm6PiErLt{&Dsf1e3=;Ouv3B76HaHXsNSyzaLn02|8U;h((l|ucE~TJKWT0*fvP|uJ@-ih*&Fv@_EpN7w1Fnr;PPJf^zY8 zyP$vx?Wts~&xA3*Nh%%4{dE!HJ_FN%5jqoJzOj~Wfe%Aq{9Zoye{v*M0xVy_XvL2? zoC=;H5eEASGu|gspq!2e$P%|66*p%AySWCoMn$bJ=|}*TLY(XceuQx8j}jxAR{`83 zUX|=1u*$kn!rCAGby2=2^w76!IL@M7xSeRQn|9yzZ&y7OJc#6Czi#Qzv?@Z96L%7z z+h<1vh_Gtz_>s=Jf7TSVj3G`q*P5-8mrLzIT7OCpmZ1s2x;UjV@C%Y7z+P8<#!lnf zhn|!&9v&qwUOD3kwng!G;<6rd*ciO?Fe4m-SOh)2c)(^DzCnn9`gz8Uh_@h;*7vN} zK6~||RCQ0dB|8GS;R>}T7lF@<{0PD6WFp|{?i}H+#fwE9f68ezo`m0$+ZXw!`S{{R zu{_t)Mv;J-_D|ExiZiwfyW2+zY0uJP{UItsy*}yqh3JP8!p9=AwDO$L#29{%1HVYW9#aRW+N+eSR_2|8NYo4hacU$|1`oMQf1Yo z$u{TVY9UATl5IvgYONvp>T@&I-mFvFLtmvEe~!yWiKjKODwSZOH_uuK#j2)y_-^9u z>FV>(-5q3F9*>B(`1+#tpHlAL|Iya3j**{<+41{=(q!saRh12~j2j{@y;?IxZ;#AS zCsLtZ@5NNh510aQU4!g1vz$+fDr^j{Rh|6cPV_zmE}@b_J^{@!orwG_B<3o= zj%RC1Y+5wP1IhWJ$~9Z~+(3G2RBD$Rbrl)-WH5AeQ@typ7Lq(F=onhE^U{1C+@$#>}h5iprOsH;|i8lP|Xq~vFVruzJWl8i<;{4=V0sqLjg zv>H&xtRA*jSpsFkC~M|3K`Vm}h^lsI&m-2=8Pt9giSn3wue*k+!WkQ|r>52Re0JV8)Q~TGMA&Ug)ByhAsxvt-$<#4TM)yGh;BBRK<3%gA`0$K9X+^}bJ%Lt zw>bH2(c;?i!6guyk%7ut~gw)L(*Rk zGvt)c`%=0;A#dI-CbHRPM@xLzV1LaX6I5wfSD5FIa9~cS9)o*m zWFvRk9q!YWc{ug+$h&*em?hun z=6o8k22Ot>0W!5;3GFsP%Lx_W9|%ank9pUz3cFY9awYsvq498sV=1s0lYP4mP3SY+ zj9!S6z+0Qbajez`e>SN6E^#25>1Mz|cB6XmqZ}3XrC>fL>*~?`Kw2ZP2vj*6fpg@% zDQR%^u&E3{uyV!wHegAV3Viv zz-fY(NQ;A4U;B}$y zvs?GF1<1b7wYsiJhvlx7n;>SP?m5Eq(1BuzA={b3hzHFcknCXr|5qT9i@!HL-Vy`c z+79<6RaYdGzbh$AC4mCv)b@CSvX;neV3X~uk3b$h=uoZU<&ps&vCK_!w$+e1W;gEb z-Wog26K<=$f8*gAtJ14+MiLflMgrDPuf$s&kB_<Sq=LH_L$tk?+y zsqT8*9H##a-I|`&@eLogo2^dm3WP4UKTtGu60?Nfh|~1;WZoYm_2z=Xz+)ig`d}{D zwN@AkcffSoKV2pRk`HLaU{gK2^W08t%zb6=5*OeMht#^$g{g3>5q@4-v#{dpJZEMle>c^7(BgYFjE5mC2_C+uRO?N!ldOf5F@vvmiiVAbeO?psd~_%< z|CLXK^e|3M^l9gf_5EToNPOUEvu56B_N~{9@d&c>pWK`&RMZZv3*x$1N>|qN3a%>RyPIo)n4&R(Q+dj_jde@(S^yi_N0B+(WSPn%->NlV8;MwW`H^CRjf zg^JcI-!Wml!x%1!pBs6Df;MgN+toxdKB!I!3gv3*CujXz{I?w0={>ZcL~wh@zCpPY z#2_bxqkZa^NiGV2xR1OG5_OC-yX|uVHbEP!jFf&E0;g41@$c7bk3nJq-5~x(f6}Fx z7ZeTilQ+`TM*u@73-beQOpA2w_yP0b@PwmGB0hU9Wm46r3n=0%s9^{g>mBMweboaAk<|^iIMSHLB6Wt&da0gy0$7wlK?~Sq&b`BZnuK-}y0U z)%p%g-v?4f#OW`X!iZmo72-YVxcqepM+9Xwh;SbW$&M0e(v8J-T#w(|e@2G59rcgR zw=azKVWS`XWE$7h8;TZsJs6x8YoVgk-=j{|kPFTTis#2QS}44=RIS&m0A!b>&fn}# z-5kyjzwaT!@V=16RxLEY)>+zZV9oPAI2+165K1Oi30lX@t_`9c))wPLs;Yb;|9$m^$}5JE$0V@Lo^Un(sAzsX7iwns#OY{$ z=b9yMsoGZyKNdr*cUOs;!EHjIrIdPy#7q_n&owo@0i8;Nhy`>ce+G!Qlsnzx+YJVr zzeyqyUqHe`$K&zMg&4GJ;qIPUJZ3b`2GOuc>yoD_GNkvQtfDjm%Wj>A%;Yr-8zXrr z_Lro;pcOD|&0|+{gSnwq1qsi>Vws|BA46D=THQIefzHb6QwrG34jOlx#u=&$rS-=% zhrlLZDYulMgfOA;f3aZ3SUFQdRHYIv>2TLQ#0=CxUh=$5*sS%62Eze}GrYXmQ@58h zil4QSO~{{bur`nfMPvMBzp&hRS7HXU{|u(XYp{sf5+%<#JNiIOiisXT_>A3KWA6`1 z?GYl?O#k3QLVQe+G*Ylf{%qb+2BZN!8?c z(gBgg;qDTOEd>`Smg|zpjFYl?L94S)mCK6|+FyeUZY@7K!Vg+gpQ~$LgBi3IVl5pd zR>Rv40*oz2M=k5%+aw{AdiM5n4!{(6b)rQq>p?IBl~Aj>9hm9vRO|F~Ek0(w;6;({ zB%WcG4}4iJe{t99D7N|@v$V;4VKSjIFdYQ!%sP|VJ?TPUq_Ewf`8*4z7%wzADF>2RbPtf5N>M!E?2ra`!ONXkPSFVyEu7T zxBMzye~p(*us7!{ui9sG=b@IIBrk9;;BxvW9V7DDY43d2Tlp-n?5$EI#JXOwK!n#d zg#PLA03iM!!*%As=~%5HUB2TtA}3(>@smtZcOm;J1M~8ydSakRFc$-cO<8@j{FZ{v ze&@ih*vkT+ajH4+>sqf4KDGj|$VowDK1MbAj=#47@vSo(SYv zRjqDKJ?myZZQf1j=VDE|)2%LvWhD7n@Ks#-7e7+hY{3bA*C3*=P88?xP=Id>r6Pep z&N ze-k}^4@3G5USY}gkyb5Pl{|Jf${Gxs-T@*RlGsZ9tMk$htvWh7OIVl+>+|;yxR7a; z*L#Z@T1bJO3qsX~m&t5ByvBHRNd7+fP+7$5dTXNf9a%G07|p*S%6%c>_l`@Wxq?ts zEY5$p!ZVUwG+{dY_QgGYpW3>*-8}(kfYwxb4Io~1VU+Lp+ z!`f}t_L<0m_ZMOuH2RI7h9us~InLKJ@G&X1W{bS5vw%aDl^!B(l_?DDXZ}w*8)zG|l>|@>f7HIY zyDxq4=DZ_F$)FJ52;^08JyPz3tA3ox+=k0#!EGP6w^j5}>*#mYxfK1h+h~?5I~e50 z+y};q?gu^fgno5Pxq^v{Q zqxenv^VRoA9ecVr^pTg&*qd@2e;xc}h}G>NOPIz^r%0jcxN<41fjUoOLMVB>gXd@` zD0?|cgWYdL{P|v%VI&8Qu{(b`{Vx#v8CK#9Vndit!2%RflHbXDk~kDU&EK{2v7d+T zvuIK6S+>k=d#;B>)~{yX8D%$}fRzrDE{n|-;q&gq$^&U>>7GD1v7IW?e>YH#fMpEO zh2q*xhPohVZ;Fk!lHfMqF|dQ3gna32sft`CluzV3IAWq<&6FHTAY7fY@kwr&V(pNBS#?h^0yBrk7r zP@8+ddr6ea2A9?y`4Wase{L59CDPb=$F^gXnBzmCKqhYgD-a!3w>*d)e_&>9mJuo| zmilmXkfpcq@s3)hF!svH2Ry8|J5#biC?r8#+PraRuAunJ`UleVnI5b!DHRna#ngYz zEybR9=UbH6oIP9@C-Bq@rKOZG2wGyhrQD?S5J+=-D*Qv6M_Dq1f7Jb0Rb9}Wh+RIE z60^|h*n__TW(#+Twz6YS-3(rZ?9Kr*_Q26OLY3|0u#3EFOYcXi2%dPVb+e*TY&-EU?Tbuz=olCp%#lwSuoY2iN7HmZ;IsF>=w74}q zRgs3Syr1^(Kx z8Iz;P#*}}(PEZ zzrrL`PJdV5CBt^f5W97gonfv|IYq%kKsbo$x+?yLaYI`(p;een-lq2ZT{+hO%)JGI z{vQQ(M-Vtpg(~@3!DetqI1`=k`a~``KGir|b(;%Ur2xymQr9ZBV>-}I{xqrP_2a8g zc?CV^&!fvIe@^Y4KQ_*M(#Dp3#{W)a|H_T1|Iu^4!YT~mb{filTS&d(IFseLdFTtt zN7BZ?i-zvyO7JvR0qCm&s0YyD#jJ`9`-lBPcJGH6e^9eS0gs!;YU~5 zP>0Cze=`j~cAwBuqKuAErIFW&2lizJUwJB#B6^8gt2LEVt}7Sk_cv7|ahZzotYUQh zS&M;mMYN`izX5IrHXQ~?CFkI=B!EFWm>$`mQ_~#_{Ok*Pm{T{V;~KljT%VuluB zs`JUL<(%FSt$rK!r%snVq?cEYZE{UhaP*CmXb0`hJ0}Po(zV&o&D?}dpZ-HK2j>r0 zf5dNQ4vpo&+|CtmpO#w91=7YSa35?H*dUlEX_NKt)&pj%R+sj{6}( zK`6>tQTkpeTk6nVb5m$e05hNa`$lQy=B+%xsQZ`s>&C(4pfk~SICNq=3gB9grvz!K zw5*&W?}tGJ@?usvSp)f@G>g7lUjppEe|etc-xIyY%0vdp^eWdGrk<{}FHRSFZWUHl zoq2KATmHPvcUxcA#d<*USm8(holdk4?nfSJFCk~z3b?kOF%K^a5~afeE#cDHQ!f>a z?n)iJj27Fyg>}uTT@dug-J=@_&@Gn|n`POze?)K18IY{sjV$Kwj>OvNl@$I)!sQ!!$g-Rfu8U+n z;^yX(@4IWr`aTPY>4&c_BV6h#)Y_hI=qWmR=P~=UTOxo}jMq$@rhawJfhrphGR<8m z4o#uMS^)A>?bsijDm+DhvM8J57{?ImIvf64C4$oC+m zm`jxNL{T%&49c(zh%`tYg#RZ{@jWP6a!%hIw`{4cq_5JFs&`DxXSSvad1IgwcS5DQ z|F3I9-v47NLiKX*Bq$$ZxtR$e1xOp%D~; zlXs2HGTA_c^LfB@G0gl;e`-s493`I-aXY|p(RM`xBj`3bp3hY|+;6DTI2_ezF3Nz7 zUvCoSWlJ_0byrNrWn^{iZepYhE$-dF@W%MaPyXJ{?Z)h0j25!jyKlyo_>_f$6pjc} z))0OYQj*gRLkLALYD>J@IS5;T91%_kRZXy)TVOx0RqHxDe6h9@e`day4NDu@Z}nOa z?Srf(_f4gBvIn>aO6BCE&)SIX>GJeNv29n&by>1e0^Wyr!2h`IHbNGDV-XaLuTA7O zR|vtiYg_xUUUQ3|2fZ$yCiyV>pki3I{T-^vitsM+IDe4qITjGtA@}T<8?@j*PwCPjm+OTO*lzgW zNQLB~$h|_+)%lqa6$u3)q_L&mGJHAi<0<^vZ|~FbE%@(klpWf2=8W${1R{mK0yJ&h z%bkH;Qv(pUX(0srZ5`*`OJ(f`&FvqmJr9NXp~#TdK5{Ypf6gH2V&->!8^zKpknRYI z=@TP_ObjLT^-$LBLur$=S~|no9&!z#D%j6x_gl5bDA;$^kbgrazIKob9*A^8WHquq z&gm+9?4W74kr_0@v#a2=LM?@Chp+FqYIAMt2gSuE&cR*xPI(4eIvI5hrP}}`LIf@k zQ7>x}`oaIee-iZ4$<>7KS>y+4Xu#DrhW6Axqw)62rsW2j^>lv1-i!$TV1F{R<#*yB zUL1Xp9#Kpg-_)NAD%nVoN25O^|1%4K?-B86t3s`519 zipo78A0Q-^VKU&9xhPH;ZE8|i?XG{5t0{fB#ytN)tv<{Y%WqgLoi%&Iv^=M! za@F>pRQ5wRgD5Q;GLGUQ=bs;V_MX3uSr003`Lp+WK;j{$rtif%+i>kNkujgERoTI3 zokAwPDbjk^3hQA7mgl7v*YT;sNzt|BEU!}Ke+KkfB-l=8bjr^Q3nIJ59COR1MUpA$ zlgr68tc&Ra#W$F>8pWl2eU2%|(*1Nls6HjMOG$pdHHfB>kJElteDh`o{VJ*GHvkv|6Y(%0NLEvIK(m7a;q$e@`|H0v<1Y6MSM)&ay9Ih+On{Uv6+)OjesE zuXa+brw8Q=av^ckA^=BBHvDXDi$tY(P+8>{TVr{otav5RVbh)a^V(i{t@o!)Jg8el za3OYU$;9yrl5$U#V!=K!_K*&KeTY*VJ=VW25bWvn7Id0BRC5-H0iYICS#00Oe<7qe zoldgi%Qz2u;XOhdS08U$%_K6EfRSNNJ}{k*<|#Ltduh(Vh!VtB#!u8riR;mLV!RD! ze71a-o%Couf4N(TM7w(E%%OTvbdrA$*|MFuT!0a71TLaS`bpw;TaOs8R=mLS>yvJMx^o`)(Uyj!9Mam5RTBW z8YJDn_xmK(bCG*mPVr(mGbHzIMorgqhn=^FfkO7il8){(w{Cx6f5Gj1>M?^LM9*q7 zK+738goVPkdmi{@kme7q(J;ixv-B33U@5zRjaic>U1!c;H2X-m!*?evQ&${kx)gg} zOi|u3LClu{W;b+Gj>vB1GAk)BWuyb>MPH5q;4r;6s(J} z9kM~0$CIi+f_EMFMs6b8c_khyA*58RSE%42(h}FXs2?>e08CO&p#&JqR@NjV67y4T zC)*_N*{P%hv*sPv^uT*7ztS2o`nT25yFD^GMWpfT5Wzb#fAxjU27c=9B$82*9VGi& zt#6@C(UCXgzTVCtt-_;rMx>D2h(X=!#F%*!Xz-PJ=IcI+{KzzH6 zpk*OaM$#!2Lh1*sR;z;0Am!KBc&b$K-Vls1?#~Xph!Qtvxtw)GyCq zDrpV#Uo`T>gF7qI9UtKtbp3W|e-%lSd)~bq+D7DGf3q6oXcofg5CyG%k3dCd&fKZnjVs9h^-#tfE#h-KzW7MNZU4 z%KX;|f3N3i1B^xfl%GeYwB*#$RKe(kLLo|Wxt=l9!Aho4GLz-2pUc#a)uG5+2BhvX z#$8`_Ih@m*80}BF$L%u;cnXxqN7C5s<8M>h4{tp>C=+PKMNyMR38G|vh*q%kaNCDp z9mLcT>}!Gzo?eFy<^YjfhA7CV?NyFCDzz-Je^`zDZ;VLhE%PH2qfAs8BWu>L`d~Uq*qRMK8STY~{{DD*+jn(-z}OSFU}1hHZ*cTzHt_a<3iu|lf8;ue9$F2Cm!2Ik-z6qx!%mh_f4n>@ z!U(6dz2oIgz$c7Y)<;*2<2rI>K2R*DR5t6W*1V;ayXMT})y9*!g<};JWadS{!y-$O zChPE4O@?<2G%Y$G3hCN?VWu^uStVTrw`ZRO^J|l+`SUdoB;ka65W-AI+nT#>$1y{S zRmb27s(K-Kt zW1mA^z78y#E32PxP=+~R6yA;dY~2qbnpJC9P1}$6QS)xLcRzmBV8>RXe_7Jqzivp8 zmX|V7-_Os^e~#5@W*4xOb_M@J2!gUu0^JyWX@e$}uMGj7%J2-q|Jzw{LL&e*dc~q4 z$Ei~TijA)kgtImjBB2M7S0d$zLixm^?oLHmiYx28geIdCM^+D9LJX)BW-GI-T98g4 zc)3CFZ?zDKCOhN(!KKX*f1HZZ_Koy1-zmy30_} zYoDMW+DWR^j|Ev?syp0Hj<37hfNe_a0J{Hm|J)+pg2 zKr<8|+Tr;mlhmNn0r0bs*`Zb{*P^7u($}EQi_)1QR#mefsVnR+{D_gW0lvY*vyJQ{ zoIp2UB<`%*OeKyT+JFG%j($7Z@B;+JsZdV_W9w*DLX+D_GWwoJV|P}OM`PYmcA2qf zWsBHiqrKOW6^6<5e-nj+f2pF8Qst#n3d6MTDSo+Tiqh2O4b^lh$x)%m(FynU)kSl= zNDJd(QNI9w-Q{2WkW|?TLuWp^SZjPiTkZQs;Icc0{|jBMIv^{wiB+Arf0&k9$uk_a zS_tD0Zd3sk6VL1yx-IH|J!sGwa;$Rn(IlV^CBziuOP>c*f4@vYs@H$_3slW-ZRIEG zGq!3-(nU2S(cDQiL0MXfSZeR|nrLyJy_I0`BEi3*|AQb((#?C63-{+YVaH6%yY=f; z&U&*a4H#r+yrwSaLUO8>AK0F+rt=m=Ne6Vza6lnz{9SHs)~Z9A@hv}Sa5%Ohe0vtn zf1w*+|6zR*f5y5N?TF1XL^5lwR%cZ9bhCU)Pg`V?P<=Vqq|dqUBwnE65-gR^L$3(| z-!1~Y2_N?zYGvher|137Sb58SYRz~adYCd_@uN+?g7+z0Eq-I(inf*H0Pi!qlyEoz za}A7SbByB<&J%BYyvxiZ5PZ=<-i!3oS#B#FaJJgJETskCix+jYoP4ycLDXMUO{uHnIdPiOAf|_2~X0zIG z-FJA)Nv2KHze6(wF@;|zk5%v2tlCXeqg!{ZrEk++q*UBUduY{`go(& z5e$Ibe;x(&3AuY-*Rc*XJ7Qn&gg0n;F;Dg7!o-;$yfiBlOvTEYumCRW>eNt6(zH&S zH1VFmQYBnmXjb15XE>$25}0_XQe)+3)uUYD%4{W1z%opw&c;~@e+dhusoB3t(FiAu znk(F%OsOC{ZXSMz8o682O<;hxJjuaorrXR3e`o$Qt&`?j>-NASr@VfJpeJ}s#+Dw! zKnhT>wmW;LpwdoHqr?-=Rsn+bVQo&{zz6;2M(TCs^-; zf55(bgM9)~w!@KNhlX3}({{h@ zq5|W?VXcio)tX&s3;Lc_JA0A+DROh?f1397%YHn}2glseZ)5|4Nz#?=3E9pddb~VA|&mE7~9=ren55zRk~*uk6$MD zFOO96RCKf8F`9$TZ(wnhd$jy2h3nD5(;!Zn$4)Vl6o6xr>~?z$FA9A|XmLsXf7@fM zlyv(jQiy)}N{WL7M{{@9kVj+g9W2g*81JD{I;#$yYxX`|GO4HpsW+R1$I^IK^GI(; zU~S>M4G>-|K zv|J6}eo}^o^f}6VLS20=RxTlue=i74Fj8)i1C2|GgI<&yinUNuw|Yv}k98tVs{fM2 zfK^XtK85MPWcV2{Ri;nTlbWFx>O=QH@|6u~$D->hrNs#e%Fs>oFKnz`B|VT8Ftars zxRa#tS_95G|0`bjWb#NfeddGdsNGa+xzXGXNYI*VI0le1{Y;Z@Z)Wk)f4tl4zS#R! z<(9Qbi(?{Mtd^7BzKv${;Br_XSMg%_pcQsmJQO>q>PrpS5NwK9k zCpiULO()-Qv27@tbYFLAWZnz=a~}P3%K4T-SvBxZI=_8*{EfgYwRtl3bZ^tYkqp}= z6r1Lohx%%xQ4ZWAKjM=(e{pOrgX{sPQu$K`%ehs`$`29w@Z>6IAFin@Bt?h6c8lg~ z1k~m5=`LPJ0J~wP^jb1+x>xM!!y7gw9Pnak6ZGws4T;T?SnehJI#v-+7P7(|PUm{T zaw7OFlZP2(a(0A3C0kXV+%B){s^INygR20TKJXeBH^M0HlUl&3vVw1n z$Sd8(Mq-Qd!wa$ooDJZ?giO#r2Ai4tpO-Y&cgE8QjZ?MeRxY^1GtLTESdaKT$E7<5 z4U-Tb>OD{sehDQ$G@rgeZog^hd1>8^Pq2l9Y%L@C%huH$X%4%kkwNcZ03-vIq8ZK@ zr4w7t2=&4Ml}ajfe?G2WFB3U#*XMZF<3zyN9#s!-?OTH?!cvzx_Qw1CSJR@nyP909lK{GLRg*{Nbl;eEdUMAwe|MXRMWwYvJr0_F57!1) zg^$M~E>fUhzJVrtQ?er}Zx}q($#jxWTN8W%D1(nGvJnoBI&qpojs$p5Csv_u`9Rjx zR!g`T#LkpE0_v|4VV&2QCGgTN{9lsZ2ne-$U4o*RI~y$;N~_a z9#uO#0lk1}c0xJgF=8dh?4=*gAL}56*7zFFG3t7(pG4S&RZDwMQj2$I&sX!1z9KA3 z{{s~Q6ZgcZ1p{rjhDm7$HuBh(#WVvQz_vG}Fo`S%e`ZD@{H_+KjYsY|y@921vP_t| zP|6aMEE57DA<*R~QZ;TD;&Q&70GX;4-RvZ2e(vyw;BqC6i21`Y{sS8ySKm3t%dR$H z@@77Z`lH%HAZ3jxhPGW^>uXm~@e2**<&HrTDf0%2duve-^~DM-?S6kNKgU`n-2l(& z(*u1>fAN}7VC~nz*ulyG)tJ7_9P;V27%Nb#r+<%~`1=s`~M;GxP?V_Xs7s_j> zf3(xTdP8kcQn3g*PB6t!c$zXF)i}* z-%I+%l$FsGt0G&!#N6ZFge6zLQRCyaHg#Bt><-kqj{A)u{A(icp{O^-(DnhB1ScZ)6yre<^XU zVij9(N(y*&c#x#&kNhq^a{<*+A40UgRsFOGmoxJ`cqgsgNiKhFCBuDCtbNqnx@b6X z%Js$~GTVLM5Tk?ZQ4{tLVrW9MHBrG1dStBYKlYzG9LR8GNTh5uS717Z?%PRrq1*VP ze6#>S+tr)0R>)L5(>xvgb{Mu|f4F3G!qX`NM^|5$;aPJ(^;-xWu6#+uMHmiT#heRmVk0Jgk`FSGbM47WBDe+=*l&9=xg zsIE27|0Z6cE7}oS>4KG7U1W<~0gicT*$$~4g?^tfVl*diyaNbwp?$Ta3s+rLz(lu5 zq}(K&ilHx(nTv;;w#S3jRiF<$mvkZC#0alCnxQYl@to zf)G2zb+m@YEdj^yO~w0;4|9(5ED1i(?)R>+tw!d0jd$6whBE`D(DEHYp^-35f}IkY zv&C^wt^4LTdS1nxH*bpNLzPOaOV^11fkhSqMn3#Gbyq6%i25Q;f70p4@urcIVUNbr z!|ZzFz??ymg>LSrltG8sRZ_f+LQ+dxTt~)(JqMf_Y=KdF9Eq<|o(>-TWnZ1t;a&kD z1outteCge(lRDs%5|HGK5rwQzPNBL%m`=iaMCdVJ60dw7lm1}gOfn`E(42L=+pY+@#PU=ff& zmHgFEtp9c}X9mBpCf)NyNVQ~S^ZD_V*T*v}s9!o-w@2h)e@h3F+$ox569T?TIiDxp zkY}W2?+CtMOXh0X5?^7bP4|sYR0CJ1dhPQZilWyF^*VCm4Y`mEvJq|%SsVFNFno1Z zM+)-Af`9kgEY5eW(2Zd0hX%6n$l@>D>N>tNlJO=a8{f0DCNy)C%OKtES0~i~O+d20 z8&n(q^eUH+K`aIbIe!tqz>p-9M<;JW+Ijmv^nH~t?_QeBtUP+?x7O{=?Z@$3f=%#L zN7bI7C(|zP+o2g2o?)=ONDY0hWcSH*>omlw98JzlcQIOn1HzlDEVR7@-q4qs4%BD^ z*D9enAl{V7n2>uIk5k-1-L(gLtu&Puoj5GW@_C}Yg;e8Kb$|ZFqVe+8SM_Bg8Vg{R zz*xoXHz1|P9hqu&ZKx8} z6BaZFfN=VHWM6W?fwAOG2$vjMGfb5(UI6sp)eIv|18o(^g2k27j#t73vdaw-{6lwXX)lSP5@v$UBj*iei~@D1n52Gs)9&*a zY2Fx)>uQ#hp43P8QviW@Ja<9aWT_H%hR&)__`7Ur1%J~RbZvy`PibFykpwANAxSsz z_#<4RB>-D|Ep3ILYp2msFK{^=rw1{4EDE(0m`vS{uUWT;NahbR8g3g^&9&)nN09xV z@EUF>jSNTX-Vo~Z>wrPt=LbUE*`p+BlPnz63U)n_{W|L9AxL!|ri9vhQi;xmld=Zj z?0t_EHh&wHw}zXjU*ceD2H-(eslrw9;R>5d#{7O>yX9IhGs7Ddq9#FWjcV>*K&05d z-^t0L%KmY+GxWm<^a9PF{0YYIfFhnm2j3FaOD(-QC5!%UOh--|bzN&s*}QE~co@ zYgxsBDdL=KqM&)k+7O4674H1e?XfP;?Smajy|uqMcJu8t8C?<_lpiog!>J1!4H!=` zpB<|pMH2HD_Qgnd7dE${zda&WIMfef#{nBe=VcsL4#6JTU7f4RxCs(k%M;)E()KMN zXn(Y!@S~g;!5)MUYsI7Ky!6z1bV;oe%*r9Kp@1dJ^ih20vD2RB@_+%K=8&T$m5%_< zSC?<>oh0Gu>iAU8!mGA~iN}-C;9EZn&YPPnanp*RUu5_r{#f#!~*HcDLwFIu- z-0rBKQh;=t_tw%k?lwS0XB=X4b_9~5CiB9DbCHN#ZK#rE;!Okw{eiphrXg15` znRw(@8| z_fzZ1249k*w((@3ihtYPRLRwE3V${p=3{yvbm$5J-p6WTN#>7tDcqN$ZG1g97z#U< zUTh7A=b5&7@41^qgrG>&l}J9%ItZEy{L5uJc*@~3e3EqoTCp^l+#e&U=ZPmG$&Dk( z$*P|3#s2^xOE!%03qfi+&h?a#@?caP$4~8o(BMPj3R|9ev3|_>bLRG&Ke3a%w4Setj+#HOtX-S#qZ} zca{mN_PnzI)O2U&KtUDRVdAGHnxEdNPc0N)EInjEPllIiG#h>%Mt|AUz}cMghhCat zY!&b(iLf@`CFt0%3HIOk?pbe-n)lONAVgzIJ}0R^B#t;^3w0K$sD2yU!udZ%o97>29|TvAk%uU>YF zHHTpO_@%TS9hdIhcz^9#*N#&K%7_*73)WE9ymWe@pFaW!(_$Teer7K2JvsLW`>r!i zx-K@>qsW#-A_Y<)(cu$$r-{a-yy5?ab@{jumrnf$i3t(M_fF{K?>g2P6vy=J0K?5; zM`Gcx20eX}7gr;$Vr)C|WpSOkgVKJe-QOUs2PPi%V81p73f**@882g7x1&ujfn8em z;A56y)INiyp5bt$OQtl0a@*^+EvfZLY>S~?{W=xFx0+S+2{p~}iz5lyjg1T|JBQp5 zd{c_aN8JXjOn)|>_p{1Bw4Alf=gY|zDU@S>56uiHeFb^#h8?bh=BN7B73OtESft*9 zE~QUX`VsrGP^ueZqpyNK!Vj1zSX?R*rdtio`6_Q)|5?Sx+EvoNZS%|`uWhJA6n3fj zr2Q5I7k*8#C-9m%1_o`Y2;m~b%;(@E7Pp&+1w9t$^MB4rGI{|xF)A=J#LlNn$@6Qc z&uv(57m2ba;VwBNh2m`ZiP!7Ba($sSJZh_qHewJ6GWtPlzUI{^l-E%(J=KinQl(F^ zJA^XoZ2of z4_QhllwymCAPY%DYUi#{@+|B6dbs~9=4uDP(7&s#L33=~tjg;vG?6cfeaKQ>XvW(v z-;%3bu0gGjpH9Z<{E{i@#lMF@GfDIvP-g9r&wu~&%>_9IwU3X^X|q{b4_0`GBb-i~ z)$XdNN}MhWWMfc}=D-=P?wx-&f3mTY?e~sCFE+Q6l=!jrQ#hdrFoh;px*!U=0Rt5g zrnnOnhNKX1v2D;SCwWp<0e8A$#n?kKB+)!VZ-TOP7rfLC*5UhZkTXmhhx!wT6zK6-M&;$2r<2u91wopuQ!B}n+3hUN$ipCgl;|JxX$!LP~h*M|LzxZQ1miHf3aV%Y_MI-+#D{f z4MOuhMK1Q?BkY>Wr@$Tq`h#*aMN0FH9DmIgr@=nSiW2KnWJ?M(K&3i*i;=wT%@&h{ z->Qqgcv0B{Nk6_5rq}%-AIPVxsE1iz#n;2{buj0n3 ze23J7m<;V4H;r1^VFo-HR2$!n`s3z-kYVkvB8n8DTs}XyC8ph^g1ESfPYseOEPo|8 z#;sNj8>F)Ay=?>-Z;);LCIc6DvI#=Lr(0)Blw)F~sIni-tqZHp%^4KKoL^;^dT*#J zD_Bb2LC_(~%NPv#99zEveY7GPy92vZOhQX*pf;oCNip) zmveQMN$A=a|2mBFhkB)0XlcKTfPbKO@f`Oqo(o6uZGOp(XK}AJQO16FL;`EIVX?v00caemE}59 zYQv$_(*9L^a{DtfMi(y|li_&kWa5Abno_T2uZe~ZnWS)$M4R2B`$iVl`hRGKpQ9Nq zuT#&X^!j~=EQpsZMvR(7Af(EMJV(n@!PobG>?uo*w87B&guDw5GMc7)G`Y3%mdo{; zyCBKPgm6ErfT@eqzrpG}eMg-JX=-R5B!6_CoaS;E2iRoCi`>3k zE{N|MPpo;*_#b5-Egg$5aPU#-9{j>x@ecDzRaZOaqll0(>MqZ=oUx1W|4nZGl$lj$_!$btw%cq3ISolCLcr01 z>2kZdM2;V;y}rR%BC^YaU&dhTo(?xw%&St-z;zBa<4qpX1iALSDdNp28% zmmn`S4BzCw~n?N8GNl^5Z43eRc9$Nkj(il2=CfXnbVRohv{zub9N%*$ve$&ep+T znLZkE9VY6&P$j#$qjm@JVn(JrNFMWws#l%SD@A$K$bhg@p6SLF74o1dWGNm^9-|9e z6Fg<1-7xIX>=+8|oQDFszf$LJ!V5;i(;w#c4T1Ye=6}+pUjHBI4i@+(-MwaRz8n}J#a*-4MKuJXGUSC$`4ge z7S?SBrHS%JJqc3+6(WwM6OI(n-dJ;opzJybL2(OF^Xk-~_wVSz=_A8bCpb~S1?$r} zK-f9=#eaEdnAq7N{e`Ktb8K@Q{5vy@@rE&Y%#zW)F9Ns|RcMH%D~N3KaNGcNe4R8g zp|^d+&0=@jn*bpnGORda#3Save=lx2A>QGM%_Hrt}{j_l&Q@4{R2rA`r z7Kc`nYoCCpM#iH4Z7u-Y+JxMh3|@MzI!>Yq#(!MIulr(-cR~<%E4vnh=AoX?lu2l2`=#B@prI)^N92#2kE=|T;w3!B0iKlQr z*P7=RlL|?f3~L~b<&qP0+NBTD-Y;n4nj}l+jrAWq?#n67zx?@+-FI+&`E&4_wa1EV z(tq1yO~$}kHDm9|fbpU!w|D1r(%T2-)=0zyJ8jkg{Z!tAE?rQNDuc9!_qqYYc?+h1 zFm(iplzlBmrXN`9I#B5Tk^!#dG3NY~zmpvGk3L_ED*;r6To_G#av1rwKq;o(b&Vr~ zO6omAb&)8HZ;q6{9(h~~;KP`8&^K47ME-{WDVk5yDtOR`+Mfo~$-n`V* zEKroorFN{qxp_5xHMXWFCJv|Z354{{?;h$bgV#&8esH5_hW~fK^}+Xw^oEDzx$a9@ zNOBs|79?5-#=SGC3?9n-QRw-NN`EN(lP2R~?{?=eOtVOHKBF+?ht@NEY#subrt9l1 zS43ZHt2?>b9kUf=m;B{yUFKE%)e1+__*!5zdVg*k22Sa1?9wGBvY8gIT_2+vc%v5{ z#*xQk&w@q+_0pheSvv?5wn0%7R0RFuO&6|KLWl{qSi8Y zfhP&pI2Y>jgD0o^`hT#f2be7%e-SeM2XUNMg2&nnInp@#!Z+n-5xe+V&|7u&=A9-y z<{uu=#prU2D%jPQX~*nq-n*hM^QnivFr^1iwF|o5l3DLxamS2iuzz;k;VSL484l6H zY1qFZ&oE-G|0+O)_x`LWViP?`{1YwBu>v;qG|ZmB7rP0jxq@+mE(BAk+y#W>Pk|I+ zR*>jaWG=The4tW|9cI^Rvk|Te_~*-PCP7?>BaByxApC{8G-A&_bm^*(v_y&0e=FIj3-f|YWgwGgzec&0UXq9RU=lp-zqDqu_T%9XvqdNBShO6v}yno_b{%9KnJB4n^sA$Vi zLRYHIpx^xuqw09*&+R(JtZVz;V10hYJvR zh(yY}qnhzJ)hg}0Qi@xoU;ais9*~U2qr(ZHQUlmSvs)q=GS_F*i!67#<;6`?-xh>N zEx-BkCHl#rFnt32RXkUIa z^LG&~yvIk+@%@>sqIxWv$3Labx%bkhaU640j5zu{wr{5En39=yU3;^m88nNz8#8NC z^}#L82se&q+$z~6n^Wsft9-&qmE`1m1H?ii^46K8b4yu5GV6n|nwE2&*`Ygx>n=A6 zfe&}+KYvxu&$R!e(~8Rx8y5bw^)fv@v=5AK#=gKTdPuR4AoF5J-P$~^-%u9sZnIHvu8(?wjT&hm_1%!Aw$0T44h+JpiX6MofWDjNq?+1@r7-c!++R`l^qsh`&j@(Ja z1hg9Qn5yol%QLp%XL(qoPU2` z<$`@=@ujq^@5>!P&Qaw@pr}YKVL!!XxXRXo*Xy%9JdHMYLc&A+-%NkITM~Pu3X~{6 zHttHS=%yS?Q2IyF5S%YdyV5ha&rJ;HFSop2lr{0Bh4sxQV%G$q?8D z|M!~z_OS-RhcIMaR61bcU^2z->}-*#OhnCzynEC3PPl2`Sih(@*ae3Vx6qxA^q>gCHh=hMEVKDGLG#-X zJ$vq?%KZc|4deyv4oT%BOsO4cn`rjSQ*UyDx5YtFYy>`#pt|22(E34K#hcAikiuO( zrg3gZ52p+%YXi#aaJy0&MkuDEzG1#?s2l2>2l{P8Ur0JRnCP+B?jt%qd+=hl*;2?SRv8id0BbZ+DJt7Bdj8-z;~+KyT_%S!>s1=a?X4QFk4O*at!Hl6jR!K@7oW~Sq3@7& zItC+PD*kMS3A9uo$98GiL$}OozFgC8FDqzZ#T7R1wKMBtH^Xaa(KwB;#+WB6TJ@nV zjPOz%)SdG`y?<(tng6|4ZFrge`}*Ulv)Crd{7}7R#-R!fI@dQT&SM4k>sW>%9&6pj zs!dqc0+I^44e+2xfuO4d;RNk$uS>)?MVv0FBaO_nAb)|Zv{Y{9l-7E)l(Mwk@%mB{ z!EB}wLmpE-G84c({$_{eKXzznVrXA^x5LWDh{`{9NPq9l9hBcLcMt3*;@JO%nrpEr zzZJ`nw>=L8#6<9}QBW$C#;U}AfU~Ge_ zm!Vi6-@P^R+X=FZ(FRJ>>q}I%XMebV)Je@a&wtcLKK6EedD<1I8h=D9A1EH^5v!~B z8R;MY=JR$mL8-hUst^cVBbCQZ}vFWrHx%A z1VI^4eDi*LH0SoHBy zV|jy3efe7!F*PDR?k8W5W$8|Z@BCx?v7Pv=n15xg5C3WB2SWXKlBuyq@~?~qg7Q7) zujO--(rWPwyycgm>{&kiu|0c=dViqQzcbc9TTB8F|Nouq{|B7wy8=kTYK!wQq&iQu zb^E;(;Njkw{bGQ4>uxhV4po|r=<&3pj?V-`W1@`#;I%(X(<+$baG2H!hqO z>UpzB8joWizL{d(MoFnu9LvJ(UtHP$Li%@0|BpBimijZ@A)Be`;(`sN;Fz;L5=(k~Qj|=yIG3rpR9C~!0)#ZN zD22f>@aRdm;6?zXfPYP4|M~ALtYvk_^e>U+jduUpH?+956fnqp8UzxDAdRy&7}dsY ze><+)^zxA7q2#kR$yX0z$@>4y0{Dp$BL0(l-5;B_qgqT$vqSbV%Ud3&_zLisoY1m* zB>!9U|C4p_&tok(73hR6SD#0a%lzh};$Xi^8y(RA$o6tJnSWP9VBd;d;7)Y`PhA7- za;w&EsU8P`O4$}pJ#PF-w6?9Bnf_Zur29u6#jmBdXMFG64>pw^R@U|+28TtDrNLb` z5)V}Fga+WddoNQ$o}?ek%}eTPe~Jj6p+go3_An8XoGOw ze$-kobEu=Vf`3W@PoMZmq|%Gu-!JU)8QQ-_04B2Hfqfq-`3|7caDc0N$g=!(sZPRPZi$N2h4{C+kXdjONij9zh&}nd;Y+@Thqq$ zzl}b+|5%f%jop(s%6s||--8SmgLZmSnMzyAY?h0bhP>W;`c1k}PG=-jXIKNX{$JnX zduw%HSO%5X%XPw0&Wq+`VOFX}wICJy+rq`O6+^|$6@4e~I7#&X7ETI+N+Ibi)!xIq z+TK%gV1Gg4*m%#BY08Vml99Y8fg3vx3o^EzZyN?K5|jr)z!4*XUay*udEPNSth?`i z+pjx^`RU)%55&KYbS|n_t_6c)xqJLL*h(!)9H=vA*% zmIS?i0R(oz&*U5Veuy03`=nE|E9BpdP5S?qvFYBhHrwy|xZhT0 zua+TXKWWs@Rhf+0>ih^h6v2Cv?JY0WO81v|Mu`N9{!7O?!~bjIds@RJ_7?X&Bzk^{ zoqr#(iT3ta`D5`xwkWx%S8ZyjJP{zXYTHl?6KS1pi?vVJs4z(EFD#jQ5&;P6clVesDBXV($~n zxhvr*1wtA|c+LAK;UG5-o$=*gvi2YRe}5Or&SfWk27_jMe!8TscG^n*SZLEzZ{L3^4bGb>M zNZx#TVk##>GA51pDKm%K`6kJZuk7<%or@vTEc#C+sFxzo`0Kf z_+je#cGk((qkjyXytuLXZ*hz8AKd!V#17z$@EW1|8`}te^w^iXaCJy>>a~uM|`){c7O6L2|`L6 z^~LK?(}yU-{neXZJbMMx8MnW`P|lUhgR$$|Hyr%`$R!DgeBPM4;u_#k^M0crf4lKb zQcO!zQi8)g(f*#kiiZJVkPi5qLHIvizu})&ZOfBr#4u2)LB`=((!UmsxsNYf8rE)k z72849?oieM1T$GqE!?i9+ke?EoJGA+G4rY;g<=`4=G7_k8_uK{#CnS;5WkafcZ0ji ztoM6io8{Yi_`I(dA+?skukpFXiWH*rW30V^-P@R5Ef-M^r)I^M`(6CBIFRSl$oP4! zG+<4f)5CnJDyi2iZ;Zl6N~J47_WfK`{Rpp$8MyY#vBwAJB)PBKiGTVL!K<6l&|7Gr zUa!v)@A`mrH&H)4njueoH!k-+U%wwid!QfRtN1j{vCq_QVI%1NVN(cYz!Y?kb~ODl zG}TLsxygL*!(ziXpVV->3WV2V=QVlHp{$GH@eZ+5E$-YtSH(;42xD}bC}hO&sBF?S z0%wGMTGBKZ{`T+A*?)q1wyu{utSGBWX0bxyY`>w{ZxVC1$?~+=52SMaK`tPmkFvn( z0T)FU%4UNb6tT3V;IMd+=czmMZY&L5&sWpqcFV_}@Ao{z!-}E;+uN*T=E}-PkGBk{ z+AFKdhZ&@Hu-r#oX?{PgRKPqMUEmxsq0|C;NbWZyc()@35P#zB$9NdSBJcLnzv)<0 zJi2#3+QkJuX=u0TMJDtNieEdZ;6^3vtleClPo<}Kv_47h^w&21)u32%|6Q?%wRj%v zqit1mOkvm@{JEKVlW``l@dfMQJga)~ZnXi}t4FK-g0tuh^wjT&;WnKvuW7CI5IU-Q zzFC%@EA`NVL4V^Qa-u#E?ErBes@-SYY~^inXn!LeE-Q7dSNii2Ex8)KpR2xT+Sz)# z?sYxtbzDA*=JbtjOV&-#uu<`psb5ClH_hXXa~1r2dN{N;-R>|?K(MnS309Q)K+3!zWcm(|-nMX}1Soq`dU1rv#OGXh{=H zmj$3(AJ-u51`Q?kny@(^K#II{(-y4L8%Za!`2&zwyX-9)$Vg&k4U)TM2-PhJ6m+DO86l6M{ zZ(0;e<>YN@SF68X=FSX$JrZ|4cnZP?#9qXWMdTIf`oS7*3G}=l;~=~x=io4HMW|uU z+kfvFE=D(10o@Ow!BeX{?hA~^;*{qtCs+gF4k#gU9dw+eU9W~#%_jHMzuu|l4%oU% zkug^g=6K6%xf4rBlR#%8(ixb<%Ur&R+`;Vx<}1muu{7QRS7zGSH^L|Y)UMTuPjwSp zn*;2mM0q%zuRRJ65Th3CeG*87<2StT(torygmv8z%~A)tO!D_#SExzq1e1ny6+3C7-Uj3T36HbDnq#;D;d8hbp&-S(?)YV@_)t& z@?#TM;I2Dc}$$& zS-UuNAwO_V(Z0k`>T}^W{^tJW{gbZx<9bMHby}EYCS5IfLs%)X> z*1>b3@VLzgP8uiLLO;CA*dXbXJ??KtW61^bjjqo>Jq%$=mC^#MEthxI)_*_vLOKLP zzlj=p?6t*5D4ha&61Vko>qY*DTgg&O$R(r%DpitQ#R`hqV#T#boo%~tS>@5RhHpAz zn4gg&6!;nMyWV@kFJ`9g6@eRqriHb>{ETJ*r+ROB_^)6Kb<dtj9^?$cA-NtM8i_3N_ zKP7imGW0#N3;d;)b|eXt0(su@4oX<(`+t-Vnm>)Ebxf~#Mi@Bcb(2g^infLo6q`?P zPE6m>+iM`wC)aY4S0x1hNC?BW$i!g((txt%diso*>rfP3{rd7~OoDt*rX(SzXNN+A zMUAVAd(|3FfEx3qJbzK}xUuCYEpL!t7p&yW_Acu{ZOfO)5y7%y7c6w>Eey~%dnT&N zCqi1g%g-l9x-{7t50)0-jnk?ca)=U$KsTbi{D>ZSS}BN#?pV39=UWxusu`~r-N=C* zUWywhtplNAole5G4G448j5~iCNd=W^l0{*}7Gcm2s#_T+NGKhD z2zBI^aQiWTCp3xJL{reuIk~Um16e`1pV|9Q)1;6$3s!g(%(Qx)&$3dy zK2p4riu-`RGB(6;^sSAK;cve|{@iCLUnbe(2it_$5*1?f{;y^Y&jN@>NhOn~yCJk& z>eqm9&KJzLuac&6_O~actH+Ng>fg*iQmD|~;r63!+JF02)}P+-XI5YSBI-XJRKmA5 z)qMYlyxZ?r=w6>sz20f_u&fAfl3e>a4bRh_{#S*%ichpkR9QA?U^G1POU^KbciBf* zuQw^yhAYG|hL5|Pxn&G_D(z3~rrYz6bl|CigTYhUf7d3+2=%%&s4 z2qV9}5PwI8dZXF}wQI*1%@1;(wpTD>8{$O17Si>8iRAi&g?eMZx{YHUwnca~aPCga z0;q+X-ocI?JoSZ=*8M>B&zTxY>&2$P?pkb;t=E!za^4ft-ZyEEA$r(ThBc{52W^}F zU!P`LxcBt(X)7rrHxbUAPiIfnw}p1%DoQ^H4BpDw6Z~Y5yfHO@b=p?%>d@ z`BnoE5aH@WfB2k>tdglMjKXWU@>Sf+i+_ueS{)O(^CmiNu(9|1%Z9(LF^Io(&S&M%dIGsH5b3b zE`R;9B4O)l!Q!;?rP93dNR1FcL+Nrl<%O$T}9JUyb_#7qAo z%|$N&xWArXXnUXWRD#uUy(488iXsear_Ot?knUzyU#ixo#3Pf>>Gb;SJi! zs+%2&=z_yK`7MZqIh-s=2}gJe=8q?S^?!Og7db*s{N9qQ>$yL*O~QVcd6bWy!2Rg! zaO7b=@*2tKskkDk*D*SH%jrB6+n*krC!UcI!Lw#Yb4WaNJDRpE9zzml2m}THec}rK zD{_v7wLbPSC?Xmh7K3v7tE$oZnN)?y&=FMs5q z{$4}C6}C-*L`u5iSOoV8rt5ZI-r|(uQ@))8`MUp8$JvzKV(mHey;|J)8zCm!O?R)4 z<$Ns71tTV$s5jWRS35gW{|e_a6-@e_-0&t(Ely1B19hCoZ!C!kYK_?IV?DX0#&F!F z0o@?nd@sBlCn{}#s(?M^GV*8acz^EOY>Hj=+KI&|F@kc5eL0GZU4IzXB_F=D^Azhg z;?vgW^eKo2NhIqkJ<=kMb0Te}6qYs`Y}Sr^zshNV$oT8f)?;mS*jr| zfT*88@SauMi}(_qYR(qFA3N}4JoIL7`6S0+mP{X449sFzU`jrL88n_(}7>r244*pD07_Y(GY7xL|aOANA& z+kITzg%5*0v~sJ9tj~H$hCtf&-16yAiggF^X-9_;0_T+s`-cB!UOL0DPJ^*+8_J3`J#O*~D5pGN7L_2OW6XT(#u{n6{Pc3yK@ zqZ>@pnQUWDD5R#AoBvU>Ecq3>`yl||0iM5=QjxRW^zsnYJ}Y3B)ow@C_1JQEU?N}D2&cFz`;0!1y)79^aosf?=WMyF%9#eYiE8j90pL!jra z{Pj0ca`hX9B9#q|W%YXd!~~7vC5pxnno?rYBDaf{)AocnW*7BFjK4WOKh8!e`}yQ&%AcE8rhB8Dw$l#UOCKX{zOb%F4Ve=qXCieNF0mGerdjd1sz zAQ{eK9w!|Wcsv!F34d%Hxf%ELjaENx$6W+GAiQ~>%INiq{a8GIKzJoSB9};7he}f4 z^1~p@y=oO(*%7&2eDU%a`z{qV+(|zz-M5Ss5^E>ii)&@=VTj)Auz6s#HJWm*_V!R3 zHS<`>0V@A*@uF6If1}-!o%`7tPHJ3?VC`hoS@#;F@t8C-d4JFcx-xZ5Y>e>dk2~gs zwTQ}6=VdbOsh*rCPm|=*d+nNpv&wwhvyG02EJKIIR=(zIS7Lm;Gvm*fr&% z*4Z!q_eZFTZQ47rQ*zp#kZUjT+L(>;Ql61y5|3`xpi(Y3e9TIpnT=WzjC5YL9v)J-mct`HiIHChD zO#ZYiy4g%&d53d{@km{2sk;OYpNRE6=gHls!m~qjs^i>#R9_^CMXX<5amUqQ!7`X5 z;?i~Y(1^Yt>>*4aA>UXzr9>OPdGl{wR3pM1s-@|ol!~t;MWBP_OST5ZY70Hf%)`{JMo+63iL^2z8 zho&#bB!6Wx!Fzx94QNa3?n~Cr0$!w#VaMjL16YsmxOJ!WA1=5I)i91paQdF!_*u^Q zdH)RSj<2cv(?eWe1UtmNWWz|VT0v6n#$7=%D^St%ra{pQ!t-;e=R~+-u}isv!e#D0 zTg%DWO^IC%yoBM=`t4|AGX6zMj_2vF` z72vS(zxwbJMangbvQKIj&zEhwqZBo<}!LWtlvdRWVV4@|$ zLD{a%oIvbP-IHj;3+^d&IrLCgl)}L8P6q4enazqV;eW1GVs(P@fMAtY3?ZT`iVJO# zHMLPY$?1aDbvRg1vTDn-*Re_5E0rsAO8pS={EZ0JQRREv$-|A>cFQr)J)CRReSaKD zd*jQd^T-19MW2(@rMiyyg8M zu5){tS)st1S`j`E;b8(I&`qw^YP|G6xC5UN$$iQT>Xg79@x};|=8of~?0?Eo7QPD| zVnHY?Yh%Akx@>lGdXr1@LW1~1&hV3f0a_74LXh$gC(Fe72|sfAxt*802Zm4^=mZ6D zImv=Lv93)_wNoL6xv2<;f5vbgeatxyn-8I`TyHyl=2n&zdmk*<&XuVxZR)BCTY>TH zey}-hFcC?Wg$3SKA;ydtmCkhG`K;PNPb*~$95Eb}v6Wf}Xn`9D`OjbuxT#K1S2CsALJ2QDx%y_9 z4L%zv{vafkYmV4K;RJ08XM~d!8NZZUtW)a>Z}~|*Qn8jBP0tQWD1Q`(#-gxVWzaG$ z)~l08!;74V!{Ho})eIZiA+8`d9R56;o)cW6u!_c!6|sZH`#RQKX;VHlf@Nv~eP zR~F#-E#%PZ6BHT*dv5Q`2^=RBdWFYVak?2UYgmNrwWjBK^{pLsx4tAn764%-n}JM* z1;5-*JG)*qL`Z)l>|-uw@u8UPG*&r5O+otGLudwjkuSxx>)KvBf~)%d_zrjqFK0MQ z^)0gv?h|~FBybSh>*W^@l&nEPp(QwcjtPeckI& zXqa{)dX`WqSe{!vr}5H_I{oFqvCSUG(>6L*kw&h}4C!QH z`*euKA%EH7B7~8CLQnMicu~~&fUOjX5$P-WXW5Q0YSIMu|qWC0-L#7yEs4#~khY}sav2oq$Uq3dShoOaj-mN0Ep(~XC4Kz9(*fUi$J3${t zgpv@!UtjTLtgjBZWuQU55|5EBRzC9_(NHJi3x9QzH!QQotUDa?pi1RGo)*d)*2d( zb@gh@jHq53ESv>hL9m}bb|$(#^&swtsTmk^Pu-w9<32x6T z=+tF8eB8Qqt64)J(Q#xU*zE;IpRS_E-$Ulz&j?e!DK>&>eJyuoS<1-u})>*2n7meqT0{ zjT|-!09E2^EyJ;*0&J!Ru-S7@zsSS}2F0d3_xI8)Z69>43#t#e4iy-w9RO5nLTB=> z*!;@sT5x=Eeo|*ViF?%9mTA%14_uBa_VX?p3B!C(_(Ue1GPLJHH;)U6m~l4(Hh(A| z5>6#^ZEi9N)fdc5HM(o>;Mf%-ii}vvKXnrPKI^HGu-1#x{q^zI0Co;9af4)t4h*G3A)mw&q@i03bvoNs`aW6S z-`l|bt42^)l&w%Wua-I0-C0hV81b+%eHSbgxelEfPOS?!`!V_Rv=rx*4u6inT&u6_ zPC5{IzewX4lsZ+VX2+Gm)z@ovXV!ytb8xV|MPA!$lUj40kj8}daDMlu(Ii8)QPApi zQJVJZHw5}=<#M4n1@;6eqlAgDaEOC*^G++*7lk}qEeR{xlc5Eb7a}ipHzQYqk&4A zOd~4j0K|Ms6L)zK@UX=iUKbx`0RN#k&y=eL+h3<7tEt{g_nvhprYCbGC(jyA-tpDADDFesX*YI?!p zOmIp)#_H~}4(fR+7Z$uI`T7WFB5;MX{idtaxv?#!T37czQz1HbSs0yZ?-}*gbm{Ao zpkWfuOn*>FiVFDf>47LMHn3kxv)SC51}v3>(q}4#VNe|QaT z1#J%LP@y>L_4^$b)pmb5xde~NBis10{E8Q|0U5JKS$qpV$1*85ntGnpzkraFXAc1pQ&(oVo<7;r6F0S+Ny8`?F?xHUKGu% z1-U;2J-9wK26-_5Y$|t`Q0oqNrra0~eyyki14u4;z1(9c@>+j})IskFgezM*%Zf8r zL1S;?T0GIKP0IvMLUO&_>DMWVnSAt*WX5Snws<+8-Z7X*lU+i3Ty}MsQZaR&9<6$T zc_{UIx@j3nw0+#W?H*pg3JEjwb!wii=w1Uu|= z-ojRuHkNcOi-vSu6E11DXdQL!vZKG>i*zuGS*q>2PqS1{Dl|^t>83*3&Md{!*;L7u zXG`n4?`lOV_vIiPe0Bcu`JwVbXDKfqK%ShD=78x7BDR0#!xc74$>k(4=o@P7r?!GV zEtAuY35ZO18jU-$XU-(v8_t{fk`;@O@>rY;r+vLoNR5kJXPmPP5ym|@wMn}43o5v0 z*uzAC&`C&yR`vM}VVf>&Cz2($+%BQjAe#&RSME3`JX7iUV^Krk3t;LvL~!GvxZ`)W zm+pK*B5{AO2lN3M+f^@u@+Q0qCO)%WVtf3MN6F~Cx;C-QbbMC24LydTIlS=Q!a^`R z^@xC;7OZPu;LQFsUf3x3Lzv{b#is--Z4ldqP>oo=glmUS5NDg&h`J3C0DUBRr9V-P ztYR~kA7`h>N+f~DHBb%%fF_5(J@-V01$NOdFJ7w+W2XX+?hPLHvCHbtT)K#k(BxIcD$if%jAMEtj>Vb%!hl(?d)Gj+quUR4NN*6P30WRd}#YGL$MOsf@F}PV}pldi{*G%vkd> ztN2Auy8=JJC^9zs$QUhD*ys>oDTDCc5?_J?v~sDjjA>`E8TIOs(9yLB**16S_BHB+ z+ZOcT{ArV!EvX_Xr>*+#lL~vXgVG?Wa9DrM9D?Um0>GgQd9<=;Lb?TgY$raP-^GfI zNPAwnvm*U8o!OJ42CH^FEKrjyN@8P098X~su%#&RB0(YHhs>5bPPQvg&b-&%jZlO2 zQ@%uBgI|lXT?qae4h3=A$7g6?Q&KKtpc!s8oZ)Avx^KnBit6l5B$&bsqg8Q`#D0Tx9nQX}-BG z(Wdc!Hhc5*4`A>VOjlV*97K+OIFDMoyb)`<>`vQS`z)$=aM%u8e=7$$&Sh_`O{o*N`e#%W-?#E;sp0fIxFr)S!p{ zoe_(*3Kr|4>6FJK?z7=d>iq-hLA2AcnmJmDib`Mj$JkdwfzvR_R!CDE23-hTdRq{3 zx5iO3*;_hhHi<2|B-a#_5ucuH_lOhC*mZtpZqwk4hzGN$J5l=SxL`0bi!6WXqeO24 z-?D9RR8d?`$7xP}S>Cp*Zhq|;QW{HfR_jSM+WN0*D-e$tX2>{xhMSpvO)XOj+m7{} zo2U_zKTNm%culUFYGoO@@A@)f4Z)2XWk}X57kEV4yd;b=S9C6HIKYZLMm&?c@TZ=3 zGrUYMHMp%$^ABN6n7cWITDRPMC!zf?}QOLuft`5bxP z!lB2KFY_g$l91AU@NgUrX%8_vi}-&y`^Uh@x@cV#?pPgn%Sz5lr zRn4fV6i7l0`^7Q@v}#wS%gCI3@Xhy%p#0A!9VR`m%=t}E`;w~Oo4>VN-sRxv3}f>X zL5fFXL*kHU7hdC(7>Iv#p!m@}dbFUf)l??C*t~h*Yl-LOeem~j0`mJGuQ3LgsRe%P z3~eth`SqQGusnfvQ0gX^51*R(hSJXFL&-XwF^2KT_KYO=JB94DH07BOu;0#@L{4om zb=My9W)f^Usz~-l=(AURs$Hn~4+?(2eEVv$kZ0eblI9=GItzb;SkjFA*yY^ews+lb zBpe7&Q1tp0elpW(LKAFf)KV{j3ms+B8eAFrwafJx^-0f{BsYkNz)6<+)q4x;<61BfJ-2#gXzU~^)J&dJv z$hh6(mCmR&&km{W#JWO};*}qRp=wPspn(+FB1*?T6AVtf!Nv4w)%+}fVj~Zo`~(wt zfpXjanx;Q+gUPaGTFc$NyD2|YVDf%PApe#VOTu-KeF1-Z2G{oVd3vZ*1%!0Uf~oRQ zJ5nx1bo4%e(dVr}IVb$2d0;>K5Z6<_Rgira%NY@`P|7964tDNcFc4uTe_)gBSAL4R z`^|hl^b zDN+tFD=vRqMwHZ#M=Yqb5ptFfGj)e77c?wA&ecGV%(+)zM><_uv+R4xSY{TMN{Zz~ zA_fAz?(y67EZM+Q7}_s6Fg7yTMuj?ZA{aD0|8a+?c`NPd>Sxuj5wF9IY5AGgg`=8v z{*xY)r^iZe2%xuJnILVoN2;kfQ2a;d#p42Lp>BU*uWT3Pt`9$OBAUS2O}c)vh6#<= zasyw*FPQj&F*Ho&$IPR1bH(n#=M|ZjK28(IiY2|Z3cA#;4Q4q7Tz9$S2+W*_Sg|fM zeryx_&N5BILO6+|i(Npr57(#>=i0-17*yra7D^jw>EX>MlmIx}>8}6HtvsiqPhzcrqxaDIQuKUvD93B9nF5xNd?I4oy4zlkYl->BOh zd3+3uLGKy+?e`a|fno->mo!_qc{jpg&V8P&GaP%z#0+<*B@dQ&CpnF~vj?B)<7|xv zWgP2tFWaUF0#He3gmZ!v?Y0-RH^j2mY`xL;#28y0vdtq(*C}~XIojXg#|d!hI8Jp*34OT6zvo*WIt5e}=`uIqnrz`zySEItOO?E+&8@0@-3kEcoOJ^W?2L8X5a9TUSi zeVJ|gEyum&&+O{ll(H$yV=ad*I;k3PjT&9(1mvcvlxtOCaD7{PwB1|r1Jgw<1n`J3 z9OwICP7d8S7vecz*aJJr$iu-g9lH75Ati;D$KUV?p6vir9ZYwa81fVKDJvK(?Ezzv*~GXLPpVJ+0bgN)GQvUYyCVjH0JE zkf?UE8<#0HLskFUat9b&9c#4#5i>;&HEH3^0C%hlz5ajLlw#e)O-g;v5(xcHdxGEb z(6&&CgYL-}h9{#fjmgWXju&7|sNwX8Hg=1N?dq}B1YsM~Hwjoc4k!gq$ckUDNBh@! z`g7lIhoYFUV1v9aTeXx2O$RJ)kZ-3KgX^Hi;qCS*sv8fSl;LQ^8|QvE%UCx$P@ZV^ zkal{j>Un<~8X)U~sr|ZUQ%LsbP1bbGd_RHZ)7z(xM?l~8-#$PQEd9op{4-~iJT9ob zwbhX_rzPOPI-}Jz${wbF(?~s1N>hKNmx6V=&ph>P9nh*2B@vJ3n>i5inh~hR&P22M zC2zE`c;@uSy%1Mtm1PU<)qKkZMcEw7twyYK`FMY1ohMD9#-0;SkOK*mQ)P!SSIYQH z)Cgl_tzVOtCnTl3i+vrNg*Yw~r9|fH^QyerH7WRY_%+UmQE|Av;{9IVoBOQOdMgIT z0C(2u%B@(l=Pz=}nX`(i2Wuu}$Q4tv=`dHQ(9 zWwd{pqbL_!lhmoKh>GdVdvEq<#N5w>bIZpExNdP zQ@$rwbQ=v~WV}~+tUQK1t?etlFOAMyTC1(1qzZV;dRD+hV?hCL{v2D%#(oe)6NLCDNYa90 zrJ^q@U-i#dLWdjo)jE&b8fw6)DnIG2hEC9hrNpu>FfdrANOz5IyG~HUezB>B;LLvv zh77%xK|DG*?K4|#sLSmP8!!}q4z@iM(gFPv!|{P7cd&70{R##N)Uq1s@^VWyS|E6m zjh(SE%A2sjso+AHCL_rnB&T^gO)jVgJ%sMhE?(vfzTzf?RcTYD{t%1C31cHh_@j5BKRETfQH~X|lvvO(9HR}M!jga7WRY$8 z@S4hOJ8|eBr;xIHtb&^B%uSyU>NDh}D>5&(p3`{QT7E`dhOYgbER7ocan?xYw%(4Q z*zXu8gOHjLvh03=^EFa3HQPVj>f9~H;Ut}fqSES2!nV7R{5q~!E3>0wgw_lte zMm<)zgLwy3yw`-@%n?jsW~}? z0h+kM1vevVG@5`>l7FFvz>>Qm72|OH7p&mZsoDln_3MhKj0;yz-h;M_v<~<6yo1Zj z_2-uR(F^WpkBRF;I2?{H)3vAYX+P9uiP|ATwl1T#m_m5{)TpIqS4)4~GL+$rg$DWa z)N+n47;Jqd8uc&cH5ud<&M?FLu>0jZqG7M|*1p{% zA*7xCkTHc4aMeP^#U+~R!&_Ke3yfDn=nbcA{dU!x!8iih3>Hz+lr4}lE^uA5R?39+8mBMKZnliW+jw1xJEp1?xgk)xX z2O)GQQLSyioQ+X6ccqZ|GZT527@H;}YC$I-#>@^*ZG+0^QR-1I4j5?tV{kGA%gx^b zw8d6CNg35&oKAmzAg+L~+Hqo)L{mqmf}SO6Z4skH{roFJE7>vElJ(qft$=&$aZN&& zm#sF<=e%b@PoI2Q(KRys1V7PREi&3Y3Q_VWxr}gj|JHfCD!iPEKc~Ll?OR8q;QOJ^_ZGo3=mBNE zK|K`@N&1PUGf!-n^69VpVCUN{+VdS%@>usVTD*xV=Wq{TwAR3 z()l^9v%Y^B!sG%+-IZvx4~y4Y+SmypXeNJ9Fw^bFu1Dra2OYS!GTTcc8X1uI&k8F{ey{-j6`EV3EhrF@b}-2B4r1hYM( zz|Vh#7EvS#^_BgIQ~;@C?UR%=M(s1-NF++(R3L|EP@D;986i;sywmC@d0mx?d>8D(@2Gy+z=ioTQg4 z40J&Gj1PxKL#8@plS5~wzEd^hlNktSPq=?bER$W^sRF~{hC5sa>9 zBPry|%jIgKftHUdz26{{9O+Aq=(aXbhJ|KwCG1^-fvh70r>L^pRvSrE^o?O$0)Kya zlUXQAS(D%5&@40cpQkk0twv>O>fv1c0r_PSW{tlD6lx{mqCwB*~UQ?~++kTsnn8}R%UVfEDyf(?yo;Y&<|{`5SHXks&Y z8Iaz1u3gidQm(X>m?8X&5i6g89O!>c{vVur>k7~5t@ zj#l5oAp%w(CzxxW2Ub8fodvniOh7n+EKXK9q7btr;Q>sI!(yTns%DO6OMO}4wf)&z zs>myVjGdyU%*y*{mq7>N)X}xsP~B8+9w7K_^(clmO`#-k%dsNHUf1c;pO$|i#*EDC zgv}xwy-e?S&0nlkP~}%UWllOY^%E9L^?2LyQ{3+Bz4Px?=A;? zn9i57vbriuSS59Z8(KGroWAZ?1o!W2nVXmw?ZBKVq~qyw$I;y~IRVy;|ENwP;t- zqhD&>gS4V=gtQ3~doN!e+lLM5$QJ?&I*qQ4|fK8z~i+Xj0219L*uDJGQIWq(Ca->&@E~ z2NXzZap_`N+Yx_oZC+g$9(0TE6oGC8V>=Q^xB<9s!Ou`vL)J>PE1TXqSZ-D1T(10T z6pjvsa5f+NGQZ+CtSwjG@Ugqt7@!lkT@zu+syg6MQ&FDbTzajGGm~Hkh??59KLD;s zj2Lvlpe+*%F%VCKU8E1drB?x{VS5FrsIawz=m#@ns z-N4(AkX`q>%;fT_I!1(}O10hdBtRu){|Uyj(K{H^lu|e`%a&2LLI1$Y7PJ8?RicYj zXSvq7B%^-eb^C@6ayzQ!LW~H#P9Fey1a~UK5nyDO8^wR(&bJrZZm`r#LH~Y z45R!OhVFNEc32gYE3Q-PoN=~+e)8q+2Fu=La}$#(vQRp5p%^e%tc0>x!B=|M&#H5l zwc*{e-*>B&@yS;QDjRS2rRiF(1pYiOK9KB~m2^?RA$eG?IO-Q2($`yfNzItFwO)d) z6Q6%KBI)*RJN@-@zmSz(X+)^|r*6n$jM3UUU_OzqE+VKN+W6{p&RT;ZOU`701+|Y>s{WP^^`FvEt;)u3L@FeR zVcs8ZLtHO$?(<2HYv`~WY)q{AhW0*sm`6cMD-*xX)ePwo>E|qe=;feR;=}t(N(_Ii zjUEb`kk61~W|QOjLUnM}ZR_ZQ8iD)LwPZ@yp@jE2Vc(Mg9~$ti(&Ac@>`7J8hvztt?A-g%7LclxOU+ zPr;MxnBj+p!x7BJ82MTiMq`P%BcHt^BGUTR4ha*p?O`f|c{WCSQH-_)vOQ zi%D+)w>RYqHOZyw^@3`wgW?q9dZ0Bsf;!ZW(BW0*1g_v59IE5;SxO@@Q?Y+?{u&6@6F#0t}12344{9%*6&P!LB)6`GFgwP6|?-XX3bk+$a!-XCv$guM|`T% ziDfsa>h9IWm8=ZMFT$e0s-T z!hF#Vk4o+gYW`R{okP&L7kl<{his3(C{BREAEiK01bAX8ut!HRr|yrZ?MtK+a-d6@ zy7`^{#1{r&bjggLT;yX!i7SuwS-?gvnIk(KM-8WwR4F@UKxX2G>hwaXT*3Qzw%pLX z9l;9KXVcFmGYRC>f~0>L3n9uG7H&+OSG$#5dn1}Lk3Q*l7PD-&xoM|_rteJh{lYH} zrqEN;GCD+`Q`(hWpVRVYy*kb1 z@@j6=)~xA16ngRp8PLlVE9}`e!5Ad4a``%y9}A$uGQ%Gc!?%CVK>0k3ffr!XP&@f* zZRUiccEK1C=)N&vH(if&nykbl=3E||GxZiTP@4$3xqs@Ie3@(-39Z$ebFcCdSaW;W z3Itcp2j?cO{kF6;G|&Bj*LVV4J$Gc>r+Re{fj?}X{(kJ#fZ;9?PRZF|8NI;Z)5+p? zDkKK(XM_#S?Dv0#MwTB1GX#U~+t~0B#`~MBS{WbyT7!FN8Hef`_xW7DTY&q_%#&jR z+y3Ra2^?6sErqFtzFH5M2*slqhSf$f`?k_DuG zbZwxkj1@b3la#1WOa_D?nGfn<7<~y;my1nzBQiiWgll2U+Mk5|(K5z?WgS4)(gP%@ zutUzJrmBCukq_kXzVttDJ51{v$vt9Br1vEkXeoz5Af=@Y9pf6jo~X;Yvs52(e~oN6 zF>R=XyH5D91Vkq2a}cDUHj^N|c1WI8d`4(9XWlFxEgKv|UFU7LSj<^Y$G@qwE^5R* z!mV{+;ZufNoX*fK-+!riCbPhQFq3XSEv7RsF>`+-f2w4j+Kk8XjVYykd9sJa+*8s0 zk?u0Sd8$*#?bShk)!cZ++vnuO%wI6mW#(i)N$hm8J9LdTR`qUmX!DhA$A#zX2HFQ~ z>y>+0NB1q4M*(R^7rSilq8%7r6q0yC5B$C65xUKQ7S9;Mqk4OsjaR!UYl{S)+fg|C zHX469ppQn`wd@|o?jDJR(scfqdT&;rU=;!>CEIbb1`L37g<0ovh>WvPbD$hK`~;Su zbOH@cEsatcD1^V?&7oDhtyGRuji_)G8NZ8c+&Ua{TWoHo+5AAm-#wx0h;qUf zB+j1c?ON|hi!+k-@GHtQHH6b*CD11NAEbYwr?+iG8E#j1*XtWJr`B_{zULfu>AQLO zX5wCSFi@hcWYzW?p3%!$J=CX3B{971^=`%Fpgx?^ZJ@+TkCnQM;#8#zj))u8%8;IC z#grTVscneCZ^31utkmUE(KI4p)1bg|it58Y7S|S*s5dEDRnoTE&KRjx>9KfHBA|cw zw0>JgArm9_iNRYMO(t!4Au(#J%a8FAiTuYtW={lTWk}jE`s%m(>;2VN3c|ifsl)LV zq$4vH~q$lEWY_8=ly@k7U;u9#u--fk%{^(g_24SeMEZI;IK4IC`M@h z=k~26wtF8PffVAMN@3ZWY9Lcxy4`=J$9W&>(WH_4k0UVMZYeKnm z84DLZP^%p|FJyruPpcDt~|7CA0t7 z%9R?u@KZ;eIm18G($#f4ch?1oHXM~ho;>9f*Rt`4q8jX7{kYi#TN2%Pxmc^=>`G0e z9iAl_?bHQZjhwZd6q=W95%Y+GiEFRa#h|N9=Ws2hn?`#Pgt&IF9Y)0#bI6W`eQu{t zGChn^CYF`shO?W40Vs{}0=9q1zfnF-cnY*+C@QEzU7B47u*h)|uIivXI)v6wH>Ej| znX>~r9Iri{z@a$3LVoLMeqq%-@QFoJT>qstcVApv!dMD~YEomdb)~^v^l!6xq^$B@ zVas-vbOwt?5OpFE|B{m@p)9^FZ}-RbBfUhSMOk}pQBik^EyNb-o3(!})J3qEsYLRr z&GhyDQ6v1r)#53D`J=fU92WCs(RNBufnv=2V}sBk`TUtE-5me#+X_DHt$_OACo!?P z^^(fDxruBN)?%A%=V0;-LVYy|Y{Zn1TF0|hetdkH3#kSuN{xn+2&~ntw`0&Z%BQ@R zncQD$*Q=u`$@m+ctJZ%d#+*M!+xh?tii}cC_6J`KsyLwoB?&)3f->EtpZ1BZR-I2# z_e6OVOLrejkA&ygvpIuR@Rr4CnedkV?EJQtSGFa8*2?H${3J;WApjANye`h|@zLa3 z$?VAi=crV=;(T_uXk>@7qP`ouK7*t9CGqs6b(=*o{g)HVcvpWLj_n#e=gh_s)w^g3 zN}G1g%x3QD4|%9K?GELN?;k!-a@MrLdYf&`=^Q0n8{sPaY`QQVqD$iu#;$H$#FpN$ zv-2yn`#6`YXdgEE6T69KV89l`whSC`6H2ONy(ui2341!bA!M1`J-2J42)D9wV)Oaq z6`UWC2-}uQ8D@V8s<@U6a!AnHSb^gMiwxICtS{QZj>ekca%-e;l;oRhP1{;!E~n8* zUbD94-w%U|Gj%}liyGn1@nOg7G_?|QQGDPb_DvZvjKEaKr z^~rE=9&Us1+PR^Q-0*1vaSPOx(_=T0eGzX4>W734Ba461?`~1u;1YGLut+Dh$d&W; zUCeIt))OVA)|Zca6Y)W1)?%}sifHF?1{f0A^^cFTC z|9GlPF`d+*}t;&@L5zFd;6=$x^Jql9ewfpcG<+TXk_k1Xq;k7WoxUhTyj1%IJk zBn*HdV>xmpVn1T_LFNhKzR#8NRo`HA-~aMRqjQX)rbMX1ywpb2e^+|(^UV=6Rx+1$ zd~2=vBiNU-3c!qTL!e`8OusIv7V-4YVX%`RX#Uwbej(_mq^!mAW_ZBgG?5beF zjS_a@M$?KjK}%);o9|;(a2A_6U)yzlF)~_lW(WMSQec0q6j1O4dywlEn>931#a{`< zgbe1}Bv+080PS*{I|a%jjCysP+14Q*YxpUt$5uGTnc; z;#m|pH4)TVAx+F^kmFHbdTq*OHBQt}Sg_Y()Z=cH)SoNsYXk#d7fvp>v=(}2s$Q99 z#hCKUj-OrPLJ193rFogR_=o^Sa($-zwZU>Ra<|xyd8lc%xv1{(txVGCqEKC+LU1L) zoZK&jq4B?=%4Aj=E|4tu@t+QN%A$Ww^17+i>%T~)Falm<_rCEwZ;A>nhe!BkRF}Di z+nd{Y_)3wE&cu>Zn{@qny4o8d3hH0#V-v?z~fxo8Cj8lh+FlP|VKxPk0;hul0jm!R1 zwtd--w~Aqz(v8WIvdQfD2L9v1q4seat(ZNmq#LIEKu)$(NiVX77b2SGM$5y)^I6X* zj>4CRHn%Z`8#mqT4{?LqVOrlTQ3(7#$X1^(`+nx%bW~Mkm5yy+BT}N6^=+FEFrNCI zLO$T;R}+~eTn@ay%8Fjx+XH`-EJjXQw-pnc=${;V5myP-aW_wuQ?um&;CsSGt0C2g z+PFAoadh8mV$U}xqSLfm9aa>$^QH_A=D(uX+*i~}EzcQ{%w^ai9a>;2kdDqjOk3Ul zJ`a)^j(LrFqJe&aS`x`LclQ-hN$m8pWdpdLd`{m{;a7bzpRQ5`!-0QZ_0;f;XV9wW z=7C(>Hal}MM@QA5^#GaF(Ngwlnlz`)?>IKIeGmAJ-O?C`M%O|4`k^ntziDY^QD~nb?^Mz=yXzC6KYucv z$%`{T8O+eakAdr=U^ss+qlEKd3H&!vzDM00^&uzltYG#>{z;UF{YjKN)mQO9AGJ## zE}K`d#KnrTZ=}8lv_BmGq_4XIp;37N?0&b>xe{Gai;e;(^8|x4b#0y3zp8Z~8^GBt z_)^E83v5A7btogX*(kT9I)2dz6QkSPxR`swremX+V{%kP1%Clpzym0yo^Ge}ByQ%s1#6W}3X+&zIr!mAP!dmk zmS{!wkWzV`UnGvFY6#NM3Lr8tQi|b^2P~cnZ~WUxS-Zq|HMb4Z=`FMz-+>rnv->KT z_AYL%R`>Lg3upHa;{Pm=4w8-U0W8GOS-peJY-2Wp_fBNSi1T$sQ%|Y$^g1nwKBlY^`$U zL{TN0zlc#08atY=cBRAF+yJ?UA~J+N_+A$V0KJp;!&(lXAKe~RGv?RaZf4QNjw0=9 zd!YLAGYo2X&A}*1qQnx3q}@2mz09RmL7RA*@PL0x-Q$e(kS#u|X(l(Mn!zdYV0PXd zZ+Afb%ofpr*jcg*ui+JDu}rySZ_&^lAxhy$xR_7O(B+sDf1%~%Z_^~ZL>xR`BTQWl zJ+i}$42)OiUh#;o8j}%}m*wDv#UG}h!6@JiCnX$w&~?h5JG{$zK4S6bgI zfnt9V7UF()OQlpQ|Jrt>#<Q@s_r|I9}E~s`d zP-jw;2%uAlU_(%>UN#SkWig9mhW{YZlEL@+oLBbhkH8M+sZ&AHvC=wymSu3{@El75PBj-WVD_^wIW&T)!WXJ@sy; z6ZT=S?EV4~%>M$Sg$^-~3XtYFv<^>mJ0kXQJl-VX@N~khKs0ETVcTV2V3V;|atEPN z#n&dJRFvj*jyQcFJ_3k57oz-@--v&d%LYFTV9}bF@NR_gW24w9f`YIq!{f5k3p;)S zs)DDe1N(-`+{(+LrPp?zGW>0oAH`~ z%*5R7?vVV=T$i9is|oT7Xo$dPn(vdazpk^O;u=&I65MV=f7n&0CyI&spbmfA9#t#u z3FDvykJVRqdrT$Y)6sQGrZXqwQ)r`Cul*&VIt>jBO9B6ri{*CX0NA3ReP(LT(j#N}6El}zrk))nhGTzyK$+Ob z=K;rGs8qWY!K48s3@QUV<^@iQSfFUZ2x!PJpfF zwkTCPTZ~t|x+(Kxd3TJd(L?0Sy&ex;Au8*-@hFx|%Y)%3O0q+Yae06F@MI_jJXm-) z;?|&8(^B)mMsKx-Bl0u{lzndwr^eQ%;6SSwUh_sPIGiA;Uai{KLivkU?u7G^=pk&s ziw#2fIC4!fvvi;Ox>xslaa>O*Lfm6;G7CL@{TV3-CVPM%`$J#=iHVv|DeA`c2pysj zv}J`fmc-sjeEdcBqc4Aa6iW8z%(97;16{S>^A-zIIwsFP^K?OL*_A^a7`df~aG3Ky zNHkOW>2-aOzzFw!vH?<5sE}Q7%2=esnvFcXWFd+$1_jU zkCI5hUyk8N&Y~oeqM;E6Z_^2d{gYRtVLE0oxQB(E91JygrM@(hW#Z0^N{>YR1zJu- z>dw*XGMRc}uwsLu4aX_JBS!tLY-7+;aNv0Us6idi;Yxr1esu!3#~{t=nbhg%+ATin zPrm3h-ikLd~e0Gf_hH~>HiS=x_>xB&n zt_ot2Y}v5dL5tZ5%{iVOWvQElSonE+b|1o(m z2`RtZeNj>~C5=idHS9%-6=qviiKNXjC2e3I!zIy}KcH+B(Cd!m6T|aU0W(N;9`N^v)ad5u+y;kjQ%$b}g9V?&;)0(U~S*(@d`?g}SY@1oH zDMo))aJ!I2K9wpR>r7k|{Vnw%S_%s$E??*8x2qbws#fVV)=(X{F@R1sEI8DfWMaE4 z9Iv?&9U4w(6481~+hWQ((AaeMzW0&{roj0O+^K7vpUSgAqBSKM5LaId=R-x-%I#L*T z6W^*VBecA{y$DXZp(`Cvg0qYLvgN4&c!FRDW3&`gwyAUO=RrUy5}>e~8k=-tavk&< z9EnVd+(x@r$W>t7=E@~*k|QtWbb5aVE?e;y`xe@T{}eFXDD`$zX>(CmnRU;v3ej|P z_QSeY)#hE*q*$GXStCoDN`t(EioZlP)EI_rhH!;fl@7}GDsY0AqYFlhGz^D7j!Loe zHhgAb5nR!7*~hh}kk7ld-YToASfCG64q^1t^U;4Yn|v9~BQnx|UkFLJ2YP?>BGW=} z6yLgcR3X&Q2K@iHu;;H2FekJl}y?C#G! zYjt--Zu}S7v3WdXQtK7{z1#BvZ3Vj%1=wtv?QQzCh+Ezt_}O>x2S0xWxbpkio?t}S zqLGG8BOheV+>D+(~r(VlF1m(0ZRs~D5AFi{fyG$!xS#(d zpji{PkcHZ&^&L6&rpma3hfczBol<+As#YvJg1^#g*J1JV{`aaPXc!3_#<_gm@le4? z^*)V*46nBc7QioP_N@JV-|fjLy(%8Eb#c>`9c@z$!ZRcI-4R zN41FW>xs8+)OL0Fkn>|aUtsWDQ|H5514HIf|iV1r38C+EhD_?z$d$~?#Idn(`RaFq^;RC1RIRGqK$T$QgV}!o(RHTPW*6bm6eIUksYv_mzu;dmO7j;b`yv@dE^0^gIF zZ542h`AKoL;BXc{Tir?y42r2o1y^2H9Dw!z21W#hF9k#^ z?6HW2#uRxGsK5YB)Rkec{MP{#9iC6{*=&F7sdQf5z4I$@8$oeadjQQ7=;wWvNzdSKzG9Kvk?Vix ztsI2}b7k@pZ2(h1tiRQ5#q9)!$K1N0C~6F4KI|D*j~2PoPT$zOG8hE|>?-eG4XelM z0s=Rz7G3A4%w4R&$r}A#C8S6GF{KeY(MpCFGuF9mBgl;V_=u7I^PmQ6^arb>oe>&!!0Qs=5K0KnnZ zl|W!B%^UEi{Yu?1;x}^>Mob214>9N?I0$Rs1X61!eQHDPfWIeokID~5=q6{z^scCX z34(a@SMbrwXXcXc=4n!7@2w^VY!Sw!q*={qgAEBgtR77)OP~CuGRz|9ei6@IvMF4AMeesL-ScPIhydcW>hfI67I~W>HQ5 z1?ATIT*(RI(7a(e-XYt)zL zTKtX^d_DBzSX)X6Mx}23n;+Rm2l7Meh&{)1rUM^;Y`pmrnX3T6$zDJCFSzh8K!aSn z(DO(1{1UI!3aI1F>q4C9K}kglC}Ncg??Kv6oDzfqS18fBZqj-s^{WE!4Ue*axGe#; z#FhA+gDEDzjN|#t8?*Vu6fss@Z92{!Y?3=)l&c6eth!0yS~nX}wAOxLvl4-E@!8CX zjvUq>>pch6gI%n$pJm<)B8!V1rWEbIw9t{afH`SNy$!|N_9nuq#qhlydRgn()nom@ ztpuaJc%$-U zAJi6Dk9CRTk2$LinK>~t`<%6a*m7B}u6lV@auu}-4Tvd_Ecn_ZJhH1#g99a2=JVZY zhqUb53AD?_ewoR0^+8IFr&cFfxI0I7$C9e;@v^FaV(J=#$0NA- zt9u_rtLex-EIfO#T#+5k&}U`j%~;E=ZU}Pc@>IJRgo3KuaLdLXJvwyHxSuy*SCO=f zHdGponTwJw@jR$0S>lHIcKXIZ2(moB=4wcZ2Cc_44UKB`H}rj#8`b+aJ26aoVCA8jYU*&mfi_I7 zOdPuJGWES-zGln);F$T3d>s0q`ztGry>We#s@F~4XvsvU{26Qr?;WA2c!6Bpfzl&L{RPbT^^imBI&QbtK&!!O&gsUAp%(l+_(7IxO(h?r4CwrcGlcl~x zg#?$_U(|YR)H{9DHQlj?PjVua^0R-yik;bXyuTHwkaO|vfLF;`MrF0R7*DOLkUf{# zHowlZ%xB@VV5FEN`ExJ2pNk3RNyG*1p@k~kVlxMuq=6ijsZ@0<2-O-d67qN)7IN8E z1SyNYKMSO+tbL<@CLx@T^~!bgdk7wbvncL4m8O)HmQLrQ8A4@miVWsn#!Zy0nPG$s z7kjWm>YWmb5|LfF4ubJ6XA$o^yNC55Mz@XC@Or(imACL&F)f~AA?V^B9oN-g7NdQ; zf>SKJ15f+zuEfEgG2Og36U+Vl>T~J}kH_PokUY!Z4hXC;G&FYD|P6OUwQMs7{cM~*BrF|C{Y?neBoDB{RVCS9e?y3+rM2W6PH%I z`6vGH<!IEWXgEgauhZE?CR*FWm>3At?4aziIwBq%tVfjmqEH9}|R$%1iF-7<)`P3y$ zB&lVy8z+sD0GqYaVR?6x7T&KL+d{4(=5nIfmLG_Jp67|W#ox1|AE^}Qh(vbe*d2=F z1UyFTR^I+$BhXg#ZotH0cawPaZ6Jlv4oo>U-8b``#X=>)rkNo7e&Wh|f|t@7b0!PH z_ls7R29{WUXHm4a*RzyaB@P=X%qI0Avq}5eFayj ztECsK4O3=4z%3J^D#6L*^%Gt*fc%cftkH&lAV*%UO7O*G!-DyeSi3J!*TXQ~pEIy` z(5Ry#_hC@W=AVq3&wdfQzVJVmY-F1HJp?{|qG?gBHb7tW5^?{*$t95J-Db0;juoA@ zFVQ;r?3`F(K9vmpFm9z`dvVO)U+6Ae>!dY`XGh+`Z(#QXBze^q~sIC_i&-v zZ*2F43|8Gp^S%mk`kS`FbmKlKL+K5+vNMB8wet6sX^;pL1pWC#VenU0trNlg;noc6 zpY`?ilZ3(WdV`*-j3bZETDCv0_i?Ph;4q8q;v(wL3+MrWfUufMCmpr~sZySQafOwH zFMl&69-O5K2ZM^x4cnE9q^_AFZ$aDAc685HXTfL)4OKyNWoy>*J81I%BuvOUPRjmP zoZQQP=xReeyAL;mO@}@@mZu_yfDv|x)ZXizIh`KrBFkTY5DQX$QI!vdaF9f>G8Xd^ zpoG1xJii}F>ow;;d8K)&OQG9;Ly_C|ca7g7S=ahC2m}6Vl|2THN~2I1a^!&8hRrwj z$Fc@3mu62m|GH+Y#8UKXDZ_K$K|Mi8Bji5D{pjjX;Lu z&-#@_(nd}cd_(6_ty5rskWd(b;5#ik|F&IN!iC-5=+{xzc2e|oMol1QUz$~jfNO(# zZa1e1(aG;sIAnN=;kH}Y8FOE#+d4%okKIFCH@VtSvzql|toEJNvq zh=TPaW$MFwm!=ZtvjX0G_X_LME-r#&KSmG}3`lf(n~+5~_lF7-I=* zan_onikuL2|4x?a38#I~qHAAGh&S7JyS;B5*q)D4exAu4n%Ucq$LTtv;^(oN&n{#B zo)xH10@tK_(4bZHyA2v(6WB*Y{Y@|?CJr?$jg*KHkr zG=pLgPFiFz!D*dgT`1*e{Ssw!;@_ONt@Ku^j08Cli9^fARCe)uMaW$Tqx@FY<6~0<24BJl8BYWW z5?Jk=pDtB@qcC3=aiGL+`qidjw|EeOLo1;{L)qwcy%k zUd&Z;^Ncm~+&A9qd^Ox5&Q`94!;ycSiC-|&rdI1h-wTUQx7@ z2)8@`y(T}%_lFe8R4;fJ%w>wn4AOXKR}V1*&b<|=9mYNvJM|ZZ)nP@wTW{b20{=-( z#_d;sRKxWPwokH6O_Rxpc4;wuVCOFlMlvf-X@*#iYMkO>f3PaBm8|cwB?E|Wy1+8> zLxNd|J!)6$Zf!Xx5m)Q&gay6nboz08)}^oapH9Hxi$5xufBynFpH1)e6NUq~6x?)? za`J&&7RmhLOQ1GG{ZFA5VT*r?;!wPXlLnlBWdmnU)_R|}{v%s)_NVF2LI2f#b2IB# zPbDSY6Pi9H`7qz(Pj=+gmw9zMuC;dmblO$2x)04u@Q(0r3t;yFxdt>427uGI)h1nk z7UhmM0ef~$A-{e~rANyC=rgqR>u=xjK`f*@Dts3VgU)rc=Ri5K6a=6uo>tRG7}0Bg zZr%WvRUwlU!|y9wSMV7Km>Y-HB=G+A-VPa(%*COrdnjEqg=jhW#u;$m9L!a)U#zFuC-#D^vbAP8f}+vD>+MH7 zjp2l;ojpY{Huzi?f3?m04D3x;wSG$FWBI4G|Zl&Ap@pSqzB<6CYpNefPw^8(8?p9tk#{Yj*%KmlSo&= ztRFm7l=5=n^&S&iN~KgzI)~n9x@Q9?;l*aOO6=oA8kFXWxra{7pC<)MjAP<9eG+14 z+@?3ppt-o2#%axui}IP?f)LPu)1AAYCU6oN1D<39uDUo`m%52V>uQ`n86CqD{n?M@ zAly~XJ$iaOM@tWccR&zPrh#^1DLGB{ytSO8OYD|xr`J>`b#RugZt9&|E{odn^ml2& zJ}Eaqvka&bQ{mS5#67&<;xk*Re6wUco|@O?P!F%S200<_L3&cw7VuLUkYv#u z{Qi}*D{SM#do!@+euMV<91#Sw6mgc31;EQt>}odafR(F2$|UfK3Bkl3o-9vew-rwn zeu=Z&b>T~o=%;Y4oq)rCpb3s*lSE48L-#p8DPMD+9w6{SVWmdQn+ua$ zKjrxf_w3m^{3EKmv~sp0Jh?Uu>1>U;;fU-=CSOOOU+RTgO%DLl{Rz-^0&X8xXH(rQ z;Lj*$kKbOA;%AAt+O2f49lGiFTPn=D(sV}A#Adc``>Vwb40(xvW{XoSv3RX95-K>E zpT60tokmy=UoW`s3A!wl0AKmg1z4f9u4 zD}@X*UM)dxr&i$A5JED$Ve;&v0SN~`75Fkvs+~PbMSb`2gfw2=Q`beMXIh4fRID-g znU?+&;iCU50C-h@F@0_Q$C69f{MZQn^ zBoM{oesd^7P^wOc=zMtO>$K&qr{TH-&o+38v)+W9`Wrvc>gkSvc<2*-0&Ld<$PRRg zPHG_hvUL3gp$TF@r;ukFnVH%`2V4l0jCJkU08=Z_#xQ5G9!(=KKM${)-0lmVtX$Cgb@HV+ zi~~6pNn7CFmNHUzg$E5Fhuh(Pdo9W-YCERZJ8GIX%2C4g)I@slx2x#-+?ZPLR{wrw z)`jHs4|bY=j)i92D$#i;a_WlA*YzJb)4;w}zO{JtQD*J!2XS$^XH*{kk~c}A)vB#L z733d;k@#x`V6`y6;*$VJGKZYh`pz8TIRkzl6RZt~BxuLo?TXq6EdS~Nf@#GWIK*LS zRQ0ddNr-$e_kqBQMGQjig1LB^{fnj$+mj>|^XlAxo*o4F|JL^>WE+t_c17hr+_&>S zZ6KPv*?ov-`+%iX_DY@~48W-n+{NfmpEViZv}Rhj-`H1%L)qy}Ez7HAc4aI6VbxXx z|J5K~_d8rq`z_ElQdalsHLnc%Errx#1y{dwDKFP+<;NN=-D#`*#m`)>+qE)X@8KS4 zJf3QQRi?LuacJ0f#n887S}E9txP!N6%Z`zTtdHpwJYzfZVTy2Q^yG z4-ys@W7WkkHroq1Hcq8du+8AX)0Z2Dp|gN=Ks zYt+k(E2<%t!W`C%#w-;>DvgRQs>g~xohv2;uBpDBiOX^1JjEBlHKWrIE}+G`R(bMPK^&R8 zo~tdX&_4YaxWi9)+*#e5Umb6ECCnKCZ@9~-JMw(}e%}jH{F?lrz>~1xjHY0RueN$b z4BAG>M`3)3f6W@+r8en-?6M?{CJ9K8jFy0-75sop^?>X;XvHki^KkD_p}rAZo!w!3 z^t`dD@N=i`Pe(c1U#ig#jmdL=p7i}*4aJe0NiQ05<}IF*L1hEY*$x7k;q{hBY`j6n zeAMlsFL3i7w9vZ<0XLHv=?(6lV=}^MAWF9{IFtk54SH|O0OVi3KsRtO5$xUfkm+hf z6&knSNG+Jn=E;1Yl){06{}9e(LfXd7Tu@ju?pyJ*?J-3hhjAB~7cq8!Y3*{)hOz#Y z*Rr!0zJUlc%H3bM`eT+L_y^81Bb1vujmVepI4>kg#>U;6238?UNRFo0cbf1I(c(Am z^OuXOa}wNcLB|gp%$9GxI* z>)qPsFSqBi1D#I}HeBp~d2Wx`M1f{1!Q^tenHv-5$LAe#cbfdrzo$PCiXx0Idc}>3 zIa09%zef5>G6ns{hcMU=c}Cx#-gN3U*`naqX6;-m{OjuWLq)rwcFp_D1USVuhUplH z`MU1B()-EnU58qP;^{A&zmt`9R;8b{;e9{jmEOr>IffTv{vh}HjDJLr)d zHPr`*9P0kpLdNf2N(=u))&a)^`8#9_e=BN`mkB2Kaf|5a%+ zB$6wQWHAtqX9VN#^@f4`Ti)rXM5GqjVcBCSu~3O}`%WRn2)rPfPfQ8{Z~&SfH0%>m z=l?cDIUpat51Y$>tuWSxT`ad#P@=d^kKq-Zc7%OGh<3L`0V7C^@NB;hMJaA?)6Ooy zW_zIP>31~-M?nKAZwQ9Q{2EKvAaZ4p) zrS6WA?yEVk=FKl9sp*5~|)Yk$Mz@d)=XWohELmCLV8UfO>e!tvH@xs$t3aFhR2*}O&s zGcS%l$KW=Z5q36sRFlGj^_C#11i-BZzVK zrE&;&;((NYdO3Jjk)csXWpFsU^iP2eei>7*%L&)(!My4(T_{qi>W(;ng!q`pUII`J z>m9@uKiH+eiwzTJUEuOTCFt!nk|sCY0OepWyp;xIp=)!?p(ZbwP?XSw@?0X3$RFAI z@Bn1;>=uT<^Px$*i1A++M@S~)Wk#BJil!kqcHc;Uw*i;5@e0~L-yuwv;o29amb`+`^BncBZ@ehes{X2oeyvH_q{9$Frg|SmyuMkAco#14=696B6K_VRffIl8cJQC+AI*4TiwMG8v+DoZuJ~O31ushNpemo0MN|D)r9&BlsST^!@T6lOO3nyxftSlKlg2YH><%&yQl8q7xo974&Bx01M>*F_EI%(5W1r$-}WF zqzhd29<}l3t=AF6>3RTsqD@2)lUA$Q+&r#IXo8~N?6%sD|9BTbzioLc!-?er>=`DR zbZRzy3Tr4>1nFKXqz?`1XCX2;e!t3q5w!_%m!lZoov1~H4Geg&qkMm|mK5TN4 z4XE)??*ED!zv)21AQ7^XYSzOxe{tNVF&ph7QOf1&>?RGRTpx_j$k^&x3{9hdfpZQ6 zqDkzHsumd~qm@%ve_y}jgN`V*O7sMTXU!Oovm4>ZWE*Fg@=rKgkYSYL;kmWy_{2c* zy(IrPJJrCi^G&)V69%cJ{|a6r}A_;;V~L@NOM{tJZuqyo(d0@=As zYdJhFDbLr~-c@SnYC4Knx3<02-`+FM@Dej$C@;wAj$oWMd4IXCh^0Ay@r-@-E{)YR z*`-j=C(dhqG(wfN#Z<-wrO$%)S;f%CjWRrFWDj;Zz$-Fna+QhA@OZq!`tHweey73V zerK-Reu-|jP(Ozj@Xa{z|2a|r4CTz=e_AA+&KGIsB_cvu8g?9R=ai?w1ni%{mTJUl z7V{+HbZQ)wmqiXWWJ4*R1E%+7D$k|b3SD9G--2$g(hQrQk> zmMSKu6t_37)6(<4P{{Q*;;`Fn@?MSaPw>98w}v+SN}f@KefRn|FGK*+uG&c_{Xdq{zg`-wulRB51S-aX_NuFR-mTR zYBmcI2}?&!)U10E%#|y0M{AI>|HuqF@BiB=;?Sg{{r~;S{|rFnzwsKe;V_WEN#`>& zBwt|q*4@YWMv8yAM6f(*gWlbH!z!zb78&uA{yij;Fo>7aP*n^^LMsABH=k+ zZ;ye~1rieB>7_Vgw5X}yWQ3gc|2n2-jY_)z+2Uh@bo`rtez$Jk8A-$FIywl&Fugiq zGQ7b$my(7*o{PWd=z>A)mrfP$czL4I^jf=}EfT~zg7o@x0QFll&R@IP4p zKf}Y1&cCbkjkvc1>*nhS&O+B3VMdCN?XjIlb+nn)bESELE_HOb zVjXJBX0zFU>6MCt7$^O|cZO>Lb3~)h|JJ05#iH0{|7SO*!T%pGTFJWo#_m1H{2c!1 zaQ@<2lvbqwJkL1vaZ7)|hiXHUXv1(g@z?QOk(f;n_J;t|{W~dNYkJoI8)o`H8n}@D z2}}{Yo!2@7pP6Mx9MgsLHbxq0HdE0e`3;I83AgQkKO5sutt12ZrpiOBUH|1z)eQ!~ zAv(31@;5X|5l{br;@AJb(B%C$G&8OqY>xsrZ~79TO!|uN4M2zEX~q0uVYj0v#|`2H zMIxp-LlZ^@l(ijM_J-18U;s){K5wjlfr0t;zf*+&M+4!%M=D|B-LA`Dky3%LXKcyL zu8@C!I5bu@F{9+U;_-XS?CH`yMcaE<;LQg-4lY<^k0a!Ln)ti!;G!IyhX12W|3nS~ z{U>E1^a~qBQEZB_4GOe;en%7%QQDAaG6zdDYMiYWlh>_;|C1;>!7`o^G&0yF;3 z`DKcg(|_v&{tpHuyni1015)-q9!c2oxSf1|(A#Phgf%JvG=FTd@i$tn7k?N)I=L%5 z{a>#9r2!oKpI!~+-^9!)z~ll51EBxTX4#)|^NP439|2W8qd!;j5RxJ_`(M-CSO1-0 z{XZH||3w+0^lrbcgiSwo1Uw$`51F1>QqdleZaNn$moIsQ#e$m3_0`T<=kIZQ?z)$M z`2T72Q2!e}DH)Y-z8?#?;NvfG_%IX`%jQg5tWeczd?RUdSNVbjcw9 zU9zE2m=1mz0Ofc|d9(OhvsTjPE^Op~y~JPGS*I|<_@DBj?mw^tNNi2~JpxXKO5U%J z`Bz_TGnC*m|4&Jt?;n5~n$2RJX$zi*g_W(^pV-w^Q~s9#17<#Y#S&=BmP2Hdpxzq z6F{@67Yeht8~I<(9@iU!giY~(wgq&zYxf&72N_-@+Qt)Z1bI!I;bk*__*^&wUs}Fk zET8x7foMBl&)<&Obh~iSx?M~Ke090pPikmy@$ccr8Ee8~p$W1_(S4?qojkGI$>y|w9lI^yKltJd z3q-4$evwLo4nx3Ho zbCVBnpAkuY{m?;wvWSnIQpubY?)4VGocXO3(yd$0l-$@C1o>O5)p5W-JSF0Cm3@Op z43DSUaP;?>a4O&zbdU~z{kO`?-iCK>_xqW?(2n1jL!dWtSDrLWus7GEOEnI=gSJOx z4JIkpXqGo|-}0X-s30P9qat(J5v5@p+4FSAo=*<<(qE6VKd3*)Qhk14rI}#? z$N$FbB>Bq#OsjDG%gAPzph%&90mflxmtYlRjmHtF8~FfORr;fUDz9`QW0$#p6%YLT zKL|1PAFB@54PMau28w_A4}6PS9WH2U9r8=Wmq%@#Zq&qCH6Qg)>XHh5s%U5R(aKapH+2U|StX?)PV{n7esl*UNojPjGsE2-I1bDPKO zaVAiq0YIZ(Fj4P+o?c0A7^f?!l{H|_ZR9!HfeH=MP*3F@jYNjsR?79~G^yI96E53J zC!Y6p&4*IC80m404JPy_<%m|5 z0IJU3aO8)<#F04I#aJrcOw`lOF;eLD zq@rW$&+EE2>dcf1wgjPs*CCo@U1{ZV3CGfyXo8fT%zZnDVhRc^{6+RslxUKtL)SQ0 z8n;Dsk$-dUBjQMVJXa__i@DmR4~SO%YL{!bKGlGDX8CYZ$M3`D(=)AX`SM%c5Izu5 zewr!A9nm&_=rNkp(c4^HMLt0+bIOqLBo zA_>;M0oE+`ps8v__ov70K9vjGD5KYA#j5nJz_+&Bgr}#bbJF&CDOXELk&KJKVoMbJ zKUABnu8+K)7f9rT)E`0MLdtplsWl`sELOgP;ZIf=SAYEiCHSrJzFNGmF{oOf%zCwh{mz80;pC9g4OU zYJ@){%d{<2h&?_ZAB9c0ld>lXE?l%uXoL;X<@%)tlJPp|6*QU*i5#X}Zwxxll9?^! ze%KTNwpACfJaI=(Lln|+ZPJ6daDL_bIxJ+Tg#*XP7IX{p0QDE+#c=1w{Mb-pN&lMhev+?)J_4f{#0_BRb7>Of?! zINAsfY?7~7x#8;3$fN4#g}TJH7d;Gr-QucXd7Bq5Z)6+u7D+OQWwc5dm}+QIB7l#7 zZ=Xc`OrJLJ7k1aVXx#7+1+Oz7pJjRra;gr8<5o5aqu<&Jo2IN0JFd>ueax}szhllr@1elWv1a38Y)Mjg31V;*yHuoH>MLy3 zZ)ibLQ*W>L8w!mo-?P3xK>4^{ovE`mCePip0Mlho@`oHPsRvgHcgc({{xT^s`M-&T z$61+*R-x`mCzh^|GaIb~ zpDmABv}4FY`P*&vjJsxq6NZ^kQ%5~yKLWRFis}omKCps*`@}O!D$H-SF|6X{2|fV^ z0Q?nVI{;pwc$@0x*s!@+QYVuyd8mYjQ{cT^oW+jKI*=sn4?`g>?3zc)v!?w@eB zmD%&`Ee;L5(%_ZVXD@{?AKhWp|BXVgcpc^_#6T)&KG%}&%8v1$YY6L zx#Q@xldFwK_LChO`e3-rEmSXZstv`P0&Y9rUnAGuz3}kZZK41f9D?_`Z6e#ToX&u5 zqdEfqWj3p2No_oKVo8Ctd~L7SxO0F##@n^Wzi#V@|ip6j^*1K4_WIM#0 zW!0o3WyWZrAj6Yiz1}KjDicU<`fMT5d^@kcH(@S>Gx>5j` zG?rn1MW@sHI&&LXS-0MCG2&qTbTcg_bMLk~&$@%4JN%&sOJjMxw4^Ontt=Dpq0T2# zxg}CqJ*ihNn2g8ehGW@6ZwhkE6sd+d3>+gjTAa@E(y}82ZtQ;R-H%^ky&YQ815a*i zB#*Y!&*&^>E6|Miq%(K~KV&g#oO-f1HPIkf~nhYDL1XYirmyd~cxgbU%4QvK*Dnf)D>C zW6rkLvIc)VBRis#xXU=3hTk33zJlD^A8xNez+dmlVy$hByPf3a{QcDaN&LpNVciaY zE>Aimo)WgEB#F_e^7>mwQ#e6sI&BcEQ5?Oo^Ub1Iv&-c-?${lWDdU6eRy_;*53evl zIZ^zftz>ME8Z(eAAFt3#j%JT$rn!_rWE0@N4C0v5IlMdFV}fOBd5~U(zj)oR-J`Ue zc5@jRP95Ym?qhRW%W^YamaTnrq8tH#0P8K5gu_$nqbo`ji-hzCK0DfpcLY^UE!Um& zHE@0%pvFvCw|B<=4)A?s>OiS)C)>Rj@T%JPMf~B(PB@Ur7JSKktXL#RVzA^mQnSul zQXO_V?c;wjnUG!nMm+rfM=?HD=dJk!U(+74MWFs1^EDahf)>BPz@ByYv1%fJF(ihR z$I+ym;OK&RIA5(3N!I9X2#idi^G#$qxaeIp4)lEiQ8+PE0%KZ48O=0 z*!H+ncJqj#Ee6F>qMXN^4PO&uJ?mb^Qs%rG(k)KszwB?q^Z@=%`t4jzU}J(eRh&c#9O3qSpGZe|--ad?so}0U(kV@D zBEvDf3%ejiy=_AIS~AykXgbcXX`c;)RwfJd5yrd~iSynC|Ll$Ig%XAdQcic%Oz-UL zX>zK#G_#}TqvC?eNYK^AJ+{&K4jDolZZXgEHeE-MEAo7QDNkOG-Wb+N0b-EjK4lz% zt)36G0k*qY0Z|~FKJh>M+WcdWezh^Z2fll2H9KtN*RTEUt8T94on{gZJBMSXH_nX z2=<2Kvlk$L@0~St9Mw}NcZwBjIN%95WFT-3LZq^hbI6d;v(CVeC7}?HF`zK^&~Z9d zCUJRzA(*AON|E}S7f8gJTHGxwv=$o~+|twyPQ0h>aI3l?h1T! zAGT8`l~#bhuh%&9hGR4FN7I~g>D#OeSyUf|Ha2sAdV)9~$J)ktVlX>9mwM)g2Li7gic$8YlS0i?v7-M{MD#!QhaHuRdy`=bCtAI33S;|eSAnLvu&^g zWn-NIpA{ONJ==YENMrCXp6aM9gT)hxBQogY(A(OCXnTtq=g>mIK8PiAO3xN%4}h^| z5c{Ul(O@Oa5l69xcs}9jneKLTPR9l}b*Bn{kzb@1YJGOscKfo!>FKyh$YMD!j6zVl z<;~1`A8uNX^Xr3dSk5y$=JuQ7HvM3RW5QEfAOq{IhB&V^*4W!VudtTb@F$zDl-Cf? z6&XLF8D#wGfs%i6IJz0AI^RGJ-EMno-CjhWEq=u0?}=O^ZCQEX|6zAj(P7Z$X@3rX z+TA-!Nk|%o!>xeyawj)3f3|0S!-gq+n7u96m~)0pZxcl9aCB~ixV=>QwW86gKAZ0N z_bQwp1UDy0k=yJl1`1DEJ8kP#AqT( z;!h>v!+P?JyK@b?y?grLvh>F3m}a+slg-`>q^~D;1PMnZiDI-j{mifhH%RdYbCZuB zz1v}KvlLcs7eQkFhz#tKwo{t^gg@2hv>IP1o2)&B*ITOQhhY_eS!ZDE zLHb_pIP@)o>2uDam2Cd(K8YEahF{PYA%y|xo`h%t==+cEy6tC=n$By@Zn4eLcOgf9 z0=c;m&wDQ)E`0m8WnlLwqImQ)mUk(hp8{DcHhv~b@kKE@S#6lO>p;NG`G>|OSQ3c} zpDgr6Wh1q(L+Xrb*O~n=TP*y4?s@l^;DRSXzmL!GqG-FBZC83QPoSmm97dJlRgTAE zHOm(r;P^fjVXDwk(l6SHbn?aN7SUiqm5G!y;$G2Ba=bvYwgNuGY>OR1##Y7O1mmxk9C4?S?czhbcLjag`%Ng>1#I+hoh^@Qsv{F`U?!L+$Q29{o zGRMxJr`@47WNtwD4*Q1%IIOJhwe%@U(*Viz!%kyzh#16?Jf2X(2$m-oHW?)LM#q;pyGGo9#`SuS2k(ds>y(}MoO^wfLDeUWJE$~AB%R-lil=wf+=E!4Hxxn~IK$tV z?RCpn$8{-45WHD`D(L%Krew)6*kh>F!5%eRW=>XoEi{^*dD~Wp>lUZJw^8y6TV0;p z-HKBFZw*~C>3>gg*Lix%nOwORo1Tb*1;Ho;xHHKd&pHpH(iKefrXoQggTe6y^V>YB zTSQ2w@Tlh#;N+XmtV?0gdWeuor-e~F=^=ZLmr`5_6|v-hxaC;m6&{8+u59<}WbGLp zcVNdN@dSxZ_S#==(>Z9EM9Jb5E*fNT*~Xp;59}@&Pex_sd>fL8qeL?tivDHQd?}0t z*$x4|j6)hiUs`ZF_swT|Jr`7t`11?UeJjP58G9?&`hb>BWE$F8eF^VNwxtJ zSj)IvZ4ZCTW(^$)83@7<(|vY3>+N-(shWiWjPz1}Tn@?SkC@zE`H4+gHuRY+G~X?M z@exI~kDjLW7oa&AAj0fs4GDsz6qAtcJTGVy7^ zx?@Iv{n}_5CH19IWch|&K}}Y8L7t zm*Oz!u6!=N0V#@-a}wr_Q!h%P6!SlzCF`}mq_`jzxCQeV?FK1>iI`_w3g;$nc8?|= zUr$>9ob(GHp!1K1oppp}ea2{Kkq9F>kL{QvkgR{c!66=A1HBbB32*t4T0ACk@I&2y zhEMXPY7(wUru?%Mif5K0r|-fy1b}ygYdxyvlZgazg!_4W%6S`GTA@2{(A%BgX^stc zx#r6JY`GxD(Rg!GP#`63Y+1dfx@^mxrU8yG5K1T}XCPw4t?--A5DQYxslj`Jc2Ikk zey)~|Amkn<_CX~`SrE;nIJ^B*ZXpYQRr1rklWQf9J>`g!^D$ey)Ol+k@tc{fQdh*M9$6@8m8uh{M9mer9B==`}lKGm50AIGqiq4SXoRFw7 z#}Zk8+40=yq*ng!6l#~MO%sI!`0}j1EU9>s4J&eF{X&0@z!vPX96*?mD?pgooLA}( zOA*R}ppb@Pt!6YHCYR36w(Pxs@2r8*7m>*!WhhW6%Qf9wxR?{toEm|dzD~gO(klny zcei^!YY^K03dH=fmm|E@vZ9xt@$2)h{F2V%IpO z2~VrmxNHDMwglY49ys8n?`t0RTqym2fugdaOI2xBu?AhEbMhvQ=We1IsMF6iAZrZ+wZcR*=PBPKfjo0s(q_?{}v6pKp5-;I-J6}22Q`z~{7A_6Dp`5ogF{iLm zZcVS}!LHMZv^0^WVKhX4J+I(5--9Y(Ukhj~rDJ}DcOx;fu#_q(07&MUX&EKnD z@9nZCyApIGte~sR9Nf6?m$L1B1#NXE^4sR#_sSbq__ZTk3qsg>q?f(n*o}O7_!_i! zKRvU(P=-(NqMwfB@P{7HcuqC&JUyPh4y2z_(QJ>;7Av!ScQpooXlRZ#nfGQwXlfha zaTt^!gIxvO80x=M&lA{o3c!d)60&lNeGp|!Gujxo>f%LF<>d9b@tDNzLgI`fnp)4C z8e2c8bDkK@G1*kS)J4hWo?;{zVm8GArPhgBNXVhMHj(@zTNO_h{dKe2fE>GMGKmD~ z<7)(r<%k(P^v;`qtX>(cTH>j}QiyYr`n}e#TYNvTv5JKt28v}gMP=Clx7$-v1YurN zj4%%)DO7yOrGrmm=-BOR{dyl7EZfbIpul8);?_2$9lgSMkb|ygk*h{Uzge+#sHZwa zjQ|ustiFginl>t<&ie+y))` z&f+4TEH(tYI`e1wE15Dbtm$iEJH4ZtvM@k0R^937Ng+JKs4pNx5?f#e!}PsY6n&3$ zr)IFewQF>n(<^>AkpEuMtA#yZT=fPuLQ$EUl`B7BT%`Zpc*5!S{YKsjIlb<+yT)W8 zZu4Fo<=?b_Y5o5#ZQAb|9&j&aTP^MEedCIM;(u>u zBgd0n1Vic+PpkQZNH{!Zs8e{mrDXk%3#MecRND(BZ9j+Q_(!c<=a*r7yp>@DKoi(V zhu{Flt^ip2e^}CYf16?f49{<$t_I$hS`cg~LJT#PFSTkAZ?XA^$q8k%cZorx`5DQf zp$mWh{HDrSNEn?19Yh{&uP_wNBpeOim0GjSUGdIyx6H|D+u(fvNpZt@e02Oq!TCL{ zq@-kXy?ld-yRcryPp15A6F?ae=ch< zPrcuV{`T@tlNca_pz?lq1#Q2SKP)vcZ7LfFy|Hj8NS_3*D@YJOUP-77es8nZd>^-b zLf1TTm$iF7*DK`0gkm~BQkKT`k0*(+=Hj=z*sE6*lh20@KiBKzOum%Ox{^YUz*wVO(U>MH*|{Sdw%17w@GuD>breNpj+Z zR3qpg%O8DWN^D6^8hkNfwzK4 z%LA1MkO5=J^nuvCw#72*Vv5B-Z&Sb57!Uq_>QGs6&mM*|?Kd4xZ-W}#74(KYm-P<+ zMt&EOMo*G^7|BuN*4GVf?&!AhS@)8zdxGbD{Ze;F`; zzw4<{2M^bf! z#6JJZGW()eEi)joK)(hN+}qRF&Cx$^)}tCx_`S+@M9ED!Y+|wSXO!ob&w?ddL%r1+ zKO*<4(v^25`fa^cnp##nF6q(~f1)(f8QL|a3i_wM>`jB>vcp?&dpHYxKA3F%!1}C5 z_Ai#YQlPGUEgfhN_#Tf%ZjtwlpkKe?sY=g|g5? zo1cE@C%Is~`@tS6QKnc^D0o+%+^DnsC?o?wPQObod?Nqjd6O~JkncCM@=r^BGu@2w z9Jt{8G~o40Ap>s12Jf&W;nRllWhPzC`19WYvGF4d6l2FtMOB*0Hgd1xJ&=f#GnyID zbGEe$b=_s27sim{c}}??fA%mXBTVu;9$Zf(oJp}Oiv>1H%DlR-< znL#>r zKsZl=fXt|LmO;gxFTHX4*U~i}y3*5}QQ5nd`c^!MbeLwzp1L#Fxp*65A!mI2)3IA5N!dRxAq&1#WEyUMZ^%y&n+#>-Zc{lw0 z2V$HJ7b&~AbSsib$&jp(m+8scOj#u0B z);G@Ar3Q+Je^3^J{0DHUfz(F{IYy%)qW6%5^3+TOg)se|j}kWIe<+oThsXz8Gz`&;|6=p~d(9e4dJ9dTF@{XNgZAtH1|!qaeJ z7p{}OMr_^Gs$H;IEXT_=sc{b#=qIE{aMfFFpoWx;%uERRsn`v zzdpbHxcrRrRzueg7?RJ-4SO$d?Zb6_zd5;}K;(HbAy1QtudtOWsJq~>eUcQ9*|(Ip z>LAh<_ZP3jd%pO9SXvV6u4%^fXu}fW-mEK?R!KB0$k#+SBj?*BlRo=kEkhSbkj_ zKf8K}9x@|}?={oUQ7^UnIIXrvD865>#!OdB4aCM2jB^CYZj3hh!iX+6SWaCx`EWbRA*FsA*jUtYCsW^la zSM(xEuEM#FFK(X)IC$`21AmkAf4;VDxW`!js2~$XeKII`YnU2}=& z*LXi-T@^fPE@iHD^ux%J#G4^IFaJdG_*GX102n z40Z;83>)_>xzN8iIU(*#e-X*XU&^Mk%UgW`TG+TWnxP{IF&PP zLv`$1Fa)U%;;}bP>)>L8Q1GO78ACb!uRe`)9=u1zi8-dJU0reEf8W)dH|*yk!Ud5F zNtvwWu|8E|9!jlF)p4PG!Ub26?6{#G{Bxd_6o&JE;k_Y(BL13i96egKefIAj^RUzw zZ2aMe~5ehhwvP9%dKwp8~7pShpPsE`)!zvRjyj2*okv_D`WaGNjCuI zZHwK{eIkw}FX{p#9bac=e-8Se;43Hpyr7$#mzqU3=K{PiH|VVgp= zSUCmSWS<*Xwe#z6+~zbC?1&0M8RrvHhptImu@;=!5P2XUe+O**d=|q(1%G3&o0u1< zU1Aa{YLhrHQ6b={so6KPd7#etToE3_Yg^2GQ_^%gc?$(K6TN%qF~PZB;?%X#fLs-? zgD_OZxDB5wx9CG~^@jgF`0kX9=}~EA{A9 zVIoaa;GT;JN@`zR$xKX}3fC(N402`udfTHfmId*M1))p>kDV6jNH?w~T% z>pK%@Eb<}dC>w8mF)PVOK$GS<{6K+TszR+;Nl;OpeuAapWZU z$c%J0tRR>nK9ov=L8S%?l=?d>!fF{XBdtuY_Qq;d9Z5M0f&2bs8 z(5Xz{63*S&s&dA)nrEufN_Sxa0*?)1rJ8OVf2zRzaf*3&BkbP8IJEnLT&@0PmY#0| zf1`_kzZOd35NZuB>21z7;9C^QNGC?rcU(L9qy6Zc2oD01F|8VXqr~qd@I~pe(=2C~ z;wPS7rDXMi9u-wVnYTHC*>(@6kIw`CY&`NYLP*y=%;DSKyF!+Hxjl+LLYG1Eb3a^D zf4b-v>s2M4ufD}<$ZRfi=@KVeQOnBY)qy+?u<6kyisM?b>Yj^7kwPQ6+hqR^+3;)L zmT3Mqqv$Gy&72wG`+eHz0AcqpSDDTF`CwhG!V*9Pv<0p6Vw0!9fgyVP)Z^Z#Zh{G4 z*P1j{D7T8???#^&T(#rQrX4`zdv*0)f0CnqH@02u)NW&kmMBKovb&g-t$Q#eePT^1 z-Q?f)YHc9uoo=>ZB8acB&DUWU?DL0}q7z*7xUARGk}ICSP_{&5MlDBDp;m4wHLiyoRdCgoJ90}_}h1yT2l$BwmYRFlcuUCPb>STjy!r_-2 zbY?r(iZ$S8WvA?Qqy{fZb<2Rdf4Wr;X8(+BVVV}zg)a7b;>QI%We5;sUfmS}#M)+1 zaFtfW1|$XyMdCP;+FD)cZL*gitW_nia%h$t5Bg~jB3X*_l&0x zQ~CV5>M_N_)M~R+i~)ctY^|{S2f3!2`cUtGsd}7B0I?CYTdUJu!NSqsf1XNTypTqZN;TAh1N5}h7 zsSL|*$nz%}qFrLi-0?_ef0D=;velkm-%zl_!kDr-A@RQRU!FnDFPqQfrZ=vsO6ZY} zYD3lW`6n^7t2gz=p)QrB&->-HbZGSU#b9b^NXk2n4;Nt9FEFS?`x83p}f6w>+BvoF5m_{y@ z95>WpkrRm`;W5-7+a|$j!|68Ly~?-OI!;-}_++grHplhr@Mf}fEhUcPVQBy1dw*O; z5E=JSKkXWNnratW|2Jl(DalLN80AW=uM!DN5}dEH>skQA14jb1 zG{h6&(Xtq!9MY_{e=sy+J4L`vi?c%vARYPWn359QRv#{q?V zF_KO}LZv~!K;&>lQFr)preA}loT%|dF7FOj83UI*%nRWg4zXvnLo@XX zGBk4%Ky)aDe+a&#`S9?%*?O*k%W|br=rfRr`A=Nww`YzHU%>|Ufc9^m z#Df#RzY+I8z$Ziz7o9SjgPsIj5b*0!4nwGlaEnOT`)ROF2^;a>R; zP8F@dN4%t=wTAOQUo9pQ3l_V8k&`P_2(daT(mjEN%?&-Dp_s`XLh9N`N?K{XyEs2X z_SwxMf3WzrP>#;=&WC%qSqIW{$I(efksEZh;(hfYu&mwrFO_JtYI8gDg!!$IVwE{mYvVmYGj&+h0~<1{@j2p)>`9fBR}y)e}-cUiQ`A~l1XLH z7>=hZ*Iv@z97ft(=9?mCO}3zLyEo(WveYl4!;!4FP~R>g zf8{Y51Vm}9jA5z3J3BQ5ZbdA-M?HQRDo{}+TEl?8TetQ*X>}F5%?&gD2H7lpzd=wU z9nM}QrT_0zHK@$r>p$KCka?G_nT|S~;7{?wg%q0Fs2Em9?HJ;#Hs&23hpm6Jz$r-3 zShk&XKdyc_uiDHGAh%pi&upX3=bi9Ie=eM@fz+{W@8hMG^9K%-Gr_^ujuur-L9W?Y zb47E!&G123SqR6GA9b(&`8C`U7ITH-+f4rar z`S$c-&Q+ufCxB(uE}po#rm>$HO*(lV1b*^~{-;L$Cpa-d6X6A*#V?$_HjbxLo!LNF z#!cMNMuxsz6-A1)sXFs+?#u7EPJ!qncjwit8s3qtcSLa%9H5zZuVjkEeR1Ogu~Wi& zVX;_dD@WEQmI1l101jlOTT{o`e=2QEWGcQDU7L{+C5EipG)Immjuz`1Ds&n$|KwAW#JyQ4dzP_Dw( z&TVpy@nIwD$7-t{VpB4>^+j0lO~41r`VEIaDA*iczU1pV8ts9VY8K0+{|4YpL5H|N zh7=caV0e48qnRudzCV?Ee;(jCN9U69U2V<_#fMBxbn`DbsjM$)QN+;@#zW3>SS;#p ze>Oy;ET#3TV8?B7G8+)EzN?hXq+~HwWPMJrlabW}nMYj&oRd++-%UD?d7T=iho3$l zRK@zs>=?--AGHZ^^iYyR2CL8o(SkuF#)oAl)+90+390)O{!mmxf6^f-UQoeSj^gVs zHYL{N*heMkJNM6KTW!7SZ>I9aIHwJd~pcdQqS$bU0 zP#Z|eOtxc&MhcYorMSdM0!u&8vJgg^WELW8~2 zm${aM*VG&MxP9w_iNlYlJkcorK8T-nhUl~F@Z9l@EF;PI%Q-GtyZD#NHxY!Rr-mhJ z?QxUHamuM=>AK&J7J`P78TdC`|07e+asgJ)Im?4|jOYJ?b2% ziFoffyw16Dtd=rryY8a-UoM2|b`|Ta<7jvBpZksfBO<-r#|{Oe-ByY?a}he6ZA1h$ z>b-CjO(M_z_BiI-?e|J6Xa4XNr#=?p_60yGv4GDWabK%QftzxAY6n%a7xkwZSq%%; zaUJF2MA^K{e`Bo=q4{62B6E~%sV{@R0b{#CGtJIl1D2Rku&HiKq%3 zuvQPvXf%qi`eEyuChvylNp>FTh{;8kJdj(q0Ug2HOO~{`VilY>jv>@mi?hdv*M@YV z&V%nUz7mkXmEd^xGPOylD5V+-kv---KdjAN`v9$Xe|Q-w^EEDW-NjN}@GqYI52pS{ zOy`@Mp{;-#GM%-R>_(Kf3^<{m8^nVJ>gv?%AJh<`!hc&VEMQ6}R0tsdO|{=2&nG8^ z6B25Hps$L!zC21inQyi-$jmTad4xSSUd4!P?@Uo{SqEewQXFN~l_vUQXLQ#L-z{Vt z4Dck(f49_IsBW7`LFzfevHz#JA2|LGN@`mRcz)r8YV`br+FyO+{iaL zHs(}8C!;M4;M6D-Oz@)yh6oA&ZsuG)YK@3*R&nf^SQ-7*&bOsx+7@MKNdD@zC#HP` z8xT2jQ+2Cla+-R)b8LcSyJ3(X>G=E`gC*|7f3_1R!KCckI`*2Q7yG}g{POydCC0FZ z>{dvEq9?#Sh!~i@#?wolir|}ztND-abDfiG0Dv2Jn0R>CKo?F%CW+o!fu^gkweXR~ zeEYmr!#H1!%jnr+EmsAx`7&~{P3Aeh-r-jlS@rmGQkW^9Y*(BRO`%LGaxt7GS{bB$ ze>0z_Db`R_I<4EaPTl2x&kZKjJLPQgGyW%soiTxvwsxN#) z#+O@YwGZi*-NO@d#W0(_G|Etg;BbGVet3>EyhNKX$t74q&aH|^}ETf03U39F6{GtL@*i0(( z-#6qZX4NYz=t|`BJn)R35PY7-&UIEX2k0M~xmJ^x>r74WlJ{$9xe@n{$|cb$e>M46 z-9Yh?I^UU9$>oY@tvr3-%H`D#RItb9PCLJPeek}V$?(bMB-5S=wM_J6NX==cH<7VEATztbe$DEg!XU1Un3nu809yx1LGG)v5e+sXVIA1jw?RKiDv+a?XRNmcg2xW_^Wb+<=I+ULK z@H7$m!LbGEvAf3&bMhh8e<`8u@P(3XqC5fE;l8nl_1|hUl4NbBW!R6fFwH99%>CS* ztP+%=2VnZ{rVL2LC7J<6@ZF!0%2IfzN(_heA)T?6;g+FU$yhDds>5-tZi?dDD%V+7 zY`9KW>bVIlqS(S`zT(k11YbXL1E9bNT3WSe|ILPOq$;DPwKDHSDU-bYc%nd(1;uB~KuWO=)vAX1 zEZWq3Teb{EA1gYJ#fIH`$;FQ~Iz{U!8G#9MfAx#6Xg?A^#9<6Q2z40wF>iCkvO-+v zTBa~)nd%lnfx2eZen>o=$rCC2sU>4Ai9x5mUcp9wy>EGKTTL76vbifQzPmrc^P1Ok zm8ZT|MSl5!d5)Her=rwv^&F4;eqnC4t^c(vOd~}pBfV@&wPQ}pA9V8at4pwvK;c38 ze<|mbsoTd2^%pe5ZKE;oP%$z9UD@hJIDqcOC@=PMWpJ zHPVI)dUszmmH^&jJ4R!f9x)Of#H~)m9$>_Y3>m()r@Z@YVbOeqBREbOhya-@G#6NO~#>PhxzvKp{p$u;DwsDHHD!%a|F_8x~Xlp z!M*J^N#Qg~c^Vz^^)I$vKp4Q={;Q$dhQp?}dAmZ+VFMrP4A$)IRmbL}fqaL8f8ScY ze)PNPYG9A7u8TxM+7zSKUTr>}fE-p$)fcKL0rr;cRqT)8LTU^FUh(suioIAy5{6KQn7&> zmv-cL=!B8yJ!xd&G=CE<(!u(9&RNwGUGcMQZDXoh0n$+)apo4wcj*FwL{x09NmShK|IB#F7;6o zhL+n9kz$VE(WY!Az;v^Xj0;J@BPJY{f|aNRl5>mYJ-&1)YPD#t_5F1*e>ZY)tQPQr znP;{awo+fk#!ND3?fK6Mvm_`24NIRj`cMc=+@!RT?+oTcP0O z#!>u}9v@2xN+Zjfi@*_5f97ij<$@E(6#;e`Mj54%z*dIwh0Ysro>Q|<h;T z?x#?@uW_EgJD9b|4+D}yu$VBawt~*jd`GaS5KlFu4>)Jr&NR4@;iSp6H6!5?%UhgI z(OuVMH0rGqf9K9UUUkFGk0|sPXSP>flA9Mo-WQKZud8Pof5aju=-II2I;1CI=-Fxk zHdgihc%WAk=mZAGC^SFv>%v#xo?m6}W(MC{HMfG;en)phnU0@eCMb%$aC+>(D3xcJ zVGpv4fEd<9>_A@nC1E$VqzSG7U}IJ#ZYv~%6`P7{Kj-)D7>urXL`W{Y<7{Mwhx!71 zll9`3DR*Vke}#P{8~J;3I-0IPe*z|%J~FE^Xg$xta8NVme^V%*xQF$LD0NPk>clEW zY@+sn)*XaereZ3ANBF3Ff08Q({Sb&P7}dlEmPS5Nw>!%JnOcx>?B&a4{#ahgk3S;L zx5p@L&>BzTb=;^*^@gDko2m*wKmAbFXW1yJx-3&sf5V67-MO8`3JO+)ACn@(;wxMI z9JO>7R}XT7U+Bk4nsw~39?QOSK0bv)!sp|ao=61TSzxeat+>F-h6^C|e{{q!oD4+1#RxxMm zE7_Rcf9X<|7xwcL1}wuFZJLN)cKx_mh2ZeO-XWJSOG_ivxceeGejb&GXHVyU$5H5* z+Hh!jx0~2MH`P5wmhIhMxM!!`MBF~ z)b~r2xrukhmDezat^b{6qM~dg?gZ=7Yh99)ytN&@0$BLEp;)Xg!d(Mi`vQM|hXjLq zg+YlAt;1QZVBn#O*!{IyfC)FzsKnli)D?_29l1X}7dt=`d2$>&WzOCT4Yb>%wg zf4N_y>-2W@ac^iB_8(bUpNp4`_EUK_G2w3;xY8>aq?+RD#f_I8EaEJ`7Cq_+O;8d^ z6~x-Bt+b0TyaHgr`u!onTSix4-B9LQOQ7_S-R$5#?8{Dik$%4VbbNwimG6ZW{6`8N z77WMBqV4(n+y{i%*&glhLg(HW(N90mfABYn<)eq}uKkQSG0+EixW(d2^3*NK6`CQ( zl9MC6+Sk!YvWHKCJU~y4We@?kX z%RBkgk3gTFA}aJcjC90TW(w8bU_Cdew9NMOyHzqzoc8nd0S#Li=U7T>*|a(y_B2xH zu9-GJ9~w28_@*{fVJ$$~H974J(*0oi%`bXLWu2kvAC8co=FbWPk-2q2g?KCyglK_8 z_HejLYZeO&iO^&7no4hMEH?<*(1od=RJDgd=8<5^~s5Na<#PI z<4Ka?DV&sOf9&@fW0gbve~_I}Yu~u^h7o1uJAbE3uq(QuG>YNMTs+>}Sq4t}`HzJ0 zqUB5a4+-NqQaSXVP-pg{i*mHk5Tr3{$E{TB;)?khns;x)b&jd^WqN`&bI?nyL9^EB z=9_I2QE0{r;8EFfd8R*HV~3&8vErYIvoFBE`KePA=PVe{PjS1yf87%r_Dq~t`FJ!W ze#mygvOT+nv`nr%BrUXrQMHmaxdQB_;hM|3@>d{&*bsWYvAjd7a;|H=(qcLVy5fJ%gjE2!e)bo+Rg|!;&&3J^jq?q4K^@bi8la0`a9|=$(F4bkpv-x-eLiw8-OA zE4%X{b$R6mjsa=^f3%6{evQW==dhy+q6#y(%lcq+6Opd&2cIlFSM;=ZI(_P8&w-%EVZ_ee{UwODoP4Ptk`* zZM)n=+FP6a=gKb-Ow-Su*F9yvGH?0YaiXPd$qn3i^x7eLw>lm(#Z#7jQiHA4H~m~Xf=93vB`VV!?- ztn5NgJzD`(Z3g`@?@b3X-`9s3N}+hQ+~Ch&Gz;a7VX2@@V^KBr0iQOVKoJ0N47C=! zsotl8)byg?r-AqZ&F^kWJfO}@IX9Xx!%;lOFPi+s9}Zl=Sk%?%!Eeau^MiB5OQWqdp@pjeMYNhp%|MluEC$?G0@ zEsTe{isE)DVYSEN5pqxMg-aCe91M|s_3|t7f0izRpcxQm&94y?)U=E2;9DFCSA2x# zoWDw4AJ5AfK~2b@YMWCSUtdT`PKqmrwNe0I1@Z9W4(Fn!=0(cP<5tz)7oX$n9#8e;Of`xkG?TgNIs24GDEApuv5Qd7PwLM;%}T z>v%k+LOf*|;oqd&MUXEL^sT&_CSHUG?$_l3yyBBrg}5p1F;Q7DG$M{LgN~~;mv0|Hs#BgI@wT+4~XdEv=UII$5l@*h>4B<)ha=YX7zxIsHwf1>xh zdu_)x`ShRV6(*1f zBO&F6TdvV^Q{&XR>G)G0+VO}K@y5Y}5jAq?0XymQe+5@m?8?;hzlCExep2Q5HvhHb z<*L*pQ#;HP;GaKO&Yfi8%m;?%f2mgMU@h@TX>1-MlNzh>tzfvh7&M@K%p~6D^}2A1 z<+^{s(OWUJD*!fUGB(X2SDs5%xVCO}ySGV!ZLR+boU^{oERN{HarUaFk{9TXC)SQA z{XD?IT704hHPg^w1WtO0`u*`RB}N4g&KRuAX7>p%M(-~DhgSQ-fntM4ehkTM}3jL*sZx|2tZiYH4`i_~B7pTfv$yg5);6He7F$R)OO`hAgEd;25h}me=sCPW!9$N+=oxr zve!W=!zcw;L1R3%Qly54Lv|eT6dECRd-o(AFlPzUM4*7w9?R7A+RY5`!SzIYmi#Vx z(25ru87o<-*@8}{w$&s%9jHWM&TKX}5)n6E`EH|Nd`zagoldv#;rvYzgICa}h~H?_ zuIg1fX5Vrtb2&HQe>+EkA$QABQ>@}sQ2_o8qrcGJ6j2}APDQlq?N2uzOf2P<92ohX zP)9%&A{znbWl#kk9v$d>lH(qNZnui+-FRc*kHVFu#lh?kJGkP{YGV&wDv3wRNu%`? zSRR89-as#34k$w;@Ob02&5_4;0Tsvxu)?C>D=PwZ;rZBie+UXJE}X&Vb?Ct~ogY$E zJLYUVd!99-!8vN*q4W+#x`OG6AJ zkldVuJgnFwfAxr?RI!|5^;fQS8jCTp(O8Tc5z|()>j<-{7?M3oW9l4Ixvl^PRtui$ zsTL6>oZgwP$id7H_+tUU8hnWIdo~lSx{M?lEB@W=uyTJ1iEm zi5q6Ke-R+Y%{Ham{PxS1Glo@+qWfn}3@UA5t(Kc;K3kXqgZ;w%d_orU@q&IW!6WAN zPIuKEh&4b~tJ54oaaqVdyPnk=z!shzboO{vmEJn_eisEiT-#B{w(S+@8nS-=!k*Eh zW+U#vc3@Yj`d)c3Q#!73tQd8<{HxvSQu_-%e|^J~#t7-#!v5T#rfyZ&eGX@PrPou# zx)R~=U8&;F{{fi~T{?}pSJx-8N`b`OExtT#bWOy=!7aO2ELV!JCxuk4Zc5MA0r6@` z%8I{MGIu<+Nhq@G<^>)>DRwB{a{0-p3O`#k_lNuG;BVZ=PJVX8Ia0e)Bx0ae)JvBX ze{<i z!2T@1C$fB@u1Xj5R6t!yecA|R+dgGMmp5)EEGh^F#9m1;6lxFJq#ggxE7er{UoWHx zwL9)!g1SQ_QP_qAjam1@Rq_bJn6fGwfAiLs-hj2-ZPOR1FyXuwt>j4%PeBHD0X0M_LOlJz5 zcYKY*7f%|M8#SxGZJ9m{8i2MVd4^Ojm2K#?2Ff3*UgiTPoOeg5wVXS3S^yKbn=0upWie~Cyygm3H4qA&X9fJ# zTos0X%9zy0Ephy`xM$n4yldw?(O(|l#cyL8xdPkKF4Xxd&a``O(*)C_f2&ok-jfBK z51rZod!qw$lD9FtYwb)5v%LrIES09xs8Ua<{)}s3SFXvPc~U0Pg^f}6;zNsArd>yC z0f9ELdKG!c-b9si!#FTay+5w;i&AJT{tR&wM|}NaZf48zUl#MG+^boC@(;KFC7wF{ z91F!J>@?kPkd5zERyn}z6_@Oe3As!r zknEz^fpxQx`DZ7_wNIh{()k;SVae_ZEZ_4cWF`NY(;6{RznxfaT2}M5_jt7i`Z%n0@SKP$aRIedeba;wS4Y#As=trjU^R-Nb&m2fOHN zeO0%fm5g{z>buQRb;~OC2;+*~% z;`Hydd_J9qUIIUs_q_b**=gGsYrC=SpCdfhAm-r0sB#(&c9tEfgUl#>8V%K6KO@el zRlbjHe>lryN{-r%EMzLxSM6WDQm3zJ`$(3Os(lrzCEgzfK^sHGjoevo?+=!+coP*< zK?wO*{K$J9ENi5Yz|3H%V|rI#lwtO-hk%;^vfM0y3|Ax$bo)Q!2TT(J1M4 zf4jsSBKx z4#y*9t~FT;D_PSKgUPebi6xPlNpP;8@)uLcw?pYZP}aH7h;DE6gCCqPPn~H)8=fj} zSd-L&3pLX4@ADDMk@@4qJ~-4DtC5+tHC@TqlRIIu*x00q1BEMzr#Doz$R(7wf3;ne zi#y__bOWmrtng%Aq5Z9D>HwXl%(e68Ck*^nAiryQT#@y|^&T>u)PSHnP{Mwn3f=Qu z%ynAihC&)ny4mJih}c*vQvy2X8T$OX!-%}sQC#~a>;lCpuDc*Ql0!BLXKayN@+n zDe?Xw*SBsnnc-LYofQ~+y1Ikj-`H_IajDX1K?`$=YVLixYwy8;vIR6t_%n08Wt`#S z8=NAl$AGOnR5bhYN|U>@f9G1e8+B7hiA)MLv^2@uJHdDiS~k!bv|NIh-N$2wEJ`P7RQne{Lmz9B`-)D-BJaComhkk?h^)ffDi z1szWb5}EWI=%e`@;_xGQc!Y*2=kB(n+fxqHmEJ@cbXDH@*kN)Ue}xC;DQ5yH%VnoL zM=a){9loJrvpMnUIlcCkrUUyamb2BC$o{ZYN-ckasF`IrIL!Fcc~~-My#P``t-tn1 z(c>VppaYdx*1N*8hV22Fgzb;>#KZbFwdiS4=n9rZCR}vI8;~nmlQ}R3Ga2cr{BVs& zY!x;-EWcOsCJDouWIY^n-ha=NPhYks_s^0Fl`FM{DsV$l$w(2ApYTzWk!D!sR20`C z&5hyrJ8-60d{m;?q@rvWB)Z6=kI8lzeoQau*=T6jxGB+n)+VRYEA5frNPUkxtQYh< z^=4N>D0O_&^9s7PEEbLCcc?2WlpRcdIKr}P|6>i9*QTVD-f5&)DD{)^);*cY;+ARAB2pN>#;}$9Dc-u_v zOacGxn6ZnMQSD)W?$<04ziF1o#L}ZxF$EUn$LYPw{N(O#N4zzPljpwmR)&F>YL=&K z@9~K$nLC@v?DSub1in4K86MruqE1i(j zV6+zZd~|3K#<3`DxS1+?|CLx)oSrp*mtVC>BKv6qKU!(lK$5E?*|sC@{%8H^3UO3 zMGY1uktkx4cYo<0BH&%oAXWyuVf*pf+I^D^s`}i!uq6yJ6e31JCjrG7-lR@D8p zqz~C$#*ND%rS!;lTl>62Kl?6KMI-}k%%hhH- zU0X=Q!VUj@**w!pfRM>_h%vifr`Jt98NR}p6IR@OM1P%Axf$qSTWcqnDLv6EGCasQ z=oX1#u6XVEJ@#r4Zr*!=e^diFE*LwP4T{hb_!I_6T#lSCPf>`p(jxG0>4Zw>LD!V_ zhdnmb(}Vr($oIC&5{Z>LR^4D&R7ljW@}afWRKi+ETlS*x$$2G`S0$HYv{>;Zk^bU8 zCNPB%ZhxR7+>}4S%T5C0MSBR-=^fMQWhDaCG|2_q7LUV;2yYMhW>NXn12D%k4zbrj z#1Co=C$PPvbJK-u#S*wvvCMB>UMdMv4=9;a{v&eWQxoF>R0Fa;;2x z+kdbYI}#Nsr+f5JvTg`Z1N6_epKW`&P@UZGj?B!PrX>DC{5Eku0B*2%S5*Gl>h`B2 zG-;EK6K0j=L`&<$wX(oBR2=Sg`7I(Vh{I#$ulS+C{rq;+P_5;S&JX1y8AfYRd!KfYk=00IY_~9GVG|G z8w)2rsF4o+L&At^B@sEKf}!9!x7Xh|vbz%~Ct(-?YgU@7hVbrMRL)lSz>g2%O89X>BatSz9i`m zsAe-Dl9|RP5LtV8(8S_1c2!Ihp8a#FhNIs|-wS%LIY%l7Ee$|@ypqq+R%hjBz;ya?8{y|w7eQ)lIZJN&BhlZMicJXit3bz?O%4|ubLMqY_m$EvnJvNR%%4-Tfb_4wz@)T;l^Kp;sh zXw0L0BJXi)F6YN-X`FSQr}%74s>181twQOmFMFoRcG&ehdE!gAUDr$hws$n13FW{0)CjJ{VTlVFyw{#zz6N6RBNhto_i8I44!l=5ys&K9IK))YynQ3y)Y4Xo2TBtoT2 zXE{zBe;iZ7dfrouT5yBlxz<_B7w$}KSt z$8dBNYuGtABukE5}o)WEKrRR-hbNR;;iU z0~OLOF|=WhMgHcx=yUsmJ&tBca%f%x?1HO)i0cD513=@c9jtnA(xNELWpt5$@`=+f zz6Wx$X;cTkxE+$*Dn-JFc)NZ;8!i&ro zNOponVSg<6oZzHgaLnDIBi|(I7CTu#JgK)q1Z-##d)GJZ5p=15E%zRn#1G_Y>+Na* zckWDT{lxEB*KamQhye(JPL7E3@VQd;Q92$c zY=U+16UGjJI|rq#&}bw4tCU`(NnvNTz`eJJnJvn zfPATXzdrA+^+hS8?_x(89Q%>y1&12)0|4@8N=1P>g?J)m#ixcbD{SXfP=A>|=2`2i z7KY#CZ?9uj{2;-my<;qmViD+Wpj0tbAW$*jT=*kKfbc`_kTr+lX?3qn=PtF&>NeHl4WzqDx9ckoeMJW z(ceco8~&AGO6Xo=e&OFaoX_*Fd6=k%tsarA2C_F&KJuvo z(uw=|J~!=qVZLRBnmEroF&cT~h6+Xl8VD34dJV zYjIGnHtO*c;!|w#hW~w4B7=7qft%}csa+)G1nTN9P8?K>t3~d%7PhU;2JYVFcm3yO zSVJD^HUIg|1u-eJ5#0~4Pr$KLF9k#zy`s3%(m~6s_7XHeA)(N-UC>R zuiCTSu3S?C?v!`?9>`W@*6^P6bbsn+5S(;$UQeK4FUu83HWUx&skFF#NAKJWdV{F| z>i(|5zq5M;|JUpu&54DtcxhkHa-?BZzAS3`l@hK-wvs@ken=9`|GSTA)z)_xn65DZ z-bUh??s8s_g6W9=zo2JHy6i+ zoq1W@7{_P!PnTxB-wYla2}UXG^=D~g3_ra^*y45sGp4g$lRaM@(@YCZoF>zYIn!FL z3tb~(T_q)Hx%Q#ODWoy-YR5`l7HwO(`VwGWa8s;s@n1xaIjWw(iU(6RWSNE%bCzkO zqryS!cs?_&;%h3bMpH@n!+#p^cyxe|%!?Z@9ujpX!xwa9ibfVK#LWHOK6KqB*BIv~ z%c|nhrAyxrasL7fOB&ivla<;S2Zj%uz#jG${37n;TQyYM?)^-w|_h;`#oTvOU#(u zLpSvVzz`H{^X$Ags@4S4Az?9`)y_sn0lfTsgUNhvYxjk5dpEeg;+sKu&zXDef{b80 z&ohXjnGdApygQ6P76A+h-%T#Ajvzvq@(OhVzdYg&gW&?;8&SgwXhQPaN^40_BPsxV1> zt39|~ffFpg2!9BPOV*;2mf0Ju(DD7s*JmLcwp5|PxRtnvbeF=LhjfMJn|)c-4+T-F zFIT|II2vcGZF+;=#J^63`y=OVLv?&*cH;5BW3MUuX57N}UkZ!aqF|RGj*;Prlz*w^XxAKHx~`_iP`LKlbGQD7 zhB5eGH4K-vb_>GAz7#QQ_J3|DCq@l#ZGAY14+Xf1B~V|4}G5k-EUynaLv5 z2Y>sYJx`~9etNgX7~0HklTG)zMz<3HX4`2U|#|9_v6 z{-2TlpOOB5HBz5YRG)RP$8_m5P9ppFVrKAN$E#;FLD$mn|HtD85a_<=4dR-Ezdfg- zrF{!`>AaVGAN?odOh1PIYK08(Y5xTrYBpyjii^!Hjty+c9Z$z`h8c?de?D#0%#i;s z7k~O6A^8WUw_oYS^ie33o>}vRK7WXjpw3QVDw%#e`72omOfsIE?Dr3PA|oabWw^7k zDmyal_J@R2v|RHa{w5I;+Th>;3oA<78?^Ro>yU3cg#L=Y5{nf*c{Z{AA7)$pwr|Z@!>NoJ7}=bPf`& zOzyU54`!2f_~EAjKo<9T*gJh^g~7>Y=M zS7Cl?#|s$O64!`4Y*=Je$@3DL+;=ij-hf+^cl#E*+Eg-kHsI6%LwK7YOK;13IZE;%Eu z|IAh+?N`y#jdX$7g_?-b{dKXyZvFz+c0mzpw8*tQj?vcO9%IYFh&o5(WNwzCRO<=pk@(Wp}gCH zOu@Zp{;=bweVx2SZ7-qUmOVfUQiTE_b+3ino|xs{y4wgcX zHtPE-SouOHgdcPsqUxf`8;rklTOmlMC!rs2aesm8TJBClH(YL% zROCvP$BMO56NFJEWzsl>v@*}6pb^_8ULV(A3ARSO_OTiDow#l+nf29w$8Ki`OQv`-Z%)c8_K?0@_6lg- zq0lGJbbnYg!Bfawt|1$FCc{3@z(XThI6*PKjhmS#u4Fc(%kpw3Y&H$RQ9?~f^#Wgt z45!ktXxZP)X;a2rtyyMQenZVel3o7=PC2-`9oA{31=d@V)S1m>yxD+A*>4q1duIHP z57!es0u#gW6jBPr5d{v1(*YS!ZE78QHL|QHe1AMI_mLZD!$FDKc4S>Su{l`Qtv>Hn zI&Ai=8n58ZN8BQtDm!_xVZldb+ZL^7Qr2W@l333&jWx+u_!ws#paQbDGAVj@xM1Ly zhDil2_7u|3d-z`)la97lTtiI;##gIXpb-uPRJ2|DpfsSV3tIc-|6Z!L^u^%3Ff&8? zf`5#w`g(O*@HdP-w{Jb8oV`Y#-XiytNX-){?ZUmCUs^ANK?FPM4cZFqE(cu6_M|L)L&0RWFu&x5 zkC^)1#^4((ToytN2l-vnceJQ~`u;;rcz=Zdpr+3MsIX&v{XtTN2l!BDvK9xv4Sx0? z-Xa<~B-87HV3bomD$+dEcD25XWZ_eg6Dv(!A)YZ(&o+GV?VD{gE)gO{`J^**SW|iq zuEDHqMVF^OncF5aqbfT{L)xjE6v6QYeHw+?!-BG;rJzIB^LSvp1 zROyY3&E^)iH>xEnOYrI4qt)uJQ4V{!6F&gYblN^6eba1QwduL3b-YfcBhy*BY~a4m z-7suUlf12(e@c46{&mh-UznrQu7B^y?g>U|!Vk6$8T#@GLqW@bS0YR$%d>@frfxyY z{Btkk;I96#IpxKG!6~|g+dXjcE6j9eA3BQnY2csCn{ePA-x?IzwLtgkb%~OBkwft} z#cIPUyeS)1;m+@$$~FwY3>SUxBERAyB8M5_U!QW+$7wWw6mAAhOXO_z;eR#}{^^Oe z-=qFU9|A0Lt0Q+Xn?z9dkh%)&HqD(X!&0AWap{Rfh_MsL{R>*(8QH&p-{%xGw z|A%oleR1R@_U7|`yCA_73NsOoyXOHS$e|@usmqS@iztK{Fs3cjy98Bm6O0 zsUny%>nE2>GP!seyE{P^J0RN!w>w+fsHxU9gguYv7&maMNEwSiYkyDE2fuyVr-(sA zoSzv=R#{cs>r%F=FNWleh4qOKT(hXZRk`~0qNd&Uk$aGX`ud%vwSPWPuo5+(W{Jv` zNE#*o{lS0u8&sqS|1>0pHZTs8B!WjZB#OdDLmO$hypi-p4W=@7T6oEx{9gf=U;my` z!?n{beJBzR#py|ge}4`qMo1||vEYFnn!l3~Z`(g~ERR4=5v^;J@QrQ7vY$p6?^H@D z6jl#|2M&S}BiVO4UkzdpuThR4nxprPvvKo~F8q`@yMHyewq5D|sCmykKVrfX4Hbui z-z#jY(TN@waXMUVc7)~v*=Rf*y^!Bc-B%Hw__A@k-?bqXIe)G8UwI}^6PX0c+Sn1O z#*!;GGNwNs(=<~V2roB!zHlZ|Cu@);eUzZaWbh< ztT(EB8J8Q-f2RtHqi}B!&&vg&p27X3`}vFrE_SwgWm6#i@78#7*hDKUEB7*691btv z`SNwqOb@$2ua^f5y(wY}qX}D*c;tf7*XMI@4Zo4~5r376Y#E+s$rrLjK9{N7$Jp$P zJCzF2WcpQf@TueV^ho6(UML1lE|QQJZvfmXmFBEc+}-^`wNMQ$i-1#V7SHn&1C>sD z2sgV&nAiP_dozWKG`S|=!|qt>2p!KxoBcixyUl^tO5hfo#|{jYdTrX9b{Fb+y@W=IvbZT8j1T>k?Vg> zicu|*sBlr)@e-&Eg90XzPRqgdehRh=wHwK#p?^X9dKd_684E{5C=y<1i&8L{*Ey!- z({jy5L;+fH0Ou4ln4AOa&*Mz_S5-jw29c9>3kA*TO^R`rr3pyIDtaoUL9NfXpTVlS zVYd-(HX~y!au0S!FyfR}c@2^M2**=Q3!10jSE$N>{5&b#ohg4$fzDHBZI6a8)A=ou zwSPSpaTE($inXHn=CTXx1tUNv`pa(x+bot|A|MTs0ErR?P$+3>sx`QhG_mDy4bG?s z%$1%9JL=$d^b&wEFx@Mrusf0M_b5*&w`ZkT=f=hEu~Q_u-@dAGFicTt1CMii_9l8} zzHO$UDpu*cPE@<`HjZ)FTxh+kRM6(5`+v_zd{RC%YWYA~VpKBCsJ+n){O!LBJS>t3_NGMXAjQ1$5^4lj|EHj98wr>IBw|D6 zlKLIxj7R6T!5wqS0-2?sH#4{%P0wE-u7Bi~ z=!wXTBTm5P3+}2IsnjbzzGvv$Q(Skb#JT{NA-Lr>V_eUNITgiP8+?9T3z4bO%6A9h zF39Es^3u-KDXuMI-R!rzS)ZRDj;A6Uo@0yD)=Q++;?wbXKwAZ^V--7wi6%Jni(wuE z=O=;0o@BS@z_ZwdIpZ~5{Hx2XcOURHgstZ?QYNJR-g61V`+BUQnr1wjfkO%Tm?kVz zIpp{0??j%47v4ye#tKjKHq~651oK>q%4Ye3sxVa^-H3BxCr6lPx_R{?k)+)Bw*!9y zuY=nku0~$B`>9W`Q5zR3*MEO4h$KPpPb7*|L|2E_Lr?Dm<{gc{=XQ7}B{>Vv>qr+8lB_-eU zVu59Y7|iokpfsq@a8YIYqp>7Mg<@baU^wA_8PK9#5f~ zm-cQPw!kO6B!r)tF@I-~*0*YPKKE)8epMZn!7~^p+xx}(1j>B9$IZp>S!s1G{h7Pl zynR%rU;y}$*KXh&4(@n;fqx#3+Eiv^DJ7VPd$U(f zGsagLfXs@CIrmOYt2{QdW*aNai`O z(+`(i|27va|2k$brENU~0KN(ym*L4R@8JfST1@pu6EdzDYue{q;5jm7)|WSTCiR08 z5LwEP#7qeWVSh^=Gh4)@#wiOYTX&>`?~d`&)XL6NxMx$c?ze;$3d3^Yc1IeaVBMWX zB3gR9fX}S8-tLS2O^zs_g>^49bPjc7bXDbMZHv8eFUw}djq3Iu>vW7QwRU}`uHJHK ziiEL9n-0jXiO_De=|mlmV(9q0uW2ln(|?+s4iCskgMafS%TrfE{bN+`GCcG7W0>S* z4PA}EhRsLA6y=rOUIy8#m9J4z)oP0pmxON`eYQnKWr4oFU}PAgBisMt2@nlhKDOW8 zt7DSHtitDVnzgB#x<%gmgUU2#r2@=vUc7TF2&zQB| zIV8*mWq+ffFEHqMrWZ)N$G4+V_np90fcqmb_ZsOc8X^&_D{atdn*tm)4n$_4i&}Yu zvtK*2?Y0%UW-5s!&{57t-4yD}l5Yl(?=DHt|BBW^f3+~6&{iW=v|WTlK?UJPX&ouPMD`F&3_RUZPZFW((vD$;EhO0h5j*4`QqqE z?$$@p7v$PtZm>A1u5PU7V?u+bS`R&{zxLNMoRx+-Il0mJ;E6=HM-Qum4`#9*(g!0h z@v)iQPOm-!jAo;dWR!?RqAeep$NCBaDtL3-5LfwSK`xIZu)f9|zMn=0ktgK)zhHK2 zi+{vAH*n`ErwfDD9&yF{7RpoC@~{F3BCwdut^lY%1@N7Fn$gD|MmDJRj?4 zv*hgA?RGNj)zwJjWq$I|Q^Hn_RtiU^3mWtnH0ymaD0v2`|I+BtW8j*u_xqw(?!Ih% zHBV7iqVRCFSI`GuPU#D-=l49HFjN2`jX$OiFT0Y-+7PM@0%2ygBRq}P!8tfNZKH`CR#!{H`&VN>$ z7s+>3XN;e*CA7OJmLHPDgAT!!_sv>WhhI3;=k)+qtPY~Qb4!wV!>&S&k+| zH>J@H>afu$l>Gz098t`FyS1k`$8Og6S!ObOua`aRNZ^aTJW|S_LojQ~(!HnSTotj1 z2bSRA?^r%P9A~t+CzT%-n0G9ENq;YPHLms4W_(W0@TzpLd|D-9t{{>6zT1q$`?IRz zU!bMx$}2Z0QGp13De$dF3`Gd7k#;!OzDy-4)2ths(wk;kWhX8r!^Y0 z87BzX(Z_C81bK=woD#)!XfJ+)D4sgO!y-xwR18isE~W)D%{R5zO|%J-s(&$>t@zro zip%L#p<7P17K6zu|2Digl#7yBgujGMIwsLo(RVqRXK$*5FbE1 zEpjZ~bE@2}F&WSk38bdcH%$!)NvMhsJ(}Vo+KH2fW&+uO+uEgjSY3rl_i9ZSfKzaeF%@pyGBrgZ4i=~0^zm2D3vPzmMs;lsy8qa>52 znhY4_i4F9VGt3OYGVLGCHgS||wlNcM44!9wY~w;xz;EGCi^}IG8xfBh3d7>k{F%Y_ zRh^=rT;lnuhNj4Q-AA4bemf=p<@~iGx871+p5t;~aO$?h>B#{zY=70AGc@sEu$>zL zT4~#4FT|TXE}IzQm#i542X4hlFtD8dcGE0R*xD8!|G$$Ui+n;>-{P`zm#|quW)FMyHJ^Ot0ch z>4V)kx_pnCkL9TVnoer$Rb}R1gyD&;SG$|3K)HtyUpGsbtfymTZhzoVjyk{fcL+ZF6Cn%9&byjk zsrq6ua)gMu0{e1#QlF+Bnvx4L@?|nJs(-hoSF~=y$-T;q%(S{dn}0zO^0wk)!=n04 z&fP*$wzHyD6ld|;7mAU;@Vh|Wt{T2HL zhv&ss6!~__sr>M;FoPEW(@m^!Zu~i5`TO!W)Rf6gFVf*15nFjw|0*=k!SL?o&>WMi zv-u#&OuEp|NZ8a%caPq;;}B;fhtJz8^(R!O-+%K3^YXI!V+z9f<+AyR5OVRUhDaIX zoal4%!xGR4sN5stP&e;tu3wX1uh}B)Io2PW_V3&?GKTj-M-OY;D+_G)lq0;} z@_##Ss;|UJ;Z~@-#PKQm0%g#0YP;4X#Xzim5=`d3Hsf2yW9ucA(e1ATQK$YR_SvTB z9Ets&w@==1DOASwo8@addtKtVmfI0C#&VpjHgD=?%X$0VQ;#ns>V=;s^#i47#;baT zE8T(<@s^+U?y&S~ZA@J-ZdpD~E@dJA(0`nEqima-QgQ+6bQgDcPp@R<8q1+b_c9!S z-fbbh;rHKw{IFXUr#Ybe*5~EgAT$c%z7Xu3a<{Il+(X=~a4w^*BwllACmbw}J!(uA zt7_d&%p28Wd7_&T&_sNlLfHXRg><^Yu691@_up*u_t&q#Qn1r(d``!$U3*?F*#11K(sU+a9O|E}xkJs^jmQuse*q_sh zVtVcp%N7Tlz}H&=r_2@f=wyR?B!7yU&Ag44^kY-fR*M9Js%uJ%}tO1Sno` zxntJtn?*wn<)^8Rbj+XDY{vm&%WFQI7&o7Hvcit1>k3u&b4RJmj|Nubp?_BDhChX> zMKNxj0gbkLi+PrJ8%9=ufryGBx7)3D$OhrxHCoik9)Olssv3v0`9V)P z{ev66(|_n8wxc0Ff0?tq%3}QW;Domh_ISmsor!|T%D1Jg_<1m?$p`mv`>Ot?tOv5OOnPy(T56FGM%I$QqoPtas|2n5}!>&EG^H!vBFl4}bg~+Ps{51AST? zp`t|y$oj20sY-t3yf2NNcRj~d*Pz)iKULY->Pb?G9pW6mN_e|!jce&Rf@5#BoR__l zy@hM2#F%=ukT>2hx~#3yf`TtKg1$`pq3^j$vg-1FQSR3*Sd~23nI;=>H(g`SuF{(2q z+E!J~equ2=qb~|j-gnH&be7}#%Y-~i zy=V%Hb^D5|a|XeexEA!$Ol%X^?wdzu&TcQd%}^haR)}6t-PbiYN5Y73jVDme^z|~g zn=I1q^nb{(zT9LIW!+Ry3id5*P(WE-UMP2^*V~7aC}wK2v%|PnyMy0+giCf z30JYjmS4GB#hF|B0#2j%(S6v}#~0nXjamjQbANbu+ISvN!bkZUH>Qacf+PQIGa}Gz zM}wW;Tz415H3c9~%v}-FH{Oyd4voeiKGeSkX9(*%vuU+Aofv|eO%fnjgUf;^)S#uc z85(}-A+GexJrr}qy=`d>~xOS5P{pr#~Fg4ww5k(w<^<9%~ z6m{juPtUkfZ+&HMs3t~Fme}^Z-H&DFpfe6x66jEIK%H}2bD{f?-?vrLIw8j80)Lyo zkKH8uJHWucae@0YK)KngSKH3dxprC&=q>_g~+_l&Q>ORS?d8=p(#?&~*n zKk9EtwK1T#bjHA;JC(_y=S)g7!pZQVr)k!4XvMpl4%d94#V2tBZ$Qm+E!Ugw&OIF9 zN-HJE=z3yUR2kh==TGbaBl?xR>3=P`(&jr#-GJ|BMGav2X%7qWIMh0Py)3jUI_f0d zb_pP+-riKglJ!@;l8>S8>$wOvE-5?jL#M?nqRO{_XVLGTel9?@>@ z?WEAGb!K_J3UyF||B036nNaT)af#FN97H$ZW}LZA1~5SWfZn)hN;^MURe!hpT1>}I zY6+Q-e+K;cOXUO_v0X}|E>P0KKD-aR-qBc+*7;~f>n?q=-}@|?1auGk)YDWFbe$*k z>Ud*G5XX4q+x{Mz86yzB=~z5vQmQF?-xh*)P0 zd{~g0$+p{`RMha!XNAqV_K$K49H(1 zuvmEb@FIUjd^V$us0L<#w;#wM-7XYqpf2$+x6avlE}-&mgEl~nxqpSy43AE6fSjwk zeMB_Fu$XWqTBx3H-ODq;uQ=lOPc3K-b9*I5*kc*4480oh@6mZbCh&YfkeY|CjWMy@ zP*#~)zOfn^&bvMw-oNiHVteSb>owjYM2WbLjh+eK&Y;o%4m)Kh%aTNM6$j5js2Ll4pMf)+|+ zKtm}*4}uD1(bn1{3&8BCr*y!>AE{5$ecHq*4SV|;!B>6x3G1by`##8%crzu0&*2-s z#5ez7UikZ8hI+F_9W~!ej**1=6{uK#kg(4YUkY>%27ST za2ByJD$R;yYOOX_+y%tieAw+b$2=+!24l4ZXrH5gk2UJ($_<@IWloqnRTa+ud0cWD zzND;&Xej{Ru*R*K8ELotiFa+>XZNDM{<4*rcO7vTW=WdqVTX?sEfQxG58zxCR(n2GEs-8Dlhpjb5a3>1MAOHwR^+i4cYJv1rgL+Xge!MoS zo*`lez$UNt`V+vmDUzg$c$9)e6KcK(oiJGzLWg-AIt08WU+MicEqrKX(@X!GFm%PJ zoi6+iIGdrcuFGt~1%*nxE?tovRzoGQ63aA_3~Fqd4}X$m7Sn$k)_K1PX)UJw%rEvS zMA3%UalNW%x9Tcu(P<(Hc6q(gU1TuXI1T+#vL2_N8H&WB$91H5|~$dR5z4;ZGV6d03@g-Gp*ctt@^xrd)0c2v)lq@|9Hgab^+; zq_&NxM3=VHwapg)fchIO(Kln_G0#@7|{G zf`5K%7(&v8&`b5FAN@C@k7c}{*=l>HaDsXuF_(GkYp%ZflqW&v zS1#yy4y9@`d1cGO^zpMwe&A+4-|TGu0LL~q-mzFy^y!uFALU*#+{(C%IgPqcZRXfi zrkT7@A<3@+U!iKBD!(Po=*XJjy(?%9!BvddTqK4z@(qi}}F#5s9P$ne`dU&Ee(6v1{qLgJRmY+(j|kJjl1wB_n^>VK*l z*y}V)Qo?NGTt^Fu?iXgSgoNd)*FoJOxq4Q(={CDX9OnrEy;_2cv=wow zMk?F^nx;3ZI}}a69Q5?DG@y)O)vBY@UH#(tP$R=*B_c*PyR#;2_g`?`k+WX2vw!XM z_-l=8^kF3rHg~s{y$)3M;3ouqGk>Q_-k)0a)$w)|8nmTszy3rpYCmCJT&|mJz7&y< z*kT_S*5;gp?stp5?8@hP+y2VH@9e+DBNxG(l^95~FgqRa>7c7lG!kA#|(XgD@E|yzk`~4uR z_I4|jrnI#tvrN7HSIC1&+Yp4vMUy1AmYziKN2W@Dm zMyhYD*r$3;Ql%>!u1=GR2XBfDRV@vYZoC(PJDK4|OK+_0fpAm{Z>*i}W$b?I6wV(? zfxK%^0ikufWIuz*TJ+2rNtt{2$i5D%nWk%LXRp?|?ZW$@Y=1``H+w=Qi$UL<1d}Dg zNiSBgSRCW`B8|^pv-T@{BFve+;ssS#xRtxPwP;$j1@hSfn}&_+M*-Uw*ZLKwqgJcd zQ3LVaba4sjG%2Yv(Tfo@s1Z}#{>Q*a^`@t4iBn*3WNbYez_PDfNXEPgr9bl9t5cEF zAApnJsVIP*XMgZ0Hy!tKm9qq0RM$O>)>uQmrG8+}ZZ7_`yH!|QCPg{%luqS=hhEQn zy*6GhH!hNHDL>&k==OXXw0Z*Si-&gd4DEwZ75usfm~Qr|{`9Aw3|vZ~trnA=WVny7 znxoF@EuB;p#;q1*G@x7#Jfy;p#;8jAgmR)c&VzZbVt+$k-HHO}5Xhs$3ioaWgxGncnlERQo*K46`weTMN5K<~L4Ptcc0` zmAm;#kK$=rv@OeFB_eJ3&>N=XI`s4)wPcntI%}x=|2vr$T0hk{CDcWIOmv|9HPLWT zE`Mx~dFPwPxN9`~$YM+1A?KwPr$hYnCFa9BqVh;K#3y><(?Dc4kC13-D_0KGV@ksI zL?o71{~ivjWUQzq<@UJ zh`G<6H^M6H+;W`FYYRvIAVxS^7`d?W#(n8Z`4V~b>mQj`9h2}ecaE84Q5n%{uV-c_ zFIIYTK+oHPVXcx>VI3OVBWi!{=J6JigB3tUu9r%u8g<$72A>J2bfAX73g?0FvnBMy zwGKyqKc+90L3e%d$zh52VdW~*z8&*{`v$^Vk z_0XE@a-}XZYM=M)*t|e_&Mr7p(ii@Zi%#(9T=^sbq}0Zw~!CnTK4*mQiw<~|#P@1f-@!=9N%7czeIQH?cv3VF8wRlKfT`7xi z2RpJ!xsv-_wbCoB*m<-ryArB~)}Y_sw8$+radooH8Xo<}w6)3s%x*fz4Ic zS#zb*%z_R=Qc-~}vf(9H5ff)!afDz|<+!V~>Hq?y zW7Io`*hyq~%T<_UvaG$41@)%PlcwUGEsJJ|aAmlD&tlP*Qdqkew6cfv=N1Oabg86{9 zbB?{PdF@Bed@&QhiuZq_o@C1_ewWvvO%i13?_|yT=j844_SZ|?5qzzRIP>N`GLGcO zH=67vH?CWdWT20GD-QqkG%Rw_px_?VmsN!Bp{cc-TXv)Wj-7mnBz?6$?i}GcTg3%m zhTnEQ9}=cC*VL{6`TNI+>PP>0Ub%7Qi)0RCw|?=xQg-LWJ{fvrcCBE1n%ngmL$h#~RxseyK z2_Ms&L0?I?JE?!vX;qw7jf_2i-tIqOXve%jT$eTSiuJ6q_uV`qh6=!ueSE+T{glWdVd)uoOk_!mtE`CrPrKgj=0TLjn_U~9A1`m zixNTtTE`eTT&a}1Nxj@(BdJ{vsk`bVvk&A=1cqbXYzKeg*!`^!#HcOLr&`zk=)Q1~ z9Sx*foG`L)EhlB|>gB0_g&1KPR)MoNtq#{~;+9@y<~>TvcX*Ro zM&X3|1=@dOf9=M*_BuYo*~R_`n0${cW0fBlX0OGTac}DtRb3H6$@||#kg)cwb-gQH zqSGEUuGcSw_`V;kU+kQ(>k9q0LWi=LGAribbdY#h*#;#}>V8=U-0|Ft{$P3_v69^h zvA<2oxBgsEjVm z)XJIA4Zhh59ZzJK>s_!#i;yaFN%=R9{?sj>#~T&=@{0bzk*V8-fWG$KcpYo8BXSLA zcIEojtNM3&n&T}SxLOAhtSSnz@*eEb=vv8(%za5}h2imnB1t@zRx#2}xuPUrZ8W}v z$W(s^C8fksqTPvZcxp&%W`6vX1CBL4(i;@tUN>q>C)VtpY4o+a7V+%S*k*yJrtTO1 zvR<@K%KWnbCVT4px_>>#gzI?V*Bzym=c45O+=(n?lL5Zps5Cn5&}I9AGpI8|%RV(j(gs8}R?8d5k*#20@E+LUeJo3NZ${lUyCAzdNQrkt(AllRhC z6ckZh3i^zEg^qV zuKI-P{0&c1^y8aI>v2_$FE1RW^krhH?qa|&uO+>GvByt*6UGd}KjIPJbj?lBKe~o` zSbb289sN@WAGM_x45sPuNYUMSq)ghn<_L|q%YWQ@NdNoR!$6ysyLdy9Rx@49@%`3g zh*(ddfICZN?0^b}@}mf@W8Q3$R|tQjg-gO3J%I_2h5Ne2@swatxZ=L|&figK)s%jn z9NxG>Muj1o~^ z>Fx(GP^I?9J>|7MOq11>z*Wi_Py1GTQwg#*Zjb4AkmxfT&7Vo@fA^D~e^LHM`vD;l zlsfKC|Da&)4TjP*%9-Iya)cP51Neh>F;L#lmod_~ob*U|C*D z&9Ox1Rg&$wR=xVT;-R6R>18I5OKcS}pVEQ@c7LMqoGhTRUn=$3T}*#@&^+>^_tt0P zQxVOEN<5CkEbWxwV%83^tGM(t#+O{wG`txh?wW0MrxSLX6q}soWEyHMVoREiX&0#* z_s)-xBOyi;`ELC#($mU-^149F2a}!T^(Fhgp_{g=IH4z$G$$NCM>r2ZTf}*6-QAC8 z=`1R}6@o}HiNDG#fChim*WR)zPYYQlZbCuhzVzcxi$4nN;>CT-q3G1G>UUu?#Aci& zQn)1raY`u>I_E016bdAN=DvN^9_@O5vuthIw{~3*V>|WP5M1X@I3cN=;qo@VpJ?i9 zeU8`9Kbv}*O3;J>AZnlx&mPgU{F zVDPU<_RhZ}*_?Y&+p%!cP(5?WiYl06@o5$)pC!;^=Z?tAw)%8^*1Nu(U^lF%Rc_nP z{K3E45V(KTkh$O8d_pxkJGwe8@?dUU~?}y^L)nZz>}79 z?rOI_c+5!uzLDPtnX`jmb<=r^vz^ylhaxT5cZGjvMe{Fcx2yw#Z0{nX7$#W9R<69X z1||~rp<=f1NGE=sw!Yr?L{7hyn1f1KyrX1&Id?Yv^7Z@R%;+t=Cw>gq=WG~WmGWqb zjkjof#%tO;t{K)md(|6y`OgtrV6vUQ94^D9@T}bz440nz0!n5{4=rG!NZhBZ9-zi* z>v(^>SDyB9DI|Dr;^T1}ug5LPomZF_&-<2|+8l*ecACm3lN>%U1OHlu!%0VSaqFnL zV`VisTo=NlM9PL$sMR!&rrzb@jKV6|bl2!ug2nqIG6}9RD5o`$!Ml1uItwI+7gyH= z#-}^@94nsZ(f+F?F0tfbmZ`_*>elRQ5+Q{sZH zr@%R&lltj4Bf0fJ>3_Th@Us0z1(pSm-vd-9jd)iC-63Yi`g5)e<`wXv@+UVk+#ZDU zGEqUlETK58Y=IZavb#GC|G`Mjo-7fqMDK=#%sFo|>B_}i)7u4IQLv9BrZO984!|xm=8MOps?ve4Mx=b9AZ+;H zq#MuvNrnJgJgvJZYKyEqk9$0N!AsvdDT|@zsAGCdv0#H^XLRI?bO9$4;gJOov>;n$H;3g2M%}NDkdB zN}M{~1#4rf9>X0qJFpJb`k!cz?Re^2AxCyUX^)Rky>K;Srz{*>5!5IOOc+ zXkIFnymyB*yW~zLz0R#vPv}a8s7+WUc+TZYc=87222fmyQzI4x_7yyLDimmE5sr+K z5BpG@spB7ArcHmq>63p%8my0@5#=@B8LL)&IvT=xNMzAm#$~C~(eu9fsn=fU?21!U zC9Tv@nx3KN)0d!~C4!|tMRJK}^|Sfnwl`bW99psw%vU7mb(#pfkPE!*sW{O-T${GL zLwGk#y3>&1{hf|(P~^ewVpi26ST=7AUgj{a+}iQDoy%rlsn>sB7{5jhD`|f+Zx(5v zln--$)T^C!I(KE(`iG2rq0~>WC~9f7 z3k5LZVWP!7$?Ih=5ITYcB}nTHAx&BzzN|osE0+Zx1HNY%FEILhCTPyb|3~B94W32s z^&nvMUW*xSgD-#1`;+H;!_XoIZKL9}`Kj#Xvl$G;3F?hSjl^fFj428^F-f_*Zb+Zo zp<>F;z5;C?p^|a_sIeh0s!*{|g|38}Mg&0T^2Pmn<6*$8_P)-1tPG;RSj#SSqu$=E zzIfY@N`lf1^r5f#@ix-G6ak6`9xJqjD0FD3V#bg+_KbhcTx%?5tJD4M_zRpD=6N~5*TL4ncl^J)KJE#O z)A#1t5>1NA+ocTdqG$g726=xz@CkdLh-xY1X=NWcbQEdhUzlz{Snl zX5D`SYohZEpqc9DZkp{?#P2f^+IN?w;HQO4K{SmcPpn!&_ds! zg^2#tINEgU^be(4jruKzI52;aib4$Z6QY-TU;_58J)su859E-b3Ptl7)erHvMjn6T zAMPXiB$%Y|CaI46aGnT+kzV|e0X@j~Y7NJp;8NmPEy33=%1gHTkQGirxrO_9MvO|c zxR3|=R+mOY>*TK02!p9Xpc7G{wDgl%3HN2xw5A)?T5VcnKgc0wD^)Y!LVui!mGR(C zU-kI?`ef}3-6?NaPTUCxskPWHW?)#%$?i!C0#|TX?8Jwo273bv6>_pq4yg z@pG$fTA#^>wNtn4R_*Sn-V1+R~B<}*@-q)!qH&f--8w~z~vq5_>qQ(AiFhPfGnL~NxIpg*Q) z#H#+b7?dfTQNcfM!Pa?? zfcn%B^bQ7C^t~K<{#&ARe^!62T4rd)H3ln#QNQgULNPP8QclCMvyDXHn4$2!a?OBH zR~5|^0ELsuhAM9XvL3K{^I#v3pR=sDp172GJ#ZCzL+9V%fAlCifan4vF~icwgBuzV zz_tX*Y2n5jaXW@w3u$l{yGsN-hWuCE*M^c&Z3u1a@W+VHCAjjk@2-EwCF_D^!a&Ew zA@m?!Rq6*kpmcLwX{!QJJNOOVmBw;cC|+K5a8;`4qsWMCk~N;1L{N~*_OROsv=8&_ z2I@eV)l0@ygWrVO6ChCM1o7U@Eh*#_%n1)4K8pL5#YYtmS18aTU!oyTP-Efs=r-QK zSt|-T4ATI_Ek(=n{!)K6P_nia!*pbflTjZN3FrRwA*{|ZJb7G&6);Kd_LtgV{-rkL zYJ6rlLvMVmyDV6Q-9;H$B0fuC;ys)e?*=PZY$4EMx7pQaN2)Eb52L5^a@m;E5W!Nx z)*A2SP}yj|BA8NJzB${aQdC=<`^CI~_F2~!jfiu>N?pxgKTk{R~am}x|c2lgZFZ~5T3S>$X{$kZPLfzBH;u*kiP5+9nnyl+=rc# zM4?zl)cxjqs{=s-NQ<)(OBecwp2YtVBe?Z+9s)2qbi;pK^8UZ*N#48E5vB{X66IT3 zGkI8=pYPe~LFV-%n|p;V_sr=5|qf+qXmxs!suB~^D^N$KOd zlYH}&38sHxqmp5+^O`7y?40(u3RKA|^-YFAjzo#t@L+fk+i%NlQC}>F?e5gJ;z^S) zUAn}=IKe{|=1kgml0Lx;Jh>Ht=ig+%Ozz_YP&}6N0`KEfX2bHU~Hu4kZPS`4IO)at{!7o)zL4;sM+GR)=uJ&an4nx zoZjar2gol^N){Dh)oTkQV{ZVzi;gL@RBM0opW`*%@0zd|D!q3)oTbq1w-l4pZbML2 zFIEDQwln5&OCyNp%=EP>1tN-^&e^33FE~4$JIKHk5U~r7{n%0ib^~qiQADaA=B@%{ zH-MnisEOjb^5n(*p^M0Da16N{u(PAao_p1%zo=b5V$K{NSv5_U#&&I=3sX)4*W-UG z_^D{)d_ElhR76Fj#x5|k(uyV-m{kR|g=)6;>en;TBH!;>{n9J|kcYXX7P&O4!0q#c zVEiboF~}uo7-OrPH$Iy{0DA06wg~#D#+lvc{*ow3>q8rWLXCm8v8H@=>P%^@M4_#vv^7iF?r z+rwZWiKOusd7>I()Y6%20eh)l=2u91I9RWmnRPuJb&P>orSUGLpC?~nntEK7oaE*U^i}zy_ zQ=g#`R#H%*+S04_&y)QuiKbnxVqC{~S03(Z1e#Sm`geuNo;t9l@s-7 zJM`84YqF>SCG|g3HPpPew61?Kb${{=5soY{cC4?ky;a}?ns%tw@uZx zzEWk}b^U^C|Zji3n+n+B~>THDvkK@Yq z8*YmijpVxf5ck2}lrR*9BKZoZNDF8_cKO%FCe(!jaMn@=U*{`*TrXyw-8jFJ*~m2z zB;VGkxj=$%l;hY^9#wzkHuTJQ;Y%Z!4;W9ykWFPJI5y#;loppf7iav&x`p`)5B2V0 z`IProVq;F#iXC6XSv)&uUk#r|iFsir@mw8|7|pO*(1c?T4r7^G;jUxnG3Lw0dPLck zot~v!S%h#`kzIctabA)AuGpxZuCWNzfUYVg z?c-+{50;lfZ|DuDd6_il2Vf}iUg2_xRW%muaO{Qgw?{J$F!43pL&_}gGh{ZmNxU0+ zTUx7)IzP?_(fFBdz&waNN=C`3Rj%B@;5|#n%cAHp{63Q!wW#RG>`8r>fiL&d@RCQc zQ#&ah+IfEsIpnc^D9vQHsF14ZXsmi{Cxv~l-JVCO^x;t|d6e${pc5+{A@9-lLdWLT zntch%YPxG%MQZ*TM?jk;l(m<@SAGMx6})cTSUOJk(8mxPR_ulFKF6)or2lm+Wm%#5n`s8f7ZWDwALqE@~;@^yWdwodKWbctz zB`hD0sy#yKcPtI_VCRd@j}U{=>4$M78&A@r>@kZeC96_4<)Q@bUk>AOJw;0>$^=6(&=KCT!rj-A9HhvAMriTMpylC>La@veTN zILn%reOm@*l0u;pU1=Y*pB1Eglcx}eYG!|U?e%r-;08W-!B9=ifs!%*IRWq0kuJAE zAPWJlt}nZ)FyQN!|A!3=rPqAZ)o1?&Yc-T6gZ1K!*sJCApKFY6gq7s{cs{(vE^OW~0Ycm?MVoYh&TGd0pQP%Rx?9M|rO8TQH_r8irCS+~$Y!`oo-)?c#`)EH$VWvQX~h;irCev*mnd}U zXCm398%Sh|rI%FT@Mf}%Q9FL=5zv3G$`oHD+pEyUf8ZI03UgY#YfC^WylI&%aH*T; zWU=VSvmtDjKK|r~Vi5R#84AZ>_3f>k=Ey3#MLpcrAW$~Ma7AqEt@ra81hTJCU60gi zBfbGM7rvbbqQ6;KdvnHIh1IiYEQ;FfU8@#1=u>|`XOgMw*T9DgB7DhR+}M9qqUF_5 zLj*$+H(BJUW5*r-TofrT$p^~wfZ>@_(Of=~5i@CYpS*+XL&6ql`hR8$kuyU6#Tf9SizVY!lwl*-phTm%zvX-X*wmu{pSgb@k^)hWi%yD1NGR_C?cH5@1u ztI^?9_4Oytlv|IqG^tQ^U+9LEK~+Aq{B)MVTsm0wvEh%lJ(zT=Fyu*#UJiPDi(@yR z?68orTL;LjD7Tj`PAW^&|D~Ic%TrY9bh6vc{?fsJ9h$??4uv}kLxF8Q@ zqG?^i>fYl)Q;IDIxU%X<<2 z+5rs5*EHDK`AY0YzusKaP7&mrP|j{{B2@wP#;PsmO=B@QMY>vBG!_`ExgpNjsu{NW z&1s3x+rqI6Ja+Xox+8x^0^41?&csEy^&1&^)$1kRFdG%0QuoPJmZU8oIOG!6+3xHp z*ETqf1X~|}8H6mNCxZ$$Gk|~&E!%c-oNf3<{AlJFL4rDfr3*D>5{12y<~(&y_)G{ zxsO&GWzXzEIs?pZ<`YD=BN(3r^W1Ig;un>gnWSmK*Alwj1zh0hk&O-U^KC>WAF{zv zn8J?cTG@DJak8Ct&QTXc&4}O}z;)m)9hy1JXI00yV7n|82s^PgdtnsMy4n~yya9&A zi^j)hbMY%#4IO{Nwuz7znp|VGwG?huxj}=F+Qbt%B#f!`#Xz&x);qtC2s+6B=7H{! z!(M&4MaPMD%q%)vouzA(Ecb?LQ+TEB{`c%x?b-!(axw*#r9@bnTx;eAKMa#aybFJ7Wh$Nl8PPh3;l}4+vO)Cm z+A8{?AKszGo-?!|*gVY@ONSvygoTzxcM~FNAiGDThMOYl|1nK zs$(n7!Bv3e*^=e;)rC05=O-0Ku?LQNOvt{a!GSWCUWByN-x=Sk`RMiw4-+UB_>9Om zz<6xON~V8UHxpD;fmmikvUu*-Keb|4TE_9lJ6KQw7`(Fo_HJlaqiyoE$eZV0QNecQ)cIv zqDFbU?NVK=ek%C_xG_j?cj|8RjR78z`v-P(4)#!gj@agEvVdJ7sv@Y9icR}n8ejYA zdCH+9GPw_DWxZHm47W)0v}BZ%j;X$H2;zU+0U>}44d_nfRSgbKNlvQzJN%M{ml{>* z>Veo>z1LgGBxIs*=^vz>UL&CC9wz6W5(5eO4RPr~61DD1kTS(<-M#EIVb&`u&-5u@L)*wYtX(l^@20j+TK(*|bifqsqS=7#v3!U*>ppO#W8BMrY~ zu47V?l>mO8eP6FkogVzPoSEC{0F$U~N+9sF!6R?P6W+eL2k2*g8g0;hM;Q3kOS9O{dNESrVzD}nQ0`cN(lP#!v6&=0iEiN?Q#dq zA%&fAKaFZi|3!4;IqJ&@)#`sbp_~I~{jwq8%YY-u5-vGdlD83w%30IH6z z7YrKh=DYqtExGesjW0!d1aib)58!mRjC=J2S{;aehK_-LKLYRPFrF*zramQ~CnHSt z3X&fM%^NkE4>61SJi-&twD%K9pBZA=#(R;8-B;Bgts`ipUwbYUooatPogYu!N#CAk z>fDJ{w1K3*Q?69Tziu+gs+31}fYb%gi+`pDu)ONicw!og2NJRR8e=D76x5q2V~WkAwa&-r4ex(EY`=QMr*J`J>3%qF zEnLE2j5)?yc2oLFQU^BQn^CHtQ;biw8$n23(eEDASRxh-5?;RgeeMsfVVVu6yX&E$ z!1;x>^b#_Gb$#_Wm}!yRI=8)iy|(C9bQ0H~3M$H5RZZ>ZO87!)z60KrWd0 zoTV~Kb3+6mA=}J z2d_+Rwd`|?J4b?5U zU`{P&4&8=v-F~5igrq>R!eB8{u5aUsgVHeJQaNy4RBIT|4TfHs62LQ-P+FM4$sTU_L zFGAsHV_Wq(p9-BHU++zRHgc)r7*TrBQbm8{*Sk-ufVfgN}g;66Ukn5mTmKYdJyLa?rtTqKrl*wQad1~}b+VRlMZkb>Av?^P!V}PZiqh1oAO6 z(y4=a?{S8Ctpc1Xr)K_BQ}yAd zFp;wia^GIoAZ*AxIVC2rAuWl_T@c!vTJRd)L5nl45@6z#PLcw5{jGbvOGmt2HyrK* z#vYADMIyLihPHVbXw%aIjLu!Mwlf;rP3PWZe4r5K1itEa(8C0_r+AiMa%X?|?a8Y)!ps^-`rHa{b*n@-vONWHb^=3C#O0XW z!yj%g_p+j|AC>Y^Sb=>`0VBZ3DJFc`A0rc7PQ&4XaHy?uj4T-X8vASO^WmC*F38|g zpZL>n?PYDKa;u>45;)RWQ$v4eamS_KeMN#HLcD$HdOGMaKnBKfQ>e^Q{lF_JQKZf* z*V-$u$}1jk@-Dzpv$(MH)6V;W*0x&`6v=RjWS6h{xoJSb`D$IgdaclVV&FOB4?+Rp z1CFjk#$gjb;OunCMjqvM<7IOF6$^1S5mr))%2}LDQY~2q;6M^FYMy^$o|Afnk@7Ab zgp~lV*>U>D=s*@YB~hver)ZOS{cvI;m4gK>{9S4{s2^E)q&i zlGof}X$PCZx~0?Mb=IPiCcSQ6XAstv%m`BXUC>rJl}AP!(uJjV9N!`;LQWIt@SXi#~(BrxsGhahp3XmYI|%Kvfk z3yua6|I009nRa&k-#nM?H_!dvZQDQp(`!6tCCQHR7yitWLciL$oxE=xKlGTXP|;b} z>a@Go4GrIOu}3`k_@;ULnRX}d#j7p^VvzRaLAW3R@$|)h!qGqU-Rn@FvuZIp{^Y|@Srr^wr7yF_ptsT8@9$0AkqqdInLny zKq?Un6A7M1pjQgF;6zGLpUx;?@ix`04wZKQDEvZ|vE?C3Shs2e9Tdv$OLY$4hp@s4 zZUmT2x$b{PLuyd;OX}<-Ba0O^sRStzNu|v*=g=*_CrEWu=f`R|QkUO#-X6NETV!S4 zR+T$CT1^$GL{hU6FBUJ#MlfQF0LBkf50%y5&x=IUQq45M*---#A)k+tWi8v7W~j|h zqKvNR5aYhLQZvVmcYPL15{M$~_lRn!K-Q~Is%jzuZjE75}#rOvvv zZZRb3&kBuGT&?w;l$5#Qt))JzmTpryoeny1?HN1rF4Q}Wf_mrdcBUR__*qS#A)o2p zwAxl@jeRygPgE#{L_&5^UP5%tCQqSVl{6w|v6LCD5~}421Mvs*J4}m2de>@;rVHz3 z;23`;Mt4YRCLPN)g{W4*bA^^&>V64rddkq&H!JuKA$_cQ{wFymTQpTzvkXo~B{{+- z#Rp}i`lSILNnWYPIG2G%Uw-<(?0;>)=-Eb1OE*=dDwx&~>hN|W;_M0JkWG?z@%XH; z5r9eyrksUW&4uEklPN}M#`*wClv&38wM>5miOqN{8ioD{oS43sRbx&Rop5$Ag&{3B zlxmWb8X-?{ulQ?9J>mrESYp>*3Unr}5JN;LH6_+051PXBbA4$-cOEr~Y&uGyD6ld2 zyHsj17H!g)Ogi88^^^kQTMe|p=28!Gh`+=;ONUx(G0K4eaUa==Es9%Mp%IC{2Oa7h8C>Qz?HS@G{we zmUOyqWJpu^M!1W&i8qLsUbD8Z;@gjdZ2;TD+nv#(K%m~+XvLr;wf>SrZs(nYj7vqE z5BCQrVz2MWeic)DB!3rEuXStmOP=OZP9*RX3W7F8BzB>dP_#aQZJ62Wwc7=Xlw>8S zuv;pbUlNRSeBfaG$NMDBm6-110+2ZbX zfDKfD-GD~=_*7xliosQ|Vi1n*s&&mVY70IIMDCok+q2WST^r_|xDe=^gcj*7JQ0** ztMQveG2w-XxzF8Iq*zk+Tyl$vSGN9u;+A*Rzk6|I-=Zwu_*0zMzi zDwvgdZP&-2 zX`kS^v5C!ti+6uh(8+#q?hz(1s6E*4jc<4bWrQ#~Qv9xhfW zFTFfyxU_r&)7E3`d;CF3zcVK#vZ_s@hsmu$bypeMsYJD0qjasYSrHdB*?=-rdQo1T z@m}Hwa>$n@#J+4eLN^xEFCo1PUe@$8{4cwZSj_KaXpDbj9>aJe`@@htcM|D9(GCbn zcHr9>Gi|jN*OgWlt^EXAmTa|A>se9t(&!JFQuNT{m$Xgdduk1Mis#mIH8|frXpMcN zWoz<`%an^$n3JF@G|sh48`n^H?i$!1E~|6}P{aNxd>VoUqfd zlAe0w|3rUHrhJTUg{3+mqF=5`b^N;MS>ErQJZ#FiYX`Zfx?S|<&!2UBZ#~JNBMai+ zXJp$pC17&e++Vz3seig0rQ>sKweUA+1pj9jxjYpb7-oa;*{a%Y(IL=6S|AEbaE z-)l_LRE@j11i1w>kR+xos_-*x_R)TET%q~sG!{k1CcZ$(lG$sn1c zn>*o~1xN1>=1LwQ2~=G|(PsuHfUj104aLCA7UN@_R8^vx#i+;slV%Ag)tYBn!!%(D z#bm%kjX^FU4`LScpSOntiD}G~7cJhPfZl)e+>Jz8{%qiX(wxeu->Nb|e!(LO=HwDJ z6xNA?6<=Spl2(Z%eu=t*gU@{T5hh2j24R3{j+XVTN%-XjIHUSI65@8VapG*(R^^DQ zp>~eXB#p*0R4IcJXvc0d`rxdSQTex^mH0g{+%24Ljrs{D=k2mtfS~%*g$%&Y3AleR z9~mcuC!xocw1lEUQj>i0ef?pBd;*I(>VCN5<4y~Gg9g5VN@RUV7xQ=^^SHaH4|>C- zK`z+~D$n4r2U@qEFH|QuGQNBC7}Cd&^jK3}avtY$Si~4Rt!v`&>O|ob^Ifv4NOHB_ za7ELox?$e29<+lBq_QMAY)TdxyQY6UhI;&hVrb{{9RQQz;G|BzLQ!E~Cb2P1q7JT3 zS@6m}`{{aD;!F6DGOV-H;X%NNnL5+E77TQMqQ}?4v8irj16%VQ?3u11mrGUUGN~HV zBK?7hTZs4ydrt!gg=)1bOWvuiFDwSOLUDkaPy;j|aJ_z?(66Pz@4FH2gCYDmsy;6jUg8*++xn6dG7U*V1|SJt8WZOrc!OMWSEXh|(3R z{qQZBE;u>28qcfNuI2>_*T)nMHpz=Qmy)(nBHfQr4?0kCmNz)fcSQ_4xg;ERhs2WH zr%56Bz&y_E{ZZ?I_xf3KdFX#9gRBqXPiG%BTp;3r3p17KD7;&>*sr3TU}Az_vH%L| z0RpbJ);JZ_gWL>PffwvS@*&*7q$pYaF;7?mN<3C~VwxyMDDtcoOOKxA#LQXt;B)qe zu#egpX(&8HOXJ$1(IPwM_gpCcZ+Vg#*yyq(XK6h=k@xv&Ec9(oKn4T) zZ>u~Faty(LFrQOfba~(hL_f_fUj`ARSim%FdBQ$=q-xMJtR&!T<0DHs~T`|yqzuMlnU8RXM|=;t#nIa?XVEpO6pD!cBZN-_G{LVf8IOl-jcU8 zl)E)+W{3P-zQUH$yJO?`XN}D310@ znvL~)FBfm6c;q*F>V|n@UtBCn#u9JHDa+)^kg?y_oE*N$?l22^cG|~s+4oxe_X#4p zkH1-!Dp7AZpHwRqy5!1|?Nf`I%Jp5)@BYibeoU8s7tnuXXDaiS9RGg?$Y9Q;XBJwsucKkYFdFReJslv0)NbL@1vBm%sg+=U1ppXg8x#vteD}`NJ>$ zy7Pg&u0b5?zjpfHJYgiiV_-clDTy%0OLhnjQQ_M!4%(U>Gxi%a|NZmFSQd{N$p`XM zr_JHX?g4)R^cInSWvgfK>q5Ubk)51kt2=KwimA;DYx$`|CR>`9W&3xwBKo~T%nQ7q zzrZ8z=vT(jN$&^!ovl7XvHvlANN?};u_PMbzz9Jk8Zzc#>?|hk$f;YQXhKFYo1V_G?3XX{4cGpC;Hp!v)S#FHjjY# zrThR-K(N2t`mZ0ulwzcc`8!t-TDJI&M^yVb{Qnt*{`_!;IQ;S zu2cLB1{q;e3$owb`*&k_oMcindNnAtaq!*$Qk5xQey1zzp$tcfQ!MWFC%knrAL zMLOH#P`N9;r+W0Du;vrttJ^kB51+-pOQASTn}xf7RX&@brA4>fjmRoG^u?~Xycu7Dt>14bn8MaF@EvBP=hUt8K0fbbQ*AX0P5+{Q zx~ZAzdaawOFlbyp%;SP*ak8ljmTK+I2;cC3{Dqe>z010Pizo1Z97wwoaR5ZeiV}fu zKf~7#km&&BN~EoSfUL~Q6wdsVsSuE-QeL$LAOpQH=oKh6CEl>)Zczz25llX>&aE)g zv)cbPJn#kmkzfC1Rzmo5!5P{))G>!`kCx>7UIojxV?vKRV$Op~*(OT}f<$tODTT{_ zzVwGwUbRR1MHI7+^9w*Dx|4YRHjvBzNHNo~<&*e=;(f|2k(iNpkMF4Ys>!r6haho# zSP^Rbb-HW5@ACgn$9o|5BcR6$p`+0c}|$+6b&7aAvj zwGYDWeuS5n@cWKm3DphZ3i@v?`sR;+MKLbDj$B4sJ1b}2?Shdx_bL4doJ*tVmc{z>qd6JN_E5|y`Kyuz)Ir6gz@;D#E+;9 z;QsD~M_}?lBHkTO%92`!bl5cVTD|^|$w_;M@%m=Z!uBj*u9!4untd{m`v1OH@gYd~ zk7tW(_P!W5+}*}k*q1W0v$6qyd2COv zOINQ&^8-EXesiBtn|rb|y+rL)Q=ji&kcV>K(vH?KW%m)Cz%UcLCmZM`Jl^Z#WY^kDsehhl+pWwP?4 z=xm~*sFgq21M2GUjvy!}Q=H&lJc}kekNfMKeJMTI+Dv&w_Qxd9kSo*L%P)Nd`iGU8emR~9YeB27=K^d7KKWGHYSPQxqW) z%xE?}tO#BocN6OS;oFCKPb%1caadCSf!QnT#MpfA=&L@wWziEx>~Rf9IBmX1Xw+|T zY}R%kv^x!i9eD5Va(S@TJA;k#D^UW!_y#fk-KUs}{Tb5(-x2);4jJ4)JcL7xX5k)Nb#Sq!nUS{d;lV6|wmLhcZ?;QRc zwCnHRS_@pn=FRdN#}-VX|M-9UBmbZNVSv|5{jntx4ofbYZ$9Kw_z_qOwL*W&!-aOg z+o5neY8BLLZOd7I&j_9FX@w;ujg}~+48}{Pvy&`|T&XUVaDRX}DBqn`oSaG~W=Xf- z$W(40`d~$yy_;uFEsi7;)da)PGG9dl8@wmM7=?|t2CjY0vG+#ehdbr|G zaWV*KXfUm9I6ZVZx%LID--4TBy^*=3kO5Y4Nsgbs~{O=%v`=i zLB7tuc8oKW^XmdI2E*xb{5gcPHT9`YSPG;XAUHCs$b06 zNDhbgduzCV-SpQDPnm>2@tmN>E&ABi(_Uwt`*%)9KLtf?X zoT4FI=}~T+Ygc)ODUlM%E2`W{*~Z4moV6W zo#g;z|75vWa+cKltf&}GzYtQ)mfxcJq)#Mko>6Iks#q0gw|oAZ*Z2tWi_DId&OtH= zBF!g?6gj3f6_TtTZJIj0`E}tT=tbvSZg#L&Ov<|;&E*P7UgAeycch8T<*|3*kTvV{ zKsowBK_Nz9-pwU#DJ^h=S`Lkpu@zsTFs|Y1?$a=+q7PmA7Wsj4V!12rxm%WeKXwXX zENpOpf*KLn<0u^4Y1tFD=Oi}Oj*2Ihmhzmh77)z2lZmk&0y2*o%M#giDV%k}fnO}L zf_+=sT&Q%B#6Ea^=?C@*6QP(sS~c&n^>k0TOhF zoGUuHE(Qg;NK;reorfg3xQh;ps%%v1_NA_G5Sxl$jRgjBD_b;N17$jlHF%8x*Kj0V&82>Msb^|8?E#PV}EsYnm^xk`QiRb$kau@hA+81=s3_+k#4r^RQ;x-S{;Lbxm6P#`+N{%7a^s#Py%KGs%jL9) z#72K~G!;howplF~3YVown$`H`$>J@4p)(^ml|#dLf~}KN(nc!mk!*F-RU^j=N9OEM zG8@?f=hhRI5jFs!ft=rNVJ`{nsUo&wr91_mm!Jk z>byT!t*rM(p34BmaKQ=ej>ny`S*MfSOtc>sxV8mquhKWW0`C_Zu5lh2vGY?E;Hlc) z#hp?^CRH(|e6o=(uCl$-8Q6b1tgm7tf}z4k!$wU1a6RmyUr~Mpvr*$^>r=G+!Erfw z>|}KS47KAUUo@@hA?mxxqSgX`s>u$<>rr=DG?q9A(55#&wE8`r*#(iN0FO*JQ8PD$mH}%XLgBRN#jL?44G(82q$^P)kJPDj0pQS3I1E4 ztlIrag~mu&?Mfg(Kh^DjV(Nzi=Rr+|0|4~m0ozU+K|+8hYD_{gX8h=IMGMAkqG;AkSlbU6GnCtE(UO-zHQd@81ND@tq;k&t7%D@QzCK0G$a-*-;G)x|8<;~+;W2y7Ro9D26e$lrCbegsls#%SM@H9rgE_bw$TA`l z)3!U5gNdmW^iev0Y2TQvCzO;cui}F&Ev;{isxobIkY#o{eq0|`?Tvg%Q9aRV+Pg=T zFHnftK@q#qMCJy3g`oiDM`-T!gkYXbo%tE|ZNJB;?4+=<2u=TzB^>E%tv*~_?%GNH z^e3Wf$#lWBqC@!(z7FOV3x1^2Nd=z;JJj znvM9a62EHAwh^F*y@CK!Ny?i)rF0x8n<3iF9kOyum^s#4)liDZS_mLHY+B%Cg_UC5 zTX6U`kaQlpsy3Ql-w(MI55tgTTXb9l{A)zqgZl>&pZyO+3<5nV8o5F_^{1O}&dwPH z3CR$*hI@~HA}=~hmMg!usI}FjEK=}H5&QWG?uTAI1E6x?ky3fV>{m+=jE@37cm$=#7q4-fA;j}EoRCv^p_wP zqasd!WXOONG|7~IEr&0J{^50T{|5lVsjQKEY9xX=XlKncm+6ikFD> z_)czdF{saU67AwgVQ6d1EQt~l? z5@}VAaxr=LOEW_q@4FZlCqJtaUx;YMbn4mi0L+zj9-hXr-FI zOZeQnll`23z17daaTHm`+^bxt@n&#+yvn4i`Q`au@3wS%gQ4TUmb)ODKh%W9@PG7w zeXh-U-+P3sjUNye2pvfQ2S&%5H394!QPht)C){=Tv%0^mWdgHH@%%`=hc6tZU6Pfv z8I{%5%E^i!VQ=B=>0|VpFx@Sa26$lr-xibrsX4FL`>c^x^iB@K1ifDAHQ68avGodKue-FeXLzPsDPR zhU)`v6jagkpJ7EV`9#<`Mm=~=D@;-#2Nkz(PwxoMZ~@JRp|jwD219#$hl`4(@c4?Y zT2RaK&0fD~D-A-=l5t<kb(%A|BplUBNa}fd5cXl^t z?xNV^EzFJD+?I{lx;s=L@ zkdi{sRHr0=FB#Ti1YT^kHY>9(H<>nPrE=8svg}1dP2!r(OAJb7n4PCxq}ol{oe~uY zb{`CW!8b5qgu%Fh@hRqIHLKN{{&0%=RI#%n!>bgGVk#8@_gk|Z4erxICjC*bcw z3by)l@GTe+$%)5mQ|$zQqj(rXiFkW)5F4RS(|?f)tbadOZANwZp|)S$0b3)LQC+3_ zV7(BbVT;!7K^(GYE{9lJeco;6JpL53h@0X}4&7A^?(N+&m)X9D?@=ulq>6sj695^G ze2Be_P|z5H=j*|rfmDt?b2U3mjY~9^D)r`#^J))ToFDOn-OjOpCJ&3pj+1%6at0^P z1N<#smbCOHr6V7Y#Eh~K-7tD93Pq808bBy z@Ssq1;TBz;xIaDap*OUITM8bLw;z(t=}W8eo)zyZhJl5zv@LkE-}v~Uw;W1R)Hky6 zOz>Q4|8-$iko>uSDhGRF3WRN1+?wn*=&e;@R^B$T06iPa2MIpBA$^_Gg4q zsD_FUCUZiFuf?pwaCwE6D=kNLYQgt{3)}JqauGPN=XTSBFzb1%=T+;sgzlECtaW=V zJ7;=$4huAvb*CXO$FgKO9hz%Lu2Nv%HN1;ov?l7St^A08rkEsKCufDrtmE%gQ-3E` z(qx^Do`FfDAXiLjsjT)Xjtopcx%#$SDFA)CH%=PBSrG3|{LK@!3m%fPM^$CAuw1yi zB!`gn4*iD>nTDtAWy)~h5n4%Ip6%g`z+v4fi^GE2%-t+DcXAp@c3V@d#x{j~)D}I3 zBaFn+49u{9y8F*V@mG2fU&(oO8XhIuqYF&QBZQ=cdAFR4KK;3wIstU?zc8UXYxAfU%g&mrm^A70l~D9Mw|QK*nEz```6gvlleoAi*g&2av3Oz0=KMw@}Yq9e;FC z6u?1>)YJcQk7c7gsd}bLAVg9ftLKjE$>>9gQ;$i2QCxCM&%LAP{aBG8p`m=kYC180 zAk-h#f#_Ys)yH}1VpA%*&utTSz28;r%)>)@XLH51sLPoP7W+SJBpDH=I3wUt(bZq> zk0uh+x)>g-F^P!eDHlxBh;zs_yIe#Nz2^>yrO)OwJA*@v+caV{*kWv5*9w10&71+<-sFN3FS`N1TO zrPmhOgf|9}`J4NbK(u!s4GBH|eHr?*=?<6atgu6})N<6}8#v)51^xm4%K#s$YOf~Q*&k%54^kA$*KtFKOf2CJ8L z^m(5LnOwEBb%`fN=Oj2E8kE51q*cy)Bh!9|PrAR^1LVmZ2wY#(2i-Y^cHd0;6h|d6 z=L*R3y~*^D9=YOEQ|R$enT-d2uFL{HMh zeEuw+1xw*r+BRWc<|_(l?lL2PJa)8f`|w@APu?anPcCO==~5NTd#e>RHBF4Qx@WEw~O!lo{5(Cs zs&6k6Hw4;qJXEz>B#%*leqLum=7D@?wy~oAu$sYYOwMTfXJ9=|kGCubv;s zDYQhlwEHFM_PqZy7l}JxE_poA>Gvr{9)LWifD+v2n93C=>eJ?bVJ$l`zQKBv`BfZ; zR1`F2L%Tv{AtviBP~W*xuh@vA36hdN+(RxkmCJOh22waRd9Jj4T(372ixuBdAiIb1 z^3)F`GRY6|0;+4-EoW&TdMT1g_of;MUlVfn%h$A?Q`@b;T(1XTkCTM3qeDx$TpmXE z7uP7X5d5^xLSR&X68CtG>5GI<-&lyNGwv>(ha{ApG0Lh#IgiqP4T)4xl=Jth!8LL)Mp zEyAQ2x}G(3(_maqJ`M-&ZG7rUf~2r)){m|gR^x`{5&=aoQE=&(+#VY7vb4Q##t!l* z7S2n3UuL)I*u-p;oGYm3jOWpE_3NuyZ_~QPm24P8FWn zQjw#;$3*pp-MElzp|Lu|kN1+fQB{TaIIiBDejik=88+0NYHG{MYHnL^yFW6gjv~?@ zY^y~aL$&LEbu7pY(ZFXuNYOl!$Wo{|l4o>kD{z-~t8Qd8jXK-68^J@))^4{o9c}aS z30(Joge7!v;Js>T96LD{oVTNWO9hn!OW>@A-B8v;2Rc&J2(#7~t>4>hnYg>t30sK1@mvg?5YHR-+|y6`bJ6c1;llNq&HO9>s-g_u;%I0czYq-cvD^fiP1EL)1_Q3g4g5o1j(1>vrSbC z(TdxIdVS8@qzb)>BjSjMRm24|rH`P0+6pTsy3b4wZ=Xz$D?8h|7Kt&`pRBs*s_V-` zm}K^gElxgCsNh|By_QqAr1g#Cm;=480ro@ZkheL#13MU`)3sZ5pqhW*f7buQ8SCjI zu-L?9E5gj??rP};_F`bqUr5waT3QQ{UYYzdtmJ$5xr@lCJSlZVhom)zS945%+;sxE zE6d_Wr-rHh-{DL8K}5?DJ@B z=rY}QP-4%I=z`?*-m-3j0Zo>p*)>mpojz_-ZlEYLM1BvuUhpXQiJTbU1CYv`Tt}P*gL&WbN zvirQpQ$6m<$V3z)k~tlJYMneC^2)PBMcm1DuPzcT^mod!PWl0y^AIIR{cWN=sV1=w zU@4ZEv_h>4{%pp4emt>5v6@Vt4!hH3V^;Y_AYf2kMT*g+i?nvT%@qw)W}OL_Bozbp zxZ*ROYWm_XoAOA^{32W6_3;>OYtGy+oT}s4{ICHDBsod0<%`&Vq?YBQPj!PH=-mv3 zWK|feWU2b8S7YGXl%afVQiYnPY}bq-wC{b<)IH0wH}$u*|Jr{AssD)1!T%vT2P^Sp z-4{l4<696#xmK>$DriDKWk3`751>Bo=e)Xz&_r%8r2Me`)#lIm7>hg)+3yrNk0Bp+ z^n{o{{Zs`UbNZrxh;?6l7e{3yLl09p6;u^;9Qt9fwVe?fW9Z@#$x&r;CBPm5EkU`Q zJ(#!=QlHTKsB%|THB37&f5;b?5FMRud~X+DK6`{uVdV{Ov)hZ*Zdx6tR&=NwhkTx@iMLa8wxW#NDAhkw*`u<$>N-*_*q7h#B%NPH)%=swB6ZT|0d zyJ?<}%U!$n;~X90GOAuJOi{G}UzY82MIsOqZq9wf=GWKH z4=d7nqQ78&M=KW3!Jo1}Ax^{-fW5qszxhf&Ynn;Y=%;w4qc-S0Wd5mqb&(1b3`Qv7 zfuyW$&qlyfSYy6z;GT zPiQspirKY56RMpId;KG~x^n>cX zkzT+6K@ zO+gotAN_4NaSeOex01-RM6hG$sw|GMSlE4+6uL`P9Y&bO@g3%=bG8O{_Aej{WowGe z_*0?fs!pE0_F^oxbtBPURI3eMiP@JQA|B~~PYuheu*u*Yihggz<$|usWIp|4uG}l; zL_B(XR*r5<3(=}Im0h%Rm)@g%fWg^}-urW9K|u4!V~m`C z8tD^F>u|Xem05KZ)z?>TsMih;%^Y*R6se@*hfPeYVPz%awlKC7xz;~r}%EiNNERiI1^o#u`~mOG?QC2hmpUt9Zp*Us6Q zkHXjoL4Tvq3noc^X$f+B*}qKOt}gw5`du`oRWdg6W2VPpG#ujdRQli&095q9dL-m4 zn>2;R#5&duA6AGgR6$j$$@PMUh-M|`UTXikA>W?#r*zN-?}@QU<8gk@dIGYj*V`&q zkAT$ZZaj%iQ9v)i@?;ff`RhyVxu99&shr_+>%A8hgplj4)RO}T-X0VYS!=m}%fkgG zOn45FbwW5KomZo)9H3;)8Jn16P%(eEQY0Jc+U!%zzl{jvif=npLT#^fywhgi>A>K2 zK`CzH(XXpC2lXx->D^MSA9gv^?5VMxz`5S_3dCn+r+uPb`9mivOW(hLE|J@DzTVD|)PJ~l!!2X76?ML*WLG)a=qIT%R*+54 zS$kLb1%u7)cHxrI?#A?bEtHqhRB9QsmF^Jh7n!a>4C{n#INnqpYvNaIozSf^QHXVS zOFbTZ&eb-r1`hGx*?ZYMp-HWnzl@iPhhhQca(NK>vo8ZR`-eA0*`cI=!vsO$ok*3{ zxIvynfAYyK7&rO#s%yEnkJb;?ztGv}JqW)SrS~P?dyMV0(B|A6F6Q3hX2<$RLKZwh z90SuRuL$fn%X_EVtVQvv^Fj@0@0&FW%R13P4d#m~;d04SZ+4<7Ia9} z~4RWNzz!pS0P09OL@Pi5rVP zm!-?p$yqrxr?%`wMM}mzNw=&Y8<``~ib6os*+=piKvtvmgkh0d@5}c~35K~S+I>Sq zI2#R}ZKH3_ce~u)qOftnwRp?AZA9<1pA&XQYl~U2S+#ww-51{VOUUm;%YA7+J5n|08Xv8wZzO z^v+D%UY)}m>7?`_{qPa5Up(xSPtU6NOQbinfCdIRB4c>OZb~eTEs2A5pznEmy7hy$ zgw`;)Su~L)S?}j^Gw<2Cv&(jJLx3oOAeqJf-wq{B47qga`5x?;*kJf}1=wk1|5&0Fqn;j~9I%g;f&~zt~ z=)B&VZGCs#y{B*3Ol6sG?e?X%PKvK_>6+g&lK)++=C5#v4h#pjvzRT?1}?cv-rucB zvU7k&tuK@fg_|0-`lA)G<$~E==1w}lkBe$O+(XKLod-FTZ(6ONQib)rRkuj7+EyP7 z&Uz#bf?Bp*f?y7-w^;JO*7M7};(wU9qA!)@T#munT&+#){llIG=LQJe6d9bzV^eW8 z^5U{DfFdbv8Uig*y`UmF#0iMSBSgex`?Bos(fO;Ud5?Ys_qP*~j&L($PBV`RYf%=h zT6x@mWn(-zwHIi{+grkE-{XEgEws5TXEXDf+|qgdr-60ba}>!CG=!wL=^oqOJ8sjQPf;`w$~Bk~Uxvmm*N6Ju+GS;FETzihq8W!C@yq1Heds zps}rLmOVjsKr=L#*I$gM{+g*F_`Twe$V<77gRvc9_WP@~DkH&(?^uRT6LH5U(${XK7CxK-)FDp-^;}I_WRpYa2l3nnlB37Rk8j2KqJ#MBtpU+qf`9$iKXpW<)8u@| zK|&`b>tvS@77K)=$Ms6aGZ-b@-SrQfk3H3x2a@;Hj2un4*>AQV3E^X&I3@%B&J#*h zj+lS<+OUs>zp*obc`5HlECtv{rnF@vf)(s!eaLRuf061D4sL$n<57DExVr&=l~|t$ zQ}i8jH}9=`LrWYjLgOW}c+hWc4bgQ5s?!*zIKAX%BU82dWrW@szwW=CfQSB*IN7T6 zmRvTKw{|}>-qGcKap22b#!~EgKN}KLMQA=^XGPd_ztG_RI+v7)zov-8fv!> z;$S3DKSC-TN(K^}rC2N$-#rqiEZ7`9n%`kPdvfIXw_&aI``yr6Eaxtd89?RfTNnAds}_aX8vyR{I@MlJkGZU$1Ugn)gznM3;iQnRL{gCL3oayYbq$FHI@I_*mo zlD~TMb9Ddk-`VHA0jK^bk)?Hq$PfVCCMH?a;$^zoFlLF%<>7-fE63i$J)T`GMkgc~ zdwTZpZuOZVZu0?$t*zPHkl#MAEbl$)4cX0d>O8AfV>s=*v zV}p?pCJ`%o{wpm`#KcU0{q1>=Ki#(A|Iuxm|Le2Wq<<(Ct;Cq0>~hqHwuv{t{nO0I z=-(&T-_Ji?2Dtw<)&H96e@*qjrutt~{jaJ1Pnv3#tkeOlac#S1aJ>n#*a#YzjD=EE zqkrU7It>P|Qy=vlDoLr^?Qpq!dpC!o(@7ElM?DU};P*7B<9f4yL(YD^9=RS72#8=B z=3rvraftZ;R#Tt*r>0K#pEdQ%(mBX61>Q!iru{aTru!_}N%KRpIrD?bbW!X={iV>w zez)+bNr~{xMX`E)mf7u4x&482ZBB3U*!m$S1`w;?6L^9zs&d@bi-4^xe;>+oFUR3< zK<{GzKz{(LieSut672{iAv=C}x}KxfI2f4Bd0A`C*VKTv^EUrFq({$-MtQUeec%m> zs8M=Sh8cR*Er|qZDm3W6D`fT6> zd}dtBQxz4G;=!kZ%cAl$F~eH*Py1z0O$gtUzbutA0z55$qZsKIvoV-z7QPwEL;T2< zvBS&a$0_X;0+$*xZc%Q|^5!OzE@a~5=suF@qAx|)IN#98u8m_B(z z#rEy6+jJ%3cm2n0G@jq#f9dN;OFmkBALH+cGV%?hlcM(KFNe-Z&-#ihFOaO&r91GQ^fE6 z$etO0BtXlEwd_j{WTqS}h?^Ix_Rn{0V4|^M-m+JyzZJQRr^@#t8KUK3sj1#+FE_b7 z^e^Y5t#lks@;6gzJ_BT5(Qfws{`JG#uuDGaRWK%01Ovcdn^=4NQ5j&p@N3Cx7j7en`Q8 zuZf%YUV9niMVj9yI1J0j$U)X&wHlLC2D)a2oY+oNY;<5q9tEdh9rUwmn~PCBPxKzr z-V!W?d9TSu7y^|eUoh2YvPQA9%3^No=iF=KG?5NkCuBxDB=UTq^rX{tC5tYYn#$_t z)@fHYj}hs!89!97&s>g%S&xd8ybq0k=$+Fs)aS+7;~rdbY>ug_vj3WMF|%f-x4G}c z5$WjZJT6`ptvKm=a=GLo`m+GRpwIPqY`R|( zfLgxGbJ>+>5&2Z582?@a|1PJUxWm~TA6Z|UKvs}QXt`3iyF{i=4a>7z@%t=)6Ggv% zEP--LpmRQo;8D0%vD%hUe>Y(O!i^;k+p#i*@ua{)xcu5zX(m*Czd={&WlR`G*8eVe z_27T!iSo>m=K)g7n9sJPX7r!7#fFlhEZk3!(1fmH^Ru)3n~M`7D$Suq95x_9<}uk} z+m}H7g}PTaqIZGx1BNg7$ThZbgA0+#6d#JI%yf8Q8DO8Zc#5~6htVp6tvDj8ksc3R-=7$PK7WaEc) zpn0Un#OE4s==%(4u~6Cr5KvAQcj=svEPV05Z@{7om#xfFvq!U{Qud*; zqU#YbMTcGmS-j+ZG<8(uPyUxz0)zRt^Muvr(eM~K)Y_9yGHbs1R1k*l>NI%sp!0c= z5iS(YPjJxiNm3o>cs${?Hd~@QGGv*0y*h$Ic_`Q~Qc{k&pjWV=)kfP^=q*<(Bq|ev zo46nGDt)WPBlOAA>%(z>s{cKumHYnIu*$N#+j)wZe_TVdZK^vLZG=IA;_ z6r!5%Z$v$w-;Q>t_ANHLWM#kk=GNrVC1)v2{u8D8n#hcVw``70?G)eSx@{FQ-ESe< zg4V@Ob9w6XMPZERY^?lxb$05S&vX3ev_u>UI@??L%Ds^<%XZLz+N}*c5yohDD?OAn zC#I(y7%`fQpT?@62nc+C@1b^?%W5XoSTVn1vLakUwqzn%Aa*m!Y=vR0J{S-oD^bRj>^BIz zQwZ%+vkAh^cyPWpgJ(0H^n1nWs;%w4x?J63CUd{b}i`0@5ZTb9QB==R~ zbbKj#*-$$i{{5}T)v=(bKh)2sZgvg@2pXH3K$67M_kN9k)5#ptjrTFAGp8khAgMu| zj0XI?!H7^~&u^D4;c4e=DXZ8veg13b+FYMO{<#a_-sK;P_UeC7G{4Tnm({G{STqW~ zPji9&F->F^4c9ST7mi&mR>OzM8!8AR3>NCXr$) z*D5vJQ5frgvr)m@Y(%gOi9qn=?TMzV_Vx#7Q1t_?O_b^xpy^J)9=u2gL*Fb{Tvwi^ zRy5WZbMW;*#&yw)wW`rrp1SvXD-|caxX4xTb#S>(Oi@_hn%y*3_#+WeeT*qO8pWdU zRG>v{`j_HxSi%}bRfaXcvY9HISktR*4CRN>?sBVtH3RnZC|Od6Lz}Gf1Q9BWl+%oL zp^lO}_i$FTMOusI)22ncRd@;^Z`C~*3Br}C=Rd#8qHOfVM2Vhkh5CDylvP72)rUAv zk@95geHSm!rVauaFeUk z`tB!x#Xjaf?4baL=YmR2jYhi8!L_H0VSwfv6?X-*Q=LiQ5;;3J`Bo_DoFAuiGpD!U zlg-3H`NVsbCddV++tbXs@{Jm$+7-2hZDU9*e}#de7=HktAVnHova@j)rX9KJerS#x zF;5?h1yA2au&FHVgG1K?X=p-7hI91Z#z4h?(C4V!*7fZqKOv8MygPufASzaiu(p8MHN#c6FMRP2MB!_g>q?$~&7}Wycn!r{|m&*B@h zZ%c1DG>tL-_6}O2d4GnuK%pcWg?FB4$>Y*{od*z$*^vHCE?QJJw2q2|^y#F7=N7U2Xm7f%U-q?wURKyZ6jn zg~~SiYuUs4TNcf|`nw$RpVbi&u^ZHXLO2iEE@w)8hO1qv0vDZCFA-fX?&xs}OpNYD z>0=m;Rl*$iud;N3G+Q^{9TRy$?@Xo&pHdZRoB-wUb#0bTXLMIKpl%}ZzU-WDm5#t-uQ=N@jz@N_vJ2Y9TrDF zQpW%aP9q1LUcMl7E&ZfWtdVhl8>Jh%7XuKdS?RW!pJ&tb3%ZBN`z5-crk|`G3oHr`T+i5zq!FP8X3v@Goonr2t8=b{j z^|c=@owTFy_T`S2)b3Q7jkl=_sCKyB0-E>3rgMfyK;Ff@cE$>^6%*V{UQX(WE+L;> zD|myp2o_CIUYXajpRN7;Jj_`UOyLYj?m2~;VS7GY6!-3kWD<~J(e>e5&2(e&HnU9S zS2T)*Xw3>9okpG_<Sbr<*K884ee7uEB@r%m^ps_gvgmnCv;iqhvv`$>Y@P!wCZ_j>2^ylcZW(o8kG&#?Gd7GHCAE@cu2{{5cA~hgEE~ezw|E!z&gUF|%2ftMAqI;Mac+@L zB|-k9*}3hI(@)Iz7wssjc1z#h`@h>&ewT8Kka&JT=TetycelG^>7;RlCn&0SH?B&d zB)RTBTecPpFigF~xs0orRN9R-o`+9)Xw=QaWOE9z`tPcjh~^I)jn=;u4TPfLrS=

RB4l0FouVtSXFN3?AS>6$Y7QGu##$+cE zS6B`>c5svyYfz1X-n{=T!EyW7d*=YZGa|E8LP1I_c#3MS%SIvpBb=ehO!dYHCaS83_c!_G!T#@(YYU!t@ zV*{gJ7$!ht0$@C0WR&Mw6fGCoQ;9F0sx1k(>A2&X}J6p#ho?jHnT&Wa+l@5y8b`59X+q=M%abPOr0sq4{BNG+wd;cWsU8?wt z?(&s%Q3M|W)#i&A?8hA#^Nt$#9(UJ%A8LvM+@1CXI7%&VcG&gjXV6q~*B16$@R7lK39X_rv3TD<4_7=2Cz043=bont_5p#r7+Phq zI{QN+9F1jN0&Qq`&yJ~*sX5Znn zyq$~Fj$krGHk55^)^?LYCTlxfs*a6+qEdgy5M0D-9*LZs647bVyyueoigR!RVm!iK z5cyFl#j99y8k26##mv@ptvajDm%q?rF);qNu|Kjabn26yDQk%d->|ecgg7P3>u_c8 z0`94H6nSg*|<3i$)8&(#kKT0`TA*XEjVln&XkWDkRat_`7g{> zf@uRMXS_69xJ~yc2qG8VZURYv7LOMTZt)GxUEG(oa2ZA(YDWUL@L7-cEarHxqr~E_ z>THBM!0HYkqK1;^1WZp=%jB2bfn{_H!yE<=PBq*BN058^TB^YUsm&vUINt1^HNV%P zt%`lK)T%_y<9PBiC>UvfNzj85&cXw3^hOw5ruLsV`~LSql}D0GSxd`*@w;mn_#?So z?`0Fi=bM7qW;hZ*6noePRiJT6Gh^5EiN#9icjuo;rVNCcw>%*J!@64@2-GVL>m#y- z>ZJ$NWArzRfdUOD5sutdSFezrPJR83ZV4Rajyiv046)yuvP(KWYehUQiMo8tRk?LCDu^FnefZ9cwy!;)CtJ{MyyZEy4MxWzos*d&`f6W?eyG zR)$cVh2;8+krx;ebvgMs*GIOM;*Cd3ncrn!lM3X@{Y*k~&+{Oc4pVjqr;&zuUxt9d z4T6E+r4#`u1eA`)SQZShY!D4D_UmuC8?CU3ef%iZ6N{f1ERm#td!7n#&D14a5-m9C zsAgXfU%Y%3vlYGE0sne^D3&$!J^>Hdf(_DmJELgcOw+C#UO_&rD=x|~1>9tTLd;@g%FtN^rkxYmE%YLO2bwj7dUb_ zTJV``W5_cL;YXi;{=sH(Jod8)9rGPqia}{m>*lybKdQqrnQ8!4K&roU)38nf%ZVj7Dj9tI}u3U729g<$cN)fr1=K1TyD?(F9=SQvv zygSGxZ(l*bVlff=Igj~%so1CDpd{D_^-{}luA}8O4)m8b5RH+ae;zd#b>%mrUp{{Z zX*B8t@SrqK`89BTLW$756$M;GQco)e`Fr7jv$vl6E1Ya{YhLW%s~_3bFn#rBeC4#` zq$w#~HsKTbRY}bzjcqQFMRm(BGN)8icKL(;iBfHqw(bI}_7AzFrLX*&l#=MEi|<-) z6^04a*R%0V1ewiZe=x#FScqe;t@b3*pjw%t6l-S3$>H4BUWJqna^+~XV{SC?7@2n1 z4Gc_oSY@o2VVj-ze}rSO4B4NrRc*SyP`zwu#W}&~LRMTHx#K?giPBWx`vR&2QR1U& zUsgjuk>=2~pW7gz(?f{iOHB}*=u3J`o-d$W!ShMc(GA% zLDGHg$F|ig{Pe?&yP-)rH5%euMM`Vc0f16X+)B54hwWgDi%I8&7&6^?jGbzVvNOUI+AJ-9+Ef-0%i4%oxtn(Fgx zKbu_Paafpl6wsAaj?RPOrytC^F_ET=?Yz&V2{DQ2nwJMLRy%Cm+8UoVfbWmirr>@? zIt82Ae@-9tfsTUfN*u3RPLpO)h{T!hD@39^IX*I!n79v1`>#zCcDbNf0AZU?pYP1D z{WpB){oIbv!E|M{R}#7C4=le?V`5WY($Q*A|seQkMBqT(5k;Z4q%% zq^3dNYEqZXT;;+8-`kI5iApY{I`Kx0!shc1F}cRIP>r(eWqNC>WgN=r$hda+Jv&~x z-byZ+1MST5N7ytrK!B>y1UVR3Y>!MT;*Dl}+Bm~4UPVbO6GE3w((-}Dvi|vO(~EO+ zf3Es1h0BA~f-=Ib6j;uqpdW^lUt1O21S>)hcHF1l_rKnA3vxp03+r*xT61bOz3mP1 zNJoO}T?7~^!Shg{!KPdtdTd)csRkq^FSEuVfCJ9Wi!*r3`M8?P;{-+m%KPn#*_)joG3|a~gO-(OiOqT)hxKh6(il!o>FOsI{ zxBcrwU4v%^>v-ct3rf|%{zrxGuHm|a#g6wb^98^O9gB_or^dX_GU0Q8{PITY84on; z-ec&GY4VVSoHBF7BtIkO3pGp`FUSpC9Ol9zCT;(K2gtqS(2UgKsDc-Hf4hux6pcS0 zKvF23o-7+3)(`kz1$aCDDS-I>`75CfLxBQfiXAOq{lOTxT&UPo;- zy?_6RFTR>loM%J$pk4Qo<6>`sHK+=l-Z-(W)L@#*)c;VRI1Y5c^*;X+$rFcn*=FxG(?8~V0TpUxV5aFcvXNb_ za%nJI5mTKia`aBtQAO3+1lm^S{wDAmZ*QGYTd#djk)OXPFH+%2hi)J{uR`wDxrrj)e|?+EOtB_v?dlnF z|8A7Q?1t9p`ZKH>8`pm8xJISc$@Hfb-iIi|=*z}2GSQk(lxi7ImvuNJw^Y)vMfddL zF+5YRxS+f7fxB2Cvw$R8OkASX6d%gTA~Zq{Z^xt?Z97Ey8e9`kZcMDdc=k7Bf0`b& zhbe2X7Kw`oSZM5Ie?Z4?PQ*FZes`Zid)rTH|pDmnSkV&7htFz#rCFXIfa{(yQf|Q-MY@nHM z(%9}>`1OnGs*SBM<|s7QwSO=tEy1BCN=Q3-N@2RO&Y{i8e^yeB=`=QVCVa7ch5`-e z?`#iFAGQ!2@yYv8?*Pu=Mfmtujl{C($oCQVtx~UU2|XxRRv|KtDp&_QuUrRC$rn$N zF6|uS1Dw0qXIGE?vdYm9I3Tpy&?!7_A!i4?LO5ubBQ{$NDZQK-@x@DaFsf3~+Ssfe9D$cO;hDPNLI=5Ai43iT@-%-`c}p$=tzhT1I^t+aBucW>kQxFeM`RLAnWVOt1 zviqI*dHHp>0xx_bs&}g>5>!LviV7O0P9u5ZG;{M7e`Th2h9)z}d+AzhuYn0v6YR31 zbSvxlp4!7r+)eY9q#kNoEVcAJMCk>@QSuz==}Itlv!j%2+v#n>pU{&)1>LaL)L;q3P4U6?jovAC!-B(bC)G>gZ4zq5f;D$G z=tF=(Y%Po&xmoq&WiC&=c1Wn3HdYwqtO@ z#apjkx<39=_LrC?uoTHHp1enAcY~};#>8YPqUkUd&8?Wt%W9(H@$juwJaCq6k5V|x zW!UhX4F$NuJ~_+;EV{8DdvzOJrRjrTF)Vtaf81SZ_sHzo))ad~JiPJ-Z}llO-Pc@% z3N#;Z^upexe*8vxj%x=-v>2CYnW5^E!J%`f&?y|Spq*s*{m;W4VSi3KWTpol&t$ex z5vUmX|2q5`Necy5T*+>9iT3+$Z^qa`wP~-~Z!+add2Ze)n`kn>ZND?~v&!Vni*Pbg zf6bD{laEA|UsuE&G}94>yg!6=Pn%uI%np;?2@-nbH=ngH$1hb%Z7zo`X?=jklAlZ? z?fcgHeVQV8^c}TDllInAhif=$3y;~`sdhywN$t>zBvMr+=_$GtTe+dSbePH}&&$nJyvyN(1w>agVaCzT|e2DWt zaR;O%xwi1U{Mj8OqghuF8@%{zdZ{Q%2>Iu1WAOP05rel#?i!j|Ar~*dA6P;T&tI-| zcr@s9dVcML4}eQLdIu-_mZ)W~J;-RyZST2ef2P-}{JjuTqVebI?%JzrzZfSVyW7TrX#Id-Qr--vaNF+1yX zg5zk7aVKZIDccJd$1=}vfN03gfBMq_jPP{*sx5$6uMf|o{-`O=)wueom3=3R`}n-iq)#Nd`y&Fs=v>4x0}i7wY4Y`Fx4<7Cz1t`gWcmzTT&p|ZCLs|B z8BJVahhBsod?slgoj)-KFd*_jbAE2Uy86bd`Q*&c{u`<|>P`pBjkxUye<4A>KRpID$ZaKHZPBuAT!EQG#Z1>ANn3oxF zq0}h4sC!p1j-oAc^>CDjjbaa z&b{p|lRDUrT>TG98danWoV4R@&nWtR@o&b@K3Y|(x(QZU_CrA|y!&{^ zD!{GFm$vz;Vv9a(GnD5i>$JSviA}x*%Zr8pw(*cvcbH2$_hBMb_%&Pv4J|7|W{<-O ze-@NP;EnK%NNh{5e?qe{5Egv>?PQkYk?fkUX>d&4Phg(js)MBkzGn{-guhC4eNgv4>`s63OgRxRmT=>2dhYGu@;4_37?8jxE#4#=%=sF&r+dF}dIjmC3yg zAJ1X-ukYO7SG~zrviVdFm7@nLq!Q}7nhe@mgGp&?eEuoCe~y(k=tv-b5{hCS-i?84 zc}3;$)#gM!YW$*yW|vz$#;}1KCri8g_<5o3MXDd@)QjCcdf=hqq?ZEJ`fFf_4rA^wIc@_jY)G3jj5`4MZ+Y5`N^P>qbsv4^KnWsPiV|3LKg6!7vMiB0i zydmVJ!;Uyse{6KDpFxc;>l#}>YDo(xLSJB*e+xVy;^-5+ckC6^_!;SD=GxtA;NUch zqx0@qf1VpTfF_s1AD`jN$I7E02?DTf!r!fQeQ&dI_7DS0FMRUw7<^tpv3VI*egha- z$ZJmOgeM)2EC)xU42m} z+Pdd}Lmst<@OkI{L1(;|$18NEhx^kDY)T9{fA#H{IB~me@EP4fk;RhS8H$h&X$h{o zws8r&4d&6}9cO=MVx-kTG2}D`$Cl2V7z35}_y>zP@2K_+H}poaJcp zPW;?_S|iRzXGFFw zyxbihBGOZg^!S6LCIj(Iy8^+C_#bRrkwxQrC4nrpp#4izbrS-YhTs7?9Ti7d6$X!l z5}~O)N`IqPjGz4V3S;YsNTUrEy!$wpe_E3!dS~%*n^3C@L@UW8Vm=Fs(L_TQ9P>&g z7TQkj_hlvG6N^2iNUdbwV@IKiagL0Ao&k)HvNylWsnfI~y}M6C9d`1?xqlik**r9N z_2Zi5;fKQ*CP?9_C!&yJ&;^EzNmG_9hT*fe@3m8@v)YW^}l-Mycp#FTECNB(NsP4S9Os0Azfk+ z&7hSDzJZmvWMk3~u{{U{C&_g75p|{-i(EA4b#dqvG>(vh3?JB;D88JZArheQj)_Q? z_;38?RAN*Qt*|t?K-Z1yxcOG7nNNl4Sxne}Te zSk0o|DH%H^yGyhf=&_wqe__^N96Se)fZ8?M3^-;Cy-$gP)khoP+5NZ|8msrZ+hb6vJ#M{1F#46qx&e~OnMIYzlK*qtn2 zvK@OF4E8xB58mJabdO|SYG2U+z9UpH(7nx;;qzBq{@!|=-r2v?+8apU@X;k{#8>9Z zu&19S(;@Yq0|v=gM`{93$4wh;qNx-`pC1=d%{^}*iDxuXEZa`7#vfoyEYccfCc&^r z2=IOXN~-dkI|hUTe@dfe^^0l4^F=|T@7_d1oboTT-%=~2r@I^ZUj#pUj@&jzX7PC) z4}kE#Cx}s-P=lVMb${*n1P9>h?L}p9jO_Jt(r4P>sZ2O?n?87Py)vj;X(zk!NUQ3l=OukPs3A1b2tvPH-ot zmuz47M;f1v(y;fuw~9m-VSYoZfeezk2RfdFK%9H(%#8 zzlggNFpu?G>?e9O$|1%JYx&kF6w`T}Qhfm)5aM0ws+;(&JKdD)F|K0tbbtuJJI757__7g3?%R&5hxgS~Rm zwe4x%f8a<8nXJAlblw5vd}Vx3-^X|=62b=MZnCCWs=tsk;LJL3kNTK;gylIQjGs7m^;y~|{2rn0{XkrYCehPsVyo^~(C(%x<%g2tc9Sp$Cb*a3#-Uk)K81%AkIqY=9-7 z#>%*k!T(%J&(qb>w<)U#jF3l~&^!xyLX$A;!(z&@RIA|nS=R9FCJe#$@ z)&zX+lSM`JnCP7PY$}s91Ah3{>OZJC0@e1u~_48bU zG&u0>GDPS%1O&}4MANQyYah;JQ*=9C%8&Ex@V77a6rGln?z`=#D)LP+z2L)Ue|uZP znT!zsQ{wpq(|Cr58MFr&=w2d5L+&R8I?umtX^C99Tt$GO{~`-TG=Fx|w$!|J)QE-k zYAl&UgWP_CQ#8Js|7W?bw;m7FHP?MGvu3Nbl=oWSdpwYapRJP(r-1XEl%IWrXXqV* zeRI%$H_yrqX1MiCFho4<=%Qype`T(c)*)=$bGQ!olB(git10->WmVAbkokfcWGW{f zRLro(v;7l1^yU@=gy6&Zj(1ncepGe9|Lm(EAsc53d3~F~M0XDrKLb_v*a!W?g^1wX zdnmY3uY_Xcv}x*o?QUygo7kxE{REpW=b~Df?Lu3JLKd%}QCM)c+pheQe_6&mZ@s#K zW&WMGJ;?9_cZ&5$pM2UD@|H9f9Q4+hr7q?jL@=-Lxx84keI{h3C4Ukp%eI#s;i82aaIQ4{nnm*gJ+D~r|IjkLG-858|Oml ztBpu-YPhEVT?Qo2d6wLZe-sgO8^7!<@8G5UUpd}$e*P)Q4WVs3&|JvMZ$rFGYX#Fy zf_r4L9a#E5J3s4loyqWaF7v;gkb$ltB*LP*DhVqvZvI{Do<@v+t$@cJZ`w}Ad|zC1 z0p%DfEW}T*RPw#_=5IGReP8jP!C0KUa?Zzj;mTTGrwD&#SKn^@f8x_aj2w(F?W61< zk^5oiw<*q}X3L8$irta|*G(v0^ktAzIu9O3gW_*131ywM-+5l3dOh^Clij{S^Z*@A zl*j2~-9BC1c&s|8#o!P`W95;Dg&x^Ifv3nXf>AZx)Udx;N8FGeX->9DTl9FjJ^wDr zB#S7TYlnn*=vr<3f4cDKuY$&ybJLha)-g@l#d)9C5DhQt0&);c5Ug|w{Z~Oi^)EqC zl+2Qr-$`qxx)*BkR|(_!R~Pj3VN1bx5v0Vfr`1_bi;>0YIJus|p$Q`6!F96|;|Lu$ zvD0@os7!#0mXWUaKy<#)k zd?~<;sIg<#be)TlZxU+c)Pt#Z8E<$kGsMKb)i3xwX!UI{-hBb5%T@24HP+x=yQbs$ z8W+xnJiXNFrU!mkQuC?s3qrXl+%xoaxhUH5EDyQUw>C9!z}3$LHn${4E+%)20IJBJvv>(oTO{5>KL2)jY z7Rf3uPw;&Y@pXfUg@+{G({&P=+Cbcrr;ARM)!N5JI;TD)=k}sqPquX?5R+RDCNjKD zMt=7}m|L@=8!a;p5e9?tmv?o3KZ8hu6I^u^X<&VbH{__lD7HcBxopNKumDRHhzoq0emD>9PGCAMxv^YIM&VVl-f7_A>oOjZ53L%}ex$@K#l~j)+N3euR zG}RY_&GyTQiA>SI#b#c{v&fTvKBp}^6)rO-jJ{GqtyI5K;K6Jxb9ssE!ila@5yvW$ zn6GODqhRtI(U!$n-*$8C@Kb4VMzAF!vp$jVM)_( zf9%mLRm9e9V%-q-wmoh?b&;F`=)|c|DCbb>V$sY|2 zWMjO)WtAQ+w>b!OV?ALhbDbD~PV!ji3wXs~z#UcDL}z>tukyoaaH8t7SN0XDEGA#( z-G4c5LrMI1NKva0lJ+aFngko6+x~soe}syv^KGt$+lw12!BhtG)*GX_dBP$4GVJCF z8*%UGt8j|T0DG8x#=IG&!VPpXWX__m?kvX$5n$ZD*p0lE;X_UDP@)Z$xH-IFe_-)X zxJo$xtp3T9RTRcr*}ae9EgyFAZs!EsGZWM7%F2`QZfa#cO|5P_e|7-1jh;Om){e$L zZT*|>-K%)dk^oOrYhwts;o0idn0Q$xeu^a+G7T_B0Ei$gnYC4!UL4#Nb*4$Cb@{Wt!<4wK8rh3h>8I1rRwZB1z1scS zTNZjETiFJ?g&yzUB{T_1AoMnjMZFgvshh{;Hk65r;wpQdp^g)hYIkXa?lqJzZszB_F&SODsqwuVTcJi$2{E>^f4nR{L#`EE7a%+f zu26isOZMs0;%u4c%}t()er^R%Z_L|^pCevv`P7%Ip_X@3@Bfb_$k*H`?53MsE#&~ehwts#?L z=-goS0W^c#GOX?(fBNiGA}kwm3C{kP(CPfwo#-}+47fV@+;2ZJU_||OlZ14^F3@~5 zhLB+qDRuWwSFu=yzUQap=U?1zsX*jkj>(w~=6kIec!afLus9kL{W~?$8wGja458!W zLeKBcWlgs)Pm=(8nftVrk;Ij?Yk$c@zpYC<47xNf#+i*Df7sX>FbqsoE1k8nXQ$UF zpJ2Q(!~nAWAm#hWd=yTcer~Mua8ER_yB3@ub^Zz-n+w$gm&I#yTuCG-^3^Ggjnc}1 zWDPuxq$a8&?#p5RhwD|Q&=H?rdWY`W;l^dpC`Q*>Z&KN`nJ5&T?;RKh)ZMkCG+B0& zG0dja3w~?ee_!K^wqNitxibc$JpZEQyp4fugR4yIz&eCHCVu)EpHL!-m_{$lOG5Hu zdt~z!Co|cQY)Vxd_uU!`T4vm#$H8>Br{2ct!AjA4!Zn1e=U0hql=Q=>f^%IM;8cty)Rw` z;;l0#bLla*FdM8I7C=5YKO^*g+>)V2KHf?9ZFuheR-0a`8!hy)CwtVYa1&- zH8n!#e^UGQT*vV}y%vK&EgQtx-F=t`iM{NY=k>jVwAil!C4%el(;6mnOg0OVkgsi? zWz9BMYZ`i{G zf3I(i(`GCMj@CVWyru439ikX}LFYewr>&N#)42RAev&D`)1-mP`hG8E!D29vV#+`D zMwTPV14aiMvm5NU$%aHWvKmgF!mka&pnFZ1!oQ|w&&gJv2HxOGOx=5C_h0XZ2>;h+ zHO;y&a-UY?aT?0_e*b;NMJ(iST7y|Uf7PG~k|{A+A&&vjleTpxw=mrN1n#ZB`Pe;n zTD4X=m}KCa@9yeV^Mk#Gerc;Qi;QMj6Om2?`+8ee9gqYu9aM{Zx&#yOUI%a-thDxn z?Q;PC-CxO#0`H`us^%>z$eSM-e2eWd;mh;sG;FAz&xB{fFLn#h0 z!n3a|63b<_!pkD*BK&`2#HMy zl1loO#RW>OckykLTZ6XU4Hf2W^n+1sr(lZzcNzoQg7ZqN+GWwcn< zHW99;CQC6NCF?T;+o%<*YhZ|;#{8n8D0dNeG)uA3&6<4ez`7{Le+s3$Myf|i+tE`Z zh0Oe%2mR~5=3vMt1zjB|ZJT7&D*<)+Qa(A1amO*}@#4kCt;OJQ<;X!C?QWp^U`p-w zK7S&Ufy6gVvd!%8*BoOBCZjxQuJ#9lXqh5wy=XbUi-}H(ajjH?$9gPTJd?Bzx7&*+ zc|N@nQSh?yO$vPxe_r!x)8Ab^Mi?}Unhuv+HSC4b5fy${mET4Nyr_wV#iPlSJTv2= zZ~h|iA31TM!vJfLh+nBE4oFLr^}$*Lup^#s*)9zX**t0lO)o2B^W9w z&Co=#UbE{^Z7%VuIvET~8mgluUWu%W;eJ3Akn$cu6j)7T(t_%*7)MheM4=`(paO}+ zF_I@MjPg@!tRZ$}y(c+K>K#9~x|Yb*IoTK!H-4N5kRRYE#M9}BdyAc3SmBr7KV48h z)m)t;h~;(*fA*EInk|d${qk`x+o#Uw`mGt(OK&8Ff8Bh#+vDYOK;5|FRGWq7t`n|) zWO^Gh?B3~f7u^Wj(UOSI8y>C}6*6W9tds`M7{T4)OtPQ{;^0F= zTM|=I;-}0Q6p|iWvYc(xJSw2EqRq~yDyO;Ahpdf{e`2pPl*AZ$8jq*`*dmPbkj%uB z+aAxdfx-ZQJfJvfY(!XINIducL}}5ZvEDcZ__RvU|NMy8B_=ekKxV?aGUhIoBgQ*T zpv*mroJUM1;e(yT^iYB3?g*EeOh=n1YcC7erN=f;SDia*vf%TTiD9J%AJ;BVq8aQo zXrTcGe=)m~&D}lo1x1|=dOGD{P$&KS&n?1oq~Docen0ENFHv9?kom>i{>|AY7qIv8 zl1soZMZfF?gXo;&HW=BFh#~#0=S4aNnArzOx8kGgk>?xp(N}}5{W`etG1+dr9I4Z; zBrZ=y7IlF6fp#!2{il{#ziP(^&t}uJT%MJZf2tO{4yhkoUeBvvi8-U<>~e7xh&i*G z9em|KOks-O9?8sY*W}&v@c)G3@%l@&9$N0H5$fxVS(!YFF;6ouEgyAx9UtRmZRQvy zR2eQk2)oEm+L0GJ9e}8a7FkAY)dbzKM5FveFtXr*Je3id{`-e(D0Vw0@R+w-0LW%! zf4e2nlyy$^3g4~dZrVCL2}n*~*;?`ab86{htvkgj=dAy$ej#1<_=v=If{Ayu#U_=% z-=hc+l*ne1!b2hiBb)!j0P2U2$MP~nYTd9u-U zQ+`1@25L5kuIyaYrzMTWwTw=oa(jZ?e@gBs_Rm(gKzc>}O~)av;^+DcLpv(VZn760 zp5XoKPJim@z@@vi7nO>)?Na5unBqeImI>?P1x|W4kF#Q9p`1ry8ls91Bc%Wyrm=R7 zleh-EyHVB&zC8Kjbx!cpk9r~FIUx^ zDMd-uIFLpPGO=|UUo~X)8b-7xf0J(ZIMckKFQNbb${-!yC ztF25m$Sod^^X6${rL2~~@RE~uJ|Oz%YkN_^d)t|HX)VA5o4I;6_LlxpV{Hr$!dm3W zV?|D)(D^K`R*lV9pO!UeT9w`?6D}hDL&E)QfACNjVflhK z5f^(MtNGPp!BAc9m0p|+>9OpM(%l*$M+uGw^SgzTSeiPMPMd|@xB-s+%{sm3TzuYO z`haKo10n{u>&d)^*+`~!e>u|>$hoLLX=T2`FR!PIk@JAx_LylK9(vBZbQ5JDd;b`t z8}GfLv)oM7wq)@LcQeTo#HuxsVz%RvS=`=<$ zf!W(zb_8V5igl)TUAOSv?dHXY)dC{?y8T$YyOPpM*6E*#Z?)BDfAbH9lUaW556w(g0~@w$=@oNYYH zY|Anc_>N9mJ#z!1+mbK>3Cmg%+shtCEZ#Q=2l^!ja! zbQ8$atD??t&ceR*BO#yZG_Bh9TS3R~lnD#g*qJJw6|Y=s_gQfU_@y6N|Hzar zU&~KK=$yQ+TVFuI*g-6( zE6I6&Bw+zL&MKcqE@j@_Lh`)hG2kj1TT^-W!?m08p?_QdDxauzz&J|q@I$P2WkL&r z^bH@M%V{*h;LJ{M+DH_`3MWC$;jw*2up?#!7lDQUF(xR&nSNafy zeeGeTjbAY*e-b2af%%SWHIKenedG0w-yQbM-8X4GiGgMJw^(`yPF^g^t!vYE?%?)s z>Jt+XcDGXfi`O^DcNAs&KDU|T^f`#u$2jiUbn^NFiu+e8Z&vdb!TX*ZQyX1FfL4bn zM}c=waKalqhH;y$-Qbgd)7Ct}-=PpL_=seEOZLT&e<(*j*n{M^tjxtZSQ^s0_GUy3 zWNyhZm;$8H(vvyZ36z;8uk-87ffM!xWC(b7?-+Uew_m4SoG*EU1*M}D>0eulQ$vSy z6Qb|^9Slh=ephRc+$uD^u18QR8d?}LZD-$Wi_Tc z?e4i4(XI0jNqphE4+8VYwDX@CS?3wXiFVgI-d~-c^sGd|P|J3|=9YW1KTc1py90pS z$&OkO$ZQ+<@D9wQxXf|=^Js7Me<)wHm8Q(_@o~vuc&6l&+KPv7_9GSUkCR)=M%6NW zgCpuGmrk=lEcf-?f6#OYRc-fn)`4(ZQQhYl>X|ey_X%Ii^t4Of;*r?Kps5NS^4$>;u_Uj zYWHpbMbQBUllIB{Wk&+t3e8W+UU#ws{Uj`JVuy2oEdN|JXU~oU;6qv+9hbml-d6&~ z8R7BUdU_vyA?1U5Wp{R3e}D6qwvO3Z$o&GHYLiZ1Bnb&~XRB#AS^!rhcEEx6Q2sl_ zRjkx+3wBoRg=i4S_Le8Q>pfw`fe;hC!ou`V^;f`xoCwHD2JUG;SzMu9QU`S1U}vGx znqgt9qP2PUrykhvbmdj4wH1Vwm*4)udvfbWf5zm9*K1(iGgMH;e+$fCOe$R-5T`L# zx<-zz4_*H19ic;W2zaFOWFp(!LAqm0t?-%~7s(YeIkhKIwzUx4*)?~t&FTK}0IBb9 z_wjXlJk1q>qHHBR_#TdYZymBIS13~pgQnftzvc0F*3a;+T%p->Y{{azJn@FnO~EWK z8w~gQR~d;5aD{-!e;Y}wDH}?_Vd&#w#mPZh+L#JgVoMjg^`3Q0M z02;S|ZOgYtA%7;F8ew!Iy3cqwOD!Ntlv|(e+A3l+0*CW+pF4Z5IS(t}88pcK-ikq` zZrg3!XCbbA(8NMiQj0ssghSDZkgFjRh$&=&$v52!QhVD{f6Ab#kzn9oE5Fa>Gqtz< zDGBpZuOu@rr$=&Vq5hzZB96m)L6a!|b-*)tRLZ}a)zAP99_-4QjRBFj#%~eRV*}k8 zkN2k$MoPiy;l@{1q}qYf^emPbsv>Wv+%CJDuPiG~lfekL{iQ@Y2KZlFq8{6nDDt35 zm|!@eSf-Fme+`8na$6Z>1#->Hm)+rhUG+fS?=uE+R`l9=RqS?4owr{rB2oUru!Gy4 zdKiSYM>-50^rEz*=Ez6Y9|v=k%V3D36U9PNuzrV7RMfO=Dl^%(dkfgB@mEVc+%d&& zxche%6aDH-!%?h~;W&0}4e5cQ*nr>MfisHweQOcWfAs2vk1}DT(it4CWP2+MQD;b# zWifVHpOU`WKX@ESE3k@v2-BmFnj^Y{|4Q3^)L3Cq`s(@h$*e62r}a!-s2bd^Hs1TY zy}~ThBbo!#&D}HgCg~)|qgw4ZJlUjnjR)q07wjKSmD+igJA8O(bqeZU@%l({+TDK? z)T|jUe~~Bv{GK%^u)-j8a58v5X(h}soCH9U+e)$?=`Hgf1eq>tJYqnuCry_Of~=GH z8FV-k?&K{#n2%a-X5aAXSeNugaK7fgW0`4?r}rF1#DeNjM_&0-HNQ7zA{{8dBPwq3 z*_&(}7`8;yD{q=Tt7`rZz1c}imtOc9?zV>UxWSQ1bWLJv`)`^gu`rcnpy+Ks=Q{B`(rY_F{S30 znUQbd1(t|<^&fmS`AE9_-8!X3P2Ce2jMWgvE1oU#&ws1{bR-v_zMy$rye&61)hL`q ze^-PN)dS#kPCIS8!^F6%;N)sn!(hU3Bi*N^Pup(v9;X}FeC2c;ZfJhxi1kal@#7dR z=8MPWnxdCRMV!8JONJjUh=-Z_!92f~yKh>PFo_%{O?8AOKB-qc%(bL_OD|C?HGwX4 zsD`fBfbKEvR@%hB{dl?ILbI@b)}JTYf4H-h7HTmE^~sJ>@%5dCAj^|*#iJZd#V$2h zBm{?G4eUqFXfLBIbgg|}lno0$&W16)@LBO?3Z|uYzZKbI>mTGvoDi=|`=1g&=#;M_ zC!Ow&8k4j4(VPQsy@?IURjw`)-O_mMM=t}$bk3}wKL9$gG)f<$520WU>(`j}e}WVH zjoJYJW<#nbbUSB_OEcfbpPAAMXJ^JgzB-HY4Q*DnIZ-xKj4|%Rh zBn(34N&q0`I3K{O_q&5O1dD8DUVT+D*#CinL!aK2BKFq3z#piMHhFBfSwJc~2!&`= zjrzmk8O=h*>D9{`Mnmi#Z(J%zq|PKSsKWldOP>6(bzS!e@3?Lew|dS zvJ;>J$qSHo;3PzmIpXc1C)v~=tl{q-!{6uXU-zU)kH}$q@ZqzRXwkwYk9O-Lqx%k- zx+b&|@DMBiPLC>!`heDhE&Lv;_neI(EAjo=3n=e z#vx#m?B%|FZ!mPbHUwErg>N+vIgQizHG@uL<7j!LHL%~L+DLqp^um}5nl@W-GQx41 zH%71!i|61Z1lryCULa52P~x|2Lif6bzDMHR$M|U2|M4q2_a2pge}+ATvj?-V&pg9n zEbHO!Fd>{MizxOSss@8<@XK}|z$-g|pl-8vrp-mXc&Si?3I=DAsQv!c`^ zbGw{?+*QYCe0SF;JeF`=Ra@bZvxi1cUr<+32R#|O$uoA z!tIcR?7*w9f8V?YVM*^R7K; z*0;2)5LeV7F3ldWnCjC2N7H_RdeQrEuuWxARPeeju=0`s+FR@oRFY4-Qq*}(^XPfy zQJpI;)`NJp!2@HEN$&2wMj+5*_?|-))19AouWV~N>ZQ9G)6d5JLbV9^ukPI${tz!1 zT5jkIE>$Jwf6UtLABNYHbWT#LXLAnvq(By;ukD)4WE2a$dEE)}g&`9aulQWNX3&EX zQ|%Rj6`v(ADPB1FW|iGd`8a~Q(0b+}_4f7MQvF!Ac~c{%<XX)f)a3SAJhx z|B@z~A+jJgxM{P|uT{d35$lf6zSisSX`_Vme~;X1jdvIgmtE5kZUH8)&4ezd zxWgXy6GnShAM>?M2kmY(L2ZsdEUw4wVy+j5Bwjd7qFar*?b)@@%tJ8TN~iOUjJ;_} z>Tqp5Auf+Q!MydQvT1ILI$h=vr(?sTt53FI`(D7?-mZ>HqW=Ihm32?1=1Pu+dcj#t zKVPxAe;sfv7e3cC1F}p`P*I@ z)NRYfu>uskg>3hFqPySQF;&7V+@|NLrC)ZNe-st?OKLZ>YbH%m!cA6QMo~(efslOb2bk6h5Qp#FWS!XHL({lG~DYdz~^7bb3 zv8#7-Jke*;B*17#<+hwQ$5gXy8eF~yaGm;72zV%!d7p{(tidaIp;V5#=l(!Ri6+T% zrSg2Z1nDv9wq3X3>{p)v#&S9y>yp6xe^4jtyZjS*FUTGfiO2I9URN7yXh7XG-q|!I zf^X8kAfsNC`Y+(12m5~rOgO~0e<{p1?t;y--~bh;%h4|1zeC*F;diFZM}QKoH68OF ziYt_0@tZ3>f*TL6mGO|EF0JiIuN;i}WXujHEl69&_(V6%?mST5AGZ+;E^EC7e-FK8 zFbDj0S+4SiFl}piOXfG-^hW!^=3k)$009rt$zf$A|GeO%zM})Rhb7<4H0Dif9s63t z%)8Y*BAPf-he_zg`bm#(dd|0_aDRCc&dFIC-}3#5pHUy!)V7!OwFPnvgS{q45yvl+ zdpMxt&2FNNpS$ityV11dAbtt-YVmesLdNZk-*hynFTm@4vc7j|4l`~$_No+bZZ)fQ zlsk%|s8{n2rU5+AuZQu)z!C8R+pB^yo$_1tA|;W7n_j?R*jSLE1iS!+1%Faa1=D3L z#c<`#vOfd~13#K(7}Vx=ms^PvsIB~J2?AM9e+O}$SjpOKDV(~{cQUX(+yLSXUZbg$ zp&LFBao5R+umJ@OW$|<|@jutj;_j8}Q>Jt8*FA$W4Oj1U40#E5Re$jheXxj=P@g5e5=3bgNUBU|!8X5hFW3<+B_2s;p_%ZO z#B5mdI*^}i-N6uPZHX{x-MJNzSAJND_0FW?@w;PLSL>9XEmXtU8O~tBd#6y}B3IX# z)3|`?IzQPcJw~qA^FX%eyqMmMQPvrzS3IJxfqe1v`@S0eKwQDl`G4ovwBB0n^2Zl> z6>new-hw^gssH2hEP4shn#bQ4cdpO7w1WhmD99>&r;r-E>-$|$QQw?y*P>U_7kfvX zZ*jKJbZ7)Zyp8f^;qoF>T^{pF(6p&|`>t<1W~S!$6{yhN-Fgn1rQz>9AOJb`Jzw{Y ztx)vugsC5$^Q%iOL4TpDiJfQAoJDJdJFT&|SLT;>FC1zZ1n+F(${sw*4_Oj+Sd;s+ zk!$MFrx~`h8q7z`tzhD~h0)g_`?U~sRJwG*;-V7u)vj+{{paXMDm3mry>?!Zh})Zq zO=qYU&E$^aBN@2+ZF3gUApn!9rKmP)EF3ZvbDY#cL~^xlv46w&!keNl%!=qDm>ILI z#lRijH>t!>8uOXo1;92}h&hAZFx~S{|>R)ee1u z@iy%-9&QuPm~MD45l#P$k|TzTK%LeHUWzpJ!kb>7u151Q{ezE>;NGw>brH0y{No!7 zoWtaY+Y6=}$A2A0p3gFeM4ZbI%eP^l30?-3zo7-zhyTiwTRk3M>zTQNsp9Coi)_Hl zl@_oLeR5tG(vjckdN`BJAmjCiQUiWV2z;}=hB--qoC46*GTGs=3&F|s>M)kce#OM~ zE+O4lIvg?`j@0UQl-Ta=pCBv8h3$E<%_-qIx_?$cq<`kOboSF^AXdDhW51+UTasF0 z=k}zhUGcUJyzIVMzQcGFRvCSx!;{7&4z`yNHS@kYPLLyJO+{xGSH5&ChGP&I<8d;n zZF%8Bp(2BAtzqDObx#)}57D}PWsv94Wuhii1>u_dN?aavHO=){%Xqpulr*L~Gni%+ zCF$+o#(!O?-@#ZA-|E|D?wqzXOX8j8Rc!|u1N`pry(YsHUa#V}@K&g*(0$%MO*vv<&8=5wuo_`|#KaRT?tI_{Xxw*=es-3iS*xxfG{^QFN zcO@bf6l*td(DkpM--8ME`6ffKr)g&w<^Om$tRV{eyOI*Ou#K3meE0mze1`@$d*uwC znR`lx4E#d**u}Dboy>qW8u;^3Bgj7~wV<^adWWS+z{u%)Rb5=-A( z4$4xsvot*fwg9&8;1!B}k-Kd`e=P`FZR@ZCQs8pI2ao&`?7?D{!m=u52vKHSUlhAk z9gR?01!)UENLKLCCyqlg$+KB;On+>M1@7<4H-=t70PyPr>gx5szt08M(ViX$iJ#jf zkr0ZVc{STJ&S2Nl0@3qtZ zzgs&P#K$~TK6h)#NNGeZIch@Te)`i^JJ407JnKnX*2lhlagO{%%P+AKL2|mwP6d1 zP5`VkE}yub(#6=sPZ0?Ce26g)4naTGN{tRl?;G*|Y+{{%o0#E$HL+#4irVf9^k(q8z0SR9R_5XfI6VDOa}TFWYNXDCLJhr8Wj;1 zTyFoF%s$O`_g|V-Stm9?#VQM~A?iQRRKY1gI#HnEmks0m4=oI1VyLY@JuQe?GDzm@Tu} zhyVD2XMt0Ml&kRz<^e}<{(NLgFk6aPCI|kxTNX3G7P)@x*?*$spN}jXX3L;SW`sX? zOJ$va{}V9(Ct&_h!2EwDV1%d8^WvJCxOsfHBhq7BsJ0VfbYVw|QjhmS!(C!BZ816d zW$nZT5{^LQ?oG}H-b|u$XE2$^Zv2r`At4nl#RoN&kp$x7aA(o+;5x_zqM(+F{lT-V z_b;SqXcRJe0a&3wMONW)e#f%}^9=4*wn%jD!FyCWeSiOxSx`$-9>?#c9?M;ZP{$f~ z`r>|<2uyYuU9M>vLo2IZ0C1`Aa->u6J|M3I!_PP5zG)2Fz z#QMU;#ukmMsuj0`3Hn?z)7O874b;Nm0%1BiQA%yZ(GWU?802&;a(-%i`3%F1)EB*# zNvYHHl7B9#sNCYeetrBSd&mbiajUfNxr{Hovve+;r$@wbKs1_6qf=m+(CZyI2mtO! z%?t`jMoA|QF~yy`gJ==c(?c*2{-kfQtj+-y#{!!C-9`JWmvD&dNODZBu`Hr;s?Hu{ z4kUbjHTP#bSx(bx8iO11q<9Wt`;sVpe-J4SU4Izq+^*VE+MF^yczwevo&esB-~Rx%2n#~ffmg3fm=*J z)_=2~dq&2_V~FX-?=pGtT-0-05p`xV?LhlvM~3Z3Ge0$QUH_$p{;8pYb&Q{aC;K`& z5%#iBo11g%o9=s{+v)1-9Tz><=>*vP|68No#9(#1H@Vpu6B6%{kK(eG>l{c5Vfm9G z<;udK(ItcFiZD-W^dLR@WDN~h~zu?V_u}1C?p8b(oYQ;=l*t4|sRSpx%p56;K=5GXAgn#I{66w|= zXiQg60FN<}sdaA0$`8kU`PNC1qaNC?Jgm>3Yqf2Pn$PM}yrOS*(mRMK1V35b+~T0& z#AE_RwmdY+z>OH@J((BgHpi_eLth%ren)1siju3!$wi>?Ovhu{raq-A)aDOt!5F`= zUcFN(9CIY}cw8=4Zu;zJo`3!a#?MEbldiMympw6*g0#K7yPNIS`Hgau!zKR%DmPEzC<(XYdERhJtADWHeC;erVK}cr zRuKiCrNXH!^xJpj^w`N7{TT5AxtH&H!w6*tmo=}1V9Uaf|Bq9thk0*)qR)8)pYsgA zXNQh~`P(;=&tplZ>3KG!$zHO#)m-|&eZ9)=dj7t-@IFIY??=54J94@gg!sL7?||I| zVbz;eldR@1i!e&31%IBrron4g@f@u^SimTbw30ngG86TNv+Zvd)Hs&4ffDR0)HE?+^wfnbFEm{KbDTh zJEZ@s8MC2GV7X(OZ?4B{v0ReLoa{NZ;^&!kq{+LjE%aFZIF?Vm&gjMc#woYD@xA~hhcu}Rh8 ze1@&q`ps&*YJW9`i|0{Vt<_PVpke-%ZK`a41H;EKn`@hJi)k8$1JlNw;5(TFl;e8I z;1zOS7tO1~Wu{qO=X_mfu$nnp*ktx%U@(d)s6#aNjU?wl!Uf4hiHfvsf7B!W-4#W& z(8Kl60!_IXf#u{_JA=1?|AVI(I>9i_)1Gs0w@yQN1%EiLJG zAtvvw*l?yqc%*Zgz6j#D5AK`4K{d{^cA+e@X_d{}bk0y7@7PM>1Mqq~;5bg`@md9) zi1EEvjc(>9wCW^#w!1zgANDV^GR(>)s*VHcJhcKkZCKnt$hfqOP^eu#L9XH_V6k5_ zbES~UPk$?PJonZsjee&-@ki|&IONfW@D{)WuaREkwk4Pq5pdWzk0O?X2o}IL-Djf<2#^c zdn@&E_>9S*%tPn9&%ym$izohd1$qW$Rl^xPA_WRrZ}8+Ol2Ne9-4aLAZ6WSY0N@nh@Tdautm63GPkIR#JGC8%lC z{C`SKV=J1qarwg-Klsb$XadkYQ{ZFgK;kOlKS9-Ky>K03`fVG()H{XL?Nc9UFUgJ1 z1|jkNt<(+ZvBR~X|5sj;lY3$eO_s$mqDy*;9|B24D)*V~f%a9D(D7wTDO}oU#)8yO z3D=%)HZFIfD=*(tRGhXSmmR$K?ms&BNq@ z%-OC5?J3{i<|=QtOkU+hH}1#92PF7+S}shn#DWlmZvp4uKIk+kD>S2L;eUu$%sgJ2 z*?+;L1Q(l!oSw*Fz|iw8DxnfEzwazZqG~;LDq=ZX@O5*P?AojK4vPf+uG8d&i+>)- zuW?yGwxa#Ys#qNiw>Q_J))_GKZwNiEb-MAP&)b%}w*S1?aFW0FhC#eX)2&pmw4{wC z53p;@RIgnSVNQf7TMh+-7~yFAar+R^r#wzsZX9Rxl}1LgE!+#lEzlAi4JPB(sBd{0 zsA|BU#r(czh;?WVw{oz&G2W@*20Mbat03piiy*y+3s?TWJP|&3i?`)E>}c$ZshnQ- zs-DKzP1zT&-vrP0c^8D{`6uFF;xpQ1_@AU-AU^*0nE>D$?AbVCwwo_~A273Uro zeQ-%ywO1;ZP!n?=w?mK4n=7$$qsy5*uHbL!5)rSMg7rxanaEGoD~TFyFPaFF&#}9% zQLK;Y+(Ue7>_l_H7M**y2VqhZ2Se1UL>58F`4!BwCvR*Y;AmXa`0R5Oqx=2!)aXDo zupAaIm%@589LB!tvi=KgZ-1(}^@IL-&KMLb=>mN-0)CXiUOEsOB7TH`{PTVLcT8_W zN;512k!ws_KYqg|OMZV(D2+se*)?rFzI6E1hIonEoMJ?U&^hu+QczMz&1jFWPjr59x2>3ini9g zaXX3=AM>Rpa9;+S(~33(`_XI+%bU;io|#tV)+}oaX0|Ni*AjI`lesr><*Z*G!rQO< zEQp0@U;FL}f&A~Q`@s4A7a8|7vdXh60Dbwr+1AoAFBtQ4U=I36=v1QAMo@^svE1E+ zer1#2`e`H>Zar(CP=7kz*ur#?f@_j@_6B<(VD8t=7dQk^^n~#Wx>Mz^I0`#eMPw|G z3G;9MNHkd~6}Xt(3C!)-1dML2)>fxoo9Yhe9cCaNNIXF0UD~tc)>sv4IGUs5iUDI= zlOFRfjI^f4ct$;5;#)8OVwKiZ?;4J>>3kAStEbh^LbCkT!44HwVy4N?xXF^I;m%#7GKs@7y$#dtCf08 z?zUUfo`za$*<#%deU|C=@Jpt*vea#-NZN`6q!n@dYIHI=A8$-sc?!lj{a#xOp2^QD zq_KarYXwV~FC(VELP!3`vdJq9G3QQc`ZsFjPJnn?=YO=P+^U}B8Yjo)c6>Z#HB)5s zRcon?EXzw|MW^Xkii=(~mAY%Ta(>uQC?3rS_<+LnJU&HmIy{jngC0i2+>U=5rh0+X zcs6()jEi67E+-H$>MC{I9i!a1^hjtyCujtR+SsgfoBMoA*%*{M_j1D-UtnTcHri;_ z>c8%fqJQAhtDhz~BC+f77a%Lwte2?Uk6q>!nwB_bUtXI(a&smV2;QrHKe=ttHPB&D zp*t({6B&C@o+}~V_YUiKNz_a#fa>@iq#5@SDV=>u6|lBZ&?);dDNtM9YPjfi zsFy>{tbc_)8uwR1npOM-8oKRU_exI7VVve#o9;M$CoH4N;(gV8b-anq{03RZ4$T#x zpML<)kLP;+x=WVDrm>NHZpUt%jZOnLFlM7|u3IW+Sx?(a@HTYr&bFZ^-ir3yIb>N*Yd3YO|sIaIrNm87xB z(^J`MHp*4B?8{hdHwDBIVt8iDbl1BINCH9RZyG_{i@>;w!+FMblStYnqeedYXI9&F zj#otlymT5JEocQ9RRw#lkFP7+KYxJtXIv5|;6@nuS89G*%zkkr<#Y*{Dkd<%rPJ)( zOES39UMnzw95tSpeO;4?B1zACmg2mFPeQqhPm&m*Vv$vp%=dX5HRb-w#@t{>j)F$} z>+e!G7=0`#nMBvB^Z2UT0%&8mouphXk&I0F&?E8gOX#~OQY$sj8lo#~Jbz5GC5@6~ zu{{4f2nf$<7V5eQ(~8w*TfbxU1hsA|YUgI`8)XH6rkTrj%x~}OwBuWKGX+kjidB>w zJ~Ei=rGC?ETeY>m;*4u3U#$AtUc_#vs-b`QP{(#mK7NevI&c4oVbPnsxwn(`AyM|x z8<%o*bH?c9KXR>n;X35>RDZ!4DgEO(oEpb84%-z4!CQ|EHcOb2Q<%;I7Aop2x9ycT zoVzC_G!{m zjz`^!t>-{OMwrJiVfxJ<%@;087e*@1{S<5@k%JjE^%Xacfq?W_*G?9rd$;PwJGlhE z>fnq!{)v{9%U#0j1t^h$-y7F`!Tk@5d5*q&pvpoH^7pIpH#M^PcI^eRGd=QN&k)no zfg}zA0V=|YyQy=HkADZOzB4}Ng+y&^(iQZHO6p@NG2IBFk`YOtHeL*lwF?$Ssh0X= z)v7%P0B1=xzQ1!qsP7wD9NK0u(5Yo_x&-$nC}=gk?&%wxLpyu@R`2t0TBQuqMG!%m z<@>!#p(yU!q4-F}h}%l@n>)6*_L;h)P7eq`v(dODl(scFH-C*ms@2UZ=6@`X>}Wov zV_-pKOzTO@o9r3uV9DR7T>X3(W+TazrYB7!#FfhBfj)Q3Zr1#w-=4Exf**4A=EWE4 z_}GsKN9?TCnPOq$V(w?CU7=jY+jcK(hk{nfvXEx&Zu_2ci3ytd{AUDOG6yRqKttP= z)yJ>zAEDQ`X@BL)wQk1d=j{Z?u>)}{b$vZZ>BkjHJ%z1J_HwHSFiqt=vn9lAbxG~H zHJvkRQl6JjP+uot;T676vqf*MNiXf&e6taKaI4yS&6dMvtDzes9+sB-e&IgYO#5xg z5C+rDp*caE;mezCYs(U+xcZB2A$Wxbu~i*-g+RSrtA7%IE*m^aZ0V`5=^7|`dNa`T zAweF$b%UEM={l=w5Hg{+3%^Pcyrhrz6VmtU54gC7j;s1=&00SS>a~xj>h;XqUM4r1basA$C*T#3ys}Ylxh&5WA=6WDt-zS6JK#O)R!Gxlnz{RG z1t5D^ZsD|JVsq#WKX>?f&A4zMW!+nIqnsxt;(tAn-3T)n3IDEX@uO0C_TntMW@BQ9 zqTeG_(V7^1dk@4eUsHl{E2NT5d3U~p_N_$Ht>tF4UZr&6ionW_6}tL;>MzAE0GG!) zUiCGjMrmDL9}Y&$YRlv!n^gZ!=~ndnji_s5AXu%!r80sfrJ*I?nSut6rcq)YroKbv zbARmy4|z@o!kGZuhmHPKzf)o&V1gc7b~dxR<^lKvx6)MUF;jUcA=(usy(Wz)(Y2Hi zQBR{NK&3n+b>ND^O&c;zGZ&RKhS8F##CA_F^HrO3yv>|f;I(3%&s3^-Cv7wi@E-|~ zE~=PyC#@>Zd)>}GkpE-C9sM*167LoA+kcL&ZDr+BGd0V(Y=zt~VYO1Ocw3Mk#bw8u z89%5#uFe|Rt^J`KF}*J$;{(*db9o+I949#-uyx|nhk&VC39h<7@ds(wGPYm7n72@Y(Opx zm>x?%+r+Zaa6HO9X7s;LT8MD*r6q7EQ7#R=ue&q1<~rXRc{4;PALWwdljxo8_2$NO ztWppd%Q{&wZ9Lg*Y<<9pm>x^jUw`vY#i-T%&aW>WlmRf_%vhea`vJzTiDOkT%%79Y z^_Lu)FzteRQw0Pj7hxPsI%#LUa;PvN066Cu_ts7LW2fcSFq)HO$`?B*Jef{Uh)ybb zyj+rffu75R_&K!UxehZu)Gv>b;`lMGaE^!v^b;o-gEN~B}t z0(k_RjQSLB!eFg5YH-K<%6~o4o1-CQsv_kyZ(9wi5C-Pq-zw@%S$sYG8mZ)2Q8 zcPC56GLvagV>UfxY~M68Z^8PY*(A0ER~TQiI(5^$Qe$#MXTC_E zG)(B5D{o~zTbx#Qzl(QZN8)g6)WbTE%wnvpVa%Y|xmPyVgn#!>p9wrD9bTiP-`}t^ zUo5)|{Z3@+NR0VO6NYSkXcl7S(w0zsUw7X#J@WyR+qV@AtnJW0q$oq124~iHJwBxI zSZDHs$}8C{e^mt*$hLo>q8g1rMY*Tv7J$fk@6EmL(8v`o<@UQbp!EbUJDqPzTVIX9 z^a;|%MlC$dt$$lic13dIvaC!d=X6oIiht*XJ>k4k16HXq+Ug&@bs-H0Wiiax=&i}gVj#dg zJqrnD1K(V;?PltEitQ285?aryaL`z>g!;AiY=R+=26?c>5U&5j*o#n7oa5H zJ8bPZW1~sg#Tof>!$&9?$00#;V-NYdh ziE%0bxPM{*8Tf{o7UagW4N6!jZ(Ok`fyh;lMt;je?%*BfQNioOOhBPicj$wnxSc$R zcicUos90=bHD6B3`j{0?2dEUszBQTf2Kru23*_cx5tj4v#HDh1i``!Ot&m_7IZWO) zi;7ZI>c79a*m8vtknx=@5mv`ZcJuQ^6$#m!)_)=Gji|=M_w*X&j@y>MWqQ;DQ<3LZ zUh#C5EahKd5DqiBTJ>vuJICfO*A+B)D)Rb2LZhaNz*UF0d*EE!AlXeHDwK|{6rSpsGlf~i?PI|)N zIe*E=^FlMc{eF;?)wYk{TJuq-#X-B1CSQ0_u{Z#j*$&WN-k#@60Ed03FV*+9QeK$J58`6j{rfHe)YQSF3GeJ-Ux@EO z^y}d1jOFm`mr77tAAI5HcnzD7w~4=KC%t zPEKEb%x9SLyq>^kEB3;rKM{VU$_{L+J68qj_s3ro&i0KHHmYE5pmbtncdh=BEf=P5Ww}WdJMXh5xarZl0u7OU!<);spS5)ff7b7h81K&TzrCVy4 zH0!D}WNduthY5Z5KN0#(z=Bwmq%Pum$&W9nScA2dHlfUasewf z&6b^&R0S?q-i_*am|MO2W6}k3Xzl8@3-|eTmXNYn+z!8+E7gl8Wz-hBGeX|VYW$9~ z&2l|1MSu+{u|$pTR!+7Y7iQYjeogo!9m=xuNvufnc_|olaBWJ*4UfJ2k(#DfJQMaGTy_besZaS zfH?J%60OGKbZ%S_QIQ>+abJW}W3jecm9XIa`4-oCZ_=al;N+gq@(IDonA1mTUfah& z1m*HkW2ehq7P9)(#B6%sJ9*`rU*LjpK)A`sg5uV}qp$T?gMVEs|FXI^r~j`MiOzwX z&wn61I=>P+Ov(u74+&EGLtTHx%nZs99MjVDh7lxY`LDb=nDsx#OP)ZYk$=eu0cor> zM}2dq6I9Z8;PIMDDVPmEY1-gyu2w}m@H&&E@TD^F0}GjVV7d6OS$dv_nvK>ubdi{g z0G?R>K!&>p9)ISmo5q7mdOBGmz=jadrbl)_oG;2ztbD0188dM}{Q>9iwI0BHZP}t0 z+dC;O=Os4`7=J)kwho$ zB{Ah+)w-Pu9l$5vWMv0Qz5MBJvfr571E!l*4^AV~8Z_E2;|D)Yx^>uI$YXQ8ev6>c zawNVzP-PAA6I?6(e3zGh)EDI=f6ShLc)u939eSEl!F2N_|31kviij>*kuaGu8M)+8 z;t#)PJ%1TQ8I(Vyt~KpbS6VNXik!y2x(RX-AmbIVuOP}g%_4Bnr6*&6er^~8uS*4S zz-gGh!}4)Wj(8=&vU*qv(0rpMck(fY^D#i>c!|e!EYqN8rCp_;W^=x9@#X`Obf@$Q zj@6^Dx`hOL75mWKzdZCTuP{ z22#Pr%`#Sv3TxB(bsnditt1|xvN0^vS-Ad1KirP7`TSyYdirl%0ED}r896L}y^G?f zR;;(T1QU>05JM6CY2CiJvAO3cic>>sQrHaW5-5s3yLy`|ViPz_e%Yd-2h|hFyDYA_ zcz-rNiSJvO`1Yk^nr9Paolel-#ia*4o)rqvP8_+u%US$+pi)pdo`yG4OJh-ASeSWW ztL9Quh}6R1KTF1Xoj`=w{vF2wJ+GJj9}^o9`~-=EcC$~1#pm}*Y0QqP)AcC=_UqS% z%4ZMWho27IBAdT0x0?$AFUA3az~wive1Cy=#Cue)ai>X;vUG8{Hj)O2bt7xJc4;&l za|W&GtT7T_^*t*M=Y_aDewHuVZ~`fYfW3Sg$)WJb-gUe&>a<~XHz?`b!Acj{@$$|GZ^auv z)AKy8u@-M(bMb+fPgTmOzp9j3t;RccrIQnZ2q@)~_WqA}2HKTUIq(JipbV#^VvEfR zG5r02#>c827s*D|%Yl;+3il_i^nbQhT63mcPc2nwK?&>eA;hQY+}NBTVK_YeTlgWT z@l~tB4G*Szy+z`9Z(4hV@7+Tzf#B9wrp=ATICM%z4QMu8-_SmGSzoMJmoGz@ z@aSd1Hm@J7AJZoq7W6fItJS`7$Po{DDCI(X!fun-)q;%OA#eGa)qnVJyIDOAF{{}U ztEhJ^0xr8%YN1vcJdw~ltlu$!$C%Q}4(0QmA^IqR1>q9);+KvdQmt68c&u_0D5{^oKm3JQ|f8Qh>bpH^=$N08`cLiCnkI6YNIv0N~IDB-^u9(-OnAq zz?@0$2?mnG|76QnDu1CAJzOFCk6)v&qx+Da6uZciVrM4$lbwI;Puut8v&h}B0>2}| zo;9O_;de;$pX|ZQvxI|Ju)Mvjaz+)ShV{aZ_eA1`#s6eCoczL^L5-kMHiVU&TZ3ur>QLVX`Zh&z*4~ zjmDiS++}a_ZGS4E1w35g)Bf#0d4UXUGU5ZFMqIe4uN?#t)3ILhQ=c&CevpiOg?l!@ z&>OWmnxJB(TKPf4WMEwi$HD3SA8p}Xy(getb$Q6ti9GDa{eBeyG>#Ia8`AKhF5FPi zqG5RF8?ni%mNuHs^9BW*JlxeI4C$vMEs4Yk%PF3^!aM#tzG9%paIlw-e?; zjH_{DIayrr6XXtqDQuPLtZN0bat_5Z`FHuk4{fopMUf|-0r!A~TR)?OV&ru?>PBZr zxDw-Ku$Rm+@f>*V$-7^n{z<%aJ)JaaG{l0Rpu#zVD=e)Twfv+}sVyyo4;VFUM3D31 zvw!$FZ+~CHLqMylL|)`kN#Iwr!{?658vWPSiD*ytl%mfU_l4Hd#lj;`+pJmtwL5=& zIr&evuRepvMa6Lldlo8=Beem)BimVQmqDUDhnRe5hLU`FHC3#lu14dXxw3Gv41|Zx z3_Y*=A9==8lPk9Rs~D(lNiB1gxGeT8Pz?UveSaLS=&e}XPs<>^E?ULorOB#?;8;Ol zUTetFvpC-af4Qy#8tq7Qg$L@0KRHE%BLGjM*c)RXmI>_5q>7g6+)Sf3{+28T{Au|y z=cD-6Z;*d@U#ipQ@7G+3bg#dn6Gq$2*Jy{}e)x@V)ACrY(luc4#qp1(B1H+)UcWL5 zf`3%-Gg#dS;3xuLsvNz4_pXJWyWW0Qov;I-8cD%N=x=j|luqruJw$DgxujgCiHqkk zs7T~0Xr77tC$C|A8R+7)zFCa1ZNa02ILZ|$N6eNi8g?AL9x#`*mF52sGsa~XbJFoB z5=qXt&ntgsSs2dPB6SRCY>RKpX<~~)SAW1l`lBNb=^%K=LOFs$wh6LmZon9hWQg$R zz?VQFJ-KhY6)Sm?1#mQcWqFF?Pa6#dysrjV+5v-#t^SE|g)crP6AXghn#zEz0_PxT zWpT_2jQcM>y!`k_$beL|j-O8xx!lo-G#aum?}fe$N8@fG9TG(*#F)nEPYYw+?|;qZ zZfQm3kS#AZH=G7~iTHAg9Iz4K71q8a|IukAeAIM7UkQcYjL0AXQzBepVIFaXiq2Hbk z4^#FbrWwg~?)g|I(5tb8I7u7>PJhU=#$)wHr0uzU8l(mFTSs{}dFxB0t8>MD5;09h zkP4M}hyMsj2N8=l@l9zWmfKemM+*n#uLfH8_tR%Smy@V%wHS#X4k!fdZ5FGr+V=4| z?T#B->mLxxzqOvBPrn%@3wjPqz3@FIf8^d8lo`J=!s43PRGq_CoPTF#*?+^c6oh?} z`}WdVa|Z;+HtvSy+YlA}L8f zSe|9aylOx&RIOAs0S2{G=6@;I+s}6`x}^q4Wu&%Hj-g{c!z`RJ?C=UP@2j8x3FiIX zAmYEoxvP2lC_LY`g)t9!%;p19v+S|`PS7w*(y{CKB$M!A&DkGbLGV??pZ58*IUd!b za{!B;?B>p%$7$34W+&@sEQ5}EGpnRqp;6pp!_nQfB7~waC=P(S@qfo-rNpTY;9K3q zVwmr_Dt|-|mRGFPwXv}o$Ko8L&N&(RJhT8j(piS{*0gU0?`{DU{|G|=HQ@NC3=oUg zYxgId=kH4PYjhlk7MwBO{y#al^nc#+|J!d7lkb!k7)$4g`#zm%25LzHT0R|0aiRFL zLn(jv;)0qk-$|?2;(vlmq1!SyH}|VJ@A3E;>#cZApYYHk@K8rDN}#L}F`X*L^pB2! z{!PsbV5GXvQ-6;Bj~~H#I@{*)^8dxiQ(?q1-V6pge@2)7YIa~Nrnr^DW?sg&-*Gha zlgOO-&qntb;Hpw8{k2r16HOsUp@9YS7x(tS-oC!R!pQrbWPeA#@A#k9JEa2?-4$y? zmK&K|PH3g)*8)h*sQ!3-sOipdBXI4LfULO3YG>iA^5balxs@(q;GyS3BQ3eZBxVbO z{KG~DuMk5Do541~;J?5{=i}nt{$pua!*Wh*D__2>ShR$23}K#FHF*~I$<3RUtg)N6 z9W*>7q?qG)g@3&lhF70;N~_Ae@u69`|wqf!Wkd>th`p97IGo5#fLS zLz(o(0dhJBZQKf4FP7t05y#}joRD4aY%lB= z&RUP=_ohcnbRgt*pTgV!+cTs@-DHNhRZ##_$Wg5;Lz&}eHU{~Ja2DW-pJwQ!=)_Sow zOn=q-*DGJbUdbvgf4^9VFi|WWS5TsUv`}Ky8&;xOy^a>Jgid|+zvDWB@T(_F?cS0w z(3H;Lcl=1k@6nks_sit1AC;QlP9YnO-Iap>J>Mdlg47%}%MH{Fa2Vx(fp=wN<0i5+ zr>G!k5ArRFWBOF4q+ljGNeJ_f=QwkXkADH$0LSpi9Pe4i66pThKel)(ElpKi{#hTb!QEdvQr{X(`1Ix*=6^#^!rR9B!2T$TQYIR|LPwja?2i5Uje4P11$$=P zQuEl4Zeq|ns|f0oyA17$0qDeSSzMt)skp5A?)ltyf&N7VrEv!tz2ca8 zk@cV5!=D|o?YkH!S3^%6=#$w6*W#T3PdX+WmbG*3ytsnHN2+8gz8{w6-sXG&nswi+ zLoMaYSABJL^Igj$E6>yLPJcc+45CfFl^3DZMtt~1Kot!XO5N8`RDIs?|Rf##Qw_O^)Z6nf6M zkA7a0sVz00^uFaLAb&tp_~b(rDvI+P4e{zoNa{s8=Ruxr^j`Y8R_W1g%&%x(@)x49@DFf)_r$l&hY!#FIptuBW`728M@`kYyVF<9Wm|goj@XUTth+5VLphjJh;I)H*-?+Pq7p6vI&a$e+{47>;h85)oS;eWP6@o1)hkTy;+C?5rjJ zeZ2I7zNyoCtNA4raWd+25j7$yehn!CQp9J8A3l9lD^`|PD_5Rp8#5%!S=jqsy|KSN zqSI2AFMmU*VLFufX}UyRO*|~GGwh<#{W2aSeoZTm8%cE9_pEV?V)oPT`Y4=0 zuamlO_clW^ZC}7fm4e4fx4+u*Vl4E&-WC|EI@P=X<3?Z&&hl`c)OyEsD87JGQ!k|T zaepf)bg}Vxw?(D; zBa&LRvgbZ_*xSidZyxcG_3s-q;r7W#l30GxP^_i56JkT~*u@rx}!fIOc(dnUE{A*g9yB$caH;hP4 zjZUNBp@yoMC@j$-@+4Mag{6Yu_jcI%P}o0|ldzBZ*lfAo{Unpu|ADGly{+t6WwF!a zON-V|uX^TB*3zYW@dUoe3vVOlXNDao@7)^Y^Z<@38PwqDqQ%_hSL@E#l&Fu8WO{M}FG$_n?!6Lx zA(>DTyvm2?IG}%MS>iGbn$a;YoXk?(W0#hI^c20crD$!`092nBN^o z0DnqGO&KaryB*!Se!cw^+LuMAM!#oIjzMr}-(ovY?A>D9_kj<>ca22}moDTlvfS=( zpq76I?+o|Sxg+oa8H$z041VrBwLzn>mrXF3q&8i0Q~G1u4<#^_(^ATrz4caiH0pnb zeaP~zufV`ZXIPp^uUW=A;GhI&1W9&0yZ!#r{_^huynQTgri{;Pl4k)l~ZLkA6g|C2ht*zKw%f`*9L*$ua#gaG}*C)TrgL zQm3)u;I`{f$g}+$uiNRYXq}LQynxTWRK3NF>KJLXBoI&w_}XfDe}A5wx#xewUAo`G z;1999=keW6=XHN*JN9N#WHRVT*Rp{|f-j2M%rD9DXErPY*YsYT`Foty8N6nq^f$%6$TYX1o z9i}A$cTf+P-G$3RRnj`EIA|fyZMH{s@#; z>~E261_4Vs8CF-yc)Wk{BXR4agrC-R1F`N`$T^kDHJdrBb0S>Oaf{c3A$?Ac@my8le%iYErpawfzT0?k-^&DLo8L+oW?|?yAOK3T8joBUbm~jg zR2%Iw^alL4xeV90&Iw9gJV#hp$wGwmTHUB>%<&3Uv;o*A(U8iH8wG(p1S%$A%xz_Z zt&XRsw4?D-^RZE!XYU{fddN=&l?FMj60Qu6v!T81`wV|+twvi=al<6n73;NI1Ird8 zF79mEVv#~7clqwVs0=>0u%ys{iD+TS6Gqm)%mRV5^;5_8yPaj?eERU+&IWfhmsjEBCPDI1ks? zRt?X#DpAb~QaGAu{cR@QkDJ~74hpQhlE~OzJxTnjmO9&FKW*4{-4`0|f5Iq47CR>_ zFUf+v+Zz=oF?L{hcyS#X=Qn;-Vw1wUpAjS4EH!^aQ}uAWcmlp}pvHytN07wK$<-h` zXx{o7Up*Y;fQq7)KW9i96cqNWW*%z0=a%07BP841OS^|jvJvons;vH1XXUS z*`8ue*?vE@Nlla)&3Q5aKw>=Gt&n)!%Qu_|mfd zU2m~418*GEr(Mk-I43guD)hXEsrKCu!Owrm?g{leXp45v!wD;u%ttfj&om+!TQ9Tg zDH%gNh@-8TD+R@^Ef?EJya{c!>s4*F8&AU|TE*LiqWAb;%RwCp8G*imlox0UY=fQ= zw#W7j2XZ?Ox{9{W#&{<#5aXA4enw1ATl-*Dp(9^XhvQ2blosY%j_Q!D^f71MbNqk* zq_s#ri7fccT5fw`JiHvU%c`82=ulMX~0M5z#*O~P+SHv`11{ueMQ zzLm6mJz%)(SLI@nMB^>rA+O1!7SrXVX>Qlqxa)uB0>^r&e zrOD4SW=K)V6Fd;_hIZJkt#3(_CN+P_>y@aY@ghR?^a@EY=N!J!N=!O`i{`CTeyfr% zCZENUWT)DXs<6@DZY=LjkCnBT9gE4?Ad~PxJcM~{3Y^9@jkoaSi=IF4q2WhVEJlO+ zqO~58ff5eVsoKKoo=ccOu|yMkKSn@RyPC)?Cmk!tZu9$@TAdWG0_W5P?v{Tg?pknK ztWX z?*o5j^@?eFpHnvwXGEdOM<0KyQToHV2JfSshlyVMLA^^|4>&D@@Ekn6uPNUo?o>)j z@6Xp{&Ai)3)vf#TB6ig!eBC&>Kb4z_W%jleK#9UT1*({Oua(CbfbwRDfKJGTqnJ1Q+5NBBuamyub2Dm$#pR1 z%^2Wb{n2g;)g;L7mW=y@)>WJz*GACSd7{hXdA$TK>KDz^2xOn`qi%kD{+ia3Uj)r& zSZ?y<3pCb>O+S|atjyRqInS<$Wr)3M%D~?H-1H-q^sXro4S`l0y2m$Dp{DYn?}TP;bBO@HC}JtUf4;A__VH_x zVpGg6o|5G0GD$3NrJP)WC(7#+++|$|37#K%J+rEkP1Uf{7l42J+4o4ZVeYcl&=%sNG$&nA;fz2+81%F z{~AN3T$_Mn^+yNoB+>ABif=3y{%O>Ff17|$RktDeEs@l-z+XHWQ67&`2<*`w*fB1@ z<;ARovp%)R(GGuVzzHyk_|09OfF!5O^}_w%Wtw@}ZD4s{ogb)OQ51|#eo6*o5}IX@ z3YvU=k-6l?jNCMmG<;9lvE!?L`tWj%)mG(u&nuJ+XO$-JgnkUZFC0TRqu+C^j_<|2 zMpLjaKBR&^mi&-5lJsZ$E~Q{M`6kz5=(om{Z2`d#Vy=G*(J=8tO3ba`cmN$WTp`m4 z!{;^!ES)hz+(78{86NM7y95?cnd=gJh(an}`wVDxv_1pS3VK_1-@$S_R$tdJuQ-V~ zkDrI&B30Q9`I;yBPH(2bE#e00_B}8=D=1Y=s>3MwN+pmway1QE;h8vFrGFT?G{G^UL-(tTkV++F@Kz!lmk%4r_S+3I}vnZgbOtv$u zCv~xDep+&OS~HbIRNzXyaD^EbzB6tc`r$&R|7wBaQQBN?vr1pF_GBRP>^;!+GYrN9 zmgQ4x$I<}F+sBRU@t(zciP;!*np{Q-XvIy;yP$t64TtYD+sqXH+=8J~7<_p4iyjg6 zlSbP1qAi%uXS`oZe%hfN31!m%Y|HH+42p+h)jl51|15ns<@hiT+I)kSezaKsItVDs z>jwnb>H;&@{b8*3W%A#wHbuyXUUxL0E1zR(D2=)~ZIOnWr(MLk#B@l*a&P=5G+nR3 z#S(uU|9yAX^7%#w31Y*<$@AHZ-!N>#9?l^8pqqM~scCI&U(BrMB;xNo+awk2s?2n2 zrZAB4%5drq;nOhQ{XXK(^3|}50p)KW_&o`MsOk~uePvsRW8c?Z#|2s&1!O1!M~JcSOuhYD=4)Gz zLYa9a96o*9CpZ*%X<=UMVFpX(0KhUGl$Gx+A2t02a05J*(~K=9fsqQ zq{iVVROsw0fpi2z4>QAb#BTrl4sywNQq$WLDrs&x%=epBD6MT;_7`h0Z zdr$wV4aU70YlpY_?5>X%YBh9zkP5I$u(R?DgBKuqi8z*>b-5=ift(4nbe?Dg4sQ52RN3Z&dq}7GvJuYokfaH`;?Htc)6LevmC>>;7CJ zmPpgA=KZA=TvCD*4qLBRogf;Hh)Ud;DqP#X8eYurH~j^dXV+bCRcQ?4#>~{qJr(fp zPND8#2CXz9sji#DA^SR&pWlBl%O{mDb2bO(a!ejOTsjrEoZ0lj&kQNGen@M+*HEO= zoOvUyWkZQ;1pUS!#`&F0f2v471MB5&b*JLT0Q=bTGfT#V$=YBjBQtqjy4<=8*-TdMrr|K!WV91OatIvB3rYwc?7xy_1I6gfGxp2qOIab-F79B_?pY(#dZSVEu zemyH93<{K)K4vFRRX!N+(3oJ@Hl|Uq5IS0Hm{CRpM_p6Y`q=r#?t!lCA`vXikuV6U z2BHOCSbZ8NoW=13JG5o3H9uRh#*VcbI70Od%;OSo=j5m!?V|ZVb)WBmD@D&cG`I?4haHd`!SD$a&Tb^@2sX89U@CmKXA) zgaM(oXi+Zk9!Yw6<;`@qzQA;B9=vLq65|&0EVlV3y@dduTST4ehyD` znwFF@ok$ye&a;1C2ux}T4czw~)hR?pq=%DRYUK-IJCDha40_1L4AtR?m5}4&>5&x6 zlnb?JRgo_&cg$0`Qg7RRA(4ZU7g5#5lGJ^szc!L2*~%&Z;@{egY#_fAk$$n-``tnj zUCZ(8j>JjhnI8x!erg96<-JKSkOAUJ6yGVWNg}doh5X|4>4!`%3Ue;{1jVFGT!oEyokRmy0^i6hsG+lDS%9Dkk( z#2{iS%~G$(35Vz6kiN0=IeHifIhlAR2FUI>mnuAD2ts_$q7H{u7gkM5Sb`xyYn5$rGltLGNy0%)R8sVain?1^|zQ1~4z- zEQa)IH)vKTFSwVbBt6#A<;ajbOJC+6F%#Kmk+y9BYmODhBq38{pt z`d_FYvVMx|`=Lrh#h41Zc&_(KN1kYUPO%^Bh_-)SRl4TW3APxA;%O<+%viz0$s>7gf6cC{-r?8Z5ovMg#ghB1>XBA7O{}K z3k@;2Kn%-)h2%qo$UxlCXeKmVNiL^oF_%hh_8n!_Xm0khkTKYD?kuR^;_x7mprZJ&Xj7=dft zd^EMtZekmSGeZLJ6SC3eyb03i_g1HTnN*W*F;aywA1_~`CCYn83Atr0Jl^q>tz3V% zMeSSb4yS~xnMdo5cM3Zo0vkv*qC%bQx0K$GB6lmReI}sP~_)pOs69!*KgHoDOfej<=e8)J%bk3)DyJSSz6j;EZQy*6@u-3O4BraRWhxOGBMvxG@R0@56tNBwk4tq-ZW-c+;ck_CN5Nkcduy& z-at=FP@*ZBJqM%2HYw4x;vsO*yAo|ewWRkO>{dG(8z?x&Qqz=qY}NC%Cq(jxUC2#D z+mv#44?GedoUH+dw>*_KhQ>auF|R2E{ED#Uk>eIht)@$y65}$tKi?f$lrewR)uUXb zQGN!#`w{9g(YsJ7tk&OUU<|$~=epH8lapCGe>2$j=9W2GyC%-4x2YBZ^*yD}{BUeO z{Zx!`e?6W90`4IDrz8SL#djQuT=t+`mhyCxt&iMago;LPJQv;CO5Xh-+2gG6=YIIi zpWOJlSWvfAse@mdh$LQ2>3e^R-+Gm3s;iL^^7 zwzfP2wZjimHYzl@S|9GsDzLOP+y|jy$7gw+hN37|v|0rkf<8tZ`S~)51uU*bYPJcW zwqJ5%oX5#~UN@M^S4%S#lJf<_j^mtM;1Ykoxs)a%I#o@~*Jv>Xmm+@;64CnHuKRM} zc+|QCx{n;T`%M&TI-_Bn!A^CQ($A6~Q%Tu$1^J;4-?Fm7&ryaA{Q1PA4)SH>)!T#o#t<_Qcyh5ViXlyXFF$aH)0L|`}BB9X$8(@I!b z%o5A0v~jzoUZo!^)x&HXL9@K@=5dQ&!{T`9x6Dk4zm#Vxk9j86LY*{T>Khl9pA#pB zeZS&62{B-dc>nBu35*d}EDm=R;KzKIe3A3}9S=5Q{T^_qh(>>m-w5GFfQooVXrnW~ z#AD?PGuSx#Flu{vb)Ls}S>WUVhd4Agp?U2;dEDLt!zy2|C80)Cb&%sSg|5 zIQ!S?ne&EhcvrPJ#0Fg zNylRCP8G;|u(AzWo%(?lt|Co5ciH`oTW*MOL{)?L!@$&P10(0h;l=S$ZujGbxadoE zmvJ51Mi+m_ix+hJRkUX-tv};}j5oac{;w*|GAgRRZTm`hs(_?0f}nH_IW&xPOLuom zw{#;lG)PK!cStJDP(w3xcf5E#*YmvhJzwU-UVG14|26w}9>-bxP0(_=GEK`)h$iVT&<=Fuv5QT!xM^e7c_fT0i?*DY z_(JWJLRvgl8GHP4nt?^U_i!@b&w$62L}bIFwPE_HKT4f@~EJnYbYmtw{C11(3x7Zt=Q6@&3e6&cLH|s zU+hCFtj?FP9VuD`A8y5X=lhskKrf({JNlM1Z#aGC6pW~exY%-X`%UdT&$F0mE$OV; zAiH~Wpec^Ee#Q|}Kv%JD8}EMN>G?Am$r^u5bv~^$3R@|6^x^BaehrgKft<^ei2%_V zQ%F}Udt8MwLIVK;tW}4bmj_ZxGFq;m`FqZtbWIj&l5eLjNL^a%+W~P@ZLn;&-CB}c zJWdc;>Q2AAF~5iGnTm_3M=?&S-sl&M+NcjKoM-y&lBsUJq$tYll(t8ae2aUNsYrjw zEH7U0YW&>=fPM4GlrPqZxbQ69i_cyV>_}2ZO7B|bGJTv#=eP+$>awU^F90nXDCp&s zoH50n3Z)SXa-`P2I=i@M#<8hCnk#0Mh*v2Je?3G;mRC+`&`#uwxo|y9(keNzt)Sap zHGa88xGNT&#a~Pm?0YvH0*(;QY}?n2N{^hUuSc^BGIt0MWiH&Ve^^N5!MWVe?!aYq z3H=U(M`V*3k2W0-?I&bP;$fHso%-F^tmt6tvBuV|{ioY;bulJy%Z#(slg7T)XX@dA zI<#?FlT4bq25YrV+aHz#Bf~;_K7oAXi9s_ii!a{2-c6)Ts3og#e}NP3P}K!45xTd2 zJiFocAXmu9$HJ+S!oYdfiRpj;+J7|nFhlYboxOUomiHHxjWFfZ&_G)hv4*-mZb16e z>7P6LpqGSp+h45@3H2W$U1~?O+q){AL6b4_zG}4S>0AUqA-)TZZ~B>I{L$8@abSns zl;5(7_L@^Wgzegdk*ap3X!}wgCGf(C<7I42yK+tP+Yf`($nW8FgzcBU zzsPWIHgC=TMI7QUXS1=4X{Y#A=RXL0Aa0|vIM~2ULsRg7>pNWzqyU1@#rxFDEk1iT zU&xV3$Jvsq~7#E0r$ z${<*dr8uipiHLijjbT-~+~L%vhN7|bgkUMDma1`OC@*u{LO++_8|5RDZKQ~CSr?7Q zY-6MPws|;9qmRE)s6dF@W}#A@m$C(wY3BV<=o%lOY~6nb+uvSo>-9j|FCIOP4sVz4 z&psB*Y1dQB*;jk5Pz>{o7ZMyQ6M)Y^ZuS{u7g?-+G>5n^2@#TYcYKZh=I zVR~@BMe42E##?8zJY^>t{YP8B$#rj`1=Ty|8~E`rqvW2Q!0INq)V!0HFPX`I`C*)C zXD;-x`!XsEo8fU*3#+_PG@ULgVu|@Ao;IrSY?}hj?~-#cX%Mc15!GJykku>o zeRhrms2XObx@pU;Nt0AQov)+QBhmfE6qf3W6m9{YPNTcq;$Z$wi0kpqvs)$KISoqI zloK<#ZHTfmb@cIZ#}Z4C+3pFY*=-)X8J0G#ivf8s!%tBrjM#m<2xJ`xj^so0x43^O z{Pv+dz1m{9P4t zECoK}&X>pmBXsu@oJ)aY|0*%I5K*7TOy~?7?Q-h=fY5vH%eqQE^(G~gc!a~Yir^Ex zi!--dRHMCx+Q9q&l40`SUL*vzh97@y4b*b4;3v=86+%+$lOgbNuWgp509Y>xJG#`C zr-IvcP4N2hm>F@$Sf%3Ao7XOiGz=%`2*qfA&Q2H5C>7KY*}z*D9o$1Q-2@+zfXTGR zT6{eEMF3D1P)Ep8rYS+w1(r0u7hZAQyCiqHyR>nM5K9V-wZA^bzlh~xF}QziTbAau z)~5mX0f22FLH9togGf-8zuF0Bsyy@na4*-MhoQ3 zq3?`&X>C&}p~`jqKNR3-h>uO54JSLE48C1PAvQ$6p=Tc}+G70jvwWqo7_8odP;?Xe z*38W3n6F;L_=(9rWTQMRp}l{wO1}7;nC9MW1JO@H6oQ{gpYFg<-!zL&nK{9Y!R}!; zetiH(KVrj@BGDGjhIbN~dy}Pn#R<}a`K|>{XIqRH2W}v9D=GT`bBD=KnyvgfI9B^b z4FYzNG-SP*6&ek~lB|h?j?<+eb%Qqx3R<~K&2ecxNsPJ^NA;jtsyctVa>E<}utN=) zu1i+7Xg<(C1N!$#ozKgX+YCdkF_BCkf*a;P(^WoFD8G4_ifHgYQi-94ShEV z@dAuK&cv*{?ni<|K$A0rhGO}CX%I^?hMexAm^DCoU;l?ciEMhUF{N8P{{(JDYwF5n z%)~#Q(&x|rdCHialecF)+%-f>1mGp=$QoU-`J|~`e$yOgk4=>hAK_Bcvk@ByeAkU+Vt@X?o}U0osNHh8Oc zImlSE2nokAj9%)=BI#YEFQ|hbh1#d0xfkv^8nR0F7fFE({pd&U^+eYBS9yHj;ivBy z;*P9<9u$AB>TW~UaRKXVN)$4YuybwIY^P842PQnZ()+5Nk9NK{R8_Vy9=@Ha4KiEz zJ}%uF5)CVghnS5&44{$%F{}$Nv`)Z{xV>Ta;eFa;JAI`Iz4 zJ-Djr`4S&duPG6Sovi`SVY*j1QenE7-ozWET;UfGahckkB|@aaetLYP$vvy)O$@{f zEk&?>z;1D%SPD>{;X2!J_g6w8y+sW&nq1V|neD9Gu+e~sd=Cr=!Psbx1z-u@{X_b; zpGto$^^PR8P2_Xgu@)b-eO@?ii=QVwGcp1erfLEMlm}E zLCXXy@YUl^pj(-64S|_ffbL50gQ&_Cy%MDom9J065dtzsWkUD2vp6FTijueO| z?}0|q0UH6%GBLIl#cxKlN|SrPy$!Wuv9*bxeA$H_62?Lf^@k-ggjx0tZ4;ym*`F#Y zqM{4202PiF3PXIjpY#LodAjITZD8vHaX9}nBro0PW{82~7gAp*vc7TgS zN5{iu?h)EKY!_R-UGNCp<=W)N@t6jI_FL^|DQB(`113wR_iCg50R_*meOrHaA0UmM z!7j4U|9`&L27DqgtVhJpidVf))XI?8{!)>z{L$$ZV*4-cT6a_y!xs_r@dJh@0xBRZgX(<3(Y zN{o}jL4#cfl{)jMZbJspJv?8ve@3t(?$st2lciyk%-4XOZ#*a?dKPl#^f}n(%O(UBvcAH%oY0Us`qMZn-*+*O-6)ixi=L3(HX&YTO_$&&e^b?dyLAJYvo)`ja=2=m3WjpD+yV z&Wvi-MLe3SD7V=&SjK>C4~zuqY~l*tmr&PZ6NrC@ZJ60Ewif@e>&znU zLbq-7ZnwF!y;!i5oYEn%JK~6qo~#=#6;mHH$LG5I0~td zv8;mDr||vPc{G1H?UmNA2Q#Kn zSpQ8CawTWv)4+XKC~4p9`{e)FAA0wrKT73jc`4%;(qmN9Z|DW&mF-~l2RC+>LG8w{ zDAF3py5+a(cvzwe7j?wt;r!@G^u-duYnl#@7aQ3}JIH@+(DBs=j6`=Xt@^Hu9?(U< z+e!n{aM2GSU8O{dy+;NM$G$8<$@ZeP9 z;oN8%D*mUCw>dkm_bx2R34VNn6O}55`2GD4>0~9<5*4MtZO?S_12`XqCpd8ldXmOd zasGL)pPYYj67MeIVB_B^=yJBbn(P7{t;qDAw!1NTw#q*1+3F8xJxt~qo!qCqWD(Vj zG-c`L|MwCH+TV{hC|{&tUHlU8Y(=rM$!55`50-X-%NZen!K;;9lSiaAdfEYVVB^cY z0r@Y?ib4W^_QbaO(|Sh93T_^ky;03a9haLyNV$Jp{T@GcWC%H8(AGD(ck5nmvo1M0 zT>t-fmmCx@`7ZuNYd}7?U`}1oq#vGAQhxpPPesvEz23$vYP@vobw!muA${{pM2#w2 z_PwSk{{8#t>nIn%DKa8G{}`r(@U6#4m)ilflv!J+?Ts?Hm?rY6%j<_~>DKPFIxisq zzS4g)_A%jMjikRLOOFDPJpneuM^*;e#9uC)@m$ymJG+Ib=QKp zwe8X<`uF2<_w}RhQRea;Z~D=DX?4KhbbWt&)Wa|3|Lj>H!he~7_NvxP*TCAgAmhfi zfM{kJ%gU2`eONJokZoIn$q~xu>lwT$vd~?rW$XDIxEC+Q0iXb(uSl+wHoMb%TswYW zoJ9}PCSjG`0HDhebRPcx*v=XtLUz9SQQess508qm!1DW0iPw!|iE$VK`ZVrbwyu8+ zrdP_Nn6txY`!rr`f^;Uhme*L%AxX_vT<;Gj&2GuqsF#c7YmhuDVB5@(1i9G*Tw5X0 zsAV4B5ch4j&>PFq$LvS3r ze6^k{f0RGhY>D@u{7EgLN^#X^dZ~YoV%Y|>NzT2{>u%Ih1!eP+qJ>ry2(;RzCE8=K zle-Q|P~jXq?)$WD<*lz?Y$(oYXeGL8syMd& zWfa*5_v0OcvrhD647@)Kn)=LY^y)myprWWTEbeC;irA!Fzl~|7HW;2=a_&#Jqov6U z>_^S`+Mg6l%Bp%n@MmLCra3*7Pdla;PB1kt6J1K;yoG(xdj!drv2Lewgy|@atA7ur zq4|buuqPyc)*=??26&4yNtJ(UyeJ&>aL!Cv8zfj^PX~UzvEQiD8ZAcc^~1>XK;Yw0 zcxVojcxh2yJ8=nvP5-{9{f9%B%0nz%IFc#ojmrbtS=+T1?0AUhF;@>EZnY~gq#2D0 zU#RcXt_eweqCey`HUPn>-T*H>oglq?%9lK6J!TfD@3ABdKvR=M1`B`K#+y)ep)=SU z*k|u2!d`I*1YMnyZY+pY%h|&J0}ErAZeB!_`|qf0>AQaYBvP#3Ot*Yy17ZZrMPwBO z>~uWIfy%6uUwLcZ-O}79I?XKoF-S}RlCOeubaj5$KX#nblR@)6#cW(hW&4BP^Osa? zuCP9_pfeTNx_BQe>HL50aLLm{r_0GlJ69hu{-sKK>RBm^yV&F7>*U9sbYJl%F4%qA zc;zNr&utqj2_e2D@lTlVOggf|YZqjWDr8q{4t%ruq28-9#j>;#<-&pzv9r;MAD=B9 z$;bcDd2`F^eS6sjU#g(5L?+T3?UcM&5 zcVP*t0Em;klXicMp>Jm-^CVA)UWPs6&z^VBv7UW*KW!k$VpjLr6)crze(PGW0-W7@udg^nZ{UGCuSN_8;4u|YpzcX>8s9>nh&I9a_k1O?pPDjl1v)XATVm5 zgC-aHNoOKf9mw5KdFgZG6fusX~Mc;y-e2(P5gFk<08$if~BaGtTHCoGkd!Z;E z1~Xuv%h#>mr#pC+1gkaw<>&$|S2H{xG-ziStUsE|P>?4Kf^Cc(3B;!%Op`WZhx+jO zw%=QJ-xMbh-C3F*o%g;99@9jGJpC~r##v6{oPsE?G|Vkb81PY zNMZRch2@8K0J=7d905KmT1^vWzID+!RpJdv!XSS-0!}EEVBzn-p89j`!<(kv5#E(@r3`QFtg5a%+_MZs`AXcfbsK;xFq8xittp$wF(a_^&f( zK~o-wQT78$MX?!G1Ppvy{Z-JrB5BwZni_ktZ17KGR=6)CaC!~NLeE}u1Dv9B>qLm1 z;-QBg4C9-_RhXm-#naP`WG{=wenvOogR_4RR>pOxi+Ki|3DR@5Ns;KY^qK;CyLf}Q8$Ki`7NW~$L-i`ZnYq`g`1 zeiA(CDp)o14a01ToR}1yN=EM}lF}*?5f!&sHecK47vN9&^~+b{)m5i zPDsBuanX=1I9_}yvWeNCWxv~qs@sg0Qi9B2%lG)m4Cq_WlS{&d5kLrFtB5R(Ke&XJ zVgLw0JsT2u+D|79F$5v6{oRC(}=}DHakoDTL zZW5+;{ZB!~8f|HBEv43`lcGU8o{xXv3@Z65HbR-@Sq0Fd2zA3 zah^t&ncFOLWM$00DF7CP)f^YIxZl>Df_#Z#cRLKfg18w*JSg>HZab?1yq_N+Mfft- zQLRF!Li7qg-J=>n*lX9LKlyA6UWZgZme(f_;*}rv&K55^H5)7ChWYdQrU5PqWYb3QpiX24I5(W&i z#h)ox?HRQS7XVOs-MBkd3Up!t{(fb5w8WWAx-+%QcntUo&HX|S?y*=M8-$Ouq zJ8nM7n_$;zp^s?I2j28cWubo{NdH`wAQ*dVyX*WwO3<5)@Mn1cX;WXM`W2d7-U?U{ zZJ{m%ieKCo=&usvtn$k7cA)5SzJ5dKBCnqcGu@GGEU)&E+>4)Q&GLD#**++SksSz zi5c;=CzX1qP`DU2^xl|o15&BS4|*aXNI$d)`%1HdL)V4Anm59?*5eOWkXcsf3;sDwPfw8a>! zGj%~TaKxnQV!-4A8rL`@yWZUyKX5fF2nlV;}!G=-U2UZL+ z6UZqkYTIrWrK|8B(oj%O)%VhwHC3KH@kAS*LTDuA*ax>150obsg{X_0ZAKD#a&4q;!PHqH8nJeedujD~8iVmAZCayADQeLiK&bX(&<<9XVc9}k?JyJ^-09_ zVP-))|EG3CPNt|h50qx>^%q1>9UnhScUo*J-y&45XS08@C&ZMn;zWv{ltX}7fLK-| z^GiwiR-WBQs>-PDm(Sjt@^v&jEzDD5VyAIy!AM|k`Cd9`feasB)#X71?7p8}0p-|b zCTGNF#=>f?u5@isi*pK)=wjE~qomn!stZD#|BSb%6kU;)Bm^}ygka0$2j7?{{$sqm zXBxzE$e4c>4>3~|&1U&&%|YR6-g?Io?0hPzB-t&!rL9HPkQw*1Y}k>H5}|>G;WAQw?Z)l7wjiO>Quha(-}hX*3ga zdwc9LV593Cl5x~<*smg=_(|;pU&8m=zgr|kkOF`3(hry0q#@A>&LvS)8bwcg7TcYR z_7}YOl9W8DDyZ5~rG9AA_H5j{R7)mg#S|ab90xrz#r_0-IN854Yt|lRQ_o6r)8wCY>Q{Nz)T#pfvg zi&WnzUWKpAd#zywpJcUOa1NmgiJ;GVYqdwxnT1<+iI!f7g)3lEKne3(l>5|ohVLIO zaxn&hwkSbKcUP1pymeMZ49*`+ZFeX@4&i?Q®rHV~+3x(})4af{)5Kz{F3qj0WN zQ?<#+2ZAHu{#}-@caAT2aCr0V1Df0ezv`6xG@_KOKBqiqiTW(+fGsLXfu7Va^Xcl; zv2_h4?z8pCrrV{MUUNg3ggwUOxcm2m!MJtH$6Vr63=>i=W516*l%()Rf!jn|3z2`m z(GbG^_cz$5vILZ|x@_-(;;C*Mm^NF(6?A)_Og~4910wmB9n9HW&!tq06e5pN_$Jfe zTli0cHL0M0sfkvQPbB& z=xW`a6w{L`-{Ii*kSmWek7r7}R9>!;$IJI#ASc(>C;gKJ4N&JD0%<=cG2XFIZw*P! zW0yC=w7l|zer&bq(EHL8&N~S$k40DOkSJ41h^=oV$0(Z#^y(%mPZ)Sf$D0QTV zRjVzugmGKtZ_@;EP_fzZIEt}(Y#zp&4@KW5U!EpXnwl3L>a_ee3ESFdEGwMlMrqnT zwi}otNvlyWRqHzo=-F?=BvhIbd?04!L0a<}G>uJwSiRsNUfc)dG96VEhjem%bUqSGz_SO z=!E`hNCGVXy{8WstHe($nOVF~?}0Nb{7jAHP&9WJS{ z&E4EWk2W0Vpsl*$L@FT5mwLk=g2V;`l}L0!i2ec7@>8W+A!5T<>pKMFl7PCcJNikE zKf#O{NOW^&C}_(ofB+(uL7q_~XsFnHxpq@YMk|{3Smixw$oh&mNJqf^Hi%`c;#(}V z>@8=ew6lS4B0zvRd9 z`CGeWP%Fyrj~v=}1+85!ykOrGJjGB_?P_$U3NkbL{KGFJ3WntAWJIz1MPWr|jo`jh zU)kFuT>%xWv)!SD(HJd5V%f6$Jj2jG-*%b+IfJ4r*}f zMvG~=+Mk`MOA%X@un$BN#mHp13 zAreI_2nfqAA>YBp>8kvLjZUSI$Le~%0G=4D&h-)dgOZ(Vqocj{Ql-%E9L{0Bz-~Gw zeg4wk7avWoNd4*lO5Fo%K$tXi=wfdc%peG8Wz?uw68PbNkSWWHR7lF9yN36wz(KUA zbGPTu5B{EW4fy$h6wCkkl~Ik36bi)0VP}5%o~C{ivy!_9jX)5Ug>rfMBn;GcdQ^j8QQbeFIazOT*k6ibcIx& z^)T``XMN(>tbTm=p9(os(v7BBQLX;4bwESv#mDf6m+Yf5TlerO?#KEq_P=?7<^8mQ z6q{Lp9yhMDK(`K_^nZ!~lwIgb7*$Mt9ykRKdADNcw#BMtc@IX9UXS<4ZLT6K*wvgf zek}Fv*vYM2cJn#mujq?-UA#2DpD2Ws(ZqrNYZDh6(g>%5ipE*Qy|P&mW-^ln>t^HVHb(>8{!S}C@g)M1Oj#NyM-B#ob>@Twf(pxj zy$p8=6X84bO;b)?Dxk~zYJUs+JZJc+Dii~Pqu`Ee@iw+S5IfXqo5?&EXDz^ggSP1S z^MDX%7xaSEMPbp)ZmKA=L>QGu6;8`*OQ+ZFrD}8I-lBH#QvN%5Y2X8|7c1(>e($`7 zM8}cgZOE4!yETTHISK~61z;$}{+q3TrX-bDg6gs?wBoA7EC^jlE@&vI_T+rE%HK;> zCJ0c=2~Lc+%iEkIo*GMZ>q?qMOFC@vUY4<`OlhEfi9P2C6Wcf|aNhj&S`ENm*jN=KX=pZ!gO3Q9`D^8Plx!AJ##KB65oriSo~ z5TkU&mh1&B`7J`D9RPt$rM#=Qf7D{&Wp698Gy$Lk3h2Wk7^iZ7Z`a zE7rKW`Yp3H-g?0&E{MTzUCFr5;2PhJ+PqI6*Z9qAamgxS^}c^{#Z^pysIP(VCq{pW z&gD745r2G6nsOmBR>R{zQp^BJ8x3t`HCVVXhPE))E!a+Rr_p}gm3(FMBK{Yx{Eed7 z>y;9c;2u%yOfW zz{^GlAG(xtGPU!sI=h`x5GW&z;hhR7V-F&;3eZyVswtisa8;gaf`>a9X+75HK3 zyNXl{rSO1Q#}QnA1|eSLi#R!suPoy0&B*ud0=Ipd`AlTJ2vX}^yiVL@-X~GQnux#9 z_e5gjZD@B=%rWfReAP-Wgk=b2Od&VlB!rGgi)f zoh?<~$tDTj?nh$hI%?DktxKhBy&GNqr(P7eKal>T%4}>=9Nm!dG(@o6tI1I_fTmxV zcpq%|7}So?Ki|-|MW;;Q^i;H^%FB8~^PkmSO0<`M3xXY5mu>eRO%7hd_mz5Q@Hb2g z=yHRvXybkQ$78Zu7$2Rx=hl8<@WsIrUl)B&4-mSf&bBKq(u5qT{Y}Cs-Q{fBz`YO{!J0`O#2`Df>+YKKKI*HbBcS zU!{A0vlGY1LE7!`UVOLf;m-7Tz8jf`U*!3?uW#|eLfsG2?>h15a`sJpvM zxX9DrQ95LfMDzs8god}6ckR(qhCqN{>hbb_mr~1EPZefU<+gUQpjX_1x16U3?`{=H zFUkmPBw}bqQb&azP;K*c+uqVN&);|AC}HWs=ax%+R)I_W1EY?_6v-FU1Y=6LWMqhl_%;uplJx z=Tu@sx7uG8lV+t0VlXH6*J1ZO#>aktfI2g|4#n@UPeiC;WV`Z}VK#3UQUdH+zQX16 zXZDKV)3?h-28g?u&4pykCgKmFCD{75Ri+0X;}?YeL-%45#&tilV#AAGy>=1wj(^S4 z!!`(faYXzPMr2xq+668xq6z!R@^E`O1F(DOxekQP0c#Q6yG+9g!mOqS4FEcSwiSEC zug@1mMkkcwQIO-syYY%o?XrA|aAoGltogVC2=7gxlunw8we%ehzf1Xj=8($>Kg6^@ zbT~#ar9FGfTt*uIf9lP90h3wT7m*KVP6EZ=ixLD&bJwG!b6G2sN2H>;&rxV6V%%Ng zd&rjIQgi&d7mHX0siEWPx9kfg*g-RGF}fFvHNl(6Zy>2C`AqSPq~?3mb;%b}0t zQ%^`cd%0go#Z;N{(8??R1(#5%2(|VZA$>2{zFQnmYU$=;R&-3i|A=1`n;l@SlQ)cI z5=MpnI)IDNS*AAF1&Vy&nfo>-R}+KO>oFGow`zasEV-g;oh67=>X`I{|njA(X9_;KS9ND+rArT&bzEt7vBFz_3!^VPnP$HJ~UQgE%qx)yX(EY0s-F`n1W%rpA->qGG9zoHY>p6Pm{wv?jfM9F}Ks1$;h#VOUJC(6i6|?2^oN zlRQzdp2}`Et*oit5C7f>T2WyPBnHc?p4~~%xeuxBD zi)EUZ+v%p-!rP61j66`un=R=napG+ppvHYRPN5AA#iKBe8bUL(0&&%TJ zQX@c1Gaw3?v_&x`n^eD%HAFOe$@7XX?$G{NH6htL!n*$xclMz`6( z`ZJRs$ht(WOjAR;nqHE7*oO8a*HkPH*G;`5kKG?eHR7p%`HrUx(II8Q2fk;cIQ4u^ zkv;Wg3vL#0?yZ+pr|MOP?u^K&6GV$xIBdxYy@`2}^1rt(*8z5%R#OUXoUDM3+x@dJ z=x78C30L@*ryNsv<8v?aF9xlKz9G|+yZ|lwW}U@v`ETn0VM?|(z(HGT@7oI1WpY<|F_5j*OXCl)@jTTa#!RMxOu{$Lf#Hv?a;v+VV@;kXqvn{h9mJ;-C z%(E*XxVDyO>Opnts>2+KE|GA>RG=6h*h?!FpkM8Oc)(j%_xY{Vo0GSB-CkOFyHsIk z7+|-Jo2lcR!1$>Jl!F?!gBV5E6q$v@72u94#ot!4n9auX z$%#FGmKZVM4WL>erzyiCJkEHq1dv)L*)WE2q>I8Z=^Ed>=mf|K29HxotWu+@xEL+M zI73Oh($`}QZSYi^fbzQIR7QreM-~OE;`WBVRB^Do6*p~rf8*q3ffo29K7rRgnp`OJ zV9FreLm=={KoLJZHu=Mh{}`Lebn~bTs*NpwX&H8TLOyhxHb!l7p+O$k8{E^&P98wG z)+hc!%4sq_Eq(^-e!iLG;bK*+&gW8lsbz3GzHJiPw%R>q*Fmnh4bJ^It~Z>_EN{^! z6A!lf*wL6_q_1+eTD05D?mKpRcJu*VE|k#dgc%hFj%lG@L7N~|S#Bp5e_hGtx2p($ zBD{DMD)>8kxT;Q3==tgnVGMXedNVTA)G>d!#8m0fe!pd}d%=OVEQ~IfUcqiQW|3|i zH`Lr(i>n-V$ycIM{_}lHH}2Gz7W{Nrv^Q-LruXG}D(xN*vwwFC@qbNY^C8%7WXYV7 zb1YLm23$J~SE$dC(E#7ov^wl3In>90Kq&_xkj&g%CT?%+UlCNX-e`VUL8MZ^e811H zD{vN=)nh0@ipflRGrJT+6aJJ_NGWS858WXSzQ?c}T4kYKOcm$vKWZmN`FcIq{a}Yn zp0zJEKbJ~6($EL}${S_oVtl9ha9PGzV$P~GDi*RFj7e1O?ou}_m#)LUx|Xef`la^= z0yb5hnJzbjPHt%LThvd$#Wvxf#iV>nI4fvw*)PbrFC;r-UO;<5Vnu z#hJz*gMDft{(ZRBpB;EVp-W>(e)9n>KBe7+sUG@}7BbWeClGYp?s`z7wu-iQ>1B%Kw;P*97Kcep;? zUQYb@yenGl+#gpojPcr9K8t^P1!n!a8iV_=l(N+HVS-;={oYnu3VNINlZ|RNH;Z26f#UYJj7-4;YazLR3uNu9(;0hs zeL7@8CdK2M;6ds0|K!ueN?4-2>_`?{oFfetXG5dlK+7{d$&K9Z6X{NR%z4qpRKR0l5)8IM#zPI$X zP5JsGArFbp82G`R{-+)ajm$k6EnOyc7d@RqGRHsVny*!ODRK-#a#=j0keT3VJt=_< z8^Y~AhN?t~=Z8%Q3sLT9QwJ!Y5DWo;KM0e9{v8Jy)I{R4n=jz9@ozTy(I5F%SP6r{ z*d~DBZ)1ir4??wn?XY90H(~5eC9~L?Fy5dG8Z3QaFFS&r|73xF_cm%zZ17|5T}+q?gH-a%6IRd}i7Nlt*U1}3q(Wzf z1eSmAi+s`RJD%Ah^kltM@0;zd5U*Ug3T#q}?m z_%s#azk}&lrfQC~&rk5ji*?bWN~vB<7zwE%GXx9ag}!=Ow6xf4Q>dnK_+#dBu?f=> z8mVzRQsSYiU5V@2w;ebpRhhnr>$S(>q9LFP=YO zaXGPa5&eKu{|8V@0|XQR000O8jDa%E$5Z*qAqaBgQ+R0#kB=v#Sd=v#T0u^){Se{ft&lLd;} zvScx%#mr=jnaN^i$zo<^iLR`1lRtGw=>8&PM)+4={Wt0s=B(0s{Cl zwpNBF<^~@=hz7?igR98@e@4qtkBQ=iK;Y&5zAuRojl}yMg68Kn;XH)cPZ@p~P*jep z!gV#7%Ac^BsItdD=MgzVb$XGj80oWeK;Xj zrXCe$`Y?dFBd+W7@?|R~=1Z>S&yO^`pX}HBc6NeQWo0)$By`@Mf9`fBZzCi`%dVD~s7s z09&Q14-VB4SO)nSf|*Dq2G08ewfzOG#lA3W6}&Pwhz{i=uO9DlY;5dMAS;+ImN8L) zO8%I{3m@5wRZ5XSe~O0XBZAVfkd^|7pUbBC>Gc}}&Rg@gzBsstO6RWx0?6R#kuSR^ zj!KqEFvWb!RNupc{8XSf;rZK^adqQj2w;Pi=W_yDIGKmPjQh_AkHJFt&LeE%%Q^}+ z#7Ijbf40F7paM-N9p3BVD*?aY=hOO8GHDn~%p!aMT`C=0f1&#Ql~bAUj_T@0CIxj& z?y!9n!8-0S$yG;KgRloBhz80J5)q+T;5aD}6)V5yi+oaN1V&A?sGk0XIxjX60@xGP zb)^rZm`vDY8zY&Ob=!PC3;`K`@ZFwKk3adPZ|!-*<^6_61r-P)m8XnI6br4pHLW`z zFF^NHrJffBf3JA@0e|!^QZ}rYV)K|hkIRyD050Le5TSV+wTH5Q6Y{~DkMjD0(Z*9D zjZmtEJ{AizF9=-wbJs^K&<~%Zx>BHQ`A8A!ZkVsTIE17T)A5kLt?-K;*N zDH3=vgui({&U_KLeW-waT0l|{!s!^*DX4&+mo527UxBPbG#i2wyOA|>BiPpTxwUs- z>$eYkm)AX(dsUUhF53+I4XcOj^$v9Y@70N)f9X^iw21je+YBn_q4+VT{O( z&_C#}q4ypgN!C0!d+*nI?(M#P{COAb?d?sp);Ny*;n*&4>FwNUGyswZ=hufcp5fT< zP$<~6%?Ru2KA)XZxamLdc6=V~|I+Z0ZXFzik30y}1|L$wXTqq%f%c<(rvyEG+|O?n ze=sE=M;*cyV8Qt}9 zSMadl%1Tk{FXG+A^X4Te#T~yY*-Id~_smqlXG1-Df?08l*5Rzd0@p8C;WK^4`p~zI z&T*W%R(-U)IWD-LvAn<26VD@aLSlEZ^0VtuX8FnR%?q;%vVOXgL*KxO7M&(he?wl` z(2^iDL^47^N1j8zLy$&V3XTkZAx09*)R(>_6^d!db(e!Qp)o-}06m~A3ccnNQtTD% zRV5K6Q5DLS%azNjlHX|0e1FJ48)7BfsHO-4szMZywI7u!6LNs)qL7Q#Xj z5hIo?qAF07la{qy&?2j-$RvLwe-o~R`fF5CGp9-Urp2`2P=twwiG!(b5tOND5vlgX zlx*g-ur*7rNJ)Xm^cz!Vlb@OZm!ya23qVTQNnx+>YG$uc`PW>rwxX99(G;}_^&+tv zL8*`9PsilhV!qjm$scCcOonC(rr5I$c@yPgbJ%%bijs30RqP6O!A2};e_|5il>lhk z#VTf@vnHke_F$zdA{_F~5^nX5sg9Wu2@#=EKBJ83-_!Nj zahj6K;%P-#)SZe>Jy%kjG6TNA>cJwzs56#W&d=za-2@yx96(-wzq#FC8Gp(h8;74B zo765_1H5&7V5wjTVooxxf9PBf<2$oH+0g2C=@t(>!}FX&V5(sV`Fla=nl>rl7`wUEsuwUa- zESW7BUd#-vwq*?QAVG7Q@$TNYZzx_Wd=dL=^DLYouxf5r0+0(MM?>=MpO zOG=SS(NvgJ9Mq3g?Mk3a%}URfWGvh*^5+$nH0ntfzgbk9&n;R%4*5RHXD?`|K) zoz%=6tdf_?X61wke=~)stB|>nZl1|m`205lzo^Q*Bo8LzO>N1eyNC+FHf|iJeDTQtS4vk}=P&nEe-B(BR(y@?swqxtE^1CqPLCni z88B;MAz@ks>J3u&lnlgndb>|010^S_pB91_+-wzW08AxB(WA1(Ib` zE1R>ON3EhmFWP2;4~=v^sX5mj)>VCq#pk8LI*%3F#nnn8LlN@!3AhE=bllw5+2?|9Vi&(xs7|GIOhY_`_WR;C=I{HYwheZs5yrF3X% z_qFgf@XU7A!&Cc_@2xu`ayW92o5a)g*5wd=GLcE!qMg|T`pMA?-{=SLaJ=D_k|00f#*E@FLas7ffAB*Xim|=l{7{VqU2Hx9;9lEB?z-Aj)ju4sKt*cH zWHNvJNly8}2KNIFmun$h#?1EKcU6KeyN>+A@*=jP^)-~<Lc*W{SoWGuZ2I7fB5v*e>KR54}m5h zK>xlk_2EDNqW=E-Kfj-10zQBPuaJOWX8_25r3NJjeEP5JkKf+2@yQE_i2;A*^=u6c zEbWY~?AMCpOMo{ZtwmMsK72qUdH;SClOsL{W|=clP_b8$lwjAhvY^$`x6(DBb+)kn ztA0*rcHpIjfxQmCvxT{(e;vCs7vWzi*n!vYx9JG+|4L$S#zm+iDT6OyWov-XOiNEo zPsk00kB`r3t8d6ICn)@PcHm!JgvR#v*6eh2PEJm=P7JhGwnlVc+1S|V=)ci@`$hvy zL1X7)X|LlS`$|hs_upl|7yo{jUB<-Oz+6?(!~!TYU>)2H^vs-e|N7)#bE^E$ z^RM*GjDKcT`Tv|%(aykDz{&#n*q-};#_R9Qe?I&#NznZ-QT#{Fbbkw)8;XMa1|890vECW6umf52amzvopl@Js&pFYwD-7AA{W4gCEeCdjAY{P8#)JPk!?F%TL- zfftNn9TMU*C`uea7Y^Wzi!t_rjPg4pG(_PKv`>Z3AR$4QU%vk6L+*wo%z#7{k{~BH zL;&p>n0r(Eb&#B%oUYSsWaOIp_~O{y+|<-`ShTuY=GA(me-xkTrPZ{EfdAnm1n&nB zgr6Tk|ND{`45`qESJwNwkqPwwK2P`n0tOHAFAq?qK6qz$rTzr@-{k=>fAV%R{98@_ z$h?5Ub5g*E|D6GNi7#^w^DjloY=9z=Q?@{o|DO!NO9&LsPybSs0_#W65GX4UsDEh# zusdV_R@DFBe{?5`cfGv*^_$o8Mc~8L7|+Sr;Y=~|X#WM^dpxBuk7BVB)66eLhNIH@ zjFm_|%_3w_sQ3>cLRob%J3fE*Edx?C>HegCZTPJHEn|-ZefQ*-a=E4HbfLG5Y|@B8 zk*??mZ^D4^@A3bro9d&gM2TAMpu%UfV2$~KU%D?he{(EIR1m-dYk(ck-Vs0jv-#xC zAXO>%H*?z1ZE2{gyum0{`S`&56NtQkpa>c^fj>G6Sce?2-AvT>q`0?5h^}9f@zQPQ~xMctF6c; z?+M6~5RwK?p#9V+s6U4Y-wA}ZZ#X@<86+?{LhzwJMK>2v!~pt#W(NQXDN0X*ZGd=N z>Zb{ABZIh8?-Uz%nxTc*j; zUo4st0vt_B%6Jwy@9fJuE~dc$Nk72*X+iX11pLp1ea(`x~4a96zy=-Tc&hjZoH6uV3Zcn6bt;^EkL z+fSJqqtbZj|1(e^B=|bIU4D>;7n{AR@fET;!n-w7QsOEr8ykUm%+-b?gM;7sxM3qW zPFbsu%9;%w+Ub!RBXlra6#F={Pi4H`e_rl(Cq(d=@VuUFl*VKfEC5E;0)cS(@&!`) zzy@Zm^>GEtUQ9FQD{Fhq3@0=6zjE!_&0s<%sa*(xY~#8e>(s4zK8-*0(sZjpqkL1x zVzEqn0CsPLG}T=9Gc2Y)>R|(uZUy&Q+f&&~cP#5fm4SHiKKNKQ{2$Aj`J;!xe=ofx zan|*ppN4_$wxmnua4Kuuir}zqTsE)h{A@U1q2q$3Z;R8N1|_DO$*X{4y0Tu7>G>^3 zIGnOjQHeOi;cRuK+)IVylr7AspeN|7USARzG@75{vhB`Pqn(l6x;J$U}p2MRA~Lw9k_aLYg{` z%QpRSd%=GHHQ~%b1Oi~bs_p2Gaor8iXN^yXgX21P3}^DJFs7D#7;4uxzti$l1acY<47v+bqn)h!sNZWBjiH%I6YC%YwkKI|mcf914JNpXB( ziw{!|$`&}m;JO?jGTa?bzLk-uNnx?FWL~kC&p~|H2JGl+dp;Ool`=lQj-gy@G`BE$ z==m)>fC~H6LjjmUyzI*o>L z_s4?*Pc9OR&jj=LdLwD2=1XT{}c3z0M;a=Rr@C-vnV z`q2Lw=^$8dXxfI3*&RP*Df;HKHn%=kuG@J%!^i7`d_LdLyUTWCf3+s>({Rb8M7!Oa z?W`rH!9IC^>1L3syr$sk_Qb4i$+FpAOd#9|r(1?WGd3AC<>|C}Eor9De92nf?xGi+ zMCIGiq|4*=7ZOj@>Al-Yio#C?6hn^f#9cgEcE)1|b1C{}>?g%9dg;a5?HML>W$BJY z(6Wy9Un{v7Ppx}^j83$PH_hr!2ZBvVRD|LIJIO0W<4Txln z&*i6xQMN@klJp`fpLAYAv|^&+{L_c$qsz$%f* z+g@(fkL>tYV>F{SJ)ST6%UyyS=kKiyDS91pH4{KPN1gN*%nmb_hIsg;@=o1Z(+rGtrZR${{l89 zoUPygf2L_Wq9zV?)`OWR@H($A)^szg{#4Yx8Vu~-*xK6bEsGs+85vQ#PJ{G1t$FOh z%4nJmjps{ZE!7XI0I;QnDP*?0zV_?x#c5cA3x{Fol$)$H)#7lDQTTq3V`T>E%@ztB zwy0mJ6E31L;VY7tyO$f8`2}M^y$8VaxQ^z@f2M^y>@h1-#qMg{%h($UR^L|M`Bhv- zFPX}G6PYt#6y&xm2Olo{yvH-PAinLpIVm3K-$ac9B1Butr}iIQU)q;-{mJ4Cj?!Xj$#?GObAh+KQm+q(4Z=mFd(e(TL13shV;U}|s*EEJv75M# zKy;8w{=oMaUdlqU#=!y&xvY(%<}mn;f7Sn!7g4Hsx7!EAn=q|zP=r>9PR9N9VbAdS z{-VuDfe)JYA`U*t-xHp~5zqbHn@+Buq`(Y;_pr^kwo7~`*8;|7ibxTdHe)He|TJn zmRjCkpCDf$8F2`;hg3)s3|AtzcOGBvcq+aOV4X_y(A+R^JnrX|+?fk88q55p`Dq*3 z&1TI!F-z$(Mon_5B|YyqM0ieUn`^EmKCv)2u1<~p7#RiXIGt6`CyOzg1NWnF)mhk7 zi@GACUb!(l#_82*AP{*#db{^}e|yTjbAPjpG&n8H8`TFR(m7pmTC3IyHh`U~HA7F` z4QE+B?`hJ0Il5c6i;Nr9F|Z&O`E{w7Z6M}A9U2+{iC=jkDI9f);)tLq-M zffA57Yr&<#vG4OvwYst7f4pp;mwTnB;j@O}d8IZUR}GE$fI+|HcCx422$%c0QM3vs z{77OlOE7Y<4^*u?-uq$S_4>6R`@{vE_N1Kz%bPzxO+KGxA-{@{TwF7AUPTneMHJg$=ZWqY5PJpo8Gl$Oy@Z}*^f12Aw%k_YyIV431 zX#izARA+l2n)!FKQHF{Cv&2C+`yP9M&W`fJB_9Bypm)U%15BEX@$(yVH-&>`e|f$MXm6NfrShVfpR5p&OIuL%(BDW z%5G6X#we5-Q0pH*f39O$Ht6s_>U-;!x85=`7`3HDk)nG8As@%cbkP{8M&>C%Ah7i~ z<_D0NT*bzb$PBRS_XL@hIs|&jrr?Z-oGjHxE;y|X<9V3Sl41UBDJZ;wSa5V8G4U1t zyUN_1;V;0MY^f~(FtfE?&Cj<+9ZM!%_oPNCr<8W!)26IHe^TfM1GIr^IQimOhAl@J z0m()NgH$1KwOfhe{g`CQ5o!k(wAkX>I9f0uLbueRQh+2bx4Oo=RkxRD)z*Ppxj?03 ze0>q#x3E>0JsgggPECTVS5nE8hHjsg!Nk6vV5f!4-y!%7x%oDNadwh;q9C;at*cs~d?R88iwc*&Mb$mCQcTQAzcv z*tt-4RpE-)%MHubML!!3?Y%{OQj!5HmB?t%t-+u#qlAf?droQ z$C>MtvFG{Nk?&c8x~0_9#2(py8PzBwkd%xftI?hre}YMLetGn#o1?v_`^R@~2V8;s zq38$Y8g9aCbRswo(~ChtciCv8xIe`172r?siAR&R;JV84&x#U4KBa8#-{e7taWZ&G8o;y{En%q|a!QdKwJKXkjq zQu|7EEmtmI5HUhFg#r)BA?fcfx* z|2L9{3hw~Y1cTktbTg$gjpRCGOltXl(>YBinw@SjapS`$#hijpzWD9oX_rKy;soC@ zBWs*)^3#VNWa*SZT-SLh&L#7y4j3%QV93WzwL0@;m*1%%M8axl)9U2Y;zuVgoayAb ze;!~siN|h`=Hy~$w{scwjr)nHA-Ai|-6_`-=Gpr)%`v-^m3=;4vQ(`j{(1++-JzX6 zgl-8!z=g3biDQP|A*fM)JFZ&aAN5W;)2^=>z5 zh_1a0Jj<&PU7K^d#(1JDaw>uw>N5;Re-ggF{mCP|-DWV^<3S7<6b&MW{=kp&aJkfA zLmF=M}NB-gFz zLCJ))0xyR@rOW`f{a{?bha8O>ep9f*AT28rUC|HZNG_0bFnoAw?lYO?!l=OWG{flJY{-J`|jw9&-VRgxeYHNdZ2 zl4WH&(g^&&sBH+@I(vX@oc{Sbf#JklXbF=qC^&4zal;neYw5P2{pEIGf2&@yq_I4q zejKrinPesDCoCG1F47yvU+hOvSjx72&A+y7&+1Em=g7Et4KnSe}z~FjBWAJ!{37o}tH72t=EGoF~ z?@Zr@^Gat=OK3(m3p#-~EeF`mNscFrXyJKM6>!IY`F41Be+U33pWb=vtG4G4 z9F-;$@ZX?}y5*?VjMlSN2Agq0uq7{d?QgZOiPOOn=*uXPzHsd1?|l*>wa^ttzUv8@Abn%5D?v*Qjlc zav)GhXk=E`-lx&te|VO_2}6EMAS{;BYIfPxjhgr6?UGgfoEfhQHwt7CBuYGQ7Nv2~ zFp!R`FxV)QIFvB05H&3Rq(8g zM1fSgT()M-zM1rhrdcJxs{}};B5i!st9ZYfOlm@2vBWaae~1I3yn@uNTA8+xaymKf zP|58}chjU&uDv?Vzn>flNQ^SL-Ez{D@q_;DRf7G<=9$54QL6YOMIUkkNN&#y z>`haOge7*U8J5NyZ1j$d8J4bhB!QNFXQZG?3r{F;vC^LCZGk#6_^(;xHYLfj@R2S3 zV7@XZ8Y1b8e>+D>b{Z-!#j-SNWU`l8hvY@k9>jl9ESy>#iVe$#j|z4_v&;3UbT4~8 zA0OXp@HA4O;0F8`;)x{l0 zLaVo}T!qc??I=;K9s2Q8G%G=Mk1os+q09hTSGtpwsOn0V0NrK6q6Wu_!TCD0TQeH3Xp;80h#R{9=H=Ds-Qf-f5v!K>f7s`v6FEZ3i;#p!5-W1=ZP_O zDkEby1`7i&qRPW}uSEQJjcf;0^5vTpscd*!8l7Jx!fPIPtv|&K#VA!}?XM+xAph1q z{DkyBIzBw{h`)|5)|w?&8b@+o1|_rEmHTXGg~rpYLce(xsNcBa2Enw ze{%+l!1z6paoTzAD-6@r_c&D)9MixtE(R4DR>GsY83jP|Y17bfGK0~|H;4;L4>LTs zW1EcX`4xd-CQnt=Qr@imC`=~HWedZJ?Aoi{8-AXb>nVnm&i?W*LuJwIL44pjm_>88 zDz#>Fv`?ELlF`&?FnFH9^`p0=ka;sze+gun8sB0zq<{c2Jue>>NUrf#?d$^XBuX`f zsj<8XXSm+$8-v=^a^jc*@Ojw1+|1=Mk~MEdK<+LWy=@gpIu?FO;H|XYNZftCqB}pU zo)E5WEy8`Hy>W`6PKr|ooLe;*ir7h7HR$9%08>26yPeb)u#D`Oza`=PJ<$jdf2?tO zkfm5V2S5nroq=r_0ou{ymock#$QKfM7EO(pN7!MPz0&m2WJ)|qowmzCs*|ENkEfKA zx+Pl0s_`uZ^1(n9?`-$yHpU6b ze>(#5wUN}_Z{zJ`9RD?|@PG2kLX^tP8w6DeoMiA&lS4gy0oZRKle>EJi*CO-0r{=) z)q`fCa(TwVO0(01Q6dQhs-z5Q9Iq4Whjkm!&bw5DUVs6GeK@@5gAVz4-_2 z$1;qFCCc|FcOdM}JHgeJF2xCpBLS>gK0Aam#&Bh`oZ2nU!P*6Ie>k_7)GDtAhtoxR zTl&Ah+HQ+-#P^UKF5cidlHfWGqy6png#gt6M}!V>u!2=RA5<;OM>)b`v?RD6L33PL zz6#_;WLCYTvInU|AO}^&jR6jw6vx~O-&|9FtLqJ=8Xj~R?om}d>&5HS8JP;)K#S;5 zl_^#ry~!l;*yTo1f1D!Qc9d@`1s2yyV?l=NV5(c)?1laybjv_YUeR)lVxQd)`9*M{ zy1O{k-VSeH&I@X{u@#S}+pkY&?b`eoT}b3{MYBY2*Hg8{^$l3R4-&EObD*>t3E%}uX!fo#~A;b7-3mUVln!W2POPJf0kyBPY@vbsW5DxLLRqz z6!J!rk_S6ci=A@=OF`-^)9V$?D_Pcxy8lV|ysf{*UD>U@rB|SK??s_<4epjhvOh$w z#L4?|R;g3CS@(%>u&PWFW#01;?DEzkK5vnx^}N(OFV54)-ihMui-xTG(~ z_A;noJ7~1qf0o}oSl|x5rM;@@ z;L73NwjHJMvYCXxE}bgZ)_w=hJo0SKbM?unge{Cn7{8p`Yihd$6on28iOC~l1n8ua z%h(Dee-+cOVq7YB>FBZtQ_PtB85ejKQjYl-KskP8>0=D^aC!0d7kf`ZHp|j6{1SpJ zYkzfh8)LRuOxlX%;of-N;R=!iM8R`=L>?vE66`lP0$ zn?aTD$mvS6dZ;}BT|lD02?n#dN*5`6&77hUCp@qeFr)PcTtvv+45-Vzpd)>F?R zKv2Yf2(OT{+zBp5g(8}3P5Ne{(nF|r<8{`0SHhNF9F=b;nih2y+yKU=Ijs;}zm{<# znW&O~J| zR)fC?<*&9r5ld7i(d)ST5GlmJ+%EO2i|4GzWW!=vniZ%{-w9)la@v)450XmOF#ri) zJXOSw`^^lARv+p@n+Jz9G5Z)$#TbsH7~p!?hkv9GMz_2+mr$tH80!GZR|0^(*=*tO zO+hH{pGWw*e3h1^Su+zg?!TbXwcVp9LfmLMudW$(y}|LQ5s6IAtUDq0A zC@#59kHpQ|NYszqW*~_I#qZWUT;LZF&ayi2+Fe4fjZayRVxWp6)(~o1q`D%)Ez|b) z3SKU0nq*OLv7hYJX9Z{4wCC&mAHS~_h!kMlxKcQr!Jk4tKj{mDAu%<>q0)tnh<^}H zf&@WBhzO8HN&vwa0}1>|+7U)Tf(|~>LqM!z-r;PeyvS6ocw)ywjAnW)UE{WlF3koc z(kh@nj3mal2g*n`>#S`ZqE?*h=%BEGNrXZs44Z9GyQo61KVmmZ08Wfb4wPe%b8)$pIIN+7NP7LIdjq(xQxQbEiM`~>L0O2$G9;H0Xe{UKhuzxgaTFDiQ z4FmD-xkuM>?u6q98bLTm5;j!B#Tt_gW(GxJIp6>s z3Ufl*$e`>aGwg9=n}3NHI6SVoLIl098(hwMFfXE^_zqe5JQNDt8W6zi=v&evKgLS2 zQ|#c?iem+ybooPvzy20uQh!QEB-zta;El|xbcP$Z#m6pZcYQ#i;(jW67R2vwtFX^fssErI|QR z8*<=Af)I%~#W&7Q+B{7jgGz2|;~R3Ec*nmN%H{~?d%eASx;lPG4r=9lH01SE!#IoSRjy{lRm+L)1n$1^aMRyR8 zxjsOZ=_Qpv^`3Cc+J6`NhB2bv?Idb515JE{M?lPrE7+89Wx0|1RCMihkvnaJl*{~` z?U>FqXE!a&gOI=nZ$iwnZKyxP6?{n7*#p2d2S8}CaSp99`?&(oVFHaR_k4I|k7-07 z2>yt&sDK02=(on9-9d`{08ESf=?=S#0Xw5N0~$oAvL1iukAE%ypB?N#INR;e5GT$F zVqe(p0<&xfF=p+I=0*#F(!1&thxcbV$q_HFxM;pwwjEB@FZ;v+9)Xj^S}M)foI-jK zArysB7HQCsO$(0iaDTSjf%F`Rk1^PEiMOSAsdh|)rT8Bp@nzyNb<%8k0YUO+a_E0X zS?G)L7W0E9@_!Crev5Gdoz?d8XnusEAC&GWjmbe3fI#2{tbh4Oj137fAWvGkL`4?T zOzV|$?w8__0+1Ize~JF|9_j)VZyaE+1|m`c+0OesNTkryrqQ9*4Ho$Z^-6s3 z&Q4N~tP=Rg?e|OF|2NnFo9q9}xd4MLE>}wZxwHzO{!e@yutX9WgM}&s2o#}3XB4de zlK=r7L{Z6&fHpo0_g_NgP{@EwW7C}FJM8}nm46FD1wK+?B8&I_|4}a98qjwqvYW5# z&7qJRku6Y7Q{n*AeV^M_!1#w6mipkWyRu+I0@QHHA3%f*xzqgrQ9B4wRe;dX18?#J z6&50;)HmNh>uCY1yq;!{A{4BME!!S;;|_xS{*-JRq#uC{wR+t!=ks}YUWit$sbcknFjjJt!_lJDC7QaqEPthB zmPnzc)0#(pkz#Rz$;_{Kla(fiWTDr?;__jy*DJhg-CiiW{j9*0e96=V7fBbSWgsrx z1yXhvvNUU-cSpIok~HUqpUY8cy+k~v;xHJx#t+lkl7qF)yEQMCyA`KQRC0~&-(;EQ zm`tYn-!%6o11ewEec^|(Ib1*1m@iD?9_Dd+JlZ>j(EK^ODC2lBKzbv2UX3jqcM`rGRRY;lYMEv;seQsw z3%R7vUWVIfLF1E@}uv5#%M*~E}qwm$bS;SaEXU_ z2ikaEAfL8MdUs=(*DV<4TdlPT>qbM{?G9J^#B2hG$jCqVeZwD?dA>uxX;c%h56znp zCd5Az97G6riCh&B#r{067n%Hcb3_&g^1qEfm>|BnceZ-HrD5yMY&BEdP3L>$p_k|V z1~`^-ffUYEu`=D#7h3p$K!04FIhHQ;3sJQp$b`%Ak+aE-}Hluv|3#$6vlYoT9Uu&1%xL` z`{hJ(Hz{DEXIZI$NlJ^N+oF=cJ|2~qwE4%LB~r^AH*7`h-fu-pOYP#dUk*~evx!$5 zRS+R&hZ_d<>R63L1%EV@-*cdmOoi-jx8~JALX-3fo|E30^5|vzTbzTAEf*sBAR$`h z6$FZldA|aPzopmhf>A2B$KyQ-g*d?ZVJGpgou=(6ey@F~lr7TAtzX4u-%_pG8Z#3e zulW-Rscs&;!Exz;;@Rkw4DYJM*+| zQ>sa;TizoysZ}a?e~-+nTUjjDTUm|F6DR!{wZ18WuMOfq!1a2+wR=*(R)=r{OH~j? zyZ@w={^WACn+%+mqrz6edF%M$^yeoy`k7yf)i&>eZpln$2@bo3XXcXPQ8_~XQ_?&u z(}l9#r$>pO4}TeFHs3e0Lec4%VKEsils+||)TQ2pZ;%aF#8S(MGP3QZ*&z*{(khd= zU++%{`v%hHB9SXt&Im|R8t7yLD9|v?74Y!mJBW%uYrb#UFS!2c6E ztvD{J?xj(R{{q~`<%O0_aMdrGQdI-{xA+RfO)iHRs*L@^co&FGbZ9R1z&+Gqy0s|F?c&&nJ)pU%643wk}b_~(L~r~h{(z|%9#+c7p(R8 znm>iZsedf=`~0Cn(c?fT*~dS$IS=SH?88PN-r}hCFE}%mynzDVi=*=P6tl9%KD1W( z0;#q$2-wLLn|>VOgchi(M@6Hx@TMV`Eitlm{elS5o^jr^RQ{}Msg@`PQHLhfw&zRn zB^#%MxB@JgIBixBI&hlE6CnyuHgA4kkH_0tmVavYc_hggU-pgnpdeJ9Yn*aL$`~-H zcs}1S_JZ{iK@H8F?aR!!XCt55z_x>TmuhBYnau&oW>B%x8U*`qdrwm0!RzCNsa5gi zp0Tg8xq1bBQV;W@4(XEzo=7MfnFK1I)1}J_Hq+w$Zc%+K0Ine)rM`*J_uQ}m#i{=8 zAAckv*;ubzvaBWNJ^+QH2y(4F$Sc6#;6;O$lOvNz`hiBPVZ0Rxi)E#vY`_>Y6;zTv6u0M`x5COX8dMSodShf$y@=!wfY)QKa`1o0oFD%SLP8%}M@SkH~{bO7NtZg!68)(KBTniA;kZZMy=$IJ-Ch zloGL}C^g6B01*ujBvI9%OpA&!2?C3x2HGM}3pUp_$>7Xq zLpb4v77BboQCe%#!8hMt@7f^~u$d9t|Ije}BtLqCpH7*scv7vvbEbbQ^?z%mkC9VT zT++xGv&jJ`tOwgbVoLrG^QzyZdDx>vw!CYNztaySK>XQ<4`qt4iaS~MYUE+_+`@4< zX6urvY4yKEQ-Fd&Hkpp!DXSMK06MQG|&tursDt zm_?#Y9V|Sb6sP-ggQxFbY9Q!4OUd*0T>XW`GEFl8)w^pm?(;R zo-c>!>WGohwiQ`Z4TNt3DQlJKSX$G{8aAO)`?cIo3}`4dX3}`G(0>zmqrl$gYbBz} zl&uNi`}7Ibq1>V=@{xyD=y>ZbAF763eY)&wT9(fM`5?nJ^c@s{Du_8NW56A2GO-N_e1W<^Jf~`Yjb@ha*aUM zoLjfER;MBnF;7C`Ab(?esht6o7_g8zEw>T14+Gjg?XNdmDWG9y9^M15b^S_;4Qv3nvOEI1Lch$EN>p^-J>=>Ug`iNX0t1Lf zGOQO=rzgg|@}JT>4~ShOnEds;#!~d#Uv8mE>z^JJeiEKO8@PUX|InZRz@8>P7B*T*E)XmCD;|1whx#4h0A0ewuEf@lhLNFwt z4j^ECjgkK@Y4_WH#9>a;sGwR9xU?x)iw3d6phL(vvws{V3~aK@7)OMZNFh-8d@hzm z00Vld`N-r7e$}JwLw8sBWU*525Kt(t3kUE@b{iT|`}hpxt$>?L`5-w!!^3WBDdyoR z9;3W`+HTix3CY1Ktr*u0q(7El7YV)%Khda~1%Co^RU8ByW8)PNt``}JP4uK(29n5a z%LAh3vww>~7C(#3GI8+|orfroE5Li84;7@?L1c#VZm=IeMmTUI`q1QC*?Hs)6?t4T zhW=nbA%jKybO zSo2#{6VIP5*Ir{EP5k6h>+s8rv5-x-2O?Ec21a!6SPs(8A*Kjbwq|fJ0L#+&GaNPs zwe8}ISS+ck8NhzZe4AW;$^e0Uk}#|n@_${W=#hH!_eO~<$xE^IYJ-T{@Oksw4FK^n zA6@D?c3m`)$YNr~s1St=kTSu)wr~~=QAXOu6uAxc`@cr-8x*06q$TR46a7IU<1&9z zlkJW?4OFYdM+PH*3xVT{=KrLw7E#qg00e*enBqYbP6p%3-ud4P<;wsQcchXj#D5CE z+43aFWClR^;mCaXs3uI*YA<6lgY^k`fdX#RcI6TxRUWQZ@fy3lm%t75KstZ*pFI?nGxGpI#4K`wq`D~ z7Ih1KE31|}1Bz%Umyy|Ziq^%R`+q7C-xbNw704c%rl##dp+rbB)Zf{GnJS2X2;mfI zQ7$de2GGcEYxRALpOh0yQO`#=Jx?JAA{+6ppeq9caX87Xad8ow$3OsjgwOQR`lJ+d zjP@TSXNM4gO2U}O7lB$`86;zL0Yul=!+VKOqoYIeKY`^u;EzLLr^pdN>wm;+I>g=h zr849D+n*DUxz;Qb2sycdwI;}G|ZRW76IsP*% zKTZ@zW}FF0TLJIkFn=m3_V|pz+?${3m+^1qIT6_}(@L8;RnsM^Q)=G_ObVeHj7AyX z4>fTqU=`BDYi)+)mk-2%d3Q~O>;c(tD@Cu0((CBv8m`wGtN7&CDE~&1~q-?IU z-(5xmjx?e~vwt%Eq0!ZjV0(wugw}?uoJBJpH_D(bHCIgM$}AZDi7=)gl*lwj37EnO zFmT<^`52r|HN2Q2*4cs~kqq=dZYZtCc1#YazX;6YHYwYV-!&aliTsL)o*YWfr^c}7 zX0;IX`1sHgi;-*cr4yjYr=Og?!!r{J86YtiT)imvf`3&T>Vv^-Hak2pCphjDG;|fO zto0R1yawW@4`{B(-OAjpfr2F@DpCT(1@TW_+1!&~iE&z#NU*uwO8QZhlw_GHbP0z$ zD$xd8e|ph2kRSvA5Y>W@aPjwm&IWT((7bp>pUKaqa6Sex{@#8cB&7@YtxkS4)|~aA z8i_@_bALame2V`S+BRj-DI?3bli(i;il-_PhdR9`%R%8Wjwhr#XVcaIi7$iV^d$R_ zgQ?r`yjhSaE~T6-z8^pd-yMDwTHgZy;qZgNm+2|yDuM-t3I?tLg4)jp|8P8dBl7~t z&p&57BarL9`~H@lQE>ir$vu$% zOMl)sfWkxkfbEEQKlTeMZ2NytQG<4o0gK?4xAptS-TzSF_v6KEAVh5ca$@`6P6+=O zl>0a5=ihBQ2e!#?EDrlWj?XFT+2ao%K3KlL3*aA#2cm)fj_44D``^_9`*Q!jU1an< z;$NKL+oFI$X`YuF_I|T}%lb}n&fkC)zJI;#Epssa*INiq(!l}5LbU-$0kFi7NXlKb zvQ^oj$b@`VNeBos#6%&83cS4^bor;rK1bn)2u6qDrEc zc5rhubj>%inty>mPS1##(Y~`+esLHriU`E>bfzB=j z|MGzV+8^YgpAi4|6UP1#5+%nIQsKkD=7dCP0{_=Kh0}k6hRC*qQvO?^K|evERz4Bp zgZ$r5jubCQZ=p5_>c4h`7sUVc-`bDhElY;)tvgNU4fSt@65@OF-~DStItr1ac)uPd z@`C-VT%vfh-2Ppn(~zP7j2m$P$bY{|O%yQB^{)+~KLp)%@z%Arrnk4Z7aaH1q4(>A zd$Pnkc-D8FE)V+G)?X2LyL<;&Pi*+VwIR^ui6Ii$@sXVlHRZH3Y*ejK53L_pioK zV7fwHUS9hrNAO(Lh5?D%u58rpU6&KFx<7v4+dC3k{QDdg_IWQ4pPrZx2eLnc{>P>u z%w0!l7eAM)S$fsHTT8Ve;G;E z_ww-F^^Dge23FMgc9WBx&!hIQCaXCm;-&XBnJ(Q^BO0f(ifEyWZr;rSW!w^

vDo~`e_n`d;g_V#*b-H+qc!ztE$TqVAL zct|NGk{bF~=cb8xAy^%t7NO+}q+rGATJ2kBYN-jv=jPCxnww+Ua#+MPKgiF&?rZug zMRH0*p>*pEE0Jx+y;Q=W#D}3kpxj1+Pe>q>BT#C2+$cidC4cE$ebz1P|5aRulvtl5 z_a1E?80q!ARq4zKfk>r~3L?#UA&6q}dQzg7vtfd$DSakmF&OrTvTN%aL}VKe`s{vggyu|LJcq zSq4Vi+sl~v1Y`(`NNN zfCYwQ)ejQ63yP`)K}pRd?>uF$#c=-HTi5Lyu-MIuwa3}|fV>2?L3F0)mb+EA?(2go zJ+j9m?YGB@-4rwCWkW4>9c~2j2K{OwsP>-%z7c)&2q7-3dLFI?^k zD?FU%c>`XgHbsxF?S6B@ZBz5iRe9dB`9wH}^{V$PiA-4Q5VdzU*POm3m)Vkh)9x*y zcQ%`rnmSM;J|nO`?&qi@k?+heI0P=lhp4FTCE*vjTv^85f`^UK#2F4M$Sqo5e>R3ep{+ zic9O%PA~0u$2@PV$s2O7F7pv7-Vep?x2Dp*{g(DbdiZAL3(E|_aoV8Q)UeAa;dPyZ z2Y+_BkFnXOq;8>T_u}Hv=}suqAox|iFi1Wjhj8~7Z%4%&RgUZRLcvS*1lJ?Y7ce%; zh6&lhLCiAY>vL{kcXT4TZjIH0z*|ij&%dIq-L>p-8HBQDFXd!hZRO3qj97Wvv7XdS zeJ?I0as}DC{((f^x9T6uJCR}btv;jngnvTQ&MlgE0-LUs2B>;HA)asR+r!g+OoEF- zd0y|gY-R~VsgV*iY=7I^q~*s(X6!T?T>0|ykx>xr*+b#l4iL~}wH#Lsb(PkyV%C#L zWUzD2g+^I))n>ZvCO4P|lO#Z7-uIB{7V6&ewv=D-(>K7EH68x4o|0sNUCK@1?SJGF zFru-53>!~VRMupwBYVAw^b(8&#eAFp2#$@Apy{~4s(B<$DN}*RG6pS|Mp<*%l|h71d+*cNg^RXBgDwTakuMD5 z#|B3fzpM})qUaCQyj(B9`=%mPW`EaP;Zs-M>t5sod8uyrOT_B&1+C2ay0vsLo##cU zPI>DsCRftcMD+W-3Z?PZJ=Tn!pp6C%nP>xHZvz@R39p2Dc5<&ryQ05tKg2GPQ5EmYdA(8>!0v~8h^X9t$!XuC^&PO zw_Nmfmoa1^Zi{vcV5^%~_L`sjfOW}BbHyL=@wuG@@;vLH2&;Hqjj+I@k0NvxQoqDx(GseXhfsQapbjCmSx}s{M4mJ)`?^( zpgfwxFS*{HuWa62veEht*(?oRhOP$XcU?O|>`39Pk_{4qo2`Ta_z`En$uh$ra~klD zzNRjPo1coafy zp>-SUmgFd7ibO7lxn;koX>nYnfJhK9x)t4zSA7P7G>mtdRw(uKd(j>FtIBsFj|Bs2 zJ-RjUkyszaV}JPWsD!Vcg;uUpsk%gGSDA0_5`+VqQgL72-AzzqwK|) zbG_eimB^1XU#|=Fd}rqN7={wiXMrP*jb%tp_KK)juQ+s?nTTc3BLq$Kb-b+k6tiKq zq5Xn4kOt&)Uu)g0nwh_q$B{adfajn_ne-I!yKVxj-k7UH=nceirp$)e%B;Xjk4cpC52{EPZ=~@u}HL*Q!e}(DNELy{OS+ zK-{Iw@D;6{%1@J0F>EkqClVmD{cyUM0);~CMN+ik6I)WY!iSW#$8Ari(8LPD8rC^*BCia#jk0SUmzSC{EPGt%iS5sM zi+^TgO)cWezyphQ zX?Mqkji-b1KbVV)vYn}o-s#~z&13JFGWy;TZ-NETcXG_enh?s2f6f}&f~36gEF_VU z@dC7kehpayA`P5^YOnKI0PIiMHP^ML)qh0IyvG%=09Oz&DtWqd`ZxvsixZO^XF?Uu zo}(Fq2E4}d23$`MVi^Ux-$0G7`xsebi-tPUSrgP`n3`oWAB6i!wk~Mu<+%78iU|s{ zMB)Bn27*=6Mt@=Rkq-11_E-&Q43-NZ4nr?6DQTiYdLU|8c$nDxg6bJF@?$T<&3}^h za)9VcDfG&@!Bq~;4<7TDyoE97;cIgq2-qsj`x}0kzJ!;@dR`)Q#ut}dM4eOrc6l0P zP7w2arEJ2c8jv{n4tnh!9WW<; zCNbdp6~#B=axUMkdA;fErJiElF@H->rx=X5UhC#KPBljvwG51LUJcK)eotgE)h{?g zOW)my2*pWGzJl?5u3H}qrR4|_q1y9l2&`bRSga{r-ort!gv*K?M4&^pTCTU{yw#4> zEWYD+jf@$9^o}u-&G6V4z=Hxpv$;K$>B9#Uk5g;B&B>$ok?S1^ox=SIaerC+`3=Ic zMMKr~!Qk;A0`gr&^FmG+_-LLj`PUK7wJq;sp!v;hSJghvR5*6dpDm7hYI;O@C$yE{ z0foHY=nE7DX-ICBX6fP?#XLuZ#^asq;T^}foFc7l(WvAI<1VwV_k51(ZCf`IFF%_; z26VhNuf5Pp#pX_6t=RO9<9~4q9PD=>q_%F^8rz(TP#L@jDw)n6k>8Q=Jp5kP%VYRr z6X}F3OcqVHKzs+9;o)4jnS6i~aZq=_IOlyOEF)?cE_0XN>Z9zV&gJbh!HmK5DaRwG z?vuM4_mo?)>?tfd1|}p*5o-(-LK_z@P8xhhioM44A&QW3cKA;x5`Ruyx8ur&aL|xz zsE8>ObaDx+LSrO5DHtJp1@!F<6BO#>u^$Q*%N#o6p2nJXng9SV=Q#IO@)&3UVXv{w zIX>WW6h1_lnJZMGtLXXl#L5dg5ZBdMBN3{`$SrT8&ozk!WfT3$mkgX%K0`l9>pL#% zb@rw8MwSm@h;`SeP=7r9Q~8XBAl!L1S_RKA;ml_h*|IO+_tt&kBZ{YF3^DM-z+;=; zvhH0=uz0L7Kc;=!=`?_@{48*@o42);7yc>XYCJbaS}!52t05}UCUW#N6Z}gC7-!Tz z*FQ^FsOtHm7QMi$58Z-cObALb$e>(^gCdJW4xF?3gyu!{bsjQ-7NDF;4F6( zb9mB$K!i+ADtGB&2o|C==ARhaym)~%91+o7yJkNnxy9g}TThEt=gK*#9jck^M&K4{ zi(N)D2C214{(lDJGF;o|tTgWIL}kYDt$D%ej=?Z`B-x zib(FxBcv)LwV74Kd9MBlMBk^)g`S&;!v4u?kOuwgghTE7S1V3H1v$Ls(#sGwNX2D= z4ekvte2j&N9`WrjTOJ8IDXd?Ljl%h4JUr#yb(P=*k$=wXOh1kxsL^qRGz10GGUZ%$ zDPg4UcQ&J}%{2>O6e+hRDfC>z8_g_wp+!{wVa#B4iZ~$mG?2j7dEgs=*InCPYtm3e z1&Fjco-+1h>}}7~XMp$WfEoq#abY__(z-Xl5we3|*rQhT*5qkNC{~UOnQ&tZl-_iNnD8SYvV-Hi~x0f#I-4=k+k;Qrzo8OV5ZzPTP7tIWG(L zsbz}mc)ngF%!fjY06Uf$ zWKaNppQJd^N~tJZNKMwyMX~$9yW2<4w#)c(!9@KwJjZ#R83YvV*TUZ7E%sjc09;Ae zqKC@(T46SF1PNmZU4<4lFU-dx9EWMyN|sEQAjVV7L}obaqAh=6fdb8h)6785Ipf3W zqJNo{t=GM9E|m*@YuTXqA7Xle!tmdDT6-)F%D{b4VV$UVHlB7D5zD&8p@yJPuCv|H zHdNq^@F(|z_q8zax)3o=wtsQ9)gzuhR1V2MtC zz;#~1NQXBjASeo>L+y}y<4~0;lu>IwMGY-TUdA*g)s<0&%65uQS)`DFNLdl*MKyM( z09bE;==ICb?X&aM`AzR?#Cc4vDKhn!OA8kQ1(CdPP3Eo_VacQ?Pk0Y~OzO7RE`OFC z4puab9TXQ9BcnD5zi6UUfK198jY&8%tE-DT4fImtZ$9r2e*Khn$qtm8oian~zAR!369`9;C_$3W z`gS7-LaZ5qq0WseeYGNjwmGzMT7Nv5V+GJKH=t1LA6?#TGi7;t%@#)LMF9BbK}X_W zOkMiZOmzmA-7<=OoS#C(jFa06|;9j_a_UpMJ$e`Z(lMH%lb3 zn%JhLS|ZENzXmv}XezJ??-0yUml4Jr#tGy8QWYR^@JGHYVMG_I1_9%_tbd6F3>wS9MSp^vm#qA=#WMq=OCr%OEufH8UHi4dtenw)<6W&VLsJ(|hSGT6JYf(EMT!ud28Vp6tpyhT>e#qqu2mc@}6oN}q zGW?JWB2>Buhx_4_&2Ln^MME5pOhE zoDIyX_%rT7BavU^_oTlZnSDgwgdkuJA6sM&YA)f5jvFsMZ||5LMDG&vaPODx#(Fv7 zaZS@#ZU?E#VOUn3_0rN&$*YbG+ub7BuMRYfZExlA6ROrF;~;aet5f5{x0MX^z|`d^0sG zZuQH4MR*FRg06{od%ZSV*UNNGaG@d=!g#&a5#x`K*~UA_j(_55YL#baV!>mNdGif7 zkRe^~l)5t`k?()G#-pW6GcV7>i|vE`YTQz>AKyFK$CsfTWVLi?)a0ene2D}$@l%8~ z>my$dsTpIN34ddL`1-fGPbA4^!K^+Z5~mTIIm(L+a0MV>{ktT-6Z4%oG-EOuY;^oK zlf)^tmY`oR!-KyXi8`k;ThQJ|6!Wg}xkl~=8<+Uzt#Yq+(7JKc%1pbST5AZ$dXu1Hhkv;xo--`#A)Zu?{c4c&T(Vhj%0l40>HseKHguA?zUxd^J%8az*Dz!+X!FwgMBu1p+(F`~0e@>xa_M1>zuql$QFH9`%~!HlcqZM| z7@2|4j&Y}RhgoBOpq4WcwipSO+~Ny;qE`|IbTF+}kB|z5elLKAZ?Jch{?PjLWG7=DH7mxJh*53b7r5ssV2$r++(f=7}1p783mNA`w~ly>keY16*X-a?ry< z=r>E%2is>E3RkbVfZ+MZi`Lc46ImAC_EE>BZT`LM-^g165X~FyigoT;h*~%F)N`ak zD~z&&HsNi=l|Uwg|3fA*e?aHp!lM}*nuce%8 z>WJhG+E-oo&)C-#?>km{@w^NmU_`ka4v9uU9^twab3AeNs1FuWa`(aFGpE~cicTJl zx0Naop(btkD^Ve2&ZL74I)77xK{?<;v+Qu|E6w2D$bz-b=8g|YB0TNzbJZuJB9y6s zKZM&x*kZ@WmnF;f4Fx@NW|q_LeM*!S;CxeDt($|P1mEHv-*ryuwF2IZP7JI@Ei}|YSP{p3&{wkQZgY1L? z$&r%NspYiY`ACYPnSY(#E~b9lOdSGT8?GzH3sqmCc_n2_o$P_v5f83}A;F)^MDqfa zwDkxXhnL^xx?40p)4bB2Rt)Z&}DXS9(; zHTC4Dq(pVH&t1+-PfcJ=vwK%-ZFG2GVL;8nVle>=;tulao26y4||XfGf1h#0zw z+F!OE_9OBVIDgHB%d=4<{w{evB)sn~4rb!s%omSOha+v7&8JONqq_^ntC6tvOwlb( zG!7qF26y}G&Zqc>sJ@CcckqW5FBz_ZANTWr+_Lq;J1@&gJ1$RV7G9`ID&XTR3y;|= zP`;0K^{-~i$~kr1wNcY#@nE^g$oK5`q&{$7@u-M$g?~IsDfrRtU{=bT-pB5`=~;DC zf6Pp(-*Z!fOxavG?zFb*7k~IT>w+{i<{|^$Po4Rl6YfsPbQcwO&(3Q&k{KQvrJCe( z=A|^PJv-N)|ClN}V{vkzh+{j@g1>eQTB<7mDu}}tB)6Q(TV|2SRemoI>I%(waoFB< zO4t#VDSsv4ee*8B0h$w0tjSu+ea1}<2temq0`sV8<&A^-19g=boicDm8Bo}94TsZc zkwGsNxu~rc4nm=r|KOUjdGfejDA>KNMcZboGAmWr=SD@a0wkO1pOQkLGY+fJp^~dK z4mVbCwY!MBqug0=QC5Oul24g=(P5x{DP8lQ|6l!$HwLoucgmc(5BY&qPCUaE{K*dD$Ar7h2moscD-c%<& zNge+imC*;;J&6&viZHJmcKn613|<{Aj7oXmBSW^Q$!?fbX7~|j>IfsPj_w45_J)pe zDQ_P=d=+c=LuLpK&sX~o1-tHh%HG+ZFn=Zz^>OdJi-r$O8Bq2R$LP{;k?Rn-sJFX` z*e>TPL?g3it|B*{e!Z#ooXk^ROYQ+gS3;c@3=%Q0GHJvRmCYhX&ZAYdx%d-}fUjC7O{keKv?)?;wyk`JgL-%6J9B@&?SHO; zgq2VVm+14Oh+>2g7vQ{?(v@V@JMF`5hk4QA2tf1b(b%`P=`LNNp_AxLti2T5HPfK3 zq>#_=vCZoayG$kK`?W_VCcrtL|31ows6njTD0&%NXrfi4YTaxJY+5vq2B-elxH}pR zry6I|qN{>%m}c|++Y$}fO<0D1jej#iBk}Jn=lm#mAN9}$8Jm9xOSuhzZhWH1~^KU?$hVE2*kx?hJSK?o^6O| zmJr~SKq(|0BixQk;cts=lJF$0mXUn{`vHHLk6r7weRe4eHYQCp}wr%mSe zaOK4QJ&vxbADtZ-ag>(@*^=0rk#I_k1MC0nFuNhEDPaum!u?~0eAy@b(BC;~Jk<&R z1_Y_7JLb;7w1x!Z#eWMiMt@+Hs(df^=%FUicsqP^<7oQn`eX_LepWiRUJfpRk|dgy zC<*cN7IS-`^ROKVn4C{2tYr~034T=EpBXPC!kQ)c@%un_^ZP>>QSK*>n}3eVzA}g6K@!QG zpKIy{ljtr_9~5Nb_U{{b5#tgb=y8DZW3^TDMl#z=0jQ~ITY1|(b=^(jS zC5%mPg)scntl3P+al4{75EbxBWOs@Q43?A7$)AD?W2}P>ilSJmdUv_n6X}gco`@+7 zas}}6>Lti3V%&TbkAD@pY*ivkH7LdW&D~grF_)R8Di+o|6yJiJPp|9*q*8-vp=T%r zq^T#VxqcZ6lXF|u!xyHNRCaW^cZG?uYw)Pm28o)?q>qGX@gb-I7`z}F^?!K|^jZlO zJN>TqNJgO3zk~jkZM;=rc7eNC(O%EcNU&8SpQ`9717dwo&wr7MbBlZ8saujn4Y_oK zRk}uj$NXn?m$m{na#jqGQ7yL~Lph>`c8=bYM*vtsR>k+qP}nwrz8FY}>ZgQ77rxwmTg=>DZ_Lcl_TO=VIQ> zRb!8;yV|R&o*5aY?TaIoSKXQOsriR)o`IOz>>sm4p#A|48|F=fAUz$ZF9dNqWn*X+9(D`!Se0uqcIj5Rw$f9b5(f0HYGNr9mK?pa$ zT*Pf{qgM}qna7IIMb0lBmVja})h@@4lZG*n6KkUta`Z-UR(zWL`0WguOC%%K;*~l?Ziv6H=K1~ zhdXdU3RdNzk#+ik^vkv*2|P|%H^TRjK2+3Ud#)0rDVD=fUHxL2sU+vS|M=^YI?5M8 zzj2cy=N8_)goM?szUuYGx@$PUWR-u3va@d2TaU|_-yzpqmLj_ch&P_^1NH>msfu(3 zUEOS><^vBLx&}8-n^+4wmS8BUaOlhwxeQQPFE}BhVN2zxL8Jt~HBSCp3JXfr9kW= z$(A({VL%(Cz0hy!KZfV+K!Ng?0Op+<^x7_5?Z+N;!KB7AWt#sI4|^-DL9U1w{yFgI zwjB8YH>b7SWtZ}`a7uJ4d2_GuY(8x>E#@Sgty&Fsj8#%b)`ohVwV$zA0B0-YtGoK9 zk7DX8iuRMcy~YbcGgZ=$j1T^|4OGK;2UGAh8S;`PG!rx3v}Y?+W(mp*aXL)++hCFK z;P&m$^hY>Wxy4ed$KM^LxUH-glIs9E)z62>Os-zfzW4e$>STv05_kbPtf0}brrPOB zf%(f;5-aQB!*_LTsmkOWisCe3Zp;JTqf{4k_F$H$#Yi_3t0H{6Sh|o(PO=_`5WN$D zvj&JxF}6imiA?Eo~|26{5*ot;@5>Zl#+#k^XH3nyOmycd{bMGHE>HXBAq2;E&zCZ7o>k*c|9$ zqYVXqvJgsQQvDIn5*jgpEb=p&BJKV3-bEhdE=}b|>n-M^Z{o9RBvvYL&( z))d!gQ%GKFmwQHt8O&WL%1N*|ceVyc)Y~%%tWIr$8RZ5zIityhIlDXU>|^oRsiasl ziF!Wbjy#AwnxMIo)wUAMxP^(^m5Ejrq?36BPL#)`@r!KmZVPPt=sV|#gc`M#@IU5R*M8(#>)j6D8!^nYbj50$0)ScV5 zVH{HxE?(VppVxXT*EvGxO{AlB!t(j`on=atTI#F3(5OfqNV6BO6|x=!GjsM`ye!U| z&unslsgIP5xKS?H43Vz%^$?2USHlEoBr$KFMtLen!pgW~I+~WN-sC4ph(3rm6Mu-~ zh6EHRA^To^=>Wfwd!LoWvt<+ogueDC0=Muvp64OfIf{rVU}XFXm6MW6E9$C^V8OuD ziN*1GJk3IYV(7RUCi#2M=a{BlGMjP14PMDCv9sw6IQYysgkeO@Ju8!=Ov(PCqjI#t z*nu+EMFRy6s4onB9kO?}uKs2dGU6x-cbt*lIF=uR-JGMTaHwQx?imA{QtwuSvlOWY zL_x=yR!u#@G#Yt~>G06=y}j8$Oa=x-Yw%UGk5xOsyW03}2ja#q*xSy?@`5_j7cRjk z91R91CQ%snDd71?Wi8EKD1L2Z_Ad|MBu%8K6a%w1P%XOZ-P2dp(~Gd%sjap)>1B1y zQfa_N%BcXczG74=h-;_)m~dTS0)V$jzunV-h94i(CNRoYo3%7he|}ns9L?%^;quGR zD|iGJKg%9K80REJaYEM{h9M#K$y0Q{aPfam{-c^gg_Uw-vkn?t(_vs}<^BGB!d;X8 zc5}^gQ2ZTtC#sEue{deXJP|BZK?MxvU}Z%!hY+G)`)#W-GBUED7LARqQek>pZ#B%C z_V>vb=6f8;yJ4*1V?;gDK9fOk(q|Mdl9w-l_W}v__J(^e)4{tr2qQL3v_Kqy z8%Va0`I6)K_ozLExMvtaG61Hiy;34hpG?Lmirc8Zf;0YT#F*bux=-dpEACch+>U9I zmJX|fno>l*kvNf;7gCs%%M0yw0aBaVE-8auN%71*+9t4vGC1r#Z*Crb8=NO+!kRQ< zK)SY+$#d?ViLG7|dv3tbMZ0z}$R`;vQ||Q+I`$NBAxwZ|G)}|W-4=}%ep#n%D8`4* z3gFhr_php;Js2)s#G@!8?xHG#-381_gKgW2otQP9XwJw?^V9 zS$RN4M3NoJXQK~+kRruxF-K`xUcmNkZDgqp+6;?ScuK;)R{q>}OIrXM9$XV`$#fyG zRLwos$*HhM8W6pxm7%c?EizVf0{_$?Nuzc$LZL?%{`L6Q;15Jtr3x z>eKb99K;XFXN%XqjwkJ6;jsyPdGr#&{Ak znUz=A}X&+s5@=(|8(=C#&PPTn(Oo35qI}ugm5MIl>`QE#xoPYyo zH6p6#)7PnDdzdb3<;z65tfaA$DDbkmFjeg{%;FRjOhU0(_z6nZ7*GP>9Hr0=APbe*lAp4kueYm>MxBw zZ{`gx2nfBfW5ym84ZM~;tqqDvgJcSQ3cqED!#$#=Tz(TP=HB30FmV@QoG0lSZPIlA zF08fIt|79>@f%p&sbz@p0N;op*QC;iDRIXVv$ZDNC+s>9LJkC`&$bZa>q(CiLqB*$|!XF|pwJNB5gXDqOIxMD=wW#5Ls`yT|BR6`cdmRK4pU3PFJ z&t}l@XZ&*dg5Us`LJM}C@M;QK-YOHUnNcw4>{Hmase;56SIknsC2O3Um6w#vKUhQX z%YlIdt*ddn7Q}&TD3l3x5T%ApYmXA7&3LgZNA10MvYZKUen~fl37FOc< z1p7snsG7LkMO0oy3d{o}LR1{1_^UL^3C*JpEV-P3&gDtfqWcfA*q_cjo+Vpy2FpQn z)PWPZSZ0O+Z6~@jc*im1Q8vC}^q0tFyj*M(S-$zx;e$8CJ=YD2-_|6xxEfjsua9*k zOuYhl?Yx0b2k!9?g3S+{Nqmt>dV!?+58e8``pc?G$5=eNq!Ctdbo|Yih?g`_{{)vz z_)R(R_j3=>gv7Lh0NRSDqisXV9pT1_OcsO1mzO#}pHl0x9rYH5M!_!DPza4}>sQ!7P1Uq!V>RPzJE-}BFe7TyAqRltW~w{1Ua^+VH0Lbr#;4bH2aILQP(kO4NF zO3rI#IWk`Ir{2Gy1Q2g_)VYR1-Ga8Ahbq_Bowl>ol^MfFUp;7$EEJq$3FKx)E_(PM zX0JIoz^K?;3MFSH7kzg52gC>1vHj7xK{r5v^}-ad8XhB@Zxw8j5s062ld;J`cxF!e zG~v>S679-0{S<<8@zH@U_pi@j-NS9|fEGL+yE&xJO3*!)HT!CT(o`{P7oQGkbM@mw z{okva@41p)s?_NA$tbDYDnz3N*loJh=rnU)A`l(0y*W2b!$F^=NL9asqxfLom^%T< zeuy<|ygY~G^7AA@2+9}ol;A)NtKfJcT{&K!Xzj=k4j0x(bZjX@Hw61cERP+ z26w}iNu%AH&ZLJ>d>EH4<$m@54toItWYoBZc~ZzK4ySg$8$Tw9I0x5~q;~f9KFZN& zOR$Xm2chu;uP>0pJFcUCY-#kHH>+inQKiV+j#YBqFtahkOH?6SYI=jjD1&!zq` zKb`fPXWU-s>~+S9zDKnd#nDe^1J9-V7$RHhX=9tm=Gy@&GiWOC)^tacyga_vmF6eA z{BxX4M8_LiX+U4A{b*!{eNVa<@=%};0S`?ud7jb@=pPugx<@$&=S$%kZC$Cj0 z!fr)yq7p2|4;sRJwv8&FttnA7)Au^!=xd5NAQO5md;6;ok#m*w{XV>< zws5Jgi*S1J?3^JSOv|)5y+$Txp(zQgl^;jm5R8`e<^8hZPqPtdR%gyGjm6Df$vr}7 z{Jj?cmFARlFRN*gIB4h?_)PE+ngf)rxs9Lu_eJ=|&uS&q|%O z^G@}<;@2oMdBHm`5LPcpGy9!ec01FmKMcI+Q4$3B^1g(_BrhKlO~lF9_#Zv_J1z%k zL@!+hR>_;e+SWt=v`LW~WJ00m*}h!$L5y+xRaCj=s%LRR6##YP-z{i_)ma~V&z#^4 zgJXCc6aMP|Ysy<*$|%aHMk)@;x7yI}9rijuA}I(aKX2&WdKN2l#z1+i$LOQ} z%qHK_3FvyWhTZ-7xtnuv;xF9$agzoF|Ou=|A9Q_Gir zuU94G>yvalD-5L%XElAKE1UI%=7lH8uT;H?Lm*aAOKOmY4)XmQynhjE+s*mOmwsZRjEXqG!BdS=2 zg|I_EyI$%;Wv{9c@Z=~VQTZ-%1Ta8I(VxWL%24(^MXWFg!8B9wXKo-Ln-nXec zKm8wwpH_V8d2r=uAU0h3H;5Ox2a)wtHGw?gM)YTR73RwR0)k`Hzplu6D`4Kfd(eK# z0xNp{Sny_qtqgdLnj$v_Suz?4QTSFSMf~a2yk7t4&tVF9w?yD}oQef24cZN%s~&}< z5mOT5&sEF4igFCsQqG1*+N<5Is(SoQ>xq5Yqt0Opz50szX71amg8KsT(4JM>fxz$M z^GVRGm%r`~L({_B(0S5L`XbjpgyMS_eU8o)1!rwspEp1F>4;xL(ke|bV7$4A>jOQV zck!j6|D|@h$g*&=?vY}QlvkUf)dlXE+jA|8w!7{QY*gDtK5G9dwPA)M-7^;Z3_VJn zyrc4Ji#e9Khm!HnO8!%whu>gV+yp%HLoqamS?&`ECw+(W?JdB<>pW|N5mYFyJ!MX5 z1>&J+LmYJMiWq!+%t0_!O9ll8;^{e=Mm}Kg4j6HQX2G?L{dQVA+*6YJ;CIe@{kEn| zjBd3pcsqFaJc)ncCK8+OmdaTCjw=a^&%fViWqeL?d%j&g{cUW~eZtV$BM9s=T^|`f zjYkMG+41E5+fA`4<(zD+gy6Ix`^X};n@imS#M_}#pX{w^;uj}5e zmG;3I6K8zBZph>G&Nbkm?{mB^0z)KJSgah8b&pyf7)Q$jL?L)W2Jol+{dN|~WL>yd zg|1<#45=j9|K-s5cyK1Dlb8?^go&D@Qsb7eDXw~Y(w!XT@l(OekrsA4ledj4Y!9wA zc>B{(AATxcnNQr!ePV7%H>hi@;&J~?B~Y&k_pFhDV{GksfC3^kJeN}s<4;UceuG=%)JgCs(9i1Mc8PNAb z;>3;R(jm0waY`uR^O?y@{nZgD!sFDsON8h*+}*-I90h2b90sPzg7-arwYnLqC^&2r zuFfTYUs47y3jId2;FmaO${mc3N+a8#%1_MzW6BHxvmN|B6$M>g__PP6DDt~>5+gAI zgpDj}D&9=be!N@n7Hfy`K+C0VR9=Q zxJ*GYQx?E;I%D{uzx}jGT3x4e#l02-A38CPMWzM}jgyqa{IhrQc)pYZjKrTu?R0PMvfjQgEa_6U z82jO9^cNtNKG3+BN-xjGOnXdWI$|iKFs#63vK*j zS#anSe}E;JThoW_^tx}{Bg5xhZzPc!3nos>>TWKj9WmAvIHRP;pP9S=aVD#77 zloALQtOx~nX2AT#yCc}W!7GC)`hdf?GUu}d#-x`uERFKDewCaq+Ob?9+@DKbE9-U= zUp)AW373(ODhbo(1o11x2_dyVL&8g5$1bHn8>M-`vYIl_P8@H0X;^mtNV3uPGh{C2 z0J}vJbU)!=YMzgSXXWA=lrr^*0-DSqbTDA8r87Rmjy_v|C6LM!>bGYW_7`6~OVDgL z;Og`YF7I@hI-t*eonXjhjo2#90(^8DTe!x1hxzHsX~7L!OEYx)-aD>rr3^Bz^0N_- z6S%gX0Zprn(YSZK){M>iogIqW)muLD%)1ivi*e|8?pU+n)Alsl@kO_d~ z6cdaoK>ToweTkfxs$&fX=1$Zn@BS8p;2=cQ^3Z8qdj3~mN7u$B1?)THUA=Xc&{jTS zjW%wFCX4*{$1NcbD}R5ua@AY5R^$GPIE0Yr5FYh!p=?J-8C|8OI3x=<^ICHtEB?V97;=u9#i?eY zT&hpvjv@Z|7gjg=M^$InIFD9adg5luQ6Yzn^luTCb~a^bOgwe#gIlsfW6p&(z7nct zM>n|?DyzS!v)a3L4hEBxnJuV&&wd+U{_#r2pOMib!0Ly(wkne&UMiF&jD7*C0T#GU^x~`vWvb!K@a4!7wHcFVJtHi-iw8Ps*9}IgFsk|D?5Wp?N6i_Y& ze8cODf}dYsE%tMU(%Fn4Y9nf@ve6ncwY;s`vR}~q z7mJ|y<)4tQuNdaG$x+G$?s%=)lA$glj%z#H6}gdFURIsxjQy)B%7>+|-B9XxcoLKs z<*)Q+J50;jfGH@{X)&nNx4}}hS3TPjjgnhW@%GmvLSK}lIR@AepFBD z*1vllwX57$N!nxlHn;*V&cbsr>h!0Rx#PA+sd2yu^$n7yQOUVJ|lRxNdMvz32q8} zGiS0B_4NZ`2Hjf(q}z9pOO4D(LZf{c?|d+t6?Q_!^43t3i?KGY+O1{O_YY>8M*Z?y zXj<&vw1;62=pX|zf{F++*%LA?g&VBosKMG$<|;jc2= zRQFMeUv3z#+ecfgT<`)K-W@4%Tp$j24bcqUZDUxGG=Z;Am4-q^$%Q4+HthSjuEUAP zdA6B;J|!5#wmt{@LxTlQwELLLYnu=(;bksnR`0~5VU+C0Z)I$7?AA}w9qQTZJ$a(n z9F`XHdvuU<^Nd!Z)8<4N-x~YlV)?L`D>t}E>I_C>c(aU6)o=Tcf(DgOFAY6l$(O-} zgb-!Awt%>UiVoejamKX?4wLoG?pnY+4E7e#(FU9_&4hD`=({ zbv?2$F4KHXioV{}8n9sT{#e>>fecRW&!}?M(`P2u@FNbri~nx_O)i&fa||JA0;$>V zvEZCy=(cdy?N)Az-2}c}U&8unSknngoBTkWDiCmAExVMKZ>%@}vH$JTv&hosPQ|hh z69f8+z=o8ov)lNGh#4jCqyWjYiU0O%z@AOj5VOtb%WUSe!&u^?Z`}+*rI$u2u%- zM^}K@NtKI4a}95i*I*^a$7_lg++1A6au&Nx=#}PapQihF)m#|lgrJ^$ZUXVgL0%5gGubYc)NPZDu7>>KS(7!F(@j3CFh# z1ja8K22^~rk?`l1p`J9Ki<8YMe6O9#amK2yu=KSoSB7gvh)7r>u0Q?A=InB`@$*5& z&}uhCp@-vxb;ADl+jkP8IV$yP*`4N65>CCr#&~p=%+SGx*{3g!-jQ77^HSMCNWwsS zl^rqJ+|dP|{p(7VTsi9oGkN1djHFXlRxZ8>^<8+)Y)WHb<|K2BwvM!_BB2lu2#|D~()nuYL zMS6J!8~n?R<;bo2d4h3qt`F^ga*@E2c4MCG?4{?64f9{m*o-W)lVKrFX)9qAPVqjn zu4-;`0>qxEdq>vDKOgvAb%bHhX$cwv_EJLY`Nlw$?2e3gN77dIbF+B)k<>?)IqV{N z!!!Pr2y3eJ#gwWG3j6Q7Qb-b|GeJ_sVWTrvtnNPEDJ>~G6W zr=w}=M1J_YIJv5`_1_*@K>O&a8H<{YL5In<_oLFxU#d?F3d=<=0G|eFlD9!4dsPhO z$9?^$=#AWJxgcDm5H?A4SoPeP=*~J5I*&|Mv8x+hJNxRr%2y}|LpY~aU9hwmhG_)+ zy{L~!7W(2HLjKc4hk?KzmTrA3f|YJkVRn?5tHMY zaVIts_u{aO`o697ljUY-_xha9+>|j~_XFbXf&gjUs}=6=N}R_&8Jh*~Ft3xY;hmjc z7JJLjcu|gSn3Kh2_)pIum*K2?>bBPDibiqfUvm5_a{A4G=1@({Q0xGI`iduYXYugC zqaG^Hc7%qu^Dv`B=iFc3sjKM2%@%3o|KeO+&tJ8b9;?V;r;>3jvM!!>jq=vv5}}%+ zm}h2V1rqbx{v?jWef7)^qO_Ib$+N#&;Yj3-p18pN?3&Ola(1YAR(s#C@jMf=LS!32 z_OIS}O+^JwbEZ&SFT)40^b9BoA9Q{+6bv2b+s~=%Bo-TR^l2pg9Q%veh{xV-d-=-k z!Mk`o{VDu|U3$}3gH>Xn^*L)~?|n3JUJx{R^M)zZK{mr4Ejf*@sM7T!oJ>Ka-rkJP z??6oK?uP%RB&vDmk(mc^O>n}jV|Gfp#VbRiTH=BUfyq&V!`K2y1OY@nFrU>g*_x|| znhi~dR^Aa<-d!jIu)BGsmayw9upuZCPO_0?BrC^#S zS3T&WlhPN%hwck(LNK65wV?5pp3WhXwe_67*|=QtZ+mDmF^qmfgU12Dmy9>KIYj!0zS+pimK~pu~ve2;CzzvR; zeU2)Wm(EPdbs19e(>R}#d@Aolye;b_?r5qI@*6O``J=|%Oh5?!XQ0*DuW=l{V!uX< z#_2wiW<9N0T*UQel0%I@t z>N%c`h=D*gL9tQa;RES;7_A&o{o2?!3IL{3^tZ=5O3Y{=hN(B`r#BxlhBeLGYV1q& zE+ybvoC%AM^(?=FuX~DyqCOYoy`Hp!LgU{>R#PFmy3C^Gva` zX_E6e9xhT!)_;Yf{39>=m_NV%e41|16tI4@xsbOsK&N*uU7HGhrrw(2?IS$YHU;)* z-U!@w6T-ZCHf5Kn#bgA<4Ng=SXuFX*9e+yH%%xIZWOR&3KOf7#EIRIOm&dQQumN!H zS^}H{PVbKeeEY-wf6e}!a!5(~mb_4OS~ho=jYq!i*PQCZxDVl&mqlww%fwuChuho_ zTo6m$kv{bb*39*8CgR=&X=5P9A_AL?*X=bmj=qD6>!J zaJv2u6uQ2x&dO2_3oX4*RW6qEoE;9`hoR9Wnl|!%wGD*VuoCIy=2(1PTmkP};MK-JC zV7s1=^IUX(JRL8@;K&zv`OoML;yfE+E1G)M#$C|BH~d7;W~Ig1n5T1Xm(|17#P0*u zQG6S^TzM9Fn{D^k{TtI2JpeRLRB-ePKT<;K{GuiQ@nkBc-i>uA8;4ioXtVn5?uNF1 z{m`y>a$PRU4#wb-TRh@l zx)D;TkQ$Myr3$-?8-4*eWtYU1D`hlSkA9E+7z7|3nu;=(NG?1z7HY76G4B1(bSE)5 zhq|Hjl6^TQRK)H&*;u&4A3HP8<0l>TVFMji8)EWm-~ZjuhnXnOt}-Hu?p6NkOx2qs znaWlEmg$ryf&Z@AzEe38vEtowK-8{%6zTPX{9c8as*X7eOkx3F&TXDsT7(PD9sXIG z1*LKZ-*_X$77X%;ASdO~64PNR(eTgQJCqmccZT;F@svB)s z+^|H@r;1u{&%(%^@^7V4BA&<2K|QCT(E>YWBG8jk=cu71edO$^cEj$|qfPj%m$Ip^ z2Wrs27-S8*^_Bq4>*fL18-SZCH*^cVFEe!BofrPnE!gSpEqXWhfTUMosiMnER~PFd zol@lX_i~v_9%gRQav4uo5@iimM)gyq6++T_4ZB#(&Y$tkW2;;nj)ylAf3SCsocQK4 zNbo+Ooma{}_nxl=%ZGJ~0_0{-_}>lnr~r<6+y+J77j^*V_h3|q6;!qMx7xAqQ=ie^ zF8^>2*&7#x9+iu)?dr)L<7$0ed?85aWj@EqK|f*^s7%~sD3Kb15qYX=NNq{Ylq65p|Lck?H&!Dy3N2zX$~v*Nb)%a=a>vtG+G=jH>1;z z-IGlCvvh7MLU{t9mAaynnNL}lV_rZVJtjr5g;~fDuVp0%7#D+i{UIWtdQu>TDRxdK ztyo9#rTz};G?sZ7W!{BQS82mz;8-{omC3b7$e^<4_aXx(USc1FoeWLptMD}xlvQ~? z=K+i_J`UMbn=6PuxD3Y)=W;VMeJ8m6+!sEKN`8ic-y37)$H1pTsuk|Zlj3<1q-y?a z*0m^Av57~IPDb`qQ&B}I94h19=$u#l9ZOjb90DB#1Ox^oHjGdIyV^EkG&dN?Q!^L{ zBM3f-qm#RZsgskP8aa4WPnwG z8vd`;9{=Fa{QjSvT>_@`f9;zRutl){?2&RXjsMvZat{Ij6!=fTKYeePdkBKL{%;sV zI<&$6hP5Z9L+=p(XS=pT8~x9YSZjxFUu%bfGC=(|4gCmTw6k1n5D*+45D>WxlnPL) z0_Xgn;52ZsAmIQ1Dl`6l2p!mzO#bRsVzF9IXCRb`Lbn)?zhoHJ{I)!vE~l9*w@zaN zgD@5){JkF}xudFXBcg-p3FZm+iTaKF={z3?gj_`B=*aQ<@yEAGOyzo}`;5I|(E9be;RYhJxP_yl} z!Z&l=e&ntir+`5AkIF-X8bynE?4n*$MKQCj$vjTtH(FRepN)Q_Non~OcIA=CB6eDI zZsa43ul3e1u%a21@7tmn7W3i)4rW6 ztpNd?5$yXk>~fvT8#^aoDtS2DsJiXJp8ft`eP=_o6MX-U(?@5AKkwu7cRi-GMSm8= z2$s(mfPR{n6!`tHYNg9ZVe!r=6ltulw zd08awKq?xS+0U{R#maJTZCU0hh{Bneqc`m%?SKK@S5Y9^rJU5>C-?XthQW$T9(>Va ziV}M}&6jGnRIzbC@7t~-1gB_rvi^wW@V;WXJ{pA5CfDf4M~Io!W&&aO zb2-%Dg7zO|h^u8X(vyGx9pR@%Q1wsp;~`ei{ey^GcgNnVhzsP&DuP3SaG_4GufLQS=g_& z1`Toi2n=+wM4sD_4A8q9FZhKAXrVyJXF#Qq-sYQdb}UiXYA9PQKVGw`Z3F|dsrRuW zI&oEaTyh(8pd#7||251Hhg5Z6(qQ=@oX&82^%Ou$$*Kk_h*3MgmY2Gg$xTDSsaV9p zce>(NVg<9uB(bLb8#znjpN@5o=XE-5=Va@R^3Rekz;|&0;kRA~#bYw4G=qF`g<6_c z3$DbC-l*c(ga_DaH*su;rTVVbnPf^R=&Z*QZtc;ERB#i&5f$Bq7Nm37649=L8D|1l zV>jRmQ#F}(3npMq%zxxBL||c_kLB6B%Z>>eL_?X9jH8qT)uZOcBVUl+<$;{-@{u?> z<$J1;6kE{ybN$SyQJ=%;UMVmBnMWiP-Tmk4_j3hOf$Hh$6jRnvFBwTDGV8SJzGN9@ z#g+^o%_)KSU*8Y=Z!jy)B!`*a+>{RU0(gN+p=}#UId|R$%$vUFtAh~l0DoWFK<@s8 zhJ@jl+qcD%)lsLRbv_rDM_vD~&*}ML7pXxB50L&(QeSfhVNTdrq^hj04~BU8LM zu)p_Zp}`hBLeKA+JUdf?jaCn4B-AwlQ--xsW;b{qPjl=Cm&L-;0xYG)iHaN&H}JqF z6knKe>>q<~BtuAx+97_?q~R<6u{OHq5=Dt5VXT@PqNXam8d)*4zq?|zT|xM18|g)G zo8^r8$3=yi7fjhL)BY%9X49bI%*Jv$uqe3Mm7Bo@g>=ANk`K?*C4wos)Fvk(bNYyh zK?ae%!WRzoV#0e3qh80#bBoNNAW{LLJ!}a06ClO-X87eNs?{rrJ7 zaz%xW70E1slQnvAMf#mVK3_3Y%RP&RsVSpo9_^Jb647Xp<;ChyZ}$fN!v?)chLndn6m5$bm73Ue zi#jqWALI-X@fPjlYO^%}=Tjfh?iBAuo7;u*n?enrx@{nfZ%Q7nqZFW5NXN?8uy8E< z)*jL{bXn7?>h*D>*=)ebXU0`@fnDw_dcucZX7yt)VmMHJNUx4mMC@Kh2pdS=G=1uLqRa?N4)5wvWC`M=%1>F`!t^TaU!rDIybYU>Y#MMCR9*$rHPq#p|y(U#5GdXpEPFF8b&_7G=;S;VR!i%7a zqHqd&@ZHu_x)ogRYm>eEDmo=HC`HdX4FWla5L$nIu8~~FB+oJmcRAcqSY2+Jbw=;lLf5+|N--wJSRKdW%p4juI}d}`{;JnKrlfR4 zOR5&QuFx_+vlGuqb*YTvIZ6?Ki8$Q$)!Hy%*wCKUQQy?1XG-(b7hy!6J8njML?Vh{ zI8iterQOzqa_cOK`8YEztW9U}mEo)i+ zHkW)M7M*XF`cz4flYzNnbNL)?8F8Uhf%};VW~BpWwyE*wQ%{}SYx(9>q%_~bQuuKN zC()Os(MG=J!G-0@x-CDZB<%N4xG|0?vNL}zBp2tFCC2+?!PmC;*KC8eDg&0dI#`{> zYEWB-u|p_6fG>5SN)P>xo_U=}%Jw|BMHMwtX8hyN)vWsOL7qG?e$s~LG7xFq`fy^_ zTzpvQ;b)2X4?P$`sUulMX)q_1<|t%h(WTl*oF96K8q8|)JcTV)xC8&fBz3y;E44&) zHz6%TQ;X#p=b#FC(R>uN5v<&PG-$TTu~`b&)f4O607}iLWM|kTe}i0=F~6jV1&M{o z7F^A*O~q%BZ5OlQMAOtlj2=v3JM#=);QxY6F zAXBqU0N7HjNMwS`51I1mr<+akX$Ha6z%deuQ%tfk$-;N=F#~}zk3Sv3Zdk2aHEn{e zr5~!{h8nu*TLS1JDnW!|)J0NrknbYW)q#n=-|f&@w$=t}knyB_PNtTK2tq_#d7&Hb z1GP^w%?v!lYlROd9Q0`WZQ*k0B8xpps`y;nfvIYOzP4!FC;8=XX?_|W;{ln;Xw9H} z&m~jbabf^cZ6?3^W|_<#(&ReG7)r?qlKS3umE#mj=7!X44 zJ%O*#Q>{bDpOA2YqG7)zrjugL9BSl^gUQh0tE)nswf!277f;ioZk5x*YLD2xhG!I$ z0UMLDYy0Px;;kjHyXSB>{$43UA9%Kec6rNm9(A(V{d-K-ho7+Ew7=SQe<)iGgOYvV zq){AbtM()GH!>zzV8_8hQchRY;K)N+SGcI7&>I)yDJ*fgf!{yjx68D#8Zzc zU#*x%@o4#?RVRqGFgNJWio&%^nLTaR0#=Yp%YWlCoF{Q=MxXpjgM7yH3?(1f;AD+AJRu3P;^6ecOi|s=ILcsVfcar6 zg@?l2o2n^}xN_UB2%8eLQ$F<464g!Axabn}q8CyQERW7V@O?Vq`>{80gRYtf|AM~O zWH-dhU8`@Zwb!Rr1Pl($qY)c0D=vC2L!xq<6lf?$HNq&KEcjJe3!5|favMv>7B(H~ zCCu(%JY&7V8U4lBuLelLI-ucLfjLv@@Q*GOK}Ei$ath|Q8efst_IaP>>`-BKtIwRh zNB1Y4(#HZsu0aO}A4iD>5wqnGFw7NCt8+7gDq~FL#U)ca%q4W0%DNq$H(eqBniuJY z56^T(N9N!^ZZkl~NRD0_u^HG86CMe zy`G~f=l^O9Kh;W3C9VA=NM?p+JW|T0^|{{EU1*sZ5AB{{QdA3&2zO~D2rxv!=nOYC zX3ASqoN8)J9!JqSq)2PXaNhAC&UaN?ZcyNX;UJBY;;coo&f)y{Rl>4rbO+LO>g6n;M2FK()D2F8v-qhWyExoZG@RTJh&1!fW)nFfg-h5h=^;>K%R& zV*US6b&k=Q!~eFgrnc=qwKcVEw^M6sn@?@qwlTGB+qR~5d*}SmI_KV-m7N#K%OqBj}Q06lQhhcWu)?*(Z^wsEe_0~@!aVyYE#I%BJ zlrH4tyW>)QfB)PY<~PmOSVMhoDyaTTuS0os|GK!8XmeW_AanvP9l}78v^1x&K{;_R z%^obcN(>)@wM^hdTELhrX(%h#cQjWPcA|@#*m@1Gfm7|oE>#UwVvs9Oby3Wc9;a3P zv2T@rJu7}GCj9L0USqsuLuf6rZHNq)iUo;jyxihpT>y`Qf0Lu+X^m`D9kgZX;|FwL z8Hx7h0GxPltk!4>{_V26O2+$=<&lQij`N` z#@EfN28eG}-EIT7#MPz(`sgrw`s6tqIYI-!eC@vN3j;bXe(IXpT%YOKWH^gAnVJ2- z#K;XUjAkMRT6bzE&>94esZ}R<>bJV9=2WDgUk9_wIw`jPvP)!3h>4dj0}iE{av>UP ztYX%nPw{4h??q-Og25ywh%YFyVU$;t_?3Kq=Z%w*J`Dh;I?x^l`e|oQE@6ns;Zl;2 z?o*^&pGDmx_m(}M$!i=rJU0ajae06Ah=M#7Vi0CU)ojBUc{7k)>PT@+Ga?ioFzzC zLEhRo*`onD=Yv8vWZ2ow5BI*+nLkoEs=$?aP`J&;Kjg%2&%PGyIlbG{scLNgw6<(~ z6X}8k-Vt|l2RTKfaWY*Li6im#xGg-jhwcx_I7kRQwRh$4iP%|>NS~~S>1>SMHyMd; z5z7Bs%!X|pxJvLs_{$-bri)c;=GFC=i91X^Qo|B>5l^l!zz(;0W5p*|a34waulntV zAK~e#&>lGO$K1xOx1ERPxyAigVOpa<9qOk{{-Eh+%{YZNgN7EpNCt}?{5&~{_Q2S) zb)VA;_h@?nH}@m1suh*kueuq(#>FY|7K+t3e8iZ;j_^ zEK&;a*8<1ErTvWou$6PQ%wkgaTkAj_?*{2p7o)?H0(YXwlX-sWU6Yi5rKtIAn04*s z2ri?NcdMmbY(&3#zfyokr>wV2zLZtL1@l=`A+&t59|LhsoGn)p^5lG`87aH}^jBa1 zZ`Ha*4PEo`TceGeaKqH3t59d9-IAwae&nsTQg#XqGYmOe+?oY- zzCDm86UHL}&cu@V>!6o0@06}!cUk=y6_FK#j;-qxiK`OO6-(C}FVX4rUkCwG8-HDa zUaVGaEh{Qr2G0g)R$Q7+_|13Ihb2NZamh1?oFXn)XR8C(mK;<3Mg6`(arb_0N%j-UREy8YoG$67zxM+C1ao(^m?gp;n)H+l2p8RELD%ifNu z^X9KXm%cgqc*rxTE0QNvS67!Deec`swT#`b+RD7BGkD{sgoPtO@$nCtjpHdc;OjfL zz$adCQn!19sx9cRC*C@`CpvX|7^{Gu0@fJD(w)yixvI$KGFSS^t-$9J3Wd^cslKsg zG;m-non_XsQJChPYllVhjq*i0P2yo59h_B3!Rixi%-VV4c|3UgNZ?#7mM04;08S5< zM#24q2l4xw4afhErDXI_B$(L;NIjzc{is@&YNdlv($fHW07n(t{AHATHrGlw`C!Eu zVv$pd_WgTiSqGCt@(A@~jp=O_$*njE4+9?ZLdJx)d$ex4n~HO4+1C0S149zCx9Sp3 z_pcsrtBjxZ!lwRSmn$9JlOaa04ZlH)F2gb&B}*VRnzys*6|vh_c|&F#a2#xU&%&5< zj3Q9`$d;jv*|lruC!?~$5!308)7i7omtA@lp05>&VjwvC>Xsi1htp8lpA4gC9IoV=JQV^@`P{Nar>rHlvFs*;AzwYU zV|-%nnt_R@;QHM0ip#|USbC4!>L_ zQe?48vhyXbj!T*B3G{I@w(A~FrgcfrV;i^&?lp4ndf9|b%hQo5(2aYegoh_YAlkSC zbR2K#spPlutfPwyd(v;6%A!e(C7$3$(4zt;70#$N5u69lJ*+H8WpT;Ro~KFf?46S) zFk_>YO_goQ;@Z-mo27N+i*SQjB^ARV`JV~7>U$xkN85eGmr$8LC&y>&H=9YI$^(9C z`Wobw1BA#X4f=uBK$G@pq8Tinhtb>GYce0n=hJ=EucR5hdau*h(CqiR{R!cqZs@~S zOn7=o!H+HUcf(xqAP(ozP$hK8Iq{{&Kz3mzAHvExw3Xc|dYkI+Aa&!cG3w9>?G!Qz zek7Vo?K5wy=T8Z#B^1RmtlqeD2JIokZ5x3U9XVXMpZl<%=U4>-|GMz+8Ccr@w|r zffg42E^7WoVmka~M9H~lKP6rjI{p^8w{!F(JUYJrUhv;wR&&=bDjZFQwNsIPXp}J> z*gSO@X=jk&Wa1?R^Amt7MA`{!q`y=%3*nzHMD(LHbuk!nT*DDRzeT~AQ9tQkwt_#7 zy$(n#Xh@dpeKtWBg3LeGu|Z<@E^u&9ad2scK0sNds2O4A*&w7diSG~mpN_CV#pK#7 z+d%%W`SV|X%M3k?;{SLcEdT}>!=?XNBN5=RI{$Mz^a>pIpWyxp{+|&339;o09H|rn z;lHesk2q2$2s#iD+SGh#r2lneT0WtX&cT2e1H9r1#&{lixKU+&LgsJLmm_*>9Y%z!e=1tc$C7AKGrLHHEb{16cho3mtk+ zG?NB`GeW>9eqRKRHP#~4t+2u|C&Cl5S zYdEk$KaxiUt5$T(KNRa}aK~Yr;Q_xFO^GbFhqZ4Op^>J`~d0?c(ZP~%Y zzAXMBvX?d>V9;vj6*epLLOk$Ux%yT}=6>=j1usoZG$}HR%&a{v#w!San+5zE(D<#2 z;tb3ss6<5V*?7D8&;>|7jv)t~H8kc@eibUYRj)zCCr&69RZJN+A{ft}vq2WBrJu6f zn7!3R=Xo-YQW+~ye8D?_K8-s&A-#caf-KSga4NeLFz*%L8QBPVIAH=b5SB+hq;>)zPI7b$!LMs(!8x;a^2RI$bS0+a14|=pjf>AbBs<6lXW%pEfq|WG}f}|mK#9$6P5YxPO(TcsQThgiLQ1Xt3&UNRz9Y`_Gqm?Jf3IKGOzz8R-)s5`0(A$88nvrPrGIayVE*+~UN7Ar zBw}4&;kiCNrTIKx58XGS#df(*Y$b@fZqpe_r$P@Bm*oaCm%-`#@gY4F2|R zhxYviT`Pt%(boL~il)(MQp3^vekhrbNi8B3X=k-(Pqg+1I_Qa>rHEvINnK(M<2`N1 zkd9(>+TUa(fdX^D5eqqzWr`o=BuKU7ifVwZAkm%AQbdzPKb6_sCv9Xw zkgzEk#U$yH#A*}pT?1caE{4V+lG`c8s^{m5>`aJK9cIIw(4bcsRVbG{NEIzmo)W?y z3W7!?F=kzY0{`Nd7I_SgQjCtPQ+aSsj5UataY$h>zYWA?`8g;RX=DTDB`;JUi+Knl zMp6b#&IQGh<7ofyHlz_8h^?9OfP8k1$l57xpO8vpgpvxqLRg13gnhSJAwO=5b|#4u zZ^esjl2+MMMO`VDi&d)bO&5ZO&6i0^n3l1EZR4sHsrvtIZaqQ~Q`(&gWZWsBuaZRS z&4?T3s{{^!n&;59lrYcpzt}EtsV)d%r!;HNkl5x78HbVeAp)rDUds}8P=q?BMD>yk z1h<-bUM?Ez)BJ@6UVKxV|JpawSM@h7N0p#+*Rm@_Q1w1wyhsRobwd}%nf+1&>0p3i zHUOsN8Hk3w$yN&4_Xz%kjynFWEVN>%?8l-5s0MakZ!z0)c)`9~YiFIGKkEdn(YP-{ zJu?G|n`8s6Fu+MSOr=gb|1c` zF%ug{|On`uX{0NcgR1(YgWK&RbU!4Lw&p|qy zcvpyArB&;eRYVIoV3PY$T(IZCE#(e*mF34QOSiKY7Wd)MaMlrb~&f4o}+bx10FXgPWdT1+E7R1QmXaG5{ zoKqu3DSW!@n>zK}PgC5^{crgXEPjGPVK#>3oakOY>LSz(S{``Sk*AwO36;5YUTxQ- ztXh-yLH{Yn+X(idW+EjIU%}0k&AayfLWr$hWj2lS3ye8w6@gH)Ku!p+8uV$?@zn6< z0M~j<9P$rOKoWmA&PbCT3_Lw|TcGPcRE1D*<2R%rNW0U8Y zA+))KoNzWoI1R@-KHD+KC4Gj{(HeGU_chb=9hiM<4Va|}ixBf3B=r6hdJlB;~n3cT7a<@CKMA3{qJeOm)>Vn+0_}+Z;)l)i7KnT5@GyMSu ziR}YO3r&TTxUGZO^UshUdL57{g~mmXlGc(1->d;&TI6Che9*E{=Kz+ufJ9TeG&~ye zImFS4@7^C8y&V2!m^`NijX3{iqNYN7asvHg31)_XyZhy7xrw_O4;j=oVnc5+2K5na*pgrq^5oY*l@!vvO^fZ?MEw zn-Ik$JaZF;Q`1YpV8+N`EAx*<&$vJg&yI6a@QjA|(?=aj#3;ySC!PjK!Yx^%twTexM>ebBc_|}w|9Lg&j}PG zXbFcYme(6xCl5d?_3zLGT9#aC$AG;)^wWd0COl!VsS&)gX-r48Bj*E7mItbpBsW!> zm#zs(P(r_(;3QBl)mw#??xnOEnX$*4u3JMKkJ%M&U~4~na4c*lXhRC7f&7Z9kTkS< z)xKTv=Y)_Z62`e}6FoWh5{)1KL{5s;I<;OI!b>JwwUJ*(srZE|q8A{bzt%lqsdbmr zmU;u)oyY2RqCq9L#iW!&qp8jF)1Z@>r)paC3=PGC6B^iccLLtn zY+;rGOE*?7y#M4`hUt)_V*}9S^75ycPp*Y*f9>9SPfWY3K#mL7N(k0OR@xUddvw+C zLf*^|gBmhIfn78Bz#Qt2`URo#Q7!o!N8yClz$HY#{u&hCDK{UCU=FbGw@RIK5tB9r z+})`!&AQY0`NDMkRIYxYFDM>w1u} z@t1<@j!L@r1)=WZ{V5(|MipSGrh}(l!4=L+DPGm7Lx;(zp5Su z1~agX_{1`KZXAG34w`nRlxu%8#d&9}?9Y|Nm_qRg z#)8p01`gDtoiS7eOfn)AY=v=2qs1J(I>2V%rgEb#~wuI(y~^ zfi5n;&+2SdH2NFIr(J7V*7I9uCv_{%y2dH1&G${l^S5_LL+N@Sng$Ua%lwI=9ECAP zN@{%=olLQmux!^KjNT0xUoXXFoIvVg%7dEmfV0FT>Qqk!9)G3AJMG@Kgp^@qa7GeKi>thI)z3h+>$B>P^#@zsFcKbyqt7eJo`no}ExVqghLdcX z^{H~aWl!0er7!F0gUVs0*k9E|7n)|pX(t{XY6vN@Q1SfcF>3zrD!h}Gu&}fets6t* z0M}rg+`CP$%noZz2}+$y8K}QuvhnfF%Undd?q$*;V-*(x0qy$dbTvh9>J`8i-CF2W z!x|Q86Zh4G^VD*xNY=613E_Zsn09&xL9JCsf8yh9A-y~wW$iA(17Mn^shYi=a|s6mCZ!S;lHCTZ%u*VnqRMWq-|Sd+wUi)A|*D zSc*J2_!*>kdlx%UM=O*TEiG|Yhz8mk=6gvwL)r99>M~=CiCDX|P)aVwKSf{aRVF)S zMcbXY`PiEn+f^!C>Tm$D%OMY7Ly!2cT4kHy%=agl9oYp>qfb~_aja8VSU{utg1I0==2?5USE11FmJKhh!s>ir}MTFQuGneq2ixV0_4#Yad@(I zI7U(?=4K~IQ{ghbs#d%b%@zJQUiE?oLziq`FPvlO5k7j_PU&?ljgust|66)xQ++mz zb_p<%t+1)O!zTn5F1V$28XVl{_mr`}emz=Qn%t2gs}`RI2ZaCf;IMstYnfI0I>CT3ox^I!TsF;kBtTSF6Gj!k;z|3+y=$fj63mb%;`trOQ>JWCg3TV> zS#sizSI6>l6H$p%h!kMOTYaf{W@zY*zd6u5B%%T}F?s+HRT44o5?L(XgA}S0EgU{F z(GdL#EbTyV7=Uo9Hj1jVrago*nJ5A4b7By^>X)Y9Gza)*slx=ZlWybY+-MtT!#9KgN9~S`F(_ zaz|Xh&^HqMq(!uizs@>B+wKm1_b3YPMtC$73z}~A#1KzX;knG^wZf;2adhMlL$eAB zDl11O)5M+Mg~D2s-(?YunJJE!7cXHbcYXAB*c;e0^nl)L>C3osq5-aBmqJ8$i4*L5^dhk^d%c|m~iW~^j8KbsEwG8cX#5sps4jVUH@ZVX(6!GZD3_Df&tjidaBaev8=CD1gss*H$x zlyQ-Q=4Vggn7pOQ))wxF`pbX(EQPUrEPimMye29>jquvx|!;9lqs!7w|Q<~rGI)FZIcTnvShVR0R<+- zUKFTSdhxY1>%XbNd#|}YE^JuJy3o2s;Le-sGOs=6&2{B6HgS}GQHah3ZlP&S+O9B` zj+-5h`~G`uOlgFmSf#yGx_=Ve`6m(1xOBTAo1kCQWEp(m2=qemw{ke&>22=jV>of* z&E5H1TYu;OF7D`(+{@|qLv~%Mjsdc!!St#KVPno3#)bXMA+goSCLV*aUf8nw(56$L zfDpm*7J+H{7S|SYxP_l}EOO~CO362OcN#5xUgFse-G7O_<7nH4t(s-lGP=0%G~d99 zu||fI;6VnTD!CUsZa$u4#%5DEI}4<6l^n{Ak=aA)-n3ERMP-g=fo!adqymi_9isFY zFzur@8q7`M%xw34>94njNQnIQiTJyCw_S?cqif3C@We0i;`Kp!j$ zGI0BlKDtnQ&f|bVvpr+T?bl+~Tw^UW%{oTc=Vk#9&Q2Oh4o#o=30uGR49*FHTDN06 zp)}~Wl_&hh#{vYq!SwQh2tv_!x|4Q<^|<8=T2?T==2A<&r+b~$R!~_R#wN3ZV`A~; zppZR|Rfvu;Ry<%t{hGQqP}afOmhZ~xO{pc^&-G{HAM(VT>I3KB!Mka5_633j(e!A9 zr$~6Z6<94L9TwkZj*X|g+wSyS0kdTcTHm+;rw59HOqMauLB13qBcLbWb*RdF=DWr1 zvBQQ;;^W&3cKJCJO)V+$C$+s$jo-x!D!4&6tU9&IDjtx-#SkrPtm_l_R~Zc z{-;ctG5rXFJfYn|)Ha^~-tO_jLyck`381V!mE96U0+TIdTK65^F?orqJ;pNhE)lq# zaw&41egCz2hVp2vQbKK;&|qLCb+cw`W+#mAThhA`o}e`E&KJCGRDQ>!io6^%89 zX*XG`V%BlV8N!VsSIsEfE?Fp#EC0?C4k_J~=HZUGgZOiO!xE%%-m&b?i+DrPm2HH< z@*}pK@npzM9By!8hdIa8l;htIK4OPrefu!J#7wy`(2ypIoYPyiFAPzD$A;28Ld3N) zglQN<{-x!T&-iM;p zmB$Dw5H`?akp0*3NunM)RcHTPDuc;|E2C#PAUuL=F<65i{@)}-%wE@Xf>|J*jv54aJvi0Bsy2hLs26ZGx)zeJ|0_cGZ@5+bjn9?wydQ&yPOigappGQ8>nOjo}{ zLJ(~%W(~%aKeCl_#pGLJ3>@0)QW8gvDh_HJFeM>Y{!iME0a`Cr)R+eYO(=GDkas*E zkRMo^1|sp^T}7Fri>ydd%*vv#LE(Qo*k8<{ea3VKBb5glke-i$EYgkUFtbVcurtm| zk4z$9X-a%`9ak(wZ$mg*g!CR#sBVfgB`qw_RWPqeU3onao(|l%K3?rY^BGyL-Pja` zK$SgyC^U}iaNcTraL>OEg`C<3>#UFBFTXy!y=!-%!1FDDAbeFxL{S2^NefgABXdL? zruxla*udA;6#TDjW}2TaS%y2;&Jo7NlJyJ(rVWl2wv4}y-oPr|#=Z6}<>$PN<2uBU zSVNc8@i(y{!QqClcGSB?`9f1r%rf(*f%K|*^kRJ}3zO?hYvW%2z5EL$;sHWdSi+?^ z&fF9C0OX>a@z_PcE_`ME@X#vfC=);qazO)M)0g2cN_xbRt#~|Pi4$^O$td&3d{n+; zHG0+0a7f0#!Tew~pG~1K>Zg3YW~kxhzn`=)4r8?-$R(X;3C+N6l+G9O8a^Ml5Rb!4jpD8z3aL@ z&L+F4P9APe3C^?3Dh*75+;!i0@Xn}XEFZlcEmg$T$R9oKnQEU+?I=x_NxhdU+9^fvtE8E+`cfB5ksjY@bjSv=ze^59O3W=z0>zfsh3l6$O3%0o}R87{67Pxb=Yd*{8 zFHU}y+pbb8H0W61PxI|%_Q4m{Kv)k~*^HZzqcY%bdB&bpQIIiHZ%q1N15=*gqh;W~ zv)x3|s6vbpmD~xd+jAdrf$nf3D>vQ5STWqsvo$JH224QRZRtoo1Nj3jDa+cig37h4 zlB(>0NyOk$7-4klPxMH86$IWc2Nh^Bb*Ia$GV>Hg)BC}6v@ylm+jVjR9J+DfZ`LmB@O(izd@Cr;r3r>umz&(RF2ptkOEp~u& z&^;D|!82BaP7xjYfeBK8{vZV%Ma0sr>pOE3 zWsuBrbIy#{1`<+8Z;?Y`X@%c8m&ky$RXN0NI=4(-8o2_o6@_4$Ewp^LdnF1s61q|3 zi=Y7|*6111Womy}+$Y!aRe!zT@ck^2u^P;U)@&-s5IAfPNd$+|rc~ptoRD|qUe5&Yp+?f28ac3rI2q$^9p}zAcHA$V0WO<_l#>gmf1S|a%90N7 zN_CAOdD?HXzs9rl&8%st=}f{OQomOfk}CaxoK-RD&Yb{hmzjOW%(*{&9_x=5dB0!k z=YVKvn_o{7-?vP#E6<+ikAZw$c9ywgig6v{?Rc|^`A>?dCXU6DueNNQbv$TYiF$bw z0@_M?qK3cd7;U>~Ly;t$S>{i-o+;^pmcr4PkNtl31hJrv-#5bB|N{}izw4!mI_eo zt0b9{fg!lN@mbuqQt56TI!RvEgbVVNcBR2$WOWT0Na@O}c9kY52%B*-I0GhsHzDy= zEE4mx{OAkwr>U8}sKCfOKG`Mg$LLGg+`pqwDN02A@80I5B)cO^P9tvBTBm-5DInOV z#-j(D?@QNI+~vfZSx_O{UE4MmYYa=Yf!Db{0IeVw>5k5>Ov7~feUG&#VXfqswkDT8 z!(HFnNn|H(2MYBK+;Apr&|mO3Gk5$ENgp1dH66s8R$%FLr*T-k*Xy01k9NNeSgC_I zhgH(&g6d1rS@p@@q>G$-kQD;t`@j*~{j_Q!@k0VzT-|F-VyhSe!YI%`acM0%{dASG zfD1}p?6`aEvn7hq9twKyfk8LQIFf4aOS&>bo~&+#fyMbhi9z?dcfq5R2jm%H=aT(f zIy-Ro@JEYSIB?KOF9h3)UFr#ged$4ksj4B|`yi*v55wCP$r@48i?3qjPp(Bs$=!DG z_vj@#s1S@~%On9`tJaL3WCzLmxQiJlzzVjhW2YVg%g7V(o=w#}{+U9#4(vTFJ0CYj zg=o?P9sIpB92L2W4)RlCGV?tBpEzg7K@fiq0TOQWRw`8}1vIdIl(Dr6Tzzpj3a07t zwIXk_ewW8hJV6qi$f2*+0n5jn!=7~HO_@p5*U0YAO}6Wq`t#tg-se+6v0QfRKv>z$ z!tI0Rkw@bqu=rnVy}$2+FB)%8WpxfUQN2H%?6B_I*EYYi%zCDV3>|KhMLAjzYZSg5 zl<<4~zx4~~x-+K1k3)}c=AIs*8or+g-0pD|4wsm(FE5Bf<=z`Hg1qpa@9r#V#c{V; zuK;#zwx!%#2(HR$KL!n&;1M!XfnJj39VfMsowY3lQqw)z3O|xFybwqOFm_6wet#x&C9fy&+CRr}RDAM7hvM06*|hrUO0r z6aepBsqG?W16`S{GZ0;U&g@n&GA9?DELwY>qtDjEorxQT+Z@fRlW%RhmmTc|3Ac@c zo=GD>VgmP6JZ58yJ@?F}8vE99Jl|w$%h}@H^ziOy+#h>{%I+_=Zf<%xkxdbB-kr{h zE&Y~Eu_gD^QD!G6TV$e@z_tStU9EC<`O@PRMypm&t`a!YsD#B9JSIe+J{h--Y5HCQ zDD=H*=wMmwgigg{t@NYE!1ci5@(D5GBgC7p!Bcpr=g|y)icMDvoT-yNE^S9%VgAKj zFdusX0Ayg4$AY8=Ww9oaWBb-vR%LPMhUl*%a@`EXkA=>%s*a~FV8O;LlvihNJM2|2 zP!jb{Y8xfQ%g$e-9iji%yY75xa+=>GhC~A-mPX;8$nMI%cVhk>vZn^V9VOLj?h7w! zBlmhCkcEL0`8;jqUrcsbPYYAMo4Te;;G8iJrlQ@On6QR=Vri3=u@)?96=}&EmVize ziE8#ox2d|rSQLs7a3fBV;!apb5VIsprx_Phz{mXTH_ue>wCeMYLoYuV#?AOJo-3oe zb$RYeT@fe_-(ss4CrHUurOSk9xu9dCG zp@j|>^M^4n%TyLJcJ){xL*(mNx%6U&e7)OGXMm~tXc0Uz@Ha1gNLdwgX9az55O#@h zrQ&7b&ard;&@3J-n7x`#;MQ?B7FO?y7dO#n`BU};7I&)B;eEFF zwWCtt$j#v>upO~VLwR}aC(fQAjPdHYFivhvSbW5&2_j(0f#gqa@%&aaRgKbcB8NlJEnd}kz#RslD|9Z6 zIKU|)YwwQPW1>+|X2SBy`N?UPJO43%_@#3p+|{rTnB6yE*YH8NF0v&B?fRok&6+4D z?F9d_b=FTHZz$)%-CR~?SzM3N(*Sm;pI3#k|2&weGUdoJdvebq#4wB=8&YqlUdTD6 zTzgie%2RNP+RGg~s$7DDfIKLmzak2&UXm5AImBms_*2GvoCwEY4A$r)4ze0bppwH_lk9!5*R{#iAI(;v6N_En;Iqp`26u z)NRE z@*|IwAsY8TaxwlQ{eBsL;wH$IfHoOabGknSidMUbExQb|%Ng6sEdQHhMhR7SES{`j z#cRN?hK}T+bIhJ5;cqr39GGo8N2R-yE>d#jgw)8SLRl#A>+me`mPNAR|vP5`;@y z#*4K`@1)n=Vq%_RVi2FU3CYai0usZ*VH5t${(b#DE81Ra${gwN9AnUXUjjx#%L88u zj2zebb-#R4zofv!=oUFfj)O@a0C|a|3grQi=Q0k^q}_wa%w&U6W(h|LLuR!ywQA** zB<_*EnZmGsL(7A?Txkc9bF(}NQ~Kel=F>Zb&1N;iOwMG;1_n8UoY9&Q7%bf6(sdr^ z|EHfJF_pwGf(R&qGLCFK9<|YaYf3VCVO2vtVA&@ViKksa1{7P`Ho_x`BbAOR!zj~B zU?>{-@x03Ldsdbe!t|JeUBq9cN~j5*s*=TEzyXmTT(yM7EIS%fJ=GKatXNd&;AGJ> zL8C4`K#jP?-o0rTT=(D)pmaGmj*KvoFYiJQl9e(f#0a<)mlejo@l}Z|Hf68z>(yLl zwEgKfAVexTnkKN8ntui$cD7=?1m&_lXDKa?so5Ows;vmHjzCAfV{vC1`Jm#BleV6% zU{KPw=M5Ifuk707G`*0P90WH3j{*#n%(VGY!#LBB7l<)^lEm;|N)ZB1493otTZ(E| z+dm)jR)8A;J+q!Y49(;-6qqs`+K2y4yTEsktO%S*>^XlObwWfBqJRY|o1k$koi@zM z>Jguho_3*pPKK?c)jP~P6oJacxvr)G%Ef$-59zM6iu=XlWV`Bi?Ev9CJr^O{)}1D} zWoi5N)Vc*?MV`#(!-oVtBuh>zH6FBsQ`PyC0pL$}D|L!QX?<;Xx$g1v6-0nk=7{qO zs#l?~0`z{%3jhmgpjV_5f#J>|!Mp)HD zs1vB2t44(BQeO=Ha4Kfpy9d#%O22^ zZ-*U>JD*7Zf1Z<^9ZBYYgPX4GNTIIm$StnyD5aVI$#oLXp=kZ@<$9v%(0Zbne~SI5 zxR!dN_z~Ct;4iQ#N1RtlKtP&eKtM$Q8-J0P-w96o|KcyUQsT9Maz=^0p!W9m(>LCJ zL|^31q{-y27lK)zQejAcLv!IHO-pvqmUWI(>$Qrsy&i09U~wjd{|){Ty~eYf5>iWm z`T~i4lfq+q7A)uh04yi8N>QCXRxR_yh>?9F;J4eY56QhSI&0H9@q}Da6`xyRMD<&# zhV*fHT6!>56je31WpqsL%(dEz(>UbDW?Mt20%Yz z5#7>yvX~$&5xt~>o-G!B_A&J)HJ(Mxt{^?j?5E+czPv~bfpkY6DN$5Maz>T-N>Sah zKhUXD2MYzXAy?spvKDnZYb?}ZFhi3dDs5cy=9rEzHl%Pmy*Y7zGgE)3#R~(0-~hp% zSHVu&1=Ug&2@(dU4^uIKfp%eM)F0W8ge8-L0%83*e)4eX(s7v>JVm;=f!WTs@7|Z9 zq+lZ&-^WpU3;Kr~5!T3}ziM>D6T2XT=h75_0RWRB*Ujd`Gb;lA=vI4NekaG(VI$^b zuyBdItZM&o36$>O+y2j{$s~9q034=Pf*7w1dDKfPs0=3eFQbIs94#Y}6(ct{{r-ZY zv?Oc z;g^0U7KjNz?-k1%yY%>TBT@C-4xdKvgmAh6Q_@2A6;GnMa(+%l;- z{nOSof-uGgD{tSe7FHPX6G7e(+m{pSkFiqYwtIVp)^Dg-FMV#*wkQ{iw)fZN##(Ct z^nzuBZ-}j?cdBnjJPz=yf~X7BFienCQ6zwc#RJG-EEYOcj+QupqFNMe*#u*0Q>9|s z_2U+@l!dYC{1^bG*7tpiE=my?n#@N^-z`s)DD(*%gEmS}=h0R7!rG-j)gs_;lT_wx zVzk10ePRg0D(<0!ByKgEO=s~xK1P$D5q&GSJ@gau2&*@J9 z1@+e!WVy``^qrL;@ERjBUHW~LL{>cAC`ByUNo0d0#fT?=!G{;EMrqL4>ap;H^<0n z=ZK?+?^SeY-;Q0dTnuQvaEl#Fm0lG$i<`X#zdgCZ-%-=?iP2VZnf?`kn%2;O(c%Y8 zOBg7{hZAbJcMkY;pW-72Sr?R=PAh`k!kh-l9vP9V?T`zaWs^#8+9&8St4tA-yPy)z8va}-Nvc~K0OfB8ONu%e;Pvy+47EF!(vFoK zlYeWQ&<>E4{=_&q?8F`{Lm{}<1b0f-SM$*H;1=T0u8G+3j)DPoI0?kH1f+uV=S{_+2Qq zJ6RFWH}dkaLkf1j?v(Pe|HfgmIuXr5K6w<64JLrj-mP-nZkd&!WK*B($Ez+O3G+py; zQcA#j|IkG8-~Mk1xVv$-_U?3SOhZ;<-=T?w^4;S8Y31aV+3H;4Xd=?*xpt9^J07** zFm?7GYnAxCA+b=P>t2gcN(3oub;6z^afOCgyNJ%svLl>*%@X*@edeOg;wE5ua3(kS zb4${@x4$VAjq7-FhJKqq2yQ~$x4(4_`fLNy!fBf6CA~5NusVFuXX?yaz$lssI1+~j852qNa*h1d*>di} zdRrA+P3pa;rJKc*Nt2@a2j<_W7QejCDbmQxmFNtgk= zubv{2;>D}9)jE;Qg&*V$DZP(I#=+)7mXRhdOjP@v|iw zTQ-FC^UDj?`g2yAE;ZYbFkMJJ&<6%YflOVke%~Fw53AnNQeB~;ao`vETfYFdU1FM0 z#aCKF*U*qR`Z&}s_~3SjdK}Q|6^$_1o6ef3>vgA)KJ5)R_bsN%d!-7pzhri!d$LOf zD-zVsOuAaRiy(j;IJ{sFush$XaHHHzud(%vVG!z)guvFguSwCk_lHz#S$H5CA%spp z+ndyUXo3ptdtKJk2$X5seA%WkSoTC6T#reu%8kDTawZ?VAk=mwPMz-e-VzSrV+!jg zLR`n=6jIdQ%E>b~lHQm1KT;m?ZlM((SlJiuqXZv~HBn)(I6at)#DBlbAK*aILDlryS4d99&sbj5#zykx4hUs* z;zEK4(-%)A%#Tk4?)YwEy7XZE^op1K3BHTR?lm_a@^O9O!L+2e^5N8mJ&fkCs(I=s z%ly@_V8ux=nH1%Az5K?zW$j~&_)^0us?N}51RJM z0`9{j`HSzGvI_rHlt0bDH+asHPrg=S{E=lywRYu1u27~On0!U+Eb;PVK^-k?UNb(^ zU8Px-Ij%O^rYRod&aUK|0(F$V-S`K6b9pv>64e*KG;_paax{q0a|>2 z@$GGtjgGKzG+YcXLh175x69oH`Hgt4n@x9K<%p6SVh%^LKbTFaaWkv$D)@(#pF|Se zlZS;TUgA7EFuB|Y_ti#wsB?}mmf1f`NjSRmuy^*RkC2CQQ`s~^0dc$Td3Mt5fZ}o< ztlYBBnj!-lcI|)!J-qvHs5&cMb0k~+6KW*75hc&2fUURcG0u1M%zsC*gkHAQbF@M~ zAU6$@pgzN(cARZF)Y}*W7I-uWsDyT8RX?UF{kf#y!de&)1O>+Ds9ycL=WbOiSH=k0 z$xR=sV|qX52JbF2wCS*uwjgjqB7Ag{)i8{nFatz%3T)%Z-?t56gyAd}8aYT|xWzn| z`J7)+FeUV9rZ>o{)ohSOc@h?q<&2g^ME-kMQQw?QOd8^3=v+noVn7Lco6Nu1{Y+Ke zeEEIpa6-N_wt>ZwZX{3E$>o!~qw)g@rh<$I2>`?h0$%dm@K^+H9mq%<=7htor)zVq zAu z+1R-?sjo?^mMbEPJ@5wPC0+`dwzOUfdrxgl1R18;zkw8d2jHuuvUZ`|Pf5A8>eS3G zGz@j8Lt?2vEdpW}Gf6l+PS9;0e3KLAs_5@4S2>^&+PFSefV{On)-!ndl*xRh8ip2P zi1)>1r3BCku2YXB)Sg8Q6`Nt{Z+w5j+-A~lEcuJ_k~ZWjp!D<7fNqQEHG-#VgsO_3+AHmnWT&rhCX?>ayGM%){Q zUzd~DZ|=!RdTkxB?b>*1CyO@N*J>pUL?F1YtY^(40Y_uXI%@`>yN)6wo5DL`>NcIy&iIapDCp z$3Sei4%q{CD*(W$7g5C#8`si=y&tTMjOd_Md57$@E;#0LI1fe6w@1vZCoyJ%$k3MN z5p`2{Z;#dS5B~Cr9`1}dekcxW_hzKD?nW%T#%4=Q~HC;5$mb+F+H!A%eIB0(9BkO zuGU1ZQkKOCt`juii@!01$zP;jen-ZYaJ2lFVBm|MC&>#=6?c$5oQ>2(N5k<;!`B@fBd|-limD&;aNJ*?)dX_Toc0HbVxnZaeaPQ9Bd5fe@*!b7HQUn7({#r8 zTDFgg+h=J|xeT+-915U3s7DsP|!q|7qGS1s|%PWmYg96Hvl)>%oF`qI0LW`wt z`9(XzW1K+iHbaf0N~s_DA%KlIQ-dTi@$KL}R1326l%&f!tE+kiDH$GI z*C9H88|0vMA)@48$Hcgo9Z1bcYq;X3_|2wg7R>*Q>NnwyPlXH{EQ?9NzNo4l^7CX%qWdyIL_eIj6`*bEZp z;RYqnt~jD2lnMg^YC7yezH`8fuolS_>ZU|RX~}jm%$?DA4Y(jIKBQteSb#z(b@KJ3 zNtOU|R6bS>i%emMg7KA-sB%e$af#BcqFI(CUKNuhklY;@w{)kG=;dX$VxCVZy~mP< zb^JA%edd&&Hz89W?G)nXDgcVLnNxTK%YtAO7_fQx27pENsWP>bdoRpAFmLY1ni*G+217&kIx76Ifdpw`Tqb% zT%SItg1NLH!rGZby7+Z+?>@vLlvxmSixW*6-3Yt1bFXR1*PXAfaZ=m8=u-r|+^92S z;nIgk0~0L9L|T+`QfR7Wz1{GEGmqZu&AB_3pE?Z(&fR57Ufu8)asF+Eb(mstqalIi z)7~Ca0!-ud;<#~ZWCf1Oyf-(E5}G!&Ze(}?QyL8dNx@b5g^_4(X}z-Km3*ZME{Z|V zv~DrnUW4zn`5Qol^sfgWeOQNeMlmu8DIbLZ%_Aa;SD5K&fOw>?gIpBXtU2ai3P!u- zDgb9Sa4h3Sm>Yv3ig$mcT*`SLb65!){HzIe0(zE3+J}eDZh0<8)Z+pw4j}NJTF@A- zEiV`c>ZqoJTj+v~K035|n@|il^J#t8{nGfiwq3bMju;R_fRj5$jO#lmpP;XdBMaqg zwi{#j|B8^dGBGaSX+rz3E9zN=I#qTIx;WZ3;=6-hIMjT)Kp;D&Ubtn)Y<6%*7#%cm z0rJpeD^^^pNFqo40m+e(?r34v)X>oQ>J%aP16Fm=#~q#d8F33+y}&<&*B{kM@tw~A zpZLH+MlIvD$B_QOb}$`>80{zN%|?EMaFRhc6E)g;!9l{MAKU$UYM32lbAF1L_f9)U z{OJ$lhhM|4J93=`+EyPh45X!zsA0@0fbj*Z9>GS&s&#Ok?QT0iJS_VFnW`i+HE)QZV`J3j}mS6Rm^A#FuAC;XG^*WfLejQe9maJQ$=4kfzLR&@<-b?Ri2;r3kRS2o)3 zZR`-C$=r|ITEr#xeAyyPk)3ESfNPgA!WJu}?#}XZzbAFcD8AaX6Ljwto>(cOFheEH^OgTSd6`M+fP!f-ps6x)Fc9;# zO=}`pXpGwsS|U@xgIMK-brtPBU)c)Id_ms|eK7N89n#Wy%abM6NE$3+~2$wnzCVXhTTGVR8Jrb4GA9={`&QHxl3okYeJK8J%6t+ zZGpdXen6MJ>L!j)M5ko-fa%KF4vxo_uL1o^9db1-g|qx|CQR#_79@urUd?SpMt7{+ zCk|#PL~xS&1%eIR)tgowR4y8N?vY)`@}915KRY?oIydHvu)@1vopza=n5m^d^f%J( zo9uzNKfJ@jg2?MgmO4&&JmE@he=eXDV@GC`I73CHnn~9%z?FLx0br-rkI3&hzRCPj z?!VohiPBK;mmuRoY;h7g@tn-v%}-u?&Yk)MhMPbP#ZD#cJMi&0ZoHRv7Y)D3e~rz9 zWF&rYp`})4^Lzd(bUPh(C9{TG0@P$*T zG(-Umh4jdM5KG%V0iPv*vPz{Yw*xs}zSI4ZXaZP=FYX3fJEc;-nJ(N~L^Os2p5~G8 z8iFh;J(e8P_NjqNb#%8pvYl%04}_@Q&NX;A|||q%l(6 zcG?q}6bVtt1o#({yS+>-fLloWQ@MR(qG`|s0eK9osm_x+(6e^WxyKC08)4>?GX z`{x}`k6{n=TXIrVZ(*63Sj!T!r#d74y~9koSGb|lqs;q~;(5Sl5PX19HfQGZIH9m; zTc_Ph$xfmK-kS-@g(Rdwe_l*h(g&S7B>G)xvUjUA4_F^5rB?B4A9Kt$j;&eUq&N=? zZ;ZuGidTJm@Da5Vj%!MMwa|c`;vYm(JW%POlL9@y(!MoOLj_M$tF2t1KwhU5Uv(Km z2nzUeBWKCmk}X9$AR`#1dSAnumrYtHE+vnX!S;En!y|NMYLs!G4CR#C0u~#;L{1TL zCC9AM0pz>Vu!z$aCsaDJu}g~#8Ut8CKL{YzCn73pxdjy?JGZ4EWNRRP0Ru%vSu>NM zQM)$@?WO{>bZtVX75xaizBruVRo?uY9a!tqtuA~ottk2PU2au{GnQ&`8_c57mL)*X z&`QsO^uBpMC&l9i7JvHOa86sZdI9xk6Z`4d;c4<6$T+-9+sH=^|v-fgoG ztvSU8L!A~ORRv!8Q9}q7Zu=@P{%10cg%OlraHrEf8kk=OFN<(fbyj`n&=R8xB#tK!F`=f1&@2w5Ij^d1JzmhUh9>UPwD%Ufn zp6kzS4SogTqlO5&eK>o0FCI;?-?WRVOd(4o!b}6Ml*S1swzw#NvQdZJikP7 z-;`CLlu@9y0HiPO@#YhSzRL_&gVt;r{1gIq_LH=?>%KDpv8XHjNKIT~y8ke}VLJ=R z8QkwLWMM|MP~2gMy+!HxY%J)`7dO0K>C(88Y%wY_b`IK%l0x# z9$xDVEk&XJWEb@ERnwqhdyQ~Pdjx3c8l6;4C75ZMyn|f+biXr<;X;N(%Qf$*oP(F^ z6cNkEGzz}_0kGI-G45h-0Qa$ppap#}Tu?1()o&G>g1I;}GLA6;cia13EJjX4exyC^ z=UWJu8Mg+`qSbshAKy0$o*KvtSO}b_bnlv~=lk~LUn|MeXpF@j2z-NP=JEwuFe(+2 z*$~TY`wIk{YD^jB5~nm0IR`*xQkn+&tCzmlg+x{>zT`-qq4K)UfC8M0;b#u{4_vDs zs8=nIUm{m$)R=)rX5(?W3Sl0>IKnX@*{^SIPmSa9t&*Q1f{e}BoCf0#k% zco?ZT8D%mwz7`m50XnRvn%YAc)or7!iW0kS;}xM^9k!#@-*mewgqNf2(FBc`N;aXR zJvJCYSc6#Fjd0xcWyW}KreCoJ2P>o z1!0zuGcLq6&I7d@!P9yK%*SHdNSb2F+0G1=6G4LEAn2SHs|u@7dAA=*JYg&vj_Y(D zJEc)p#qWC-0EDSs?Z_lVv|lMW4*7Mow!=Bp(=^_32M?ByjI4{$*@{wP%r9zICKzK^ z>F|oX?wXm{yJge5aCIG@+0E zRBMqg5+DjUfKxI#U@8j@Q!V}Nd0ZsKbKBVLw10M&1DL*vR`!VzVzcupR!wRASrU+W zmpO5I(NgA9RBLjAJ|SrRu{&{|!-hBJk_h@ErG>2HeZwt8ql=+)@14=QCO~{O2qDjq<_Pl{aqCT^tA(JU&E|!8TETujXp9li-eC39 zO_P^ae7|~}WUufI`k$^d%s+$#5YT_N2z9uX;-+<5VG#YdxBL$Q@xQo-G+=9t|A9J$ zWws~$Q{q1*{ZsNkrTkNBYi4`eulWCRpFMO`_z6Mk-5=$}HhmD?J6lmLb@%fjw7t6T z>?;eq3OYE@|7o))lvgA7&5-JPf8uzM9Q8b>;-TfYkGe6#dj0npVHR!^M z9&$9bMF7%}Wzw+$KfGj0-qeYfNY;*orl>+D)+rFuQ69l(w@l%Utk2F2QH$;2`?Bx{ zukx)!(`0kTXvN$#xsl%3e=Q4i@-7@YeN?Gu|%;)C25aImd80|23u*sJw%r+I#u=oQi<+CcvppkdJP+;ne-Eb zo%da%L*=83X*yDiPy85JRC-_Qe~G004g_$78C`C#_Xb&@0oG8sv%+AeLl-)6gNWlZ zF~c0aNdENxE&JiVstfjIhMPa7VTx8U*fXJcqjZ5Ih`4%Ix+&aougfgJF`NGAn4K6<~h?hzge|h+=M^mc`|{@Fvf0bPjy)oL~Exg-MnW9&Wh&)IZ&~%lENJR2R+3 z%iM%9#3&=2C^i$}2<+hY`1DA0>#Jk2$lf9KPDrfI5h6;1RhTjN)lXMEqE|Jj38z>g zQ>i~GljheTF%>5##HPG!o&(cdu#%r;00PU;s=yG_%4E~^0RaQuTO^_1!3J>R3g9hY z*ZmEb`1pMIOY-u!wMWBB5Y(ATeJ^Tc*A&!ND4V7uQ-#02)TKSWb+R)MGWpXOGWWmC z#P9_8HfN0MSgz2dGbF_F*TP3TqgjWrSAudc+5?|ug0lMsmJhZsQU^f0 zTD&A*A-7m|)t@V~Qa^H-$?0)YWR0LQDjGi~EPF%7O?Zxs_zQ}Nq*T?PtISSXQ*+{% z^}r5Vi{ZY!5W1z7X^EhN_~H^a4Z2ruDgZg?@P+87040QWVLMF1r1*YIh@~VEMt+rz z3YFea`_l(KW_f75w3Tjy)n^oo9EGwF2)rdWuNwY=9&_r32ZEKwclxA?h6KcJ&I>tQh~DwVSrqpr?=2+^Lwo_;89b*CGh1M{a1IjxSy#PD;Pv1{@A zk3@HZ4|nKbE%yR5Qjr9A5~Lj6F5llyf!UZ-_R9RQ5)yv>?i%}l$(I4t8H&&gkQC84ta^;c-#-N+`%HOSnq*Ee<csAnB$YjsZtOYaij7v7l#BusoaL?y$PtzPCKwRPbZY-% zGwr95X;7&*&0x#uJ$e_k<;B5%iwBiv?H@u3;x!(T6+#A;V=}_eaCDKQ%rva2Y}ijc+>+^_4j3d2H-A1OJ=Q^m-*rLBvgRaI6$XM^boG$MahnK zJ9i421 zknzu4HNxw0SI$5<-RAcc`3VL}aB#H=I+gsza>l%L0g@R9>&>PaUT~2$2mrunE;!Td6OxEyBi& z^e}90ZYl@zu@Uh4qpk`rlWJ<&LYFI1M$6^#GLh0TmUH z2;W{101s3PG45zOL&uTR-q2xn*PCrGa&JqBC2Mp*>A+OT5$_nHaUAJF?|4Jgd4jq; znlzGANon!6>XebBQ-1W{uxs`q=(cgCKuM^dJjboNfzeW;`}8!po>t4e3JyT?(&%V- z%YP8YMONS6*+8lqG-^T<;S|f9f!qnZOFw7>v}xBpT4F210jrYTm@-kWSGY=bh#ksX zgkjBUh(Jq|aHgBw`w7Af7qVE$WzXQkC#6Aw5;+vz<^B55&k$^IOVpRqy+u}S!Hc|) z_cCIwt@etb;*2GtW0O!b4IA}AQzP$WRijiq2VM&&+W)eamcuNQ0qD)L5n|a)W%(aYFDyN*Ji-;w#OwmvMRi6+c*hdK*q@``&vR)^XTTI4G9l zX;4r7pmY3v)I#f>nsTF291ZELa(T@OXvc=b^JK)|4vL9QvK2eR!SbkPZmL~mD@i`n zH2o6~L+vctxuyt)rQYYHMu1@CuJl(_j3zs6A&TH^!iL}~H`11oSVmH6f9`i{wDJ{6 zW{30o+B*-t%EUQM#dwr{jM}82k+DGr8PtB7rs7|oXrCa4_^RHBnr`g+49)FpLxKn4ZPQRADwq=E%hCa<`8e5z zF`Fv={nXH%s#s#&nsiH<0=#eZfQQj!^NA$J)E?*)xo-h^uf) zJDvCey6hxHzN^n2&0g@4YfB*3!c8ZS#sI8VslkmgoGJx$w=i?lNdPttKzAO#kx6ab z?^sayn(%G(BS)g^`6*!NPuoWNpljs=Cd!)sCuW{Jta%E@aexIKZ_TBJquzb<;a*1# zzHCKqE1du7sVRGnwWZ3&EC%djYr%ZKA&PZ}*Me0Gzs=POTbs&*#Ll2>iPaO*)JA42 z4{qvN`SbT(zeN%C=y&ENV0^+GB`bmZ@pEv{&0R!!m0V?IXI2w`KoR=MLZ2YyPAON+ z!P!If<`jsE zsS&%)ZPv{$>R>QNJ}DJ)E5MOzp`}FBW+U`X#4XmKNQRxEYkW3m;S6um&-?TXo2=Gs zt@JG0w|$4pD>!@fW$WFvzih{{5_e+F1C>BF`H^s4*P|#!Ta|~p$K3IIXM#$V#Awjy zc~jav*&*oVJ}lAzpyMr+eczAB&{_)%oobAXON3c4=|rS{$*j@SsEG4`lvc+4KrX5* zFX0B!eZ-h+i}Sk==^fI~yeKL7`+e&uvdZmZNo(DXJjP}Q?5%cYpOMBIJvBcWx7C8k zIsStp-(@y>*k2o=ZIC1N23?IKbeRE4GAhV|FQ5p0Mq`Hyu)AQV#)R|)(XMLfeF8zC zqf|JdrGc*idlaBVg6i(==~oqXu%Ae&4XpMqcyyzH6Eh2Y5te&`Pe8z5sI3tA6gIFq zcVX@AyY4wPQN8l`#iqX#v{~?@fKVteaR`yZ_PDIg=D$VQ`vZ$Ru^4`~+L{wK_civ?C&_FoQflyGtwa8z-3WZE z=E}DhVL376?p9^}tu^jP%N2N{U47Ie%xBaSd0Uxchg&U(sJUruVr=o!~BQkpuXi{br0s0_Z8iAifXUb%_{zzSPsZug&cnt<(hDwt?f>}=Z#zN0L3%_@WI!@y)l zk)m`^Y7T}N0wpBzTTF@tPH#6vG>oT>?*IA{>3K9NC~h@P}S= z`m~*shRulJsv*t3Go{uXaYpE5a0-D(4Zs2%zB%>rQ)b9s-2FkTrjbb@y9Acj*i*F8 zo!}g>JK%($HPS~mhrvY9DVxBXV>|#q`27NOWFk~kfXh2$2->6zVr-JW6PA^eljg5F zL;kCpCbd#faG~+EEQz1CT0owGjj@x52k2V#u+3)Z;|!V?LR-~{o@J=}y_t_DQWhTY z@@%NEvy$v+&&q(jjO^KHcdnR+`pcVqkdEpM*TaGA1hCX zUthdL$uoFfjt&aOZj_jV|KSS&ytz!Z5?(St{Ol^-NU(^`kZMtHf7u+ov83D;gRBT! zln%cfoM;*w=){w9#THkC6~lU@A`taA*EgO9{kxr4Zu5u!V8IT!HHD3@@f|{9fMS69O7jxo z;8+j1&cF5kDx8(yIU>RAO+zV81q$7vUAWyVZ(X?5P$KGVU|zH(XxU_hsEI|@l3WY_ zL8`&``8CQFF1zV&2C?m(TfzkszG(tKG9Suf2-qW%v8!rYn}1A8!3U8d zS~_0!rgMIpISs=Cuc$?;ED@84Q#V2H1>nq0L^1`L&h?($n8Lpdh2`5N%%J`Ud@Is? zg>fVXDv(7GE{kRQ^$LmrEXH3El|XKWj@ScBD(QL!T?G|aQ251Ki%}4&{#{fp|FA$m z)WT7nW%$AGAf=gwU%yDOagHiC>xgqQ%-PNJa=j{0W!cnbPS9Jwn#!he9ie^H!>6Jj zMMeu@1yOJujVdBtu4ldmsSD2)h>GyQA5Q%8zSw}PSTA+dfc|Y!GsY?wTP6Py-BiVy zxAXk5(64E7gPT})815VIJV3OKTA0fHH7GM)2-$M)WxgIGK3rxIF)m*6T6YEBj$Sp~ z93IcnKRnGrpt^mIV1teTdO@dwB9+8HbjbQzC86J*AzXw+4nH=4pc94zd?Kd)?yBXDUpuo! zvmuoBMYhI6Asg*POd0KB_acc9VhIMO<*j)aB^Uh5LpxezxL`xGGEu0bqB2SrVi<-1 zNdc*D=bmyZ<@cK)zqw*|AT(<-VhQ<0910&>1`Y-iB9EQ{K*Sg~?&=u{64SurGQg7M zz!>_PU-sV#7a4tn(EMu^uA1*0VLGyc7Yu5xD|%YLVG;5x#!+ySBCNv^@H_8}2w6YN zr^H*PDVxlgdnWe1%X8e7{Lv4~5_np_tq1_kxue$JSFr_}4A$)HYUBlVJ2G42dKgh4 zx>oVyyH4(agRPD_aaJhPPuPr9YW>EpRQ#O#0U9p?<*}e9$6R`Uf46n2384&D-dxw~ z^JXyQ_Od6bjHKQ5AY%U{CSDbet7lEZjB?mD+P(tAUQy#I;KX2q(=izT#wO=%U4Jm? zhhH8K7Fkrtq>WaKz`qiMAYAst>1Nn#!_|m6s^1*|`ig{zYqJ%hC*~sVX!|G$0o2jI z#u{^Sf>?E!Q^sgP$pfEnd2!2l!`%J`%*zvRJ&dWbD(ifqcnFA6UKnoiBj}8>Istc(E&c2K{B0SXnKXnG0GAEU3isWl z^>Ql>2pw7p#U3t!k_k2kL-17Y+boVgkQZP61{r1*a1Eg}cyRU?lgR9{k_XPgl{?)` z0!*0x)~Wx#sUE1%0IAn#{mXN6C<#N_tWTOhb~xZ!L|89$Pysu>>D79^?dU$TDrc3~ ztW6erOMl#xiun)#rq#Ob;=_(TduWkL?J@-k2rR8lHWv2qjOK|6r zJVQ`4iEQ+!WXLYSOcp&lyfIG|@mUu$?#3-usx^d3?97KFw_ zE}{oQQ{!@cv_T-ZPKoB_@E^&k#d#e1i9UcGZ2tH4UEfFCvW=v6I0Zp`4HKlgl8w5PH*| z)ea$|+Ua-wuggvl z11I_Z%{vrSRWzT&T{U`jW?6k2zrxH$mrvqvcv`zvXsN3Ws_Snmm$n0G2RTsRW3TD~Rj2>u-NYFYXTDMcUuY z*#$C%s2=3-=_U}Y7?^E0$?A9Z9ulgi<`nr_AQQX&YcQyz;U~`7C_@D608liPP&ZCP z9OD_r#s*|ClRiJ-ez?Ohde1ujOPh;Fx>}nneXmH{CHZ%s?l^n5BYwCMZxd zFOq-?vJid03L`NrH_NA#zq2(IgARcg7bfGw+R-FDH1{H&xt0Kuw{oISG3O4xEpjd+kyuYAhzYbA{#l` zV$*=e(`l;%-!&IvQPbf}vtqEWB279A6+Zdp$P$JkYhYb1;%8Mh^^2CVenx_+){(HyN6X~L-uTa!@l8Ar(B7n+9>^P=W1m*`LqM#m@( zBI8*J22RXwu5%H8%)$*GfE#e9sLAR=u>$oSkUE<0H8|f*#1I#)%Uu(lnQpw>l%$*Y zH7;hTe1ShtwvqJ-)d!7QHz^y2$jmvV?+DMXbrAc#^a}GC0K~eq=wAy{Zxuk{khNxy zS3+CaIkpSDxK-!3ery$8g%HstBJQHn*dKK^bxb7^o>}+qq^wx9cCV4O)mo9Q(tyK< zhM-1p`sk`&AZobWKV;W{mm6gp6LVX-F>uoV#+sT+_yI2ne(S?Nv6Oi3VVRM{+4@4MH<*YzArbnw>z3V_2othwpOcu|VaXm>!{k8f+&Y*?7_*aU zbp|$KQrN}+nv;9{dx(v(j?a4B-gV{`WN3-na+ov%0mU(LA%D2Hp5cp1E{dS4l~fO7 za18^fNpQd(x$J@~G*kSrRUOOL$@bmuc|4n&2XYGGSr77jRSJFOxbtgB`)VW}Q1|$L zW#MQi(0@@kfbnR~N=H-2A}tLne&qk!3Qn<=4m16!4Gv=1n|Sr8YBT}%ywXCrKTh>lS;2Zo-T=@$fiCMr{0Zbz&YA}gkbsY9m zJw|Qc98uU5gKPJ?ofz7&Rc}_|)1DyT>;1|5E)7*EI(g?aWr8p>PB;wrJ?917 z*%wTD>n5xiM(U6fJirZsw|f_qRq$|(=b$XZq5`TegVjh+$v;@E*H*#aQ*mo+JS({^ zxnM+YOXPo-dyHnsSRkAe4DKD})SvGJ3Sh_tscYkxx@%UcT~C_=sHyX7(7zuSCr~}I z|4HY{`_+u7mDdSl^b(!CtA$bYhZJ*NmF+~BK?XN|Pu5%JH~LOUF(a2;&QD2Sb*ny@ zC7Wzhi9Bc~Ssp#%TCRUu>rZ{y69S~+FowmAu?`I~{?4KSkn$5VD-v5{0c)u62|zZD zyI55+f6&RJuL~|M9*c)2Y4jyGTs#%{tjhv z^#q5UXDOHnFATLehlC{Auc|D%6(LBt)z_9Y#y|6*+r?s(@B8X;ao-FY7XzO)rp|8Ds|vy z9L#AB9xX9i(@^XzS5VIp0ZhPvz(aPWtt_=pqPmP!t5w0^R29wo4}k;f&}EQGJ49C$$eqgehvWiiN4f+*pxS05Azloeg;_Y}OH# zMPgyQv*7-STLmHC&9@Ab+I^i?NwW%6YLI1-=DeDzQ1fB&+D$fYPW0PfZ7&|Uw~N{tLLp*ir;_H{!3y}$r{R*H8?j+nTFhG|-4pDd6Y6a#gl9cFtQIqf9V z(nE~|vf;rx<4WzS0N`xTr5lchiG)63%3lP0gEwEEJ;zEqx5SZa*+aTG_~PHTmkmNs zc+bKWLoVqdUD4pjEmBWUd?Un&Ef23Wk%4|aN#7646cC2=44G1&MTR<%yPz~t$2Vc7 zau!Q+-!$fLvXn!)h?O9j$h9?}U?oWR$_#9oN_Y|fE-(!O15|zBE@3rAEqxzbsW`Yv z^?7;w>~OZN@UL75d*N|~j+LLc&Staab>y*~Ty+OC$!+c04EhAPLv{!f^uR@;6*yMa z@RW5%`ExEyz{=oS@ScrRI@g{HLq;((9{XdWZL*i1#e;7$lo}FQ-Be=MpV0b43f=gL z7+yS2*)0j00hn@$LB4wDMVv#DV+K5v+f=%n2iY6I0T1dol#JhfHjJ;3^{WXx0QPXWg zA-2T|k>cK0*^4_7^USo(a%TO=F+GZ5CrQ=|&Sx{#0N9RE4#K$OqV=jRm1-}xz)0Yr z`Z1`H-B1~7Eh{7*X)(pSu~c=F(_q6LFvrT$&Aixve-gfwcR-!gc2na{7D$KD%C{7{ zkfJ&xWFDZH-G-=|O}kkXbX<)TY8q}w9_@TACm7`6sY~s#86C5qE!uYsyMj6R0yAE8 zfSgB?02=8Oag;Klay$;9Ha7V%`v2%ogN2c*`kZk6x3xdlo$c2K7`w+{x;Jkgtgw9G6$Y~d3*1@nhr`cRm9Pz}>Bq7a+7LCrf1^k>7!+2>z1Fc~z!Q(ZD%mr7=9^NYtaX8)*3Himb@#d9XYgC}4w(kzN z+Qc8$jt+D1B6o_#0SS-v`uz4`8ASTk3OH!Y+hW#eEf(c!J{;P`)#`2;QWLvqP)t-_ zBNMZeJxOM*EMXs)l!APK5kT}9l6sKc*5Dh;U-n}J3pXH>xj@EH6{sXloEo>S#Z#N( z_&mL~Yqz_d!lIZBdrz`^<+Z@mt$zVkqo*h+|E=70_)_j=XII8D?z60WG1u~Y3t(&v zcGdO^a3Q_lFU5eB55>*b4Z z*5W3X7;)j~H_zijiFZPrNVK650ARP#$1&cGxS8_=lzIMmaZ$Q(@fXK#@#ENy^V^mfAp|fhjCHV$Y zN{N)2g3BXEY8Nhvtu3>w8-V@EkAdwSqQ7&mCGw(2DMqxjSxy4&`^PS@&XuKl`aHcb zfx0V5nP|x!HC-Sf0BkwVDWWucL@F63#b$3=i)*A-5Ft`VDzxgL*S;ckP-#0Nf4gZf zr~e4--R`+1(=loLbKkvAyn(n<91z3l?GIxQ+O*Kpfv}>Blb>KA?OuKw=z{8*KZCz~ ze6vb<$S&$(@~3jDr5%2=%q52jox=EQ-@p!VaY`~h`_WeO_#*fI+gQ(Y^`fKVx#a0j z=4!v2@(6uasT^wpJHYJ@jD!j!`R4dGRis)@` zO<%$v71D`aG9F*DSawpBOo`~~(jve&-!sOuv8IZlk!pT@c}S3iOnFC7Pis6Qj*2hgnjg|O6Q3dM5{r*@A0 z8n}J5*LPop9l2KgE$p=f{dYG-BQ#nBS;*`K<=?m*|A{w{DmYW>yfod&`{f?$DJE!;F3Z4?~izoAf~!@VDE~*g4tAwOq!$Y#8cwB0mX=@W@S*e5TX!W(Qo zy?~2EG>7wYlZ5sEM2qJ1O(LamgV+)k1wNjfX7t(bvs86h~ zF@TQ9*nF8d^ut|sWP@sJ{PGQbU*3k>JS(6);R7*+%QtvpJxCJfv+D({Y{t0I$}mip zVOt?;1^Dx8iXz+Uz-7vhaJYZzKGqRyX zj=p$0!On>!p*0vH$tVm(hNUni_$)LskCA+<7AO0NhsqxYCjRw;70S+i*1%q zoQWBixY`$Tw|=k>4NN&`UR9qF7cRqaiTvar5VgHS54Pu=t0&_a9uJ5B zzQDns=iN?!hXKP=c`)6!hafD5bm5=k$A7e46;+Lv`?`QvIcb z|8Wv(&$pZ$D&(H$PqY9JWQ4xK7{Ft8(9*dWr(1!{?Nb=}Qo9Gs@$AbLon$&zss8+N10~{!IWWF zGwtGP72k{=`*j-rn-A()=Ijn0oycPyRr%8Bn*hmc@3Z z+JO64uDbM@nRj;ag2pR=^NmyKd^twfRWOe3T6jKzvJm%uiq5A4jwMQkoU-Q}9XX7W z*00*TK_u*tQ)zg8<I<&4C@-B|}+ibUI zT*+kD%|>hcC3MQUTgNuR20-f-mH$=&^XFcoRU6gLbw^R#k-~B+$^E0;nYPb!+yt!$ z%H!BBl7s7v-SVwtC-=?2)gD)SUavK4%`u*N z2{P4}LKm9A-DP-NPtU)7>uYUIQE~}w)3x%G508cs^f6QCuQ9I#3<8|5rQzUmmiv@& zS8-4jTsYJN8vgic$q_7?inKX%rGO!9Nfk~s!w|{Lq4`y%R~k6}+VQmIlHwxF#(g)b z{%9G}2$ra4TOEtkPvqYB7dAls8JR}L%riHdU6mBAI?VN7SC<%s|Bo3$I$ z@XVrgBKs_1g~Kf@O=$AwEr2v{r2t2?rTl#Anw~2OA;be%2gq7ZAI>r{1QZzMHs9Pg z%XUj`*ZIFMZ@{ybP3^QC=2skHKBsi*$l2=Pe+Jr%dQiOsNLX|QMvAS&kC)3{bJAX3 zruHU+wF$Iszji>JDrB{LUrbwBGbv64_17f9M)Wf&`8RPGd&k>QYMBYRnp8NM6L#!n zvOpkfozMmqtL6o2zwumV#A6-*_Q4H{#GHcyRY6D49PB-RI1rx>~ znpcaD2(xDco|NW#1LUdL%9;<7I_P{yBm&uKtTb&uB0<25Rq%>Ap>ohF@fdVgw}osF zx!A`1vIgbB)6&U>_tm7Rhsbg^R5lYi*Q3V*dfM?WQ_zoOU_8C#PjhUQ)zg5R9WKjZcR5)4NJlyR5TD$m8HAR$X4MZQ$qZdeq7 z;2c?Bew^p}nR+(?_lRs_xl6jvaVJzsNq%jg?DhPXj_%{Kgv!vi`&L*Ncxax{leAT! zk~l5(3B~H9K982Or!h=TL~m|0%~%6cm@69foa5x&_jWI2Z|sdlF8i1i)e%q?rOpip zrJsHSz+;*Xh&Gv!PxiiJ-WCR;3XjTrPG4e(WZ(YzbBYL9$5;dqHnzr;~EJ__sbwQ%{0SSRLrG zGwj-z0qz*juqOsxk}x3Urjv4`0ZEE}r;H!~!hQd39Ph!j9Y0m?Y)yA=!0#clA`$JtNmTFnT}>NQ`QGXdW@0DpK1L*h#r<|7~$ zLV!~Kee7?RpbsnO8K1$%H%e&V$U)TJ>Be5i!tTGj&*Qn9kEt*sxUP-w1`Yqr^Uv}n ztJiLR#Ug~S7rqns_aG$xTupCY9XXypCb2?e>IW9;L_Fk9y*FnOzc#Gg?5+*~H%rAS zT=*&fRqxcI`wrr)9<$WBF0f7HMyHVORJHGD!wHAeGbKw7o;75cl+8jiT z=Ksw(-KptfLX-S#PEQ8{lK!8ZQ}3TnP^$kY=hQ-fnN^Pk(Du*4a?5Un0trl1V>XOb z@x)RK0Q%gaJhgsMd?U-r;FF{i9ZSgmL#@zSPkY6*GL6uakK zr>h&w3^R`lC~x8hVc@z0k5VW3R%|)~&;Nm(mWH?ZLRza}SDS!U{O7x3o zzBIF7G5TjW18myHA{G#Iq3C|Ne3TKM7*NNn!-}at-{+xl;s4@AU!qMVlwa$~J_`7( z?&jvw_%oo;zh@-Pbe%|8(I9~5ojNpPnyQ8E+=j#rKo?iffFgG5Gp>l0rA4yIklr0n zpD>>@3V4CbE5_0q8zJDtefVWIds}>;f*9mIxKtH98a3>Tfw_PZJ|dUGCRSEaD2om) z=qgk{D@6zq{iel!BXK)pIy78jnz!7H$qCl!(K`LPx@drIrbJ@hhB`Z))G?y{O=EI? zC;Pqu(03u>>4vUfiCE*TD7Ez*kGAEgfs zzRmK@o-?Sq4*j8sRw|gH`b98}S4^7-+-=y|Cqa;mKsZlz52i%8%QSd-mMNFU;l67F zvF1M1G{w)gBy8WG$PZB`_NVH-(h>K7V6&Yh^{W>8CI#&u_NJo!iSfBD!xRLRt+D41p&8x2ad7%QBdh)lUD&_)-?2iK zI6SGl35H-{t{P%%!i!JfYYaI(v9s_C=C>Bk?}P{mwx(1K6$+NNxhlXuPvkZqr92ez z2W~fi(Ur9I*6@f*tx_$ZOVh$UDRDz508qZFGdC5$bols>Jt_qknadK*!xC!$^0T#4 zLJjAD-Ua%;U5*B6m80FE^1~eY+Y8i4R@J|PtpfaNRSR=UuhQ1!FsicTbJ&`*_@D-&PeXb-@f-o z8%ec_*zuJ8>AeL!MtH1zs-j25Our8am>kNey+_?4ps@I<02qGF(VY&}*i3yl~aRz&vh3DRe&&i0TT&t8K42_M(1@3^CHoYE#VWh2cO5*M&C#e;4WqVPp zW}~bn&tId>rt#S7bw0o4CtrCWBGT9lwvI$=r*ToYt9tKnRB^&wPlrhMV@Zva|`zCrzsv9QJwWT$Ky!uCr67Wmy;e++0l-n5bpx#j_ zlncF++J+IzkK7nl!OjT{$(M?$)W!Mlwy+Og%)9L!RpU5~JZiQR-QxmyJAp5Kgmoo~ zUR1G062+d`mg4hb_&SnQ6%o44*?YwAZ9kuR3)Ve>LH#igchrtnh>SF_Q`-fPW#t(! z?rqY$3mE4-MF6_R-*LZ$Y$%LIxvD%1r8xH7tx|#Ntmli(@$T{0LWli;Y!5?DI|O4S zyl(ke$tM#`mk+2gB;}f%=O$CjX+`|s6}o7j*~ZCZ4e%!G<|!(8u)6bYpJ%ze{-zL8 z1O_e)QCULIlV8`OiSDHCy%hOvn}QLp2o;kFAg&l+rhuP$Fvc&btPTQjIMLvgTn!JY z#i?13>68C`+rEHt;w!m&Eer0gQ}Myk;$qB ztoThB`i=Iq())rnc+cLYZD_Ed&K172XEA(#NxXW9PuF?OmS5LOTbu+ZFriCxt>*{Iw-!Be@IQlPj3`h_j- zS*30uGG3Ht+&Y1_-L~DRxuqF2C`TV+VF@=D?4A0*mvT!UHD=7T@M29uIL;S$=HiV+ z@6pf>o7J<+%ew6^zn;g==MOF(8~g{vTs~`jWm-+V+nI_HiG*LT?C zlOIsrDKHKALCohnz9AO>1iEyPqpdmfG|To3k#gK2woj0ZVkV`{urD_)5;$-9G6WRD3GzfB#~Ps|A$%3#ybjy|1F2TrwD_+r~DCB3+z4hatzc@bY4fuL9#3D)SpJ>|JJnt zW{S5!0K~gYN_ZP(zFP(&@K42;98QFg%-ZBhdgOt6jXg2{;9-s9r}s?5Mo7Zi4vq~_ zO81&xbtx9mmSxc^HGcnvMyxgWgt6v^THu_DTMN4G?iUDy++n;oK@l#pUnQF|+>2W? zbem9taWO|k`1+hXJs5dfBEncOs!LX65-`&@1+=_;xaA+=xx6y*1lqmaWBJ{!1-m1@ zO@6?~dhK8#amAUGyF5=!C)uk)*qA3YVPzWLe*TM4TkM5ZgsJ?#{Q5W6^1gZ=A)GFW zfVfxW_kgoZ6&xZ&ypPh0AR$m-#i0hODoP^pxq=5T#FqeF&X1I@m8ZV0ZEa+fZTqp1 z3`pVlF&007mxtPOWr6baB_IE6syz7g-8DnsG&l!uL}=Y)aH2;|zx4~b2KvXj|A)zV z>CPiQDnljTHqf&^9AB)i(bvuEVQHl_w0VXW?rAfiz;bdA66;D5650!#&?FOpY3=6& zcMVd9k`ETyrxO?|vFb#h2&@#2DS|opB_`aJk4R95x+S>huKumtIR;+NUndD7awC`! z8dkI|_Se0TN+BRQ!_K*JJCw=7HPwG8Fytlc#+qlt-56}kJE52;P3r->g>bRh3k(of zF3fX%P@Nc29itMdjhq@jLj{Ja^IvnLS_B)m1KL_;vus*gdXZW>gWc`Af+F_FbPL6I z=GbJjL6E+tsB$fh0RzA1WoHoSuV_Z9D1Xu<+~XL(-oT%%Thhv-)l%g_O>}LRBU7bh zVQP=&i`F(Rh>K9iRO=J)K?dFwKsf}ujtGjL$5%hdy1xCMZ?! z&H&seXaCe^3M)15d!UFiF~g~70M;NiE{gW` zb&J@@uJO;L^%COl zv?i&-It>({3lj)dz|p@7+(p$%AUZl_PV8^vdcF~Pj$dW+$KobLohAvo-NRsEym;pA z1$a$ddr$`n)Qm}tpx`sR02XGXpH~CLK~^(i0z;b0>^3s2MAbq1?H%!Dc@dYMsR1MX zm!T%~E*Ih#r2U>G0J;!(0a|EEqsR^Pbk{UVXjuUUXY85)1)BXv^DNI8kakQf0nNR0Zv83KN^4UNblk&Zn!mtqxx3|EfrV`DbX_GDT9|v#8awi` z{c?3)E_4XxZD3Y6<2fu~yyH^qG=z0et_bS_6IUEP%D9|D4RcmF_%7lE<(I;@tAP9>7Xq8vV*Gfy$cmnxL-?UPNN%h~awZWSm&ip-#px1^T zO3|}v84S%Lt1;~Lf>-2-p5!rN)rALPEpCrzQyUP^49nEZkb^yArV*IO{Se*Aw0ikM zN&n}jG6@VB4G?1pfPm~PLgc?HUPkXB_=^$ndZS3&zM<+9E7d6Y#&1^{2;8sL0NpD0 z$82Pp8SFtP=3G*PVoA*QFSUk>7Px)fjEuiB%Nmvnj8e-%Re*GoBTw*g=g5n=gCw;g|6kjUi$VZ{p_s6!&5VzOnNsFyZ z>)YHU7r>x-|Fwv?XQpeDc7ABJ^p)!^@CWs3}1@>~fG+pO|m)>5fY zDQ(G^4>38Dj`8j=y@^5~Q%)L8Sn5R_^A-yPvqu6>w4ErI)O(wsMaE_?=+Z2`T_-Gy z@ucgBc)1)jnd^z@OF(@(Hzx#^9b5F%JT7pPSGB5bu4rn`d4>`rmm=qdHPpyRNZ&CJ z7=XN_;^OWmHV`Ixv3KUj4$MjqQ1yRWk^M+a8m zHfYkzT2>vj1yX>%v*4Z{|4w|C?d!ZSApz_;)(gcTx{S zCLtCNb@e82^^AZiX#<^(H0OL zIwh`pKQ~?duxxVCO!D}oQ)P3`S4d#6H#+HCr%=xdlgp84%vg+9s?k67;4U~lkO=q7uo3$lX<^Vslu7_I)_lXy~;VK z>v%z2JsLyOUs#DK4`Mo!;c^vsv(K~t-JqTMWTA2hs(SI|Sv;Fi@Y`6xsqXr09JXVS zkC1YwXhSaT&Qw9nF>e z?Ep`0%MM&Wt!`5-7chFfC5T&B8uORot&FlBmt)VHK1#@CNuPrK%>Uk?{ zK5{USHOa$pXTO{a^Qy}CKJXI%1aUS^g*kx!FZz{JS3?TPYCxz^`zdO7=9|<8O%_tp zo5>wo+%H35$%F74{i#rWt#H7bMxFO&J2wd&)rZo)9D)>CJU0hr?24RGn2VM~-tqD) z7q^#8n4@giDq2Ja=)9BxwM+em(L%K!c*i@VW{0vNjLh}L<;8w%B?Re~fevcjI7sZT z8Rg$e3NZ=d6X^&x7r*09PgzeQ;`p0QK#ziwLJr{@&}DHeIrdY3FM0vy%p&RdJL-TL z%K2!^f3LZ6Y^wi3?W5NUTWY-I9=~6<_E=E|F~Bkjd{o!Phbioc+5OzbD3TC@n^Z^0 zM}<8h@AE9dt@0#3FM?koe8>o}k%|w+oVk5a%`+r87(drvTm@@I<0EXrn^ETvYr7-PrIhcE!p) zY=#W#=aM%oq{-xv!qg+UzI=W3JqU!1t2&FKxS9N+(G7B1i6END4Feag13IgJXck^; zaEyLlYfwxMl*|LGcVdH;y@xS^MYZq}IABbrxp+GS$tbz2#=3yZgr^HpgrpQpozp@C za42-fztF4IJB)Sbk(SatMfXzc5%)3DMvIp*iV8=TjgGdWO%S6TjWK5OE+m3`ER&{U z@V&+wS_8$p-6~YVzD9Wwb@M^mVvaERKgjgr#p)PWdTwC-B%P_J%x|}BO3TLevjk2G zL)i^`3&IhRlY0QMNjNF?DT~Bqc)Bg8?*ugb;fgpgJ|j|5B=KAe72LNK-<0)^>EBK5 zO|wBlInyx5e>&|rh$Mr2ys@2>D&Ma4YunW4bnG{ zj>>!N=vH#C=Z3%F-_k47B1Y|oJ?&!1k4+#a_#9JPrjXy+@4ktt8i(orhT05G_?bDQy~}v<%yS@J09* zwS&>P7PVRh{zQaoUqrR@>ljF#u9aogW5{J^8264`Z6AZ)13AL<-+m2*A02P~FLTxd zlP`m|%KrJ}PzNdz^ymo-wMQ;$Qq4g-e++!#8UI0lhhqDsO9`0U7~3E>c$Uk99w=M; zn-%~>I=dt;4t$?_GNlLV_?IzA$c!R3n*V8v!zam3%NU}YYCBEcn$i!e$_mxGY4bC~ zt*)4b1p>p?JckT-mpA|&_b7HaUkAoiQ#`Mt^fb}8(s#`s!nWT8`i=Oun(>9oiup;B zXB;DlUxo(D1%7iR%9o;lU00|#m#nO9tzZLQy>eR9A{C|xa!*|h6ylL5taCa*>41jU zU|0v$sdMb4h&Vy34U?LFGivwysu#j8JZ1xRtSGA)|8)$6z~&Gcw)OO{2;pd7PiS}n z$v*YtyC=r(7U(?x`CR=09xM8x?sTP1jWrR}U(cu0EtTLm7`d$lVmF|*5udtYAAo_K z@6v~7T}mTu=!;Zuc;77>&V;Qkba0(xd)MY^3)~upNV_M}<33E$Mw_`pGbf_>%J}VY z-n0Za-3z*YIZ|EbwuFK|l@ov59Q0JR3%5V`HCbjI9A%s_UCM*_hx3PHl8<;St#JHH z>y%0Bb9u6W+nous)HtjTy8Xd+EC9ru?cead0>*9}DESIg!}cK-J4S1@xzv1Ed};?) z&w+|<4fX>qJBjDyZ)d@^8ECo9JAZ)bO^vY~JwCQuNkn->Z`M0n+co&>%?t;=6bO3U zFv_dcYZLiAP6kbBOK_VWQ!Bif7G&A6JrpSkn=(R^C)LJ#jd-9sncu0S7?P;hq*g3#Oi0UY)@*5Kt= zB#HS(($2Otjc4jm~6f=qM6`Cco4O}ca&bo?PD+jBvXd4(Gu(E82=#OpN9eM&jCR22XTN zVPeeqiqS&NVVK)loQ{=Tp5f}7|2(3ZhI14oSFE-j%gua!amvJ@63S8|Wv%vXUbND7s>jxHR*b!!4cxw07hi1k8YghU9kQqDAX6!q3FwdV*RKxXR}oK9Tzs}y zFu5doDWfZFfsxLl;H6OQ;@z0X&|<(4@6mYA!D+JwQBAdq&a~f-csaf&qq<#ySX?Uh zyIAuACSEv#hBb>CW&t?h+A>LtB7ezK;rA~;?Y&ppH=0xa>uS4-^HNQaG8u7=`~$n< z)BvJGS=V)mAxEr;LXDh4C76s~Df-6m8^gageYrU*JZ86|%MdRBL958y!sVQ2f7a;$ zR&iVy$-)ag@W?#KIr6q+@~cuHI!)i5?S=Z96G6yW?lPAbP<4wF>>%SnyF5gbk&@7J zy1wrMn^`8+{2)JX0$;BdHDtrcIBGuZQ61#5=ow^HUv4fOPNxw(tbbXNgz3pzcWQBHrerHAGMm8f~g#P<7L^&5=?2>hOB?eAA!_@as)1(!~gXR5R- znIY^fd&cP{KwJ>sqQmpZQv|^-$Io!DQGIO@KlnNCK+t`X?9u=7-LFsc&mV-s6r18k zqTew{DQ$lHudJu`szwJGs{E0ambd@@OloR`G8d9KWFsCm&sM$}3>2X{PGfD^NG75E zxxj1am;|LRnV_t0ezV$0^-^IK+0yUh+HkoT~Rrag1&+!xLPr#~yg} zV|_lpwR(xI6I)RRTVlBxPTLBG?wAj8@~~MHk8i#-w!;JUjTUWE%-H%9$_NN``)&N+ zf#FWGLvHFny5b8-aG^fv*Zey3#phV~HG|v^z-z#MP}3v$oPAkro)cH8;5YeKewYS> zzBqUlfvm9)1%?kIslTczSgIZpc>*rO6jBJCo@}Sh%R4mxmO2B)a#SuGO@aZDi8y~9 zqUX30zF)Dx!sTgvlC`zoh>yXW?7v~cqtPg1A|5a)?8G~J7>?z{r?JoD&0YHd!mYvI zfa-9kMUe^UfZ`NUdx8UaQ6D@;R#v;lHLBJKG>MWzJ*Cgg3yBh%CxhP)gS$wBiKz=-OphAPdDL{%D2m{sqC%*uOgUBsFV$0GnXA zg+&ykg~`y=TgHgCYZKcmf9El{p=gb%tG>6tohy}KauYqC(#{~Uj$?AUwSl%=#KoA| zee;j@`iZr=no?x3k>IpP=Tr9OhFF~o7P`>d!Y&F;TXw*^U>0uiWL$lj0iKNg{JD#k z;;^oFLZ-ju=w)EI$zd4h`mjmO0P*pHc3NTF%R43aRL-g&L7FVZO*2utWENxn(Us|h@N|y0029Ip?BilSCM(dnI z;pfWMbm?~e`uTTPtDfeO)5vHGH{p@+Yjf{T+w8jixg~zduZr{dXNjYNHXx?y(_!-;3gL=pE*`+T0kl< z_V1J$Jb8vljgQd*nH$!QF+>GI%{Kc1a7MA7L+Pnf?zf}*E-<&z6mGOK&}kaKW$Sm< zM(cyWHTL+b`fbJmTJJOXy*>^8fLFP-!J*teg|`H1jYidx3sJ~N&$=JF`g+%q;fUuV zPc-GD?DiKLcW-L>yvtHGgjwd{Y|z@K@rt|5*cbKoG92#v~i=m+p; zphNa2+}B30^b@=8lB+c}*t{(>n}G1C znTy7-N^D=D1RU1b^>Ntj|H57bF%6@(qp^V~6%y%V%|BAq_fY$kG#@q`MObCRq?du6 zmh>}tkfV-#QXkG_S@Ii&WQi08m!n$8Vn3-6+{g$rV%~C}bZ8zVMp!DSzRSkoR+kiH z$FNlv^ig*JEaE*%E@IVn`|wQ@LgaKM+4num+<*^FM8nLk|2AzpB|gH_!DP^)bw*2< zu*y$j5!a%*Lqrgis$ctA#^bu5G0eDcLV!L~cbo%XbQUoMm>lILc76d8J9F*j?g~ZX zl~6{POP1PwV~kPDYBr1UE2m;qb6Je8?~<>ue$kl$icuncmqFwA@N6i;8iN>xmj&Ej z*=3y;dC^$SPfgrv%V60&f7Mw}3w(JPE6L13-To0SbjMETN&M=fUpQ?_B(xh@)iSay zq;+YIxSPK4MUmYn=(0Y{rN;=@m2a8U=`XoL-l$ivR4!jCZGyLumyGzsR4)R`BW$&* zo|0?^(3+cOa_YkD<1P0m!?BE>qb6i;yrK=Y2(^S?{~m!DSUQk)k(0GlxOijUMR zgCe+4gU%`?h1U?jH-jXlD4Ey%IAnk@Fulm(O{*bgEz*5aXG+O*sUdL%YpE>5UF@P@ z$2x0-z(Xt7-4|kI$r7V34fWE$$v1yE#NhDr zmq^Su`xM9-G53iW!zX^?wA)vEoJSb)>n78W;e1s@Wp(pm3m?~1xKgBMOTr-o$o&y~ zAQ=#$>@PglyLDdT4Sf)o5?DkXY7~imZ(x)yAN!C^^)Fwt(79ADSbr-1YV9B_xl@tP zb(-LCtzY+zqrkUifRjd{(_LNCq+ z_;wtGVtsQi<51I3h6pkb6xgCQ51itRmb{~t3s<*eYeNxp$0KMpCCdl7NRCuc<8%r+ z5OkA5mSecB0-I27)L9tCd-U6`U--FZ!bt6stA}=3*iy4a9M^IKT_=_R2e5ZjHAmjO zsYG%$!k}^~Zxe(cXC6UjgT47N<*vdQ8h*z3#^0jI=efBEalJMWr-lViLj4hAuh1mk zf3T0jU2piJrf(d4585Zla|~<~sS{k!B20dik0q0S0-ac`gz?Nif6cjyX2YZc7InG@ zRm??0uxWKpaFXJ*S_<6d$FIUb$3=9F(oEnk$tK20qaM>IleSsg4r@&-yzl%oXf%G! z7-R&X(R?u*|Z_|&k%)tLIpldH0YhdLtP$!r$wndMNm^)?fUsGgEOD&*E>qbbJ6_zm*i-tT*Meq~D^D;`Z%c=s17o_@N zEh~ifg#5}))F_gQ<27lIaN19W&?4HH3bWDMGpAitg}&QjRo@1~Su@Hku#2r6Y*$E| zGXZ&WEGH>gWXk9gM?w)YX;xK_vhtfKpCPfoG$gr{>v%Pk^#k&`fS)MFac zXs+YZepcG2tGy_g8~vmnf$KBLGR~k~OA9Mh7FD-WI6J74%jf6AB@2!@3(C{%AoihF z69ttcXest-JmPQ%E{nryl+W5|L|6x=dI2j5tb@&xfil>X8MLV+QpHSTWD#Unuz#(H z3=or8Lx^EC@0?6?OKXs6+hmax4IC6(liVCzGma=!QA`V1U||WbYDT!RDAej=Sby1Y z$t9u{(j>7`r}ch$4cw9|(m1KgxG<$r>5uYle!6ye%kgas!#dKlaH%?{q!=B4d;(tW z;+E4f$M8H0M*!-b-$V#?D9QTS&5&TYrv7p$t$$=nV)x3IXUThia38naW;V`a09Su)BvscV;XC36(< zjs_GCAtG_XjPsTi5(^O^UJ$TFzF?uD&vILbs-Bf9%ZxzBjVLKdM93^TG6&lOLNKd`5+%KZco@NOV4m1e;s7zkp+p1F zn|!dX%Yc13e^woeJ9s3j+H4#KlLQN*OfEP+AlXd_jYcbvt^AZ1WC!@!hGfXPuP%s9avMP8Jkcxos{_&ZIRWLji4R4|QAgd!B1K@$;Hxkl{9$Tj>~$P4s|mxLEOz;knuOL)RslvjRa~>ce1JG&!L_fbZtzPQ zFAVK_uvH^199(Nwi5n;|xE3br_|s`3A!#82uDawhAo2>*;$~daj z1&RM`gKtT}Cd*;M-2nZK8Ua`J4pmE*zbO5q73s(yx}!Gu^_qfHXbr)57>bG+6X>nz z$_#_xBc@&VY{*#$Vx;&oqC{y%{4e+(!bTASQ8fEmPwQsXVDF&BiLLP=#^YiP;SUZH zu>2lAQ~x+NiHt0fG-Ed^)Y~J1`eB|xUdG*PX<>l&DT}cm z4cm$vzpT7a!~nFTv!H`RI}i8=p{YPejT>#swIm*l9UfT6>^$U3iWSI)9gSH=&}kv4 z>uO*xO!Uk8T>OP)6{pyM91Hx)4g5?HPe9{Rj$wR=(w&BsPnuE|C?7$&`ItrPKuyO+ zcOMnPS*%IY5Cjd0r3abd5u+gB18gsxxY0r(!jd(7(Q)$WT zI^p*$j+Ieak}%sC8Z_j`)OV%BMCZkiRDS)AI#A%-R_(0gvFHWzTd<-^LE=!y2nf4d z{+(G>7c+@itjK&pRr11 zvot;*E&#!OgAP1=dl5~S!ImH ztnx!g6{=1g?R26euwL+aTWG#=JZN3^n84!amay^y(6!;xZkB(PkXh1F!6|xX>Kndb zW9U4As%ssv93>p^+m4Gq&!zk>UgDq_V;-aC&wOI_;hkb|nuRXsHqRO|caw7Ut0CTG zUSmM=8IAi~`Rblzapb^!Kst6A&fOdP%ngj)83q9>A0Q-UjQdT!_~M(AaP_K*V)te1 zku+lvHaI5~80X)jUv{%-mAscx~@f^@AIIKuAr#M&92 z2mG*hlBl$%u&zGc{*B}43CaZhO6Q!M4p`)p-?c~4;<$%QxH3k#RjN|^?e5xeMCt!| zl&1bYcOBJZ?t8N6PF>HR9xTcH_?CJR3~U&Nnt^lJ145m7x?e|Iv3vjxG;d`b-ZedI z%j>i4MRWaf3j2NPmrgr(-E%bycuiYC+<*CZA@kt_1Nd|x+0N_a;o;`@B&FYu1V|$- z$6asZ>E^->X2IS*?JIqJd_CNWd|d7ycsqLDtw=k*E#v;s`M`|dr2?IOUxMlOlU8b< z4$o(g{CzLe=M*X-NY&p85HV&!rL``9jy zHLv+Z26lQ~jE#O-e9dZ1R4h7A0t%aA(B#BW~`Vs<$otqJn`-K~afL$kjLBpI$wi9~_vB%%4?F-%-TppM)H zeN1NKs|n7=Z3n=+6r4us>LosDq_%L84zP+1fuH_-k%s>QFEnH~1sv8hi(ReQQ`Q;- zWAMR{XXim3OKTW`kSeJBUjZaFR8d#V$aP$*}hEuLb32HVOcVsbvSDPJ2|eobuRU0eq}WC_YKK(?n}0siJK z=IoAWY{Ab)?QBu~Jgr15N22;6+AGF4Q0{*GL4q5SHdJ^>h_Gd>!v>UA`hRa~5n4iF z#d#1V`DS@24EcSw<~ecbS=G|9HqnRQCMwOfJH$p;<#ep?E?epkSlUTJ8?~OAq|ESH zbdU))X)D3EL?eedr_=Ta82Cf?eaoYaYvqH=O$)RF>yaQN{PHS3zkENCeni*qn$-p9gV#&}j9{jfXzCro+eR}XV%!^jG zc<2Ip0Quxcnz`dKWM$O05%|2yO2TB=tn{m(F#Tqam{<$&$k^Gt@955utc!)*=W-bk zVAz#s!k@?~w0`k8G}Yn*AOqr!2s;U>3>VS;&ZbI3WMr2 zX*wKSmTin2{XulQfo_~aqlnGC%B>D4g@O)Je>YYDCmVcb8wTeM?$LT<34$+_@XTqQ zpWEfYVBv@yn*%;C=qVcJPvE2DWIXlvuBr$0axf`zW54`vhAPn%s-97K1xvur#@=g0)wQn0)!hs4HwTh15oN;#wMOMC)tt`$bW0=dBhDMGLcz688GlLU}qJK*r{YQ!)Nq>qL0 z?JXPGU$cT5$6#-C@yXqnK(EGjb$#2cibn)>b?uVvzzssTDSlZxK{MKD6)DPm;*<~x z5KNJGXeT;xUSS2i2UYD*@n(V^I}y1N5vV&J5Py2IELdO}B{!$ZivqonxREN-^Fhxml)}O-%>r07%Dn7VFAkl(@_EAH-53rt>mm2c zzA{h}Re)7G67W)MCs1CCMO|Rz(DA63v33PPmY(829*B&(9Jj5#q0~j0YYQ_JkOs<0 zT8Cx=(|f~IJ$Ol@=&QT5!gf}%V?<#ll8(R=8XzcvqWKdwt}bFG*~xxjrtOV3EzD9N{^yk~M{`4WSusY~GMzhl8;+*Dp}&s(fLrn2y22V| zrNmhD9juqYqNn2PWxW~@vOlogh!~TSY#0v-+F!|9S)5YZNS!Rx+Gw_$U!WK}1CeBg zefmei-N9FQkWM{7MEjPdvOC2x8M@UntpVkqW84*lQkuy$SLr>*ISgAu5e1fsaZR*8 zA(9}Kp5n9oUpmRPdryq`=L#f)pcPGZ#n?!D(GYd8qz+t4k98Jlwpvm3?+|f)@7p&P zYyaVVPK3hGaKVB{8cg7xQZ`rT+Tw@}ry(%%S z9Ewt(yCDyh%N~W1tU>MF3DFrj{EMM-IYQ%(Y$KA?>7=Eov~{z_SLIG=G1opy0_~z8 zbGlbjsB@f%e#ymt5icMzhn(YW0|p8^B^qjMWNTWcf5LV;LVNs3D=XNa>R&tGAGNK? zLulI~+|@@v$IIPXo}I${264~loe1k^Y=W0I6!@maV_^#X>sFd|++sec-spxQFLGFQ z2!ce`ZCcKj-9n$tn8&dr_{$Iku%0&%fz>xleuJ$%BpS^RkPQX|G4^2t5LuNn)Iq&! z0+lKVwKp<5ZlPI6&Uf7z<~PI!=7yeac0O=*qX$90Z#R0ujb`@&9LmabF8Y`9pRN%? zp639aIz*e+&Ucak7cIC~6UKx56d~5JAsMScy#XG{J0#%qH0+@L!d2W*_5kS*)r{a- zX!1EYNbZ#zH=;oyaSBp806~l@-)0o@E=+wz+vPV1g^-SYu~(**@e8J==!$lpmmv8O zHpQ=!DLaD^EV>~12som<8>=vl9n`_LE%_WbSk>qz#V~>WfcsjrQ9ZlffcHqP<1_kaV)u7~Rvm&l=kAxU?^3E~)YnFu$$wXgpAvVFORjnoD3+w;qr>{!w#0 ztxaYqk_$oZedmF-YI+rcM_m8^J%cpY;0{rbDH$leVkXUvQRZwwyT>PY{-33VAJ;>lCl{=+4NBaEs%KeHfi+;cy8=5x4b zB>oLnH5Y|l3PIpfS+sO*Y+aLUpaiInu5=M-Q6x}6Rf%icB+y`$ z4mH88DCc{qP84(DLNwxsT&aVQxN+7{GU!h9lFXc0ej?V&6LM8G)uvp-nP&dJ1*;qu zLp<%y<=cf$(2&H24MZ8*8!^H@xML^t9ksN7WpFD9I8UE7&3zp%>GFz!Xc@mO*79!Z zbgl08MO@R>adBQ4Y7PKW6w{t)NvcMbVb)dAl*q>Ig#ACny=PDrQPeG{f*>H_B1#4o zfkVzgB#0!*l5@^E=g=bb1&NX*gOa)A97I5Z2uRLR0m(UM7{2dK&D6Z_P1U@bS2I7( zTGjjX>eGFH-0FMxK5KLNQ||ZLGMvPfZhk>)YG=mOZ)o$f60U-{bqomwxE!-i@~POr z;@Ge}zIXPX;Z*R)8J2L-aEFGV5^q|prX%_pzfG@NY4mqdOo*fZ?tQUilAp4YuZmq7 z^bE*;GM(8>R7)yI6&H|V8_v-fUELL+94@Pr)(cULP+w$L`s49ofYpOwq~B|mW3XvtY5sJS(%Mi6!x7&@lgB+q<#V&+Nv{3z3cs3|?IXD*KU zrBrQ|%;fz0a*qp5oT=C{#&C>+bniay2$^e^&&YvdwQV!SpZC-TeBE9#=!~Nk{7c!v zY;Zsevd)5bYQ6+y7_ug6xvcVR1VtLykqC&=Y|PBB9IHT7lRsTCTk99Uzbh2b%yA#9 z-l@8*U|4Ra`4lWImhh5UKgxECH@mr3V#QoJKZ#mm*8@YdH(5XN>I+Ic;ujw4$eU3Z z2fc)HSdug9y>Z3??!agFm>Q*4`ss(9+vo2$Y*DGEM1CGoGOtoR5?XO<7shmUdV$2) z%sr!`*-k4+{j+4)_PNV5d){ZCq_ifetClXT(>h`evzAn&zsol29Xfn`Mc(sxWJGw? z^3;Ir8LldC-mj)&r>gg9@?VznF-6{9v=z7dlrq*-AJFJ92wBhK)F=keoa!fc%}rPf z+v`KWOyF|M)1_I*7~g zMr~TZYCM14AK?kU73a6F$D@8Y;MX6C60+9Qo4KgllIM5X{f7Yy$Km>($!M@Jd_Vii> zpXhz)5&MZ_z7PKl{1pG@2SF!QeyE*{AW2Hlm*BW(!v>ar(S>Wq@5cIirL=2uwQkx* zHIN$6h5JTB_W*iZQ5C;h&sIFred?&i?QJi8auA*8Fq80roH(+X;`L)r5{7?=#Zm&w zis5cEbdM8;#4pS5kTXdgEgSZ=dTI0Y9W-2NeEDhy&#p&djnwL{VY5JaiR|{$Jdz~d zB~wmVGHyx|yd4{iy7GQfxXsC8)P7lsIUT|&mtRm)!XBwlJi(vSe0lW%g{hCsEBE*? zH+`4Y69|Ui@w$on6+fqC!HbH&@RnDbYD7Ca)(pT|(v5vJ=Hg*ckI1RNiV~V{otli< z_Sh?mxxn*yHQ-ntS)2`3bfw%DSbxfEH|t}* zc=F47U-N>a$tu61{W_b+7~jNY=_OAm!4?Jn;~30P zp}_YiieTN@MtXdD!JA}E;&T?;SRhoYMYcEo3 zy!%B}^8-~}5cZ$8ZY9nu=6~&6DF=^Kx>6?)Ek(LZYlQJf&$Q!r! zyhdb9Wh!@?3%^Al3Umz1g+A*#|0{E*IfVGZ@zdeFu0)7)uVp1q7!T=iQSxk+9oYG< z{GOc0<}h}(o&56E!llCfWo#&S{ENtCbM8--20eeAJ#(Y6!_HnldzwJ-vgXIrvhUS* zNsX#~aJ5PCx$43spuQ=CP=BOW<-@wiy;U#2tudV20N2^1@w-M!o5PY3Sy|XSXLq?9 zc=TP{QySiN&HVf{qw@XXJ^BsHakOnu>CYK4&6u9#6e zW9INvQdOJM<&Q78FKSZb#HI5TmN)v}6i!CRLJU^R{|Yt~x}b)ipEuU6cwEyNtG@Il z->(_weVBeh&ddlgOyv1KSWAeT3B6%SDy}rGX(?x|jlV&%(1=@?jOLIAR;~u}$t*g{ zfl=0MT3mylPaWHX>|e6W`o4_J`C7UBwa$ECs(tUW=T z9(q4&UhxSD5o*c^G>g$RO_jz&yV?xCr$^T z8^`|443mgEVSKNcz56~ptR}12WSMQ@*Bd(9Z3@Oxd>lwJVIcx7m+XmK3+#pIYGjU zM}-^NYF^1hfk#EjtsNY-nAPU9s)1f{Gw^aJ-GAH<@>)-8Z18lM^t)cT5UiOU%vd8i z&)yx?-`kR;*S6j{>DMnDw7%Hw|7!BZonGTE3l=TKGFAlcFiq8vQ{=3-6+_R;Y^(*F zfyjl+&?}6*LRN-NWn$@|Wo7qLHU;RgzgJCA+N9Yg!M1r%pr5Vp%6(v0U_f7sMd6cv zs&0}a+7F` zHg|c6p6kqD)R=i-AY@BGD0+<&?jPc?U)PtH4!UbBzoDuBk~X18Bn}R>Lc2(EO|fB@%_Ym zOy>L-3saxE?q(<8%Z#yeahie0^_~CIjpjen(kgf6vN4wL-08r@On@YWo@5x;H=;fM z52Km?JuzL2n_}|~3p!-CXo}9L=M*M)6(rvgT85^hRxhe?`Tn_^Xw?dOH@iC=&=JqT zsu^4dUN7aw50+bNYrfKxCdZMxgx7Q6O?m_evHr62SXai`OR?Dw8pgLc|K}$uEpcS> z+3MNw*9v<)-9J^<)@y2Evtg$af1_rF`VY)!BmBDbS5_Asu%WL?%eiVIC7-l+MX@_@ zp)2KDIky*>fy37=X8wy8pXBwD5}I^~@|I?$B!pdXsF;+T5`X^_RWT4b_G=bAb!Oe# zg^7B`St0x!%($>9;&^69zF4aXPlEz`{}!Fy1xW|1owql(Lm3e?_s2WVEE>1W`HJ>7 zi@EC44>L8N_(N_X4`S}lWEax=(~4$J4=}+3PB@no^U?-|Q5U4^QT{Qsd5?Q1&l*sD zzW*RQOqYLDg(B5LQ~KpQ9_bGH5DTB@3Zc*(WsEdNl~_|MUkTZ1U3QgO$YeO~@=oaL zSoP1JE1aHuXxGt--!Fowq!a4-rlY(038%Dpryl}6BS@hqr_H<3S_F@Za65j-p}yjx zfQKka&Tl^}@+M<6rMKN(vpo7(m^nySnr%Du91@_WRu~Sp+C9|T9lXlH&sx2Hxo7pg z_A?(d4dsosol_y7m^_2<6RzgxzRTly|7LSk=Y4FHW<6F)$gWzB)-ycD2m-qlf8I}tS7_9+HWp_}a>^&+us;_wdSfSasO7_?5$ybc zYAaF_?Plc4ar^z;ZW8REI|H>Qna*tyML~hx$50wk=fxJ9p!a3%cxv!;kKlF!BtWXu z)#%?mzlqP9vvBw)d==Bhp36Z&+wNp_DCwMd4zm@$XU?z2b-F%QF!(ebm~<+?`(J-ERtiorZY%`H`~ZQrl5*=qh&PRtbP)rAh0u(478 zQikG7XMvAy#N=<^X`KXtctWa@d3DNCi72D@b5N2PM#2Sg^7z%?u~MHij-0VFpYx%AF!SD?BX)k(q8zK*xeEl24n3te3AQ89S_Nh4<2RRfn`R4Z%;~IiWIFY5@_L(^j zht7Y5INB-1{wt)u{*BfCUxA@L_08}93ZlPB+f)BLh-N@H5+{@tHxhp$8Du0*21&LX ziBm~Fcpe(^BJ*R&b5sa(X4nIoE)qlpVv+4-H_LHUWFi>I5Lt+qWQc5pG#MfXp+$zsMOcv`@(^BRh)4Mjo;$_DQV?4p?J zyRi8vFN6Q)HpATa+6g<<@iHUuj zs)-YP5)2A$j1m+IZJs5F7TKthUWz{!QSRth-4j#0RDCR>-qEkN7pQh=8lPZPWW%3e zSY*SU@VDqT-Mmv|^XUHyc;p>yZ1c4VyBo5nSG1S+9{C4iHMa)+F=}Pwh)x(Rvf=nY z&L@P(W6LK|4M9Qj zZzujo0R4}G{~yMn+xtR?Vgj4Heo!y{Ep^_tczODb;?fB3niCUB7Hg$@Vf*I)G336m zm1+H-1L%KV^1oC8g*N+lWHc}A80=w|o#JR-VliK1Jb}=|nm^DQsSxu1kATLAi;(w! z5kUV7uk}BDskR6@$T1IV#`@k#a!W6rV=dN<{k@Zv7F>GIPOLfS4=1UA<>@`gvF6-A zoTUAYp!ZzCn(=<#{+<5KjPLXI*<12|-hOpU@1M6*f;TfCp0^9#Qqc2uy<0*)Z}+>U zkmv1bw-kD-yQT13-7Q7l>Tc=NbEut?;mu6Ut?-uOZiTn>`Br#KiMPUAO1>4|QtGYn zmeOy9xAf(9LB=;TnYX%I%D&azQttChhxP5Em-ZV9ZKqeh*l`YVs#-E4f7ADF_dsQe zlfKF-*R3a2EPqP)+$Wc~fw)7i`w!%YH}LB0vI~P?kcUBonJ0z+dm1rj%d|+L%AiVq zW}dQyVnJxS_G6(AMsJkbVO-&AfySUpS!SNGgko1{dh%l-2_|op`e9u0YJtw6N?m52 zvxMSWX!_Kx;&wvgFs^j9z+h0NB{T1Xgd$d0I@uE;31)AU=3!jFGfO< zAuN6STUwYiTlWt@oRu2x%+);zG_vmTcF%&~EZsndd8)qxNpjMvh~bM zRdLhGk+SuyOO0_C%6M{6J^NC79JxxQTs`N~@3>bgc=Gk!OCxbQDv|Q_yh}52J}P($ z_54f!;!;#173u|+Hsijj;3?J%E*-{At3)c+3ol*9U8vx_tAD+O9#5_s`L6!W(!+QG zRUi(_r~(OCTNOyc)~Y}X_ErUN;ZRi|4JWGt85pGsWZ?!#70AKesz4r|QUwa|x++kF z&sD)Y7+Vb}!DMPc8D>-iDzJbWP=#gGfZFXgUZ4(Js{swzTMcN!p=v-2PF4fjFiH*R zzzu3Z7w%RAdhnDQ(1+L6fB}521`J_rbzlUOsRLt}Q5~4T0_wmNmQe?g8LX`i%wcPF zU;%rp14}qm9azE1>cARCsRJ9hK^@q_-Ri&&o>B+)@VYv1fX~%|BaE#9oM18y;0!Zr z02f$51GvI68sK)Hu?BF5tu=rL?5zPj;ZP0W1t)6&Zy2QkeBcHR;0t$a06%z21Ng(x zx(0yYa}Dqw#?}M@FqtO!05fWWKv+QY*2zdl6Tq;xCP2d0njjeV)&wDNs3!ObCu@RG z7^Mlq;08?)4tHyU2zW{pM8fNuAPPR$1fO7REf5WpX@MA+Q47Sv0$LyrmeB(7u(lTX z3|nh~1lU^(0*P>_7D$4VwLmhA(gG=PgBD1IyR|?XJf#KF;dL#L0iSDuFEF+?_zIJ0 zgG`uF8)U%(+8`U2(FQrNwl>Ixt+hcO?5z#*;ZSW*04Hk$6pYdag>Zv5D1y7SK`}g~ z4NBm3ZBPoIYlAWvTL+XwFqsahfEjf_B`lx=s$dx%Pz`JAfEw6Z2h_seI-m{?)dBTz zvJPl~Q99ro+@J#*;cgx99iGwwP4K!7Xok;qKnskm3x2?4x}X(i)CFy@fG%i4F}(K^OdnyLCY?Jf#cz;B{Tl51;FT0T^2k z48ml3Urtp~2Uz19R}Y9{2;F>w&*8wmz7L$@IYj%%~3*VF7&r{exxn!4j;k50+tTeXs(1>w{G| zR3EIt$@*X&M(KkMxIrIm!rl5{3!c&k+wi(R*n!XW!7hw#0QO)q1F#P>8h`^>zyKV= zG6vuX);0jgu(bg=fxQjDDI975&fsJNa1Nskzy;i3050Kf18@aHQwHD~UN-$aO=2k32U71W0TnfIyNN0YW6B5qN|YFaktK86)r*sci(FAgzrc zK#cS@0wl;#BS4BwHUeZwlo23DHW&d4WVaEZL{1q2D&)Emphlh>fu~4pWAF?~W(=Mq z8I1uAQotCzK*|^cTBNoyphH?41A3&lF#yO=W59q+HU^AHlrdmJHW&kDWVbP3K~5P1 zR^+-dU_+i819k|BZ2~xuWF~+U$!G$&kOC%v8!2M~c#zsAfEQ_P0{D>LCV(FqY64y& zlTCmC5@iBjAsbA9AhO#82qC9TfG~311c)HdO~7j;wkZ%rl9_@xNJdj2h7>Rb;z$`& zAc5321(HZ>Qy_)(HU)2yp{76@nGBf%86?US$RZm|fgG~i6v!i|Oo0M&-4rMy&rQKQ zB(@n)LXw#QWhA2+P(ccq0ac`o8BjxNn*nvCwHeSrdYb`FWT+X?LMEF5Z6wMJ=pY-+ zfG)Dz4Co=J%z!>}-3%BY&&_}#65AXYA<4{vF_O_7Kqg25b6|>;F$ZQyZF69bv^ED8 zNN;msi3~LdR>)*?V2wnX0~=(6`E3fk+Z@;-r_6ypa@`y_AkWQ#BNE#JI3dX_fHRWO z0=OUrEPyLg#sat@wJm@<(%J%eAiXVsCo(7>5_~{1T7p2NfF%e*%2)yzsci|6NNYoU#Ow$aPB)g*>+epODyAAR0+#1!9nlRv;EBU12IqDm?{gesPV^qc&yMt97|acg_Crc0phy?wThGTZW_%fVue6VJ1TRDHQ~zRGLQ z5#vvRQkVs|^NJnZKEut)ic^BoZBm{5x)FO~N0XKu?Soljs<#Tiv*AV&2DZxDtduj` zDdR7TQkbY)l^zEJ(yj*DUp+4LX}tVeXI*(TELH2@FmZke1@`9MW=31q(>WoX=43w- zgE`*BwuEo^uW|;yllp4?4lO$W;l1E5qviT?Gm6zfRpeFajkap+d`F?f98p$}$C|+o z#-f(@rx<$thqBzZLU{&$v1&T~s9uZWnhdGwQj4QPC62l)pO?MnS4K(lJ<=jq^-a^h z+TTq|mn-)=JUJOKk79f0eoGNUN7VkYR09r`=U*SkYBiGxI*xlB`8f<_DK3T^Ghj{W-3Hb(;Zao(cx#wk3HMVC6P;Uxb1Oko`!vMRMcRn-IsplvB*L#rBh<^m%5pI z?yq#!kJt{eVqyH1u>Ll$qXP>A^M2t-`XvdDv>bfNqwzK7?}KXtTqA$0|dQM-)u zVXtH|s8r-0OZtk{VVMFkzKycJeD~U6K!lDnH#S-(yI}^q=T&Z(3!TgzUaSG*Fy58P z-)C!8Mb}-*spdy1{8W!)uQz$gq2HlC(w~IOdf4-~Z>;@!n58PK9DHV*Pyg&6pQW4` zzEe=9YE*D(JSS~$RDYP7N>kauiQaLmL~=)qFp{^4c%L4z8( z=IE=1m+RAavK6HmW??s+mY25e#XWoAm+LdN^8LK2Yq&3qG{eE%cgk|ixR7__p!c>a zoASxgH@-S~^0%>dH|5Q80Z!WE_z4%gqc+zCa0VROzMd1lyAST}k70e1!usSP|6Wk$y>INDeq=}gO2@xO|5xc( z!WW&DlUwph*Skd`sJsQUv+_iI^R>@`24o!)R2>r6{SjfyY#%$suB-MO2|C(-r0?y{ z7SbtRJPi#Fob|LUR||uJ1D{gmE`4cC+%j+TSX4Yh{K#`w*3Gd!-5z!4jqB)eJ7(Cz zd}5&|`Qy!k*!fKl^LOrE!iA?YA%VwOP6V{=q8mZ{A%qKGY~E0aV>vm}KSOp{ybTEi z&+z;^{&VK9)uRP=nUFT2M+;YDWT-A2M-sD+d3$x&sivHtxAxd0ZyyN?u;$N=yAEVL zJ`(NRo|kmuAkCXbu5nE^OBa+=lDK_nR6uJ=B5JEx3teFE4r@F1aw1vkFxqJ358q!X zv(cxJ^m6i~XFGlnmSr$xU@>k*r5A;n@ap7>pM=(Di4lANCT z;oKgjI94g2>>g5l&EEpJ=-kZtl4+1{yA$icU3WWv>Uan0tO4IP2hPlBP(|5`HLThGWrTMNU)T%|zkRC-9U^9XG*{E^G2P8&QOPTU<)-xT&1Ekmhm zQy2bYa@+rzv5(!rrPfrDi2oGjjWqY7;&I~MAKjBbH{5JP_FM8=zn9m_OG-R4j0t@V6k$al~nsLlE&J*KXajLm(MS|bW2Adys& znW!6I*=rTrYZh^+nqHbNlM+Tkuf(5Pg&!OXC=|4V+oXx`==Iv>UtkBv7NbR`xQK-F zvjy{2PM35d=#^4a1pj06B|)Cj>r{8#kYVst zp|=0+q0|cB=(!y?PUgbgB$jut@&dOE^l=zdc|i(hS7X0li&Qyt)&NY3}OLw#q5X`cDV_N(6oYy1G zMHME3jY)CiOZiQ_+N!+;R!}tiv`x&KFF*78V(eaLhgo!0&Y5FX2r+)F#Rk(0NZw?L zIW)fV7$>R{X!L2XbZf;iS?IYeF?YvTewL!OxP1a40s{9qWwk_N$m%-Cb~~fPMI@re z?QI3w#dX?xN&@vju;NG8QYAP?=GaP&b>-+Hg(yA3n%1ZwX72ldWxZRm#?gn4&UN&W z{Z%uIH7738KjHL84BbT}k%cBuwDe%9@qEuY$i5c)9({yU+5J&KFI4GxwpM;a-RNtK z{>87So2o-Llw?6UGTCaAyZjkt8q(9ov__RNbHk>XMjvD5Zb}bBbP_K{*vAV46qhbW zOX>${u@=mibCc_NWX8-$dSm4c4rUipm#PH#87w5APk3^ywaB@b&nmQ`_-xlx_H;c5 zy?^=6rWZ4Bg@*o|S8jAPn3g}ZBBOKis+9HNDsgaMb@+L|ZyUF-u644ZjWmJZDv@>z zUo1p6d6S$zgXLN40^J2SA2ACoAw3`Ur^(?E6-24POWW6RTxo`rIr@MMM+fd3l^#q} z=-4Ox=o_Dm;-3Xt& zvih_6qR4%yep*^{`-MXau8~P>Cu537N$PK@E3K&o&#?tTdLxNg`fW?X&)k!FIcMJK zcxWgglS(}g%~8|g(aj30wCj}CW&AbV=|jds?BylvvLcFJl*d&MbQrwmTOr)@?(y^U zj$r?4nzx;_VJ>Nt&i$YKWHRE^kbLPIMxwEJxbm%`bX%{IgJ|UiYPqp&k~a^zFS0O* z-&nf)$Ced!{XoMBPLp^OjP|K4zUxP6lktO;et)3C{mYiBK-5~pwuKD<)^v%xVU=fKab*S|4 zJFMxXs!|KWW3IJ#121eXhw3w>Ck9Omv(Gk7Zq_dwHVyRpsq~U5PRR6qX>Hy1-xa&m zwrf@M@=}-amInNUy3mUI=(RI~XcdpGysjC(oBbh*?YYj^Z|Ct=YptM9sj;ZODjL2+ z_Pn!+o3uBW`bpl?A&zy$gNg=pP6J;y(~B3&3q$L!ZWxQkpY=5G4~!4kH?r$V`%ydJ zat+>wrFL~*7fm5}^izUx-wCL`#(MRoiYvjGDO6;P)QYkfUPt zUqvC)N^?`$cD|kEtl3svFcpfX^ZYI@Lv=d^2BLh-Dxa`5nvd_5rt2B$#rvJ3)M^@; z!n#}qr?d;Xy`5T{80o(uYAEP8xIalw=9Zni9~_-CMq$)$&?;&7Fpt0PSu9K&nsoJd ztFGJr)-;~o2&G1`h7ESu@uo93F(pWn9!i<{wbnX?!18JWs%lu@P*rS&gf(hwx*D{b zg%x_5#ZAw~pGM-X^OOBajq0ma2g6c0wxf547Pf>9wg=q(%?2{&&eEF0rOevUai+@! zRinNq7#0)j8?k)G^e$H<(l?U-?B`wKaC2~Rlh}`O1vztGH6(HiH(~cw*iKbgh1l{) ze#Wr!CnElA{<1o&ZqK-EwxN6fqM=965-QV-v@2o%^OEn-otueXEw8D;=Gf0I6kN!A z(>3=Zmo~k9C;1y%v01iWRRz&%E4-rk2@7cp5L;oTRM~1+okX(ej~MJ0<6T^jHK*!{ zq^jpdK=@~^tt|LF_B)jOvuD-@VF_|=D*IzerDho)w$x9#ua|L+nm=5l?f5}h5Kc@DI)JquWX+^u5mx`)BXmr4zV^~Hlz|o z=zl0fr8uA(!sIQsYKo;_q_5}4+{ci8>o)sai$xvfi#k8=VeO;)`Xp-0>u5q-kE_3p z(0=Y~P)acEOrswbAHHmz*9xKi?W&FkHMBE=%=63m@s1n)N5|0zB9sij+Klb53yED0 zG*xHw5-)ak=J?jTm&@}TtQKpg^ZR@l{w8Pn_A_|>z3myJ^-gGDJ#?J8P&;XB;pwvK zH%}Bt{*A2!8tVyk&s>1m3SiDM$mR)c`K#PYofjP6`#{V*~b*>sFw?DL|j* z8n4SBwdqEeQC|GQnlty&8dPPr$Ksw%eJ>}i?c{P)we71Gq}N12Zw6GSk`kOY<(r|f z<#AP5<1xVNDfeST2ru#o+@bvf)+52n)ipNa;p z<_0{Shc&FQZsf)zg36-HvrJ~D&7!G=^F!Z7GjYwN-Y5ZPn$|vB{}F#ugu3}bU;LzC z(eLHy_{mcop_50MMz5hDc=GaV>!yNTRtMLZ?dd2^g_q{1-V=sj+u0Ifxsp^hrU#2p zxwYMR_RE-4FwG{-XtK`xIaLdhX+u9x7wwTF8_4q>k)Dx)rU#m;#pma5Ium%XKE=1C zY-GBlCtv50?U&J}P}dr$2hmPm*0ohUx9wU^)2Z;9i0Q4>sqit({drS;veRStLb^%h z^hQvkWXe{JM-oMn(D)^{e*y*65^7}C&ZXDaXNPBh%G{IY)VV_#kM+*DwtCVl;^4sr z*#%gU=6$}8cS`&_1pRH_#MEK_3*A<&xPAXvS?nN-%V$O~4)BK2SXQWuk zgq^iuemQyVt8*;$Tf_ViW%Erdf1;@MtSM(o@aTG9KZ?xd;$~;orrM#)I9Kb?HvZqA z#9YOran+ONf@>cXG-;RY%X_l6kzVcMmln3qiO;3vcY3&!GMaYYmDd|JYyZ7+DlWs) zk%J|pp?QS3^}()imO0!p6TaURx0vO1ZtzeOF-f@Xs?G|<7H#hzo}2r1vudTXJ0+b~ zzSR%ydAt8kUm=D)oK7QLPkZ(1X6ta|VrePLLI1no!7$U#<%-A6&WXeIZkyE3VVu7} z7>s?VtM>Zp&rO>5S>tx~$<5V`=Vf@9%8}FnB9O0V4QFS`cRNXO&9w_|*y-Br*zqAV z{Gj1ycC!yHrW}lm_<7FGeXyY&N$DXTwiRUwOzV@aK9iK*TjC4j*Yj7Y-mNCB<8q51 z@+vpmZu_QjQ#wpNTo5V{JIr^aofQ+D0C}D)7I-=>e^(tZyn5_CN!OKUC8+ViyhO<- z_NF)JwjtYBeg9*It^{lG0?k}~lsB)+!Gf2sM|n6zy!_JBDTUdg@Y~DMF{{Ikb<(D; z^}MWW$FodNbrFqTKGv#5D_hG5?QE~~C1URxo*q;AS5~!dJ4s})H{RDd#LZZ{Mt|Ii z0&CpOs8YSx!DQH|-qlRbdY-ysg_hTab3q|M;c z(CUZQ{X2pq7ztl;8J{FL1=!O3rzy*ANn66;w9}*Dk^tM=VRt+1Z->L}aI7r}aQdb7 zpH?k0cW8D1^_@FgUU%;B{x_=@1ygTE=>Nv5<)5yP%Hq$@H#h#Fcq1%SHYvZEx?+s% zIbK(cw7%|Tf5mq{&Pvu!hgeH_gz#rlrnN|qf7R6sAF5epS^GsJPScnvncox2A_^J8 zD;`u=r>A`7g!%kqOw$XO4sU#e@I}^$E9HI+$A5%=m$tbk7rCy^ia+LzgM&1vcTS@f zdIaC)-aq}$h!^`@Q6f52uQ*HT`P_Z0l0)pKrXUg3oiJB7UiAeTWqLpPv?2#xtbBe2 zF@)7KiO&jZ!ptTdsujS6Hnf~MD16Zs)xpdX-TgGp)a+}J+?!`@@7~G2yg{?GjLv5u z3hj}GG9}^+t9756-R4%Ec#)BE32@_2GUMC7XmgMMN>JV3@J5^dNVuA3axwfmj8H&D zlsi0*1=Yc>NPoEAY)S!lDk&=wNd@XIS4QTgWg zXU-5MN2q~wB`B4&jOa6ydwC*p?nlhd#h3wd& zKwiAaUUlXhdnHg`cL>H(4a--Hmj!-V;xSLGKjn^7a;QC5jpZJcU^U|W`I^R6-&(?t zCDn|Fg3+2W{iq$Y`bnoTpG;6jhe@g!=?FaToTh7F7-vmdQd_F0S$8uFwzm`;tC_(UcVheSP92SACiZ+vn#f2 z#@}VsR@{t#$xE8m$V;!d%GonS@iTNw=SerYVJJoWFYM$4^0*tUw_5$bMBW;cGL_oe z*=LSqWY&bag1J_7%0cvb(>uK>DNWiI^$%%G*+baF>fe0-^y1gAxo7j@0?exS4(>s6 zM7wVyCF{w)e$pNyuS}VH@7gIJF}Q?2z@jYqidcrLXB~YaS4#tfYnXL^meH~k(IR6T zf-Dc^8d=Mv%8lGT? z^;wUr$d3m|(Gaq)J^-?!@&s1*2z=YPOj}(AUA*|7iGAG{?5#FPwZVw_$+Kh@0CBU$ zGII(%xQ;Kln;&g??{(wHvXar;R8k0`I{Nt31Ixz*=6%ffD0q4o?mpMlyVqcx5@z?p z`9mPKa$+$0w}~ej)CD(Gy>=qlJLl9%zl4d@Jv1o466!6h;)=c#Xa1$s!rkh7S3xwv zZg`rc>n$yIvFxuqvU2;aT@Whih~yA@qlXe*rVql1doR>u#$6eyRP`{pPxQq*4Ly?i zE^FRvl_Gl@qBXI953J+~@*na$w9N4)Wg2|ZGVQH2m(6=}03-CQ!!6M4(7H4<=MC+i z|GLg3`N_p>a!RTdWV~f{>=r(j*MpirtO;&28@6g>barOck%*PNik3yc(h;?Wcp(ge zr|weH*i+<&ZPa(Nc&1yP{-i2;eMkJZ360>79e~H+l-?2erM>jcDv$Y{dfWF{gKlJx8O@anlbk$a`b% zHHIIITPIdHA<*uNRl~$YP}Z-Tf~o6ZmK)a+!8rfC(0lb7=DM~`1U%jk(&>}8a##G? zS!;{y-+yUCufklv%-t^C-bg0%#K{`V&rH@+B}6rRF4aNRrkEL;zj>b_Ljy^C{enf1 z)o|Ww4`F1}Kg0VX@--Ips49*7ESsz-F^ZAVdT*%9*GJY0o@y!?^HAU_1k~`6I>oOv zG2*?lQO^y80u!oO5h3I-!h)%`o6~bXH=Cr6)7J{Zk?dUm%Z)A6Ae=8%}9_R+RP<_W3i?pN!;h35YF;^w;$7my`B2yxQ3yz>O^L3=iEZqEoCT?5y=~cC=6BLDin> z-2U*&_ArZxR_3(dXy(nP2;QHM-vhH~_T~w$8-)oR@D-xmPwx(c5inxO&At`_pYZkIHSY- zZP3$wCNoMPNy?xn`FB5{Qa!ssRNRs5Ux?n_xK6>LaY{$34|OkAs398V98vMPyInim z&t77Z<$S)xc=W@%ROA9jQ$0l{sZt?-s{&m@%%I9jwhRZmt=Q_W6cjtlV9Bt@?S&oF z{^JRqf^o=-ip)y(N5#pfWD3t-pmi1nZoRho>kON{)@L=O#z65XCMy2(ac_Tzp+xa` zg7=51cqDN-hM-7D%*z@ve})!d01x{tCPK#XHG&?LTD1C8XvQyd{7CT#n;) z(AoQ!*k6}YeYZpF{>bG6g=n98Cfa^jXCOb`sb>2Gx)1Z1yJvM-)^n8TDF55*m7q zW&S92x_v0)tW9M^zdzXhLZ+X6Nx6rSUe2oF){TTx`TngN345>UFP8siM*;=&zdMfV z;#=8|4ACN*6%sNvx}$Vt@mT*2|1T#-bQMnZmlYJ(?wC$wx=_lF5~m!S`s*(9@(;#n zq_>q_1Xl#Nb|fY`MNR$zH$rE}Qby?yjlp5of7p@CJyERC384>|G(MW9PNx)v>bF!|Ibb~-$e!68R*zap&NP4!Jnb4Q4W8Tr}qihh?`js+RO`^?;(4?X7sEHJ}&{Pv?jBIX2wP>0)sUg zwW4R7fm}32?wK}Qsx+NsS1OC<{1XfEtHO8tF)#Hvehn8$LWhD1Qs}Y;=EWaNmBX;dd}VU>ykD^ip?Kbe?>{TbZvhwin*;fTyCLG_}*z1-e;y* zB*q-M(KjY`-szmXHza0?sgA}vo)3(&}8ERD|El^}gK6lTrvI$0u{ zM7QPMjQbGwlLS204K0i6Npw|ud~;Ahle4%W8^6bxoVPC_BZ zh4p2SX8?DGplUKx!PVqx&JK1GF{7>4x9{93lL3Tci-0_v6Q6t^S2+S19$xS2l%*~F zb5KRXFl1E(<%5}&LHQdsc2S15^)3g1&MN%@ij7{|H zZ+(t~tqkSl>z?_(&G92gjgCPJu3>swsIz_z%4$Y@V5jBlBlmym9g!Eg9~ztN$Mfy8 zl@s?@T#8XS<2Z5_&3e9-_1?xl|FMHCc8#6*V~o* zU%7fwJz`8~$=)rw@QRh8iz~e>V^|eY{~UqJuii$Abnn*nIH{QvE_dbf=;$2JP(De$ zgsxG^qW28Yf=KhXd3tl%_bobyR?6SXZ?67nn`DlPsbR-sQ;#GmBmXU7Zx$Z)I@}Z64S3Kl)C?d~1dn z{&$R+NnPsK^xGXlr}LCv9_OaI>zwJA;g79(BrmnDe5TI033b)WV=fO?#$R9*|Hcgw z&(MP@iK3{83Riq6O=;4PL|;`r^qk2v1(}$4H&6y>%HRyXPOX?A53JLNj72wLkdr;?-15`XQp1t|}8tt~p_& zhQ~PA>^)^=!dsIH^+?z)ye;=fEH}V+p8G5Zy4V>J=QkEOrGeTF1 zH#@8ty~_O8P`C0s$SLa=8+JV2KFD&B`225~sNVe-UXp*Q%ZS?dojMcVCfw$F%e{fS zng(IVZdsyeWnv$mDpnrszRdd@$tdzbK54AYkH9a1`l#tb@ouj<&12n~DNi|D=+b_& zrl`L7OVbDQi(b`1^}D|hYmWn!H)WUxMu&o@>4e?zge+Z12x^}_7if)hWJE7_cpgh? zjvaQ+{QkT;{L6(Xb_G-5z1oj1s3d`RUteaOR69AOCj|A0yHj=y^!qrKTO@99o5r=r zw?_nz-!?oN{X(`ZTzI2+>j8D59ss4$s7LDQl+;?}JMiW!4*hR*onvrjQM;yN+qRu_ zY}>YNJ9%T&*#&qTe}aF-t9Ia3o+&@ES&Vp`QlSTZsnIj*a2CZ~@SAEtHSO-d zleeZL2X_0b)B_y}TYG(lLH<9n;yQf{cM-E8m}Kpvbw|ccgN92hG_$?)jUunQsG68Q zO_luXV2G`n?V7dY86pGdYTRI<+9BZVS@=xXq6x1tsCfZZ8axD!+m+v^H{wuZG#wZ! zx(5vsNAo@F#SJ|czYcfYLW{Qm1{|dhO26?o+z`7tB5V!OZ~~zenvaN83zM9MX^s62 zFC!>%<1`QRVUGC4w-koE^WLD$Zg1Q|3hqY5njGeE;?2s|A)UBO%Gak!ZJrn5IilqQ zIdv_k62@8?w%;ajCM{~+igOtzD}CmBWKOZbt1gZ+ud{BzzNK`b)4fXoenP93yWiR& zED3$24C9U1a9JfMt$WkE4>teUB_Dlt4~QbLWJur|Kz`A{7*E)ilHtj)~ZF+ zm(|7si&}JEZrcHfJOTP~s+0E;7>{PvXU3D2#StDGNm4V zcdn=x?^_XKsD&$Itwpr~s7hAYz+bAOQ$9tbl_DU57y!22L>(15#wv{&Zb#{UV&zSLM zo0t|UfL*k~CgYi_CQTjX%1sKPW2s&E(~PbN$4*_#FjPlu2e8H40a*j5h+?D~-~CVo z%sYe$<31g)Od~sI0A$k-8QI%l%IBY%r@#EeO{ObuYTGUaZbIHX74_H^Ym(R3EJzZD zoDrB+>oSN{cSD*0rYLe*6i7z5p-Lr@jMIo4!a*5a5SIZ927>0l=nKaR!M5m1weV0k z&xBNHuXRM&q`joG**Bjedi9WN{L&HG?o^wxvL;1*p%KR-di8 zH#Hhq-}Y4f&nGsOGJs=R8-gM`dp6^DC$cDqd&1ZXrb{6KT3H~5IESfBjAdU0u3YuF zy%UBAyB7nNG&#xvQO(Dgeo zoym5ZyQ&WXa@qkhfU#ew0J|U{we0n)n5_3H;6AyT<8+mk9`+Um$RqQW1F=^8gQ3uMRdpRC)>A5o zK(gteSijD9_nZ36y6ke7k_QR+r;e2-{5j!>JhTQI2bNo0Pq@G`A!D;cD>D3 ziO$*%2)WG`Fz$4o3_7WT2aVrqKQ}sLDTED^{dQz9(F%@rInz0ESM-Ul0Vd@=7vYll zi47>La!Cp0z*y=^FM<`Hq0ok=m~s-`=_&0-IxVX;3P_3vi?upo8>8jFxN|McAq}x{ zwYnQPtnB_~lw$~0F5lI;+3vfh@7OvFaL{lA__!;ohRTO~f}e|G46c(g$(XejEWy`Y zf1=-Gq>QPB$g`9Z)bX?wykHKXNrdKY?HfK0&6v}gu49DfG@W!SSz+j`y)0sIx%gZE zg~or0Ou8Iw*WcMN%RA?GF}h9to3!@phie*bw>4LOSvj{c^6P~!Q>_4hF_isrW*$2W zAm$0(%aOi7p)P-=uh^27DtbsR1CPy2^QE>{jfYV2;%M9jw_$oc^AI$Y9m_;}K;nsAMV;Nf_xRBRmqC<09_PxTPz|x>a_2u%TbqQ0Iy-(Fg6VuE#RF1rcT) zb;Y!X+dk@3qpSRP*X@!Yz!s1=Y3pY^gwfXu2gs>kdFWb2wlN2}vw-3D93J_RjJWRa_Azq+NQp=PmDwKi;xGyAJ z-aAiso1MN&l36U%8xd3z;JkOf{YQD>;c*FDcmX7}2uLnw0-@4VcUAk;nVK3tTN$YQ zinBz^pwu@%YbJZo%36x#V94G21>TS8a2SEm=G63&Lq-gZ?Fg3m3%&V-xy^kho4#5X z$e4b>?4#iKS}#G^&K`&0_JM}TyUXz9BLo6JV*^Lr}sOupj ze_t@Y|+IFz6Qt+Hc2do)d5yT2vSK0I&qhB@pac#poD=ljf(c>7cOEBWX2tiIg3pIHU zORZ5)bdZ`I$@R+`nBzpdhR~XOsQDW7*b&{dd|B^fT)B~Fc9{hKMKgMJuU?(fv>MXb z>5r;uR2My;jNy^u_PQG^;J6H$B$yLil>qgQHdrJ-i*-<|c2!}fo>t5Ld^;h~y@e&r zm7)>zynzBize)~8f|?gu-|Mgb+~1lNtX5|HwtRw1c--YeIh3dJ1%?>kX7|j3`(bnA zV$N;l)JH?ug0>8@bYX|CouaKYYD{`z6&iiZzf>P0E=rSvF#ycu6DuR0HfU2t+ z&kOR97Jb={?cD zxAZ%NH)!)|W>JzwHdq8v-p*3h1YEB=q_V9biwuWsM<(q1p#&UT1stR%YpGYsoaRGn z*VmEK{ZNW!F6#IveH?Ode9}5GG|3_-1JqblQhT5Za7$9j0J68((s;!ENO-s;%}J`a~nfJnnf-_QDrOQEZ^`aF2jshaI3SS-bs-1&{$iT^M>Fw(%)fg0vk40e4l z#<7o&y}O^`@tWze+RNNA4C4FExrzU$jaVzioLo%H7XjKx|uM5IU97)Joo3;Cad9Tos;L zfjjQkPeJ%-56$@8EagBDBw+Xt6g_^TOiqY(UCg=;K}tHtnJyps8+!)?l*sS>z{o-T zu`vh??>HUXBWP-0-lLoMpi>==ziaEYqMcP4=@HyxbGavMe~3#!gV+Umgs=7xYBvA7oK!&-y6@068Kc{bB1cSa#S6fm%90f~lDKehsSbP8 z_p3AOlw(OEaT z+Z*Tht+Cu<`9`}=n{&a_UCcZpA24S@nV5I#fF`W-2B{_cp(cTh-!^;4zivt^ck6&o zup|`D{G6z8s?5fW-7M}%xViZG`{0X)cMqHZzKehUditQj6%BI2#d{x7HpqrxzCT|% zd*0i>eGl>cyk8zY-_2|pY!i`z=Vi=V$yp0ruSl%(aP+=>e^22)K|tbvxTzvJghxll z1JvjNOMj)AUz7%m`rDa6=pxJ+^#9y_>?tuCJG-}ru`?zLzo%t=I=bEaoU!Bq z0V;pB82Xj(Ze&+a%?9p0&RX_8DCK}g1Het=)AcfoeG@%+&%tv~3Xao|P==_iqt;+X z8=^qpj3ysHJnB&n!?kqMS4RMX6i!Mj0K~GA;WW>CyG-PaN;{MoLjutA@cKduzItkg zRR#1Z`*KD|NyRP@MdCXsqwfJioe?-V>?B)TC?7emt}S`H^qmoA((Ia5RobV*g{ma8 zOu2VB?3tok-_<$-%dnp6TG0NzR|V}Robz97M3gXoe01M@#^A7}2=3q8?~sZz*r5M0 zqd)=uH@60yUi8q06bJ|=8wiN$KQsFLr1~CEV!;29YyMxQ$`#+1^Oktp?IU`xvYggp zlIr2QlX@dB`INMHYg1{{F{esWv@MDp3NjQ?hvZ?2a%%i~y5H8O!iK_R`c_XOAn8n1!Fv%m*j!zb_cY z5aGqEl>rWgb*H>`A4yo@8x;y!py%sTB9evSlaEpKEAyw2?gVR~ zslQMvw2TwZiL$KHVpL~$SiJ0zQvz%gr3MtDumhq;I42|EjCe2R&zKORk<4E_PGl7o zYA9cVh(tjqsIojnp;X2;PDrFO>7Wo6KpfeeQvqCkLaq=jbHh^G9)0$<2qPUa7lot` z&aCJxuR+)rlW>ZJ;fa?KR56T=zzq^RA*|r#%n7tK7R(Ppk|orR_!ug`-s6C5jVLBt zh3c~Il{lwO*(F#k%w$sr@2%GdIN&SW!*YBf`z3`1q7-kh2HJP2fIB0iZ&=|#(4+rU zMgh77k2C4q8CT0j>AEyOTa!wj-5Tfcoqg;0T}R({nW&?!!*1DEqfMX8`unvUsKHwN z>I~t;76nw3nb>1>vig4E&18DWBOtE#A~Jrb_UHL0>X;bV-dk;TH^FHRN=i8(bJ^w;~B z0AQ7R`a}MBU%%N2VPt%lH9@EJ@jixdnHmg_&yTpPvje%kZkYJIlI#8n@pHc%Z2(Wx zz5z|}21%mA7ZyGnFc-y=t8M0V&^6)pfYNmnqhN{JIeLQaw1 z^_w%}<2Z8z3PBw%Pk-iP`FjJX5P&U-W)^cIsb;7Rzr4XJ^LujntPzr5ke;P6B$b}! zE6fy7F#ae02zZf;ah`mmBZ@i|S8=ap##lgwXbkCIj+_&(8kc>)0;mOLXu%mFD}wMk zK-;cOf)7gdv&8pOsK8%Qi!+b$$<seHN!gT9$iz@OAu_QYs(9JMCw*rv=iRi z{t&ju@4RRKxqz~B(+u%#0RZsvqJKh4NE}`he|h^8=@aJQU0)OSAdT0LY~1&Gx;z-) zUKLR4<3($XTO3~h8Eyu>Z$ILwAKe&?w>Z2@1qI#T`?t@lMlr4X4IrNcTPU=j{33KA z;{Cx25AK1Qw!y|Y_KzZ403MVba}F{ddB_NTH_AZe`cimD|TIU9o~7$Z)gCXm6M zBU4a0k%dcz878ln+wwW3$)X8Pa6>^B2vuRGWVoIZuWK|xr%}jKhW7G+#JFLFf)5dp zucKWPCsU}Py1tku05~DRK+wr4y(Jqt6SNA>3h66@>EvB2qHMrXMQIbOEoJ4fQRU*u z9#uBcbJ2&I8L}d!Y`E=IERjHo4CM97P?G)>P7kWZpTJcRH#cPbf|oxXMO&JajvBB^ zbpqO;pCvR1fRg{`H@zG540gJT9A*Yi573y7Ob9fqf)tdS0LU}Ni|*;dsQ=+W#jq+?(s7b{L zf@OI~;dknW=pwbug((|ZuA!raaLP{q3?$ylx@9k2Mc$&={Vk`F=9xE#lLV%qa1gP5 zI%o7s@W5Rv0!TjQUfU&n_oOe5>nF*z^=~5BQNI0BB9r*gUtYWRGz%1x26w1Nu>F!~Ym3;QOJDQBQDDn_u3*I7iQwc~L3ao9)^ z_ygJ+LE_I`J$&JN#FFQIy7<7<1Vy(sO)>^$0B>qb`P57iG;(%_+|Q;8AL_k*yTWo*2z@wA`UX6M_3Y^&gle8Ld4ID^t*&JCpD zDPSy&j0o0OFbze4{32wX@5~ZVE3Tvy&nq~?t&A@Tf@=!WW(!rA%tA{SsUQ)cshSyN zNgG)$p|fDeOcx@{pw+2}i;IZJ4I-9X0PRZuv@U-sedUPNv7MfdELLD0h+#XQbKO=@ zNTpiTxN}N^>T=FBclV>xjA)I|BYU_7GobmYNwh(XNj8Ng(nx z17lUAhucau@%4-|aEI$K8Ol(~e#!l&0zwSH$b^m;f>GO2MYu4!VTB&TQ_QF?0E8x4 zK&6jvi*i*^P|dU20XX3fE0`Kj)Szk7HP{mKWuSY;ymXm&$7najQ2EU6L|JKLitj!f z;7F6Onx45<@~0>8(KK-6WH^2;Bh5Ukl%b552qBJ_1Z@P6LkFXzvse-nVG!GwZ^igRnw#GFuF{jW*A9o0$wxAw8fG~m<&V2vFher)+6n{y zfkn1lG762kwQF@`WwguOBrzZE2xk@z*6Y#VMqeMn^Czijl3pyZj z&qCb8;)-chF1{(wu+%o^#aq=?r(a(s#)R(*N##?G1J1f!J7d-ZfQr!O`tZ_^zBe2dA=BbP>kWEUF2Jp;F?1Dy`CV51a!*Lspp_qSD^+ z(nA}`*^VOKEMRm=k4uop(TTv*2hd^wE;zK(j!GXf2Ghqmwh^i506kN10eAzB6g;C+ zQt*#tRNuz3oc$jQJ#a)tU;rV;3Q$0k`N98Fa9L*)DxUYY(zV3^+v1Di2OE ze8n{I9o1FBbfslV!1S%PVVjfoHrrkuQPa*w``q3jQ2Dy$68vNUP-ht|y*9J{fE^U` zuVN;`5GBwwC4RC7nSgjM8;`JKvc!&+rux{B9Pdncs>IA;CUFb90%yD$T=lQswWyn@ z-;os<7q{54sVjq}pFrAeAf4n4dgdEiT!=OH=Uf%IY(>&mfP810HyqG#owM|Y)?w(r z$GNvTc`G?r0EYVXO=cG@tZB8o^fr%w^;w-QDDd>t>MHW z_uFNEelk;gl8ExIg%RGr;)LDS~{bRsCxr-)Yt)_!l#ic5VY}Z zN#W1Li}s5-nhN#Wqu#hWN2~))a&%2%K}-V$g9>JCzy&_DCp1jKyt;k9bc|+~aXl3a zOmf?6Io3r_3oJV*VNg61T`4n_KDZvp?krlmrsZ~RZTdJPLfpvxmIh>f;D=CJiJmGq zBO~GKNZ+YO^6N-7kz)rOqvdi`(rev)we4b;W?S^{wF@j-4NdCas?td{{~h2Rp{)Wh z>a9gTz=YKFUQ2-jmyf>3Zmk%oDjGhOiC$E3Hx# zyKT&MxJ|ma#6^w*S&Oy-iGNy_0Ws#8EFa7%fZEB4+rdt5yKn7Q$Qu=s8QwSCY18e? z4QjS4h6(su6~+2G7PQXwN&~FaO_sjG{%`k=p;VsL8L>i8E7OQW@z8@R;}zZ4B<}iS z7>&%s8+LhTPePY8;xcc>Sg9qw{E9pb7FyMLN5H2PHwP+tkhl>y zK-Lsv`53nZhY#1w<6;v&0a2Eg`3{q`1yc2Z{qxqC`5YJ%!kci{PW@-Sds|X2XII=$WQ`ljVLhLM>*n(2*&EXa z-ym3q&ENw+N1<#}2+4}ifa7+tPxKvq3G z?;R$U%d`)jF&tJ~8&k#@4c4{)0y+d+8ql_uU6YGz5Eg#CV{7KDbwEqmbatQB8NsS& z{euhYKE=9`&TS=Uk&9CiVi(v4oL*FCY$Jc|!o%XJKATl2L1fy_`!}ERjJht!SgT%& z13g^J)5!^}y=bvDZEmqN)f6-=U{_q22WdaIE$YUYkiMHF%V@B=qV2)kUZy@(^}bx6 zd;F&5FXV~s)(!RsM6m^qTB=U3lz7x3_XPxTh4!qQrQY{a^qHI50IFdp7f2X3?>IzH z^^1zn_;6GLe3{m0=kD)kr%HrPJ!2KT#uTr(hys3KHL9t1)l?uWEq z^yMG7ug+W4^SjI=P?c_ICkB&w)?l$Od4n0Ji7^ zx=SA91>xW;o;aSMvWbcdm_Xj$5Fd9?y!mZZH^4aRz0}bTwu(;q+#DXU6vru7+^~8b_UTp7F{#)^X3O#jDRWnmp>eivu;=_;1P!DIw@jd3%)W_^$LWK5+HJrscnT{x zyY<=c0vQf1Sii}-YvYTUV@9=It>)P?MAMR+^GI8*GXHb`c?NAtNLDB@8Nxnc!o?{f z27IPiyK|DS%0HP7I7x4jY7K8?$$WMjo4A=GGC1f#03b;CB@F3`*fnczcQ^0VHaVdQk(iB{G|Lw%#& zW=q4>1)Q`;U^Xo1D;?^r@0?j|?-JRYUILLYsYahkLUlwIL=q(FafyF#LPj(RG(bH( zL(!j~nM z1)i&m`INg(npx>iOfvgoFDTEl^&R7H=@J;M=ZUcazO8*|;B|4^F`VXOzR&ti-Pj@u zL!oB)N&ilAVFWBNG!iV(k%&|70+| zqaWMI+vMHE-K6dACtj^%Oe*v(sfTOB1>yn@2s61$5lLn8Jk>2=vfRBBJTtSjkY#`) z-jEGmtB88Gn~w>dy+|suu1r-SEnnNu>+E9r-8n-}n!h}ML8hcpdaoGq8pBO109J`W zC}^@&KqKm4NK)2sf8Jx(Y-0U8V`rsXMEDE~M0tk?J^*{Oc96pY#Y|=_P}cT>*E*LE zu++hK7s^E!&}RXx->L5PNo(H|7}InR)d^E@BFxHvrvW|{j7br&(#V}9-Xk^U5(`K8 z74xhHl+WwP#L#h;X|+>&57U}0?4D!w%%#8ce4$@|KOv{1clBuu25m+w%%*g`l8-Up zX~C@D=-a@++y3yD)i*?Lwu_Wb|I$DS(16?z8P?+3!i|-_(|io1suSTS$vUzU{pwUl zY_jg1-__IHgFpU#=Xj%!E4c^BjVm*Qjm5iLHJ`dDC&nAXWmcU&l>&cD%vO2xkK$e# zQ^=+GZuMNk?K&7pNN&&P5>r8(ifN-(HV~w>Qtng6ySYCpj5oz~e#3}q!|uEdK#p@Z z%uVD_eDtYK?%%DnefUYSY;A|@3ytY3i+0tcQrn6BO=!-bUPCLW`z7R&)+#>RG6+$P z-z8~$^VK5NRabDQO=fqkg_RL^o4I7(;-+3d%vEjIMUKz@YzwK4;Z>a_#}e%yg0X91 zHf%>tL7Y{zIb+k_gZC7#x!`jSs2h$=a=odu*%I%oSa9#j9(OZhFXn`OVBlW@IevZ} z&tHFPNO_J$x`}!$>)aUcL@VDvSPsJP)chp4)tlfYe4*$&wV+|>b$dns&;ng zf54}rg>RLf2Myq9 zf-c(Me(PUOL`|I7;7(sGjdcp_9c)G998o=QF>Viqvts-#W$mIzjzyVDL@yQz_4&~> z^LjOcverEIuB(l4z%_gVo(D0C+y+HyH});ANM!vCd@KGp^K2{(MXZYn{X0Nm*OsN34S z^;|u8ZJfA>zk6)}Ml+%v)H9>Izw#B;Q@1PHb!<>!ISjH|z&zB{&$6DJP^595Lw((^ zSXDhMJz45@&?LLHo@dD8tST27dn}&;-QC|iZLwrufV`kH5K7tvOCEcaHA&cxwk;gT z#eoaGdPmUEBq}WrH0lIpx&x$r{Ud^dfA2}&@oDyk-VGp0MhSXW&WXN26Ycf&>c5}Y z8Klo-%6WEX=oH9$X6*?s`AD*gPfTKx3$Vnt5*e8F8&GuM~O*#<#8NaSvuo#lm0nzfb5%V_Ce#zh1Y`f2GM2RJZeUN1=js;jbUkd z0S-y`xh#$-ys8{Lvqto-L<*Zi4}vuRbEl}Pv;-Jm4w*EN;q1~$@yEL~;=Eb>`$X&Z zooVkf)L9k`Mc|}@{-q@Us++L_g>xGPzuZ!F9ubsDuG(u$$QmF#ZdLFHk>a3Cfaw6tHDk^n!6Whr-84}i)Qkluw;WoYv%eMI-AA;cJPa< zi)uYQTOV3n96h-fkP|tpp?f-=RR!GC7d{=+f@%%@#SxxURHKNUI&GfKU{D5|mE_FN z{N9#~!k^45_9fl$vo2KJENU?Vg+U31JRQBe%qNxF%E_?Vv?7kPfc-$GUHE3BktS3L z6F-kE*h@4BZv{-%-Rt+O3S<5VnCg0L$NHT}6Wz*}xE}CyAq4ZbLxJ_ql@t{&O9PHK z;F`+o-e-yevt)zU7|4NV>J44qZQL&F+UH5ruztB5NLsW&4XUaB1(Ui~7YQCIgrPEBP zWh^1ENDZp(Wf|dYqwT}XvWIk(bdSYoOY0&#_lNUfwN5a2Dk)H#WB!XS&f<=MISeOX zC&dKzriCfDrb5%|EI}LEzX~G;CEWH zthgNJx#S)!px^Sv2)0ENl%!dRjTWzpD#Z5kPG`8>Z_cRPR}iCA+Hj271@*C$!~i{F z4OUxVcwdJL9l2u*>gz<78`th%^5#+C@$9E}bwnH%fyH>wr zvU%)Xw#0I4_8w$)Q6lXilqmuZ4R{nNv=I791ZH>w45Sja&o}cf@FK~I^y^ctIxWq^ z4$ewIDygsyw(fB%dMeB_(BNSS!4GoAQ%u>Z*7$BH;vg=@hA2oyt>l2(?v8s6{tg=- zI9Y)hV46@woe7O<25~}MmAL?q37_qSL=)yWmrAnCXp}Z?R8&G+53#x2c(>iqbwbb| z^OFGrXg3j0e6#jxV?(sBk>7e?g>}bO$|&q(Yj~*Nt+&a=s>3BlYcc=c9j7760+npY z{e^e7iwxNY_0I&mNSx?^Ojv^TJZJ(Oh=HJv<@8561mGCb|56=N>Bm# zw8R$g-_%mw^iof%x>XPH?)+I>F$U!)TUtK_6uvQsKcSOJCzR+|0JWoUp1d{D`1CBW zBI0J=sZCuzmXET=kEIN z+!rF2fxblkZZU)TQ@ZNrt0km(!qn&}GRlRUj0$BI!z5_1E_>h)y1n$)OV{h}8z)`} zAgHx`&1QHPjmriLGf_j%#(VJ?*dGVZK%Ct1Aor)(@imz_T$FcbJP)pXgdqZ#pm~v^ zcom+_?fk@W+DvCzx1}lnNNIVu#!51}5~=z_d;AWXMK%m-YT3(%RnY_)?AUnmQcx_^ zM_Qf2yv-S4k?Ng#)$MU>Y2l{9@A!@k_@rfPmMfHBggp1is4@&)p@Wo9{>E6sA`==} zl0V`yhI3`ZEP*Mi#AAxH3@B?Xn>Ek-d^>tB*{&Rk^>2&%44=!#SD;YHkC&JKrz7N# zRX`Y@laCvhn2VRs_I3AjyMMS`5p(=L`eR^U*Y|TkdFgfwc=A)4 zwvz1;;Pd}le|^(k<9>VX526fLi2HoHpJq|0UsS@%87vNMH>^c#;Z|ns#X8+vk1`=S z3u|A~I7_eJ)%DMMBKi0uctVsj9ktP*yq&#<3Aiwr6aV)={+BlR)e9%b@8_cnfJXpf zv20*QJKOxuJ1fyl<>f@=r;6|ZCwVo{JJgv4aqN6Aj`HERPTpR_-Lh9_2A*+cwi)bQPAUmJ)Nt z;g@G37mlB{2xu^mAI0ki>{C}&eax8<|LotDXd<=nrnEzMh$Dp?dl;r=ERaLrrzXx7O z63M)8BlLa(8}s+$y;!E#U;)4iucx{@V;eL(e-^qsC%mRz5bFIuHh34c3Q3GK0`^`v zyt1n47>abwsi%Iu@F|%35-(-+22^Sv#|+P5YSE#%bA5M5IQyA!82X$j+?tb^(UQAl z%4?~f^JX{ep6kA5OlL03TGB7-K7esx!;Wd_LZ``T@}tU@bt<9~Km)GxSc8@L$pgG? z*^(Dx4x~ddn2&JMw9%L}<)XQ+$|2tu zSxA3rN^&aEsAE74WB@cDO7=UcFc05UZjjB@1Jq2S>zO5D21=dV_G!|U5*&irTe457bDF|i$R_?q z>Chi7XTo%(Lf(=@$EY<1SxcrA%$)RY2&eKy1R$4xQB1qJg55$WK~PFPe#?$AKR_iY z629ab zgbQwhvdWrWLIB#~0HG}yOg{$OZxS;{Kd4GMrHPuyoA7=;%xDmO$Xt3}jZDMZt32Vj z$5P5iwdrB)+tV29b1IR}V&5X$0=-r?+WS0$2@uvMS#Jvs%3Re}I$0h~nq)Q@7XJ)H z_GA?SEgKy3tcSBmoDW&6g%*S-9N3gUUorLVGQBD?d*p?cu%ZOp zEqMYl!~uWy&z&7%d0K>lDuBcWgd7v8e!ziR6P8~w3%T|o92Q0YYX7Ef{9VP;ABrH~ zlD35A8W2H_Ov<5%%Bkhqx*N7cz8W5T;2@C(4{9t5pyAzr*I!Ss=~1?d*HvKS7H zVmn@VXHqSyBiJO-{Ba3aY-wv$^|l@-UhzkJTod6 z{OPF`Ga)RvA-;lVk=R>Q`2wBFcqa`_JS9;)2jraD2Z~+)kwj|w*DS9e3O}DMMD+9L zoDJa4;P3MnJP)7Gdr~h?cz++i$zE=+zn8Q3OORflu||6N)%71tN(~Fpm&2sqt{&st zKAwJV69L`6zj1oGf3Kr*`(aiCo;?iHc#mYq#tpEx4x-QI5Z)ilrhsKi%pF-{X}q+H zyeeQbajW^KP~@UaW>xsbSXgB1OZvd`iDCgsL39#d_*jY-hK6T`<+Y~#CX{EQT{x*U z6}eEd4q5Q4hlqv?1zZ~ymx=YBz&+7v&T7m@-ZFT%Ppt10tDgR&KW9|;qfy)2{C-D2`{{Ay}zL}LIi{Yk&C!yt= zLe#Q#gj$*XU}MJo;r6?7rU5KraX0{I9jsVI>-ka$7j&h5KoiB0&{2Z9rsmFCX8${I z%R2=nr$cyX$qE>AD?$M_PO`X0@K-Q&DbOn&gCyZ}>1vuB4AM$P_yjjA({M|vs+P5+ zIK*iau7Y3;hV#L09LR%uk%2(zy^z%6oM_YeAlBVI5{gAEDXaS{Z1d#jEFEAD^W*#u zDZ&5|;sxrt41mI9L#weOj9g!VF$B85aP^uB$XPi zzDL|wp_u6jG!6uLWfD94+9*JZuA9_r!CDkd_ZPJt<$eH@px!=!vdJ8N{w-i62Z_(i z&prZJvNba$f7E%h2uu^pS2GVPlq$Uc^N1|k{(6oj*ObF&z~qnUKm&8UeL0Mz0K_x& z1gG(*fxzewpYI)dv%(mb*n3pq9mos6F(RgitVMur+t2=7{9GQdQ$64#GUNDfYRuyX zvQJmX@R^4n-%$Z{{5oky7$9$ekPZKFHkhFK=bz})*DM{#L=0n7wtvItTb5Ch2+I&Z z#Qh%w!9VV6i1bJ%Fb}v!9&8RaG~vV8{UdJ97|R=?hR0t2+t8`%f5QwzEg2YWhM@dwxN2<-i9{ILVyMe>wF|cP>PFID;f~;R;zq z>yVeOSkhJ!-YCTESa}TmZU83`=pu+&e0zK&;NMyw6naaOVQ?$J0!5O+vB|T(wXL z>CS-SXkVQWujjFSE0W7?Fq<<1z%y$1zvHur&YSJi=ZzxJW8+F@xHuwmonnGudpypj zQB_ayhK8Okxd`_Yi0K!&uy)E7{xXKE3AGs1A!#E*wE}v1LO$j2FRgPfUynr$vU5irxwlOg1Ba= zY{5r*e>@exVB`iGID|96>+N2S_(gf}s}Ki!2v0YNC-R6Je~@lHEPs)Uw6_oUvO{D# z2`m^3s{p`AZ>|?VTjOJp%?~xTDXZb&kfpTDbpkGq8V=~{6pKj{N()98V$y9O_OVn3 z;5Z_NBr~TI9S5^gvEPy%VX{G>F$;T!^md)|rbs-L$0{JebfFv62U3}|kp8iOVcP@E zqs!!y;2xPZP6u=9rD{A$8t*yo{P*o7?x6lIfde!JR-sO0&8k6_AsvnHWZA$s`-Dai z1gVA&?Ugur>>*B@w=mCR;LsPYqTRb0X{_jZRob^$AL7?m5=+=Sqs$9ZNLW{QJs`r< z< z@c@~R*w%7P_vv=l3nR?!N5pt*5#T5|v7DZygL5axa9Kx}Fbv*8W6Yw+V@gXjg(fU% zSdf|o37D4{zrUQ8RYQ=8960});)i%M3oB(P`Wm8VvYF1HG|Aks>bH$*n3*-%j2qg* zk8fuy*knE7e971afvTYvJy{-P)fRVvkpSEiJrkVh)x+?=JnzVTW^dM;P2TPghQ|gJ z)gvu6>pv`MQ5G^=q0l@?iaOR4bg4bFF{Bowwx=EOjbHtKb^V};Dr^2ooT4wq)w|p9 zn(B19oZZ=x_W7-5DB9utRX#Q_G!#~P7?D3rwJI4|FlY{Q-`D0?+z~;dJr|7?!UkBy zjOkNuq?CwZlu__cJB5*dptu0Bm3MOWJVyu;Nn<4A!Ch&_a0w;_@#hywfD{zqHh=}D zSGu*^Y*&klx8Hs3Nk8df(e^qekAu34D~%|lWpBdiXK*L zBbHQW)+>$9G=Yg!gRXY^ha$rg5&^!Z+)EZ%0;qf%SauDkRYVH`R(BrruB>k=)bT>& z9b|EKstYDKaCpwZ!px^I265}>AXtJ01^m9l3yT-c$-EjR%?Jt>mc)8*cV89I1C*BI ziv*_oAdG`Mv!RjFOML~Uj0QQqrs=o(xqguA@oH`7q)TkOCriLQt*>hRK7g)jP6T^w zg84df3Xj?QEz?E*?s~K$t3`4O8YZT;8qe+0>Rr%Xp$)^^q;@fcfk-ET<(-|nv&4lpmeBx+ zQRjuOqEc}?lyN2MvXXkbP5M|A;0)Q0-mKi@lQzpkejB>RPJ!5qq5w%GF&Y4v6BnEX z;zd@?>k;%?GsVO@8nTRYWZR2ZFjJ_>LxyuS4PuE7R1mEXT4mXOGXU0qWeLx0A{mBk zaOSYlNz$pEN(F|Pt-z<>pHyyHcqW}tdM3}hzg7Jq8xOSBN$zj1d2}le6)|gKw9}ZH zZlObg^|Ecrju#b24lc0o1W5{{s!NqzB{(N#G4S>flx*B5DWX}^_<0MAG2)<{MiC)h z{J&6jebmJM0#pup3*bCR?9};~JYNl9S9A&C=tfOmGprfXdppO}8T!0BkvQsz^!yId zU>EHsgTRqVAH;nOfTvLa%`;CYSC^r)H?ik3Fr{In_k2=Ge=JL6M_8|1C$S~8)JCw# z9E+M6G<0lUv%ztsU489?%q8d{H=YFn>l>@VTFo8eq zOV3=`jv@8^XC}oZO;{|lhNcE{a|K7ENKrT{?$00r zRYmFkOQE(=3SD{?;gi;V?Gw1jGss2JdwSbxs2@x$hcaWwII&k)^-68CK6XWp^Uq#| zG3VZNvS|5}nWR9oD`sjnM&sAd3^#6$Ya(_p1@QZyd};Wg?51#@I;oVu1TI(usUL2J z2O9OA7+g1mz5;l!9APj(o9WzdRNlC2?*53-0BqcDv92E1?N0qdtzzO8NH}`qzA(!j z+5@e~l|x`4-pbxj3iy zsmQYGie*(f?-9&y!g1p{bpV7Up`QQ8D-ro5VNZRW9vtU%3(qH%9G~!iDxY{#{#{|H z!X2-@jj@%w5av}t#lm3fSLOuq#PvK^oo|r?vR#rIC_ywsPB#<2$tW(43YTbvrgy2I zOsH3Fr8_Xbz@n@=l*awKW`j8jbWbi{fFrS{v8Bx8w(d}9YwgdUqf@Bqy}gTmQ!ozv zt?fi^7l90CvDI_01yp-B6oQ&_%|1AlkpYCgTL z`-vvZHdX&3`Ye|FthBA@!MP+c~CZy{i8xjs>!n~!o zTb6#SaXfmppQUCs@JgEgwcxlv418FmrzLCC2OyNZy)#D>2@e#9^Wi~2IJP({v`GxC~5ija8h( zQK|Usm9O_@;j?chIV%#Y(8txQ9&320mWED>A)Do zdyiFVy-K@6Z7rP}NqnuxT5V?c;;&C^FTkOuaSS>IUCyX(xV0vHBp)&7qP2w_QA16C zgKX)1cv7ep!WqBhIZ#9EmvV1WG*!nZjK=K9X2F8_rd46?Z}!F!iOM8QIB@2%VLaNg zFZGg;;B|L{YBShj^0c(UpS6n$8i<^4_bbmbob=PEXE#%Z4TnPg&@$74vs?8 ze;`t48VyN=l^iskcC^VE%XLWT*v$9V-YY7PwNyqvNRwX1sEZ0-b%10y%LXZ5t9ffL zX=thh>UuR|nTp!mVk^VW5{%SNL;!hoE>~+=E1S;^Gi{zFZCbJ_a~kb$GEODYv!Vj4 zpt9W?PV%sQi*T=Ze9afg4ZCtz5TIH`n(saFMyQ+JJpm355i@7spY($#n&(4?V)%;9ZhU*pKegDVHYIwF+QIG)FIH6VOo#8!u`W&m zuTiic^w9!l#>?-nYVGIP9a!{CA{9BgYuT_ORBZOMSOG_AL9V8`_PO9IuZKT#V4{Mq z%Apqd-QuQyVy{E>>inlY{z$T;Zmey)h~@2quv-v^MRc5+YQ;^b zD?=G-!(3_)WXXAV+Ogc4faqf?kA~=(=Z)6k7Kgp~TBmd85rfMH~Q*pDCEsi-lMKM?Ok;NpMq@|iAy!)plxCJiUA{S3QUgUetQ6s2gn zu8q;@Tw;EIrJBcYrEMe=#H?mEl_O=x{#C(!I}2!Nevt=9lHLZQkX>=S)k06j!(42$tCsS1 zYL}axZJ+6AQZ&LN$T!#Q=vg0=N7O$%8H` z0GdQzK5_Ym1{f3GJD;Z9w$k=vD4xk&f!6!N4E3IlkCz!R+<=K88VyWMikkGu#J@_Z1hpdBCwqrb ztEJ&eZFGLAgIm>0WNXSzj zcN};V6xq@XO16Pq1`%8@68;AI1rK}IJLroPh)K0DIqjq3__Zl`2r&b*u%5)CH9B1^624ecG%ItYLVIEH>*@$4^3+pxdg@C-j--lH zHLX%*ztK@uA(oaI z%@Vectx&IZwR!A}wZirrqe26Cc?X%M7(C=BR4>k&U>ZB07oot)t4#XVwqLS>T1}DW7o6myNGDjixWHt z93bAG0D(-;Q*9}!m3*qHULKio#!fX3j?5YP?FM8(n|A%Q3vvbAugHcN1GdJ}xAPVe z7e_ipVtu@EoPnNwM6m+XV}WN_NF9Fbu*gx$sGb!}Zu+m$>Q#jptAPTds-<<=N08 zX^l0_oxCa&im|#`q!9;iZ?i&GD*ZOM;SPj&B^edU{$N<~Rw(j~6?6 z10|&!5}Yg8Q`@U>1tqG@ zHDA`l`uK|1^<~YS!Ws<=5Knf}O4PpFSI2LL}xoyWnW;0Gra#V1z(ban~;sZAM^%`VI0#y0i zwTMv9AYw!ekUY~%-oFDXzz7>ny++w@=5QL%FJFuR1^IOK16Y$NpGDf4Nc?3QTb*wrl1FXCfh_93&NJ|m z!Q5b^OS~Tlm8eHt0aMB~WqXZ~_`m%(s;raXhhE^SmpF|!SJGYsORaS=Q1u+OxGwCj z<3QU)0<}paV0zG%&~}Zu6H;vrO{hRY^Wt*^g3Lki7~qo~3)LBcLCCE6C6MQZLlL}8 z>tAY?%iz$2mPRXUMlimlq|LCRmC){kdJylFYw9_yVf%Ft>mG#AB5Y~GxXJP6kk8H0 zY$;1NJ6o9FE%?_{qq}E}1p3p^?&nn}YCly4=dZ}JUWx);YS;40K6B=T zLFiN9S>*P{(+{28zWGWG zvTKN?08Aw1^oe}Lt2sw@#>s!xoPtd%nxt2epBYQs5EVcZrbfv+mYt_XQsG9+*@Ku} z38k#&(5|LaV#036pq^ue&d?#o#T$#34X1|_$2fiKW*E6U-mV#@ziMCPKc-JQ=~{cJ z))h_T-7IU=Z`)DLmc}&EO8pu_#=kA~ue@ia0IdM0h3w4JC$aV>H%upl7gMcRdpCs- z0NS7|u!70{*pT-VREF34HTUfo@@h*)%_8UV2{a)$d9}5ph-yd;jcgvcZ|` zD1uM~tDrjQi0HdS-1p%tSfE&R*1uMrk57MEr2)+Ocr$P#o$CjNxAI|TYj%8arXeb8 zJW5rjtDzVoPKBC zfi7{dj(A8@=|CG8iBjh};=^QC+IZ$Xx8dJ%iHoyCI%JE9rbGpgK;`R>=JX{1q;lKS zkdBmgQFN9CcLttifoN&%;--|4B1^ubXnK=_j=WUWt7b zHiQKsNL&uT)^;c0oohx4JM%hg#i*N4(3dFp$=PaNCh537N1iGJ|5&Gp zo1(#jY};(VcwvV76&`4ume|8jC2@L57b~!)b(n-J+UDpZ%K$aAc}l+PS$duKu7Rt) zc1EGeu7!UUX9LUdfTM)LU7dm~gjoGZO58C#g4VBx5Dm8%mGMyZI{gXMMOCJy|2Ejs z%2vU&t5fGHuw?%%&aft>w?fLpb8tiW*{tEpzIr-RI4G!1{DruU_9) zUF$dA^e?h|tp=p_9ftySj|~f`>@j`x_o_JjTO>5nU;VBlZ{?92EnyvbueHAPpKnjX zE!OzHBBe8yW^hMlLvj#dr1Cdz;Zmj$7F55wb34$_*OW;%G})n=Am!nYvTq?`Z0m2{ zhC1A)drf1OC^9pGRu_i%v_fvH1W?*P_kqGg?B`KU-HU1^T}i;OfB+ALzrF?2Te5Mj z(hI(@LyGn&ZrJD9!(`Fbu);<@U|Z^KTF!_$c?XK)cs-3`-7EP%`j>$-44N<`(U^mC z0$ey|MKPT6AT$6$R0){Kuuj$*I07DPnQYswJRY46{#O>0t$gg?`3;>uiM;X2kD zBtj`}Bij6YIEJBTaMjolZ}4ch=>F!?)0^ws!-%8~N|Nf(PF^!Ntax zn1j2`r4W#l3>3qAZlqO4Nkx8dc2f{4493=D(l2z>`>pP>?3)aPvEyB(>;uWP_I3K z!uqm+0$zl@Co{Gj__O}`wv5v28?6QtYo{05!V?LU`@1O6gP``x_cOw?FXV4@#AhIA z2NImJmtWeI=|OIl(BZdZ#IUcMBu8nUl(MU5k?}9I=4~!nI@x)}34tQ$K8fJR_w}58 z0>zM&5r=J8J-v6x|8In6FCBg$LzDmnq}c@ogztaUIwTK1O#i=B&cZjc2cP?YBYU(f z%NT5ACOpq&iW}J|<(f+H{;BHRdY4gP!+M6AOHWCaGpF*6o#11H#bfcoHaT;*h)ctDE1 zdxb+49PWge*u2u!i(1_u24zdA=K!@HCUn%(!rZE#iIP9jRVsRZ|58wwmrywG5;@0Rzj0XCN(x--eW{RgDg>oGXqT}IlUoKUMy)s9;}7#&#(Wi zX3?aYX&5Z*gK(#dZ7w1mTWQZS$7zndEo6-w<44)*-sWXk^BNFa2W%my10r+TBIi0qViO1^HkYk~m2DO0ONlH@e9dF6LgMY#rfx*KFzJoS^>qk#! z1~f?Ab2&#`Z8M`aE0JaS`OPkcdcc8{@J9`yP@F@J)WHACnlSycHB*(}tB!l(4Cm(l zMT-3U=BcZQo(Va;i{X^9|KA#e`*q>gmg6G^*sG960}xKOi;U@9ak2jQBdtr6Ps8nN z>Jd1;`tp*r$Ej%f?XX;*-X%8Xra>|%k>%Q+{-UV-yqmdFSauR)ojGprDBU&?sM=a(J+gv9j3(uW>m`eV>JZ8J5lW?E4h=~` zN8CJ)Q-DFO8KyGqZ;Pu9v!J2242K`*?WkD~dHf_9$Ast6(*qnwg3md=+Z@%2 zz#(%VD>c132=bdRWAg&Ij|GvVG(gjw6`)z&&ce55tL$eYfwwugVU0f~vt@(dP)0Z| zvMSRHVaxg@aG0~GKO#Hp>9Bg2gTWOcwtn7g3PjKk+&J!}bwT>+S1h={EeSv9-K-?Eei+3RscS? z#2a|NYP#`(FzsLW!J}lJ-B<-y zfM+_VmoxIGvpQmixr@Mdkgs1))JMD5&%fYr9w9z`eHjdi3wn8Y1ccDa>qmZ_6nVRO zx;wk9qf3|XE}n#WzrF77YQLY|PuaP+ywAY9+)iYHANGR}zez;XtoM&C-E4X(c^-HP z?FQ8vuvGRLwnj=H#~fOy(-cV?GkHv20vqa7$+;qyzZoX#`A12~Ef<2M4 zf{_(}tBfZ(h^|6z_cq0xK6HOVS`rYl*9V9XGs-#J7(YUe4^vvM^zV%AEra!b9MH}L zrOfgU@`4kNxmK{}VnO-+MAX+2)MOD4XDSa_9ms>1n;j%e80e8G=?q05O&OXa1VD8W zH&k-@XIC%{oTL=WMlB`{af74debiH*K^uN3`qDJdo{LkW5-Do$BEWz%wGYXB`wFnEoEfW|Y8qXDs(2gOH;4FVK-cyh4O z^E&h)T(q)&pr~wr-hel3OKhnR0$@Dtg@k^)<`OA#LG&i&)6*8i+yhkJx)|W{1jvCf zA*A^RRdF`#e!?!lJDIl`OO&5qwUod%5HgoF%;$02MqJA_b@* zkcryF7)7kr{ri%e4hqp^Q%{~=!EbklvT=fWabEpvaaE$%W)+Tp$8%2g7PS#)+IHil zM!|aKKlcqmcTCI7d9Zn#gvc5z5jf0t|L&1+KT4y4jgp^_Terh?c~#GBWJcU*@+*{8 znuCG|dgP2|hq&u&Pn$!q0eIDGn`E8v7>W9jQbizRmDyV366~E6mA2ibvKxn7+vIYN zoCGXTnJDjGKkzM(a|8@8uNpEJi=nNkClLm!K98SdaHb?sLJ`FGsmx$bAwM<1zU^6|_O0*AYkT=@&8IqC>UUDLfI%S)cRWqXH23RQDtFw8qsY0LN+qYm5 z9gNzxYT~)cJxA89OJz`y4npRdBPFpXpnb;NU^S8GGE+4n)X%vku+BFT!GgS20sWM- zDT^wxQC~4yS->uo0P(+;cWtmh!Kkn=i}X-WR-2$tfxU*4Jt2S!L&wsVwQJlf6wOq) zg-x<3ODE|p1mx4>tgBf6fw2F9F~+0PTY85;~*yUrd0&kN}5vECpPMGW_! zx`u6xqam$om_bffnB7xivMbXP87?qbAe?OEM0pspP(VdbY*K#}lSwK#j2mW~@h#Yq z1y{Xlai#)zc$vf7hKaFsSP_R6(cGsTI17)ia5|cG+*>B=`Vw9CwSjfBMT0~x$rKLz zFk5t#*?-GNkp;E~NKN#$mK`~;wn>L;v^kW<3x#23>byGV-ti(J5g;kp9r1CznTn)% zOI&}|)d7dWV>vAG621x23-5h2mW7JpRc=O+4Y1Z-6iHUrECb_CJ)(t$C#j*;m*SVn zOA)oyOt!JS%zajiDNJ0X>PKA=p?&>}mhkSEuw%9EffmSnLIx#%3KwUnXU-oa5#(Ow z4IHe|#Mgo`=6>6d22syxRod~Tx^13fwrK6{-}^s3dmbc@JWTfs4J zjFQS9H`m9=&&LQ~t%8+GaY?FXyi8gcze)>c*a0c1u&0HOs|w%<`61Nrd8Ye40_kx#=^3Ay2D}?iqiHK@1MKK}@>6=}sfA zGkN_I_Rc`VB3a}uK2&hW0z%%!F{hXuV8wU^ z6WYCQ+VJ@*tVu7FmwF+Lt}Pr+fa^Hrt8w4DQAxM7dlF}6v-&Bw(-m?!6`+C7W`l(3 zJ(cTh?Lck4c)eRuGt+blM@LUx;&j_J`<=ztToLqn+;3rQin|zHVOMoT*Z^EGRrK|^ zW>{7@QN%=CILxcC1U+>p;L+SaaY9taZ=847mdtK}R=?Lr6C0AuV*DN2Wihzl1vLTM zVFH=%;9ETE;;@Mn#)V_nf*=_8J6450?p#}|J3&nGA$HH71)Q&jO`#VC`_=h8Wx3vI zW-PF|J*@b7@qy5CiCOkX3RP7(FOrZMj2B4R~kqw1n z!KG<6vvMVYp&1w3?*O=qMtSbtdC2>YvW@Z84ctRzKU?CcEmB=X4a%W_gmt+-=@*av z#e|0B;g>iWhV|uRuAVOoMjntp&i;$H-x|UI>QEkiC*H#J<4?Kt7{EMo@=t559`wzm zY@P`&FC#s{Z+CXqF6mLCn#H}sKBm@IezS9=eqJWWRD(ChX_mHZgJp`}?#CaG+zrrG z(G6p3+JCN?C|N57m0TgFPdzq0dhA5%e!uJ3s1{ecV_1{0au7i7l%NS_>~b1y6;14B z%Q9co*%EYZQHi}hz<>x?HifiFERzd$Hv1H(&N&PyYWR~OEJ4tTW5hu(oa$HRMC|b#YTLC&k!nFY!+stw zaw80c7rdZ9A0T4HpZxdBMxh6^JA#Hg9dtfD(DdKwHIyz=MpMD^!3GAi5MGuL(+m6G zvIM06B9AE8%fo~t4fm?adv;r-jNWkNaZ zg7-T8+ru=Rcs>#=1^Prg?8e!sMi?FS+ZUQnUzw}cLz!ELF!4*DdwQ4 z>XJH7pO@$v8^&!2k_?s)fuH|a-V7^N)=NHIG>$_4k|0gP&pj7sFTr^U_aDn$E}O$^ zTY=u$7ocKu{0p{S6j|#fKbwpSi~7A#rqB{cb&{U=?2e9++3`|6dFgkOHEqfxwm0MB zE(Z&dJUM`GS?I>kY|s!L-&E^O1l1F*qNwhSh>)f~0GyXvTM-2^W?<5|VtUN$5h4eUyJ4Z}z<9?7@Jd}XD4mC*zh)E+;B zl&HpTuX4}2@P4evw$UouN-&|+eh0(_$0Y-7No$e|;CUs*85J+9YlxLa=-j$K{~8U- z6_TV#hH`bSc@HWU`1ypZ|H<#@l0EL<%_#WN(*JFbz!-ZX=yFBv)oCbN8{ilKC@t*K z067S7&M-HE#?7qJo=ZnL z@JM~}&=0{qdZC(Q_xb4CRb+#h2O1gqxyEGB(>rp?`X4s(dIlof_pPv)eS^xFGet05 zOp1$PNmL>O2wiG(^2_s=saOfiv&9deKxF-wPMC~D5-s;oW&-`Oi`?;Nj;g7Mpe%I8 ze`vbO(8h=m&oEJ)f!hkwp1*WbEMeyJd6f1lwvp3DzZr7Q94nS#J^oT%oFvg;d&qp- zXzFRJ@1W!Ch2SXJB|jYy@ru>iXF#~sHFV0E=zN)ukKmZe!Gh1K1oy#Ip-3*If$g*# zA~-ES#WQa(q5&eM|3axg`QOf4W&?MoNEH(}Ogx#RbQ)=~SjA8R1w7>Vh&9tm^ z8`$p)guuu96SC{iI11)FpR(YJFCle6$zE%^~2;! z5?$r)3JX8Hn-dn$DGArD|M~LTud!&Mku0r&8#ox=pAT7rl6!K8Hr7Gi4(yRe=cCW@ zOW&!4oTzPiBF=+BGA8+D;%?b`afviK!at9khIHe|nmEBV(P;Tio~$o{1mKTjjL?HtsWvQsqm6 z@9@WY)p6edA-(Xsg!Ef-Alx4wH?H%E9u$=1(ufOlraLTN(_{wmPzz|f%bOh&>U z(NgUi`}tDw4IF$C15z@%PFM@ms+GqUd^7C%;ONUW;h(+KAI(M$pjCrct6B{UUS%~i zCi_)|2W8`xWKgEuwJLYZ8srA{*!u)Zcx{=4?I3k-wTYF_OWV73m%ZF5(>p-ZddhSf z@`9C!>35&x-%m`s+$~nz*9Rw><_4rrr)?jOIMapAJTBfU+h2MVC*)_<0;Qz|dfDfS zbUvVLNi&M`+d&c!9XBR?gv6g;^Vmi_q%RRfs z@)Qr9tv<6g$bOk7&T!#^hpnxk4Q>}Ht=gMu5FD@K?rr;k%s=lBRfOb*+e$cL+Q(Yv zWy)8}n`gA+e>zDXAo*rb1mxXxsRju8yv_);BI41iEC|igfWP(}Yk|k#_|B9S1>dzy`Crs_&5xomr2-p^cA*qLQ@LVO?kWhhG-4Wds|%pab;_Xs^M& zY7ySOJ5Tt30CIKMu$#7Ujvsf$Se88X-Dh(RwuxMBWb$AI^0Yx6a&7s}B*dlI(Xb?!?=1f7K)a7?v( zQ;ez~GoM}wulL!N8Q~K$3D_-qwkh#ds>nKC{Yc{hB|eQKYuBG$qP?5PnAW(TnP9;T z`_|d`asmg-_jLbV$(JAI72b*6?xj;dL;R@^O+ohEvC={{lYMyl2LdM-1&PUQktA6F z<2Y{)^1-y1Dia|n0;i?yuiJsxc7@`cnA%6Q0efA5a$@`$0+Y4Z3P%zt%+S1= z0e^lrKg^6j0KaFktMHdR?CM$NjaW*f2jaN=;2xF)W&jHEju~qv!wA8b7I;t|1wf3q z=zQ|ZyM8B-@hZ9#9RN73&V1nASeSsofYF&r7IxG`Kf%xb{ypz~ElO(vm*=YA*zC`# z)=?S6U=!65FK%?Ds_j9^r*Y4(QNBXXgIX@^RXRr{Iio3`@H+{JRX&J!Qeqz>h(`~? zPR{!M-2*O8|LorK)lK`ZfP6IAlM<1Fe8MxOkg-9$S!rme}pRQ)A3wX7nVrYUB^Q3$@wRT#zRPTv|5*MA`5oyB{ z3nAX};pWAJ)y{y)$?)Yb(5&RCN65PL_=;FB z>0QIjk$u;$FS;USzZ_4KTE%JI=+#SfUDwD$JFvf!6{+e2Rss0o3w6m|U2wtE1dQ~au)8|#2867KeV zzG*h;f?1~r%!{=C{IHHkh9rPC3XS5#G%m8Acb>5?$AYji$1Xd=0}_P$Yw+ZV$f;iw z?n{`;3Frb+qcnm_{}jq3-)(IHiA#SR9L>?B(69Go>KlhlWvfJk?7ApJg)*d88|(Yo zYzKdXQT6N+ps;ewr7Zev>DohW#feuGvP zrPVHFhJ|%ai%P8QGi{RT(=o={sA_#KMuWss7}N&#h<)7cMufPH7_i$WLYEbOY4?+P zP>Zi|)238q=m9o%9GQjZ`zD{8jUQskW&iS*1VqtWEFj>A4>}@jz9E zWL?c7TJpOl=-qwhUioyk*(pN(2PPa5CGwzB3s&CeQe%(yo#Sn2c#F8Pt6aMN!L&{qlTgF81bJ6nLvRj$yT9v>uR&;m?p>-n@&M8UZV&TmON z2X1a|f4heSMd>#3HO+#x>WW_=B5xEhW6h!Iwk5?W_*gCfmGa{q-$bVSy+R{N+2}L` z!CADL{V4IlwDM;Hg=NXmaVE`?MY$2oqb^1rCqh8D2`Qv-YT?NA!$zsQq?kpPH^&5y zIqqtv@QLo-^PxG$1b&2q0`=L~WO)s_U5g8+!TCHevmr1+5Im?8l9_+->SNc#7u0usr<=T{PX!L6g4HBJJtUh>I_?7Yr!aa$vZ>fotGrr|4ISdQ^yHsD#{XMP z)L2|JS92eo!_j82G*#X($Z+dl;C=XGG@@DDH0gK|!wsN*`#%y2Vyh&JwojJy)7vXM zXWD)59qZ__CS-@IjRFs(mliCF*AcFt&;b|qe-o1rGqUOvn#JmE({xF6IjLP0E#nuh zIZF1N{f0B@Oju=D7I`sO=4JEjdF32o7dP@Z7-H$Nzwxk=(-h6q8AsF)s#4QrfA!W` z?%v! zn~)gJxI@iYt-&Rbn^2)^=(&EX~$SYZW$Zu52XYy-o+ITqakRumQGx{sf+1lg{ z*zQ5rLIvHA{T&L$0y}gibyb7S+V}&xO+G)LOA5`9K7>U#haObcCQHFWutaB)G?E%7 z$}w1=mUUJ@sYuPPk||FOKQO1%$^>N#R-b;9Fgq8NFxMkuXoruN|4GG znwO%8xqvfp46QnP2WC-JRXB@Op@+2Tvy@foRK9ben)fkPHq?OLf zF+YCDjo6{^Gvn-~;K8V*t__65KyB>lT4@j~o4Hy*@4}rd7fz~-v0ZGtZTlNXdW&BN ztfksIL8Kv#(4T1&9|P0&?II8%=9)BttkQ#q7Q7lDwNfWR(j+65nm%C1tD0Bp zs#+rmUJX;Sl+6TT_<`e{8z-dD;Jn7j9ahy2Vd$xFI9(whjBNYSKRK3~5Jm`uP)ref z6V;?p8yvEIe5jKl1)%1zebE-gsZA*pClq#zvWA&gw8Z{ixhSp!PQZj7$h)2!vTgA> zDtf83oHScoL=07Sq4D+bC!jgf^4DOkgA zXf#6-;fbEPwl>=mbATi;sFwOVYO;nDi7DwS^jILSVgL`Eglg47>S4DysF#chemzES z#$2dYG#dI$Tdb_ovX$qS&eDhgUs5-LOLAS*c%(Qor9QKY$iAa^CKUprlvV{id|C^q zY|eoycO$1F%2pxl!BD z>L@|2tlOy`7;+~5lzgOa>iaoonlkPdFxQ(Vr*MuHfIJ$mJ&UXtdi7^i$D;CF-whG8 zjv#`|Wgj9Hg@?do8*ei2#H*n&hHMVvCp-+xPWwDo;4@tdi+5%yKW9v%)F1A;NUZ^Z zKY&qpdxHd^o=%U>!8NVa3KcW+Q&L2ex`I&~MA#bVX8gv^=Rfn-<)MY~`J;0P zsBp(Kbfo=svoT7P{`CPWYf(H=Jr|bM<@Ufj&STwQABPULqC{qa7T!Oen3q{iLZ8xh z=!c$2IQz~I=dF0`RK)HJS*?dr!UVwd8v)KwC;E~cBV&mi;l9_5K%@7}ClmQCE6i7x zMG2*nMh(N6F}M?Fmf#hUU-bhPd0Dq0+qQ!FK^XOc0?RM7RZpQax=IK9fv=M9?)83p z#amy#HeXs>`TEf%J13Ri0bpMO_|Y*tf98ivb{csA(z@w{k?M@G2(95@9wBVzX`p`O zEGDJ8=y~UTF{}eG5TH~vlyDn(PuTAArwmg>&TmS&xj(Ca++Q{r0d*>Mk>v}pa)+)p z4a+s=84-HT>{zO*V(O`u&1WpTzwVviNWZgIL4qYP4Wp&F$jv2*7BsIU2yj5kf}pcp zIL>X6lMVCL?P`I6!f5WAG?*170uBWgnY&H7bFaC$pR!~J=Mr{7BE#eb z^Rl{ZMQ?DXE#!+3Ief{8++n-T@`FqWQf@Nd@OMO>{tRxba!^GKbWGEG=Pc{AVF?ZB zsA#1c2mV96p$=mjs(TeS=~6EzDLx@LdupCq30ScSy#H`0{z9KSN%sl}0IFDa*ykua zLmdUvO?S~slH5J1_=E{T!j0oyY%UqY=bYHsDF?X_>oy%{s;U~^{QRsXpo81M zE1$|*6&H!l=CkK}{8%6}4Q4w&4%HCKJo&_*XRP5z#%;k|SSGpC8&dQ)PiUh$P`kUS zUkIP!Pz0|oo2G-VpRn@30J~73+ZQh{e(V>2l5m`5?#u2OhZak}V8^-T1_&!N5uc@& zz7~`y)rDXtsDuPG5>Jx0t)2mhf%PElkdt^|EB1faBo09S-BDc3!^hAdFyjLJ28Dyb z3@h}&YmIbvmr1;W{v5buL6T)z3;Ch!Z_^Hh*}e{5a8PAysYfP&()Oc|r=}{|KdzVj z@4OXhn)7_?ATyM4#M*URha{N3A|ZrZ3H#}MRk{x3!VMwg=Qrw2(q zL2tT^Qj6?-zVo3=qA% zLSxqj=G{+I%_~r-_=+kWmuv`TO1+zQOH{^6MnDMkc>W3ie0qU>NsRpHoCZeRnhRD~ ziKM6cUl6CR->!PpEDS1a@=?U%lZ~Onf%#HkY2RE7HmP^- ziKQ&DGyFqZaEri`!PpPH=I{eaplZvLuA}du0uZ|ZE}6#s`6#4ga8_ z_6@}gA<)CclB~u|24V!LWxd`yWXHwb!6##X+15ZfF^92#dD5DhISU7_=8f5fmw4Vh zmCrrLwAkc@k7ktUH!99s`bDxOCx+o2Sr@(Jq&6gB>6nWo&_8r_tUl?KEd@SKiDa&n z^WgJFcjz&)WEvy>JtWmS92zMfnKO>P>1700Xo3``etO} zMSR(dd9qi|-nc5ccWD{}23sWHYl?joUYS$qKd&3&OTx$t#gb}OFr=Q!-m$%Q=b(0d znFnffEu!UsRRYNv=O6PAgKh{}en|3TQlyTQjz9mHBnDB4SKbVOZg2*1CwG~ENW~Uf%@Vtytx>) zYU7s@^%F^Nto!so=>=S6x9lQshA7kRlWZP&ZmP|mPkGq{;K6J#y_W&AOH!^A$QfJ* z#O20=YxFemslBr}0y{KL#vjE#v)1`nv-_*RXS_HIw@JiXAk=B$boEqNsTwPojn*_R zY-CwtEVFU#e3Y6LJ%48%@o+oHO62t~RmkkgiV^z966*GVVn_{6Bcw$Ke)|y&z`lN| z7YcVm)rCeST@&~^v7Z5+(F5J6QwYN!aNv_8_QDhiLGxZ~ME-aR9okukQ{wDBaoivv zgB@7fU)w}{mcCmK9Q~O}NVCYeAk9ROLK{^11U~r$D#6(HJ6oRVP{<-wQs&eb$srO< zM66##J)N6mP*BKP`cAx5hr&To_3fLFe685^EIK!ngrph%Cz8OC+8UJo?e-K`Q9m_| z7r#?RU$@%9inhl!0rKvrP2lt4w|oMjpT6Z7?@3;PMWx$WG8*HT+ece<$m62E%CxAb zH(ZZ|dg13GlyvO+TTVpC;q{myzuYH9zKOK_X|B?7h>25MV}KiX|Ej2irZ~k0WQB0z zp6j4zJa$I`x&^tNY@wHzlK???_wLiyr{bO{0C62a_e+BZ12yqgnfd7W$WgA^eD0jhpJb1r z`@>O6!X66yt9RJH_rZ>D85ugcKH`K{V7;DCBO`$E-Q<1mlIMxe$_4St_8#8qurf<( zMrZE-qU)T3Gz-=?TefZ6wr$(C&Hl=^ZQHip}fY^M`iP9SG31vBhKXdNVLCAOEY5o4>0Qr)SABCZ4<0)&Z11n%ymTB5G+-%lW!% z;Q*=_@#aD0-MjlC*IG-oxuTcVY0_gX-EJZ8nU{OVFnsH?y5^U|E5&1=pD` z?DSM)p%5MnvClGG#i)=AbQJ{Kd>98u`tW8oPqpaJ84jSP##a7$!T|SDUs;%m)9d>7 zW^8@9LT(vtijz&+33ae^ysr^X<8Tc%ZXv76`Y^8W#u8SpJXz;dp$$z>bMsH&eD1os z+ksBQnN>!6#6DWMsZWf$qL)$NtjIrxc|d0w_nokp{%}!lJJI(lr{$RXTHVa3M~lz1 zUVwa5NfW?X(en3bO-D-0miu`sl%H8DKq0V0R>OqUou{CxuA_U=GedJrP1@_lzOs)O z6-tPh=`_`+MZpYI7!J=fjDa&Bqtrhwfc=HK4b&v&jR^$!nkHX^wR?&sl#ccy?RN(; zLHnd}4Pv}A$b4d`a6bZPJbc>Sajgy;DyRKdynEu3YxnwS)8j4*j zb`W+aXeJ3Rq|a~G;onV=2i1~IJV4BNndr|9;Vo)OTh_2LJ8!8r5nzaSZb1?aiBA&W!!sdmT> zbR+9vn$&r~kaBGMKxYy)1g8PTX%;BZn|EN&ndAhb-$h`8f|thcl_U(2Aq)7#MzHnGHONf36Z&x<9S?yuq)1kO6qcc zI;9Oc;qduJu;GTHcVYtiI;K3`;w9jHQM~aO7jjTnM=-)9hlzC&lHTXwM< z!k0IB)FrwDSo>*oR~eJn+qaN^+uLV3fr1mFFRJIjgqHsSI>q%PmXoVxl<6G<@p`3_c z?xfaiQFs1?-yu4GmSfpKx6#lb3=`w^tnc;CD^PsBLH0LDP?IUG*c%vDdU9jpdoB^a5C1^Egb-*8UfLWaXz1* zkXnOkydK5h(G$Lbca1AZJ1h+XR&+W3U79o87 zq`aS=>7-D+j=K}2QSuo8Sd(#(N(bHh4hI zR{;@)1;UQ?MTa2C#v;JufNuB4SU3KM3w5Q}ZhCbn2Y=)6(E3ig>5V73nf)K?zB-iOAGM*Mqa*@z7t0nWv83N{hjMZHA7A)j4 zZ`J++M88P12D{<9>Zs3&q|c~5(vd5K>0x>6Hn&o3?XlOgxvT2vF?}w%oQn%PylM!v#13-2&D%&75ncVlu~w z%*}MuhE<+X6h}e0+5$@H8LpXd{fff=o`HhS5S`q1vy#a@CG1-f}=)`thJRFtOLIH{iXQ0U=w6zI5qSPsswREpC=Yxwr-Jh znpJh)u^fK`{`UO&s^*DZW;~8E~`R`y)dCGVRR)!86L=QuNQ~=C0*2EP!T1@N? zMbxkjMs8O_`V2twjqDvGFjqYAQ`0ba$Ap?X)P4{Dj-hYp?jq3E*3LBi1#KE*JOTQ+ zRq$vrAM2}G`->53b$j!J`WI%eb=&W(>&LuZn_hj#-Va&6iuT_SwHEb)XPykg!Fq0c zKOBrsLzmazLOb}MTUY(~16g{b*#hL%OMBw5^$EM-%^d(k$F_rJVE3}MtNnma4~VMV zLG#|-`0i^0cN`vV$0ts=vdyb6H21>LXKscuU%Q7Mek=E11bQQuv(?dqn?+fk@3jE!jaXd})Uf~HoM2X7gtQMM{UYz+dy{2s;B}4^myuy=&IP1Hteg-7zh7&ba;%OJnCYJD)s0 z_9mToobD@_8(CRdQT%XAiIuZBiyaZryCa(gLvT%Drof(-%ecvxVLe1$%=P^jGE^4( zjmGElFg!Baqw~NpKMGRH#Z4r#R28*-+l2RmhUNfB@c3-WmY}wdVgRPSX)`PinAr*0 zlR69 zuJZwmIxE`m(Dx_vGdIvV1O+m@+n)8b3y~Yo_4(>_@NE0b+Ao-**14$7(F{3TFKz z(tm9*j&Yw49PsEB=+s2ZC$CS*I+<_(5YgG*uH%@|hV{cc3M}}6OI|^EZd|bain>^+ zYato-I}uk^RZv|yU7a_0J>LrcrJ!MoYD^F}9v$=5;2|Z`w(~{KGy3f;!rF<5x2*_p zn_PkL2-N|%pF1n^RrYtjJNUR59DKCg@TnWtS+BVA3q3>35ojX04h<44sa547-4_*oj+ z1*3-X7}-}_Ci0sPfKDn1Ud-4FOHTxZz28<7heaV_nDNbo+UesjF(FnfN}uJG5S)`C zNyB4K`;i9{Szr(4E53IZ%!gLwE zPR912jYd_uB&q2Vk9z#v1y4O;up4x;%5SE2h{d(S)stk7>4H~nqu)k8=jh5DYKrx z4ZIw5Vu&*YG6Qt_#QL20;Hbu+T?$6P`A4b|3AGvrSt$I7$53XgUHD)m*=R-@DXor(BDlb#UHE@I3EFNLB}Y zGQNuydA8@>>Y)SAf^xmt5ZYu2+9LnUE`F4^(38A+QabJMgW0}wR3c9k3n-SXc&n)k<|IJ0NV z?u+#ucNy3zRvgSxt`nXf z@@!ab#M7!blR9k{4Z$vs7b1BE{_u-i!gY-UXPpPffDel=ODO{oy%o|HuT9d-+{#p5 zxOB;evI_( z%1sC`qr2EmL_mo<$gOj(yL%~)F_T)%7FozyQy>zI88u+@-Z1O4vdtEkZSi14J&NMRbwEur3`pE_X9 z9<9eJ!5V@`^zQ7N4H<}P?}-@;jQL{D1y^n)O=d$>nXv#U>G4bj>~@YK_e8R^7)x+< zy{Nkqp8=4pc`mjzi%|`p!=XmyVlV zP=C9lruYC5>|-pWb8K!KJw~?1iG!pXXN^G4wYRLEsVux#qI?Y$yK^Pf`fbGO5p4ts zB>&37VU1oe;QKs>>huHBKv}MH(qpPzQhngMMf93U z50iyNz9;FDJu!CTIezMmIrLp~!tzv(HEwdpVNykVkz}Fe6{P&xVZAiR+7xySRp9R_ zk;Pi(*|ip9L&*&=Zblm7UEzA9rkCfb_@Fwma9h0Vo1xgePTu4hC9RuR=u5 z_F$+U=qlJ|n0q6_;N#q{Zq$?YluJ^%Aw%8l3$L9wtGqu1cwVd zS8DWkjzK>WQjhRv-9WgTGnV9c$58Xu`E_ zgHmHA@DTR!yufg@B45?`chMS`OcAe5E8gKku~G^wVTyOwt@R0zAAU66+h=2dE*lS1 zECo>{lm*{tZ8srtR83^UDw{(yL8eX~AG1r%;0ZPQ=CEuGi;y9l*_4Zymw{pAz(~Lv zo-MbMBl1pJ|KSk!cO6kPs7JH!1LDM|HdG199XU^JJHV#x_h>_;2RLO&Mg$lKhkawe zz^2802dT%4YS5!>o)(=_BI4~tI}OH6FkF7XcJeTuP_f?C*M=E2>tPYhB{8 z_qY8$t}dwz$LSqKdvN4A|A#JV|0V#%%mQjd9)(lxXzl1%sBPtBIeoup=A;DR3&qfW zK)o?bNiN)l@O)yx94_0{Ly)bZM!jA8MG}mkC&sCtK+>^Jr*@*ZfY8^NKzY-0IGOUH zqVa9w4*uP)A6pLbVLwbU!botaO~YDHFst*gOZ4rFW-2ONtMug3Yt*!eCpch@)%1dt zG<>R*Aj~GR`IdI_?jl4iqP4?+CnyECTF`z;tO7^Kn3>v>+ z8Er>nrawY%B!zH%{SA&*oqeWbsEW9cdc}}BAr8sR1dh!P_UgjGcp1$V44^wcG!1Fi z=C6-DQ2?f-ekOlgCH@0h(+Vi|S60u)k76{$5=6#1V%pC<92D~8p?p=6f}5PoY>o8G zu!}Rbwuy9r(mIbz7TAZ$Yp~SeiZ}KTA%{oPK zg|#U~ctl1SY;kL@R(lNMD|6xDSEg3*j_VZCI=`vjcyliPu-y=tOPc=mxqo?2 zA;CAMqjBz4a%ok#QyO7JX=^B9W{9K012MnAMqj!=vy6Ht%LT-TH4~H99wufUeB;*2 zL2&bh9@yGUg|qg1hY?+|!YlbqI>Voh%MwJ38svfWIbJTY5S zbdVlEYdKekbph7y;Tt2;G`3;+lZ^(PPOJq?#bE0Q{ZQ_ii6C&fP*i3^l}WPh=Qg*d z(=?ti#D|zIP&gpu#L))1>bVdqqks=L-j1-!d6kc(4(6kHv7GiJN~|LKn7Nc$!!~W( z)1whK;O#z$%&)|LV+#TK?d56R1krDn%wz6o9yE}WrvM(2+^Ce4le}rpn;tqt`HvuU6)-Bc>OS z({{GtAmC}DBRS#pk^(t@2{Zq0Deo^15_?@M4>2O;ACw>!k53cGP*W3MR@{3()uf~q zZ)CNIh5(eblnVt{5m=3Y{-6%wlkOH+)Pzv8@UT>9!OS8(DM#RFd7bnu1(2oDQ#{Gl zlZHfZVI5h(YtyOdl)|_cV%tVqSS1TfpETkE2hAj#Kno!}>tqXGlQ(Nm?WrESt$_Wr zpr)d5EAXVsRD&JlX6X@>&+`e0oly^Eql68*AOM2K2gFTWE`iwL;2V?;3nMDM&WpU9 zm7W265VFh+;y6dD$cgPr+mMk==EmxG5KtiEd#N6tH9ZhxdS@g=V z?iXHRXJlI2^9$)niAtl@d{o(ddkRb@vEluY$V|4xy6lYCD?CjL4UZh=b-(V(y{Iiq zs{vIv0ZfgHbJDK1rkNYX>d9EgHEK?*{H>N3M?bOI&&SUd*rFgZQ+YtZ@i>A9 z=8KRLM5{w9t7m!U4$9t*5!naD4B*E#@Cr9LYjmKs{NsEmdeSC9t|A))(L4??+Q4<< z``nAGLmgMDoW{0o$*I#i&YadS{X^P~Q5Hug(|*kx3G5;4H1S8gyqJ8zu7t9zNTkjIIx-GH;8(wSPl(E*ZCATXlGsU#;L!OM(dcxNiGCO9!x zT69Ff*XkR=V2JR5e)$chWazL7r21k+7;fw^hoaUWOQ#)2*U3j!Va%WUyKT%^WxI|V zXM%;FzJFQYAim60_lRo;FxdS>WHH07@II{)>p}sX$~^pr)q$|{xd^e>LJdXU>En&q z5ukZ!+bpQQe5+|gppjo*ij0|jI{}=8Bny_j79S|`s&qYz7T8!JTZ}bYnjdWg4VJL( zyWNX(go&~iTZ{0wp-~}k2fHsf*2}o@!tFcly(I@+H<)@@q+uY@jYaB%k<|#U#RADM zJqv1`nbV9}kT9n-6SkvZ9=iq@Kg0##UVucq_c~^PP(RU)03FaI79T>y8Gn9ooK|H3 zX+xpesDA-VtQBi#BzbP!Fl;c1-#fU?WsKX~_vWZw)2_T;XNuiKH}l&uu`Au8qVCJH ztdYx#l*rkrLh-)TPt18jH&>^O2VK7E9Y>j=l-|RitLhq{zqEu?o|WsIy9#B7t^gK$ z75_nmui@cE@CDx%(fgIOf+uQpWg`^bVBq}XZCc^7*h;BLB&0W^PZ=b6!v*x!pq(1M z!7c7Q^PYs1jbUYAJ`gNQi|f@W7>Nf-vBqS?s+7-`{i@y94lS$-6^S^LQ}gB-yNp^V zyk}sw$Rjc)J{W%L?@zks5?L4iG{C$1(i1HBbtL02`y6VXDAU}UL0qrBXGH|`XB_iL znTMkJ)V|PYPT_9YR8XgE+QClbh+*-Gald>J{ku^?0qs2+^=NOUQQUy;fzU6_m5MmH zsr`}S%4Z^0q7*!~cZzBuBQT6q-I$|WwW-%HJmED_cVmS;#%ONCFA6PtF+dy@{%J*r zqB^A(-Qk#jOpy()Yaq>wuqJe)jNR+69yom3VMU18%A`CiWP&s|a%bQ4$u|0E`W}(m zku7;xZ$CfAY%CRG*asGA{dE(HtHP=r7*Cdd{bJo^E0FXqE8#*36Hq;;KPe`y#R^tw zy-|Zqlu>kyBUGdnCgbTxQvf6@w~)DKoH;oH1W73R@K-L~059&J3TCwfcN1k{Y0Bf%r$XpLllQ#JKpR?r z4CsrI$ylSLljR{ErNr4V*&O?Fk4bnr3A6~>{YZXzI&5bJHB61AX24vCZ5VPY? z9gk=`%=jAx?Xwv9Tn;MtuQTk09^u64*}(}}zLWR|4G~ek9*ABvL8zoOS@QHro2$?I zA-e$Flttc}JMD$kIUi4O7oTkDQs^d(i`gMI1&3o-nE?CHIiiEhO02wk#x#Nqjf$K~M&JMzO!DeJKr@(kR=^F(jc0c3 z4r7fWOpc6*2{jXRi<&=V9w19k%U(Yn%=m+dGRqLj z#-=dMmNbK=4c;t*l&icMvNlG_mla&8S^!8i_BGzt0cg=U0~vj~ zi%(q!;M6$PoEhmXa#O>xnMk+GjI#`25_MysO>^$ZdQSyr4mo!qKeo=%=uXp}83>D4*aG&0#7D}RDL+{>^ArpAo`}6hZ&euMtw)CHFLvba?Sv3n63kwVJ#S7Kx zf(Vk~yNHyM#eiVK{C*#91q1#c<5!y}mrr!eU_hk?X%Rz@Rom~qIp4RsO+LV8eAUVx z{61O!a78d+1NO2j?1|@n_b1W-uZK&_qBI?*7|N$b!l5Rnk67-+&H*Y#){1Z(RP3 z>%VbpU)Xei5B!hE?!zP2&1@M2sI&_7|Ewf1bal6OGdFZ}cXM)gV_|S|uvAxr1_J$W zzrB$#7B!sPz~kR|wsRYJ*)IQQqv+O8n-?@7phB7d>hscZR=COkf9S1iy=@gd$+W&T z)gzG^nT9V@;3ht;MKBOL)L<~;J1ab?1v5=^&+^JT8chvgegw%^;%AGSXAc8Yf>A}5 zUG7Y6!R?h6v+TcPBVq6tbyIx*ehB0zQ@-s7YQd_ZWwz=&-wQ<$-kIhD;2yNA?%eEY znyZ;QBiDF79jaWgr}Pe15Cu0RDX5!-xS?MA7@m{8O2);@z~I?wFiFW>{Wx#&|D8Sj zeK?gH=Nso*W1e>1_%KvH!3l@gco1ZvChvIBkMNmgVUQ=ZN*1wZhW<$5_7J4nM(`nr zDtlSGkc;}LA8iE36`_~`62wN?I1a*)CZf2?fIC?_&s+Ds1r{Vmr0c?@UC3wt1+O6e zpDx124`vsQ6#46eB&d^C~8*kNP6~h=TLm{Plk}FA8>-1N7Qpu}`y|EU7h6G%Ep5xjV(73wO~$SR6UPiR zZ*o;ag4$E*J1?d`R&i(`riL^%g<3OVHZ`V#UXyp6QQ!q3UzCEss{KFnSK!oQK~qo0 z5VsAmq5>^xen(aU#BIXq13kfrDF-Krl~Wh-E{!RgGxIyaUP$X*FK*85WmyB-iICjr zP$d*prPkS?)zxNgbkk0+OR0~jatdjX4_RAu$Kt!pX)-BE%Pi;~OLEU6u@ z^k}#cP_i89^Qh&LtZPTJZk#1ssx;>{<_U@OAsrUd6+Q+6aMmN`QI!>Rx2RxO?HPo& zmf<(8!@n3N>uO?+qnIGGQvNQb6PL>;=+S1uG{&dQV5t|3%V^i_L>+OS|AS9uY1UH0 znpP{$Utdf&|D1V@dOBmecI`BIfIbtnc6#f{h0}oQ{FY?0t@@T38zh^|JJnwB;tQgp zN65Q))zYs9oV{rrR`#?idR{1gTw6styo};~pBfV8Mt%(+n%>_)^&1dGwTG9F`+kH& z9vsDT39e)A<`7)v57677kQnfbs*E|#7@3$dEt^$`?4yi` zN(hDB1>aIev9j{BI`+JcK%Y)n&eDTqX=>igy>D;$eblT2kEv~@EwlpUKQ}Xvvc|Tf zi{CL5_vq)u5>2E?EPdn-MbYCx7A`Ge#M~*qzPFLw2oh2w>I6hTo74ccml6}Uzfi0l zOOd7kh_dh;k%G8;tGOv0*U5Q!@kWY5CWqnz(sck7+Jf69!7XsjSP_AMb)-HH~7&(+YE)0 z#F*@)SHKrSal$M^3Jw$_C9s;;_QYQUb|DLW^iF?%bfniKUw z!Nx6>;V@^b=aQ)4H8ondOPOx8&L#bsimCGk2<|}SR4ROJPG|X9R`K@F=DpH5uL6ny zMJ>V1q(tCyeon%FI#vuf#Rd4Xj=XI=KK!Nn&r3g>(#;OPJq**6@0iuv`C(Q#G@@2l zX%=)MT5byHp;bPaB$0B6K{KF5ixkD|aDqov4vNShtgqtP!RAP>q_+_#HOVTc-j5F* z%*xkmJ+eepFM_bZFa*kXbqvf#`fqwmv{PrHY|`_OJCnT86|SvR@) z0lS+UuZw{~K*g}8<^6`f_cJAS$aGBvqV06OgndN*iNG17+{ovKnbPWbfBl|p#_;?i z8XJSgtL5w8aQMiR=Y3ZRMbM!VeQKCz?vVO%-dTZzx$t!;>JCN1s|Y({a*q4*J3D%1 z+lqAI+oe~AP7o2=^Z5APvKdW(wXu@l05HrW2_wX+5!O=ueN&s(mP*=~v$N|lb8k;h z>i&Q}V}H_jAUElSL&dp;FqR~0Qx$Y{MjvLr*RbHXjj75co{yr}3&?^p{n4; z{=RZ9!u}w9(4+RviW7q4kQPC*t3mHm;iAZqNzRnOhy^rqP&ORpF#}U`5YWu`0}>GA zO3Ha&3~Q<-WpF|K#V9!1Icd7vxVxj^)o~KiM-tE%7qKZqQL7u+hQJY^Q@B-kF?kNP zh6B^)^{|q~DYGb2BuF3kiAL$fZ00sA&Q9~cQVFbQY|fBbzl=-$3N?52gx&)3Tk4+d_dY}vR({0@_)`Skta2|$(A9gX4E=Dl1{ zAI62GX;^K4!_HLewuqgNS908R!+eTXzEXY{(pygqYpWP@Hjweq85vrJfRTZT%!ShC zyJ`u99%-dFmK|N7MBQwr@V$i80#0i+7RZ{Dw;C|xH*RdELDkzSpSQ~ph8%ydD;4~z zJ$}H05Y3D$U8OH0mu+(&0bqo4xDC0h9u9bi3_}RsB;fB1&yX@4VpIBHaf|pQZUy{w z+hZwN8@nb$AJni&yu&1x9)ZE(SNWW;6)6`EZ1#2Z+PWDq#C}<*S8r6 zSA-W~$d5CKJk1oHoByN}Us(LmsNu(mo7G-u6}^i%ghr~?FdU8101(EBUU#P!Ml1pj zAb1>6H>kN0m_zXw&B5!ZxPxSH)0FM0X}-T0CCftQ!{WG9r$Bm?+VRGB#O(+}GX-L9 zi4@-(m_RiifbYHbdJYHbug@Lrq#UDh&|FoSWXaEtcW%@yQ0DL%QmIkHHjd9(aGTq> zWe?P%Ob`^|37gkU0U{zzEaiG5>F6^!JRCWV!}Ns}PSVWLg2rbF_AF!g6m*wjr&7Qyy{90cL>!=Mevv_+1H7)S&BN(liN08xUIwa|`r){5tEP>X`u${+mg7dWU7q%H!Vu6p4H56ix! zYuQTTe$MwD3NFjo7Jc+mjb7zdb~g#NU2xs*-0P110NfXRjQ()EYaJt2s)Y8m#H3MQ z)ytdyE%xb+jZFnJ?RAY- zHj>sL#lj!Jl`%pY=8vEB1wxP7t~d5hXm)hi$bmZli=OwRDc-ln(cqwO)=i1fS)*5W zUI$@<1DJpHy9ZC(n~27%qZdrZ$947YlxZQp)U-(wQ=>eKl*Dwb2vOIF`Dw|rf`bTg z!L5c-;@BZWNn@n{;@D`0sZ;Ovsz0Q`GvLyY23Log80B2T-0TLX?WJWf6a_|GpZd-$ zi72Ige&9P~s_Yjmlsa|yHga7L+d~K~?SUAt2F#P+c~sd&hOTXI`T4szdC@8C^m6;b zUW98Wc}Sn@c)yiUKI8YFc>nH!8OBM!574Zo^1j9CpQ5OZ?v8ukq{vk|j-n}Di)HKl z2hg0yZ-S3YchT;O&lv$o%v9UWIb_ZGS}z2kje<|9r8h0;i?vovn15^h;(_9;_gg53 z27oMej_{Tq6efj%#2s5$?@;_f6E3SvTIZ|FRSEb1n&Oh(w_?$B)-|Ctis>_M6r=Kc zeSh-0{d96|UQV1&w-1~>W{Ey5eZr+$^48un#Wn|S5N96G^2MB1KR$L}*bs4f(QZe) zmnMqR0wDf_l>8=O%GSeBg>k6NYG`Ce2DDwQP$4E=So-h_Q*Qc%($;IY^T6PjBxa4o z!VasC+I(+Kb2j{e(F}Bx4&#_CTnQqZhEk38$t+{c!DUE1NmfifQL!=X6#7e-u3b08 z{nDOa!rHG)VO6qZOw!yyyqjS2V0X1`zU?3+Kvi2ZUlWwlenU=1#m-?+g(HFZzN3;xjul7JK>fhNjk zc%fR}$m(=I6pa-?&j(rgD}IMGWr?|Z10BqUMhtat`B(OUm+t zb4fCfMe5SL{S@;f_!Dl5C$q67oQ{9UbMwW$9`p{BVGH6s4|hE}%yL}0zI7cy(XqQy zHdly<>$Kpebi4Mv9+pEaqy~F)`L(graedwy6}`I8Xf76uMDMtR23Sg7sJzU!^ZsdN z`&bts=}+^Nv|ubKzFLbKX*VAzoMcbE7A<)-d$}YPXpsanSP)8jBkiMbdwKHa%F!90X)r6FeYo}5dHWNcI5LHE+@$eZ>M?nI*?P&r85P^k8&a+j zK%Bik`auV;tF%Gt3y`*8%uAh`SRKWP((|=;C+e$M*}+?kF{^Ur;b**dxE_kpe47R} z50Wbvx7{yLSk}=^` zvn5UjxRScNef)*gG}C^pOl>+|86ZgU`dWIsl(&<=jb&6E3{Y|X1xmW8V`tvQa-w1W z8Tp2HF+iR)|a zQ2M-niWY@)8DM8K9n9iEz@)b2Fvl#&$x*e5X3#@vu%V%FQB(+LTH$g-n#QVtchl$N z+We5W8HEeMTkSf%;(4By!q@+Bv)fL-7JrAqbNSiEW0LclRjnU=uC)@EIzz(4@k+0- z74eJ7zU|QW<`|#SLAS5vH)phis%K)!)>$BZI7(oHIsjl<2*1di4~aASX(_2cpL!w} z_t4#CdtGY2!@pl&3En3wY0Eq;ip7l}9S6E}`!HEb6x8X=K#dJ+*l4(w|N-8~qfc zkE1on8_+Got}%X1>k!%IlA~hPCBs(Qrfg#-c{1lm)qNjkow$E-mfc4Q`{P?fE^ly{AF^33%xiYT+cQ*VqoK_8b+R`S-5gGWqmj zK?E*HOQq`@ul_N(PelM2F0>vz)`o0t@-P&we~6Z7xs^9|0mL{vJ46@sL63*@9p7qm z+TGyA^>x>DZz;$(N56X}Dsb>;Khr9#w~1A5U%mwAQ%2v;{o73O*JE}ZcRl)P`XFL! z02`~+hhs=I0uYPrf0|voKL+Ib)JFBYA;RTTtKyAkqm3D*%IwLF0NV|SnZD+rS&NDwSeJ z!`#EN{Ef6aG%qCRe~P&spPc#5HgVQa0Cj@tD9ZIT`ZEWZp}tDK1EsswF?X)B+Lr&j z8`h2A{gW%V-}`tj<^Jiz{n;d8`|{kg#_Q;zkiZ+*qYOZ4kI{Er60$yi2QT6)dC75? zp!fZ~=gkt*=6OkOFnLuoJ3HGx{)K<_ApZwRH{agg^ZxQt$IAx|lD|>%rO+CU?1vev z%NgT2kNt+AkKPfQknozYkDK??e33Vm=Y|)mZFYyfI$h$Nn?%N8*CH@&p}DVAdGrV} zs75Ln->b(0JhH2K8@2J|#uUCHOp)%E*rd;6>nXkGyt5bR+TeBYdH*>pz1n3g;L5HO z!unfn3EinLKfURkn+&HI@YCiCIrP1cq|C881)MsEtlq{%mU|z{n|^oB&HkV05@)bY z2^ky+r~wWLi1vRIJzl%L3ICg(wiEPtlY>LU0R7Js^KT*GsT4OoPMjYS!^Pdf(9PV{ z&6V;0Mx^Ju-pR^LXz56e+z|iSJ)a*WJrD;7DAnNqItduGy&3+yd-^{z6&}Ex-rB1X zop)V@PH{4c@1K3ZW&OcRB`MktMZSp?NoXF3+<`{@Ey6#u{VVM&#cwiC)({8<>5vTP zR>4G?Y(9PI^p`^Z4dFLY+ZavU>}p`gXE2iB0HPo^+O*u@Qf=oM81Ix*t<)P%`J zP~h4V>k<><*m-FU41Q;3RW5e-tSYhNCM6PjSg%NXs0!(<-{L%h8tnjaND2JWDL11% zsk6=;8s1GbiiyRV7~>GA?guqx9)?pIYOo=k}@fx$d2yGwGXkIIuHFQQK zW;olw_u56U@DU%1TK3?<4l65VjwQ@36fY=9k%qQh7HGp&(B7|XdrJdoqMP0mQg%V^ zj<>-mojB#LTD~DEwf_T%;m`z!;j`{?)hoyb({*o-qH)X z=Tsn44nR2!QAGecg!(p%QNF&7cjf|LUtK&`$l%>-Ie<4CNF)LiU~8RW@xFM_sC`ET z%IYIWI@cl-d}`Fjz{)9~ScR=0Y1Q2|BLeBd zb?H7EgB?I5$Dv!{JX&y0;)s;(z@9(nCnB^)uK<_GtABZ_hAKzEIgQj1) zTcP6G`60u^p_n4Ae{@MsywoXyf|;G#U-XJE?JPnJZ*5;b;2nKOzO_l&R;By>U)PmA z$K~A>qt!BDsy0 z0(_ivvc~VsQ8-90l-xkOH#RhGiUr{W^^9GNUeW<+4<-jty-zeg!h^;MAYGd!T?A|f zX;CbpOsq$U!^X*sSx45DD|sVeVJcT5wRLigP02zF!}0W)Nnu-@{j*MIF%_RhS?+J3 zA`Qz~-dr~fEq7kj)iqHXlxQscR?7)2ucQD%6FR`d3>lmIj-q_)V5N zm^R2#{4jiOy#*sAb>9@_bn+w>yaqTPl(REBz!y*OPl7g7Fn541s9+tri0SL!s2cp3 z2Q#YcqC*ekeOXq1AmaqHse11U-i&RDGkSHBvXcOZ>SW;*PfddD9%?fq*hrp=IzMF*Ou_G1VFN(C%BL{f9LQ0 z#1s4?y-HSwi|s=$s1T|^94Wri$UsY=ZzKF3PcM%r?AF}Yl8?HHjf+|j^h(-3c-*YZ zlx6}znhuPBOXJEu$ZTgsEk(TBR1_6f7Lz;(TkqV}0 zg<=g#M5n$SGzp8BzP@#8!R!bMH!D3aG2R(W&CNeb!I4NyKd_hMZzxm1US~S!#9Q2z zH3}{&Q7y6?jRFW@z?CQqVhQ|p@=q)>X~q+-L~-SZhRObR7pwx`6kQDiM}o?t#unvNtO^!ym_&3A_R|z0%Q=P0 zL1=^fEEiBbiR3yjv_Xj~8lkejV?CxtOLX&94>=k?4;JH9j)v$}=`i`#3%fAq&FZhz z6-;v79gE2-^o~%Yj@QwNe*-TGlJhL2iSKVT6WXrZ$tiFynGs21PwJhiC_PmJ(O4AH z7d9N``nS4{%n^=@J*sxF<%5}auK{;>7F#(3QXF7tUP2VJXuTtxYvw6j6|6M&(`Kdp zoBMXEmRdYSP}2y7I3RXCABdqt!{gs=nJCqPf{r~%*G{tSeQ)x`5`;m4;Z1z<@eJ6> z0h|pYsVKhQb@E(9SVfYAK~$n?*bzrWHbS@?2ynDLGo-#w@4)Q|WsO;H#I`~^K`;A) z@f<+7Pefi_kdMoHQT(ZN>$+yqrTyes9){khjZqtm+4hrMs^E9k4eJ@MA4X0gcg9O zT1`Fa1-7oDcp{sz2um=DQ6q>SlpwbU&!vmL)pBwpqlmgO_>$BkSyvagd?6wQ4OAr_ zrX48|vj0WbIdx|iE!sA=?Nn^rPAax-+o>d9Y}=^Vwr$(0*miPr_HOO8b{_7_d|5wW z&9(X%qeG`1Q)wXrf%`mxTM&dTJwI_`1k}Fa1d{#8omhuI1m!Pux`!uI`RUenhDnquO;X z;gk*X-fMq#-a7e53_w|~fK4!437q&$f1Bwt-`rYplvlREWmd`9*IpVtLR5X_cadjT z+n8i)$UjJW#q!b_VeG2Rv0XI=f|exJhV;js=_t<#_m<_`rT4beZ7k;vX*QdAz*z=| z+H!W!LS4FoyKfoUBt7Akx)OlTFhFl_T3KAnsX5lzbONm)ejQHNhtBWwcyu^7bh@3B z8ozf>K4P|7?1-0YB(xVSe}No}0_wm#nQ*kd=A^9Usr|E)$J&ITw%GPEie&n?2hx?R z^oal|-3ILddnJ0N#azBi}0%f72GNxe}UJqRHv9tW@%jnYq~+|;g1 z?!!kiyg?Qu&Ak5$(j2$MGbWDtgkMH~!^CFq=wANNY^Dj&9COwod=wm^Ho4=NLANCg zonJRCUN%{P^Wz5SDfXI^)=^RI2S%%PI(dm?Is+ZI*-(HGUNb~2WgO%#;z5NrT!^#P zg0J&|igmtu9k{VrGXQMTBUm56Yw(G-)#0(&o}z9n!~D$clqx;lw|8#rAJ&MQp{9EK zH)7|3qU>P|{Jjey4AZ-`6!RGJ6x0`7A*Witr=5!2EjLE-WU{xiym#k$#Hq10(Bn%h zg0TMer)zsjt)#54No)I+AWFV^J@nTid*@b{dT^VNW&}K$w=AH`CVMoxJ`*l0DWXbV zURw^!;a%uASsZ)hB3_hp{Ewv%0kk!%Foovf72~e#cg-vYltuMw$Sw+R%LZR{2}1e{ z$!&2H6DF@_?M2QM{EDqU`y)(OO^dQBK0R^@>fT9 z(1>4vkUnTj6(zua*_yVdfq#Hv#eqbz^iEK$+91<8!f5mX|83nxX5p4{4@^+rTcPzC z-w`)(n(m6_t>3=ooNQW-R#j-I$>`(0oGs2(TuI{;fdn{c7UhDo_aj2EyB*Wfdk@75 z&b25?Qd94Pb&!|_F>tSc4f4mQz0%xDWP2Z$PqU@Z_YGiLg4JGb%7ANOqs>p^TXZpm z!8yRxhPrMk#5$N_zL z$3Heo3CseeQIyS?)zNS@4o-2?SfdepYa0)3S|HC&*LuCnv!ml%-V%?I@USSctFsf$ znkU>>Bmpdrb81#68cA%`g!^>%4K^>fb2Rxz3sGn*t$XO7lggYWhqfe823(E@1aIp& zi>)IGYq4pBzf+q+EIAg9_OSph9g%t`RAQGVaVBP3t!%({L~Rxs zzRCu@@SIr862MofdykkHRNNyF@&$c!SJ>;0VFNn#JlzW(T-4GACPKR;2-UZlwjnj- zs}KmU)w_mZqWD-6HHX0UtNEdF|3X)iM|$~&+Cr*v${jf60D&b|(7_(>LVg}BVIXT6 zKcNAo3u_a++|7+_#da(SG1N5w%E$y^vG)mrPFZsAN>7-Z_KDnPSWRv{UmV?%1lGqh zt^m+{T$Q(uF>N1a3-eAxTIci0Zx|lCase7H!Z-%=ZaAWaczJ};54P#Y8F4>r7b50b ze(LC3Q#!IlFVRdRCelpQY?jqA=pS1>U0VM0m%Oo(=LTZw69&2n6|RsyB8S2XxV|Hd zyLRTSX^8H&@y_p>-3~8{If4I_iB-N2+yIN(R`Q+j2eOMg+lv+4<;cM$S$qXjc8r>7 z>tPQFlaol11U5LOC~(eyL6rj`#@LGg?ZWf_TNjXPt+ahnUZzCju1eL0(+q>4jT;F3 ze!VVmXq?Rlw&ih-@=-058wzB>`cA=8&{^e%lRD0WwZ{&ls8hO@`e|@1;+^zKSp&Mu zqp4hEi}jW+2PV;JLH)@^uIxOSCYC@d9`|QD$6C^O2q21@v=)?C2Iow~*myAfe;IY- z8Hk>HRS$izl@mayzRJUX5!!F7*$7~+EFNf2z(U{jVY&aNk;w7^@YXZIQ4(+u{WH0~ zpU;_KO24AvKZ1E&J+L9};riIKngA$&3I+{*dCWADt+{nZCdJi3qX%8Nc#f;mehuz> zLG8(%aagO#O`_@KSSWM<`3K>T%a*{D!p@P65&LUC1gq?yGH#FH`QkJY!3xE$H$1`2 z?q`Z(<&g_0kcIMQv12G6j|yY?h=w*u2&z_P(oX|-1tRfye7O+*ujZ#YTIEGc&xH0OxC({m?B95cDI=Gg;>UtrXj0L>ut4Ik!8}CJ*I^rE=P8) zZfpeeOgoVNM`~(qA8)ZvM^Xd`pl8WUb#&*h=0xXY<>toG@GZde9_WA{Ffp#mSE1Oe zCXp9`z$Uf3u-)VJ1+I#Y8w()h;88z^E<%SsdUh{K_)R{e6PquPacmEO@`Q0}xlzfP zLi?+(QtRvVE7PT#+5uokk;l~FtlnQ56a7Aht`ve`MkwgP1kqxzFe&JKfH5C8C;dgy zq1vnw+>~K&KIJ#WO;gByc)k2?M-x|irX*pZ26FgYnrfPtvDh|?s19JG_t&4yez743 zvby?CfYL0kt^yc$_|(NBcCD2JX4z;VX?eDl1$}z>MkuPW%IoVpbl{dM>lwE*fqb4N zicj#X3+w_@Pf~%8HB%W{idPt<=%a$-&g zbID@%)``rcexQlDk&hxm4~(wgZfijm`HA(Kqx@ZmL~(?yOj)ki-SR;y$yK!Af*py#`1_RLhW^UGcb$kLdrx z9FA|Ew7sI)#F1*gqI|VUj9a3n)}}R$eDtwU}xPWPDt^pUf7rf*ZTA4olk?`%y|g9jUin(C-FN)s=8;EjLD7zC%V18X~k$Q zGk;S^puAg$5*x4q{S6f;GOFAd^|`7Vu34aa*&EE?z{>W5CvWERGc#Y8LefvYRQQNrMg zpp-@jC&)xXHP$mE5*P0~n+#%RVE13z8wk*9p17*RJR0y6(KC(nT&~x> zo~PH(CK?TVDKp|P#iWa(RLNsC4P6TMN7h0GysO8(N}sy_ zlS+cUwuJ-=ZD z`K1XkXaiWtK<>zLQ9R?B%qzTtqr)sRB=piz@^<5PEOb!MS8n6RQlbhZfb?srEtGBj z5t&wKWiCEIkS>szhv5`M#U`IEs={1EBioI}$EAmUwYqDu_GB+^G2A?TwElVc zY9-O*MX7ZwWV3^1qln<-m7?(Io8T>c%CHk8%@N!f1orwwa+|=AI@D?e)DR@UCE-YR z5N);qI)2Ok>sQqdY3WwAV|g`IULD@L9=8dnfcY@mpgZ)8Q8Mr=iZ5VLr`NL1b-}!C zaRb;&w&!phVtvIeRrpcpx3~{p?YtFZturD-mrdIS z5h7?!$uDbcJJEy4ES%(v8-R6N##WFamiIB$UxIig%2S2(Ru%9%DSQ#X;!Blz1#v)A zACk>H=Y{_h&Y$|VSIY%>S)+OSWUH1sV;(!DwH>4Q+ktlCMIN}*3caH3bKhS$qbM-$kilznOk&va2d4`W@4Jg z?GpwVTqB5gpSVYm0wW@=?r*W%(JRXMXoOQ#lZURNawf$hD@&e3U=z?&#-X%>Uaw$lQPajv)X7iuC{j;`lE*BNFqX`Cqv8Glq`}aN6Mfk7K)|*jrQb zYSXQl*HOOBxMZbCE=y`N@w8qUg;W@27_AIKa&5g@Wy=0yW7&7ycl?KP!2$)KQE{ca zebg^y50d)_1_n9{Kz=;B8(A0bb8c{>srR7ntG^7$zr8wD|BT|xWMooy$u@I8*cCfr zkLmo3;tSjt&;sZte``YByf=I3lNu#WAI+c8%8FLk^L8{N4XV8^$fu*cs)YB9YDn0- zpoBZ~vCh-FGcz%Y&5xvmGSMd}RwFA=@0MXU_@Rku!QSH^<2BVF>N z!X7*-qokaem@JuQ9IzpvphS)A=1Z}tUc^ionbiE05{hco(<^=t;_O9oGXYF{Qt1Cf0z^fDfw!Jk!IQ!qieigrNgZUn(cz* z=9UvB+^C7d66Nu!^tVlNPNt-@KKUPNagUgcks1Q(_g4;ULjY`ydQL|H`k>Y*p`7IW z9YDGENnEAGgSZKb{AfJ3=)mvn8=Czxgghgzm_e6_<4@^)|1J=m!hxPLWZ{tPjWxFB zvsPnS(JVP~vdYRZXCp<#DvIVfl-ykgrI4QRFJ9gtcRmGlE+{AW@&YKvHgK2rQXCUs zmoBy`^V`Se<2%yk7NNkoAUFV`*zuVM6ac4q@O9zJEhi@D;r-82ym0R-VH&goZ3s;B z6+{ahpu&#k;iUohyV3IlK&dh@_IpZ^-Q%PoIiK++a6FtRf(y~w-#x`IPvc78(3ztF z{8^B%Dto(;71#q245XMnhAA9+LQ{5PzVqOfS5J+N?FQcuNL5eHUiV2t%nDY95Ku`` zFFJEf&2U_m&|ukeo$`v7qBM)meLJvx@8c_N>_W|U~=tEITjz7L;7+bs1(gr%B-8#msbQ-n`S595qUBd)PC zw^K?W?PI~)$fWSB{VkTpcp|q~BdMqJBh^m*gYL9HtD0p^{0YIpqN-+5NfXiC^t=F$Z|BvodhhFNHDErrH6|#)ZLgB#ca63STut@;H zybO6lC*pEk?;FR-e$;*d9`Ogh?uknw6~r%3PJWW93?iOD09!&`@8tGFzq7BQR59NG&4H^U9ELO7nCC276z>uIPPcnTL!BHmY>YBC9gDNg6zD1KpXN#?;=E9Q%(Yg08?{xvjPi zWYz$J6-YIm!SmP#a3QiCFNtIp*DoJyti#G#$i-X@=gUk3vDizp`nZcL{%X(1vj@RJev3 zy0GZMCAR3n05n^2KVWVYylixF^bEm(MU|P7g8L-yM|;#lR-?&s97Rsw5hw+;3Ty-R zB@N8Mv+S)w?5g+K+=U&fCYvbz==&0NE+JHF~EtxwVLb7ELYPa zSf5fCU?{@X-uHkK2j8e@g)sDl$Z*p`vGXapN-=;1fcTX3`wNoKk0yWQ9&jgP^xh%B z&*Ss8*>yH-mE~^lnk0`n6#0tJ-*axJ41IL1z#DA&a?JI?l$gKySxs{B-Bt1jQ2(n} zWKL{&p{Au9kUk{|&sdKj(m~rz=o_mHFojsiYR zmqw5V_%EBgGk^l|3k#Yaob45PsUT)k=KKYdx3QYcj0Sm;jaK=L7#01{UP{z{=zuVw z&JJ*Tdl({=yFBtq`n^7MFik1_0!n*3OF!Nqi$feWk(p?2uQv7Y^_9;VGKE_^2~Mub zNtHYh-p@ceOOEpExlMLqc5h7vzTCm#7@(=YGkey+1;&xqGa50Mq||W3qFLr4$qrk! zl~t)M0R&uZ3HFdSf57}Af(%91@Jhg zIH#^`(E~hM+f9%GcA;1DH?K>|OA>oazwGO%x{jp>us2^ei&s z&aGE5=u9>@bKHVy&|l(f=Bw`@HeOmDe3MelG$H?`bA$h-bFP{{bk2Ha393`j95dho zz%Cof2?81b)1_?29S8u^<075M&Z|{Xge7A@rf+Dg5i-z2z{*Jcp|Z}9EDcC6#weUk zk_h5hSq=Iy(B{$a8mq0_II5{wj!V#tCOV;vGT6CLWXeL&hCNP@YsqR2n^TZyVHg=) zm=r9Qvx3ZJ1{`@|E-HRrdc-KbM7{r1@*L)%bCrYK+Ab|V$^gv@52^~)Lj55<3u@Ac zi^s?Kf519mQy^6?b4DM`iJ)5eDguBXgHtZ&5a7tTy~?!V4YZnF#c?1xT>4H;VW)}k zRHu#MuV)ET5F9Iw41fPtC90I@du>6petHhaQ=@Qh8Z%pti%|}T_DxQrauWY5>c{TG zrZMToSt#%9{tSRP=jHjH5LzO&OGOgCMvrfl2)6JwAjRJS;Agl!2?Uo0D@r5>hd`f6 zxi)?1lehOW(K7$*?ERh;QjS3xqWjGQ*_GC*^k|cTs2IR(#xkUZd}}&YBR(WzAuR5# z5CN=L#^VH7L0(U-M-V)Hx_{_hI;!W-vER`e>ek+v;|Co4_FEu0DK9{bYKn2Ad7d^$ z4rf*@v?XPj3M!nmLZXp6vF5Z&O3qD)(s^;TLX!xzd}ai7S&iP?lcAJ>x2Db$!YWc; z_(5~921H2ZC!2?K?`^x(HxWvrAJHI*P{1~mUtkV>(e=rOS~PNXK|W)MW=@jr1NRZK z{DD3pKcAygF^rhZkk%G~zo=xqIwZIW*>gmH1yVTB*2H)7&3XfE0)wOM8vT$t*Z(1N z5ca1+*{kqE)s(Q}h*87C7gOsQD#qw+RVE}hHr}m{gjfmde5E{We`VBif7C^pXpWoh z0vnEgxF3t12Ro3ONOPZJ9#jB>GrLJsoAPx&0H<%}G~kDifM<&5R(Ml1TtZGuP(`ZP zbSaV;@h`oR3i)lGK=JygfqzaBg}H>hF3>>5k8b=PDDO`V?g9E+!Q`V4>yxRhEDsI7 zZ^8x2=oXVKmW!5=P%OYX74?KZtOV(KVw)d=W&qA6v)Q` zU_vre2@)N_zT>P9Uap_*P#ni2Kg#j6CmRwivcNigQc2-fvu(89NAdjCnQj3F0cFae zW(L2yOalEWUA&{Wu)K>%E>qcbx!O96KB*W5*J&n=sjiC01jB*|?Iv1%5ZdOVd;`3u)dWq@adYtQrpzW^6ItqL75 zAjQghb|F-_WGq!O2gm}Ji_jN&W3M`r$|Fq&rmLp#%oS9$vR0lk|FL3I7)e7N$?tUX z2z2vq?L0FxzX(xay(@}i3^;?ZSuAkINY~lk&X=f$NrVG)q8r8@_5DCFC3|gFw+Pl~ z`YK8mAvrij(l<=Xu`E{|I67+Df8V+x}xxC$&xMv;f*(D$a|_*4Bk$6Ze7 zWa)&yG`UmZ((Qsv7BK9^FH;mw{QB^PxDu{&1}yh$I>`|@6vNu|*`bYEwxxchbB)T5b@&~b!kz-kq-EEXj z)`ZE#*!EQZOe9}Z6n}Uwt65A$#s&Spxtx^>p4!w86oY{!o3C1Q6@E)LWRVW?0!)Ct z-YX7s7TOCD(JKu%M#<(x%_hTCo8B&YYNe(TY=m8{Kp22fJiy1x%~r1zpW(ldjJq+| zzlYCZslx2QA1zzXM5)vUGsO8=?dwb^dMBoPW@_VLtv#P~{e2{pip5w4G0ls>=sBn` z-u#yHQd&!lTU1?uYD`k#t=@jE%F9!yl23zpKr9jnCu8)mbx_Z0!A~yR)}y=4U0SHV z46kye>j|J-LHwJpR{$?`$2cw|0AHkSy8T<)iLMj^d>ZzdS+m@z-Cbeb!6@5z?eEfr z-O1*kXl*=&B_X)Y;DEQL%y7L``T{SzWzwyG) zt7?jtU6CEhm8d0PoS`LC4z zk+QG@Zf^>ZX%c+ey|vJd-cJ2HFS7>nPC{6uNucHf3&5&A*`3M9E4iv25z3ku80<|~ zVB>TyNc2|cs}#BOPK19C=zbk8ygO5&wu|eAII8m(KzFW~Q@@~zSLrw5NmYW7)D{z? z%DsW$!~k#=DnFan~09&rnJT;_tDJvsDvP>{uoI6n{in)$wc{zgun z-#rv9zQ|vFZE81QD0%kr{q~edVv*c}xmi@n0{*nRx8U&cedki?kJeNnv45Y#Li2%# z0XGf%ywdY=urD`he4jM}#JB0BEeb_7PG3eqjsF7W|K}N0_%&>#D=)2Q0kR%Gn2iPksoiEj3cY+LYNeKjCHAi+1G!* z;1O9&rL}NLp&5fsldWdCN0ifBYi`wz(M~R6#oH=P6^%OjPsnwN;C(0w<_UoHqoGYb zVj2d<0yL96oeOrwks~ydEhA-K78lQ4O_O_#4K~P1Cy3_LT5z88r<`TPc>4{e2zE zs^j%khYX&vH@eypc9&H8aw_=Wex7{nq3eq94U;|i=+ux)3S1!?8v3q?o*KUy z1anTCAlz|oGXH+{in|%(_O-+2TQk#}3bDxy*YAsqH&YXrfi#lSDH&@HdOCc^AJ|R5 z?S=uv6LJomm<&A8w=K%7>2hc+**4A4G8Rq^m4+t0@uX1fN>GsV(Lw>Ps}IqMu3g+7 zNOx}!VbUsGk@%BYt{Iyp9HAZu)%RzCPP;9V1a8;x47r;ivy_Ul-IBp#%Mf>uoDIVG zb>D%|x;H9hul8CYom<}LmL4&NUVb(1d(p^IIA)lBMAC|{Xy;}$LDw+q zQ6a4cR#N$y9w-ZgHcbGQ=;H}NBtFcqm2EeL(+xAq3RXPY9&oL~7^omC!&HsQfM5@& zA6r9Z5_&9$JZj2Rkwtk9X`H)r)9?D5S&TN2<8rWj(HS+()0#%xiP3@psWV|XO8}}{FPU&^9V56M4n(5El(4+oX0sxk&X8)b5&!*OfER>bqev8aRw=Hg!a6HzP=5+zn{$?jpw)pt-?x#-J z;o8Z^m-Mz_*yn0WCExEUU9Q1D*VE8IvBOL1sHtV?Soi?XZ~n>Y)+3VdGERN1lt;uh zw#_4}$q>AS{_faDj6&WTUEiavngSvs(0eW(mIlRWiB@5*!jdX#cAKnjkG&c)TBcVM z`?jr8Q_sU_wZjYbxCJLdao8v+&y~L7rho~Z_C?bu1!=fc*;}msn!mDvBE2E-5_Q8D z+Xjrg9UlN8ftVf6lei4gYX>XP!Xo0oMZFo0H_|o%`hPpGM6Na1Yj-Qnpon8<(0R3$ zGm{Jc>M5PUm{qrNM5-M8JEqAMw>j6Y!3H%a%<3z3>468~uEKqz#ly*T70C@1De&H! z%~+VJBBE7oeKlXUw~wnK=kn4ak+{lfP^2DPjG_k|IG3I9?dDADl^8(Z->q5)Q*dB! z_!gdSEOS9PNz+>HVUrkV? z5dME}H+L-#dOfCi?RB&duDLZIQgx()-wY``8nj!dKGtirn8T<5xYv(GdXE?Nu^7M< zR_`1Rmm(PEavVvb6P0E0=?L2{p;k2qhI98n*m_w;{*&~WI@Xmks7LDGZ+>F<#6C#E zbSm>VQ}Ey5l5fpqohf7AuI}s|wzxt--S1RTgfOCvR1@_kB%@++TPl9IxN?cj7QpF$ zm)VsiNktXI5y0<|y8a>G;rJx+N*@2{P${bm2dm=qQnB(!KliRC>CMf%6*VQp2Wu^JE~vsQqHfRR4_mbw+3%S7Ls-@+o@y1G>F)%epoMI!1tl8>yW}AW3HL)j(IhL$4CEDuxYMv6V3q?gJ|@z zL=;dNq z8u*~>NLxRCMsb-uqM^;p9V!8BuKPITDz)NkT0>Hy)jY9r6dcbvo2{5*T}GdEzcEpf zjX%?Bb}wa^@8{?kDoSnBR%y&F`usVHxgAw>&Lyrf3Z{A6NG?Wcm4_7CnU75vmoQdt z3?TAs=qfhVh!gCE{9hu=Bvz!6x&-lLCc6|Td6iF+`iFfAT3|vx1hN1riqW*D&ETnvE?fAZj`Fh$ItD8nJ4I@>%nTMBtEjj(oPqE6n`+HP&=!Z^d3B=0)6nz5N-?bF6lRSgWAmWzA z%*2(`-p2bpGzR7>Y`heWdqi-18nZjttoBCO4zzLtrgKITISAw%C|+J2w!@&OKu4`| z8Mf?3(I%TdW_QG{S~g;HV@|P*3q3YP*q%jk#qm7E61@!`t1klWqwRhXg9bP*W@GZ)N1@;M$o;uiCv=>#DZS&#pKcVu}VaK)a-tKpB#5CHSGGPj=?T`EW9KBN+U0p=z9j7rfsch*rOv)8AGo9X6R9 z(z-Zj47moj@ku|)fi+hW2% z>=Qw)%Lxt8$GqoUx73D`@r$&89OGn@4Nq!oz%Z#M2X7Vuv}~$TYh9G+BA|B6(G(5y z`AHc46Dx&yt$2RCAl3yYqE*r7OY?M|5M~%Cg&Tz@vEIMCpf#PVG~gvh8IGJ$&|T^e zC^Dz6Pg{=62ygB^`X$tWW6MJLMuFnIYSn>a1q(NT>&nf3qa?H>TTXo|$JgI$$VH;& z4`yHkAtb|ez}MkY%}U_0z=w)##J>}HZm^rLMGIE%)c5wKz8%8Af`I}zsBf2S-6MmP z-e>BJryKfBHsotxJiZE2xmNaYS$c@idrXEILiRr;j@;H#&~WeQ76)3c3^mif&vpCO$;Y4#;UkSIkXOKnWVCNShsUaj&qC2{4FqQ3KsV={r0%vYm7`ZF~5gGi5+WuRn$3?mtc^zsK+ZMm}) zf{RFx0S4R7U#QUW%D+?gd)=+9yXWuegR94lWyCif?sZ#>`rh`T+};&dK%~S7@z`p% z5^A8H_2DS;W{%~QH=F`(sl6pz5DGAQqWFHyJiQh0AKn*V0H_Aif&G}Lug!PJ|BJqt zM}+bH$bf*r-+_Rr|BJpI{cTWFZPJP;z_I%Z5lN<_K{BPSl?E* zfV2YjzlRB2Rk6rp-PVqKn4V}p9OZwCdcXS?01_hD`VdFlUX4aQxaBGk%*i)PH9c(- z;bZ&N<5a!JBNsPjXc-!HDr8jB%<$h|ai|8d<+%}NxhAE7#`xfm2sAcm6-#o(14*v3 zSmk!*vPp~!#&s~J#nOoW`d3w7P!2PKk%qx8BymVb#H}^P^J_8)jFZBPH3vBfU@DX zn7eP;+SfcgKap~A&ja{`JAb+MBPk*T0|a5R1E?t$wQ0fe?%vYvhBs`ulAkNhj+*)! zx&qwd+02?nR=Edydrd#wgG&#@GwXN=K00WHIv+N9`B!pOJ1A<8)Pc>uBjOF;hDT|^ zGtkCOa+(LV{pP~s+X7pSiH++c;Bk?1dIvwd?sMSzdhJ=_W+hX^&CBD?*Z%v)&*!6e zTHvbTS`Oi98d3XVm*oBC_9vw2<>EW~b$i>P?B@HTe|&Kw5)Coi13BBBBm-HKqfy{{ z_jZ`!E$gY`cZ!Bnrj|T+Uie_KOR8+_l*^G6e^KD5kve9Sty9g2?>ayWs3ix@?du>f zuBajrKOTLGsXPa>De^tV_N0+ZJl7S^*GBWrZgS(?ocGkYmc|l!}NKc4c5T4gH&Ja!V!2#`T zyR{g#gd(_0T|?fD&BM14u*J5D7NyO#G+3S+JU_&MrWy!2Z+R2|-oPX}EX0q14sSCf zrGaN5QM3XB5w-_REIGsOn(*(8k*Vsm{cy4+d&Fj4K>2bS(ts^qQgsB&8a%IS&<*=H zR5^aS;K}6}5v9m55_C3<-0*3ys4Gc=B3;lvXl<{58~p&~pbn}&pm{YudRrzyCmt3G zoX&d_oHu8tf<{=YaeI1PWb$$A}l1!JF_s zO>e;)C@rNLa4ee1-$T}i2TQZA&Lq0&jp-;bl`LlCWg~$0NYy84M z-DaltzGIGxF+t(DO;eEeyGla>;bjG@v6C%Z^&x*M}&z#VjBG@59)*0a{6f5{217Q+fR z5DZ#jj+2YgPas4V$P!4|6xBX=?}-J!lr43@gr2feNXK6R)>xvXPK(X`^nG#{5n*U} z*^vKbp|i%Bv*<$31RBsw`kQq{ai<&n@i8gO>i*;P)&|b^%RBYb|Ar|@H!Ui!nK%#; zYKqg~Cn@N$%NQ`;vZ`69=gFBc4X&?sH4qxGV=8mb>>i z24ASXdY{IWPC`sFB=S&0+<+QNU55_>G*rh15Pysap@3h(kRD?@9-o*3q;o^v(C1>} z&{bSCp_-5$L>J3V{n7LNx&yXSNJMJ8Y(1y!%DCsF489E+wrA5tC(=9^RNuTS1T4Z8 z!sc<5(tX(YH%Jtu86@KLfWkD{>{my4)0kB3*`{uSt6C&vTykQ7kf_=&?Go*V6^lU# zpk<323zJ}*?sLIEXb(;X0|E(VmW^k~xIUn&SKowaU44yo7CMYk&0iw%`1eY>XZ+0xuka*`dZ54TKgfidT%kUI+d8qL--u zt++T#HCFt&0W1a9u3w?(hj^+8U=mnbFVN2KD`Bgl9D1G#rlL~VuVpF1>zg?nCU3k? z(j_$EoPxATRH!ut z=0BZ!&U~fac$h;Pigtt-%pUPiSMJhA{R&N6B#gmwdC>7R%K8coJjBWwkP;IR9M66W z^H!{A(_2x_joeZ2C(o=F;3pLq{Y@k>L&i|a90tClh@DEGxE*W9};XITu}j- zIS^d44|8&gyfb2Z)1i(PuoK#3oXJb=V4M|`Y1uu(Qa2Nn(^x}n>K*1^;c-jv zE*vgjMTtqCAL)uTgE&jAQ=g4U1U9B$TcW6PBg{>Hz&XrIMu9Ioz%d47Az~a~JzXju zdZ~G8JP3uNp@0N#8C*3L@)(}t=$YT$?f5#sw(S@LvEuE~$uRSJ3uybEE}rIQpK~d) zdgR{ja+&?lYsA*B{w2oo7u#)H`s&04N0OU;6UOGq7@gQwY4sLzckLg@8(T2)izecO zO|Fc&*hbyYKemYRfGS@@vY4x-kmGA8#&066C7T-a7P4O_ee-;BK9Tv>!5xgccw5IA z<`#w;;{e~c3k3RM&jp+&b2}$_Rz9{Xc_b(=u))f&_t#tRnWBM8}t=Cz2drD6zp z`auulGVjJAhr)5C_6m5?r0xoM!v60857=;*dNimCNk23pcx84yt_S{?!PRrBl315o zYZ2G?pzKXB!{0bX0i_qDLB1=974^PMtM}KA2hCyq8`*yN>sKu?3z}}P0-t6%S zybK9n8mV1o9JvCy)LsCb_*5lRJ7oRmNHj6DgvB|4F8uXRx|6_rHNm!thHo836aLeUv7uSGd)TsgLdU|(WRa= zS|wrguiH3tDl_{P3%eY)b`xP|6yU3EW}4H`{4nNb9*>gepeUPY>y6-^PP<&v+VU$Y z*s;I=rsJ)Uu$Av5@axj{KfG%TJ7qe4V52vk1(?g?2$>XY$;*klkcS&)8pq9nFLwzSdGlk|V`r;>Blq)CB z5B3s%H$0Nd*lwq8Bc?aQM`nw~oi8ye)*$Dw z9!-r%Hj1~o=u7=-^Eu_que%g6-smwaC)FGF-r(10rqUoTke_;00NmKT@BZi**P0RY zFCl}z&Pg7KewK9H%yjzOfBqyYe>J6>>ppufyY@FMlUrC4B&qA_yNOtpWtndsE3Txj z`M*gAT;;%d4Vcg3vIS1px^(XquY}BGa9AzdV6(2@uHRta4R7Mtc*`u++|eat*}P18 z96ZayZKvn=wQGd=03tVNGBiq+g0Gx!^>Sm92X}(CN^kXUSju@glND<{6zIM1&42u4 zZRH($rO(>V_m;YCOdASLex_xbIngGG0tA;COwn>0Xh?73($t2C$etb%#RyS zo#irFrpx+3ycdqs*-QA819gLCbM@Cd`Fho=%ZT*H*=+e+iy>DYTy+-s@_i;kpQ^AM zV>3*VZe;5)r_x5a0n^Yh)~*^2e?ezog&$ieBss;5th&wM|Z zY<5V)9=R@!^h@FGGx>d?I3BH3-lk(ciM&g|+#s9)R`%rc#l1yVGzn^SC@hU*NN>sHJR>V-m1o)CNEfioyi5UxcW;PCP#t2bt5XXSP$jxhK;8}efZ z^y|7}c~pCCSU)br-qbz~+ri!y7a-)9XShGmg>JX0z0ZyhlX=9ysl76cNp17T;h%ep z1>jXjRe+;6Iy6$T?cTxAZc8X_B%1ZKJmG4C$WBe=9k0b+ebQK5i@k@gtc9c+V)BaJ zV7|W$(BBRSt|GjuqS;e5l*B?{xYA2LrDNPv4IX=c9&x9edRR_(U~(S~!0ub24~ptXexYJ{+g?qzoVND}u)n9$UuEmL?8ZirZA4mgc0HhB_u-rba870MBix^ViZg}fEc0~F-KAi>H)v;zlA!tnOe;?NPu4;p%e~BZKT-%)^!Edx^dSQn zezi71F8zZKC%+#6KPa)Jqxcwr0fk`iHaSBy0;vwA5-DFFAz_`A7Y)dDt{lfWq5Ecqh zR^>%Z|0hEJ8Yjtbk^qR3x$8EHYKe+bcXeU(g z*n8X;tvy?JSm(lUWn?)y*!$p@z+B+)$kou;{d>5YXzUbI@J?i|)pZQJ&=ZQHgnZ5z}7wmofY+O}=mwr#%s z?7iPpbxxi0A$O&cN`#?53w< z=zGY-wKPLmQfzHn1{H>Cmk^vz8`1aaRFaliWs%_`+$MAXGAU@r^B8v#RV#8sZW+gUx~gWS5qt5x)zYHd{o%=!<#C)@qf0HBs#JlV!^(8$}{H(pm+jF_H0 z6<>|FbtJLmmu04T$$TrXJui)gm8l7P3y;_w9vy_sQ-txFm-&ZZ{T@GJTx#mZX>p*wNi0 z&NNUWDciJ{vbi(J4X*);2&}E8oD`UCg4aKmQNc?54f>8+W6}xL2 z>BITt)=?T)%L|w*DN%WaeExuRo!&C>QNrIh8Ct@F-P-q7!rxvm*a{K6$zF%O*5UZ= z2Xt}Se~A4T%7e2u2UY3Xu71k|Af$v3TjW(jj}ol$!(`RF-pwsRfSmT9O*@=l@au#O z>hl)lH0dD%b1twwn(Q)(Dm4+m&XI-GlzFr?&g5_B%8`tUBMY4zLV%WbUfQyi; zcBIRQ^PC19EE=xDaU`-G^M;dLopFkkyuQuWJn*=FX3QTGT2WX=_%a6KHN!-8T?RB;QV$BwYQ`k=|x~?Q@ z6MFf&GKD-8;s-HKZHqVn3(0-O&uCgU$acfo4nA$5lyjdRi1FuF!Y>xefgjvq8`Af< zp{N~$98GLnPthe>Xo3~ey;=L?!4W!xo6rW##E4ffXjL@WE92AvEj^{hKj#wErh~XTQ<*vgS*$(xp7r*h`y5_n(=Sb3F_7Rq`%F#0cqUx$lf6T}o!V>g^Y{))DF~pV$<9YxKW(y-twRQs@i*CBk@#~$=1Qd9r+j}fd4jyZt`KA; zk2~(wGM#DmyQF18pIANFiPJg@_#shisp_`U?pj-ir-;36-{s$GBVyXOli?#80}L9Z zsHQoy1C;atBW=u?xgPL$BYMx5yAP&whhG5i<+k-9NAd4>>LNJpg@EKeIH}DFFb7&T z^jyx6LvHvz;Vp8FJ$15_6>{?PQ2#D*!o2EPiPOI>U1^>7ahbbez%Xc17Zy@eX<2;R zX{*T3nM8Q6i17=&2|(b2b-zM0DbUGL%L|i%Ju4LeY;MZs6dCCIs@5C=s3=>jD*$OW zU$tP^*^TLtFiHPIH%m9xWPR^45l@lr$Cq!qGkdQ{tE$Dj6(`0FUP+(Az}4FBLx!^` zi_tj?ZN!A^QR3%$H?-!)SEDs`V82W2HxZ6ILDmnaL_^7Qb0DMwgfC!Kz8ld-!b>}u`QjP zv9EXxuB25{s`Z#5YOiMl<<7WDv8wa!eTclTS8V0wbNH^r$g)!PYby|MRc=#efjG=~ zUj^OF0bh)!DptELX49-Gas_*hRrk9|Wv?QD6Ht-zpIM|263oqpT(#NzoaN|#^CzrH zi6|SsMw=87A5ZyXVHL%l;#nfXyFbd`X$!*IfVY05!ZL%t5l^bPxw|ysXisDMqPxcqkHO$j^U@Lg{A((*DZqkg@=SmvxBbeX$CWiQ;hxDVo$!Q;#oI9 zXKQSmby@53#XDy*$AyX;&(rnVg?6^#*0a3;8LTJB8wx#&r*BkO&Ll`Hr|N{hxczr5 z;jeeubkVaYI;u{k}475W9Y|P`?FlWyWP#`EA0wZS+5|9WG955r;&rSg> zW842KUO)wqjQ?E#Km~}5Z~v z@pmMlIF&(S3jIjwb|m3Al|f?;{7CtBB#}6k!D7h#N#%AV(KwaCV@&-?^>!q&IF%v% z6NQM(kR-9e{?FZ>Mj&0do<$&Cgq}yB|KlX!|8dR#{Z9Y?(LXiE4Nw9+iq>C47@=#s zBh8Iy`u7xm>|Kh9pXryEN|n7CwH?d&41DxtfQ~%3(L>qYA3KSWR;IbF0J8Z?3c!iyR<6_ApJIDi!-{O9wu{%7|3EV|%<>6w@ z)+1SW2H1Kvd0NwBb(u-7L0Y2h+EiOvjGwQ8vhv6!bR;**T9x>8X64K8;7w9Y*OM(8 z3$aQ#zsrkXC5L^c==Lx7x4#?l@11`tdZ8D){6cp4{rm;L1 z;}{{Nf4`h7NGLNVO8H>Sq9dT3RgoyHGcEypjtBXK5X8w-tV~d~E`{MjKfVFE3X)$b zUDYXkMRxFESWdc(rhn?EHWhefA;OgM@T8+@nR}RE26%66SR~(gVL^vkY zn=iVk%~60(y6DUkK91X%%oE;^rBTQIy8qduW^k!i^oQs;C2gY(Fp)xuW3^|wMlb$4^CR|2gyXHlNcy~ zy<&VEVMlqYdE=}RVON#y){nZGwddP-v}A>IHb#w6J`SWY{$5Z@)UEvEOB-rhaR){z zu5-_5?q1jPEl4GIze*LQZFL$ZGZ+QL^SEJ4yWR3V;rqkUaJb2!H6agWNl+-!T=6nX zsW=|SOJzrp#$F^SXEQCE>H&zd#^`onu*PRrs*F{~veO1iTIR>d1O}7TDSD}68tNhi zY9n6it|1FmS_O3d7^C}^dRpd3$Yckz#3`t0s#uxB6&wrWt)ZkzZzc(Fm5Tuz)XDJI zKc-r2sw*3&Q9(zuNGXU2Qk{!C`3ltUi+7M!l`E1I!3sfR>)ee~Qxx4@dCI>VFdE!x zh3cGhr)hoEiV%f=1c;Kx=*&TkaLk@%hzkT^&2g~7wK?ZTsHXFsDAes-;1e%FY1M-Q z*4%-8tE)GHr`F8Rfj3S~5;Ot3@`V4n-&<;sM$7y#eQh5j;Zhf59G_1(A7IZ0-jByy zRvBYs)g+;MXN=&lDY*O7iWrHeDERs_2(u(DlcdQ-vKYlG)D=4*rl)z-+$ z+A^LM8x6Y#vtd~fpbgKTzi8KEzfiBg&+s#1^7M1syAZ}!_O>~#wz?4Gln%w+AbQ*q z=?&b522QuL#3=N?t7w@W-vJaDtDYa#rnA#|fSjp4&o=y6^((_jC&+fc6~4c+x*MBm zrpDV>V`QlNPv*u6Spd2BGusPJ%|hq5ty4Eu?bZT18lE-qA$DEvcezR8cFN&Ek#2jF zA46ex1FZ8%$e~F;@r?@X(qQote|h@Rbz^A7n@E*o$cN$!D09y2_FVqyk1zCo8g3E=OU2Dm-FaUd*osR6+S70DAmuG_e zfla)H!E)lppX;LnRZoe8$-1v3XRov0s+>tFzK2QW9_Uq+TFyKCjkGy5-wSKd^ZBFr zGJkuUO_Nen`l9)NZDq?{7Y-$|DytpG9^Y*)2a>@RpHd}jn4k&FRpROqe2#9^)Jr8R zREq0JlbKq_&H^0xxy?G}U;iu?n%O=NAha#pz&+a+qi-A7L%t{9?aX;{2D*N+`!0lR zE8Ol0e7_uvj4a5Km0OBdt5(#id5y&$gz0d56`}bf@tcO^*M$es&BeigB-)e`ieSi_ zDUp5YrH0}j+(JJrS#Tp#izoWP6y5T|?Y-!)IwBUhAX1|#d}Erym~naHq5^j`bZdDeQiMj z4uF&qV?r+{<>EBXhW+ z?ZSR-1R+xmxvf{Kr24!5pQ5@kVS{_sH<7EFH;JXpO#+C9Qryev{HhR9sf5QO3LJ_j zV+k$3w!L^Bo-9cIIB_YDbf3;@>mrL#Zz!l|CBXjZR!BTE!-=~@H~xccnnCpYgoI}_ zlD9OFKhOINpn)WZef8U|U1Y5GhO*wVq&PrKIx2|`K**17H<(CB0ST3?)riF**L}>@ zir%uv5+1D^Py!RG_Qvvfl*}&@i&Rm8i=H4HtUhDF@{QJbT5GVa@>wGD8EU0eosk!6 z3h+al?mG}M5_U1UwT)R%={ctFSU0&e{xd0G zEl%~5WVu}SAzEN2~Yin7GOo4a5u?_IP*Gs>Q2odN( zQ+rgW_U0v=O517n4f4O|s%$_?{~eP80jb0T0nq~C0omERm>H*pi}Aswh(fUu{lB7B2fXY5 z!x_P9FEaKfZ*@NH^jJ%C;*pQF&f-ks%t-rh&d3_tD23#BdR5g~>`Ka?O@(uYRcBzq zIFiZ@-P82Sl^D@Ir>Wl7eIS8vAsabWyo^dv$5%*V-w*>|W?J>MfMS3SY?^_@Df8;t zYBWkB|dFvtVx^AIri9KIB?hq zB3StT=GdcqnT@H&9zkq``J|%^l(NcH>TltYc=AKz>jCoe> z_0mT2D(5*B4Yy8cAV#&|4W8PJWVyLXT)Q8M#is>+eo|$n_QF4D5*5gFUU#eJK-?Q6 z+iQ?sL|CTDh8Zml*!nN%`b)xMwdn%IGMOg?IeY3TIWm1_#sGZ8V3&ohU={biNE2d(}5Y<{`kR{k;;g=v{e{LLzE_RJ}ClX`!Y7iw>cJxA0ERfvtQ&zh0lv) zxmK8+d?YSfk=_g~oWhyezRt>s4z12rffoJDu}7L1^c2tGLmpD$3GIv{@72< z;ElZFis+e`Qj`LOEcj}vkSLnRO5mR=OI9wOf#otXIosCh-HQ`fXp7BJ~HlcRi#HA zN^{(%YW2a)sKMx|d9Ymv(gylrEgI4p%1Bnxx`KZk{TpXeJiOI1HNO7syEyK&7znuR z)eR-*E4{?kWa)zTv%2R9G{_HTx)$2gGnabx(RPIhA2B3O20vb$*pM|7obj~tPg`7g zgR*rH*37XBvh!#QK!m&H)Rj=lh?UJ@n>*5D8pJd1X^UW5Y?3P@u*jUG@007C@NqLX z_efc(#+PfQ?ISdrjv?S{5w&rM;_={hEcAHY153=Hu8I+Axd_ ztylg!+<@@$Fin`QYAz0m;sXPm12vIZ#AiEqoNojQXp$Z`Jvm>P_J3u0TBDf5Px7rq{>3cdqz8PsOF|`RNm$KT@Z9(XV-XaxKOsi% zB$G-izkoQnM$mW)je6!LVv2yW-l@}oK9}4CSU@i=0Zp39G>oRGOGU&#Wjt3bbcIrC z5#v!{C5Luz0>%l58YB;Q(g3K}S8Nnm04%8!?RN<|7 z3#y(&ii*nX$WeGqTmJwEF;1jnmFqT&i{zHto75V~=cKLQG&khhzk|y7b%rdax~RYn=y4!p|qBbsZ0ry4SDB4?Xw?W8xVk z&rKm^tGsfg;Z@j&nP&Ruk6b-N9e_1YbN#TM?LUHn>XH8z3_LQ%q!uI3co#9|Y3%|--OA%1vA_eD(xTQdDsa>xjIrQMT;_N4IF9My`72ey?-x1bS!D-5r>tas z#QFr*!gP4HML|akVcanH z&_RIlKxFkD)DzQ!mZ^o?HyT-e)p)49Y(#t|R&IFmv^{eH4Ve~$KLC>z@OjV@3`*ut z3cS8NbG4d2e6Usg%P92uTC>1@dxVApS7{tbWNyN5NnBdqOi<3U%3=Ht5zY#{9cC>h zNhrEvq_9F6&F?^Vs5fK#*byUF<_70cmZE*UWalXPtS7GWk}g~^NrMJ(# zE;}qZRQaArlf)(A3@)`u?GM5A7;%JTu;ej*M|MT5OJC;6-(|Ef5;eR#=(y}S>to1= zE+mrl<@{m{N9q~6PyoWz?a#l-CYwSp78G*qMZ|Ct`=OXf7fRs={9sIYq~#eg6Z`>e z4MK5fo^2aHf`KqfbczTC_|d zjwHlla?TyF*I}BJKcbQC>7Q1jO8~CnuXpFv9eDONcs(W}zgm~HqeKFU7Q@Eg#)+&R z9}Lmx+RPhbe-RKLsR|LeQBD%aa(LWpU~p`c=oJ|7IV#q_^-?*Rou*M%KDEk2t>$<^@gAonZerMf>Y&+NQM zAjKK;Zs1AK+P5tL!LTrrx{x9y0?jJFT-2o&N9)5 znJ?jJBx;Gs_9h}O0ltVBtuiDaL43wOm*yhja7EorLy1$XB>kPW5*i6$nnzo2$!;CM zWK3XD#06|O^SYS2RQQJjgp(*3*=lGrYUQPsBOZVvaa!%p?bYlIa% zZ$)vQdHEsnJo5EPB8gBq=Qf!jDoZc8EjX8$s}9&_@EV)&P5>&35=1J1Ehq+*rUa;F zFfu2GdogTTrBIVDTi=xkQf8pJ0etKe?9k1nl7z|#s;mn zEU(T@cE)M|!IVm%T1sH=j#c^c(vEiJl@H<&rN*WRlBLep)mDAqM$2+jtyN7WX(TbO zwG2R+5N49(j4@`lVKZ|m$STcMO(1TrX0|CxLp)E-87y%<=9lJy*OxQ1FXCJ4Y z{E(zy*)HDH_JCN)zZ8AZZ=L_jJJt%v9~WkIwOJV;Lw;Db=?PB&&0;Ci9SCJNLAGpS zmh+&PwmzLzK++!&BevJh^hq45o`=9}%Pk~us=w+6Iv*ikX{A%{4#ROiJDzfBnND(S z%d#O-07hBX&UM5Vj@}Ek3x;%~5-_b}Y3M8F}4Jb;6EEO(|_V^K*>iwqeea{dMOl z*aw02YRZ5uX!4j1cifhI+HpIsrP{WGc3BEAZ}sCopIWIW<#O1`TjfR4(189#>;8P} z-V;7z35HbMbBf0cgA`b%#6x@V9k~X05|w2q;$YO`iPRuGSU-^3gnT@ulml%n5%H&t z8?mB?!n;5o&=EcGv`-ENw*Juhe}j7FR0sxu)I0=$sJuy!9_UKl!@6B*<09Sb*dh=@fL1Mqw02krJre`@Aw87mJ%%$jL_Opgs^(~-Sxo~4*}LIpK@tJ=!#(fHifB4)uWL8=kC)TY%!4t=>e~9F z#HyIQMAA}f=v;eOW=^okE_X<@$H4;t_KRd(7EPg>cxTXvcx2$*iVMd#QM`|JjN`4m6^WmX+h2Sp>29oF=fRI@bx~^$nMwEieRM$xC^`;kqIJ+V%vHi6-|Q z>b|PP`3B-754XVgyOyP~Wt_xJahG%G2b=DcY{j;!`kH5EVw#K(^Usrf4V$0zWDLf! zru`$3^eR}(o9cWzCARK{ux@7yl%&FPrqeGgjqK+^SMD^iNKF6ibH5ZO2#F9WSDJ(* zy|5Ld9BUd(5p6^Ds#72Y=3NKW>%;4M9p!yO0de#nFZ}83DMr9Oj^i#J?fU)$Ie(3| z$V>Lqu1bb|Yd4CdGOasd__Nm11l4edl+%jBUDhic4)M={Zh?*<7uf7Sq(<|wDu0aWIPDK157T;a;F+EfZJeKQFqM+bTR){!r`SKZDy)!M#aJ4eD`n> zV^ddHP8H&_O~-kL5$W+^F6P>4d#<*&s?f8mrE;@&(XZExJf^{iNg{P=1y*7m z^=AxiH@P$_v@Jcu!M;{RkR$x1BPjAXj}lKx7BSg{ZRu>qYNP~wia+sUc_uJtCo;k6 zJ9NsH_vmjh9cP|hD$ymt#tdV*(w27F9!Qgwion`5G#~h#&5m;OUYVA!m+Z_hPtR8Q zcoeR08kyjD5*m=LYRU87`EYQrtQSc3*e!ssqfgm3JJqqkk3(>utdESr>E{q4#%>YB zy5Zi}|4nyPYQnTaxP zxgd|h!7C^(fs=eHuh7#X=w*N_yI8F);#m>Ew#|fb|2+b@QadF;p{_)_;h+wjn%p>C zI&-^@5q(=h;(G>`G`XNI)|4~8aw(E=XCDO@|1mBm2|lieU#Y{WO>IeSFS98nABFb% ziuBjEt-Vpj>%fL-9v}d!1PnOk$Mj1dkzq^R%+Sh>W*!#s-11yLC17*PX>*iU^V(83W#1w$OpOX9di{od_ z;o_7mOAb@b|uix!#H2dc=oJ{V8Kse($%k7k` z2-9`42VG}J(<%Bq-Is)d2TZ`eQuEw}#wwG%C3|B_bDz1=r`JR&95zdb{mE~4AJU*_Zc!sx z?9oCy-8fI>Jm%FfRkI6nhqTEdLJUM=QtokmWyl3?q=9Jhy8TnT#A_C&W zM9B8ywBhSan{W}1zvcD0un-edxIVgED`CK{63l$J&<)#S=(B0P;!{eNBe{#D#kp2S$)n zhb!=Cd4Nr-IUqt860~K#A7Na_i2(LUwaYiiAM+Y31utY%5`69Dr$eygUc3TWab<07 zbu@%YNohG98pYPMdJJUp3Fyhxqx!4AnxpHjU4zkiuuU2uyn)ll^5 zv6DcOX7GMWTo;0y6Jr3zt;HNC!V_%(mUaM|2W zQchsxh`QI(zXTy@FAwcz!dER+jrGYc>MmRD#(Cbz!&h#KE~t(Y?JTVmY|%fS=PL|` zYN&i9U5~vdrfqX9>vfXT{}@6UYBOoZi1gkjo4bhqVNfUKXy@w>JOVyCWb2ggsZpi! zQrdPNu&2GyE?lY7e-H)!cb;^$RxsAZ>3weieNj_GV|x9eKh==HTNTgp6=#7Zi0V^Qm_B;oO|LS z@w9C+AwVBGF`iiqR)h`%w?k_?{LCbdUTo!(LUTeGNCo_Ypsnx8|(CwS>M@qR6C`p)-m61*% z)^9t(;$U`9$^hK#45P0AAsTk?U{3xQQ)7oDub)HnSNx=N=UO0{jb&4S4X?h_Yqgv% zh5px~>jI0@hsKI>Y4E_~2bTe=Y$I~ME7B1A=|q&ke}B}ZB{|EBN&D?wnbGVCQ{ec= zk`sh}wUq@38F}*Mxp;s9RM8P3yUc%w8+|qSHV)oGJ^f;B0NW~K9hc>M=-(0e-u+EAIz|mm=eV`u-7%1t_bb%(gtiAyqh?I%R$jej;F?ZMTLc9 zK*sl8ngGCQ_Fmhn3NXw~Xwk@dap=S5Ez%e2?uP)ZI$Z{5D++HV9CrqHS+p|D>!VxC z+1alOs+(vm=CqH-I@;N{rMAPquXIYz>^N_89QS;yF6UvZBs;_b*mG6^gAJE-&1QV- z_gHc@HwfN+z-t(_Pef7m}q2`L-3NJdh zQi;A84K>q+OV>h~6RiWrbl$X+_Te$p7s`{o4HJO;)iZBv`A~Zu@8=$^hqut)s~7!+ zC`8yU#{MmP{qCJn7mn)$cQgqlja8~;9+L<(m*ppks_x28Snm3vU_A^zS;!i(1_Wf!c;JW0*Zs=%~TwGMb!pXMu(b;e0f{WIWH-dFf%I_ppmm}+xg z7IX)ni|topDV`JFzsUR*f47Wgx+V}qTOe`EsP#bN@I&(91%_?l2EJ~al69GJJpn2s zVQR{I6n$3{OHhi-i4FXh7fqNjTlJpLjJxUeG28X-?Lg-A)#Pi?6{`QmEKh7NeRDz#SQPGW@LON`;5m1_5}Ho2U;>h_gIicsb-r3P?`4EtVHVW_o6^YKpN8uBjl3C( zT}I{Num`-tf}6(*cbC*g{OeY*2>>8ISTGr%@$B|Ar?B{W9lnk^w|+#{y0k^3BlmJIf@)d)mhi* z%>1Sv?nw~IWqhgaf}k-q6RwFWt=SEsqTrU#MEO2Qjbcd+Zw@0HToe-V^aBYYr{eCr z&Yj0M5QW%#j+zfTMr{y+Py$F}nK;J_&$dNk789gU6b)rQ3sxzH}L_Vw8)D5>L~LZax3ueBNQE2?`-s6CPU{R9aj z7?e@@cCAVziBik@SRVLJalxY#MH{{n8G-I}qLOzx|3Gm^UWZ!VX#&JOg^{mEt)#^Z zD#A&ZdxDiRL$<}B^x2eF)HG*Wuqv`iT*JN4oYmbG4J>uAdKTIOU^0d*ZzBtEig>(e zeZ|1xjVG#!I>rQjcsUqjg^*PBRGN1x%=PrDavSMPqUVAVnZqouBML@ick)%>0z?wq z%tp8-ahxPVVUd0(!va3lkbH&0c6HldIM@Y;O{|`bIimPdzEuXes(h0*8P{rq^R#TC zR$FeD7mcG`FCI-JQ})iyvieX|gik(Fd#~sTT>hen)fwh#&+*!4U4Eim{so12w(FZl z)J^=02uEK9?+Lm7u`Tcc8k9}54US8Z9cwBDmZ0Ink`EO1PX&-b;-XA`PDPhpD;8gi zbg)AGmz^@UJnEqcF4qs4=MStyc##ZQmOi^_VwGK!wlOWr>zajhm?Wja0L0@PI8n8s z%Syd*l?#`etXxj?rNw~KbTiluQ6fX{TOV|cxi`e#D=-T7j({t4Jzcm z6tO{42-zO=FC3uh)?d9`Um<4m96a)o0)&@RQ$DLPv3K|Pi(z%#<@39P%eyJRgQG3Z zuhCU6#o}|88GJc47kktSdb34Ml{fkTw$!k=6?sw;6RUKeN${`$MS=)5O&$?^Wlf(4 z4=(Ij4*`ZHQ*;hDi1)ywF9*CP_b0dQi$(~@SwlvfMtuMw>v}aJ0s;ZCK48%zOaMR2 zU7_ntKQnB=B;j|sqs1k}$4c|W{FMz> z7^y5&kGd*5=&q?muMT9&i}HR*j(e;ELzVn@`(`B5>-rsZ)IUmOOXhyPPJvhFeO?0b zdNL~)S2jS3h~vZ6r)L2^>mc&X_{1Jod}~TSXJ*IdZ=5tSUvk9=&T(Ku`y25-Z<<%= z%$C)CVU5a`1J4~YeeqzRSX(RtJ>NWTZsqVL{pUO^U_O#Df6J&w3F(ik> zK>B_T+YdpzUlr}KhmZc+ABPYlAdRrQKvhp{6yruV+iV@x7rvZ88Ms5zyQQhnm~_r^ z7}kADa$cx$S2E{n>vjH~i_vZ~RuGj>5Y-S}tWiB}kX>7i-&eqkM=vc|8qd{Vp#OX3 zh`$Q?X(ARFsE-Z=h~&RmPC2 zC0KO*90}df)g73seiq!_^=}Y;>eRg+CczvVRbkAQSaC9mO+w?MzAO0T>{rBs1JzHQN^j{sthj~Eo`s+R zlCEnT*Jb&*cNW>_7H$ljSu1VF?RUt>w-*6~TECe*&YUD{*{x8_S~h|pU^W735+w!)@A^HE78G&=k>)_A?p{N0|K zPX%tAV71)9vy6u0i)IAQ7@Ei2V;j;E;8Xt3u`!G6QFx|mERY)PUcd;O5Z0#CfbdvCz>zC!6lZL;5F+x zl7avMGHFG|uz3*q9V`$j!by~e2PX>B#}deIvJ&hjLbu^U#YQ5i$y$+-l)jhw$9pH@ z2v(x#=nz!@p_CNi5!jJR8}NA6{Cy=(77m6mspPX;q|vS^)F1tMBv7C5aNYgAC|U(JN~sT{JxwC_$>Yw{vvFHc)*8BEwg~ zs;R)B+{rxH4tkgJvRr|X^q+Z7+;d1Aolz1raR%v;3`0`=DztKli|nY!tj?7Bj*LbfUd!x zjvd*MVFmahfzP`^dcb#Gj7cujZ!{6Tau5)5Nt*dIuoERvpg-uA5N!d`o)b#?(6l7U zl7BX9yfAY1%!eA){Y5)%7+(*$r1%WoXcQhfGG;TcCmqo1B^=o zY~VP5PatLPL#$`SntHHKc!~s?`brj`IZgNpu6 zy46M-ho>M}IYV6shqfF|cK3lofD!zW6LOWoJ;zdx`cK=?dz5yVZuo-YQJE$02R{Sm zk8H+l_riwz;e!5sajrauaVlb``apv&g20&>Q8+l7lmZ(0VzGC4*IG>zQpH4?annjx zM)GJ^6{`3u_pC@Luhz@w(xgP4-7*NggBks7aKkO;{MZd;cce2t#h;?nKBLnP?w_dS zDBmpHG367VC?5V5C+ym}YaC_mEGuq!T3lS&P*w4sxU9c-F8$tH(;}~Oe!52S&2t!1 zP#Gd++yV4a&wGf*GpWrehqUEbMNPv@o{$Be!Sd0&i!8WUktA$>II`4Tp~Y7#p$~lc zFagE}Hh(^jTYob$b?-8h#3EEc=fj69Muf`ann3#@7HXp15gsOk>=&wRBs+B))>Q0) z1gQGg07Kq>gG$L$;7sD(D5gG<*ScDK-+i_sGXjtRyNmJfVpR5;NWk=d1p}n%f&e2hQ&1|kQDpDCNrEri<}Smha6+L12CQ6K!y#mA zg{xF`L7S&S*A>3=#|@21Nw~SAO1Sx15BxN zx?#N#ZGcyJ_7v(Z>~uFp`mEYOwits1<}E41o|Q9Dl4%Og3># zSPCsl$jhwn9{I*Fcqg=2yGDsOJeETi(q1TR6^`mSa8QiO#EeE%;69?-jIgLB@_?Cd zo1G1`GQoK;ZZc-SP!NxFM;=+SVP;ptsI1aTrB)(Io!fwgkl~nnFuYJJI zI`oRAf9=Pama8`KohndVui;3g2B1}{cQ%T`d~#*3r6){*jr*F_Xi8nzD0*=fs1NS; zFHt@<<0(pM9bP89)7gZ58{|&j0zDip1aQl*qhfOJJm0HusGcg z2nCqpyuo>PH0g9Gw^RyhluWDBfw&ZJ7xqT%erCarABg*!k<9Tqf{6Gv`TO5G;{tMJ z#rp-TI)#!oj&iRZP?l+`k|es4E(_8f+OZ=-%C1V-m>06OQx$#g39d67or@{|rHA5g z!xAcKX?7PFB!rC!Vcv)kSpZsW2TN8N_dxayZ(QEy6%cjZ?#PAcJhTGHjYdd7VtimH zI4nQYY^7uE)fwuZf>GH##teZYv!1b3_-{djaVTbE7f+~R)NFRwl$-=KC)V4G1$&gB?~p8{ZXLWdR;&-Ni=y6ivThu4lHJRcoQKH;F=DW z+i#qT0X{O21)i7h_}XYeiYj0VaHwmu;`6#I~Ciuor-PSsW`c@ZQHhOR%|<|*sP?h`+Ml_UvS1a zFK6$w=9-HOr~!ja(%4C6RD%ma+MH2IA#P*%tK$t2Z3n5#sJXvR;8`O%`m2xR2CXxq zSa7_5C_iNv%h8Sc!-%6tC)(2SjO_apmf|%$#_b>Li!7wm36Cm=I^XHPX-GkN?Zam& z&T6vWBmvHVw#V$QxATKqp6t50zBQX&Bhvjuf|+Fzn8sXb&so`UY&R6xm-QAQ()`<38KTgqb6`^g- zUhgKphs)gkwdkT{lj2ATk6$DCwVc9>@=q$fz9W-BPl9k^Gj8!EKcEr5geLZO&G!@` zpeEx6DBZ%_P~+xad3RJ!2Y3Y>Mzh3IcfGSEG-@5`*X_|dUK6Z=7O78ihr-ryMe(92 zdKD`K(7bX$YL*|-A$HL-Gl@&x7*fE*xMVL1E(C3i*lP8rG?~Is-1! zaR@ELe#rRXMiwfBb%O_#)vcf#tC+_QS@D8U| zel9$%;h0U_TjnC9{F$`W!-5)Skw_-uu;)AsG$HufY)i0r7v6`wGu5nGGMrq z@gn?C&#(j9qi^((btmjDM)*B0*kN4(HH zU8%fO_^=9+<+y5c|Ksj|t1U?Ls0wQrG1R(at%9UO@7Qi|%<8@>@1uN%`o~s482GiW zz)}bVjeVBr{0e?qLR)vVqTg~Fr2<}gG4xlvastOCBcV+M=s-eJrK zUKZ8F{|hq9f1b{*(&UUeqn3e7@0t$9v(OfQmEyddc*Hg`ri6O5{6tkZTT@A>u)B+| zRs7a3>K7S6O`3Q>;ss-a+Y`afGmiCyS3z>)-(P9kFG#fx{N$#f5^HDR^u_V{p2Ub8 zQU3|MXkMh;k6-l0j*<3rexjfAAY*~p&0!RhYzYhgvj4)?&fVW%IJ|GL+YUe=!IK5+?9o^ z$CzKf9g4%}Z1eNw&!j(qK`>k^;lI@ra0eQ+*oVkbbz~Nq=;ffX*8jdg;eE2l1q8A{~?8-#A}5@L^saVyj{p^7uS) zvPZ60pq?zsGWmcHS>Q%|bu?It@$8w8Z(x=#D==l0IW^Y#5*l~|2f=SoTxaAM(|6!f zUNZyt?_6i>@mH_^ny!4YpS0XPRQ5)-N$r_cv~F9uyRGf)=s&#e5A+r+yVdnrqNyc1Sz9Xu?G6g~I6h7Vzp`pnYE_O5kmo*HMQqxFoxQN&^r;&ZAq9KXpMG~- zXlrd3q2C1OxOHLJTe8-`S8SbPju@BGp`NQ;xz14zpV6l6c1|`eW{J%sJms|yxao>S z4EnGsFvGX>HL~Y31S!k>ndfjIgW9a78V5vyKo%_*uCV?AUQe3MEEtlpNs&;z$w%rl zAFHUvQJ3Ur!e-r8p3!BgdtFi))-7`Lp?k*}8c_9&r!ob*Zn+lMxz9P@6@;S`@#x9! zWFj=`Wmx#)=&Q{{!_GfQjFsnG(8sc65Y7bgA8RT8Z8YJ~G~B}fn;G}HS$!}j=r%C+ zCeikwWHW>ZtR-s41J`hl*FaIL!O`!W3|oLzpZcVl!>hq-g04f#nRE|%e*>^*jR~UN zqRv&*BGR&rWMqykxzH}Ip7DfXRLvq#kQwu4RCsv0YtQ>dh`rZc(w=O!pc2 zwBg~q@zjuGvqn(~#vjVIJmDhx87h-&?DqNi`n{igURe>fE(r2>aq)I=0biag@#)JZ z1rG^~)g|!64=Y1_P%3oHW2WD$pZdzSpVXDoL1i(?w>wdip{cBSPWV{sUt_kyd8CW=~HZ46upeLWbuBF=&R%L=^cI@u(+EcM0a{LRo8Pl9^Zz@Zeg3=HhI6 z&Hj_a>EQ)G8X^@T(~CL1xdrvQ!aqRY1;vl~3rJ06o|GgU1D5UDJ4 zoHS4~9u&woEghLE9@Enm6%{6u=R8b~g{4OVy?bhRdH(hFfN5EgRjjOta2L3r{>?SR zR?QI?1|9ku2(orGHuM?|)1J44-g+W_S2TO@@w;7whvr*t?hoUcVswAwX+sInJoXaQ zvCYyo(p^?{(kzM5St`Z7O^50>dAN>(%=WjJL-SPV($enk?wHB5W1VAJJXDpmWV|tf zWdouHfaATupncbll&F!F6ZL~PD`~~Z-xzoee2bwQH80&5u6abZJ`(EOstTHKey{JR z;3zkhTTg^xAv}~PG%Z8RlNj9H!&o!fl2MeY0CoUn0Uvo<8mFIaaX*k`5{>pRA_4b5kry+uAE zR}Q(@1zIlkaRJS+m!^p6oFkF$Rb)Fc^J)hz1sNjlrV%Lb8vQAt;!%aV46_074Y@;X zQcw}qD113k)A|vx+KJH|=v{&#((s*l zVgxPfnlf3*}QJkau~*P!bZ|#F=80P z@xG7N4VhFYH)Rky+e2Umu7s+)oyZ1XgEAx^brN`#RUHKe@GL~fxp6-kVO?esyU4oV zZ|9!C23O*68}*TEzX(K^k&sjv0V~!;5WLn)mtog>82(fXVc@l^Fxub1DzKIAZF>S& zNIWl&DwpD>^q${YvzX-^B8DocYEi8&0Tgl?Jo+5)jk|))v6K0z+Cn8lbqkJlD1*W- z{}#f1Ue7$IFmEo2K8Ze_dKw z>fBGZcT?0ha7Ift><^={1T!+Fa*xk(I7 zX`0-3-T#57G;_;=+bIrU)M>9=Wo3uSIKH|t@sj{9mEGi%ErSVj8pNKSr5lZO%cFglDazx8MyUJnBh>-k_6fTmtlD8n zb2`BC>Sg1A(7}%}y{Rpbd&HaHdZ_gklG+F42wN#!|9uDy=yk5_vyhC)KcCQ0a-<=*B6*{7bzcj*adR)lp6;1X?O#)$jr`8 zH8o5&HZ+vNoa_PFBTd>T0E4jj_8%u!aF0V?3Q9>!S@-0|l=kJz(fb>yhk^pwUd+wT zi?SmJzZ6P4TKBYv^3jS%Ke|*))7tQq_@96wioB4)4hsOl)CK^U{}W6A2fD#X{!b*) z_JZOS1PiQ{OJK(#&VyR1LE1Wth1xhGDa8$%0F~Hp=a|1Po%I+5kqn0j=MW~g3kaYz zS>UpfuhurIKRVk_jB#d)lEU-@=!n`T4$&2u95NODTt$VG9l^FofLPP3NMz71=Th23 z7Q=~d>7{4onNhY-vX)51l!tizB~>nph8{qt3j#EmgIx5YCbJe2ce5vyd&{Qy`>ecNbT8?NfrGvI&|yDt zI}Vh3uAQi3q!tghC&7tsWWq;2F|e_G=b~R!dw~ld{lY)mH5dfk)Xz(B=YoS;%BCrS z1{&Ui+Kc*RbCta5VJ{nULx_QDq^9@>w-m{ND@VMp&JAml)ZWx&G`@*N?yb)$V5<7DN8@(lv0mPwfa4F1ES1L^bDemXmc^Idp zL%_Rm;Q=9y9w{S{_htWm)oG;dY$ zlz8qo#J2FJq36V=HAZ|!`)u=DfU~>aubmnUBvo`g8WW`%G4~&s3nDroO!tc+{ST9B z)s@6sE49?>0uZa%j?Gymo$^dppm8PxNj$aNx?y^E@;b6cR?4(kt!pq$5;xLxKhOod zeBBy>e$5>j7vslKrTr(3GY3E}YG0&D*IW%sX$fh7ocL_x$;ybcnw?J!`!%C>ui}k} z4|+IVV%~r*;$#ohFh$nLQD9aT#o8J)jLs$7gvP)mTq5C+nP1BkiV5c9cJLP??WG4(U!(aZ@l1t2^-PxvUo#-svOBecGRo4{ zZb0OzeWBQ#;;86IQa;Hp@$%XFc-KSYHsGzl85In8QpBHs-z{%OoYM2?50{4QOvgxp zE}J%5rUtzk2F@Lj-Uo=wMc8{ruAbpEe2za3-Qw&nQ5bDYH3j3B%xH4QyRU6Xa-c9z zO8$iLWbFr#_5VuLI#25-_QJ$gB#0VqvQm-NnbN7dmeK{=PYH@xXlGnbyC(L+2KK5{ z){H~{-9?4njUHeVnnrvKiEFbx(_jdw-9g&O^_~_Bya=#mHJCP85N9VT$C+69tFzjEF^YYQiI>Nipp|!AC{{Ek8apCxhPCnF z4+9K9iJTryg$B!+k(UFl_mb3!1=^a4MRNfd>F!#z~s;LVmp&7T`KgDyjYaBqr=!$?!fN{lH}iaV%8zcr!}oemH;fPAj(dgRoum`O# z-8qpy?F=bPTlNBu3VXiGfQ7p=fMO6YHPeHvQ3wAAGb;2vc#CHy#d|-Gr$S9XtuQv~ zY$X2}CwN<;xu1eUdjk?RO;BJ6w_>BJtDx$-H$>?vojp-;1RI2K%;XOne(kLQr!l4QBvbT<4ahK>2TPkBfaQAw z*oxqb$o8 z4uqGyoE@=vD{t9qo@a_fu~L}jw+B!*g}6a2U&r;UNbM6a0@qWMw`_f9THGZ6(tb(03~?E>tnb!@`n>4yzN@IZvN`}Mt*1fJ|cG0LZM!I|GpQbxY+eAsBX&@ z`!WgX`IY{i_RO2)mbsljoV5Gn`gfXZR^~>05bg^77pPDS!s-;U*1k5!wOn4VPk?~$ z{P}0;*QuG-2o=*pg_g#MD8Y4fE`~mbsvF#AAQUyn5s=+7bE-OZOQcEUa3du~adOOR znImKA(GI8JYu08>yeYEbv*GAa%BrI3DYQ+^HY)2yv*KA74AUr zsuJvDC@?r~!OY@sI=#ctk&}t4)T!z(j*=!FR$xN9>~J{La?qxVZop^u-5f#dkQvUT z0iFW-!Um%fl17d2*i@QAye9e!hwJwGXocJ+$`}g|{f_Zln%k(9E^R5=Jz*CYHDaoF z?n?<@jr|X+JsUOJ@U|*dlVkIJwX~JyCDJ8mMvipjj{I{HwgU**W^P6e{;IWh+SP3| zF<|cV?o`Ws-Fk=#mel*R1CmAPRvj9g1TR(2#&_cn+ru3NExs}SsyEY3EvL-0Ax%$wfX=SVy022Hzm+ zvpHs95eYO*YK$iJXTcp^0wcNbt!)U}CO}CHE|(k-YfSwHW!*4;8${QW@f{gaJb|8P zS%D8l#T^#{Z3EGIWIo5C)Qa)m{vfwc#dXi%opKrKWElW8V`v?D#2I=wb#~Du! z;Wn0#KbgE1FXveZITw!T~|+HV!fJT{H%~8@*%Zx1KAd$vl@$~1a#NWNk9Rr ziTXHzYz-tB@3Z0F$OA&Rp2Q)d^kHeFj{kHFd5rq={k`+}Q?Z94|M0?8wfSj~dgl)4 zHnaLjk+*9Sl-(ht{J9JONo3UK94GY$d?s@`r~oF}{PW;X@822r%CCuRm7$IB(R^br zuuK=Z9qqp+H^*cd=Vji_?k2gqs_m5Z{rHnXgu$v(M}M>&y>c zF8+_ZtGlYcTnz_d+I0nfh+MYbjjb(PFQ2`7^Y6a~wig2{Zb*Jt$p?#v<6)*)bryr_ z*tK&I*a0uq&KQ%Mp9_Uahe1W0H?h6)US;u0PxRm}9xH@`6$mfam$zX%5x__BV(}ZU zU4_o6XM>!a$JQPwTcR_y=>q^Gm<%xrQr;6n3@pX2iKRHrrT-WKsg%dgf`Ac$H~%mo z&Ww19A3DP#HS78}wU7O^Nh71K$%A9p;StRZvk-4qTtKE^O8e2zjB;=N{sgnLYLjKW z9os`NTB&wV%$IRc-{~ntOUu3_UKo{@#ON^h56&KP=&mqP=q96T^!x(}vIIJ|h*Q_^ zFv4KN*(B7;%~s(6Yl}_q1AeyVc1f-w!03AarsVMr7&2RdBD)DFgLOKK;?ByR3=fFXiapAA?UC* zYog{}j-i;}4qqdakGFm*R!;e+i1OTYSXY=xyX*jvjYBRsQsQdZktd3yrUeDL2tvIc zD;P5>OB+AO4v*voze4!E&J=GQ!R}nwm^$HDT(~+2?W$KMH_(y$ydP`$nKf1d!;=X> za$iQ8=qlA|=tnALWO68F&?o!zB&M50l4Z%Gb*YtTRC-&$qIymkSM^ePWDAyQ6v&z} zTyW0ChHS)5j+iXBX|gzz)GGWm?rv)$N|<=`$Ue=;L{tQ4l#5K%!fCb6s#ylbXhsBM zG#S+jhAFCsW$c50%BGAKP6ulN%`c=f&0_#EHW9`(QzhX~DKz;hKOZXDBnu|O4sLI) zS|^QqUFRY5LWR;*>GpWx^-$xbOQz@d3p{ekQmEm?W@w_+NSxLB6E#?3lGE+Pl1pxv zlXbwREGw2VqOAk6XN^+oUACniOZ?NDmmXrq@0IIJNup_-#8|EC_A@wu^(I-jB*rm{ zVkJT}fgsSIoQPpdgeVYQusS`$YgYn?j_i?(xOw*J0ZeMm=nFz0moj5(`0D(1@QGtG zsUflGivx6ETmS-|5ed{bl(LZo9$HW7_Gk|Uxm1#BArdEuTUjrI@)FR*4rdRIa0Uu# zk*>Zh39@(Pi1Ul4GLlMQBrSW%L>VOt--IQrI2o89iCXdFNV(X6oR~buW%=*Of^ALw zBy;=_s&k(Tgaj&h9TG+e;i5wA}8>sXvTnISMJ;a&{v^ME(%A8$q+U5{Ko`h zr-k_9^;Ab4{Kzs|;#U5%!%W0jGFt@C8$-lS{O+2J^$l)(Z60?Ib zl+HA9|ASBgm&~BALj*T4a#d}T??_n0h9iZ+igrhhr&Er@xIkI_prXf`dy0m_5j-|L zmtuVg$1l*J*yIbj#nlyjSAq)K=y_jxjCe+|yi61|`S=dF*<;orvvd7-quLQjv`)Z_ z?em=G=OmK9dhvC>(0YcGAcH)lyR*E(j|LLWmZ>MX(@<4SRgzNp=^s7 zl6>9VKcfrKwHjv@S7o=83&D#fyA?PRF|a?b8)1FJ-KKPp+hShc#A z#q%CCfo;!5vMfNSM5A+DSr_WX7kiXlYK68&WTBYw|i7&Pik3T{yoBbvRwxi#y(3T@mu|4@5(HO>gEKK|?Er3oP0#FUX=wDD%W(C4Dh z7Yhon>6b7bF~Fg0EE7-pd3(GR)6&yH23ji|nM%{9;)jxZ3pIyXofF}vrPH?Ow(Wfj z(W;%S$2}r}JBzDLCUCLdHZS(rM`c>wDRMhi(>h~79Fi4Q_ie|8E?6dd#_Ia+9xjok z@8JuaYSx}Xul#-3As2lw7~tDA9eLPTu3dL3^syN;VCe+}uo?EX>4Z(k zj4yEA5Byp^br@b;#9|yqn%!SbQn`D7T#iaAE1Ml&cR#?l?YS}F=j9e8h_J9Y=xTDg z8@f0>Ejt)%x-p1)Sp4R}n(ccS^}o9jdhrF)4|oWE1!F8x#`O_SjTOv*{92v2E>K5U z4N9P{sg$IpnL_2LAGnH-+7S5{xdmxKNC1@+gac;{%d_YfhzqMPobI#b^^3J{kx2{* z{@4XVUdQE2(=HYnE4oj@4mj72w+fdqw2V_N_b zTE>VX9&8H?gN^+Uv019;b(n0G3LcjBg71p!U;5qc=WV3eI_!fpMd&MoM>)YOA)l(J zU=8xoAmxy!NLnux&l9~k`V$!oCtCs`3X=hK#ygtYhy+?FqsIKg!96{ z7=$ovIzF11Mp%Bff=JMK#)C-slN658#-Axvj|ZB~4q1?3B;Xch9nat+Y^%VI51yy9 zF2$j17RiIGX_@a5nR}4t3@}AN1S=Q{F7JnIZt>lr|N9J8SzJ-IX0#yQMJL)PPSphp z?lKgw^u++f!bW3+Lq9#L@;glXnRJ0^H;5qaO01METu?N#6H2|3&IKtV$T3P>G2vJ_ zEz9lT*waP9uY8Y+AddN^Zhzp8`XTti@V$)no)ThQL+8j9LxOO0WVO&hy^3hnWzvFM z!_86G1s#)B7u-b#Hw;uT2o3mm&Y{TqIC)dtrZ?G1@!3SxsT{meB~5pL;ccVUE}qz<2Ta`|OX|FwP|Bx#VnF|xXYGm(Mo2B|fO6fYJZ=4<|Ki`Ddj zXjdK+HQn<1c6%$84S0k&(^-p2KjVQsl%l-#$JZLh#-gv06#}kNrsBl2=^rPG-;^AT z>!l_h-W){YUU@8W!Gv%~yBrr=L4w-Wn_+?tVrmbNLa@@wBH50+&qYp%go#MrHo{;J ziWX{gx*}EoVV;KHGYO9|wv)inz>ig{sDO2LQVwEJ#ytI0N# z^^Bu{rjN&KZG>yvU6|uk6AQXf-$LVVSU_p;n9$>P2RZ+g*H2U^ zELRB-9Fs5?z*x=S_5e3a1U#sAIZg&$0jL)eouMx%(PeCHEOame}6h51B&p@vj!zbRWU8TJg7&D}SAr6B}g z%?pST-)~35KPV!5;hZve{#bu9FW0i-ud10?A@_Nyn;PH-GhnXohONgUvNrd3>!oLf z5rEAd^XVg|!hNmY$Ean|8C)wX3TY#O?K3i{Q_7YuT(;=^>&X(?#x{y~Pw9kQw!a-+ zB+y5N=FKyxZ^K;G;T1X7x8ek$OT=oYa-Xu*N|MOw_rq_oDu)f*WY%i;9?c`3>k2v& z;hCs@zd@MKD1|OH8?*kK5#cztH3>HPadi_gez^Tv9L;%X& zLgEp;Qs5BxYj!JkrtAl9L!vb~!apboFX2ov6OpV*_7inj14D|*P&(u>spB(j;bmyZ z=cs8e6myq0cX?yZuEVp#DNfzT3@@V*chc42Lt&fL!>2;EV%=)NityYJ{%DJ%Q|H7x zAQ-5u?e9F9(Oo$P2!?2US}-ZlVmWIhWo;XaonO z)P#Rwv?kg9B7Gk6=xK@V(UhGd>p|LK^%7uJq=7tqz}l-$+Rl1=t9pql?nf4li75Pu zPL6-$BSef$Ud0VffbBp5l8C%h$@lAyAt`BT9BFH{! zO5iG~cfKm!#16NLk$4*!V}j7qFL&hrjCQsa-aj6lht%#U#clqZ)|vj;q5yPH$>4u`f;mW^(vYqFGw1{qhbpc zRO9z0ST?|qNNZu5Qp9v5p99Tu>N7$CKH=unGs4d_x2Ts_GEU(kC;WUqp3j$$AzUB( zU5l&!f7{3Jf3WB|G7`lMt8^J`_muTSnmDBGL(FU#-v8$IOOxBc;>0iu=E*TSf+a1{ zpy%wktApz2rNX<^pA>K>@5{L$O6^ns^el1&D@UzoXz-Th~3*WlC#iRPt?OYLUH zs}yqUrNhBX?i_~}V=0hgE0bM#v^0FJRK4gY*&BCHLjQE@PQfJq`AHB{zp^M9CoDWC zN|pW2)bgNXa22S#^yuZ0BDsn>I&?+Nr>p(D`c8dV=Il;drcLGF72HY(8LecR>+-%b ztzIWM@?Z*{JhR4doks-Kn9P{wuE*n0TKyHx!jOA^lC6~#kbmu6o6Vd9?cA4@5lBjvO z%vZEGt1uYZsOnJ4ONc?Z(nZZ!_d&I=K~SxpZncJ|dU8xf?7&f#T6sy#{u9Xf(Qb`^2z8;z7OoOAb0`C?bF^*h!aF3YJ06i>_1m(L6^yvP4Y>N z_G%_onrSPpuz$!@S%1!(yn&}zs%GXnaH;ik$8c*59qv-8(Mbd3>~W+!KyVNYm~ zH&_kx_&}YqHqRBja+n-$^K44+)js80!LVfMt@orS7*Ybq#}P2m+o`B?IZzfk)2tCr zqBB~5x5??X6!B{asfGT+X7^Df1|u|;7g2*7)+z{HJa;A4s5x-Zk&##RSmK^0{Zq#y zV2$k%+%K02TAiRHWm7o!+5XSWQ%f4A3@fv}O^01zJv$c!Rhg?jg`k$9c!yeowyO1g zN1C;1h1Urrz0;_qEl6BiA>lu0m`#{cF#d;5YW9I{H?g0I)VgeXaLd`?*%1UmC+^N^ z$M$1{h1D)#RW;w*BHaz;$7t_lJW7iKa6VH zmBFAir*t1ruvm52!gI~tVQ0@EW-<67vsBf&bButU9zmjE{OR@yYIQo>%)*Al&s9eJ zg;5?Gy5%EFFz^0SMprK`yw0~+w5UI&Dn<8~VmbD3JRrPga==XpDk@p^9Q0#T~^wZ3#7rPVI@$ z`CG&ONrmt4#4|tFs)RzFJtS~A_f>^*&-R`_*QZF&(S85&QZ(fJc-W@odmkzzj;!Bl zuhq%F*S-j=Z120wF9T#oNs+72F_|SCL#P8GB8EDPB`vYim7KWjHRy93;{}26qCw zlx06!hSgtlpQCfsch+!y{4DEhS?A&FXki5Vkpz!Uf)X`MKT%IodB?m}|594AWVos@ z7qW8vn9p!e@8sTL+%{`1X=6KDrZ@kkn0dco*WY2imcg9c>x8VBLQp7?rpu;5xxx5H z&RcZi(<<5FET2u?o767u0(GDHwY?UolvLivK96d*WK^x;p`AI2XxqBFBGbB#)WcPY zCfo&%#$o%%OuFTLZ5Sc5PuTPfyD_cea@oNy&A+?pC+zxlYSS2nZJe~I{>u^Md(7!U z`?96jJf=^(3q`j$65%x-q@DfLY(noUS zr2!Kuu=5%N23q(9eXcy?GzXGn>i7>)~@LWILklgZ5F6}+$dIe)o zbC7EExa^tzE>z00;%?j+zo+`>VBg*lA$ol+xt8X-76t?{7sUkMZyGEwCh_6sxdK*= zVgeYssmSJm!ya>~EUg0R{M6JT?$35drd#QeA*nawbMGyH@LUl5kg~PmhNHfq#L1 z*yd{p)>fv1JKB?;3=GtXP_sWiG@QOh0%yBW^bRvZw?mJTAYL%!>j1NHkBzHx9rr0D zp$@0joM2GjaSp=hCpreOyDJ?P1p&T{dum1f^*aA+8{GwHTW4#WV7enoy)MwAUCFK_ z+S8^coXY}py2HiykbB#~^$Uxy)E_DEI#$c(^sX&;iZT#Id=Jh0T}tjPKf?^9a{!J| zvFmY5lKI%W)yicPE9YgcvnGad`o1HjkYg`M->T^1FxL8liBAG3gRj^c(IaUj{DOQQ z{nKL+0zHZ>3&dyp+0{V8MwekL?^-?3eL~(1XY3%2!>o7P`=%K6_aCgy1krn>5Y*z4&Gh=01kS_ee$w7G>z&^|IJE2fv%}Z1_*h zL2Lg`mfB?3Bi72V<&RpXG5cMxaDJHS*G^X`&T&RC%0niFu3&xrw?rmvFr6UIfn3IV zvM&Uj+wx(ZY69~m@Xz0$sasM7FPhh_XC3@iIykOiRZu8E=Q7=o5yaDu4%tdZ7*S4H z53K}s`T&j^6CtmLY8%);UCNnLN^s|oKGkPCFtDUbzx+EPMN9A_D1Az^4$o03*RHm&dsS(d9=@b}4oN(({XqLtffwO5Ix9kSENSkC9nf*| zC>6i-zTy{v-or%W(7o>k4|kypA5gWUX|>p$zwMF>_0L#qqr|`ml(&z`3leU90(Xfd zMx_D}sM~3%ctNWGf;N(zVc<}qc|`oD^Hkn}yOXe0hCY+mFO)e~>mjyYhYOY;o2xD3 z!)uw%?)lW)+}?68b5>1zW}wNpCEG*r%{a~1IgMCAkS^LR$ZPGO6o!wB_49j^=`|6_ z+Oy?!Vv00xG6sD%_sM99BBUx_m^rm9+K5)ShLhjwa|2+`>5q$2gR4T>u&1Uz>vcQf_Yqow_`uIxmR`Q?#8rqk7rqRq7Ry)zvmzOpL==I6q8`o^)JL zQ1z+Bsja+;2#yOS{;q=cjLH!GDyepu$F~8xtNW>9r1L9qe6LQoJhH(L?Rf*moAHts zYLQaH+U4*9Wxyas7v27BFvwwetJ0!f_9GoQ+)mZ!6eE?#3uO+MC_gyjk2!6Pemy25 z^B2Ll(o+1}V-ck$d-&%L4d~WgTuifi4ZOup%fgajSJ6+m4>YimE{+$lbfGmAK{WtU ztmwuH+v{{JV;Ro4GYh+NejY5fTVUEIu89Pus2i1!B8swZ68)&vjg26PG_P+jNTkyXMZv7-_s2N3ideoAo*$;qX1!P11PY z&I|p%Iv}6-WFNNpv3LKcAQq_IkoN|=Qru9vZO@7E!nS&On8Qslc6@tFd+k*JQ*K4v zP{LL`7f-}sX66AGdczk)gA3XVi?b-I&*}!&M_5)V-h+t7=6TAo$Ot$HWkmv9>DUua zfOHw{OdA4mpFz9`?4HzmM22+CDxSaJN@StIS~s*5x1KVZs}sL{UV^hJSfyC86q&fR zJhr57B>2IR4`~UlXxvDSfdYiZ5<{<74cB;BD?ktQZS(fY){XOQb)QiwA;#@T&607H zR4e$H5@}O&EbAP02qI*( zPC}1hFhmr}jTD<6B|*Va{B|eyg!GX-KxMwDCz%%^6Vncx_zW;F$oy&KYN=>B&@9F6 zTR`Sl%#v!MO9H&%xx~~Gd342w8YN~uhlminbp#bXx~1$Q*QdqYpoS5}^SODswl7X$5a7Jv*~4i`UAAEwC}T=XXpOBiMZtve+sC4VkzsJB9FLXj?yB*Vf{ znvuxJ1TDx=4KbI9sqQXw=5rzfPv}?RnLVm!dhj&pKLR?xg8#;TL~k!~LY)FliOAxD z6Sis&rWj{S)FN{bFEt`Io>M%M(#-r}B0d6jU_-W`Ib)R?GZ{4`osEs;f}KK6gC$yG zzh{Jvg(5HO4^FzT$}N&Q7Ag5ifvTcb!MYGnZ5VoLRz#cw)9Sk=(OJ0#= zfoA<$1Pyetdjv$obsC3dnz3t&)(0ecLje;Cz8QLr9FTS(3P9D#J>5x)pqq_3P^8n) z>392f(vKEjy$(+!JF>Da=l)40yq^lEY7s*#$EzxOH#er$UdwlPYuo-wC3?|G9|_Mu zLXGLrR@+P7@R&tWQcS)_T6XR1zw?X{K_OEVT!2@Z0{Ygl2#iAWia(;3*lcia$UtaC z!OJZZvkW>0D@;I1fBaH}tO9<6Zq=$WL7xGe?5bBNWdyx4tH-wBGX_$(4&;hHfH z5nVt;{pZ!Tbp$^UEV;%hSyKzqGD=D6-szi?ymBlK)C-&E;??~+l?cmzLtp9Ed}Rs5 z@B{aPtcpA{1r##hbra{c9nRLP&+nTnrY2xbIQ}(}ieLGe1wJgdo%{x#3{pvGenWL( zu})*ni0v`$L6G7U;xOAn(ZjprAATb5eiVRY>XI_rs& z7#`0%*Onk0*b`0a9CFbY3&#Dc|DwdT+!}yDr&JI5X_rSaoRHB#B#>)v3ziL_Vc}d5)~xrZc#PlMsDi<;yh5|6Wp29~7nOPuvzv zUBz*ei&Wf#86;!|6bFt`6!`e=ao}>Yzvax+0VNH*b$z#ZA@Vg|C?Ax(2(pl(>IgFt z&0rgi6-%qXUS)$nN+V({SP_VEB;7o^*>uYdA}q+9x*!`xsPo1 z#4z@q(YZ6(uwe!ku&Caw$}FHm(QQQxDo zNy3w3oh02#yo4MTf*tUj0dF}7QiRx%`oqx#b_?)tsW#oiQDBx5x{;P1bWfKIMaV2$ zPym3DU5VMwbDLlkHqq^&NlxOA7g)3{A)F=0(f$k< z*m4|kF(~Wv+R0Se>uyn)0qch;wrEiujt67?4gWJ$PY@q|`V0q=-`nf!e@{H&MJes% zOi<-rHk6$c2=rb2feAZnA9E|f`|<<@Hr~ZM@Lw{00PDN{0$T_`0nv2L=YBnR2OU2r zNU@CxLY|vm{a!(@Lc$;|X{-0@6ZSj<92*nMlZseb6ZgO@v!KI0^xYX(`c+(E1|D{h z%H*QrW2QfxPjbp5PxgdkwS859NiWNas(0H&&BbQ#E%&kjTY-u&g*hFMf_ z(TaIm&C@LS<~8wU2Dx1$DMk$&LLd-fqu0&(#_4$3pr#-!8u7(cLevnXTA_>2$cxoa zxck2Dkv!l!*z?())n^KMtmXWawCg4nXIiFa;j@dUBWKw~u_U$X(P)*|{JVMj)0w|0 z!mTQ4fO{~ET8ks~7lZ&p2Qqk)1^^Lts^Dg-@j#mj3TUfK>^3dRMFfp8-C=+T z2u|l*ysvegG#Pt56P6MjC8LvW(Fx*len!5?Vsq9$Q+d&xb9VYcZbdUrAc$oTBiA_Y zsguo_o6>yBb8-%g7)fSIl?WmnPSR+6U0J0K@Xf5+C$moLiUahQ+{0n=R2PB`(;$xe zfNg)$czJ&%Jt_z;s7&smyY6qJBXlD}J2ZLfWz5)K4s%OAN#W=p5?ziEw_!+HQSf{D(IS$qZw=ObPO=j5DRh2>vSOu4K-a$_m&V>MJYF&Gfl5Lwpz{L+<-eKe_q`3t(90W>A;`$$f z6P1Pk#n(9nXTk(rI<{?0Y)@?4wr%r`ZQHhO+qNg>#F=FC{aagGd$ad_-PP6SJkOCe zQ6Zggu7hY>0{Tz>u=;F(UPuc7oUaJqo3RDP<-upD=}4S~5HDe0l_6Vsf0HX+V@(a? zc(cpIl+0Ss4tfUBpT#t|F_XUToh=L(*VsVBPMRDjYr#-SySMg9CkyQnY`(5a$#jT4 zP8`2h@xX&Mz#3wD)ehqW->Jg6--x^rS^afh0bN_CuwNACe6iKI?WzO3OqrrgipsD- zB!P<@v?nhRl+JmM28i{83}-nl!0}9`)6ROKTB|{&;_>S=9t3HUHb#OrH%?SB@jB5W zkleV?Hq}o@IapKn-%6z{WM=A$JcTvvwwa2Vi{9`|1||)2nt&W~kuEm?i{`_!6Jh-w z$n3?S+5gqWidgC+*INu&#?l6bhzGWBvp7IW8|!z?REchqnQ5hR7&uOXy1IuFd4%!A z^QQ0WrK8{JZr)F>#UtosX8`alsU4i!#XN2 zkhi1bTJZ_IoYD>qwgul|S8%s(@tiwR2bug!h&|y+Tomsm;vo*;%F3IadS2x9QnjVgMyWyT&<% zahB#e#R4mYGZ`Afmc0`AzSWRy&znxXj_RY{O7$UkLPL8~r{iv;(RH|PZUbjwchl3r za&%U3J0gCv+m{*u*S`fuy3NIJ`iU=hr~4TFeYFCv`NB)Ha+TZB)!Nx+XZgB&u1)bR zyW0M`z5K4S-d1;AjzC8b`UeZv!PLdUQCbT0I&JNa=fh zb~GizDx`5tu`U2x)~Vj1QzoG81pXa691ltc4R*$4tEUKnWBR1?AV`^UcassqoVtU^ z{T3hTU4t-S_d9m%kF6lryLM*!fpl zC6@VXBtahJe%9cH!W4nCL^=&T=7 z^ke%GrOEc2&7V4Er~1K;x@f9VNQz4W|L{4Hr@Jdoy0$(^ zYydlPtsaG2B0g%7gz=Ys`VYTbzK97krjo++_pj|3B)Kporp9fe)+o0`q+_DSsGuDXg&k6m7RLO1ou{AUX0TnX> zjW=|rE70j?rmGp-Afg<@F#_0l#C~J7MGbBGfSMs{+D!|SXunSktI4x7I=v?T>yCF% z%JE}n$!b2Hqs-z1iHBN~wCiC%1%|bf=RGntNB^L9t7$UgEV=$uLfT;A!ws;tvb6Tp zeu9tA%I|LESHx*+fDm-(S+KNnAUUPk6TX`e`9K=p3~k$bxnf+SC#&5H-H1J~7Y*fh zoBYK^2clt4S@?|OF@^D~2|frH36cPOUpe&B%BD;Q&3Lf@6q$}AYt>hQWxK}+&oHWv zHZ8OI*4<)ieJI{IL2d*55(VI~uWv9Nu|bqbWU#&=$QYx_xY=n;17|dB=wM$_om~WJ zL`%S=YVH%3@@ILx^mfzM7I88pOrBbp0ihk#G6ugP^?G~;w;Mw{gPD^tFrabE1g&4mouOVGPG5`pXCz$ zUy76#ix{)1U8?`onS55O|H_RfZTUn2Exp|M%kTha;uC_f(d7>bN+ zmQ*v{_eRQhCl$~vMzfE!?kv~#-vo}fJ+H+4@>e$xdhp$vo#Oz{(AB0`ia220vaK#qwNPZbm50mT z-z)6$1}P1t>!_yzY}NG7)1$b=5$cT4#xs^(V3F(T2s%&%vlWuOr zbXDw_@2Sghy3pR4xv%q5AaisWY+>dN*Um6I<5?;T)OrTB&xrW`JKjW%jCZ!+F*E$}-5Vmim^ zyMm2S+sRRB?X>b|-D}Pg#I;`SsA1M3z!!U6>9co8 z3H}C^GJvJv`GQia&@5`iFW$ba`?`Rd-*bM8h;z7@3`#$E8{$C>4m7to3PVOxHgnNw z$EUd^?_*=)wWCVFcZ&4dQ$yg)hCFpSsbf0fNHk)$vhm2=%yVbdo@6;ig}pqtshZpKzM)+^WP1@XW zyO|3~Vmq_=4dcFH^Tg0_^|p~6;SKVB0*_YLqr=F4yLKyB7!zx|#kbhw$6{X-pf=L%r2zNoz=;%7O z`_EKvA)t-TJPjufX3>b%5iv@@;QFfur5Ix+1PSk{GC=;aJfAA`}>B90}F7HjFZOh;t^&q zI0N{1Lb(5Tax!wfIPy*o-@*F0?(Dd|3u7erwPa|nU0c@6 zfOLoe;jHE2<+|duX^-7ia*#-Jbd;nT48uL^v*bBxu-skqh7 zUC=$`%EJVnwvpoBCAh%M;n);SGz{t)`v@w+W-zt>ap0ECoU?%IS%xKe=)G}fFEdz)64ha zG#o!S=lkyYIIB-@)t>$Fvn)4I21oz;>WY<@&qbvP@WlXnLEwH+XHLMwD>>+ozuzLg z4zmEGYl93vXAMx#?xtfP;o{}w@pE?kwA|zlS*|n}UvE_}TWm-SnbXZC^!*0->de3l zGQ#z~P>J1MlRpt1ntC7^Lhj|CVJW3=t*dMgxU=g6C<4byZ4*oO7W-HWnO&D;V>m{I zPi)l)_Cb!R4W{1A?6IfxB;0rE9%<8^sgU6 z{tEZpSo_Bcn8=;3ynvktqFtBA#Z5ISR8-ALSxeH?{5+R8no|{wztN4H1E3sIe6Gmf z02)QB?_BJule`%+!@O<3W?#+@HbydzI?T&X4^&JSlQM?Y`TtcDmMG@7Ptr!seH`{V z?e)mBWSAJCks0B5QX}adpDLr9F+UdFMR`W7(*V4ROi&_i3WG_M_hdzX<}5G5jPDV# zUw5_n|B7IsU!IvN6DuPxWie`Ps#j#d01DH^XNr#f&_o_@b3kEatM2KZ=VIzQz(JMvFen7+synAmeZOszJ63t26w848}tsQui%y*K#6_)f# zJ5)qFnvbTgjE{}tjEmOvDse%Fjb^1jn%XkQ3 zb+#+q?~#DK`;Y0@WQx~nKanU_lPk?77t1Gzai4+;FPBEuruI?(ebaV zYbeEVE{Ae%()Kc8jD{qne;pI1n6eS&puB-_bD#?L@y^{!&W3LFs|vplVD{EWlP-zY zNj4ur5jpH(@cg-7kn zNo~p6bN7)00mB-tGrRY>Wv{<|BdWtAPePbKR#k0Qd#Ym7ZjZqL+i(SW6CG#dJ_Z@m z#FtR)(bSa7((liQ$c)b-0Lh63x7E8*PUJg(13n-G&$EeY3tYWJoM*l^(QGxRB-eG5 z^^fm?kmvWY?r}HYZFR;gI;ORF-Nu!82$VRJMX^chP*{Vwm=8XUHLdx~N2e?)9{YGw z!RsG(hC<4jNQ2_UO1<1VZ&&hV8HkPu9cpQ$%W^aTwYS}Uxn4gXxM#T%QCnS@?{})XYBM}a=a!We!p6+ zYwlzAK!(mVCRTUypGQ8~P&`k{>{yK>hQc6&n zjJ{kS&x6`~-v89J`>AV!Ufx2{&l2zxWFz!~#PcT%zU1;P674TEQSiQ1OB(f~s;2+&@)7n4kmgT)%Jl?`Rq zDE8Y7MISlNs0nj5rNt-QT~db_uAr6@Qvh}V!6t`^-_r-bTJ(2`g6KN56(*9TP$U2{ zIV>?@;F*dkb-?pCt>6b89G(JV!tjvwhYN}bAyJV7o>NR^L<6hSS7VHzhhn?^%h2sdu&; z=~_lF1n6|x$=(I2Ky!*Bk%C6TdJ0@egbLOuMWY-5eNVjGff`LH$$X*Cu182rO$<%d z2J=0afQ>|<9&7|c(~u5}%YbpF6iY7M+3DWs?@$ODIHOT2Szc-WKkX5ZI4$HR#74cq0q@!qy46}EMTlTA2gSwFw)RG)3W_n+a3eNNI`A6 z#8ts8N03YNG+`^aLH!kpUE*>8h9fMe<=xvEMzfPyL*p;3S<4IDK>OF^knY@6lGYnD z1u&Dc2%e<^PC`&g#dJlq*O0|WkTbSWlOV~EN=G%}h!oKQ1-lNGO)CrzY%wsB~ zmlkEbe}{$_YCeOJS;u8tRZ0FSOyE@yIHcnB*OQMghOXR-%W>Hl2hq*I4~8`pwFOLo%xuEZwV+0WiaNI%ai!syZ600{0X zYf+C2ArG-}?#gR)2rl~7XpX1-xe&8#nPzPIQT3YsGss4 z$!b6njU#uIkTIJXf-W4iOnI4-0kTZHpp<5R&o&+c3G{yex$C^XzEv18##ZESp8L;? zQ6Krfs|f+ckhHIB4#DHyth|2?Ke>6(0X;~FVXVvTnE!?Wu8w80PUJt#vcLM@XAFZp zjv#RFQT}NmbM0vVUImMh)9Ie$=poc8D4wh2^2IdGZ@f2-m9j?>|9%!E0pv)NAC>To zasKfR?3rAR2Nkt40y@XS08G+ES$`i@H$nvMzs3JX?7qC6rR0-h;s_$;Y~zX4##-Mb#x_2L5_29nN68$@&+sOK)wlmwj3g; z3h3$ZVbxYK!hpb25?r^qkBA3M7uC80rS4^Lrlsatc35tEP+FLc@M zGvk<7R_;;pipP3D!Ys=Obb#0qTy06nBHI?z{JcCzF#N0Y91zqj;ygO>&z~*y2vVIa zlD*Bfz~4p1x((xO64|E2ph{C%QMn1GE5QAsC!2=sc%G%P*3<_;OIQI4o5M%he>3ntk3gvKE;Z0c*9+xP4ec>fb#KC#IEEdMsEAK z`e+F04;?M-4ztdBq`soPHP2gVpnFxxlz|pL z<08SJrI;LPN|A~N=ZYD#6iPZ?n|&^<$jgzneh zBJ!op%OA*hqc2LOX_UHQbDK>Wlz+ymcaSRxp;-)6z|&Mz&FiS>W(M)-RfN);}UwcrN(vo++}6^|j>J#D&|Y ziimcPu>j0260&v{;_q2{EAZrt?LYM!aC=iCfocLErwM=FN=JWyd+l6D>d|f}7R}yu z9x-~?mgEEfpF$g&wNkI3FU(a1b}UocW~il#7wNL8biWSsi5E3<;Ux1Ybe$tDyQdlN zRAOB*FDZ4=dj$;Z??bDW@p1IB&sL&BsqY#;MnDSeGgZiGdL~UhL@%$xVu_S$G^Jr9 z4^^Ut-}pUj79#KZ35~=7+kHE1T=>X!p#f+Q+*^e&13t0>dHyPR&=v^e`OZL}X0tL_ z!^wJH#-UD)fX1bU6-*IOK@RPOLrZAoGX|ePJ-&LWHD$Nge)W@5+0D=c*jwzZ=xzE# z9l(`?cQZ9N^LY}zh0sgW$1eucr&usdRRbELxOIU#;Fm=wk!}C+@aEWUp@k)#vGw#m zHR#v4uDbiUU{bnvYY>`6t2>~1Lk03oNvluGFrxB#7GDytbJ9|>-Nwe5%ut_#C(QrTg^7q-iX zukN1461^tfeP-hRi5gp>RQ3YKQ@#$LZ~GA!QZKxrtLoFkdlAc=BCZPmVnD>-%PzB! z59g9yS}y+ZLBgxj``vww91nADRa=Oysot&kJvWl~&$G|JpBD8j3tMom`mC)xY=AZ# z^SE-#EC*Eo3{vZ*R4bmgq_P4-7uQDItOaY5j&*d;{V zqQ8y_(}|!0n@3dQfr&`z)-wJ;#R^`(g&89fMXO4bpZdd;ty;dCO!l;MxY>G5=F!!C zsFhpw&K7mFv}3fhau3kqYM_PBmjJjE*Xnq)k?-U0OxvYm z+{(s9$$@G$e5xsGTP!qAI1+=7TGkMm=CjuVJ_P!eV%PZA*8Y;N2Trm=oJz5K;1yX= z<+}x`6|q1^d)VYBG8^&lWB|;LR4Ewk+DE0Sc)^k&_Sm>}nx!pfsI7HPg(G1oh4jh+ z+bk%-1?nZWQf^h-?H*5Gk5voI5O%KxNd1tj-FW2&BpF_h0!DTVj|C9fWL81)oicvMh05z{OMr$0U$>B&}~CrE4x+= zajq+HQ|XgVBL5JwigbMnvt_qrKJy++8B@#vF<1Rx~}|AD1IRt|xn5aDfJj4LM%P(vYLI47g5A+9$zD z8Uryw?1~>m+9y@$16)h!i1*Mh0lK2~TC3pmxclRS&H+p5(cp)t*6& zq*t1J8`#vu?)0`b01UZXcQNhVP92giJ~P1F#9v2>K$|sBaQ_LoSj*F3(Ai1lrt)l? zZT9y1D1@iUp_(m~CX`NkD-6-jn}6G(lxUvF;KW^A1VZ*`0{YU(B#6%&SMUME*vlcl|TSd^)fd~=H3&(*DTdX1l%X!r@)6m20O~=V{~amNkh8@>@S)MgG%IJ*D2uZ7Q0UarA;!gzE8vANlLUcFwWUe6 z+cL+VQe&S0$477z*DL8TGGcpbcC(6#J42~(O4!4Wj7ZS09ND7!5T^5KZ9xj)uXNIm z`Bc2Dg!){pTES1fU&i-`9HZE%m4sHUi%d+HH39FG0!C?!GNgtJNhoS@owunjDt$^A z72!-i;}F zs&JlsvAAi?Enxg?T+&j!+D=DfPk_f9)m zM_BCLtRmMts=!F zy|K$wx9s>BHS|}vU$d}k_mpF`tc9sE0Mzb3&%bPnLUpW>PKg+D(^Pei=uF)*d-8@*)kOz7@@+xm0 zqSS)yOHSG4%UKn97s|SX`-ku-t7d#;R{y+6$3T??yF>}G>oHcAL_r=Sc!78R0rp4F4xs=&kja~O+Pxgv+1*8w0I{DrkU&Pir&L)kCm6WW@ z&(h+*O@4ZC{D-ZTDQSlb@=us14(>KQIrNwnW2siD!2f_rL487L{&G z#E9%YPm+5V%Pw8|B%~+Eu&JOfv`@E~zv4=0+^F2Mq9Kpxi}acKJ_M;o@9$<4%*QME!uAs)w1eynTqet7KIL|CTiX?6 zni;bvr$UH#gpJtmAq4e|fc)GaBYWGc$@~4Q<3xfNe&-6ceMGGs-DY^y1`y;5E`nYy zgM5ZQ*MRFenwMJ>1XiS2w z;dAc=F3n^K?_q7Pf~R?HllH1}EAQpTYn16K>o#o#&R#c-Zl2va;IAx)cGYOBc(&_l zi*H9yzOn%gr`odfhAyXljQrA`=8wEY1TLCzV3giMpn#ShaW(o?$f#us4G!FfX!KHbyvH`H>2N-Pl zmD|h0VYW;}txT8!0EhW5pWVbu!#4d-2ZTB>9JroAO-l>67yisX0Tmk77fQ~91LP8* zwLGuqOMc+`v}&nu-IgJ@lNq!y(^VOFv8=4s%^A`70M5V*@3Ap+?iQ(RdqJV^t(}(Z z)|l*r4<7IW@W}OgG|6OBOJv=4xkk2l*&v&FTSLfSqDlw=Tul|xjS2;@^|ke`-DkFE z#b-&bZ$r7`_(ZCw(hkO75iHq9fUiCPp-H(ed8|c_!n~s?V)ax&$K0yLuvQJ4^!*}d zd^Csj!^v&mXQyF3{K6w1mSi1Gt_+P(lIfjAd6HN69h0;%acM~Kcmwhz44F*Uw#WV@bCnNz1S92Yud_tr8ZJKkR*z%J$cvk_v$0f#9 z6d>hQ6kgwLhc4-P{MByG~)}s*`ftw>>A*yHwOZe;Nm`xfW*TE%s55il_nLb2% z6*>$jRD|POiPXfzt5@>kP&!q`?eNdyVnG1aI+$!cwnF-kESzG#dK z|LF&~694jXxKq6hCt(M-yx?Ov`*46LWBr(zv4uq(wY{9~vc30PglwQ_tC0`==sxh3 zDYVOm8;z-s!R!(hgWa05HPHbH;TKMzLFPG~{Ikf>!2$_z{F_?BQt8TQ9@~`+CM_qyuS7vTexYzq z^2hgpeX2}D$e}~Mg+Li>(Jr)s7vG^Y9bgj1BuG1P!d#(A)V@MHh3cbz2SIcjXuKEl zX5&LZCf*MFs1yyu;6cn~a`}nD5S{*be9B4xlo3|t^H6bzS2qrZd2|=W^6>Ajo4Xxg z;p8aB9q{C~ud}+IOA0fwDaW+k=pFbpJuEX2O8z25aDw!*EHb%#i=gM2n;dNFr=Zr5 z)jti!D@0#0>+H_Ko(CECG5}5T7YHB(i@gX_&%qE6eMy8v+#9fjQ_RI3lmvn+oFkOg zVQhn3G5^2}+Zsm{yjf21O0thT0NV&CLDDFR#3dmM+_ocvpDh1V9K2Xp;)Cpx!{n(% zj7EJXkddk>I3*_!+JQga#lLwfef@Xsg2-IswAwYW1WHb!Y@X?hOV)VfvBaedH^BuB z8MH`LDbmm+P%i{GbR-{XRtbS>w7?1u$Omx$lv+D%{Cd5YhI08!%^LJLr4|VAZQ|bv z#7gODZ=6Otz{DLHIsx`4PNQ*3u_>bTi9b|)1M=~~fxELgXn3=iy zPP}xbX@fDkYba=1i>&duIfQ>#ej#`yOs2diM8!Fa^05t-*&Cg*mB&13D|Nw~%3_`i zF-hxs+s?vHvmxg#Yl3z=AUgz*q6iXg?YWMUaNF~HGFPL6>vNcC&3rC`8K%lP*W%h| zkadDfi$&UJe_xVd`t9E?qfDn?k;3BrSN5J*MQwNt2n6wQh740S%&F~)G7y9m6zV_( zO_jb&5;Aih%-Qfe8p@06*a2=n)9d(+2K{j0_GShMOkT>^3%}43M`H${DGLrlQYr1I zTYh*+MN98$V>2vE=1&S{P;%Ak0dvjX_4AxJ=A#*fcbQ`;Tcm;#Loj$1R)5a@xI?lFGX2WL~Py%=(gdV zgfh=;y0(x-IR|5o`tO3y#b`K*&^kSif%Hg*Rk#J*z$`Cxz5;R-er*T>mKX{X6f0#0 zoFeL>mT{7DkvSWHz6~fC%2@SH%Gok2>@OA@Z-o75-1-AT@E9Ghl662a_C?VFT?)_! zXK1S;KP3hoMm4#%DP!%xNl4)aHR=mmqbwRqBBp{y%{_g{-bt>OK&w${n^W;Rl&aZp z`T2)^lP7L9l+6f`2Q;?@HIh!tuE5CgDj?~{ZXzhCg7$oXj3RM##0$fR_pAGs7c^+G zO6{$#{Jb*jM5cK3nAhF9EiAS~T`>X&&fge_f140Fa%2CF+9Ji;=r$8_NXEu;nVoDL zq7%~k++<6azqBoHt71p2IYI+A|M(l_Q_-;Nn+muUD?pgD>6OI=79)+uwBi!eo~nZx z)sE-o+rZZVSQWy>CV|n-dPBKhLcG036^_l&2uM6}h%wdnyJiz+Q=xSkFcd))Xo;DV z+L>EG6cY|3N6b4FJJ^~l)fN1neBEpw;|LKMAbN~PJ0mgaRaDw{$y>l`jg3BaVgsD& zitMgD#+;A7W!Z9j31dEjaP~YM@plW3P&uJc9xG7+pnhpqUkDVz9S$k;%2ez?!*pg0 z5D?iojuv24WJh?Jl^gL?G?)aMN(iX@(S@EduqJWFC8Yvf4l7;@hjIve_m8~*f~QY2 z=Bt}XiN^ReT*rY!UhCch%LU5gLe< z4vD3OS2gw(a1Kb=naDGDbtFeB%kgNI0qJF9M0&+PhhS?Jh=|3a(wh0#lELN3Fzz-U z{YwbmU^5S$PJ}O94%1dy{wWmr>}W368sdEbJQ!slb~H({8+HlUbllXBs-R%JNMwcF zB6Ute_#bbWsf;*L{}l6i!A;YyBNWgpcm@u}v88L?ME92oLaP?vb8e7*zG=R55(gya z_;|Fb2AoR8lgadsnSyu}gJV!2{l(8vwGXbcnG-|qU3*=?yK zx_28+zuO#t97TzFlJ@XME?!ldYMw;?iIGpc=Fqh5r>cph&FA9E()8L%qllNy z5vpkev8pUlKihcxXbAvLvrL^-$q_Yv2hf;q%{2HG#%syg;+RDqxvW)covZ(Gm%j$XhkuEVX!Cu#>UIM zxVbs=p`ah1%OxyCtuV6hpP`$2_Q$+_);g*}NqV$_JqnF+CtwemU@jBBM5lfaTGT>2 zwM<|T9>uKP0F*v*bax>!k95P*z9D&H&Iwq@95aS|;iHQ4frK4}3}s9X>k^c7&~+kE;)hK$>^ zZ-S8``H2mLXn92=C^^Wj)KL)_U7ZmqnvqyK%s#26z5K8#3E=2FBYQOAi-RZ{CWJpa zpu~1R(zHL0ODr4i7NP=N_38BX_&psxb*b;JHguT{fUhuwDH7z;gRM)04^VZ6y4H%I4cb+`x#WZ^!ogM@zj;4$IiX@ zA{6X7z?)r96z(w=vi9%q*NXm@{Su|-spN;&Mql3xJ;39;_1YcKxm%*az#By)AgvF& zzcGc`EasJz=P6)ceT}5wc19Fnn$>r&rDJIYBN$tMVp?stDW>U!noDes_)UbX^%wR< zuO$OyLCroHzFO}!lSK3SZ)Z;KUZJOee9cL8c<-yHfnRRqi>dYf^;?tu+!yTs+m50T zNDD9b68fKx(@)Q-ZK2%Dp#b5(f|HzffXOWb5Rg-En!J>^;Qu+1mL}y*03eH%km?>! zH1d{eV8dP+PaRcD%xP3sLbXW54*&{;kZ@E|P9bby_~_)P>ob`v!No*DvL*kh*Vb;p zfNH`o@7TAB49T5jWIy8Shpm0BDR=aaCCr^Up*9zu_{44kIS`Xk# zzIdB&LNupN>6Ibtet*;d1dNMl9#V1Be)iCO4qefBzM;O%yrCDujc4#5%84J41q@_`rxz z_*1`8s2DfG>BG8}ibKhu4M7+|StU~8Sd&DXRY@m+f`%ujR)L-+2ZZE}H{_JE{w%;i zndy_LWGv4wmSCMD{ka3Nj3%&-<3_2-;S$?U#W+uGoI3fT3|Unr)gQVACsf2t_Te9sALlou+SAWGINRKOI7l3D2J@Wv}zt;Wt& z0fslgjwHr;b-)gM15cRM}vhIoabg~L`s^dOTdKhh`>tqZNu%Xl_} zu0&6@ZNxQHXv^DM{|9F12~{*w$y)nTz-f=dwnk~h2sj|OL(GKwMiAGytr>!_-VfC_ z=XVN#!>s}1QlI(lj<>4ph%WgZ!3W>GhbHF_m!(tfR*i~s@ou=Hj>d~~@*V_YACqfa zZIT&)P4)sJU|5*lYxJ!(>;DfSf#Fgvn45q-9K8acgYX7Kn^k?qE{`H^gpZ5i{eYg1n{Eng@l@V4W_-w%$V?Ir~Qoitp|1o z`wDRhg=9+dDEI`$jn=ewv(k1?n$WA&QYwC#gx&`XMfK%zZ-3qT%48#4=7v>bg(iw< z>o9N3OZ6l>Wb+#&97|3wc0jQ5fyUsSRZ6c^Qq7YG!nkwSqjZQyB$~lHrR*Ug4`EWi z3-DwT;r6it^REri^GT$qucI4r%W!aX348x_czW^qG=l5+?+0q>!`R#H=IrY^3uj+D z+GPRgkjG0@B?X_((kJ2&e_KRm+z&1tM>&bQ^C*LJA9k>b6V`=ESv(%Yl7jP7-6;ETEbXfe_`V)WBi|Sz;C+GgM(-Yd?q97)ZEB z!AFI!cV#{0cSJqipDBbp#bYJWY_s}fBEqENN_J_H!3!c1aRK88l1(+k?>-eY^X31>iJKrTz{4tr z#v@@K7dD$l4+r)LA$5G#WGi{g>Rh^^juBXe9K0uqG+@gDeH zrk4K=e87ZL@$Q4Z84ULuG{6NK=kOKzZ+SjgAWe5s@_{DdN(dPTt;H3bfbWFc)ZmR~ zlIHItqgw17au^79nqSip^De`-V6OyFMYKu8rXsj@!3!J<9-4=w&H!~s6og?JRExUT zq(*Ndk_k>~1MAF&HvzsNXxUG6!{!8B0C#x@lx-=0kP=0k#7cDU=Y(AlP6GY2g84L~`!g@k^Lm7_Bxe(ARztxt3-v>OzyWMae8v?OkT6urv9yTNhr zF$YDgjC8N8XD5C90V$~A`7(&2G~(+cB`i}|Q6d;m_}u^xup`S|v06kE5Y|y+|7AoN zC<=X=oa1#wd6AAji|(e@?FFdjQkHksnkL~m0w4G5ltX>ke$p1+iDsKo0xYC zt{Xt!e}`k(YQ-$_Z1I+nTe;~oBvMd(9$224=;i|IqxHyW_a4l90A%79H{ zbcIp;iyG+fv4XV9ZpGY`FEcm7;y-*GYt4{Cnq^rCvU_Fm&og08^bUg@{JQ-6+}9{9 z;84jjw8MYh?PX}x$ShM`G%pu*D_*An3H%E=7lIu|hfYcHEwc*W=F_vn zdm?5~8NkYtgSmrT4uYbDKq3y!^D`+#LRDHXn4D;jq8{z7l6JHhoukIyfPMaZGtX@y z8}-9F3gEne%hy*QHP^JU5`;_sqX!QkP93NKaNJbfpD(yU6!%T|CqKTs+G^F@syT{R zxsuPS_d^TUgVjS`wd;ehmw*ocpw!etgfK)Xc@ z;3YA@CqQ?QSAl2QzM%6Q#A$=v`kA~PWgKN zfM!tG$*Y6hIv!ht@@E-7B6I_}sn)t=Kpe`ZzmlHdI%yB-QrI?=$cg!+VD6}Rz$?&w zMxn8BhOtQGfaV2Y#<|zjy$-9a=dffN}Q$vw+VrQbIC8xm{F5t~lioQUB^!bEX5gUjqi0ds%i+ckQ?npP4sQ$t(!BP7ufJh`} ze56$}a|xK_IiLOtl=K<$I#p4L$=Gdt^#4WGI|gSKE#behZA@(2PA0Z(+r}H)wrxyo zJDJ$FZQVKN{?Em(>Mu`st*Z55?OnT9_pe#2U+HM5>#&>1s9h_qg1iNUDTe&((IOlbJ zg+FQbL9{>u3G{&-6wy3zT-4elvngyud&s+InlxJMOR4y7Y5Wpn%VF&ze8Hx~6gXn{Nz z@7J68?(KBNg4Vry@b;wdh5!YZYa@9qmtg|}Y^)iEpl*FuQ#ihmcrat}lsi6&GZT%G6Ms3qN1vO0Ue)dsf6^c)FNsyeCFw*9Uvy>E*8WStHWt!J-M8{ht;wSNF9$Ea*c z3L(LGqbwE*YEW3LEBWK%iLSA->;}qbsKI!y4=ka=G2Zo?oOC$$ZOx`-N^>Qdu8vvN z3I2D2B>Pf2gr}|vi{}Imp@@SGEuUC&uAITRvk@OT3e%flHYgzIu3Q#5mNgYnXcN;1 zfi=|)fGdQiqomF|0cbu0hgw+ka_EoqUan~M>LI8wt&lJd{VT{J1@q<>LV_wLd~D3N zxkE0;w73Dz{i#qWYK@MvqBwIO0xxm5bxIA!2!w#`-?C6~xPvJrY>jwMSm^6=ribkN zg2%@uo*((_T?Kr+B5>uI+x>jDqhv)%B>XXK^QWEC=O=2p4N9!S9}oR?8(~UcCOT}b z7u898Kp2^yt@rMOP{Ym#6Ekb@g3k_VJWZ&cd&)PFeq59X=pFw zoFqlrb-e^xa^x}iTtAU{D2uF6|0vtfCI~mZ1?7|SCzzt|pBSo4=JUYkLCr3w=iSP&y)m%&!SEY@2T zl@enqdC@#lePatJPqJ9__CrQ9Jh?$SOpX=N4=W|B}wrP`?bt|kY z87zbu(+8ef2U#Fq>^rDuhbmh--LD-JTXwX971yqj&!%RustI`zCm;nvU$boM%1+sX zk+5lcr&JA45y|3Dml$g%3zLlm`(_27(7^sycaXFMGAJ>vrCe>?}qFb=F6EkPvaZz<44VDDQx&thG2HOl9<-yg>A>1(?f@PQ?#}_leh6Ggde~q z*O>fwq%x$p5-+4fl+|N9WNM3cuQRb;7aAb$b?)XD-d%yA?U$rVMOOXEr8({}R9tJ# zpK|ddMafw_yxM^cBnQzF2g7`^pro=|3bAZ|p||$}z1?pO>V2fEyx$0?tx2zn0H5`X z9GeJG4*U?ij@$@tKt`G}rE1G#}G8O<{ z0GW*!XhlhWiA+$Zi#nBa@%Sm!=vL3*kU0iq!XsB&KP0ha;NzN0#AI6S(!Nb=)BH`` zV#Ru8`Q*0G`mVEeogCY1{sqI6P8HA{Nh2%)oMY&|3#&fl{a*e(GEB4C>-Y@ z9S9C&s+2JAlmp>hdOwP@cIskhjW@?@H1`}4vUiyGDzd7EONW9RFa+Nz;&5ZEd)xIG zN)Rr0WGVT3*tLHlFf7X?#eNpe4Bhd=M{Q*uAL>d~=!st9PfwN@1VgAnllr7DWI<@ia=P_Oq#Lm#q$Tkl?@Ol380<)XV(2 zUneinm9b^BYL=_G%LS&2l2H!!fiX7kIKpD0ojkZniITgH4Wh{t=^kJvcdt7mWZ)g& z!$AQ3WLE%Po^%|a+UIW5CawY=q88iw^Z@rpn-q`4! zm6mL&q}9N4AjntBg#u8sp99GMU{~}TaPs-zzc^evDQeNz*l{@w=87NN(6_OVK0|r2MM`8K0pprZo!1u@PGW*>Kf1hX^*QS?g zSJ&FAuS5)vm*EkC23Hb9KJ|?bkkN5GlxKH0QdQ^~OHCXl$#LQh&pw)Do>X|1V3PDX z%uRKlIm1|)z5&wt0`somrX_nRyn9wQM*hNaY&rYwU$C`QJKcTg*?Tmq?tM70pjf_Y z#jO)V=A{3D*dE0ubdvY^)Cxsxfq+&?nRU0nuPO;S(xS!4f#k{%B!qWba&^_#1tsK4 zF+7$(3VQsyh8nT%v@Wpi&|Iglt-L;A3pix_5d%HeF9VRu>ZO>1`aXZDcq!6(RWyt3 zC_h2yn}JQHq!EbK8f5fn1tfq{%3-{gG-7LS#h9LH`sIxtS1wcksN3PiR6&mZia~=K zKF+BJVqOqBzgO&UBlzXh0g&LZ|9CVhk*ZUIr_H+Vlik7C?-m5z90Sy=JD8Vi6%~!mQ597bqenGM?#>1R zx)XcD*L!mJ@X)C#WsWFei{We44Q}B~hT0ey1xbzSfC;k%iWF|5fKDI2_a2N(-h09Q z6Ynus$r(U&-$FsGYqQ}t=e}1YF`=B=G6`@1>+SorjDCiKJtbJU*%9)2!tZI{?~{Ot z4M3wxZp07$&AQvh$nwJ%FSvDnjIBhY+%l1bmdU%DD&sz7cog#w&#!5C-;bxHIbCzDS-=FK4chp|)UJ*0aZ>0ePW~RTRWPTfs=f8& zpuZe*xl6&_>iX!H(h-Eyv3<#3wAS##nWm7M8I|hfi0!+S8Rkc=n&qdvUMdhNDZ}0d z96PefD`HNfdI~R+m%k8tX!tD@dcgUf}X?jQ3GGGrQD;E1{6{ ztKH2Ug%)O)`y({NYqZT2RkChydIq56Cvb;wot?a9ln>yUemWFqBMA39mL$K$K)nKj zBPi&9cX3!2xxYy^Yt$t4>v#+e8~a)M&9HR9grz0)$MjuOu0t{mruP&~)*&0x*+aOa zEZij$3n?W9`=G4M10>i1?U;fT^}ECY>EWafws*d)=2baFj#)kyx+WQhBoDw>bcO#+ zIkDWXRDeD&ej?36?!ydu{e8;s zZNw2#Ov4i)+r-Fh zue93hxTf>iM*KP>3GaP!#JzMFOm_CLH}4dtV-%e5@5eBswOLL6TOZ(b-6__gDw<6(^JQ?62>OGbmwOY z?SJ{Ra0WqPg>%LCcLty|VMBDJq!bWViY-u=Rea(*O z$+Gb03az(r@@X`V=+ej>^EoHfGD+(S7=QNe{qb8rk@%a`$Pyh7k zvd?PalsDb&U=EP<8}I~CjSUvpG0;fBzt1TG0wquM+V66>c!~RYA~Xy3*~<^ZOz?;D zBf@KNJs7fQ$U7&0!ok0Lkm*ZAYD72RC_oS{0ChT8( zck#hY^(TC;WuEkUT>)0CbEH!nk?FgZCLiiXI-|)B&`$uSryfdDs#dSJlaN4qy?09Y z7U{LDm_Mikj0Hc6`0bBIcHTJj!ax8{tRDRBX_348_pUt)$uP_lmVM5qC>N;!+46fS z$FN&i!vlUcyPH`Esok71BcPcCI=#%4)&M5o1Z5s=by3mJCq9Qdv%B;tiRe# ziX>Y`#i_*4;YD(4GNQFYj_?iYrQ%w$Qy`l9Od|k6WUPFl$FzB)Na4huL#|95{2NgH z@&ExWgu!}Z5x>0Z)ocx?_ney>yT2!WS(hbgWxpE8H~{a7pJ?r9itV$(U>xhTCW%@f zd!{9p=0AUR8b5EQZl*R>+*9rB%#zDi+4yXP&Aex0oPbhh5Cj~@-q!5%Lx+!WD$nZI z0bPKWtSXnk$Hu)sV`QCK&LMM5FX7M5SniWlD`%Np=jhSh52g z*eoM7gBXTd+GS}_Uvv(TqJ}@WI57<3KK#4=&E$ELiX|YRRNol2J-ru=j=rjjuZU0!?NsUFN2U6;uOQ z0@oxWMF+HA24IrnScWqIngl`wO!f=Ve6H6z!OH_49hKY z|6Sdoax!^M3O(0B3!vV%r$tb%io$Fc{>w{tk|!(4AP!YZr(<-= zQ2g+T#=P%nz8B=Sc8mWY@sf$t+_ zNF>6XaE}hq-6ww?X6HuAjJCj3+D#hhvl2GP@!dj9`K>#fJCUz(tqF? zOYY+=b7)UKOTmmurmL9=Rfh*?n6XODcjP+dI_r&k{GEjEZ~o&T3Z=?ej~ z0YrHO+`CSxv8pg#wM!;m!%f4F#{>IoICo)ck#(aGEg@FuaqRP_UToeg#Dau%&rmRT z_7T&t^!mrG=^GxZRi@X3v=zy7{fWv&AVN&<;($TCbb*@nPKXbacjE@EK1_=ahR%j}&AR9Y|V5csS<^xr6Yu*2*$@U=MZsy&#*hg{FW2;ERz*Z0WIa z75p?ouKzl$8}LTIUr`iXB7=v>scy}lJZ~yp-*JagY1OsNMoH7KSMcUjQrML+$w%pW z=<&YD0Dfp!5~`k>5AYuGT^283H96+I#}}M+QVT6RcaZHtjLpJRW1X)!w%w-9hAqiIW)|B^7$?qIJ+%rq0W> z_g^yK_Jij~f6Q6*wuw$Y3K{yuZ@T*dj42Tq!nOnllJ(UbhHfcVBgnEq_5#s@;u7MA; zbzDk^JnK>!t;tzuVq8yp*2O9Y7Q+ow<{P~pNc;|d`k(8=$H%_als1<0XfgD(RlRk+ zwdk{wUCA#t_+!akEgJkke?LyKgFSx%xT*7^*R@>^UxEOBNGd>Xl`W^&pS6u3bF~CO zYMET~C6cBcJv9yj8l)C)QWRp6Q#(EsTvK;R5m)x6N*;4?sv?WSq~W|7<5DiBa7E-$ z`~xyEos7SB*+)S_OO^9<3CzTM6;(q)FBEJ6ocQtwFnN+O>D8ITz2CQd_<@^y$Ta0~ zVSGX@;U|Ey31g*u)^QH}-~sc+f6YLuU~sCVmgZWr4R#$Dk|v9b#F|UXe}Vk2Fby%B z=o4x;QXuQ+C8%qvVUZ!M-bO4Q3<{Er79fht;*Ww%9}|bor~PwrDHK2yxg1mAGmZp{ zPP2k@jiuQMgJcAY_UX+to$qA@Q({o05`cu5n2Z3Ys1qhDOe>O4mwF@H#ID^oWF;Cw z+~VD$r5FygHt9&J2pa<}_!{(ZJcgW+q5o1UtX7d)u{kCLR{MMNiH7R5I$R{y zh&*#18qzest}@7GMf0GulF@Re5xNy@+NYINA(Phgi-mMVRRo3-N^+a?i)YdJ98Yl* zOgRC_YiYl6*lf+FvbOng5jDG{p`AdqvHi)ZlhhI7Tn!)vE9otbdv|c-Mw-hD#k`Z> zS1ALm4cevD$A~TcpZd}j_sW2H91d%ot*-Z&om3+nc0iCZ-XE@%ojb3t9X4zBm#fn< zX_6czr`aLYQWCV&h${%<#15#V6?#U8j1vGRXF7!ls#n|;P$0i7-6z?^p(xD)Jc+oM z=4$#IG4qUZxefNC?T|2Jy0~S8B{q5~?25t97O;A`7-h|P(MPm?BxOL2w1QDQRj=!( zop1g$&T7P`@wRe$3Y=Ztbtx5X_FlOl8+T6#d|asB9|_j@z1}@tOrLKrNp5NBZm#^R z8ND7k5GPIQ4eKjNwJ7~a%O0q#mr|^?LCfYXi%ayScq&gywACFJ1WEhFnvBgR5CZdJ zI(f|r*!Gc7cWxBLg|GsL&4r9W#Kd+vN0=#fhQroNfkp-wED-DZbRsnDldk39YICTK zv-kpW#1B1DKg&dx!I?iHGu^p08|Pd4jO@Q#&DZF_feejGwQSTN z;H+lHf`<@leINz7PS;%}G#z4^X@7AmdF>;R33c{Zl2&#*1wpr-1S@;+l zAh1f`5&Xb<68`aF_4&Pdt@}relZJ910!1kzFvOCuOc+;V__vJEXP{OX zTB&;rbfzXN2}&QN3U~GVB0nc!QU%?Zj-yJ(>9=DDZyRWnkyYm^71g{U zYga|DP7)eKMKl(cmk(&_lNVkMMtHntv831^4VpW3uajYC1MPhi8l|LkR4ig-gbjv53b2lLUq3#Cd799A{G`o^|i`-N-C7Hx|C9qWPZZ!E~I)A8-*ap!;XpBkN zA1CzmdFZ7d)Vk+a&YxNVk|`ncMPJvC$Stcp(_TYx0oxTlHxkP^mAE&5fW%l@zaf3) zrR1^wL-%$o#dUM9%lHa!p~jvB?oxhR z51*UCnio&DGpIZeAYq%AzAwHu&8K#JMFx%{z2soGFUFx*62URVT)nc9Wcucuq!3N^no!hIL}rzy2M;Z> zk?6r4)~dYXvqBdC+?IBU3r%}p0HLdM=Ifi2zYYcbTiE&(Nh45k3B+l}h#7+w%s*?s4NQGT{*G1r{%#l}upP zA@Z`HtU|UCa8cPQdW<-TDhd2f`}3fV&IP@y>+U6AkRyF%3Wbf{3?gAmQg4Mt<#vh>L(T7_jMS$+vMJOF>=LJvNBuL;XXf4UT1**|9Zz^{I?Ss5~ zFPO>D2na#(xnSsz50AgM%0T%%!iXn{(k!-)R?^}Hh%3-Bh|(@N9?B=PCS@d215dL7 z05WJqVhiwkOc`dNC59&(Cw3k?X1NTZjPl=aMW{Gn8TOg(bL3(AqDF=|_{&106uU=A zvBn8OpUlFn+qH!nj|6>-)4%4cB}ezl8ONeHwjTCyu}px+OXZNm*D%csFNXWOm?V{) zJFD9ObgEE;Q;-`>_0e>C-ahrRSt@V}C&HH;W09;dqk{U`no+N4q{Jq(E`xe>W~Fl8 zqrQ;2UYT``kH4oV9);}nqs$gE2PqrHN@E&kRH11Qyu&FRM&csj1$ zb<2*rQ`YRkl$ybJj-@_h+G*)U6!bs1kd|e+tQNMf(hKmL!iyTS=$4aHIzc^{2^UBb3&ZqAv0R zV2g^dvS-#BO(ZdB>Pj&{WXj#4i_)S*M+5~d-oubSebfXC{RM*l_UOeWU_aYR-cnYk z$cx>`Y2&F(AtH+%o*6q3O_ZIp4mVC)Cai{FznDpql3akeBeH9;i1=ID%q|{Q~b3#AZ zsJpFsZyN_crVNp0U7fm#Nm@`=hh)q+so~_|M?s%zl2Sv$UtH_{v}9rT)4F$5p1@lV zFcYBV(1h`0$R}UbVxhXhke!4bRh!>-`nJo^d=^T8`9>{MvO&4Kxh@*^@sGTXe?sHl zL7iJtt)0T*YZk0O&6UL=1Bf zfdL!#F>4W}q1!ECt^@LJ&?Of-Wi#QF4KR((=>$4jhAbAx>p&KZzaHnup*#O@)LM4( zIO$RdHq6>p9n2pV*-qZf8DM^HRRVofCwNC}SzVi6-!*y5+R+}T|6mw#RG4tfsH9aA zzcblpW)n293K;h9)D)B(0E~?{(nL=F?+~C2MoVVFn9#%%*6U##12q*TlXU22HgVoT za;#8gdlsuVh#N&S9*j6Lklp?Hc)wNQM!~th5vNOE>gu+2^*-P3zd~N#NAtd)_wK_4 zzGzbisL1t9+>Hvd^>aGxe^we-ws(FuE?PRfOJA)GC~ecCEGNCe0c@RnR8yS++0*q; z1N)NiK*P%h`2W`H@q73_$nw(0DZDb@1UX7!7V*4xB>;51T0iJA$6#nKcBw!w&vvT| zd_LB$SvPm98}&R=>Xo^>&sk-@K46-&ihCvWJ_v4WOcL?)7sh~{xqe2tqmi#UUEACH zBY>xN<*(mQ%naA+t%vhL%+fY+@&Jec7UK*$-cMP$@z>`u_mFBdK2vk1JLiA2bGDIV zK()BFz+8C3>pqTucEVpuXztG5-1U1u^#A`|2vZMeYHpxFK!K1zK&byejH6qgq%LAn$v4u>yQK2-*qjcQ1cDH z=c;tLN2v~$UuT;x!*s>GFPHTGpjg#L{1)y+oC!&C=u;MF>bA@zSUkXK{o&JoNY)Vd zQ~3dWKDEB9ZZ9!y$en__{-gKl06S3`*Sugvv4mMDdWXD27erXDw>+qQ2dVC-!*A?X zg~q!4ZnqutSyD#rDu_V6E+RlhY}NdmQuoU$QUtg!LIK@({`bi9vQB>vZZnk?{d}Hr zwrsQ=*KlxHGTrx1DQc9DF0E3MANNO~PwNq&o=h^)w*pqvlRl!u;9eqE)VqrxmOyt~ zjIiUz4BKTW#2Lf#Z2eQZtggNaFlQ4vc6nUd-if*zokSWjE{=rRZ`6W+K?y z|0j=w=CPF3Vo{Hvf>~{jyf|$=(%?a5T7PoQKh&{(_3Nth>&uW&;McXXC@4wj_Ob<_ z)n}`s)nOoTXO5+vQE?M(^vLv{{t>YyhI>X>W6-L|OzSOGODeT#-Xo=yzCGZu7xXP^ zAv#j5pEVubg9(N<&_01_YG;fQ3xJRwA*8X-0CZOpRFNcwK^_pU?nDQoCgg2v5&+c& z?W8(*;+jv4*R+t#?A|{$R z_E-sO9d;E4claZm;LwP!bY zx11N$hU`yIYK5*^s(ZvrbS$_bCbUjAUz}|96?yBDAF@dN;ns9E?xDzBfmZ@NB`RWa zTNlX@@k1y5a9;epR$p({jcTh8)^egNC5cH-B5;oSk7E&UeQv8b3EOl-kaov}|Y{+7OFuo~qdgB+$Rk-jaB?WA%eBesjwA!1fFS-&yo~wBL}q#{!A;Zm^G1aH2SxrrU}PN7|9CpwN^#Td zf+BI#Zrw`3|3B{ko16UKm^(jj&Hs7A6an6?|4Kj>UH|K%eDcx%pQ| zv%!N5{wIg6wnDb9wn9PZBK+HhPM8NofypjIT zp40vO75N(K8jF2z#{S3D!p{2ltvE4MhqAEZ0hhRioJBFNl`Yv zykd%0?a7ydgh-SzHHoTw58fq<(veAR)h_}HRV1=N4$2Qn+SsKS6m}<32!{x@;2MZq z&6bJQX=G+B0)(9#nl%^KP*68s!_=Q6X9U1f=n@%_TSCfFid=$O(9cBTK55fyO&pDkNRr7CvNALEv`T!sH2-af?R;gzB zt+l-ug$#^UM9tP<_d$PO@A(k*1WzmX)adN+_hW4Cj_Z_$a93Uwf9YHv@V7}(fe)Z@ z1LOL~4)8$(Nt{$bAD#34Hdlpd29xP&!WU|%=_TM}D(b|EY6x@Oy^7uA)gaQOw*E`BAB5k0IPqA4pC^UH$Z0?jNeSO*kmCN$p25;!Aq@ajwskd= z>1{xV`qk;V43#C}@c4N(rj!ZY+&g(&uFf%r6sGz#l6*-22U}gI`ukTHI+{p4*IiH= z@IP!d?EDj?fFJlPK(U_I5=AgGnxJDPm?fGQx6#Ndj1JMr<3tXPup%@jv6<0N4t1IL z2CBt2S;dPuP$mGoJ=8`e37{@-UIh_AuaR5LOqlPN4yPEl{d5xvzOl^dII1zG7N zj1DqW*)cyCO9}>_bzQ`%K3*0NY~aa%#ZOLoov9{7=k;{0o$J@@ zu<1X_=fu8p2?e9MbglSb$P)8aPERKpF$cR#iZPH_q*V4MN;1kdrFp7N@x}J}ANAj2 zl%0zWGkiG9ALaOR1Ih)qEXAZ^PVB+_6%Pg zC_qLl2h-we8URyz)e*+GxUMh%;Yk+?O7Jli7sku6iQd8jHXwLH45IgSQHX{R7Bqsq zBZ)(oeWES3OhgOfi9(pw)`g7}xm7cwsCPS}G@O8Vsq1Nla2llzc*lhW8UIbhFiY82 zz{sRR#h#63w`G!bwk|b>2?%O~x*{E(qlpKRbF5BGKxFq65djY%d56s(=)r(>A4a~3 zmf;kdL4v0M1b4H*{h9zO)M>%9g)v0NQP-fH3(P%}fVf?dCssi6$hg8N%A=Y<7E}oB z#~NpxqE%Jit%;#h`#mIH=qmqvdq9Oyf>^BF2(bVH&%Ub#bL5%=3p1RN4?AP@@|xJ6 zP9|3_UBq}7yb0sN+BE1{V;naU1et8x!id``)5szaAR~_rSs!coosjMzvL+ueBoT8I zBgcXrB$FW%Z311aAAwbnSv3}h)hXnv`HBrkPCDU4}XXH^&ieO z_O}i|qy2Xe>fap*?}K;^a(>NHaDmt#fLv23e^(R#3S@o1I<&i#U@84D4jZ2)_CuIQct3z6PIZR?QH^^#6g0MN&2 zC{_gA&}}kP#(4p&jT%KhBPnH`W=9VouYZ=*)iYF4h#O7`N%jo*(BH~P;t#0w&jxGf zb!1X_K$5mY3K(J(KBUgt-+EF_!yK~+oTX4Z)uVSHE@LkG1mgvz#)vyos0JA^@ZgdS ztZ;6h-Pr}=f+2ZR>=Z8HbP7U_018(r9R)HeZ6OB<>Lzs>$`U6n7sTkHL~Ul2VNVgJ3N0W-26Cd?a~ z_Gr9>9i=mwYPz6P(ymyRZ6tPw`0rR7O?s8uP{Yxcq2-V8q8AFaKD3a!pYpu-cd9c& zVba*LJu#srC{HMHWpr^qf~t3J&Q~mBRB(V^mY#Ewhm=|A$KS+Dk;q)*e z%d6K&v#?9~GMw*t5OYls;|MKB|6OCDj+qcD*?@E2DZV!0Kdd9m1rS9 z(b8|?iCJF%Zc;>ylo+(^U(c%S4szvy@DkU(mHXa9nR9DWsz{nCcw7eAJg zlK`<-Y>Yr86kebM$&DSZ9zSgj~`YzyA--N#Q`^AiYF)mJu30L)cCB ztz)5NbRbO7Z7lvFcy%~%5}c;GGZ?=b8JPxb8 z`sH?mbefJo*>8+U^bCV!Of3HsbWE48#C5+d&>8cOdQ~(3pW-j2P(4+x)J;A#A;kcE z5y}Gbzu=!j5|w`OUjEi-Oq;6%RfxC}o~Kib1b9Khf4Cv*9{n^<(~WiALaPOjCTz8- z`>kNIXu=C!i7I%US^-m)c)iV$RxdJ3C@J2muH(MxiKvahJg-GloG~H*gz9u&m5ma~ z2ZYHrph=Waap1A|e59nB{-$LL5v8 zGjESLz9%e8T&s*(DwhgL^uZ$r^HUczD0N?}R*Qo9Ffhp%b_&^{hEhLVe?5JiDOL;& zIQev0RhDF0sLuT+#fJKe=+F@Z&qKkG8n|I+XL#uW2Chm}=}Or&l55iswF+LesfliX zMg)#^((GBII$#;0xO5kX?jnI*E%LN41^gAmEtqs*ou|Ge+*vo5NNwC$K|M^>MLQxI z(96Y57Hk$gl(fJME=1_2$BB3_bB9^5IU3UlX<(kct#^EeAag?o+DS^(=2Ckz586#@ zOGKfyF>Up#mq-v=sErd+;0BJD5d)(gVub8`&Q=071DG48kbTO}zO9&Iiz&702(v6g zJ>x+uE>hY+j)^QnD|jPj!*p%mhwaq_J&3-A8FW%R+-{^@m0A}mb*j9r)L5HVt3n;+NxDEG1)XuJRK4(>mw5lo>3jV*Z`A_w4ebS^S&_&oO9g>uD!i zCuF=71cI^bW`1FeS7Csmu&`)kjj@O(SzfcP`JpA?Q}rfM_vMxqx&8_Cj6 zAv9**ffO(Bw6)_yo=C}7_{&&gY;>}u;_0TC3n+avKwd6gs$9!bllFNxfSvk7N+F@~ z!cSt1YA{mFqW-fYKaps85$MBRf-5oji%4woQ^y zm1KY5f}iWCG+QUb1;IubCB|M2Xa0-b;xkxWCQOV@knXgoUM8Co%PNM(7mHGYH2{HC z3FrnyE+E@e*0^K%Xg&cGa%av=V^Sk-#4Ok8LsY6Hf#Z(Kni)!DAW!4&3Rb|z{I8{tE$s|V|pZH#Myv*9|i26F2u zAQ6fDQ}NA^>*FP!%s|r$@F*WjC00VW1=#$n?xWS;CTV#UNz`UnypJ%YD#K7mf#P0t zoM)vtKoX@q@p&8%0`=HG**~X~DI#WCe;R-=h8$1BA%& zb3KKLQXeSsnSw$yId1E*O$_;|ce(lKClk3B>Lp=VLo`Vhaq!)9seF8V?Tzr8=BTfs zyfhcq+|ub#96h`(E+yIC6$J{NKuLwt6D2Rrsc%qBK1i{J2(A*sg<>w_ITIH$q(~Ub z$o3!2m4~0`q9nCl!>MD}II~IC02JwEE7DvQvZW?yRKWMGGp_%NU5X071bEaMFWKVT zh;JJr!KPtCU>GmAdfF7iA^*C`RrIn!GO7vQvhwu@I5LmM_^^|o_-w4!Y6xoiGnxzF zRwy=}!dl!NzaHG3t$q9~Jo;CNOVtZTD<$^&V-dv3$#4JC&8{IA+p50X25gC`P6zhW zV)XXQaWt`q1%3P3|2Px{c3yDmn%iEV>DXqvh&7v=gJYoQg%rgw5&~?xv=eCzg2q*A z620`>JXCTkGtRF=SY(_P+O+JGSQBFtq{;!qX{KBV#_FpWwP@44IiPz{IY}VUDT!hW zimd1rm8JfrUqAU1B*f2yfay-u$H4)b*^^6X0#ex2WW1v?BjaJ+&b*&*YnF%aHeSlaF-L3EuT zG{fW8B^U7jxQ9i1epr|Gl{6QzWnHz^cD#0(63AzXQq>T54$Tgz0FL?K&<$xeHjBf( zA2p`WRQ766MIK~s^NCMc(Yv$n1qTkFjtt6L+dpls8$SfPKtcC}UEIOWF{m7j7sX)|>Z;}6Y7zqjz^v=(z<+6J!@z2R@!h0=8~ z>&(5oZyCA6)uPm`02c|Q`T}gQn>W^cvV{*(ln)#5H~jF=&qWS^!9NzZ=6xMJ)Gw_b zT1Dwi0kkMz(gj1NU$qlt+Vtw0bRwC|_Hgs0MB0PnFE;(oE8Jrpf!y3rI4ag$K6qj5 zIUR}6weDZ{Mm~iR;-Z(a^ih`C6Ms{mFW(GOk^O4DPU8?$0a^>}2bT^v27oq>)pE-z z-5<>ZHQXD-FJ1IbD>9tPVlSror4J2a{*~gEvk{iHlOx#7D&DQu3ei#hmi;OLYMt`F zZn-iRc~^`V4f(K&sR49^H8Iva35b*P*%ri{fm5yi0W+2QMRi?^iInj8P`+ySx(0l> z%t|?UUPlB601w!*ldQuS{0W}M*8yOST3!;@ay(g@acvNP18QVb?9sa z&CY1!Cc-c+`6|psakun&gdb_Ey^M_v-5gz(2B&sGjc*UA*@WRpfFr3i;X3$r+$Xg= z#6w0uR#{}lpmXc`MEt50c*V-?#{2h7h8BFF&&o+u+3j)tXDg`B%&R z%weez^}k*u1P&3`tFzU?Yb*9?{v(s|=S|fuGN14Rody{7m(Xg?@dnwAd7M41ja-YA z9An`RG+8$7oiGaXN!;(ei3Pty77FUR6E1MpNVVc~(4^2>uQzzwO{)h?EuT74Pt0@8 zhqTpofN7c{cJwMSBZ_d((tebN(vS~ceZhs=B>TsgQ9%K5v{u>2M&TOwZk?7XH%b>7 z)Jcc=w6NBtg{#jnv1{i^=Lw)4qd{|Vm|n~%eg?Fmld!J>AaU$fKnRYK6}9Vx>0u#{!CwspNHn}+`5Io zz8r?a0VL?X6R@S%q-kKdO6OLsI_0ftP31S4^aW~ZofDID*Yu1$h1cg!S6r@^fJ)q| z+b2223;qNLG)*(yo6XeI<4}>}?WGTG$jHvGpq}KH><|9>0qO6dJ04oo&reu$po9I91rBb}ETDKO%LN_L-0M%a^$160*X$W?8n+ zLQXAV^e!uxr`ryNX3&o_px7ht{rRF2@l{WwcaV^w$~~=zPe??wpTC3vTISSREc!`XX&ov)+)>PC{g{m@0zi$HFk&&SsA zCr7_kp&_}#;D~ySTGiQ|bzJl7><`k{*0&y{K<@IZdm@;w<8S@(|BI@3jLzK+zjbTd zcHi2zZMUn|ueSNtwr#E2wz+ECZddJgum3({?{iK@W_V)BSs*uX zK;ri=aq-S@@#sW8L0M&Jm|z#!A!M^j9u5PakFh}|6gsRsK>lCf^?wYT51$Z~|KozR z0T^M8m;bjp(xwus|9?)0UxUN_6Z}6R{1f6oA+=qDBbP%U{*PJmnLx%2K@S2#mp-c* z^?&`Cwl5gu3ozj25Wi%SDZW<$UQES+h~<0i)tEs>?SV`ihRSJkAJU-BNw8qzP_zIM zEQQ-oGKXx&u;`E z0~I1<62E{CK=0?<1ili-E9LRC)IlX?oCKTW96gXfvjckuPPxE&P7BWC*GOH5fJ%FY zoY{Z>3<)j*dxmbb@We-h>k}$JMs}L&%;9V;05<znIK@5e=dV3n(C12 zSJ~iMk`agorJl|h=Ek8#3sLS+`jM~*)X^(tYP`{uF&juywVc>to+x5M)hoLe9!m|h zcoT3e@PR)|=EwTOE;xQ7mHRe-PX6VG9C_uH-=FP)wCi}YH$gT(dwM2V@W&XBKp*=; zlZ=5%?d&Jodhw4B)29^Jb$j(j&`UD6`#CwkJ~C0^we8~LT$MtK?Pm-L8MRyZM9j&( zk_>%St-TkMd!D|@AjpysPm9f=u;~7g;1@=?%K`okYWh(}bp_@VRw1GFZNA@r>I0;o zCQw4o8=LZ}zKc~nYSy6>lc$tRDrbzF5KZSU*rAFwGS4{dEZ%El3%r@esZEtAzY&~3 zpC?^ikl#VKK$htsT`I1GEc+#Q$2P+rPniLY#OET5*O5(|xcAs{=;Ach*dLOK5f)@s zFF^lJ__XGo+)s*6y{tT-`&@x~wsb`2$*^J#-BM>-wxBUWtFTvKS$QzY(gzNIi5`#2 zU2{SRQJ%#Ov|OF^Awpf119AR4 zMve)^FGvY-r$`+eF##Knv>2k%hBodCnlsNc8{((4jxRU5TEP!C_o1wnVCSzEIOe)I zNB=kv^!a7W?75Qhvu>y)RDYPPRkwy*{`YPM)(3chv;1(FjD2;D@BaLp;rDVg^3a4H z-{U#8(=A{s)GZ{yXDrnFbtAyX2YlPaNu8*c&598QJlQrpy-B_m2Z(CaTHVV2g959D?M3sUOP(FYIYNSq1%e9#eBFR#>Qp;W}u=8o;yQOMJ?I zfd5?ZkZHuA@F^^ux0#DLnh}nYRBs_`32h3)Om^#ltcew2(w1}#v$R_Zn_bXPEkcQ< z1UjQwezy#pL4Z4oD-mi@u|n!Fb*#|uv@niv5OiXx3EMJMgx7$K=o4_% zQVcx3s>2Hs>|y+@BTA!%9UvYn_pnT~i5-}aqDYZE))9yVX$2ew4>V_^3^jY@1VSMN+ydLPA2jOVH@5U@zY_0^0Xt~ogz-MohY_NR_#a~bFKVaqDp=L zPcdl3LWQ)Hc?BET4xUDddf?yI_G45DmA%PWgdN78Ows; ztNkL6`l1NVzXXL@Qv19S(+KhbBmhnQTSd|?sz}$2xIv1M@OCTT>t$0zMxdzBtABdy zU&khf>cN(km@*9BIu4~M>i$Q}S1DniUYOzpi(eWbU5v0SM!>WJBk{0z`6>~|K4ET{ zn3MlhMOKYf16cI{HNfthZ5De@AGi-&-JFYy7rmf$TF)iu7ZxB%i+r#RCOFxqxy)&I zGn4XuxN2N!)7%9KcgrC8WP%IlL=q)Q^X#925EKMDVwZKR+4uxvzk;nnGIAb15R`&x zmQ+KI5qZ(oOeMZ08E@CJauf!TZrJe5;ZNQi<2?wq^C!ha8z4#nh3=ELndm^5h?$0S z9KnBupqR)F6C*n(goqd?PY}s2W$^;fc12|mHED1QoMdy!_r)mHI&~g7C3Jv8W(CO7 zqJ1x38BeI|oB$SivLzS0v1N)D3$&ohzs3r(fgoLGu--nlSAj1aP&bjo05~f*KqGh% zFEg&L1f63HE0CqWpZpl)24sM?l4WB_dSHo`s677!qo4DIj5o7fCzzZrhS?Ko&QVX-VF~DS}f(rw^*mgaF0 zcqe#h#SI3H)fAR@s(VjdF^9Cz9-2BBhwk`!4z?AM{^t>w!JuJe>U5*V3;)ZYRL8y{`- zr_K|F04NW#)4%!`N^lUrgtEIflPOFMqsY2+U)rMbtGZNqGrgad43p)aE2EIs#(Q@` z>7YH5h$132k3HgnS{k(DDRrvic+kjRJVR_BW|$1uuE043oFf4?NrCT-Pv< z>^W*zd&IfJ_iW!!aPFNgV2&~(O2T)T$hSHC0qE?_r|KXiV5HG!nj;YX3L&Z>tX+d$ zV_UN0*I8e-;Wi_5IHkgDKC~e!pepxT+N6JUlh%*e2VbUpuvh{qT#KiB+KPog59gLn zx@bMaZ>6j#L`YlB{)l!Bj2G6?yV7A^UDArRo_jt5K+Nu^d0#OmR(cU0%m*xf`&~4< z0ox^s2?$1r;K=jPn*OjKe*aa$c56w!g|>)Tz~^A?h}BxR86>A4hp|>wJ(o#Ekf;xu zlfJ?BbU3L()s8;8P~foXf!eeB*?RXkP&r9L48NK;hXjMd@dIRpr$b5I)kEwDW+{%m z4at?m;9d1p{)q*cC@h}-b>ey*=0xR4=VyWC3pNs{Z66huO9*#_2kNz@FU(kU@ zUHmZ9P@_M)fPS-vGQrD7PYJLXbJ6t$|5Uk@^|z1HEN`0kxi z?m!RL-C8qzI8PF&NEWTR2&}|LHFv43FTCd0unca9owl)_hP1HoTiFYDA34E02DX*& zKht=8Xc;wm1X-{~#&c9<^4n_|w3Y$e)nDsv-2YXPSmS9-iDMC+dx*np7$jk`U}mxZ zD^SA7xj~2j)iC*pa~~t5@vUh%mAd|{MH9SBJz$G>VT22kxL?Kz{-1ngjBhs zob*O~^4rRRWC$CvLjU_BjUU6n3iN!`i(^1mzQiiUUCj>q1)$~j*>R33=?jPjE-?;~x(^05E$Z3yQ(g=PXR<21+id57=j1L&mx9hpMU zQ7G>kax{c_estAFAPO}%K~Oc1>#A|)dc@7~LbH+PrOxotHzN&68uSpJ1{$RMsX0z9r1ay^4UaBMc0D=bVJcE|o_qgn7Heo#ZY~H3C)#BUC$~m=K zI=sJ(x=Hw||A?QXqgrvn0DJCF!JGO5zwPU&7Exw6ggH>GEVAI}Cw_|_JbPDQIpyiu z0knC10x1_#>!3Q{dbdB4Gwv%<5+ZeyLbXv;4umb9+_ijAwhAMlM@&%R){Q=~Mh0Vk zL8yJ!NdJecbV_IB7G~IR1B&31UkFAx4_N$Ltx2|oMVAKd=`s-P1N`%p0GX(zZgq@G z6|-~_pPEk&APZ$7`?zd{OeL}1>#qz3i_!-4i-SI^cNfU=wQ0M6VT`5YDYn3+I0$6Eq(i>%F+w_W|*nzmy-Lg+HVKpE8F@d9mZVz zW`+C)xuy^&b;rt;YCC>F^aR=T!^K&wcXgX|1qm;cfYs%|)nv0&HA~+kzA)^TmRz}afUMUC-H&o2{R;#`Pf%<`r z?n}cu1n?c@2)K-Vz9y0f+XcgsP3sP^&|Z=A z^4{G|(~i5Lb;f4*bDQ<@l z%%VN&@$jlbFIbu_pcXSyBDN_Eh%FU60a2+?L|nwL3-0yEv2GIxQ~{$+eRIzrbZTsi2YE&iu zR}Jx{wnb^isaKZ2YF|DBHB9=c3LAM#pbx5bS ztg4NcK&Qni-Hku7PSjerm3~&livv0;{#w%Y-M3<73@Sldi#<98{7cgDEp?)aRVpu8 zUgoM454JZh^pSCew(FbLXTgyWvvq5ul3q%Dj=eIdN_EMJbvX6#b2Kw`s8Y4o;{@XT zhB|}`KNh@hmv4c$JeXo}_CSsNK=gE#1SyNP0TcB;kJ1dMDX-93tx7k#hw-Mg-+k0o zxrr^rjsOT1O&8|kjHRQPo;i*$+VQ0^c^$g?Wy3(@c(WQjXh@arvJ6W7wRenGL78Mf zpo5=l0Da8b>sCg@LsfQZHhvJvw*roLih6o}qanw=;NIZqe8B0i+VWG3AZ?1bVN+<( z8#=$fz4kp~-Qlp4D5-V-DcDX*Geo+8PJBiUQbbqA<;&IM97~&;pPM2}htKw@UiC?~ zRBCpbxAe-c{$dg97Gx%0X;*zuKm;sa^vLKo zI=nUPtKfM1ezLJPyC+9cFFgwliu~op>40GE=LT0EhvE{^y}=E>9h{C>AGaZWmbF>c z&!QPCmwH%ZB+6553x;)pjbJ+#H_19{#4w81HA=~H;W|c5{Xn7JRaI32dXJd;jss1M z3YYJXRkj3T5?jimVwN2Qu?4ga^)tzjWL^kvkWT@ju6hr&#*g7(Z~zvoC8(5K!0E9B*w|8u{|qo082T0QJBj+7<+%^ z%_~=e;Lgp`R%r|8+O^)5UFKP+oiZ4UKASpWjnljALRYL`VYI-f%gKTNLUbiwN(K4> ztdeVH6`LaU0lI~KAh1io9|PIt+BlQM;E$N$1n;?c4V+imJxRmjz*zjV4)G4b2HO~2 zrzgz)lQ_5s(eX$;Xr|3GV~Cs(>oy@o^w5?HVYkyaKsg3vXc$DqC$~kJZ02 zOiBELL@8s1o8$MR{@}imM~r@Jf2P$_Ezm9gAwub5^ZEMt4WNPMQ@r{+-aDT&>UcvA z-Amw2$lour!|%#?HZ>o0lwjEI1XX?P`+L-6LH%@|77yJBY*3h!5a|wY#`OxJp=)xI z^En)6hh^V7{q6f6)?!+wjl4s*w+!7J{fse5i*#ZC)}fCo-k#TayX;JN|IIXz`N&hL zNDM+=ED50tQ{WOTF06lUK<09P0@c>B+-qSxp|+__RaC;0oSO`cAV(VK^c`&hpW`3k zl2(yP>uQSss7a(m;F0KRQ~6QLk)%qcF6NY7+l4>zQpJ*@;k_CU%jsbwZL{thUEAXi za;5lsFX(E|?M%0`TzL(K-Am^h!}Ifahe9~9HJfc3C@>}dvPiwkN1(0M@LdDJcirP@ zanoAfjm{$qZ^2xjW&NpOz9*llg|qydQhYvm8(nAGewC?w(&A{+|KDR{PAdY{hH*oiWAuqD)DI!p&gpt@u(el+>B5CSe;;UT`&0N|X;+W*eqL_? ziu-C!9FQ#owqH#I2W!qaA>tp0#9k|(d;-RHY0u_Imq~L9LJY@W0;cU>T35>H5qaLV z#G}6?quAQpZL;`vh3_!(@GbFyt7{*zW|3XTl$1 zpG)cLDwM`kcBC*t?g*uS+d+vRlRcgTvbj2z4m53YiZNisa*Ww*v^0meus`r;xY-^l z|FYf|yQlnt(f+q*gZw-6#RPEVXfAj+-%$AZ^V`^fK3WxJ;SC^v_Mr7$BmhI^`X*31 zZzODaCfa6O^-OLqEP@_gU9?i1TD}UCwtwv#T@Z$}@5FaQYtiqhP6bR(1PS+o859B$ zMPl#uryYnI@P04q*ueT*$}IPv?RV4IK<8|ln#~DMNF@FSh3a#zMskj`;R7Qc)Yi9y zwhhg-{!q{Avrcdg3#0*=0DXn-Bh|jMKW!dQU3TPBpFciuD=*>b8Y#)# zG>#&*0hh06;7&;!ypaDLCAW*KXUT4Y&)ISlhEaqCB74JV9en@K z_JAK5ZW8ZI2yN@F>X954oN6W4e&FRDC$ zWh>Q5)&Fut!^*d0`FNx5A)0S)Swplgx>h{-k!~q_a!oK)5DosL*Y!VgdFvgDbY zbN>6mN9s~;=o}@Gn*A*bG^UNA;PO=;h(HqJv!k+%5_7K#V;;p6yLEv@?LGEST|VLu zM3+`!+!!pd!LQB?4n+-KsGUxS;|5XIpj?;XL|rbb48;f2m)*rk(nT$#Ent=}iA>+u z$7%pCICPn=6{mmXv^62bDTm@iVh-^&G<(Vnn4tA8_*9m;_8LP2!UbCmbNo6vO*X)w z?jD>^XEeKXXY!5&L`Lx}g=z^R{F{Y{+wXZvvIxf4(_qI-o|kJPo6_h25&tIT#J$gZ zhPj*kM`Wt|u8^OmAn_|3@ExbPGX=zP@C;vS8(p*c7}KJfiOAhEZM#1egbO0vh7*pOpb zRK?#yBAdH7UM*q#CiI4*RfigpUrvCmvQ3t-b14LHv#u&n%wk{}DgyOg*Q~_vBe*(5 z3|=y59?G+2ZLBcWuy4pc1$_|SPP})1KAj>9Svl^#IF!XewS7Tobk3Vd{u)Pc@4rsP zT)IXZY)_J}zrK2X>vo~R3$1`40@W$RF+%n!i`0x`^TeFyhOOT?z_<1^g6~`wT5h)- zglnE|pmR*4Tb)xIY9`eKX0 zY_9H7p+yM`6~jwq#p?{n`>qNXCb*?`SZTtX4b*Ro#(HNIEY{k@s>%2DF?Afs5X?fT zE>HM~P~?=MCH_$ai3DC!RsGd|W;iUe?I-n-qa5T9efOuiR5$hMqwN{t1=cy0k;L&T zp3olP2^cnU;K75~QekhFnx6IcuBgbgt&B39lt z$=BIhO;Us6$?Ji+?#0}J%4~(qce%2I(l-HU{?K`6HFL(HM{cvavm<)n?^T@MZhYJX zVU^sB%86J^OnukTs&rCx*dt!F!)v8(wqdf!o0nVr^?Tv+^w)3uH5#QxJu8Ag0{htm z2*tG!wxiW{lV%iXjCkAL@u$_4w3m-qIfS2F4{>zrFjFKIPokR6{3kr1 zC%o9|Z7&IS9Pi6qt=fzcGf;m=Hrl{Q@lZ#`x^AMVYW=#bIyYzur;2sq{BuQHh21=|63f_%3zHkTZ*a3J6RV@L@B#X0gbH_ zdGA^#2hvgP6u;%#HhpE{4#ZIwfo-+Y@!RQ_D%woyMO7@3Rc}v*+_EdxIpYAr!&$jj z7S}!-ehb<8yKQvS`KK%9>>8(A{=rJps@+btr23(pa$pH0dLBfNiMgP=wcTB3V9b=O z`(^czS|?ER{c$V6T_$HUoDZYjQj{fh)ESlx4y{Y2((i@KyV2rZVazq9=*+v34cN_d7j?OV# zsn&CBVO|oBd9<1h$l4cQ7P9^m*Y|$HMh+?m$QdiAZQ$ z^|~v^@y7w2i{xXN1?Nm0zDur!i!zY`w&mViC z_Q5_Rr>&s%{U%TQpYe#|ah7rgG6ki)zdcK+GGUjC&>E_xnNxvbczcOCy!JAg9$k7V zKDI=Qic}8ep%UcvjakTi3Q%j&&x>nMwgvM-D$MXKt?DbI;JOLgjycR)>huX23vP4wtTl_TW({KRiLd`6_j#E6n5hJ7U*dy3ykmjUDG?A~627IsCGI^69 zrW)ccWt{>m*=J5%`h={bPr>`P)C&Y>ixs+X_OTuOJeZVX$qw}h_Rn$E6e_zY&Pd2D z3kV)u=_Za068FbxJ$~3Aak-nTd5`@A3gxCoOy-Qd}tEZ?-`z zCtRc6^b{@GDKs}Io-Zx-8`*{n;BUScGa>Oj4jVu?`K{ue!`87U(|?t@zt;QzdkDR3 zx;vBCJJLq;{c>@@zVBS$`pL2An;9{7x=R)3Y(J`1`gT$w=nwocETZqt`U8FvetbLs z`~=-TlSLhlC+|)hI8v6^X9L*^pd0*QqK>e zx~g~xXPju2&`6KX$w8BpV?8llr9uAas=_z8fXI>Lwxd6R$Gh)!^;VBd-PXDeX(RqB zF~ccTBckL-{^;ZWvs(I&^amzw;K?q=jUfv7iI4gx*h@eO@WGSbDP}R$lg%~@(KF!6 zVFN37ddbDAv+q6rVmsQMyji@%*{VMM-l2ch)mfBu*CgznG6p0e^i0QRF}2$F%xGpg048i4PJWG&r%bgN;=5(J&*O^~ba48?m&rt{f6&&NWD6K_Rs!i$wM5v^fXu)1BW5cij(^M!T2UK@Ib92$bqX9Q^g6 zzfhi<5%7d5)d+>HRlF~@w|d~4TzHS-t@W>}hk7mll^?B%ccU1{%E*Osk+J#@lN~kC z!P4xdsqGQEU@CyE?DQoes->A)-eO~_1B+QhUiO6}q!&e|o`dW)*L0eQK@|aRCP-7> zi^>UOmF4KQ;$aE-S)TvLH`D(|{bkpwUl0uQc5)Qook`R7cm8U92`H|>Qo9ZpNZCx4 z+pTxr{k5ikTlj0`RA}~RR70l8TWyr8H{xxv7mjlI~Bl^!dHDd z?p?rRb)AL`1Jf#jF(zf+HHTV5W&LiQwuZ#OO$w)t*KycR6I&Z|h-MM2>Dr_yZhl-u zV$`@95@6Yhv^l?YVLO((R%JAq(<$T*zxpTW9uv?LKA%An@vq&_*6Lh+PxU*Zae_Y9T;+G{G{Iy+m?ZGhpf=BCCJf*Tp+S7REy2qmscJGRc9-ggQ!j$punGT3bpaZUTJJ117{Ej&Zx00dU~DHa&04O-8KEM>rX0cM8Yj~)VP{39n%DT!@ia8?l29Vo2Lk@9dq)TE!&r_w-ORKUMjxv}G#z~AWBx_{?{)H- zmoQ%n#%x%_<>3e@UgIXQ;x^2oU}`V7@^6kAHC)rVbh?razmcE@CYq1lIrk5#V5=$7 z&|Jp_8vVU&iHb89oS{l(O)DS=TAh6O%Y;>avNnWjZ;CV$B~XEebmX;{6?f`vF!#&g z5ott!$}yeNcOOStPJ-ZEUIWY)ygim4LNuzt#N#_BOF#^W7?H%lB#^6!wTeZ1f}RW5~43;)pyZ(ZEyuJIiAM6@7-8BwE%sS*k;JFT3Fp7xx?&hxEKd zL~aQmlpGNWm()D>_wBzq@y>E{mT0G!IHUfDGB8p)K7?{$^rYUehn3TYWhFi)kLU>s zTr7$ps4HZ3XfJ>wk75_SpR()oW21%QvXUiVA{$Vqq zd?i1DoG4kK;8FpKjVdh61h|}#6Tz|hU4^Z7qOm`rLcZEzip;4DDm zeARRr+HGgvT2>NEyEV~WR~cX%g@Ja@>d8FzNzI=iYdcrTsG{r0A1YK>)w9KAekm(G z3~mM<0~jTp?FgWOb)}^!l3@NKjT5|*Ap)ElO)kN%l~ZQMl7M^MQJrL`Yu5L5tLOA(J+G9avR0V}74~og#%?jN8X+_gMF+ zLRCxiJuO32ONCyavOVXO4@;$~4mF*+L81i)ZX)*WyDji5vW}hU^@}9ReAzEYk4Xl| z)?CyYeCUT~>IZasHUjNoF+mG=1#d0!2As^bjDB`f<@^j=_f1LPmpjDP2uD%HG560L$6Q9Gg zA<k-@B1;D85g7p*26= z8#RhqK%!PoB`>=9Wl)5f2g8*8Hu`KR`)j+WP0B|r&zIW`!Ry`}9$6+`IzMQ2d;TN+ zCt}x(yq;Se*d1@L>te?LdDHDo}+mS^;U`Z7n8~KFON6k>6Cg3 zT8wIrx_h|ZIh5)d+r;Nbvb7>7c1HR4(X{m4gZzbwMd|x~)Wx*>h5Y~LIn{!rCS;1?#_YI=FWjyp8dbMPG#cQU;p=V195D41999x#s5=6TLW?8m^b_>1kdL>-`lNpe4^qod=WTi*cUZ;Dp3R0{V?;hZm- z2;{%v`3TYGWqaofh7JZk#T*CR`dV> zwhMZdxZXaSj%9Mx*a0!{$K%eA^g$GZt!0B`N};4$z#}-S=Dl1?_T+a)W+-(Gbq%j| zY)eQ7OzP;@;NyONOUZ~RHd|n9&16gCpnQSQXXPRpmlG}~z%XhF!`gPblrSP0qpXsF zJsx51Dg8D*kyXN>C^N@`+xXW&K{TdNrZb<6IGQsBlUibxxc)>lO!~~>Vi8@~b>y(T zRlVLiD@_FK$TWyr2alp9mh-C}8N6P9Uc%q(^#3vvMS(zYfN|Ft9~)P&amX`I1|;W1B) zExP2d2L0&N9?0;8EG1wFz--KOyY=|OhDb2J-I-9>&3S#)gf$&1TBazkJ~&zitv~#J zz}+&Pf?xuG$I?lX;FqI_c})kE!{YsAlGLB~uYz#Z#KXgIu&5*>#nX}AW$45itRe+flQWA55_pV*> zZmjd%l@t!qd-LW`lRB{P#tI$4UPQseDr#*_EVW(Av14y)}BEa!PIEu>%ZN` z1}k|g%pYd|dP?&-QEuAt;HcF810C;U$cxqy<7Uei z1%6i&_kbEl2$Ly`1+lVv0U1pt!bg6iCl8^jmxNlkz*^f?tC{ye-a(bKGF4xk0H8Gn ze$Ft&DTBjP1;`kB6-kptzTo1}#~J9odg@=R{3vEt>K_0k13=A zNj!Im2!*~6>c-;QguL+%%!Px!==YIffLl!rW@nI}5;zB3frQY|f9*k5Is!mH*a(Af zFrzbNKgLOQPz)WPi7yfT{t9>RAz=ZLQ&-kE^^PynRdo@m&P`55jEkg(c4V7wHAF5d zrrT!@-SJ5amSrm8>zUqn)oq{+&z%?_N^dtg2IZ&5$}IHy_PDm6nqz!;V5S8xa`Aw=Jv_Gjfl#8y>eOC>a+><3}IW^yoj%J#ajX z=>70ZUCUKImA6Y<{YC$I^MZe%WfG8}ui-KK$YPn-(t^#c^6Nb# zKnbxeDmVY53~~p179xML;ca{31N2aOP0b8Lsg-|XqNIl1p{cpA^x<;IA7ii#BZ=(C zrXTdUPAa)&TZ7y96Xs|CuQkY?95D@sgL0oKcL7OwGws344e-4&f6~$kwWwV&t@5sW ziV?TQ95uZMD&?vb$a9*azO4mN{eiHitcL~Oyo}7!cw(#U+Bh%=ws(l^0?8OoO+zD2 z9l^4cLhH=%X5{^~kIWD6AdVbbSS+u#+9QU>V$1Kd@&ys|Z#QG!6zY~Uq#MpLSD>8DOk zRxOm}wdW`^Xe>{nbzcSWhKq3|Ygr?93wHIQ+~!=1>s=PGsk-?agfmECvdj$h1Q5xI z;=}G;MQmw?gJd8WS?GYrKeto;W>DTbw2uxZ3U=8hPFT zmtgPcNzcwaVl(j*o=l|JD;b#ao1!XLlSdL=O!gw*A)0B|s}3Bt-qCBlihw^X9vXDR zX9-$`Fm1hF)LSf}*!X%6$+cB}jBB7(iXe5sQoL2t46Fd&?3Q4DS$6O4KT2iOdcM4o z|I8nSw;&!m-+PArc7W)Sw5<%%K3PH7UH&J9&;ogDw_#8lZ1zu5=-~^gqzi&*95u z$k5h3rX0{nUftxCXcgqk_d*~Owoq6~nrYUcsu)7r7EnVGWdR>*W=LiD@vH20PUZ6v zhPlGZA7at@qRcAS6li7HsXMo~u>rbp*%$O`7}Puci#8*XTl%3{^&)z5wzrC?rU zK^!bepoBU2%@`)^ATt7B=7|+buxP_b!}kU{d%rKBkDN{e?1?8 zdfcO(1hxCbA`bUwvL);L+$&}N@r9rNkx=KqRtGs)wz$oc%uz(#HCl~C*A?MQV(Ad>$;L>&JOzTh=vHTMD>#)ZW3{eDeLYO6j_?c z9)1r#Q62MdqZc3AI2IqEh8|C}P-C*XJX(q;vU{FTMrI|j7aD@26A2K&21n~zRzY0{ zYHHL?1@3W26KKx%YyHCqI8pV`w0#bgQqu{tR@QZKko|ds!kJunkP*NPCDVxt6ElFj z{##gXec0SSiPGHQd-xna^OIqpH-}!#%Lc2TE*&_d=uT_eX94mo-;IklT!hmpF&;N7 z@BG`ge)dSOwOrzwj6Eh0aw-xyS#Zl~haHSj*{o?Mr!nYsDMKLOetgou1a7FR3C_d? zGmQL0=dA@4>!c>1SVz?BR!*L1E@uaJwH=s648iP@fNt@`}=0pRn>Um2P9 ztX$)#owmKR)`JAkAtzqLaMYl%m`BWC{4|cC{<27a6v-B0BoCI}-^bbMiHgS~CGewE zu3rChyT7EkmCX0B>n*4nQ}IB`<4g?%vnw}kW%FMH|C9-kN`Zg&vhv1HUf=+x{)9T3}bi-e3Kc;JoAm8piUTGMVpMgc#kY!Xi$-WZq zYYG7eJRSyA!8o&Np3s(aFB`V86(<5gfr)wQ*S{Wk+ckcxVutPJXO7e}e_Zf__f{C& zbvejd5xO7~J$cA$8OKgp0HV5uc5oFRIz})faF>croMbRP;$ABJE-opVlLoXi8|5`> zH_2nXiHgbd#w((t|Gled?=EI$jR|t}?qUIPpoIM`mfsu!=IS2)g8uZlVLw?rz|v?B z(r4S$-_v{JibF}}!c2!r0Hi2Fev16actl=3sAxNuq@(TU8%v!L?~X~&Ha+ZA)6ZBD zDPq`H+qD@ZHj3rL3>%jpkN%}tZ%gh;al~P{{DI1OtU~I>ygy#KIQg~dZz*fmt76K1 z2u2iTK1$j4bUsS^&mGK!S?0O_0VxFz!Pm&-9m08^Q}XLHX;|E78SBqRB+|L9g5sC5 zNjbevG3*}wQ0)N3gW;1TB1xpH2HWezN4GS`W9!nUv*7#EDfz|nG%H|2JW z1mHr{Ye#uF0h!IQV|PWs29jlMi8r1uoE;DM7vG6*eSl*V-Yw&=tLd9}&s1cCjxM-P zU3`txB|Ds4pL-$m?nmo9Fm(2V{H`$Z5`~olJBI!qDUv9u%a}QiX0odH>|qT4hMh@{ za218Hjrd>Iie3yBJn45;z*NFGvs zQKk*2BIF3u-YK{W)HlHc3HvZ-IJ_FOoynXq@M*j1SS$(8Zdc@KPv$9SU5es4MF+kH znnIZUMgA3VY+40RCwK)0js&_URd5@=he;h>DkZ%)F%<6+1Fd-3bYabV`5<efiEO@M?YGdE=7*$-Kg=1)L1|*3Xxmd8Nfc82rytbg&wk?|AC>JES>2X4lv2gEj6KjNSsK^v2yn^`(tt=-_OG9yELR) zj@@nngV0a|TqOwXTG9W|cyeyU;6l4`3yXE2A<>Z-&-!#s;e4r@{etGq*hFkE3bzML)`Cl1T6V zhZawrUTv0^qD8jgsiecsr#_FJ^Pv#3s~9O?uopfScyu6~RfpH=(GkZ9U0!z#zLyEw z9o<{cI4(6kt{&<3KFv2(0nfYK|8Vizl{|sTkEXsB66Fx(?cdcU9T8I3DL(%<*iq+F zOeLU>g?T?Gh?beu?)oYiuPO&GK=nmG?4mAUZV(3lK}9#>x{Uo!w5_T>_*e6xs^hQo zG{;RI;BAqM4O0*0!sYf;t;)@)L*FK%7abvO)d;YW+WZ{Kj}o zo_;GDj6=BJh+@YzN5EIP<>jl?Cz8t^mx6Zok3^yR`yLUt;Y7UXP1mYrzg=o7BKUF! z&L0&u(9@&5l)Asrwd*CtkY7W{-9l0WcQF|p?vh3(<^%hk!wQ}Pen2B{&YsgCTw9Rg z?Jb~P{kwSg9^#P7ElGJLNv4c%MO{01*K`!?E>_pLY3*MOr~+SZHCb`+86#pqi56p{ zEX%m5v^BEdZUrD&$L{s#Je(`eoJWG@?y{t>Zv~6FciP|`r`g=;$lwKjZI7z}XYl&) z+|GY?7#T7D%Xc-Z86Wuj5Q^Zb1%v6v>XLb|j%FsL zg(1ZFqf@7^3C(D;fZlK2KV5KZ+l`0pmHO$xw3P})yck*&;$I^vF6hi3e_p?(mf}3vy(T{_^^own2#Cvd&RYi zENV0Wm=YD`fe~I!3kyrANfk;sXk7<;(%Dsz8NaaA2l_*N^HH6g(De-TO$aJt)-hRo z3>^sSfY5b}{q-cf*(hijK{n)WszF~bJVdW%lZS3iM%WdJ)AWaII&>eE8NIjwGOGf-DB@hfa4G-SCwq8;RDx-8WTGS-wPpa z8O^vlZ&{1e`L0ZJN%w=Q{?}tFhSjeVC&IMWw`MhDo~;-6nl3-cOx({Z&_O32Dl+rH z#prpu-6rU!F{|&lrN%GtlfQKK+lF2`569Zvg0Hy-dLoyL{dP$@z^B6973{hbRtBDi z;QW+g2gtV0(MeUf(Msy>=-s>ZT{STeTQ%X}GoH>M?)AHx&KqRt4-^vr>M~k0dw)ls zS9zyvkzdLj&V;8ouOVBmnfE8iknT^Pohl4tZJJuVnzSioBOQ6FuN?H>+qj`3Q+XeM zYmt{Y3*?Kf#CBu6fo?@gig&hNneZM)J!kIzA+r(%%-QKKBo4Cae zt+%_pJm5uJI!355;|$w(O(0Q*EXq_#_k1024=6V+8&on20X9`e4FzGpw(CrWh>Y_Z z!AfNbd6KHVu&-jg7pPhzSS%P=V-98Au0vb7Z29pds`noHb8KeW1#*Kzz0Nf+eOf&f!7NlygG7#d-<`MNB&o4z##^XQtXOeU@ z!X@YgFgv`&E&^u@4~x^+-V5h`p^+vqBZ)I9hfYGmja#3ky+xyM0Oj~RSZ309`Iv={ zuT)T~pvNd1XtDe{3B_ji2A!0d#3T@l|v>{E?r5uRsKqT0=C@NJJmt zi(J;>1^g^+%`TIv+z#S?`Off9rVC^rxx5=}>ykj+uWQ95ngSSQ!q{c^=1V5l)%h^x4`CA#j6E5U@9V!t4J}X3c zT^Ys(=dZBlO=G)hx0A!hI zyyS>DADiFs-~ryh_>4Qgd-1q>fglwHdF4PRb63{snT1#EE(S+K{wB4d%C_U4@T73C zdUbTf~Y-}w<$d=-S`1cMo`Ck5pQkNq4OOpEmpI+brM#-Fs*ZqXtqHUdKD>*BX z0(fsGI0urD8vS`ONl_1U>X68+!g%jiaUQTfQbMKd+dk%yWfW7rx=DT>8rBelofxO` z_TVjIB^2A3@M@tBJ;gVOq;R0zLn{e-e5G}3tcnVrs#;UAK#sglA-3u~gb?WeEJi#i zSHJY9m7~Fr2-bto4a%=wv<{Ov_4Y3b{=^I%9Z^$be`veZ2# zSGI(?O~%09YL^Frs50)&`0>Js@Bu~4CFC@(PmIK_cB9wecWB64&^XVy5C)tpCg&4J zdA>b?^1Op})Q~B})-qYtOvV+J#t_~?RRt^AEj%M8{*r-F@NDm>+gOw-l1SihQjNk% z_PM{}`6PK6YJS#(_fw61{u46as&vYrRyoJsFksijX$NtrUrvxiZ?`qG{+p&7#4)eQ>lc+ z^kryhLr0=kXDeR$s29C2uS1b6NAKbEev>#aY^Wf5v1y)4&m8>)0gEV_UBEZ zCgG6F_D$gm`#JHhc#rXbcoBVrg1Fxefy~?+^RD_!_WVh1gjIrnXGfSOM zUg+pz--r#^Mz`}c#uY#%=OlQ!}3~~l%_Ju6O zfEI!~{AX`bDlQ8Py7R>quUD$1bIp81Ir!XctwGtDAxNyU*Z=^>q$-#ZU|t(AW)buF zyf0$4e0UK((69}vZ?xg15vpTp^=N`6gu^h`7PU`vl1(kiXLb)G;InKmt?2H#PTyP@ z;zxQxCs#QQ8oJj2r?^LehOXX8*;tI3n!z*3=|^jpZUh%R98#unPw5o2T&sXsHm08c zOkNfG&PsY&xBDfm&k%xX5C*P z*i>aqFB3nd7SBEaDwWjG&s)9pxh^2GTJa%6>I{+7aRTJyTns<6%YEQl{Xo5Hy8jlw zI-|l2Ff<*H&5;jv55f_O4$gXgbA4(Umur=9*-k1sXlSJn^6}iJPSwy*1t;zxFwLY^ zd&m?b!95s0!%K2aG_74}UL3l9jft+-ZV>9w89PE2Y*2G_A^-bZ+Wf;5I@{e)+0pPH z+UPaka0}33HPzT2%%El)X;qlebsMJu_4;Q!O6^Ujt6XR~(jHB~XsLJ;I!d-;Tsq^O zBf>dKh(FiIlL2hI2bY7DHPXL#x>&B{714cz0fZ%xxn1@`M@6c%)Xw-Ed;cPDjGN~a zPX`!*l$o6l^TdyOvbPf z+b|E*ZU|4~?mr)cX(M5RC2Kn~R7L~|hJ&DeTBIVROzG8rDE@@8XfUqbdF+@SO(J zs9gN(vF+PP(z4U@@#W)nv+O@9VIS@6O1CraWjUWaT|cc93HVMSvuH#g`>E0-StLLd zst2cF{DY|^I83?px94$@5YKgEv(x_BO%`DCCQ{KSN{G$It57+m-m=6m{Vsjt_@b%A ztDxHG2z^4(`eS$EG=~ju#3>%wBB_b2?RCQ?NUejReeadtx+XU>d*Dpeynlq^L&7{J z4`$Ok^JgsAy0l)$e3l5J+84QbPhtaqUKaEI#&8Vwy;4jKCx{$#0U^kXjD4{d+FAmJ zeaT+mn0C>&WA|f8{t!ZrKGgx{GyK+zT1OK>U5nKd@wA-nde8{NzrEh-r<*!Awdj8J zIMH7I8}z@45dRH?SZa&s`H!ufsL1d?U=J+q@&AwL5Sq~*|4#}3l=x3c|CIbsDXkgp zslVg?%XoGd>g6cL00MfG0|Mgvzs56u>g{DFT<-Tw7bwpEbf591(lh}7iq(8LeP+&X zblzD2UqUt>RB%r#;qE`dNWJ@`JlG}=B70{mDkW~d-h{SS_nm#Ep;v(i2YNqk_Jne( zWIpMVUGGoq4-%su=ak$ueD;wyrdY54E+foB4)nz}_59mqnP3MdRYfwc{n(u!=`Q8; zG1PcjmEu8F1{qVX9|QbwB<&-8b*Ym0w4oQ`NYR?*1s12cUn+;}jcwt8RAlKiY`_l> zsiGHEf+do*1EC43ps{r_gjA$^(Ah0xSOd$m6Mf`jd)U4V{K2bS>(DgmoDo_P7j;gA z7xrJv0;EdW18<%^RY=7+aHgR?>{mYrYt8wJgW+irf8gP&DFn)pf9)c$^hsKfg?-GC zY>|mtV-jVtju}B#>TnNHrHhUgeSj3Adk~&gp}<~)1}R3pgdnGVm#7fAs3Pi)l%f+~ z24>~n*Sg=r$z}ln_E5vi?e*S33pBtQ3Rh+*%yh^?CvM=cxD3ot2T$Ub-oK?kTvxS0 zK1^`)r__v5%KCf8=byT#`*yiLX7Q@RIXUT@Q2J;kgcF4(A{_o5 z+#c^9@oqgeEM}QIq}~bf)j2{$DX?-=roOuA@<;T_dR3uhD`ZNwCnb`+Y9z*@r1+TR zca3vk>I)XKvvfc}*;yqRVrr>OnjXM^pnHorg!(L*5^}#>(_F#KI7*Odfe|A9t%LCtE`s` z3+O7d1w`F8i`QD%XlE455cW!7&P98`vvgoqKmYQ<_C?A7XjikR#4F?$^RC)+MP|xJ zHWL{gPO^+4RC;;C$Ao2X@VGJeks)7xA(5nt+HS8y&s?{S=^x&@N<$NthVdZwaxKD9pg8yiu;)8)AR@j|n9QjhDL8 zt-tz=Vv#Lh8VrHA#OhhiH_&5F^$$$GviMGySY98G*bRCi-u8{CuKoRQO(QMf$AWEm z7Fh?a_Dldbg&qg|Kx&6(qEy(=i^Z1_gslPa`AL^psUf3%)DT}+s=Gg z%#2Lo=*6h5wI4#Xr?;mYie265MrX(TX+ut}r8YMBoMz}+ylxTij`!vY8LR>Sj--cQ z@f{Z{k~8wVqT*U*^#D8xkeDzmH;JNNiyyn8nV8^aNam0+!ubql0$Ia^Rtd_1x=%IE zJ%?ySn}^{GB9gMDM451``u`2qBA3kCNRe`9iXS8bgTX<}Q&D~fs(Ufa886AG)%I z;*d({R?C18`2c?9b=sPsL;yZ8FjT+=bXOZu$;kfb5SK!b3ba33)k zi!y;%x504lQUPQmHm5>BF9IPyr2Q2Ao~6hcL-sBU&jNR#8q@`5r&>sOisD1D_hF6p zcONIc9gh`o7JSBTDo4j&Lo{Rz4z!{ zz?KIG`z;Psilu)DC6LExL`D!9RDv%@5p{LG+BY?Y{tG(SsjwG{isv1(HB=$(6&wI+ zgWz5VOxfR;@#&Ad1TB%dW?t%x+Y?_7R{aM$O}vM24KGq=wCjPUZ0$;vFYzEU(zwEx zXhBG%Jtc)oNd=I_PG20wLGo>zJ zC9WdszLbik2id$*R7|1DqlKYF<_8e%!I11Uk|gA0Go$Kg+0Xo&cS7ZN9SYk-B60MM z+nTT8U3{#M2Lh$HFpnKDGGiMqnoZs~Qf`OGS(qQOKRP<;2q8nuTouCWaaZ<0813fw z6xj&|N>EUhF*>E(#d7+*R6dd^JM0T^fUT!eZQkub(ri}R+}BDrN0~Fj1Tw&o;V2h| z!>fsAQS~O71P70cnhy{N{a3LtyFwcp!eTO25k{23S#aY|qH-($H=3xl@p|_NTn*E% zQ%y&Q5L?F3*8NiV?J?Ec%s*C~A)gwA>V#V($K9O$OHAig^!^a5VgUs#Ia}d0KgSmT z`3nMjp{?!A^L#)wz}(_pGyyPRwgaTdR?ZrD_v^~r>K~KqqB&;h0dt?^>%5r@KXIo_ zHhC15QZEyQN3&8aDZFu`fR8Fxo)97na$pmxv!+5%d|Q~62kX6ZnVKym{PCjHGx$UL z_YO@)Zx{$7+v2OMn={KLLDavmX3HHHQ!#2PX6pec>jO$kAYtCUKmZ=7CSvT-bh@?! zhrPj{)m<;vy@eVnHuOHGw$I3-665iP5@ zs!8amH<~J0CyOeD!a49-7}5ThwUjJosWd=$mW2Q!ssWZ76)-6_7#=)4DGmbmsUX)) z5yhq{R*HAg6wu(@6puORU?-zD65i$$tebB^f7P9M=)gW~31Oc7+h-&hp*8~aV3P~v zgO&q=2U1*LB?n(|o~@MA348No$B=_gcTBDS%h%-8z*4N&-;gu)OsmsSB z^`ccL1q_Y!(@CNBQ!Te?*;=G&tYHaL$fe6~(RYD^-T^T#=y%)0mF()J?3mGF^$vEu zH4iKH>iu!b+c$NCb`ewD)HUrjed|`#8Pm?Ho`pH*uYxMwBtd$Csu2Q}S@K45Bxc`}?V&J5{hmxin~( zGWdDk=l~C+N#+xY3@JU(C$it1vf)bF)29o{h!x0cWv@@A74G>HH8U9(zRA5N6tnv# z*BBFU@PQ^+lS}#vo@HF?0?(i(=rTH*IBcc2P^7Os6A)Kn6n5Hi1GHI*3cOdJI~u*< zBiEKdECrj6?hXD}t&)QqV>p%a=&qsWCX)bcYJko>d;_EExbLxm&^6)P=ts7A*YlJA zP)plJ+Mr9t118Fv-zR3S9ISaV`?0?TEl>5Og@f*W)8Sr6HNH%FPAi<>>8S}@wY8=4 z#w-TxV{86=zX6JMhv$M-GoQ`X3TvD4gZR#%OtIAy($q#qD>rV+S=saVUB5*k)#!J| zC18BQ3neq2>+y4N(A7;?X_ZWQWoK3ce?S5H$wH4H_)alL)cvTIZ%j)O%rc^<=*|#0 z@s>cD3m-Fj-$EN~22tr7C13@FQ8V7FBANNLr{sDIo8`nN<4b32-b7frP2&+33O(Ji zDNLtOEU+@W7^+=ijtpv9!svFw->rTm0?=v3OYh_^a&rnq$=HD1<~r+Y7kMxkEti-A zx#jObxzJoJVzUwQChQucUntE+-!(p)y>Ny%>Fag+g-u#xx>jKj}!McFCmP)1ZL!fRtLw^gt$}Bq#0)(S5{_V~bhd6jAAVv81_fM;2`}1NK%kv(G?njh>R1gxhLC%zony6ZN{9@JH z3Ea&8kw++y6F-DVW_?`NV)ff1?EQho9awfIGJrG6z1|6Ok|G7ZaP;~J00{ljCK_!H zNawEFPl_l(op@3XKZd;G7+3+Iy2B=0_1|>2whMqI_^W?MfxNAa4-$oDOwJF^rfY-2O#ae5J z3{+#<$+srSuvcrOq^o^Nvj%M@n7_KB4e*!zO{&i!YIBSM8X<|Y0Y z8@i#N397n(U)gQ0vbHtFZ|_fTFT`@B&)Kca{99wx zkCwy#M6>#+OPEK`nKTKiqry{edJ|r|o*}fOWyA&+XMB@a8AB-8xZJ>}LvnF1jmpLY zgyV8(_>tF&ZEL+q36Otn2+^8N)dgHU@jjYz_Ujg@Vl<$}e)KMjxaQNjl02Z(I&zLO z)Zy`Na5r3F`~k_m_gVFJSyArEEZ960giKBa{b)cZeXwYI0PAg3OVi7mTKq@U%HUnT zt{7}tz5DR}Dcd24iTN~UJ%3SM#dX=t-M^n#m-YsDx_Eq_1MZ*NIXZnG4mKz4e_R`1 z!O_rj<>c#|t@T-PqB};-W32A@e`q_d>}tdHA<>cb#Pp=I7U1QrL}z$+Wbd?RF)N(@ zV!RO4(#_K8^>(gJ$;p`_U9BU#UaIro#lJ5&-~7$`DfB>hr&_4+&?Ie_A3AHbugU0E z)6NReKcbo`1H4l_`MJ!#z{N`a0{j{a*F=IOu*bd-VU2qZUu$4zhj3n!AqGGxibY++ zP&YHKa$mu&I`mT-tz-H?971C-cObwPYV7zIhN+cqpFg%taaj~qTUhW&c3rFUBQoe` z?Vo*P_WIoQ{;y=bj-c`#?H36U5KTHL5Rmx)e_KecMZqWg&*MevzuICPK%J}Zx}&Be zpM4FBLkm+2^FfEV>PY)mQZ~vjl!Uqx2?Rl{m13o4L%pl%70F$2)W9;q}$ssb!nPW>*BaCoZCCHe)v&j#aC!*I|Cp^|X? zY+4gW<2Ai-9hl(-XgFYfQ>8;ol@QSkdd&n3qpXq>tG*aiAayDW06_^ulTTx0qTMj# zq_hQ*NR@c6O?t$_&}>D(aL90fB{v)6zsf znks%ddRB%`Zf>A!k;68dp^r0Y9tbTJLptW6?)N5M>IfNlz{|6N{L(H6CReB^eA^Z? z3-vv3KvS$SNS@bg=M4f=8u~+9D7P+<+T$}2orlI4tRb1IpDobk%0v3MCl&oTPhhrz z;Z*Vl|K@Hz(05_3&j!SUCRyU6O<=hZHmn8P@!lJYytF|%L-eBe7#qMd;)mD!mV{Y; zgpuVvK6;S{fO7j}=sK|iIS`Tyc`u6gxw%M z2mizC4|sE)Y9+j6didE@yb)&>nIX}n+WxXRcwIqJ*Cq-<`nPT^1MHo6{KXjy02@!U!i+-`g@}N&yPlp;@@yD{Ec2R97VGtY=!ZC1~DcfT)f^)s$EZ`$4M4_x?S~878yoW(u+G zl~c?K6X*evfOD58V%X^{164_XXfo~#`HNZ+VCpCH>(1J$OMwUBLK<2gwZ?Nk>N$0Te9y>5 z%1mM7@KaX-uLaCW|@oalnT^aW+x#Y~`n2fQm%yalnu`pS@n5YCIGdUf&& z04#>z;T1rx1`gN*OUh}w`Ca+tS5Wvxnv0PTDt=v*&HvayKU6}Iou&9e?;s@^1z*33 zv2l(nHfw)nr<=2x=jM2pqsp+V&YYmPel?a(<2pcltA$NPJqnK&zzU$?I2e{kIA70v z4N?`H%M%sifj^x1=625hjmjCW7^L3!$QBN&IxK{*`dF0 zxN`^5GHhln^HZlte<5VextIQW4F7PRMZ~yx$!*=`e>-|rcm4Btj{e~e$M4H!bQAlx z+tcBX^Vr`WLvnk>wNSTEdbt>g%UfEnhvy(4Z-BSskgwOr9xpfN`wOb;=Lk0F2%r~q z8Yn_h>_eNhuT=v2?HR&ZQ26j;0|+{P*xx&R>hG>392BzQPWY7J zE_N@HFd>#eKx*!qS7B2Azolpgi*#pfXck6tHB?jvi2@9RU?52#mF=8U4#m8FW8^m% z%npPm4F)Vh-|$1BW6OZS07B%^GXRJv!^T}513^M6cx*aYk}Mc~U(?I}JK-XOPav9K zjr>*9odZlqX3&CujdgiX>o+VyUiml*ZeqA~7y^FhogpF1XW5ik^E5@HDO1nHzE@ec zo1!23VQD;f>$epFpebk6+UqJNUxVJ7ZC#ZtzjjA@Yg`v25=6%;ZhY6#4REm4Q7gs* zW%3D|oZngD|}ec8#Vl-=J5-XbLzXNdI&U#;>8#DNDx>OzPpc`-4R$B{E5a)gtik zgg^-A{V>`Ywwf^2U+h)ye*k(4gotaiBJ-An{bn3!qTecx0Kl&gc( zski>+zB!bDA!*Vh$s0Qy@F*m#6Few~9pCh9J>Pb48(EdL%5Bmj4Y{Q|?n%LX@CVau z-FEh7!=62~NTG6`f&>KiuaDH!_P1-)_|$?CReC`QRiD`2#039%{stS^Ihp^!$(<50 z0`C#zbWq`h81I(@?2mJD6On_R9vi3~Lu=KKBv+(+*K5Gdyg9?*;gme^UQFHibAozj zC(TFYK2U7n4!VEMMXe$_bopLoJkbtaf;4vW2tvQnd*q>__s`B3aSd4_u;N8 zof?yj9yQ=RaG~Ry1qi=5)6mF5gs{~s&vvrtw}g6i@pC1Pcac z+f9<%oxQubiitUSo+ik|ZvPq#>S)-BQx?h)!8!mG4JE{t!vM!e6wW;Ly*<6G^}Z0vd{qo$KPk`A^o71p5DGKdkv54Bcl|ED8GeZA__?C zfj#{#`z=ZEm0GR)S-aGD1<-poGK(V>9aARC`9@viJq&A-{M~bC!#>DF&mW6$g>=tONokW;fS4e|^lt z4IY5&b0w?F=s>Xm^&OBnnD5m)-AzRQDqNSnCOR|Oc(*A|Gw*9yOjrH_f1Ye3?GvmE z9JOv#G6CXgi%eyE)ZWxK5s!an*}IdpV%FTfM%GeoMYc)>4jUSR8o}wKt$cx~=5+gz zSp!~fkZDNBY3|0rN&6dPVj}Jfyddzc2gB}OHhGAowYG^@L<=GdLHe{J<{A#r$}AvX zu|?s@YPR%U?r_o31s(kPT!j^rRGtvV*<#=ceOwKd1*vuIU>ae_N}|yo*oaPM6Z>mU=Kk*@ zHo`hS>v4V8o|l)VA!^HJR1W|YMau@aaBV%q7ZzU>LRBfM97f|B_)`(%fIV{B1(j3T!sEH>SM11{Px39NEkCUh(RTWg0#=!yLgQL+;1k%FM z;ll7@FsMEo(^hO=V+OXYnGDC*H%nOg3mt)(&rl9bEi34BOlso)6(AwQVbL3por;W0 z&U&jt^s9U(W1I0#Kbib_b;9KkrYZ`O89_pOG4xuHIBb*(9HcMdne<$P!B9hd4`wPX z8)CYziVk6p^)!^N&fyopSCktRc5G=w0vWh>6~}y*I|ka{*qyij@IpRhqY8vp9QKcN zM5#YQPgJwyM{UWWHIEBP(9nB2uj<34@CMB!1;_=u6cYm3>oqL1Q$o-RpWb zMHo9pF$;cy`<$UJt*K(T5=~soS_s5c3@*z~ZDDP+b#*UmO!Xze&hih5_=~;2b8d(F zN3M&zAIn#rX}NcMyj-u>C(pYSRDsCko%fV6!pu0~Fx>Z?Cvaz95Xr5pkU}WQpX8tc zE(pBcyTHu+hhw}yO47{AplZ@s4RjQIgGIV+NOa zztq;By3i*CNP}SviyI?tYGnMKMSUQpCngpo)`onR5TO%*Oe*m~T|gEX3>8swS}hyn zg$DV%EPS)(Ad%dl^2vwgm7BNG$H(PfsV7vLf%Bvxfp-Ik)N|!82&)lhhxgq?_GK$Y zh6=|T&7e{#Aw8R1HDd)zVfbY3rE{Qa@@d=6;3rZ=E<#GAE2E$B!p=BS!ixlBg@P;H z8sQU!mu7hYDbTDRc5F;H&Di#|1b8DIrwK3Mk)+FLS9JJ0gxSRd9CDsHeNoLmKodsT?X8C|x7qVnr+Xil!L9I)<-;18~%-iw}x;mlW|CyTnkrNAV)58bHf&Dx+h1)j8@6Y57}BA9UaRk3stQPs3^Ahp z;ic*8g#3Gf0s5>M=bRKiaS096xX3n{FFPm->PR!p`ZjXfNvf%f8UbX(jdjME(p3(? z*`7-?7!4H;dBT*t@c#yHx;%T1k#K5`CDXKrbpGRmf7@O<2tDC73s(fWq>FS#jUT&6 zH9hf-5G}eqywXSt`t>ArKP;V37~C^tLUk4q@`ubBrI9MG5i^CONP_F8A#anp49Z!w z7|B?+t?2|SUaD7mV9P|@^Osq^Ngx=Y@&k7Xt1)uv``AkPkE>*#rrJqU5+uocRjyx}Ch8 zb;dUa!&%R^U&$RoCUuR4lk^+S#G=_^U1sTHFO-6R^`)g%bsABA*gvmB==YikJ-)^?K_5D!2b9EGhB3loJSA?8fX=86f>Z* z-4CH-Q>D%ySR$QSx$_xdiVs4gGs*q?7@#oeD}{9Cit++Q^#1tFz}hx~6Kv+M%b2#w zu3$5k*gFtoc1}?!Q+&xS#@E2rq=wIP`=G3>gmJz_zbl0Dse0kS?K?-%8-SXfZQ$a0 zH!}(#ykmvb#FigGa@fcNj4>2h*osThm?W$~|90P=L$oSBu8XtCIpZ@{*D5~hWdE$$ zq6<^!*Jdmf8v?03gxRNkn;klzQ6+Z)KW9fX zTpCkDtJ{imd(I(q!qu>a^~ymUjyFd@ez9-7d1loZ=4-3&yTPqC@`bje!yLTGo?@{> z!Xv#tzkOH+l6JkTF#FD@YQi#%*ixROi?~Pp|FT?QW;A$Y(>} z6YXAkEbw&dUO-jp$n(q0l)4UI$~^7tN}0#Kmvt`Yn$5NVMn+&)ZNC8*Qv3ZAhfYfomd=buC+{FAsTsSi0eq1Q_ zijNhJGB5-H>^Az?$GZ_Xv!8%6&L1x>N)|5u;@B;I9J_LSTL5~aYu&4xbq8&`5=DLP ztOTDM$&^CTGBRt#Dw2p=$E(kcCJ|D{8p|%K4LT z4{+Jx?fIpMt0^iL`CLB@!oAK6p7aEWbg znOs}}Y)`)QtnU#0oqNp@7d?v6BArdL;%MJLb^*06%vICpX$A3AU4cqOOKzxX{PF%^ z%dw8(C0QeqNifMad&`=fBQ*jD5z>+&l?T1{^uN@fK2UWv2nos2+JU_{+yPs}zT9BJRd56;sXa z@SCO1*^KDqMqm5-c7Tgh;_2Csw(7?h*>^J|U60j^j`HW?repjUty37(;mUuRR z>m3*|2gI^VV5B9y^0&Q{ddDVPLeS3%Wg7&BwY7+5sM8eD+uoXmDU+!YYTNY&w8Yw+Egsq*i>iaQ5#(C}BQ} zyz|Rlb-?;kv^PiRJ5o+e2;nT?Z@I8#ZkAj-E2l0vQr{yR(VY3cdQe9q4&jazt~yU^ z#=Tx}M9!zPg2%Q=>{|keh`D(GEQ+@+P=e-4O@eH2*S&z=FFo}EbVP=x%Y>mHuF4}D zR9mB$Z|M87Hr%FJex(U-h$&p&!4vC2;!y8hPhcfehJ{x8VbXNlauF*)%kL@jEUTA0 zu-eVH0gin&JI!EY+&@tx8&9lai;Q1P-G$QBG1*D1Og@a}c4JXKvHgGd_)}FEb4X)Y zVg%2Gy+W_3`%#64X9dt;q>YgoVh;@*C)j-(N~_cx$~N>Uw?G|OV6r5#wn{#i0~m;I z;mVZtuo8;b-o(rcpG6n|opygnAL(}2Z>1=$&gJ>gxm87S?bzbE+XK8!YZ`VT-e@w1 zp*0tFzwUahRpk29q=Y<<77E^a43c>>2fjeJ2Nr46B{*!@S`yz}K&T%H@0Jujg7y58 z6m5sADC(x+R*BbtM9z(VFr?Hwz^F7$P5MZ+`BEK|Rw|AxMmOsL(?+W_l!>*jlxGE= zO8cQKQlAj8&DDlmTHqRD7%GpI$E4EG#Wuj;+4tr74hS3(^~JJuMbq(ij>Pf#Bj<|V zzb(2ED%=!j_&}Q#_rXh~6oMEg3pI#C7~WC)iJ%5Inyy}KGrePtO*z%Xo4&=*D2{WV zuYvOP$=ShP$kV|9)288{21~NMno1zR33yT3k;mGNb96=O(dbHh%<^=p|(=C;zCZ?H#(XJ?ES~>BsPRKm_plfBd-L?euo&F)WqX zT$bFUsJ7f$1WrW_oR#!In}onlvAIh?Kh6Uai>m1KzUA`(`Z09B+#L6vV+gA^bbRL& zNC}|Xg8`^|v`6MBU>VvGIU)#^gz#qLC`;^zc;Ju*DBtL8x+gDWD|-^*gtzYacB25&0VUW)l1C!zLy%E+LC@3~u| z__-m&_4LO8?z4lIPDMD~@}#bxLdcimCQw8XAk|J8Ro5T~!rpz0S>8B{0*7TjMmnrl z8_EN4TeU0t+)t}IKggtf3YU>(<}~U8iPw2uUVJ!8+)D^1^t&3V7gwwJrgR{O?-iS8 zZr#4;4UZkFM^AQ~o{;nQmUk8A|7gcl>#uln2NF$y+NHHj)+?2I+`qC_CC^Mevx^th zp8gzf9E#`5(K;>yv9#Ah^YIh~xbIW6-W_nvk;-HgJ?H4ip%gT}RbKVNp)HOjVRaQ- zZ_!s3ItDNS=P7!T$P;Mb@Xl}(r9`)HJalT%I;P0GIAU+J-5#;UlU+9(t?if4$>**e z+XNc`%~w>uTY1c%d+}B+R6Ca)1uX}1%c(@Sk1{8kK96x@v>qt;W4j1;&NDX4w~nD; zyvc$7i1p!@O57&E9Kx0=-!1!cJMb|Pc^Cdzc{Q~9-IcwJfZOdA$aH@)ZAd&;299V;{`K5FGhY%+hzGFtm$jTZnq#E*&o|0xxqWDn?UDLh=l8L)3C~hC zz0-P>S8uXRCw%6<{ywPVoUCX4d5&Ew&CjSt)zVPJMly-kS{4CeXJ1-T`r} zkk#&cHEmp49S}+)c zy&5w5A;yRI%C^zl{zYc_P@|UM)T<%$5asR+{hg zm#1JYYdJ{lr1cq<2w!xg(ZXF*pjH{$pc|Il?30V>;;Wlk6<-OM`x9@A`4wAtkM zRNn_y-pm;&F(gf|j~*3yz5AO-fk&4ApC`5eAmg?aXv#%JMT<}EnPawvF!#00_nw%H zs#F4mtEC>mzsyMtYrK_zplsseUf{dg*_50v{+;*o^s^u%RwsJQESvU~zZ=F2?5RPw zBn(Ko>6F}^)OI%a)=j?dv8DP8KDIDAhj0ZRu52UfMt4YAW5Xd6Z}>`m&j1oxwx zBm+O}Hr4sQ@Wgf=H^W3Yry-g0$eZWMZpETgaXb7ATE$gwJSIGVboxhVW0pPfnLAn3 z7L!5dnWE**Fb{DhibfSBP`T2@>J+MoXzBxp)^Da@ivw>+F~vm9TG8?N ztRhNIi3xF=Jo(}cPwcofRMwSN2CS7`WB5l%q-pKzIF1Hv?N;J$)^vEJ#58xM z1(O4;%e`&pYi-E@-Aswtx*c_HCb4r=`-jTp;$HS+6QKR5y|-?FR2}GY4oU3&xx=4X zkwSl`(1Z~Fd7Mc7$o`W|(g)sTwc3Ngcxn@!unpOLtKnDt#|3fZ?Ny@sb299Jrk{OO zBQ9SexU_QXzH*xyY!9%_&8Kk`(|?So+uEC{J1EFL^Y(RzPQN?wkNx3&q|MZ+UMh}B zZn*U@2%zr;h6T0PMA^F-`j89*t`=!S9f;J22LH?a!yn1F`Nl+&s<4xh!lykiZ8~CstJTQR#qt zK(N(8obp`@eVdGSh`ps~e`4Oj<-qw_&p+dpZHb({6`}w!cXDK&X+=1(DBf65d-Z~ypsa1*vbZKg+2RUxY zBml}sb@sLbm=+)Z|1@?U@Kk^Q<9~CJP4<>OvLk!1l9?6C-aDIcjj|#u-B7NPm1K{& zW=66jN*R$=MzSi>|BdwNUi$w2J-Uwv9_KvYuh;9m&g;Bh=e+Oxz1gIfwUjn;p}OOO z!N<%8#wigNgTRh#%P+6FHXhwrU+m1t>oRp&2Fz~^Zp7Ma36uttv?+;Cq7CMfK&x#o zpUcAyeMY9+=0De=NO^C1YkJ3HBDrXaKikeGdmVcED=wo@?3m6ntoBZ1=XR=JLpL6o zre7u0khmg==5Zdjj03+Oy4B1LS(!>d(aY%mEE|wh==D}Xm(EePl^Jxec{R=sJiIlE zh>_@#3tG2)mt3Mloxst0zd1|Nl1^A!UwJN9P;Qn%(7k}`#tUC89IEVR+zrd-w>rh| z;is>%?qRg-=vxUFoRvUxwJyr29peuDrq{?(4;4-;xXfamR7`1N(pPmsThlFEfT!A* zb8F@7n-mW#1DXnaAAxtJX`jk^2epJ&=Og0c)Zx?j0`E)1u|*WCl8wjRGZr$2_=zhG zde}{9(mNFAhvj?I#%3>@NM02z%#V1T?h%I3e)C;1FMjtk`e5dt>Kr7yT%E017#bI^ zyjX$mPN$DJa&fxq->w(;4tghRA}fBEi?Y*W5u4oVb;5`DP0BR8Sh0jcrrV(eVe0T_ zyX{;l0Mi21V6Mja7HhD|uf!N}V&hJ+o|I`F*Wrg!X4liQWu6(IwX@kesK{JLc$_vxDM(^_~JhN>NJ zeCua~I#yt0)5p4UrpV`!w3BkQ^b?nmtnF1c?Nwk_2JKM+WmTj9_}?iGhgQ;&F$ z>{&r62nJr|-fe57AxOFCt4XXPYqKYI^M-9y##+>%x=PBbUOr)a*NgbDNzDAJ@3R;b zus32CX{YWc8{~>V2;kCxK=StPt5tG!lD&=U6|%^mmV#-agf<}rT=|H*xUEkHDOvln z)2SbsZ?=SJU11EBA?)bBzz;L8f7nN4UCVxM25Q`OD*gGVZRWxL;Hk7m+Euy`hZIG} z=IG&Ztgl-gXcV(k4#uQ`sj=QAv#%r7y534OKdb;<=)prraWvkMJGXk?W z*@N#|27M{uS7(-@@w5vb(WPACUH=&4_VWGU^KUJMGp%w@=`MV+>}9aI7I~pz+_?Yr zn@7eR(6a^`yTS|k5>`YrW!Eo9h?66@oCUCh*{n#lT|LN%gE$5Ay*w)>6a(qvc3i%4 zR=(ucB7NY&yE%ke#}LF9?Nmf!dOP<@GW!SHah>hqs4-%-Cuhi=KkY=zO#}zNvKRg! z+h*3~=Sx~&iwhq#(2JL zar*Ys`Mn$l|*JCa!U>BH?R8;3p}db!8f$bLSs zy4zJwU_EA+qDO~!?bWlNpE9IEY_QZA%)R(;rz>pt?5!#%`Lg&nvu8egFCXnq__!w# z-J4?11_~YlQ=C;xuV=u-SH78&am7zmP}gYwvpRRYygow&^e5in*(wB}`vMVOHtJb7@-sDP5lwt`L*9)nlJd>t|$m z5YNX!JfGsElfs5K2)_?5hHR6C=eL)Pc(JKIZrOu6TrKjUyOpKWbI((wb*Jk*C-?xY za4a;1!5M4UC!bkgXs$DC6{=hgQ`IJ4a-scHbS>;O-EIH)F$NRY3D*~P4z@TD(H5fY zJP9hiuZADL6*@>72vNeKa*Y`A#lHCp<-SY0@gkz$`OT;0<pWtoZl-eb>TbYWY4M=XwNLe3P4le2U_F73mD|--%|3qE z^vlCovRIljm`-A@s&vv_`^kcdpH1kcAM~`srduRd*_V386dHEiM_d@28YN8AnB2P; zu17?j_A?W59UOY)Wk~tZ?0rQhkt?rf{bTPPy!i$20dYO!PD^5{a{LXq&rhf&1y4ot zr$10h>!Kx8c-qZHG_5RQdaaG4FYgIMl+BHV*>eHRVzjN&+}o-f54*_TJ8^M+1(z%h zE}Ku(Vx^H~X*x36RLbugzpOWOtY}*Kx+SkdRL5ciFgSY_e%@cJKegVD6F4^Zt#Xb3 zAdkZazut}wfDc56*%CdlJIc)>(_7AfA13g_41QR^4=ea#t4VKRrw}}N%sTk~B%MFa z@z&WR@zy!O4`&VDI@d_t!8hF%+%GnyumV6tIshpBbvHUqC^ySd*BWSl?l?M>X-e=i z-MfOIadRbtt=wf%55`kMCTz*3Y+rLy@T?O0KG;NdGDkj}nKviN|4{Y)8v790 zhz5Dl6}bu=q7plQTqgt#no zxc75#ai(+Y^%;sG%hJ0#g0i+|Hqe@HyFQt7bY9;DCBt05ty2YmeS6cF=ts{k!JU9f zDq`_?>q4*X?!II`db zG$WihZz(WMlQa`vkj-)wHGt?VGhg^QN=Kj|orqH?L!4<0Hykr@vb22QvO9Danj#9` zUc7->7~v+K4v}oRI}g6Ja^`2ylr8yt^Us)-gthPaJ@`oPuLa*)1wM*RKN7Th{nanw zcB)>cOPGII6unAmWspz6#_Q4ii0XbG{H^z)Sq?oPPEoB~JcSd0$!MJhrF04o#9u`( zCCS8yX}KB}ap8>zUlQ8oC`u*DwmlWahD;)cQqpnx559)Kt|7WG3uQ_#qAPuoPmGAn znNX?l&F0h&y_@RlSve6QXzzV5q&@7`xAZSglJ9)2Fvf2r=AMBWZ&24T4(A47K;sJ) zC0A#PlcI~`^kPg--;3(!#2~$TaEqfsvzd8Ls*h&=i`E#!@#e-C68`8zAl=18e5WnU9jk4 z_Pr*fZ^jcm=w1r-y|r6tGpTgFP~BTNrDw4DB!kStoyzLgrQc*-tqec&Dqv#E#~E|w ze5pBRU}}YrycsUO4}RC#_0$wnZN{Z($&9NNq%rv+{+e^O{cCw@2CtJEukwXE$)tzg zk65&vjDYM+*@d}t*a**$O2uq(0B*!Ie&NI)2tIu@?9VPTT5rNWlSUcE$qjWvf9j>< zR#vFz)-T60rl$*BaoY`fY?Aj~WNmyt1)2sv@QC+1dJL`{k0SloRFTqQ>eJo{+bm@l zRDF4@?Bp3U-AHe|-CN2o8jt;IL}B9ff{uAbQy->vM?!_ep(iH92? z`N~aJ%G$e$w2i^QirEsIAguvvZ_Dy?<#}$nV}GPWdIE!pcYCbg7x+4FK6wTqhWR1eH|@pL|n;34gY8UGfW}A|tkh&Ga8`t^b@Z_f+HMmo#)1vyLuY39rV~y% zf0l`DO2C*Q`f}-V>~^aYE*rGEdC;ddrSJ0FUUy!JjxM9K7;6%ZTHa_1&R?&2Ul^(p zY9gzwho?V!W~@c7Cdz0QZ-0HLUIG4&vbdVgeTZU*mg{PHWMj{$;x#m;$8($NUW?cr zVvpT}$`gqQzm%Wkp`BLh6 zVar|?ok=0G1Vc~qtxP2rKAa~v$h&E6u^4h585cg$D!Fu#+aCzI^rex!N+rdxc!dWi11w^Vu@ui1=4bEd3b z=8h+F(bN(1j&eSEYp1tKD`e+Exv~nQf6qS&B@~FG^mr?1HZ_jJC4^}f24z-uu4}k~ zzkN%)Kl3?Bqq|ED+hcBb6aDe{zPQMARcD;MI=F-Rxz6j zwL6>7mT&odukNuQUv^r{NcMuZ*|n}JsrsjRSD9o+6kl5LUOU}goog(rp81~sp$vNd zt0kIdFPScu$8*CgR8<7EwDb0(y8|Cbxifo|^&&*HO0g(+lj|Wac;Bw`wYcG}Gq_wl z_cZ0@_q0Viyv6Fd7Jy#L!t_>%-glT-`+!nJtv~wRq1UE$Bt|Ujr>yzbr5~)Q|J^NqXV;td!_K^CmO! zgj-zbTvqy={gmr8G4Zds4GJhcIXAcQk-EyqkZ<42YUBfSC5@kM29H;>z0zn#;1h-A zKuYL;bYJLiR;V)Q)%h4ZsU{-jRabIe;@QO#Ecf9W#oIeSAt}g#FwqPUG;ul3z zUri(FM~5eZn5V>hMek7wxIkkgdR6t;2m6XQMy!{rnK!qtK5_mm9&B4Qt?hruY&syG zS4!A-O!KpF|J^(O7O9f9wpDM$H&+(0unDEMZr$HFS7KHY+kPLfum~@0uTfFSeKOWk z--JG}(@r!fc1W~+dX8ZJRR<*H+IK&ep5xOU|EaBCU&qvX!@2XOv!Q zDzHf59lIVN-7X~Mp(EJV9VcgzsvKW@;w7K!UsDOq`` zu^cn-^|Kc=D;W?!Eju1287R)BSn<2R+zGf`&ZCv}89EnhhQ62|m!~bwkE`JSQNU3~ zEBfIE-)#l^e3J?Sv02SrDHl>n{m->{nWt|8PRtt}h*nK6p*MQJrqN!@bTd8ReryNj zs~}(TXedu%eiIs@P`60hnD#^M9d|l$@(=5;WbtP$&@RqIeKG5eFf)#V{wi%x~SQZwiBDtl)+R2+<1 zzawB&uWNy;zPhx$G<~NCi)GyWDrxCU^gCz!b>)(^;}TW7?-M#N$;B@&3NJ**%T!xK z=E9S2%@UN8YtR;nPT!Lo4uB5Y#_-D2m!k0(O7j%Tz4aD-Z}^C`h0i$hb>*_e%k>c_ zzfnVUb1LhwO~Y&SxY_k_Q$MFDaxN&~_ZSjplHx6}wfN`JKJjPzoI^0mxRDy_tdB11)C^iYNvjY!UDlUm!T(Ew-RLhtTskUsc+QnrGwW zb4my<&|nDWXV6VD+$G6)V|5L>oVc~5OqiVFaCI@;921*dpaSPj@uWcM=NN}ek~!;Y zW6|qzmn?^u<8-v=94fE7C|6;<5Uq@})$(GN|8TCyMuni+>WcC6+-r?`oQgq~@aWQ4 z;U+5AaWlcEr{@kA$9eNfpqV9K?cGZI<@12Up>pg~q8Ph{#ydBA#!jV$btsc{6btp@ zxeLot_g%z-x^W0*M^(}21&DKEm%q>=u;E<3=1k>eKleuiJG^>{d2CS2KD7m_%lBPX zT&GMs8ET9l-wrL1!WS9#;Cuconp$3;hM=oOr$MtBD9yiDqh?YQoAPHp+O5X;PY&aVjS;QMN_cFtDLr_22dwIDOPP9Zi~P$fjeg{7IG zt){d~&r4-SD41C-qu*73G*Tr<*635WAh$8`=v}kn>(JLwDDg3o5yF2nY8x4bhWB*) zR!3u=J22m*Q}4MN7FW!^{f4Beim#ThVPF=o?n_v3XHc!JbcBXV@I|)Nc|z6Yh?|;W zzeJO?^2zsB9~)NZjlLZprG|bFcvuUI(e7o)SoAX2PB`7|^spYn3v|526>c-+dg!Wl zMhsGHkz6GwVA2|7n2k5I`2eUN)itoHxE_W@EvnSv;ve$(mT28rV)-}p2aCbd>rB** zvWU;%fA0{e->6xHFK>QT<tCjzQ$o` zg0pn*20_rcvsmKzI63Ckc{lIsK6~1y{4uUuEb}dt%9YN8*|r{M5{eQgqNHGnKG&9b zTQym(2|Y{9dQ{Z?$j9AO-z0ltSH|evfctfIPfD-CaG8$Gj<~yXk`9hEA^D}cEM*yz z>1cJHjOhlVjx&uR7?bu;Cb7C-1kJKmKB6R<+Pxiax9le^-NmEH1%s6g`={zfU zdtNh{^qK9)r<*0IWEs_ykI?vRtf;U1?M~D(D@!VW8LO|YyCPF&Yw>(fo1w{vqVS1f zNs_eRLYoaw9&PoDoc8xy^INCWnc_1pd2&kwOSj_+c2ZIbsDesJYZKh0cX!oH zoO^1O-)S;DQB7+U^fTj?Osupqfx^blWMlh6qc^bIgN^^STpnjaCREhzp%Z4|3UgAX z*i%o4DZ=9V2sBoZTQ3&F~`)i zDM{~e>b7d)yO9aAvQ0)Pzq{u(9pllNH+W?4D0cBQL}tl1eah2+{o1fpH4!o2;;o8Q z5u(?LRj7`noJbeDUL@`xW%&N1U#wuqL+y*BjuY1lZG(bn3RJNZ3+~0JufDAa8Q*5_ z@@&RAm>2m{|-=Ge*%!5gAQ60g9C zQA+WPQXH|5UC@X>-6l{YbxrUQ>LQE7AX+Dr`i#kw-bQL;ob$=#$DEYtub$hDUcfo8 z$I}hM@X+pWqTwAzIpqUT8cY-QwtEEA%Rk+2jMlxYPC2-@Z6!WHKT*xP(>vx7 z-l#`8`c;n7hGhy(CLI#4(UvaqKuz$E?S)rX`6z`Rr9cIgq)2U0; zVp;O6Gu8U?ZJMidnjw9CsfB!B#!bn*u#=3w3W~_>=>H-U1u)ko-&X9wqlWcN*~PEk z_P=+@o7Dnu;C$-hGE`Nbz`ovp&R>bp^Mvop5 zW~VskR-TcIBTji1JpEmG(M{iShC*K^rpRIK>c!z6Bh}gpAqaq`qh0}=x{fzTc>ukHo?D;&oJD)k5_jI~= zKcxC&sx-Nwz1SUll~7y!S_DC@qfUKcq#EDE5tnyfny6>CTA5DzRsNlNKhhsOWIZZY zPCv0N0n&*dFF(#$Ojf&R-22FzyL1s#p#|rAW~trq_8o$^<{9oDr z4S4+hn5!;lAWAPI|I#sXngEooO^NG>Wn0Y8g^Y5qJLK(-~YrAIH9>I%5DrZ1j>PQ0>Z+sm2Wv~qscDhWt(39nYMHx zk9*6!F`?PrY|io-Z!)5my~HkKYMG_SA~*_b*dTT+*SKrGd(D-eS!e`WmxS}WUKo%? zdCOcDH`-D`7TR0ZZ{9&i~EjIf!(7i z(aPk@cWlgG^y6$L;l)Ij=W^8Q#Gf>lEW9Lng1D{e4B_yeBg;)ZA6qiw4cO}>%fxC= zv<}Hf+^M))#{Bf{2dMK+{GnkI;-Q|1dusyG&sMEnRzf_x@Y`dr^sco0s6dPsVQ6+Y z`CsxxzjM7yGovKTMVu*@@PyFevQ1t<83Py9N&| zCKy(GSufpcjY3i*c2ndPB}MI^d-l)K@xJ_tva!#<8fs0A<~&A6&_;CXKl;u$YN|f- z;$v?p9A0^yYoBk=+1*s9+iz&a;c9>8$&dEUyXR`=qhh8tADTfmayDpWe}s#MdOw>B znlfMMkucLONwdtYmkUzu%p6AB>qA#Qz^Wvtij#-!+%aV^eq!F~xZGx*d z4U9}&#jLcjUY9NH^3_ETR-VD<+}JhXN_RVLu4`6J9HKF{z; zzHUX;gc-lMUCO4wt`$4rJL?^Eqm-3KHRE)5RVK+)YqnM6d*i?z4bF17FhNZq76tb^ zqCQy7^iOAkPkzF0iMr-RX$bQaUTD;th03IopK489FNqy(W%8b4GpMS1n9M%n@+7;V zZUBw3V9?ymsW%Ba*W*Lzl)4fia>F653ZUl_%ITuYbNP-f>;Cq<8I6cr1>RD)U@#LmuQu7snz#b=8zcD17Fy+_g}0WWSU4;do%Ql79@4;51hnJ-o}L%!i^utNf^?+}7*hR3iJnXuiWwNFU_o{0b1Ql~!I zJwL0<>Vw#8j})_gsqafB?KSWXEmkElx^})Xwsgqm@v4sg+P4%$qA|_T;53`TM-g>R ztET5_Yu?J~cV&h}x`nUvs|f71E=E5REK`C=DmuO~Ou1_dH6H8}^k|@Lktz&f5o@`u zB*M(M;?tsrGe&SR2&Y@d7H6V)*YBQK-I?`MJ~x`8Qu!s}~JYY#E$)nlWa?8;&FlcY< z-BjgH@Atv)M6{0$;|U6{w^s+Z%yzZgyb)gw_Vvg!i8iK$vT-*pwt9K1>S)WtDE_MY%KO>(>ja#uKWpO`jWNW^`l5R=rkJxv8*4TBOIo#Y6|(=wS`a zyRQvftnV$W_hflW#%Ou|^r-duiQ*2lUwYX!evS8BGfj<6wwu${@6joLV>P-dq~K1l z^g6?d;Y$A&_$=B_eLM26Z-w)|sI*t_vlSVi*6$G9HM&#j?kW*PD!yFXK^-zUu$MwpS|M*(n#$DD2wRxEE6R+WzXYZr0dV?p|#@W8POitqhNDQSY*`pm<)| zGCYei+dJ@Uqol;LPvdfi zci2qLYOvU9<9F!14ROSTF4QO_OJ+0p^_QHR3`It#ntoAK!mGCxzImF{yNWWZvc?;q)NCNcX*N@ogo+V2;hao?ux7RV`zJP+*0fc+NniTIlfA#x^}W z2U#@nhqq)SAQbTCrw9m^d@T5OZSYk2&p%3l>i5sYz_Pp&__M2}*_*sd@fD&D*5}qu`_8s)+!A9Z9^Uf8gL>h7b_S|Fnee zV`i&NR=Ontz-AMCiX0M##rXe{%Gtuz&d$Ta-QMg#Ofyf1{joK(v~YEDaJ4hDwsCRe zb8z=}wLBVZ2X8+-01aak91aXJ0NxKe65)X|JIO)u5f}il04gJbB!%0(lcj(g4I?1z z|0#2T(Su&$!8;8AMA87jhlH^PZ%a7I6JgFad2}a?*Z~f7mp}hFf8zjP-?2xZ`~x?L zolgcINJBs{QAC?uR`Fy?1ORvs8X)tBweOYU#s7l^)AG%u`ddD_1Iel%#uj$U1HgTL zG(a6m(N{8PCo2j{Ch<4u_Y6L!DEA-X_7!Ce{RhtUQa%Ot(YPLjghR<7G)LR4cf7^0{W>CsQLG~Dzp(lk4;u%+5ob^s0lY|{XM0+Kds z|H51wLqN{`x&EP1!~#vH+!z7CE(!qDkTCNcCrbm{{+54INIEv|9Dl5`Kkj3y|G-_> zDIh&*+%=sdZZw_;fCN4?Km$oK$qWAj1=F}za0ZV59s$8aQ8gda?I#vL00{N~fB_OP z1p0pf1^*uZ-+q3uw(x-?L(5N5a&mugRdRK9H?wfHHgk4%L0+K_&Cf({$^M)a078`k z-~y6Np8sINSH}?$aTL`sy1u<(0Y|a;768a1VQQbBEE+5{yMXs^tq$dxDCQy&Cjo=E z0syEWf%Ilh1|7&l2dA0@4S^z6UaoX{F{s?GBI?n}^Z? ztM@bjDI{30%BlY$7@YqD0z!@gyZu_5@7ftOU@rxNG7Q|EEdLh_ALgFhMSOg0%{le& zR_KufT{CYiF){Wnw*MYhCGE)}hC2(TB9hmyspNK5=1Lj`NBpTUKXfP+DULVQaj zNzEJz0UpZZ0uo53iyN2zKZqyj*%{^X%8DAC=Z2`0fyGSzWL(n>Qnq#w%5n%)?Y@ct zK*}~M=9k|;FoUxQ2q~=BeV*pH^(Whk`p%Mr3CTqOC7d~y`~xQs;$%>4CGu*W%s?Cf z2r1D3NhDj*wf_UNFBQ)p%<^KbxP!F)nmw3S?CsV@WouXa!)wXy!mOjv5{0ykW95dk4U zwb`+?TL8&s{81+p;b%S}Aao~%*4Ow(-k!Js5KRXFI!K<8`oCBTe*`=936mT;o;vUz zU~>ej&0)zWi?pvKlFbi6YwwiB(L;yObqq$)&;QYFKU*8BJsAX-{fvNcpvZ%>C?S!} z1}1#R99~(10&-$G0LsG#9|u{eF&_N-xAjMGbyNVeZseuk(foh0E*T*pr~me11Cf&*AJ4QCwcpQ zp7j6S$H4`x;83e+qy4EFnsU3&F$f({z6}BRk%nUXB!Z`nvyGJ(SpD%@xw*Kzc{+I6 z@VeM|c{o_{xx3mSPYl6*z7BJv7^xf%pf-UI2qW=dYEb_*Mh6nY)oZ}@2*#iyO@L(5 z15(`pnS~Ts^Uuu!D3V^f!+eZHP=J82{pT+S1OwxWR01Y6;H^Iz@L#>R?{;6rS&k86 zKmr2_;mP}Cx6nL1;HeeSpLoAtLOu#LGhsbOIMk**LWS(OHl(W)SdCvX0D63^Kc(y; zNgrL{80mB+0>X+S%jj(OV*@mFU~vi^P(rfojZNWWgg=%&MaOI9Y~kt2>)>K(0ao}{ zHhhlmCyg8qO)hCT8XEAy2o2yu;%jZ6N9AV}A|Q+?0ypa$6<9yZPv zULf7`IQL+=+D!lE*7yYgX4TPv|CX!$QI7p6iOTilcXqQgJDdY%wt~lzTKm($uAE3^ zHWdJ#f!i=_NVdp(E%iT<7T#87LdU5G4vmHP^0{pd02F}9DKhnwn=~pFZu}UjC^#Tc zg;oeT_5nceSyVy^hs-fT?+Jv9jA)$Md>FvHWDJxk;kR|1AXg18QSeI8o&O5JR{Iw6 z@1B|o11tY{eRMz^$wKjLil~z8Z#E)_A6^GnUN0MPf64-kNbVjsJ`Og%|FGKphWY!< zAqb$9fC>D!vhEM>%?>5hE5X(RO<0akUw^qQDiI6};DRqEAcQ3CowN(b7=$`zY{wZy z=XFd!$)N#)w^41+4_8NJz~zdLj~+17mF0?q4#Xv*1OJs@`55Xr8s`PE zj!W`rr{%s81sVWxLHo1UcsP;NKWU+IV4-X0sF4^#`{QRCB`HK<2LbR^F@SSO7QsZ* zL1lQt2?r1mPLzPboxsI?MN*@Hz~~s^1xR2*(c-}sh2_g;Xh8208o-N0i0Lps zM%a(bEGPt(8*34bH_?GybPPZYiJ-`S6_o%xw`_m*_;8HZO1;~1!U7PMr~VYchePUF zWQxl0glp#_AcFsip|d}txPequRN&^gHn@TL-|o@AcTG~-97FKC_#f`sn^`+}fH{Jj zhyOq3Ux@BFLyQUmg#TpekbkB;E@H=!360aJMC|8seQJCzyt?R z|K7?%O_TP|I}XkvFkrtMvi{!GJ!(mJZ~=w;FMrl~qXF%uXn%I~4vn~Z)DLF58HfWD zfZm|om-0{dgX|5O4o+E&vz87AvarJMbRr;JC~XKH?6-Mt>iD~jHuyUYN}Je&Hmep# zZTi9A&!Dt9{cxXSpyPJPfyI6dI|w<*5@j5;O7}Qy#Rq$j8BUKR_kr4x0i>TEfb}D_ z5$aA6?tuGmIo&h%%UEQ3BltH^-^Jf;U>JNj2Rlc|tz_W)t>zl9|CY1~g3EebhTQ~a z0Sq4SITPKe#q9cXpWiYNWFjEkC^8sKILjLY_SY>BIf`1E^nYABSj?_M?GxxnN z_VvDaTWrA)_;v7fgzd0eJ8F68b$TD6>v#!R!1T*ge@c-#9$e{UH=ty}kck2>zZap+(P{3GUpcdzE`t8(M(Q8PFZ7}7aV z_@xxqO&L9hM;Hu1eg5iib2+&*& zK}iAGpWgY+hkg3WjyyXLC--09XgwPJLz@{$B^`AIzutQ0Zyz`sOE5md4+|}Oi=@9| zp}m{0nU@>N4#C0tD2`{npbT!PYkjH((BC88bJi?j-fOdjY04tL7 z8HF4d18(sQ0ipgcsvpQ|jOCJ%0{R9pxCIN_lp&9cW4Fvd-Pp5xZ|4ll081J6TZp8|jK-$RJ^)e|7GvHr-b zKJ)&P?ZZ0ZI2DQkJkP>JS+dt}r$S(-m+i@674NPi*P{nLcYNZ!tcSsz<1qxKr!Zm0 zVNdu<9n!(qf$q1fb{)6L03b^Zbp*aoXC4~?81-v$9{r>JjXKbSD+9cRIKjZPZZ^y%$icfffc?*{;)NV?HdXB{`t8RY=3-+lxmB zxQEU`BX$A=oU~-HFGBvFwJQ~o-i;hw_Ukb~Dz~o}Z$>G(d2nY#EBp@M|)anNl4lgUz z@%JRjIc}r47Uc0i3@5kQviV%VSU_Qn>WTShk3-GgKq-*V{zS}9KvXv&Aj1D4)b@vH`wAnu-b*y#wjDZf=8p^S zrzIr*#}WK4;B7a~W(WIc#}e+NY#DWCkpIE%XUolsxe!=Zc)-CTc{Hi$4n{P2(@4|; z-15x=52TN(@B|ELd;?@ozjDfunuN%cYe%YFJ&so!n>Q6CU@@3(R!Dg7?t|Gr?@Do4%!&+H%BQ8J4=js!a?`~L)%60{in1AiLv z5AGic|FQmoZ9>7n(PC7i_5LRPf8_bd1>1c5z+u;mC}jR+kPP`zIRMuHe>$w0T)>~G T2p<8D01a*MOs9GP3^xA{rGEO+ diff --git a/site/notebooks/EXECUTED/model_development/1-set_up_validmind.ipynb b/site/notebooks/EXECUTED/model_development/1-set_up_validmind.ipynb index f85b592d8a..61ee21e2b5 100644 --- a/site/notebooks/EXECUTED/model_development/1-set_up_validmind.ipynb +++ b/site/notebooks/EXECUTED/model_development/1-set_up_validmind.ipynb @@ -139,7 +139,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -215,8 +215,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/EXECUTED/model_development/2-start_development_process.ipynb b/site/notebooks/EXECUTED/model_development/2-start_development_process.ipynb index 1d62e9c085..23442b3d30 100644 --- a/site/notebooks/EXECUTED/model_development/2-start_development_process.ipynb +++ b/site/notebooks/EXECUTED/model_development/2-start_development_process.ipynb @@ -221,7 +221,7 @@ "source": [ "

Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests" ] }, { diff --git a/site/notebooks/EXECUTED/model_development/3-integrate_custom_tests.ipynb b/site/notebooks/EXECUTED/model_development/3-integrate_custom_tests.ipynb index 49df9a5b50..9a3cbe7ef7 100644 --- a/site/notebooks/EXECUTED/model_development/3-integrate_custom_tests.ipynb +++ b/site/notebooks/EXECUTED/model_development/3-integrate_custom_tests.ipynb @@ -13,7 +13,7 @@ "- The function can be as simple or as complex as you need it to be — it can use external libraries, make API calls, or do anything else that you can do in Python.\n", "- The only requirement is that the function signature and return values can be \"understood\" and handled by the ValidMind Library. As such, custom tests offer added flexibility by extending the default tests provided by ValidMind, enabling you to document any type of model or use case.\n", "\n", - "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb) notebook.\n", + "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb) notebook.\n", "\n", "
Learn by doing\n", "

\n", @@ -817,7 +817,7 @@ "\n", "
Want to learn more about test providers?\n", "

\n", - "An extended introduction to test providers can be found in: Integrate external test providers
" + "An extended introduction to test providers can be found in: Integrate external test providers
" ] }, { diff --git a/site/notebooks/EXECUTED/model_development/4-finalize_testing_documentation.ipynb b/site/notebooks/EXECUTED/model_development/4-finalize_testing_documentation.ipynb index 0af95d90a7..04c745225e 100644 --- a/site/notebooks/EXECUTED/model_development/4-finalize_testing_documentation.ipynb +++ b/site/notebooks/EXECUTED/model_development/4-finalize_testing_documentation.ipynb @@ -930,9 +930,9 @@ "\n", "#### Use cases\n", "\n", - "- [Document an application scorecard model](../../code_samples/credit_risk/application_scorecard_full_suite.ipynb)\n", - "- [Linear regression documentation demo](../../code_samples/regression/quickstart_regression_full_suite.ipynb)\n", - "- [LLM model documentation demo](../../code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb)" + "- [Document an application scorecard model](../../use_cases/credit_risk/application_scorecard_full_suite.ipynb)\n", + "- [Linear regression documentation demo](../../use_cases/regression/quickstart_regression_full_suite.ipynb)\n", + "- [LLM model documentation demo](../../use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb)" ] }, { @@ -943,12 +943,12 @@ "\n", "#### More how-to guides and code samples\n", "\n", - "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", - "- [In-depth guide on running dataset based tests](../../how_to/run_tests/1_run_dataset_based_tests.ipynb)\n", - "- [In-depth guide for implementing custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb)\n", - "- [In-depth guide to external test providers](../../code_samples/custom_tests/integrate_external_test_providers.ipynb)\n", - "- [Configuring dataset features](../../how_to/configure_dataset_features.ipynb)\n", - "- [Introduction to unit and composite metrics](../../how_to/run_unit_metrics.ipynb)" + "- [Explore available tests in detail](../../how_to/tests/explore_tests/explore_tests.ipynb)\n", + "- [In-depth guide on running dataset based tests](../../how_to/tests/run_tests/1_run_dataset_based_tests.ipynb)\n", + "- [In-depth guide for implementing custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb)\n", + "- [In-depth guide to external test providers](../../how_to/tests/custom_tests/integrate_external_test_providers.ipynb)\n", + "- [Configuring dataset features](../../how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb)\n", + "- [Introduction to unit and composite metrics](../../how_to/metrics/run_unit_metrics.ipynb)" ] }, { @@ -961,7 +961,7 @@ "\n", "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)" ] }, diff --git a/site/notebooks/EXECUTED/model_validation/1-set_up_validmind_for_validation.ipynb b/site/notebooks/EXECUTED/model_validation/1-set_up_validmind_for_validation.ipynb index b9154b80c2..6a2e9e128a 100644 --- a/site/notebooks/EXECUTED/model_validation/1-set_up_validmind_for_validation.ipynb +++ b/site/notebooks/EXECUTED/model_validation/1-set_up_validmind_for_validation.ipynb @@ -141,7 +141,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -177,8 +177,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down — don’t worry, we’ll adjust these permissions next for validation.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/EXECUTED/model_validation/2-start_validation_process.ipynb b/site/notebooks/EXECUTED/model_validation/2-start_validation_process.ipynb index 48e4dbff66..77383ce0f2 100644 --- a/site/notebooks/EXECUTED/model_validation/2-start_validation_process.ipynb +++ b/site/notebooks/EXECUTED/model_validation/2-start_validation_process.ipynb @@ -237,7 +237,7 @@ "source": [ "
Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests" ] }, { diff --git a/site/notebooks/EXECUTED/model_validation/4-finalize_validation_reporting.ipynb b/site/notebooks/EXECUTED/model_validation/4-finalize_validation_reporting.ipynb index fe4e221c90..6103fa2d41 100644 --- a/site/notebooks/EXECUTED/model_validation/4-finalize_validation_reporting.ipynb +++ b/site/notebooks/EXECUTED/model_validation/4-finalize_validation_reporting.ipynb @@ -13,7 +13,7 @@ "- The function can be as simple or as complex as you need it to be — it can use external libraries, make API calls, or do anything else that you can do in Python.\n", "- The only requirement is that the function signature and return values can be \"understood\" and handled by the ValidMind Library. As such, custom tests offer added flexibility by extending the default tests provided by ValidMind, enabling you to document any type of model or use case.\n", "\n", - "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb) notebook.\n", + "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb) notebook.\n", "\n", "
Learn by doing\n", "

\n", @@ -480,7 +480,7 @@ "\n", "
Want to learn more about custom tests?\n", "

\n", - "Refer to our in-depth introduction to custom tests: Implement custom tests
" + "Refer to our in-depth introduction to custom tests: Implement custom tests
" ] }, { @@ -856,7 +856,7 @@ "\n", "
Want to learn more about test providers?\n", "

\n", - "An extended introduction to test providers can be found in: Integrate external test providers
" + "An extended introduction to test providers can be found in: Integrate external test providers" ] }, { @@ -1176,10 +1176,10 @@ "\n", "#### More how-to guides and code samples\n", "\n", - "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", - "- [In-depth guide on running dataset based tests](../../how_to/run_tests/1_run_dataset_based_tests.ipynb)\n", - "- [In-depth guide for running comparison tests](../../how_to/run_tests/2_run_comparison_tests.ipynb)\n", - "- [In-depth guide for implementing custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb)" + "- [Explore available tests in detail](../../how_to/tests/explore_tests/explore_tests.ipynb)\n", + "- [In-depth guide on running dataset based tests](../../how_to/tests/run_tests/1_run_dataset_based_tests.ipynb)\n", + "- [In-depth guide for running comparison tests](../../how_to/tests/run_tests/2_run_comparison_tests.ipynb)\n", + "- [In-depth guide for implementing custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb)" ] }, { @@ -1192,7 +1192,7 @@ "\n", "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)\n", "\n", "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." diff --git a/site/notebooks/README.md b/site/notebooks/README.md index a641f3e276..f84c2cfb98 100644 --- a/site/notebooks/README.md +++ b/site/notebooks/README.md @@ -5,9 +5,9 @@ Our [Jupyter Notebook](https://jupyter.org/) code samples are designed to showca Sample notebooks are organized into the following folders: * `notebooks/quickstart` — Quick guides to get you started with ValidMind -* `notebooks/tutorials` — Get step-by-step instructions and learn about ValidMind concepts in depth -* `notebooks/how_to` — Learn how to use specific ValidMind features, for example how to list all test suites -* `notebooks/code_samples` — Showcase end-to-end functionality for documenting or validating models +* `notebooks/tutorials` — High-level courses covering usage of ValidMind for specific roles or concepts, such as model development +* `notebooks/how_to` — Learn how to use specific ValidMind features, for example how to run tests +* `notebooks/use_cases` — Demo end-to-end use cases for ValidMind, such as documenting or validating specific kinds of models * `notebooks/code_sharing` — Share your own notebooks or document code internally ## Getting started diff --git a/site/notebooks/code_samples/agents/langgraph_agent_simple_banking_demo.ipynb b/site/notebooks/code_samples/agents/langgraph_agent_simple_banking_demo.ipynb deleted file mode 100644 index 9afebb2e6d..0000000000 --- a/site/notebooks/code_samples/agents/langgraph_agent_simple_banking_demo.ipynb +++ /dev/null @@ -1,1501 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# AI Agent Validation with ValidMind - Banking Demo\n", - "\n", - "This notebook shows how to document and evaluate an agentic AI system with the ValidMind Library. Using a small banking agent built in LangGraph as an example, you will run ValidMind’s built-in and custom tests and produce the artifacts needed to create evidence-backed documentation.\n", - "\n", - "An AI agent is an autonomous system that interprets inputs, selects from available tools or actions, and carries out multi-step behaviors to achieve user goals. In this example, our agent acts as a professional banking assistant that analyzes user requests and automatically selects and invokes the most appropriate specialized banking tool (credit, account, or fraud) to deliver accurate, compliant, and actionable responses.\n", - "\n", - "However, agentic capabilities bring concrete risks. The agent may misinterpret user inputs or fail to extract required parameters, producing incorrect credit assessments or inappropriate account actions; it can select the wrong tool (for example, invoking account management instead of fraud detection), which may cause unsafe, non-compliant, or customer-impacting behaviour.\n", - "\n", - "This interactive notebook guides you step-by-step through building a demo LangGraph banking agent, preparing an evaluation dataset, initializing the ValidMind Library and required objects, writing custom tests for tool-selection accuracy and entity extraction, running ValidMind’s built-in and custom test suites, and logging documentation artifacts to ValidMind." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About ValidMind](#toc1__) \n", - " - [Before you begin](#toc1_1__) \n", - " - [New to ValidMind?](#toc1_2__) \n", - " - [Key concepts](#toc1_3__) \n", - "- [Setting up](#toc2__) \n", - " - [Install the ValidMind Library](#toc2_1__) \n", - " - [Initialize the ValidMind Library](#toc2_2__) \n", - " - [Register sample model](#toc2_2_1__) \n", - " - [Apply documentation template](#toc2_2_2__) \n", - " - [Get your code snippet](#toc2_2_3__) \n", - " - [Initialize the Python environment](#toc2_3__) \n", - "- [Banking Tools](#toc3__) \n", - " - [Tool Overview](#toc3_1__) \n", - " - [Test Banking Tools Individually](#toc3_2__) \n", - "- [Complete LangGraph Banking Agent](#toc4__) \n", - "- [ValidMind Model Integration](#toc5__) \n", - "- [Prompt Validation](#toc6__) \n", - "- [Banking Test Dataset](#toc7__) \n", - " - [Initialize ValidMind Dataset](#toc7_1__) \n", - " - [Run the Agent and capture result through assign predictions](#toc7_2__) \n", - " - [Dataframe Display Settings](#toc7_2_1__) \n", - "- [Banking Accuracy Test](#toc8__) \n", - "- [Banking Tool Call Accuracy Test](#toc9__) \n", - "- [Scorers in ValidMind](#toc10__)\n", - " - [Plan Quality Metric scorer](#toc10_1) \n", - " - [Plan Adherence Metric scorer](#toc10_2) \n", - " - [Tool Correctness Metric scorer](#toc10_3) \n", - " - [Argument Correctness Metric scorer](#toc10_4) \n", - " - [Task Completion scorer](#toc10_5) \n", - "- [RAGAS Tests for an Agent Evaluation](#toc12__) \n", - " - [Faithfulness](#toc12_1__) \n", - " - [Response Relevancy](#toc12_2__) \n", - " - [Context Recall](#toc12_3__) \n", - "- [Safety](#toc13__) \n", - " - [AspectCritic](#toc13_1__) \n", - " - [Prompt bias](#toc13_2__) \n", - " - [Toxicity](#toc13_3__) \n", - "- [Demo Summary and Next Steps](#toc14__) \n", - " - [What We Built](#toc14_1__) \n", - " - [Next Steps](#toc14_2__) \n", - " - [Key Benefits](#toc14_3__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language.\n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Key concepts\n", - "\n", - "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", - "\n", - "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", - "\n", - "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", - "\n", - "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", - "\n", - "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", - "\n", - "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", - "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", - "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", - "\n", - "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", - "\n", - "**Outputs**: Custom tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", - "\n", - "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q \"validmind[llm]\" " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Agentic AI System`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import validmind as vm\n", - "\n", - "vm.init(\n", - " api_host=\"...\",\n", - " api_key=\"...\",\n", - " api_secret=\"...\",\n", - " model=\"...\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the Python environment\n", - "\n", - "Next, let's import all the necessary libraries for building our banking LangGraph agent system:\n", - "\n", - "- **LangChain components** for LLM integration and tool management\n", - "- **LangGraph** for building stateful, multi-step agent workflows\n", - "- **ValidMind** for model validation and testing\n", - "- **Banking tools** for specialized financial services\n", - "- **Standard libraries** for data handling and environment management\n", - "\n", - "The setup includes loading environment variables (like OpenAI API keys) needed for the LLM components to function properly." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Standard library imports\n", - "from typing import TypedDict, Annotated, Sequence\n", - "\n", - "# Third party imports\n", - "import pandas as pd\n", - "from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage\n", - "from langchain_openai import ChatOpenAI\n", - "from langgraph.checkpoint.memory import MemorySaver\n", - "from langgraph.graph import StateGraph, END, START\n", - "from langgraph.graph.message import add_messages\n", - "from langgraph.prebuilt import ToolNode\n", - "\n", - "# Local imports\n", - "from banking_tools import AVAILABLE_TOOLS\n", - "from validmind.tests import run_test\n", - "\n", - "pd.set_option('display.max_columns', None)\n", - "pd.set_option('display.max_colwidth', None)\n", - "pd.set_option('display.width', None)\n", - "pd.set_option('display.max_rows', None)\n", - "\n", - "# Load environment variables if using .env file\n", - "try:\n", - " from dotenv import load_dotenv\n", - " load_dotenv()\n", - "except ImportError:\n", - " print(\"dotenv not installed. Make sure OPENAI_API_KEY is set in your environment.\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Banking Tools\n", - "\n", - "Now let's use the following banking demo tools that provide use cases of the financial services:\n", - "\n", - "\n", - "\n", - "### Tool Overview\n", - "1. **Credit Risk Analyzer** - Loan applications and credit decisions\n", - "2. **Customer Account Manager** - Account services and customer support\n", - "3. **Fraud Detection System** - Security and fraud prevention" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "print(f\"Available tools: {len(AVAILABLE_TOOLS)}\")\n", - "print(\"\\nTool Details:\")\n", - "for i, tool in enumerate(AVAILABLE_TOOLS, 1):\n", - " print(f\" - {tool.name}\") " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Test Banking Tools Individually\n", - "\n", - "Let's test each banking tool individually to ensure they're working correctly before integrating them into our agent." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "print(\"Testing Individual Banking Tools\")\n", - "print(\"=\" * 60)\n", - "\n", - "# Test 1: Credit Risk Analyzer\n", - "print(\"TEST 1: Credit Risk Analyzer\")\n", - "print(\"-\" * 40)\n", - "try:\n", - " # Access the underlying function using .func\n", - " credit_result = AVAILABLE_TOOLS[0].func(\n", - " customer_income=75000,\n", - " customer_debt=1200,\n", - " credit_score=720,\n", - " loan_amount=50000,\n", - " loan_type=\"personal\"\n", - " )\n", - " print(credit_result)\n", - " print(\"Credit Risk Analyzer test PASSED\")\n", - "except Exception as e:\n", - " print(f\"Credit Risk Analyzer test FAILED: {e}\")\n", - "\n", - "print(\"\" + \"=\" * 60)\n", - "\n", - "# Test 2: Customer Account Manager\n", - "print(\"TEST 2: Customer Account Manager\")\n", - "print(\"-\" * 40)\n", - "try:\n", - " # Test checking balance\n", - " account_result = AVAILABLE_TOOLS[1].func(\n", - " account_type=\"checking\",\n", - " customer_id=\"12345\",\n", - " action=\"check_balance\"\n", - " )\n", - " print(account_result)\n", - " \n", - " # Test getting account info\n", - " info_result = AVAILABLE_TOOLS[1].func(\n", - " account_type=\"all\",\n", - " customer_id=\"12345\", \n", - " action=\"get_info\"\n", - " )\n", - " print(info_result)\n", - " print(\"Customer Account Manager test PASSED\")\n", - "except Exception as e:\n", - " print(f\"Customer Account Manager test FAILED: {e}\")\n", - "\n", - "print(\"\" + \"=\" * 60)\n", - "\n", - "# Test 3: Fraud Detection System\n", - "print(\"TEST 3: Fraud Detection System\")\n", - "print(\"-\" * 40)\n", - "try:\n", - " fraud_result = AVAILABLE_TOOLS[2].func(\n", - " transaction_id=\"TX123\",\n", - " customer_id=\"12345\",\n", - " transaction_amount=500.00,\n", - " transaction_type=\"withdrawal\",\n", - " location=\"Miami, FL\",\n", - " device_id=\"DEVICE_001\"\n", - " )\n", - " print(fraud_result)\n", - " print(\"Fraud Detection System test PASSED\")\n", - "except Exception as e:\n", - " print(f\"Fraud Detection System test FAILED: {e}\")\n", - "\n", - "print(\"\" + \"=\" * 60)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Complete LangGraph Banking Agent\n", - "\n", - "Now we'll create our intelligent banking agent with LangGraph that can automatically select and use the appropriate banking tools based on user requests." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "# Enhanced banking system prompt with tool selection guidance\n", - "system_context = \"\"\"You are a professional banking AI assistant with access to specialized banking tools.\n", - " Analyze the user's banking request and directly use the most appropriate tools to help them.\n", - " \n", - " AVAILABLE BANKING TOOLS:\n", - " \n", - " credit_risk_analyzer - Analyze credit risk for loan applications and credit decisions\n", - " - Use for: loan applications, credit assessments, risk analysis, mortgage eligibility\n", - " - Examples: \"Analyze credit risk for $50k personal loan\", \"Assess mortgage eligibility for $300k home purchase\"\n", - " - Parameters: customer_income, customer_debt, credit_score, loan_amount, loan_type\n", - "\n", - " customer_account_manager - Manage customer accounts and provide banking services\n", - " - Use for: account information, transaction processing, product recommendations, customer service\n", - " - Examples: \"Check balance for checking account 12345\", \"Recommend products for customer with high balance\"\n", - " - Parameters: account_type, customer_id, action, amount, account_details\n", - "\n", - " fraud_detection_system - Analyze transactions for potential fraud and security risks\n", - " - Use for: transaction monitoring, fraud prevention, risk assessment, security alerts\n", - " - Examples: \"Analyze fraud risk for $500 ATM withdrawal in Miami\", \"Check security for $2000 online purchase\"\n", - " - Parameters: transaction_id, customer_id, transaction_amount, transaction_type, location, device_id\n", - "\n", - " BANKING INSTRUCTIONS:\n", - " - Analyze the user's banking request carefully and identify the primary need\n", - " - If they need credit analysis → use credit_risk_analyzer\n", - " - If they need financial calculations → use financial_calculator\n", - " - If they need account services → use customer_account_manager\n", - " - If they need security analysis → use fraud_detection_system\n", - " - Extract relevant parameters from the user's request\n", - " - Provide helpful, accurate banking responses based on tool outputs\n", - " - Always consider banking regulations, risk management, and best practices\n", - " - Be professional and thorough in your analysis\n", - "\n", - " Choose and use tools wisely to provide the most helpful banking assistance.\n", - " \"\"\"\n", - "# Initialize the main LLM for banking responses\n", - "main_llm = ChatOpenAI(\n", - " model=\"gpt-5-mini\",\n", - " reasoning={\n", - " \"effort\": \"low\",\n", - " \"summary\": \"auto\"\n", - " }\n", - ")\n", - "# Bind all banking tools to the main LLM\n", - "llm_with_tools = main_llm.bind_tools(AVAILABLE_TOOLS)\n", - "\n", - "# Banking Agent State Definition\n", - "class BankingAgentState(TypedDict):\n", - " messages: Annotated[Sequence[BaseMessage], add_messages]\n", - " user_input: str\n", - " session_id: str\n", - " context: dict\n", - "\n", - "def create_banking_langgraph_agent():\n", - " \"\"\"Create a comprehensive LangGraph banking agent with intelligent tool selection.\"\"\"\n", - " def llm_node(state: BankingAgentState) -> BankingAgentState:\n", - " \"\"\"Main LLM node that processes banking requests and selects appropriate tools.\"\"\"\n", - " messages = state[\"messages\"]\n", - " # Add system context to messages\n", - " enhanced_messages = [SystemMessage(content=system_context)] + list(messages)\n", - " # Get LLM response with tool selection\n", - " response = llm_with_tools.invoke(enhanced_messages)\n", - " return {\n", - " **state,\n", - " \"messages\": messages + [response]\n", - " }\n", - " \n", - " def should_continue(state: BankingAgentState) -> str:\n", - " \"\"\"Decide whether to use tools or end the conversation.\"\"\"\n", - " last_message = state[\"messages\"][-1]\n", - " # Check if the LLM wants to use tools\n", - " if hasattr(last_message, 'tool_calls') and last_message.tool_calls:\n", - " return \"tools\"\n", - " return END\n", - " \n", - " # Create the banking state graph\n", - " workflow = StateGraph(BankingAgentState)\n", - " # Add nodes\n", - " workflow.add_node(\"llm\", llm_node)\n", - " workflow.add_node(\"tools\", ToolNode(AVAILABLE_TOOLS))\n", - " # Simplified entry point - go directly to LLM\n", - " workflow.add_edge(START, \"llm\")\n", - " # From LLM, decide whether to use tools or end\n", - " workflow.add_conditional_edges(\n", - " \"llm\",\n", - " should_continue,\n", - " {\"tools\": \"tools\", END: END}\n", - " )\n", - " # Tool execution flows back to LLM for final response\n", - " workflow.add_edge(\"tools\", \"llm\")\n", - " # Set up memory\n", - " memory = MemorySaver()\n", - " # Compile the graph\n", - " agent = workflow.compile(checkpointer=memory)\n", - " return agent\n", - "\n", - "# Create the banking intelligent agent\n", - "banking_agent = create_banking_langgraph_agent()\n", - "\n", - "print(\"Banking LangGraph Agent Created Successfully!\")\n", - "print(\"\\nFeatures:\")\n", - "print(\" - Intelligent banking tool selection\")\n", - "print(\" - Comprehensive banking system prompt\")\n", - "print(\" - Streamlined workflow: LLM → Tools → Response\")\n", - "print(\" - Automatic tool parameter extraction\")\n", - "print(\" - Professional banking assistance\")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## ValidMind Model Integration\n", - "\n", - "Now we'll integrate our banking LangGraph agent with ValidMind for comprehensive testing and validation." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.models import Prompt\n", - "from validmind.scorers.llm.deepeval import extract_tool_calls_from_agent_output, _convert_to_tool_call_list\n", - "def banking_agent_fn(input):\n", - " \"\"\"\n", - " Invoke the banking agent with the given input.\n", - " \"\"\"\n", - " try:\n", - " # Initial state for banking agent\n", - " initial_state = {\n", - " \"user_input\": input[\"input\"],\n", - " \"messages\": [HumanMessage(content=input[\"input\"])],\n", - " \"session_id\": input[\"session_id\"],\n", - " \"context\": {}\n", - " }\n", - " session_config = {\"configurable\": {\"thread_id\": input[\"session_id\"]}}\n", - " result = banking_agent.invoke(initial_state, config=session_config)\n", - "\n", - " from utils import capture_tool_output_messages\n", - "\n", - " # Capture all tool outputs and metadata\n", - " captured_data = capture_tool_output_messages(result)\n", - " \n", - " # Access specific tool outputs, this will be used for RAGAS tests\n", - " tool_message = \"\"\n", - " for output in captured_data[\"tool_outputs\"]:\n", - " tool_message += output['content']\n", - " \n", - " tool_calls_found = []\n", - " messages = result['messages']\n", - " for message in messages:\n", - " if hasattr(message, 'tool_calls') and message.tool_calls:\n", - " for tool_call in message.tool_calls:\n", - " # Handle both dictionary and object formats\n", - " if isinstance(tool_call, dict):\n", - " tool_calls_found.append(tool_call['name'])\n", - " else:\n", - " # ToolCall object - use attribute access\n", - " tool_calls_found.append(tool_call.name)\n", - "\n", - "\n", - " return {\n", - " \"prediction\": result['messages'][-1].content[0]['text'],\n", - " \"output\": result,\n", - " \"tool_messages\": [tool_message],\n", - " # \"tool_calls\": tool_calls_found,\n", - " \"tool_called\": _convert_to_tool_call_list(extract_tool_calls_from_agent_output(result))\n", - " }\n", - " except Exception as e:\n", - " # Return a fallback response if the agent fails\n", - " error_message = f\"\"\"I apologize, but I encountered an error while processing your banking request: {str(e)}.\n", - " Please try rephrasing your question or contact support if the issue persists.\"\"\"\n", - " return {\n", - " \"prediction\": error_message, \n", - " \"output\": {\n", - " \"messages\": [HumanMessage(content=input[\"input\"]), SystemMessage(content=error_message)],\n", - " \"error\": str(e)\n", - " }\n", - " }\n", - "\n", - "## Initialize the model\n", - "vm_banking_model = vm.init_model(\n", - " input_id=\"banking_agent_model\",\n", - " predict_fn=banking_agent_fn,\n", - " prompt=Prompt(template=system_context)\n", - ")\n", - "\n", - "# Add the banking agent to the vm model\n", - "vm_banking_model.model = banking_agent\n", - "\n", - "print(\"Banking Agent Successfully Integrated with ValidMind!\")\n", - "print(f\"Model ID: {vm_banking_model.input_id}\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Prompt Validation\n", - "\n", - "Let's get an initial sense of how well the prompt meets a few best practices for prompt engineering. These tests use an LLM to rate the prompt on a scale of 1-10 against the following criteria:\n", - "\n", - "- **Clarity**: How clearly the prompt states the task.\n", - "- **Conciseness**: How succinctly the prompt states the task.\n", - "- **Delimitation**: When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", - "- **NegativeInstruction**: Whether the prompt contains negative instructions.\n", - "- **Specificity**: How specific the prompt defines the task." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Clarity\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Conciseness\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Delimitation\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.NegativeInstruction\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Specificity\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Banking Test Dataset\n", - "\n", - "We'll use a sample test dataset to evaluate our agent's performance across different banking scenarios.\n", - "\n", - "\n", - "\n", - "### Initialize ValidMind Dataset\n", - "\n", - "Before we can run tests and evaluations, we need to initialize our banking test dataset as a ValidMind dataset object." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import our banking-specific test dataset\n", - "from banking_test_dataset import banking_test_dataset\n", - "\n", - "vm_test_dataset = vm.init_dataset(\n", - " input_id=\"banking_test_dataset\",\n", - " dataset=banking_test_dataset.sample(2),\n", - " text_column=\"input\",\n", - " target_column=\"possible_outputs\",\n", - ")\n", - "\n", - "print(\"Banking Test Dataset Initialized in ValidMind!\")\n", - "print(f\"Dataset ID: {vm_test_dataset.input_id}\")\n", - "print(f\"Dataset columns: {vm_test_dataset._df.columns}\")\n", - "vm_test_dataset._df.head(1)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Run the Agent and capture result through assign predictions\n", - "\n", - "Now we'll execute our banking agent on the test dataset and capture its responses for evaluation." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_dataset.assign_predictions(vm_banking_model)\n", - "\n", - "print(\"Banking Agent Predictions Generated Successfully!\")\n", - "print(f\"Predictions assigned to {len(vm_test_dataset._df)} test cases\")\n", - "vm_test_dataset._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Banking Accuracy Test\n", - "\n", - "This test evaluates the banking agent's ability to provide accurate responses by:\n", - "- Testing against a dataset of predefined banking questions and expected answers\n", - "- Checking if responses contain expected keywords and banking terminology\n", - "- Providing detailed test results including pass/fail status\n", - "- Helping identify any gaps in the agent's banking knowledge or response quality" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "@vm.test(\"my_custom_tests.banking_accuracy_test\")\n", - "def banking_accuracy_test(model, dataset, list_of_columns):\n", - " \"\"\"\n", - " The Banking Accuracy Test evaluates whether the agent’s responses include \n", - " critical domain-specific keywords and phrases that indicate accurate, compliant,\n", - " and contextually appropriate banking information. This test ensures that the agent\n", - " provides responses containing the expected banking terminology, risk classifications,\n", - " account details, or other domain-relevant information required for regulatory compliance,\n", - " customer safety, and operational accuracy.\n", - " \"\"\"\n", - " df = dataset._df\n", - " \n", - " # Pre-compute responses for all tests\n", - " y_true = dataset.y.tolist()\n", - " y_pred = dataset.y_pred(model).tolist()\n", - "\n", - " # Vectorized test results\n", - " test_results = []\n", - " for response, keywords in zip(y_pred, y_true):\n", - " # Convert keywords to list if not already a list\n", - " if not isinstance(keywords, list):\n", - " keywords = [keywords]\n", - " test_results.append(any(str(keyword).lower() in str(response).lower() for keyword in keywords))\n", - " \n", - " results = pd.DataFrame()\n", - " column_names = [col + \"_details\" for col in list_of_columns]\n", - " results[column_names] = df[list_of_columns]\n", - " results[\"actual\"] = y_pred\n", - " results[\"expected\"] = y_true\n", - " results[\"passed\"] = test_results\n", - " results[\"error\"] = None if test_results else f'Response did not contain any expected keywords: {y_true}'\n", - " \n", - " return results\n", - " \n", - "result = run_test(\n", - " \"my_custom_tests.banking_accuracy_test\",\n", - " inputs={\n", - " \"dataset\": vm_test_dataset,\n", - " \"model\": vm_banking_model\n", - " },\n", - " params={\n", - " \"list_of_columns\": [\"input\"]\n", - " }\n", - ")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Banking Tool Call Accuracy Test\n", - "\n", - "This test evaluates how accurately our intelligent banking router selects the correct tools for different banking requests. This test provides quantitative feedback on the agent's core intelligence - its ability to understand what users need and select the right banking tools to help them." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.BankingToolCallAccuracy\")\n", - "def BankingToolCallAccuracy(dataset, agent_output_column, expected_tools_column):\n", - " \"\"\"\n", - " Evaluates the tool selection accuracy of a LangGraph-powered banking agent.\n", - "\n", - " This test measures whether the agent correctly identifies and invokes the required banking tools\n", - " for each user query scenario.\n", - " For each case, the outputs generated by the agent (including its tool calls) are compared against an\n", - " expected set of tools. The test considers both coverage and exactness: it computes the proportion of\n", - " expected tools correctly called by the agent for each instance.\n", - "\n", - " Parameters:\n", - " dataset (VMDataset): The dataset containing user queries, agent outputs, and ground-truth tool expectations.\n", - " agent_output_column (str): Dataset column name containing agent outputs (should include tool call details in 'messages').\n", - " expected_tools_column (str): Dataset column specifying the true expected tools (as lists).\n", - "\n", - " Returns:\n", - " List[dict]: Per-row dictionaries with details: expected tools, found tools, match count, total expected, and accuracy score.\n", - "\n", - " Purpose:\n", - " Provides diagnostic evidence of the banking agent's core reasoning ability—specifically, its capacity to\n", - " interpret user needs and select the correct banking actions. Useful for diagnosing gaps in tool coverage,\n", - " misclassifications, or breakdowns in agent logic.\n", - "\n", - " Interpretation:\n", - " - An accuracy of 1.0 signals perfect tool selection for that example.\n", - " - Lower scores may indicate partial or complete failures to invoke required tools.\n", - " - Review 'found_tools' vs. 'expected_tools' to understand the source of discrepancies.\n", - "\n", - " Strengths:\n", - " - Directly tests a core capability of compositional tool-use agents.\n", - " - Framework-agnostic; robust to tool call output format (object or dict).\n", - " - Supports batch validation and result logging for systematic documentation.\n", - "\n", - " Limitations:\n", - " - Does not penalize extra, unnecessary tool calls.\n", - " - Does not assess result quality—only correct invocation.\n", - "\n", - " \"\"\"\n", - " def validate_tool_calls_simple(messages, expected_tools):\n", - " \"\"\"Simple validation of tool calls without RAGAS dependency issues.\"\"\"\n", - " \n", - " tool_calls_found = []\n", - " \n", - " for message in messages:\n", - " if hasattr(message, 'tool_calls') and message.tool_calls:\n", - " for tool_call in message.tool_calls:\n", - " # Handle both dictionary and object formats\n", - " if isinstance(tool_call, dict):\n", - " tool_calls_found.append(tool_call['name'])\n", - " else:\n", - " # ToolCall object - use attribute access\n", - " tool_calls_found.append(tool_call.name)\n", - " \n", - " # Check if expected tools were called\n", - " accuracy = 0.0\n", - " matches = 0\n", - " if expected_tools:\n", - " matches = sum(1 for tool in expected_tools if tool in tool_calls_found)\n", - " accuracy = matches / len(expected_tools)\n", - " \n", - " return {\n", - " 'expected_tools': expected_tools,\n", - " 'found_tools': tool_calls_found,\n", - " 'matches': matches,\n", - " 'total_expected': len(expected_tools) if expected_tools else 0,\n", - " 'accuracy': accuracy,\n", - " }\n", - "\n", - " df = dataset._df\n", - " \n", - " results = []\n", - " for i, row in df.iterrows():\n", - " result = validate_tool_calls_simple(row[agent_output_column]['messages'], row[expected_tools_column])\n", - " results.append(result)\n", - " \n", - " return results\n", - "\n", - "run_test(\n", - " \"my_custom_tests.BankingToolCallAccuracy\",\n", - " inputs = {\n", - " \"dataset\": vm_test_dataset,\n", - " },\n", - " params = {\n", - " \"agent_output_column\": \"banking_agent_model_output\",\n", - " \"expected_tools_column\": \"expected_tools\"\n", - " }\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Scorers in ValidMind\n", - "\n", - "Scorers are evaluation metrics that analyze model outputs and store their results in the dataset. When using `assign_scores()`:\n", - "\n", - "- Each scorer adds a new column to the dataset with format: {scorer_name}_{metric_name}\n", - "- The column contains the numeric score (typically 0-1) for each example\n", - "- Multiple scorers can be run on the same dataset, each adding their own column\n", - "- Scores are persisted in the dataset for later analysis and visualization\n", - "- Common scorer patterns include:\n", - " - Model performance metrics (accuracy, F1, etc)\n", - " - Output quality metrics (relevance, faithfulness)\n", - " - Task-specific metrics (completion, correctness)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### AI Agent Evaluation Metrics\n", - "\n", - "AI agent evaluation metrics are specialized measurements designed to assess how well autonomous LLM-based agents reason, plan, select and execute tools, and ultimately complete user tasks by analyzing the **full execution trace**—including reasoning steps, tool calls, intermediate decisions, and outcomes—rather than just single input–output pairs.\n", - "\n", - "These metrics are essential because agent failures often occur in ways traditional LLM metrics miss (e.g., choosing the right tool with wrong arguments, creating a good plan but not following it, or completing a task inefficiently).\n", - "\n", - "**DeepEval’s AI agent evaluation framework** breaks evaluation into three layers with corresponding metric categories:\n", - "\n", - "1. **Reasoning Layer** – Evaluates planning and strategy generation:\n", - "\n", - " * *PlanQualityMetric* – how logical, complete, and efficient the agent’s plan is\n", - " * *PlanAdherenceMetric* – whether the agent follows its own plan during execution \n", - "\n", - "2. **Action Layer** – Assesses tool usage and argument generation:\n", - "\n", - " * *ToolCorrectnessMetric* – whether the agent selects and calls the right tools\n", - " * *ArgumentCorrectnessMetric* – whether the agent generates correct tool arguments\n", - "\n", - "3. **Execution Layer** – Measures end-to-end performance:\n", - "\n", - " * *TaskCompletionMetric* – whether the agent successfully completes the intended task\n", - " * *StepEfficiencyMetric* – whether the agent avoids unnecessary or redundant steps\n", - "\n", - "Together, these metrics enable granular diagnosis of agent behavior, help pinpoint where failures occur (reasoning, action, or execution), and support both development benchmarking and production monitoring." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### **Reasoning Layer**\n", - "#### PlanQualityMetric\n", - "Let's measures how well the agent generates a plan before acting. A high score means the plan is logical, complete, and efficient." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "vm_test_dataset.assign_scores(\n", - " metrics = \"validmind.scorers.llm.deepeval.PlanQuality\",\n", - " input_column = \"input\",\n", - " actual_output_column = \"banking_agent_model_prediction\",\n", - " tools_called_column = \"banking_agent_model_tool_called\",\n", - " agent_output_column = \"banking_agent_model_output\",\n", - ")\n", - "vm_test_dataset._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### PlanAdherenceMetric\n", - "Let's checks whether the agent follows the plan it created. Deviations lower this score and indicate gaps between reasoning and execution." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_dataset.assign_scores(\n", - " metrics = \"validmind.scorers.llm.deepeval.PlanAdherence\",\n", - " input_column = \"input\",\n", - " actual_output_column = \"banking_agent_model_prediction\",\n", - " expected_output_column = \"expected_output\",\n", - " tools_called_column = \"banking_agent_model_tool_called\",\n", - " agent_output_column = \"banking_agent_model_output\",\n", - "\n", - ")\n", - "vm_test_dataset._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### **Action Layer**\n", - "#### ToolCorrectnessMetric\n", - "Let's evaluates if the agent selects the appropriate tool for the task. Choosing the wrong tool reduces performance even if reasoning was correct." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_dataset.assign_scores(\n", - " metrics = \"validmind.scorers.llm.deepeval.ToolCorrectness\",\n", - " input_column = \"input\",\n", - " actual_output_column = \"banking_agent_model_prediction\",\n", - " tools_called_column = \"banking_agent_model_tool_called\",\n", - " expected_tools_column = \"expected_tools\",\n", - " agent_output_column = \"banking_agent_model_output\",\n", - "\n", - ")\n", - "vm_test_dataset._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### ArgumentCorrectnessMetric\n", - "Let's assesses whether the agent provides correct inputs or arguments to the selected tool. Incorrect arguments can lead to failed or unexpected results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_dataset.assign_scores(\n", - " metrics = \"validmind.scorers.llm.deepeval.ArgumentCorrectness\",\n", - " input_column = \"input\",\n", - " actual_output_column = \"banking_agent_model_prediction\",\n", - " tools_called_column = \"banking_agent_model_tool_called\",\n", - " agent_output_column = \"banking_agent_model_output\",\n", - "\n", - ")\n", - "vm_test_dataset._df.head()\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### **Execution Layer**\n", - "#### TaskCompletionMetric\n", - "The TaskCompletion test evaluates whether our banking agent successfully completes the requested tasks by analyzing its outputs and tool usage. This metric assesses the agent's ability to understand user requests, execute appropriate actions, and provide complete responses that address the original query. The test provides a score between 0-1 along with detailed feedback on task completion quality." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_dataset.assign_scores(\n", - " metrics = \"validmind.scorers.llm.deepeval.TaskCompletion\",\n", - " input_column = \"input\",\n", - " actual_output_column = \"banking_agent_model_prediction\",\n", - " agent_output_column = \"banking_agent_model_output\",\n", - " tools_called_column = \"banking_agent_model_tool_called\",\n", - "\n", - ")\n", - "vm_test_dataset._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The TaskCompletion scorer has added a new column 'TaskCompletion_score' to our dataset. This is because when we run scorers through assign_scores(), the return values are automatically processed and added as new columns with the format {scorer_name}_{metric_name}. We'll use this column to visualize the distribution of task completion scores across our test cases. Let's visualize the distribution through the box plot test." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.plots.BoxPlot\",\n", - " inputs={\"dataset\": vm_test_dataset},\n", - " params={\n", - " \"columns\": \"TaskCompletion_score\",\n", - " \"title\": \"Distribution of Task Completion Scores\",\n", - " \"ylabel\": \"Score\",\n", - " \"figsize\": (8, 6)\n", - " }\n", - ").log()\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## RAGAS Tests for an Agent Evaluation\n", - "\n", - "RAGAS (Retrieval-Augmented Generation Assessment) provides specialized metrics for evaluating conversational AI systems like our banking agent. These tests analyze different aspects of agent performance:\n", - "\n", - "Our banking agent uses tools to retrieve information and generates responses based on that context, making it similar to a RAG system. RAGAS metrics help evaluate:\n", - "\n", - "- **Response Quality**: How well the agent uses retrieved tool outputs to generate helpful banking responses\n", - "- **Information Faithfulness**: Whether agent responses accurately reflect tool outputs \n", - "- **Relevance Assessment**: How well responses address the original banking query\n", - "- **Context Utilization**: How effectively the agent incorporates tool results into final answers\n", - "\n", - "These tests provide insights into how well our banking agent integrates tool usage with conversational abilities, ensuring it provides accurate, relevant, and helpful responses to banking users." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Faithfulness\n", - "\n", - "Faithfulness measures how accurately the banking agent's responses reflect the information retrieved from tools. This metric evaluates:\n", - "\n", - "**Information Accuracy**: Whether the agent correctly uses tool outputs in its responses\n", - "- **Fact Preservation**: Ensuring credit scores, loan calculations, compliance results are accurately reported\n", - "- **No Hallucination**: Verifying the agent doesn't invent banking information not provided by tools\n", - "- **Source Attribution**: Checking that responses align with actual tool outputs\n", - "\n", - "**Critical for Banking Trust**: Faithfulness is essential for banking agent reliability because users need to trust that:\n", - "- Credit analysis results are reported correctly\n", - "- Financial calculations are accurate \n", - "- Compliance checks return real information\n", - "- Risk assessments are properly communicated" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.Faithfulness\",\n", - " inputs={\"dataset\": vm_test_dataset},\n", - " param_grid={\n", - " \"user_input_column\": [\"input\"],\n", - " \"response_column\": [\"banking_agent_model_prediction\"],\n", - " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Response Relevancy\n", - "\n", - "Response Relevancy evaluates how well the banking agent's answers address the user's original banking question or request. This metric assesses:\n", - "\n", - "**Query Alignment**: Whether responses directly answer what users asked for\n", - "- **Intent Fulfillment**: Checking if the agent understood and addressed the user's actual banking need\n", - "- **Completeness**: Ensuring responses provide sufficient information to satisfy the banking query\n", - "- **Focus**: Avoiding irrelevant information that doesn't help the banking user\n", - "\n", - "**Banking Quality**: Measures the agent's ability to maintain relevant, helpful banking dialogue\n", - "- **Context Awareness**: Responses should be appropriate for the banking conversation context\n", - "- **User Satisfaction**: Answers should be useful and actionable for banking users\n", - "- **Clarity**: Banking information should be presented in a way that directly helps the user\n", - "\n", - "High relevancy indicates the banking agent successfully understands user needs and provides targeted, helpful banking responses." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", - " inputs={\"dataset\": vm_test_dataset},\n", - " params={\n", - " \"user_input_column\": \"input\",\n", - " \"response_column\": \"banking_agent_model_prediction\",\n", - " \"retrieved_contexts_column\": \"banking_agent_model_tool_messages\",\n", - " }\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Recall\n", - "\n", - "Context Recall measures how well the banking agent utilizes the information retrieved from tools when generating its responses. This metric evaluates:\n", - "\n", - "**Information Utilization**: Whether the agent effectively incorporates tool outputs into its responses\n", - "- **Coverage**: How much of the available tool information is used in the response\n", - "- **Integration**: How well tool outputs are woven into coherent, natural banking responses\n", - "- **Completeness**: Whether all relevant information from tools is considered\n", - "\n", - "**Tool Effectiveness**: Assesses whether selected banking tools provide useful context for responses\n", - "- **Relevance**: Whether tool outputs actually help answer the user's banking question\n", - "- **Sufficiency**: Whether enough information was retrieved to generate good banking responses\n", - "- **Quality**: Whether the tools provided accurate, helpful banking information\n", - "\n", - "High context recall indicates the banking agent not only selects the right tools but also effectively uses their outputs to create comprehensive, well-informed banking responses." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextRecall\",\n", - " inputs={\"dataset\": vm_test_dataset},\n", - " param_grid={\n", - " \"user_input_column\": [\"input\"],\n", - " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", - " \"reference_column\": [\"banking_agent_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Safety\n", - "\n", - "Safety testing is critical for banking AI agents to ensure they operate reliably and securely.\n", - "These tests help validate that our banking agent maintains high standards of fairness and professionalism." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### AspectCritic\n", - "\n", - "AspectCritic provides comprehensive evaluation across multiple dimensions of banking agent performance. This metric analyzes various aspects of response quality:\n", - "\n", - "**Multi-Dimensional Assessment**: Evaluates responses across different quality criteria:\n", - " - **Conciseness**: Whether responses are clear and to-the-point without unnecessary details\n", - " - **Coherence**: Whether responses are logically structured and easy to follow\n", - " - **Correctness**: Accuracy of banking information and appropriateness of recommendations\n", - " - **Harmfulness**: Whether responses could cause harm or damage to users or systems\n", - " - **Maliciousness**: Whether responses contain malicious content or intent\n", - "\n", - "**Holistic Quality Scoring**: Provides an overall assessment that considers:\n", - "- **User Experience**: How satisfying and useful the banking interaction would be for real users\n", - "- **Professional Standards**: Whether responses meet quality expectations for production banking systems\n", - "- **Consistency**: Whether the banking agent maintains quality across different types of requests\n", - "\n", - "AspectCritic helps identify specific areas where the banking agent excels or needs improvement, providing actionable insights for enhancing overall performance and user satisfaction in banking scenarios." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.AspectCritic\",\n", - " inputs={\"dataset\": vm_test_dataset},\n", - " param_grid={\n", - " \"user_input_column\": [\"input\"],\n", - " \"response_column\": [\"banking_agent_model_prediction\"],\n", - " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Prompt bias\n", - "\n", - "Let's check if the agent's prompts contain unintended biases that could affect banking decisions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Bias\",\n", - " inputs={\n", - " \"model\": vm_banking_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Toxicity\n", - "\n", - "Let's ensure responses are professional and appropriate for banking contexts." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.Toxicity\",\n", - " inputs={\n", - " \"dataset\": vm_test_dataset,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Demo Summary and Next Steps\n", - "\n", - "We have successfully built and tested a comprehensive **Banking AI Agent** using LangGraph and ValidMind. Here's what we've accomplished:\n", - "\n", - "\n", - "\n", - "### What We Built\n", - "\n", - "1. **5 Specialized Banking Tools**\n", - " - Credit Risk Analyzer for loan assessments\n", - " - Customer Account Manager for account services\n", - " - Fraud Detection System for security monitoring\n", - "\n", - "2. **Intelligent LangGraph Agent**\n", - " - Automatic tool selection based on user requests\n", - " - Banking-specific system prompts and guidance\n", - " - Professional banking assistance and responses\n", - "\n", - "3. **Comprehensive Testing Framework**\n", - " - banking-specific test cases\n", - " - ValidMind integration for validation\n", - " - Performance analysis across banking domains\n", - "\n", - "\n", - "\n", - "### Next Steps\n", - "\n", - "1. **Customize Tools**: Adapt the banking tools to your specific banking requirements\n", - "2. **Expand Test Cases**: Add more banking scenarios and edge cases\n", - "3. **Integrate with Real Data**: Connect to actual banking systems and databases\n", - "4. **Add More Tools**: Implement additional banking-specific functionality\n", - "5. **Production Deployment**: Deploy the agent in a production banking environment\n", - "\n", - "\n", - "\n", - "### Key Benefits\n", - "\n", - "- **Industry-Specific**: Designed specifically for banking operations\n", - "- **Regulatory Compliance**: Built-in SR 11-7 and SS 1-23 compliance checks\n", - "- **Risk Management**: Comprehensive credit and fraud risk assessment\n", - "- **Customer Focus**: Tools for both retail and commercial banking needs\n", - "- **Real-World Applicability**: Addresses actual banking use cases and challenges\n", - "\n", - "Your banking AI agent is now ready to handle real-world banking scenarios while maintaining regulatory compliance and risk management best practices!" - ] - }, - { - "cell_type": "markdown", - "id": "copyright-e7184e5605bb4f85b3d7b8306aaaef78", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind (Poetry)", - "language": "python", - "name": "validmind" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.9" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb b/site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb deleted file mode 100644 index 1269254add..0000000000 --- a/site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb +++ /dev/null @@ -1,1330 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "1e2a4689", - "metadata": {}, - "source": [ - "# Quickstart for Heston option pricing model using QuantLib\n", - "\n", - "Welcome! Let's get you started with the basic process of documenting models with ValidMind.\n", - "\n", - "The Heston option pricing model is a popular stochastic volatility model used to price options. Developed by Steven Heston in 1993, the model assumes that the asset's volatility follows a mean-reverting square-root process, allowing it to capture the empirical observation of volatility \"clustering\" in financial markets. This model is particularly useful for assets where volatility is not constant, making it a favored approach in quantitative finance for pricing complex derivatives.\n", - "\n", - "Here’s an overview of the Heston model as implemented in QuantLib, a powerful library for quantitative finance:\n", - "\n", - "\n", - "\n", - "### Model Assumptions and Characteristics\n", - "1. **Stochastic Volatility**: The volatility is modeled as a stochastic process, following a mean-reverting square-root process (Cox-Ingersoll-Ross process).\n", - "2. **Correlated Asset and Volatility Processes**: The asset price and volatility are assumed to be correlated, allowing the model to capture the \"smile\" effect observed in implied volatilities.\n", - "3. **Risk-Neutral Dynamics**: The Heston model is typically calibrated under a risk-neutral measure, which allows for direct application to pricing.\n", - "\n", - "\n", - "\n", - "### Heston Model Parameters\n", - "The model is governed by a set of key parameters:\n", - "- **S0**: Initial stock price\n", - "- **v0**: Initial variance of the asset price\n", - "- **kappa**: Speed of mean reversion of the variance\n", - "- **theta**: Long-term mean level of variance\n", - "- **sigma**: Volatility of volatility (vol of vol)\n", - "- **rho**: Correlation between the asset price and variance processes\n", - "\n", - "The dynamics of the asset price \\( S \\) and the variance \\( v \\) under the Heston model are given by:\n", - "\n", - "$$\n", - "dS_t = r S_t \\, dt + \\sqrt{v_t} S_t \\, dW^S_t\n", - "$$\n", - "\n", - "$$\n", - "dv_t = \\kappa (\\theta - v_t) \\, dt + \\sigma \\sqrt{v_t} \\, dW^v_t\n", - "$$\n", - "\n", - "where \\( $dW^S$ \\) and \\( $dW^v$ \\) are Wiener processes with correlation \\( $\\rho$ \\).\n", - "\n", - "\n", - "\n", - "### Advantages and Limitations\n", - "- **Advantages**:\n", - " - Ability to capture volatility smiles and skews.\n", - " - More realistic pricing for options on assets with stochastic volatility.\n", - "- **Limitations**:\n", - " - Calibration can be complex due to the number of parameters.\n", - " - Computationally intensive compared to simpler models like Black-Scholes.\n", - "\n", - "This setup provides a robust framework for pricing and analyzing options with stochastic volatility dynamics. QuantLib’s implementation makes it easy to experiment with different parameter configurations and observe their effects on pricing.\n", - "\n", - "You will learn how to initialize the ValidMind Library, develop a option pricing model, and then write custom tests that can be used for sensitivity and stress testing to quickly generate documentation about model." - ] - }, - { - "cell_type": "markdown", - "id": "69ec219a", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - " - [Model Assumptions and Characteristics](#toc1_1__) \n", - " - [Heston Model Parameters](#toc1_2__) \n", - " - [Advantages and Limitations](#toc1_3__) \n", - "- [About ValidMind](#toc2__) \n", - " - [Before you begin](#toc2_1__) \n", - " - [New to ValidMind?](#toc2_2__) \n", - " - [Key concepts](#toc2_3__) \n", - "- [Setting up](#toc3__) \n", - " - [Install the ValidMind Library](#toc3_1__) \n", - " - [Initialize the ValidMind Library](#toc3_2__) \n", - " - [Register sample model](#toc3_2_1__) \n", - " - [Apply documentation template](#toc3_2_2__) \n", - " - [Get your code snippet](#toc3_2_3__) \n", - " - [Initialize the Python environment](#toc3_3__) \n", - " - [Preview the documentation template](#toc3_4__) \n", - "- [Data Preparation](#toc4__) \n", - " - [Helper functions](#toc4_1_1__) \n", - " - [Market Data Quality and Availability](#toc4_2__) \n", - " - [Initialize the ValidMind datasets](#toc4_3__) \n", - " - [Data Quality](#toc4_4__) \n", - " - [Isolation Forest Outliers Test](#toc4_4_1__) \n", - " - [Model parameters](#toc4_4_2__) \n", - "- [Model development - Heston Option price](#toc5__) \n", - " - [Model Calibration](#toc5_1__) \n", - " - [Model Evaluation](#toc5_2__) \n", - " - [Benchmark Testing](#toc5_2_1__) \n", - " - [Sensitivity Testing](#toc5_2_2__) \n", - " - [Stress Testing](#toc5_2_3__) \n", - "- [Next steps](#toc6__) \n", - " - [Work with your model documentation](#toc6_1__) \n", - " - [Discover more learning resources](#toc6_2__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "id": "b9fb5d17", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", - "\n", - "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", - "\n", - "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", - "\n", - "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", - "\n", - "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", - "\n", - " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", - " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", - " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", - "\n", - "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", - "\n", - "**Outputs**: Custom tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", - "\n", - "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", - "\n", - "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." - ] - }, - { - "cell_type": "markdown", - "id": "f2dccf35", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "id": "5a5ce085", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "409352bf", - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "id": "65e870b2", - "metadata": {}, - "source": [ - "To install the QuantLib library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "3a34debf", - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q QuantLib" - ] - }, - { - "cell_type": "markdown", - "id": "fb30ae07", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "id": "c6f87017", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "id": "cbb2e2c9", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Capital markets`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "id": "2012eb82", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "0cd3f67e", - "metadata": {}, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " # api_host=\"...\",\n", - " # api_key=\"...\",\n", - " # api_secret=\"...\",\n", - " # model=\"...\",\n", - ")\n" - ] - }, - { - "cell_type": "markdown", - "id": "6d944cc9", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the Python environment\n", - "\n", - "Next, let's import the necessary libraries and set up your Python environment for data analysis:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "f8cf2746", - "metadata": {}, - "outputs": [], - "source": [ - "%matplotlib inline\n", - "\n", - "import pandas as pd\n", - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "from scipy.optimize import minimize\n", - "import yfinance as yf\n", - "import QuantLib as ql\n", - "from validmind.tests import run_test" - ] - }, - { - "cell_type": "markdown", - "id": "bc431ee0", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Preview the documentation template\n", - "\n", - "Let's verify that you have connected the ValidMind Library to the ValidMind Platform and that the appropriate *template* is selected for your model.\n", - "\n", - "You will upload documentation and test results unique to your model based on this template later on. For now, **take a look at the default structure that the template provides with [the `vm.preview_template()` function](https://docs.validmind.ai/validmind/validmind.html#preview_template)** from the ValidMind library and note the empty sections:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "7e844028", - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "id": "0c0ee8b9", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Data Preparation" - ] - }, - { - "cell_type": "markdown", - "id": "5a4d2c36", - "metadata": {}, - "source": [ - "### Market Data Sources\n", - "\n", - "\n", - "\n", - "#### Helper functions\n", - "Let's define helper function retrieve to option data from Yahoo Finance." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b96a500f", - "metadata": {}, - "outputs": [], - "source": [ - "def get_market_data(ticker, expiration_date_str):\n", - " \"\"\"\n", - " Fetch option market data from Yahoo Finance for the given ticker and expiration date.\n", - " Returns a list of tuples: (strike, maturity, option_price).\n", - " \"\"\"\n", - " # Create a Ticker object for the specified stock\n", - " stock = yf.Ticker(ticker)\n", - "\n", - " # Get all available expiration dates for options\n", - " option_dates = stock.options\n", - "\n", - " # Check if the requested expiration date is available\n", - " if expiration_date_str not in option_dates:\n", - " raise ValueError(f\"Expiration date {expiration_date_str} not available for {ticker}. Available dates: {option_dates}\")\n", - "\n", - " # Get the option chain for the specified expiration date\n", - " option_chain = stock.option_chain(expiration_date_str)\n", - "\n", - " # Get call options (or you can use puts as well based on your requirement)\n", - " calls = option_chain.calls\n", - "\n", - " # Convert expiration_date_str to QuantLib Date\n", - " expiry_date_parts = list(map(int, expiration_date_str.split('-'))) # Split YYYY-MM-DD\n", - " maturity_date = ql.Date(expiry_date_parts[2], expiry_date_parts[1], expiry_date_parts[0]) # Convert to QuantLib Date\n", - "\n", - " # Create a list to store strike prices, maturity dates, and option prices\n", - " market_data = []\n", - " for index, row in calls.iterrows():\n", - " strike = row['strike']\n", - " option_price = row['lastPrice'] # You can also use 'bid', 'ask', 'mid', etc.\n", - " market_data.append((strike, maturity_date, option_price))\n", - " df = pd.DataFrame(market_data, columns = ['strike', 'maturity_date', 'option_price'])\n", - " return df" - ] - }, - { - "cell_type": "markdown", - "id": "c7769b73", - "metadata": {}, - "source": [ - "Let's define helper function retrieve to stock data from Yahoo Finance. This helper function to calculate spot price, dividend yield, volatility and risk free rate using the underline stock data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "dc44c448", - "metadata": {}, - "outputs": [], - "source": [ - "def get_option_parameters(ticker):\n", - " # Fetch historical data for the stock\n", - " stock_data = yf.Ticker(ticker)\n", - " \n", - " # Get the current spot price\n", - " spot_price = stock_data.history(period=\"1d\")['Close'].iloc[-1]\n", - " \n", - " # Get dividend yield\n", - " dividend_rate = stock_data.dividends.mean() / spot_price if not stock_data.dividends.empty else 0.0\n", - " \n", - " # Estimate volatility (standard deviation of log returns)\n", - " hist_data = stock_data.history(period=\"1y\")['Close']\n", - " log_returns = np.log(hist_data / hist_data.shift(1)).dropna()\n", - " volatility = np.std(log_returns) * np.sqrt(252) # Annualized volatility\n", - " \n", - " # Assume a risk-free rate from some known data (can be fetched from market data, here we use 0.001)\n", - " risk_free_rate = 0.001\n", - " \n", - " # Return the calculated parameters\n", - " return {\n", - " \"spot_price\": spot_price,\n", - " \"volatility\": volatility,\n", - " \"dividend_rate\": dividend_rate,\n", - " \"risk_free_rate\": risk_free_rate\n", - " }" - ] - }, - { - "cell_type": "markdown", - "id": "c7b739d3", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Market Data Quality and Availability\n", - "Next, let's specify ticker and expiration date to get market data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "50225fde", - "metadata": {}, - "outputs": [], - "source": [ - "ticker = \"MSFT\"\n", - "expiration_date = \"2024-12-13\" # Example expiration date in 'YYYY-MM-DD' form\n", - "\n", - "market_data = get_market_data(ticker=ticker, expiration_date_str=expiration_date)" - ] - }, - { - "cell_type": "markdown", - "id": "c539b95e", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind datasets\n", - "\n", - "Before you can run tests, you must first initialize a ValidMind dataset object using the [`init_dataset`](https://docs.validmind.ai/validmind/validmind.html#init_dataset) function from the ValidMind (`vm`) module." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "113f9c17", - "metadata": {}, - "outputs": [], - "source": [ - "vm_market_data = vm.init_dataset(\n", - " dataset=market_data,\n", - " input_id=\"market_data\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "185beb24", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Data Quality\n", - "Let's check quality of the data using outliers and missing data tests." - ] - }, - { - "cell_type": "markdown", - "id": "7f14464c", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Isolation Forest Outliers Test\n", - "Let's detects anomalies in the dataset using the Isolation Forest algorithm, visualized through scatter plots." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "56c919ec", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"validmind.data_validation.IsolationForestOutliers\",\n", - " inputs={\n", - " \"dataset\": vm_market_data,\n", - " },\n", - " title=\"Outliers detection using Isolation Forest\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "e4d0e5ca", - "metadata": {}, - "source": [ - "##### Missing Values Test\n", - "Let's evaluates dataset quality by ensuring the missing value ratio across all features does not exceed a set threshold." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e95c825f", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"validmind.data_validation.MissingValues\",\n", - " inputs={\n", - " \"dataset\": vm_market_data,\n", - " },\n", - " title=\"Missing Values detection\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "829403a3", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Model parameters\n", - "Let's calculate the model parameters using from stock data " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "25936449", - "metadata": {}, - "outputs": [], - "source": [ - "option_params = get_option_parameters(ticker=ticker)" - ] - }, - { - "cell_type": "markdown", - "id": "0a0948b6", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Model development - Heston Option price" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e15b8221", - "metadata": {}, - "outputs": [], - "source": [ - "class HestonModel:\n", - "\n", - " def __init__(self, ticker, expiration_date_str, calculation_date, spot_price, dividend_rate, risk_free_rate):\n", - " self.ticker = ticker\n", - " self.expiration_date_str = expiration_date_str,\n", - " self.calculation_date = calculation_date\n", - " self.spot_price = spot_price\n", - " self.dividend_rate = dividend_rate\n", - " self.risk_free_rate = risk_free_rate\n", - " \n", - " def predict_option_price(self, strike, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", - " # Set the evaluation date\n", - " ql.Settings.instance().evaluationDate = self.calculation_date\n", - "\n", - " # Construct the European Option\n", - " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", - " exercise = ql.EuropeanExercise(maturity_date)\n", - " european_option = ql.VanillaOption(payoff, exercise)\n", - "\n", - " # Yield term structures for risk-free rate and dividend\n", - " riskFreeTS = ql.YieldTermStructureHandle(ql.FlatForward(calculation_date, self.risk_free_rate, ql.Actual365Fixed()))\n", - " dividendTS = ql.YieldTermStructureHandle(ql.FlatForward(calculation_date, self.dividend_rate, ql.Actual365Fixed()))\n", - "\n", - " # Initial stock price\n", - " initialValue = ql.QuoteHandle(ql.SimpleQuote(spot_price))\n", - "\n", - " # Heston process parameters\n", - " heston_process = ql.HestonProcess(riskFreeTS, dividendTS, initialValue, v0, kappa, theta, sigma, rho)\n", - " hestonModel = ql.HestonModel(heston_process)\n", - "\n", - " # Use the Heston analytic engine\n", - " engine = ql.AnalyticHestonEngine(hestonModel)\n", - " european_option.setPricingEngine(engine)\n", - "\n", - " # Calculate the Heston model price\n", - " h_price = european_option.NPV()\n", - "\n", - " return h_price\n", - "\n", - " def predict_american_option_price(self, strike, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", - " # Set the evaluation date\n", - " ql.Settings.instance().evaluationDate = self.calculation_date\n", - "\n", - " # Construct the American Option\n", - " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", - " exercise = ql.AmericanExercise(self.calculation_date, maturity_date)\n", - " american_option = ql.VanillaOption(payoff, exercise)\n", - "\n", - " # Yield term structures for risk-free rate and dividend\n", - " riskFreeTS = ql.YieldTermStructureHandle(ql.FlatForward(self.calculation_date, self.risk_free_rate, ql.Actual365Fixed()))\n", - " dividendTS = ql.YieldTermStructureHandle(ql.FlatForward(self.calculation_date, self.dividend_rate, ql.Actual365Fixed()))\n", - "\n", - " # Initial stock price\n", - " initialValue = ql.QuoteHandle(ql.SimpleQuote(spot_price))\n", - "\n", - " # Heston process parameters\n", - " heston_process = ql.HestonProcess(riskFreeTS, dividendTS, initialValue, v0, kappa, theta, sigma, rho)\n", - " heston_model = ql.HestonModel(heston_process)\n", - "\n", - "\n", - " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", - " exercise = ql.AmericanExercise(self.calculation_date, maturity_date)\n", - " american_option = ql.VanillaOption(payoff, exercise)\n", - " heston_fd_engine = ql.FdHestonVanillaEngine(heston_model)\n", - " american_option.setPricingEngine(heston_fd_engine)\n", - " option_price = american_option.NPV()\n", - "\n", - " return option_price\n", - "\n", - " def objective_function(self, params, market_data, spot_price, dividend_rate, risk_free_rate):\n", - " v0, theta, kappa, sigma, rho = params\n", - "\n", - " # Sum of squared differences between market prices and model prices\n", - " error = 0.0\n", - " for i, row in market_data.iterrows():\n", - " model_price = self.predict_option_price(row['strike'], row['maturity_date'], spot_price, \n", - " v0, theta, kappa, sigma, rho)\n", - " error += (model_price - row['option_price']) ** 2\n", - " \n", - " return error\n", - "\n", - " def calibrate_model(self, ticker, expiration_date_str):\n", - " # Get the option market data dynamically from Yahoo Finance\n", - " market_data = get_market_data(ticker, expiration_date_str)\n", - "\n", - " # Initial guesses for Heston parameters\n", - " initial_params = [0.04, 0.04, 0.1, 0.1, -0.75]\n", - "\n", - " # Bounds for the parameters to ensure realistic values\n", - " bounds = [(0.0001, 1.0), # v0\n", - " (0.0001, 1.0), # theta\n", - " (0.001, 2.0), # kappa\n", - " (0.001, 1.0), # sigma\n", - " (-0.75, 0.0)] # rho\n", - "\n", - " # Optimize the parameters to minimize the error between model and market prices\n", - " result = minimize(self.objective_function, initial_params, args=(market_data, self.spot_price, self.dividend_rate, self.risk_free_rate),\n", - " bounds=bounds, method='L-BFGS-B')\n", - "\n", - " # Optimized Heston parameters\n", - " v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt = result.x\n", - "\n", - " return v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt\n" - ] - }, - { - "cell_type": "markdown", - "id": "a941aa32", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Model Calibration\n", - "* The calibration process aims to optimize the Heston model parameters (v0, theta, kappa, sigma, rho) by minimizing the difference between model-predicted option prices and observed market prices.\n", - "* In this implementation, the model is calibrated to current market data, specifically using option prices from the selected ticker and expiration date.\n", - "\n", - "Let's specify `calculation_date` and `strike_price` as input parameters for the model to verify its functionality and confirm it operates as expected." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "1d61dfca", - "metadata": {}, - "outputs": [], - "source": [ - "calculation_date = ql.Date(26, 11, 2024)\n", - "# Convert expiration date string to QuantLib.Date\n", - "expiry_date_parts = list(map(int, expiration_date.split('-')))\n", - "maturity_date = ql.Date(expiry_date_parts[2], expiry_date_parts[1], expiry_date_parts[0])\n", - "strike_price = 460.0\n", - "\n", - "hm = HestonModel(\n", - " ticker=ticker,\n", - " expiration_date_str= expiration_date,\n", - " calculation_date= calculation_date,\n", - " spot_price= option_params['spot_price'],\n", - " dividend_rate = option_params['dividend_rate'],\n", - " risk_free_rate = option_params['risk_free_rate']\n", - ")\n", - "\n", - "# Let's calibrate model\n", - "v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt = hm.calibrate_model(ticker, expiration_date)\n", - "print(f\"Optimized Heston parameters: v0={v0_opt}, theta={theta_opt}, kappa={kappa_opt}, sigma={sigma_opt}, rho={rho_opt}\")\n", - "\n", - "\n", - "# option price\n", - "h_price = hm.predict_option_price(strike_price, maturity_date, option_params['spot_price'], v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt)\n", - "print(\"The Heston model price for the option is:\", h_price)" - ] - }, - { - "cell_type": "markdown", - "id": "75313272", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Model Evaluation" - ] - }, - { - "cell_type": "markdown", - "id": "2e6471ef", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Benchmark Testing\n", - "The benchmark testing framework provides a robust way to validate the Heston model implementation and understand the relationships between European and American option prices under stochastic volatility conditions.\n", - "Let's compares European and American option prices using the Heston model." - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "id": "810cf887", - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.BenchmarkTest\")\n", - "def benchmark_test(hm_model, strikes, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", - " \"\"\"\n", - " Compares European and American option prices using the Heston model.\n", - "\n", - " This test evaluates the price differences between European and American options\n", - " across multiple strike prices while keeping other parameters constant. The comparison\n", - " helps understand the early exercise premium of American options over their European\n", - " counterparts under stochastic volatility conditions.\n", - "\n", - " Args:\n", - " hm_model: HestonModel instance for option pricing calculations\n", - " strikes (list[float]): List of strike prices to test\n", - " maturity_date (ql.Date): Option expiration date in QuantLib format\n", - " spot_price (float): Current price of the underlying asset\n", - " v0 (float, optional): Initial variance. Defaults to None.\n", - " theta (float, optional): Long-term variance. Defaults to None.\n", - " kappa (float, optional): Mean reversion rate. Defaults to None.\n", - " sigma (float, optional): Volatility of variance. Defaults to None.\n", - " rho (float, optional): Correlation between asset and variance. Defaults to None.\n", - "\n", - " Returns:\n", - " dict: Contains a DataFrame with the following columns:\n", - " - Strike: Strike prices tested\n", - " - Maturity date: Expiration date for all options\n", - " - Spot price: Current underlying price\n", - " - european model price: Prices for European options\n", - " - american model price: Prices for American options\n", - "\"\"\"\n", - " american_derived_prices = []\n", - " european_derived_prices = []\n", - " for K in strikes:\n", - " european_derived_prices.append(hm_model.predict_option_price(K, maturity_date, spot_price, v0, theta, kappa, sigma, rho))\n", - " american_derived_prices.append(hm_model.predict_american_option_price(K, maturity_date, spot_price, v0, theta, kappa, sigma, rho))\n", - "\n", - " data = {\n", - " \"Strike\": strikes,\n", - " \"Maturity date\": [maturity_date] * len(strikes),\n", - " \"Spot price\": [spot_price] * len(strikes),\n", - " \"european model price\": european_derived_prices,\n", - " \"american model price\": american_derived_prices,\n", - "\n", - " }\n", - " df1 = pd.DataFrame(data)\n", - " return {\"strikes variation benchmarking\": df1}" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "3fdd6705", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.BenchmarkTest\",\n", - " params={\n", - " \"hm_model\": hm,\n", - " \"strikes\": [400, 425, 460, 495, 520],\n", - " \"maturity_date\": maturity_date,\n", - " \"spot_price\": option_params['spot_price'],\n", - " \"v0\":v0_opt,\n", - " \"theta\": theta_opt,\n", - " \"kappa\":kappa_opt ,\n", - " \"sigma\": sigma_opt,\n", - " \"rho\":rho_opt\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "e359b503", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Sensitivity Testing\n", - "The sensitivity testing framework provides a systematic approach to understanding how the Heston model responds to parameter changes, which is crucial for both model validation and practical application in trading and risk management." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "51922313", - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_test_provider.Sensitivity\")\n", - "def SensitivityTest(\n", - " model,\n", - " strike_price,\n", - " maturity_date,\n", - " spot_price,\n", - " v0_opt,\n", - " theta_opt,\n", - " kappa_opt,\n", - " sigma_opt,\n", - " rho_opt,\n", - "):\n", - " \"\"\"\n", - " Evaluates the sensitivity of American option prices to changes in model parameters.\n", - "\n", - " This test calculates option prices using the Heston model with optimized parameters.\n", - " It's designed to analyze how changes in various model inputs affect the option price,\n", - " which is crucial for understanding model behavior and risk management.\n", - "\n", - " Args:\n", - " model (HestonModel): Initialized Heston model instance wrapped in ValidMind model object\n", - " strike_price (float): Strike price of the option\n", - " maturity_date (ql.Date): Expiration date of the option in QuantLib format\n", - " spot_price (float): Current price of the underlying asset\n", - " v0_opt (float): Optimized initial variance parameter\n", - " theta_opt (float): Optimized long-term variance parameter\n", - " kappa_opt (float): Optimized mean reversion rate parameter\n", - " sigma_opt (float): Optimized volatility of variance parameter\n", - " rho_opt (float): Optimized correlation parameter between asset price and variance\n", - " \"\"\"\n", - " price = model.model.predict_american_option_price(\n", - " strike_price,\n", - " maturity_date,\n", - " spot_price,\n", - " v0_opt,\n", - " theta_opt,\n", - " kappa_opt,\n", - " sigma_opt,\n", - " rho_opt,\n", - " )\n", - "\n", - " return price\n" - ] - }, - { - "cell_type": "markdown", - "id": "408a05ef", - "metadata": {}, - "source": [ - "##### Common plot function" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "104ca6dd", - "metadata": {}, - "outputs": [], - "source": [ - "def plot_results(df, params: dict = None):\n", - " fig2 = plt.figure(figsize=(10, 6))\n", - " plt.plot(df[params[\"x\"]], df[params[\"y\"]], label=params[\"label\"])\n", - " plt.xlabel(params[\"xlabel\"])\n", - " plt.ylabel(params[\"ylabel\"])\n", - " \n", - " plt.title(params[\"title\"])\n", - " plt.legend()\n", - " plt.grid(True)\n", - " plt.show() # close the plot to avoid displaying it" - ] - }, - { - "cell_type": "markdown", - "id": "ca72b9e5", - "metadata": {}, - "source": [ - "Let's create ValidMind model object" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "ae7093fa", - "metadata": {}, - "outputs": [], - "source": [ - "hm_model = vm.init_model(model=hm, input_id=\"HestonModel\")" - ] - }, - { - "cell_type": "markdown", - "id": "b2141640", - "metadata": {}, - "source": [ - "##### Strike sensitivity\n", - "Let's analyzes how option prices change as the strike price varies. We create a range of strike prices around the current strike (460) and observe the impact on option prices while keeping all other parameters constant." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "ea7f1cbe", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_test_provider.Sensitivity:ToStrike\",\n", - " inputs = {\n", - " \"model\": hm_model\n", - " },\n", - " param_grid={\n", - " \"strike_price\": list(np.linspace(460-50, 460+50, 10)),\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": [theta_opt],\n", - " \"kappa_opt\": [kappa_opt],\n", - " \"sigma_opt\": [sigma_opt],\n", - " \"rho_opt\":[rho_opt]\n", - " },\n", - ")\n", - "result.log()\n", - "# Visualize how option prices change with different strike prices\n", - "plot_results(\n", - " pd.DataFrame(result.tables[0].data),\n", - " params={\n", - " \"x\": \"strike_price\",\n", - " \"y\":\"Value\",\n", - " \"label\":\"Strike price\",\n", - " \"xlabel\":\"Strike price\",\n", - " \"ylabel\":\"option price\",\n", - " \"title\":\"Heston option - Strike price Sensitivity\",\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "be143012", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Stress Testing\n", - "This stress testing framework provides a comprehensive view of how the Heston model behaves under different market conditions and helps identify potential risks in option pricing." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "f2f01a40", - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.Stressing\")\n", - "def StressTest(\n", - " model,\n", - " strike_price,\n", - " maturity_date,\n", - " spot_price,\n", - " v0_opt,\n", - " theta_opt,\n", - " kappa_opt,\n", - " sigma_opt,\n", - " rho_opt,\n", - "):\n", - " \"\"\"\n", - " Performs stress testing on Heston model parameters to evaluate option price sensitivity.\n", - "\n", - " This test evaluates how the American option price responds to stressed market conditions\n", - " by varying key model parameters. It's designed to:\n", - " 1. Identify potential model vulnerabilities\n", - " 2. Understand price behavior under extreme scenarios\n", - " 3. Support risk management decisions\n", - " 4. Validate model stability across parameter ranges\n", - "\n", - " Args:\n", - " model (HestonModel): Initialized Heston model instance wrapped in ValidMind model object\n", - " strike_price (float): Option strike price\n", - " maturity_date (ql.Date): Option expiration date in QuantLib format\n", - " spot_price (float): Current price of the underlying asset\n", - " v0_opt (float): Initial variance parameter under stress testing\n", - " theta_opt (float): Long-term variance parameter under stress testing\n", - " kappa_opt (float): Mean reversion rate parameter under stress testing\n", - " sigma_opt (float): Volatility of variance parameter under stress testing\n", - " rho_opt (float): Correlation parameter under stress testing\n", - " \"\"\"\n", - " price = model.model.predict_american_option_price(\n", - " strike_price,\n", - " maturity_date,\n", - " spot_price,\n", - " v0_opt,\n", - " theta_opt,\n", - " kappa_opt,\n", - " sigma_opt,\n", - " rho_opt,\n", - " )\n", - "\n", - " return price\n" - ] - }, - { - "cell_type": "markdown", - "id": "31fcbe9c", - "metadata": {}, - "source": [ - "##### Rho (correlation) and Theta (long term vol) stress test\n", - "Next, let's evaluates the sensitivity of a model's output to changes in the correlation parameter (rho) and the long-term variance parameter (theta) within a stochastic volatility framework." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "6119b5d9", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.Stressing:TheRhoAndThetaParameters\",\n", - " inputs = {\n", - " \"model\": hm_model,\n", - " },\n", - " param_grid={\n", - " \"strike_price\": [460],\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": list(np.linspace(0.1, theta_opt+0.4, 5)),\n", - " \"kappa_opt\": [kappa_opt],\n", - " \"sigma_opt\": [sigma_opt],\n", - " \"rho_opt\":list(np.linspace(rho_opt-0.2, rho_opt+0.2, 5))\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "be39cb3a", - "metadata": {}, - "source": [ - "##### Sigma stress test\n", - "Let's evaluates the sensitivity of a model's output to changes in the volatility parameter, sigma. This test is crucial for understanding how variations in market volatility impact the model's valuation of financial instruments, particularly options." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "0dc189b7", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.Stressing:TheSigmaParameter\",\n", - " inputs = {\n", - " \"model\": hm_model,\n", - " },\n", - " param_grid={\n", - " \"strike_price\": [460],\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": [theta_opt],\n", - " \"kappa_opt\": [kappa_opt],\n", - " \"sigma_opt\": list(np.linspace(0.1, sigma_opt+0.6, 5)),\n", - " \"rho_opt\": [rho_opt]\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "173a5294", - "metadata": {}, - "source": [ - "##### Stress kappa\n", - "Let's evaluates the sensitivity of a model's output to changes in the kappa parameter, which is a mean reversion rate in stochastic volatility models." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "dae9714f", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.Stressing:TheKappaParameter\",\n", - " inputs = {\n", - " \"model\": hm_model,\n", - " },\n", - " param_grid={\n", - " \"strike_price\": [460],\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": [theta_opt],\n", - " \"kappa_opt\": list(np.linspace(kappa_opt, kappa_opt+0.2, 5)),\n", - " \"sigma_opt\": [sigma_opt],\n", - " \"rho_opt\": [rho_opt]\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "b4d1d968", - "metadata": {}, - "source": [ - "##### Stress theta\n", - "Let's evaluates the sensitivity of a model's output to changes in the parameter theta, which represents the long-term variance in a stochastic volatility model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e68df3db", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.Stressing:TheThetaParameter\",\n", - " inputs = {\n", - " \"model\": hm_model,\n", - " },\n", - " param_grid={\n", - " \"strike_price\": [460],\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": list(np.linspace(0.1, theta_opt+0.9, 5)),\n", - " \"kappa_opt\": [kappa_opt],\n", - " \"sigma_opt\": [sigma_opt],\n", - " \"rho_opt\": [rho_opt]\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "32e70456", - "metadata": {}, - "source": [ - "##### Stress rho\n", - "Let's evaluates the sensitivity of a model's output to changes in the correlation parameter, rho, within a stochastic volatility (SV) model framework. This test is crucial for understanding how variations in rho, which represents the correlation between the asset price and its volatility, impact the model's valuation output." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b5ca3fc2", - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.Stressing:TheRhoParameter\",\n", - " inputs = {\n", - " \"model\": hm_model,\n", - " },\n", - " param_grid={\n", - " \"strike_price\": [460],\n", - " \"maturity_date\": [maturity_date],\n", - " \"spot_price\": [option_params[\"spot_price\"]],\n", - " \"v0_opt\": [v0_opt],\n", - " \"theta_opt\": [theta_opt],\n", - " \"kappa_opt\": [kappa_opt],\n", - " \"sigma_opt\": [sigma_opt],\n", - " \"rho_opt\": list(np.linspace(rho_opt-0.2, rho_opt+0.2, 5))\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "id": "892c5347", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Next steps\n", - "\n", - "You can look at the results of this test suite right in the notebook where you ran the code, as you would expect. But there is a better way — use the ValidMind Platform to work with your model documentation.\n", - "\n", - "\n", - "\n", - "### Work with your model documentation\n", - "\n", - "1. From the **Model Inventory** in the ValidMind Platform, go to the model you registered earlier. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/working-with-model-inventory.html))\n", - "\n", - "2. Click and expand the **Model Development** section.\n", - "\n", - "What you see is the full draft of your model documentation in a more easily consumable version. From here, you can make qualitative edits to model documentation, view guidelines, collaborate with validators, and submit your model documentation for approval when it's ready. [Learn more ...](https://docs.validmind.ai/guide/model-documentation/working-with-model-documentation.html)\n", - "\n", - "\n", - "\n", - "### Discover more learning resources\n", - "\n", - "We offer many interactive notebooks to help you document models:\n", - "\n", - "- [Run tests & test suites](https://docs.validmind.ai/developer/model-testing/testing-overview.html)\n", - "- [Code samples](https://docs.validmind.ai/developer/samples-jupyter-notebooks.html)\n", - "\n", - "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." - ] - }, - { - "cell_type": "markdown", - "id": "copyright-de5d1e182b09403abddabc2850f2dd05", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "validmind-1QuffXMV-py3.10", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.10.14" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/site/notebooks/code_samples/code_explainer/quickstart_code_explainer_demo.ipynb b/site/notebooks/code_samples/code_explainer/quickstart_code_explainer_demo.ipynb deleted file mode 100644 index 1eb1ef747e..0000000000 --- a/site/notebooks/code_samples/code_explainer/quickstart_code_explainer_demo.ipynb +++ /dev/null @@ -1,862 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Quickstart for model code documentation\n", - "\n", - "Welcome! This notebook demonstrates how to use the ValidMind code explainer to automatically generate comprehensive documentation for your codebase. The code explainer analyzes your source code and provides detailed explanations across various aspects of your implementation.\n", - "\n", - "\n", - "\n", - "## About Code Explainer\n", - "The ValidMind code explainer is a powerful tool that automatically analyzes your source code and generates comprehensive documentation. It helps you:\n", - "\n", - "- Understand the structure and organization of your codebase\n", - "- Document dependencies and environment setup\n", - "- Explain data processing and model implementation details\n", - "- Document training, evaluation, and inference pipelines\n", - "- Track configuration, testing, and security measures\n", - "\n", - "This tool is particularly useful for:\n", - "- Onboarding new team members\n", - "- Maintaining up-to-date documentation\n", - "- Ensuring code quality and best practices\n", - "- Facilitating code reviews and audits" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About Code Explainer](#toc1__) \n", - "- [About ValidMind](#toc2__) \n", - " - [Before you begin](#toc2_1__) \n", - " - [New to ValidMind?](#toc2_2__) \n", - " - [Key concepts](#toc2_3__) \n", - "- [Setting up](#toc3__) \n", - " - [Install the ValidMind Library](#toc3_1__) \n", - " - [Initialize the ValidMind Library](#toc3_2__) \n", - " - [Register sample model](#toc3_2_1__) \n", - " - [Apply documentation template](#toc3_2_2__) \n", - " - [Get your code snippet](#toc3_2_3__) \n", - " - [Preview the documentation template](#toc3_3__) \n", - "- [Common function](#toc4__) \n", - "- [Default Behavior](#toc5__) \n", - "- [Codebase Overview](#toc6__) \n", - "- [Environment and Dependencies ('environment_setup')](#toc7__) \n", - "- [Data Ingestion and Preprocessing](#toc8__) \n", - "- [Model Implementation Details](#toc9__) \n", - "- [Model Training Pipeline](#toc10__) \n", - "- [Evaluation and Validation Code](#toc11__) \n", - "- [Inference and Scoring Logic](#toc12__) \n", - "- [Configuration and Parameters](#toc13__) \n", - "- [Unit and Integration Testing](#toc14__) \n", - "- [Logging and Monitoring Hooks](#toc15__) \n", - "- [Code and Model Versioning](#toc16__) \n", - "- [Security and Access Control](#toc17__) \n", - "- [Example Runs and Scripts](#toc18__) \n", - "- [Known Issues and Future Improvements](#toc19__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", - "\n", - "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", - "\n", - "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", - "\n", - "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", - "\n", - "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", - "\n", - " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", - " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", - " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", - "\n", - "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", - "\n", - "**Outputs**: Custom tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", - "\n", - "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", - "\n", - "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Model Source Code Documentation`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " # api_host=\"...\",\n", - " # api_key=\"...\",\n", - " # api_secret=\"...\",\n", - " # model=\"...\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Preview the documentation template\n", - "\n", - "Let's verify that you have connected the ValidMind Library to the ValidMind Platform and that the appropriate *template* is selected for your model.\n", - "\n", - "You will upload documentation and test results unique to your model based on this template later on. For now, **take a look at the default structure that the template provides with [the `vm.preview_template()` function](https://docs.validmind.ai/validmind/validmind.html#preview_template)** from the ValidMind library and note the empty sections:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Common function\n", - "The code above defines two key functions:\n", - "1. A function to read source code from 'customer_churn_full_suite.py' file\n", - "2. An 'explain_code' function that uses ValidMind's experimental agents to analyze and explain code." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "source_code=\"\"\n", - "with open(\"customer_churn_full_suite.py\", \"r\") as f:\n", - " source_code = f.read()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The `vm.experimental.agents.run_task` function is used to execute AI agent tasks.\n", - "\n", - "It requires:\n", - "- task: The type of task to run (e.g. `code_explainer`)\n", - "- input: A dictionary containing task-specific parameters\n", - " - For `code_explainer`, this includes:\n", - " - **source_code** (str): The code to be analyzed\n", - " - **user_instructions** (str): Instructions for how to analyze the code" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "def explain_code(content_id: str, user_instructions: str):\n", - " \"\"\"Run code explanation task and log the results.\n", - " By default, the code explainer includes sections for:\n", - " - Main Purpose and Overall Functionality\n", - " - Breakdown of Key Functions or Components\n", - " - Potential Risks or Failure Points \n", - " - Assumptions or Limitations\n", - " If you want default sections, specify user_instructions as an empty string.\n", - " \n", - " Args:\n", - " user_instructions (str): Instructions for how to analyze the code\n", - " content_id (str): ID to use when logging the results\n", - " \n", - " Returns:\n", - " The result object from running the code explanation task\n", - " \"\"\"\n", - " result = vm.experimental.agents.run_task(\n", - " task=\"code_explainer\",\n", - " input={\n", - " \"source_code\": source_code,\n", - " \"user_instructions\": user_instructions\n", - " }\n", - " )\n", - " result.log(content_id=content_id)\n", - " return result" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Default Behavior" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "By default, the code explainer includes sections for:\n", - "- Main Purpose and Overall Functionality\n", - "- Breakdown of Key Functions or Components\n", - "- Potential Risks or Failure Points \n", - "- Assumptions or Limitations\n", - "\n", - "If you want default sections, specify `user_instructions` as an empty string. For example:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.experimental.agents.run_task(\n", - " task=\"code_explainer\",\n", - " input={\n", - " \"source_code\": source_code,\n", - " \"user_instructions\": \"\"\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Codebase Overview\n", - "\n", - "Let's analyze your codebase structure to understand the main modules, components, entry points and their relationships. We'll also examine the technology stack and frameworks that are being utilized in the implementation." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe the overall structure of the source code repository.\n", - " - Identify main modules, folders, and scripts.\n", - " - Highlight entry points for training, inference, and evaluation.\n", - " - State the main programming languages and frameworks used.\n", - " \"\"\",\n", - " content_id=\"code_structure_summary\"\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\",\n", - " content_id=\"code_structure_summary\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Environment and Dependencies ('environment_setup')\n", - "Let's document the technical requirements and setup needed to run your code, including Python packages, system dependencies, and environment configuration files. Understanding these requirements is essential for proper development environment setup and consistent deployments across different environments." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - List Python packages and system dependencies (OS, compilers, etc.).\n", - " - Reference environment files (requirements.txt, environment.yml, Dockerfile).\n", - " - Include setup instructions using Conda, virtualenv, or containers.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"setup_instructions\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Data Ingestion and Preprocessing\n", - "Let's document how your code handles data, including data sources, validation procedures, and preprocessing steps. We'll examine the data pipeline architecture, covering everything from initial data loading through feature engineering and quality checks." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Specify data input formats and sources.\n", - " - Document ingestion, validation, and transformation logic.\n", - " - Explain how raw data is preprocessed and features are generated.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections. \"\"\",\n", - " content_id=\"data_handling_notes\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - " \n", - "\n", - "\n", - "\n", - "## Model Implementation Details\n", - "Let's document the core implementation details of your model, including its architecture, components, and key algorithms. Understanding the technical implementation is crucial for maintenance, debugging, and future improvements to the codebase. We'll examine how theoretical concepts are translated into working code." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe the core model code structure (classes, functions).\n", - " - Link code to theoretical models or equations when applicable.\n", - " - Note custom components like loss functions or feature selectors.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"model_code_description\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Model Training Pipeline\n", - "\n", - "Let's document the training pipeline implementation, including how models are trained, optimized and evaluated. We'll examine the training process workflow, hyperparameter tuning approach, and model checkpointing mechanisms. This section provides insights into how the model learns from data and achieves optimal performance." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Explain the training process, optimization strategy, and hyperparameters.\n", - " - Describe logging, checkpointing, and early stopping mechanisms.\n", - " - Include references to training config files or tuning logic.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"training_logic_details\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Evaluation and Validation Code\n", - "Let's examine how the model's validation and evaluation code is implemented, including the metrics calculation and validation processes. We'll explore the diagnostic tools and visualization methods used to assess model performance. This section will also cover how validation results are logged and stored for future reference." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe how validation is implemented and metrics are calculated.\n", - " - Include plots and diagnostic tools (e.g., ROC, SHAP, confusion matrix).\n", - " - State how outputs are logged and persisted.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"evaluation_logic_notes\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Inference and Scoring Logic\n", - "Let's examine how the model performs inference and scoring on new data. This section will cover the implementation details of loading trained models, making predictions, and any required pre/post-processing steps. We'll also look at the APIs and interfaces available for both real-time serving and batch scoring scenarios." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Detail how the trained model is loaded and used for predictions.\n", - " - Explain I/O formats and APIs for serving or batch scoring.\n", - " - Include any preprocessing/postprocessing logic required.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"inference_mechanism\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Configuration and Parameters\n", - "Let's explore how configuration and parameters are managed in the codebase. We'll examine the configuration files, command-line arguments, environment variables, and other mechanisms used to control model behavior. This section will also cover parameter versioning and how different configurations are tracked across model iterations." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe configuration management (files, CLI args, env vars).\n", - " - Highlight default parameters and override mechanisms.\n", - " - Reference versioning practices for config files.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"config_control_notes\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Unit and Integration Testing\n", - "Let's examine the testing strategy and implementation in the codebase. We'll analyze the unit tests, integration tests, and testing frameworks used to ensure code quality and reliability. This section will also cover test coverage metrics and continuous integration practices." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - List unit and integration tests and what they cover.\n", - " - Mention testing frameworks and coverage tools used.\n", - " - Explain testing strategy for production-readiness.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"test_strategy_overview\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Logging and Monitoring Hooks\n", - "Let's analyze how logging and monitoring are implemented in the codebase. We'll examine the logging configuration, monitoring hooks, and key metrics being tracked. This section will also cover any real-time observability integrations and alerting mechanisms in place." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe logging configuration and structure.\n", - " - Highlight real-time monitoring or observability integrations.\n", - " - List key events, metrics, or alerts tracked.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"logging_monitoring_notes\"\n", - ")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Code and Model Versioning\n", - "Let's examine how code and model versioning is managed in the codebase. This section will cover version control practices, including Git workflows and model artifact versioning tools like DVC or MLflow. We'll also look at how versioning integrates with the CI/CD pipeline." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Describe Git usage, branching, tagging, and commit standards.\n", - " - Include model artifact versioning practices (e.g., DVC, MLflow).\n", - " - Reference any automation in CI/CD.\n", - " Please remove the following sections: \n", - " - Potential Risks or Failure Points\n", - " - Assumptions or Limitations\n", - " - Breakdown of Key Functions or Components\n", - " Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"version_tracking_description\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Security and Access Control\n", - "Let's analyze the security and access control measures implemented in the codebase. We'll examine how sensitive data and code are protected through access controls, encryption, and compliance measures. Additionally, we'll review secure deployment practices and any specific handling of PII data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Document access controls for source code and data.\n", - " - Include any encryption, PII handling, or compliance measures.\n", - " - Mention secure deployment practices.\n", - " Please remove the following sections: \n", - " - Potential Risks or Failure Points\n", - " - Assumptions or Limitations\n", - " - Breakdown of Key Functions or Components\n", - " Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"security_policies_notes\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Example Runs and Scripts\n", - "Let's explore example runs and scripts that demonstrate how to use this codebase in practice. We'll look at working examples, command-line usage, and sample notebooks that showcase the core functionality. This section will also point to demo datasets and test scenarios that can help new users get started quickly." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - Provide working script examples.\n", - " - Include CLI usage instructions or sample notebooks.\n", - " - Link to demo datasets or test scenarios.\n", - " Please remove the following sections: \n", - " - Potential Risks or Failure Points\n", - " - Assumptions or Limitations\n", - " - Breakdown of Key Functions or Components\n", - " Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"runnable_examples\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "## Known Issues and Future Improvements\n", - "Let's examine the current limitations and areas for improvement in the codebase. This section will document known technical debt, bugs, and feature gaps that need to be addressed. We'll also outline proposed enhancements and reference any existing tickets or GitHub issues tracking these improvements." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = explain_code(\n", - " user_instructions=\"\"\"\n", - " Please provide a summary of the following bullet points only.\n", - " - List current limitations or technical debt.\n", - " - Outline proposed enhancements or refactors.\n", - " - Reference relevant tickets, GitHub issues, or roadmap items.\n", - " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", - " \"\"\",\n", - " content_id=\"issues_and_improvements_log\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "copyright-ccbede139a26452183291a108b791513", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "validmind-1QuffXMV-py3.11", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.9" - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} diff --git a/site/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb b/site/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb deleted file mode 100644 index 17385b8e1c..0000000000 --- a/site/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb +++ /dev/null @@ -1,1095 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Implement custom tests\n", - "\n", - "Custom tests extend the functionality of ValidMind, allowing you to document any model or use case with added flexibility.\n", - "\n", - "ValidMind provides a comprehensive set of tests out-of-the-box to evaluate and document your models and datasets. We recognize there will be cases where the default tests do not support a model or dataset, or specific documentation is needed. In these cases, you can create and use your own custom code to accomplish what you need. To streamline custom code integration, we support the creation of custom test functions.\n", - "\n", - "This interactive notebook provides a step-by-step guide for implementing and registering custom tests with ValidMind, running them individually, viewing the results on the ValidMind Platform, and incorporating them into your model documentation template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About ValidMind](#toc1__) \n", - " - [Before you begin](#toc1_1__) \n", - " - [New to ValidMind?](#toc1_2__) \n", - " - [Key concepts](#toc1_3__) \n", - "- [Setting up](#toc2__) \n", - " - [Install the ValidMind Library](#toc2_1__) \n", - " - [Initialize the ValidMind Library](#toc2_2__) \n", - " - [Register sample model](#toc2_2_1__) \n", - " - [Apply documentation template](#toc2_2_2__) \n", - " - [Get your code snippet](#toc2_2_3__) \n", - "- [Implement a Custom Test](#toc3__) \n", - "- [Run the Custom Test](#toc4__) \n", - " - [Setup the Model and Dataset](#toc4_1__) \n", - " - [Run the Custom Test](#toc4_2__) \n", - "- [Adding Custom Test to Model Documentation](#toc5__) \n", - "- [Some More Custom Tests](#toc6__) \n", - " - [Custom Test: Table of Model Hyperparameters](#toc6_1__) \n", - " - [Custom Test: External API Call](#toc6_2__) \n", - " - [Custom Test: Passing Parameters](#toc6_3__) \n", - " - [Custom Test: Multiple Tables and Plots in a Single Test](#toc6_4__) \n", - " - [Custom Test: Images](#toc6_5__) \n", - " - [Custom Test: Description](#toc6_6__) \n", - "- [Conclusion](#toc7__) \n", - "- [Next steps](#toc8__) \n", - " - [Work with your model documentation](#toc8_1__) \n", - " - [Discover more learning resources](#toc8_2__) \n", - "- [Upgrade ValidMind](#toc9__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", - "\n", - "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", - "\n", - "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", - "\n", - "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", - "\n", - "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", - "\n", - " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", - " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", - " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", - "\n", - "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", - "\n", - "**Outputs**: Custom test can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", - "\n", - "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", - "\n", - "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - " For example, to register a model for use with this notebook, select the following use case: `{use-case}`\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Binary classification`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " # api_host=\"...\",\n", - " # api_key=\"...\",\n", - " # api_secret=\"...\",\n", - " # model=\"...\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Implement a Custom Test\n", - "\n", - "Let's start off by creating a simple custom test that creates a Confusion Matrix for a binary classification model. We will use the `sklearn.metrics.confusion_matrix` function to calculate the confusion matrix and then display it as a heatmap using `plotly`. (This is already a built-in test in ValidMind, but we will use it as an example to demonstrate how to create custom tests.)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import matplotlib.pyplot as plt\n", - "from sklearn import metrics\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.ConfusionMatrix\")\n", - "def confusion_matrix(dataset, model):\n", - " \"\"\"The confusion matrix is a table that is often used to describe the performance of a classification model on a set of data for which the true values are known.\n", - "\n", - " The confusion matrix is a 2x2 table that contains 4 values:\n", - "\n", - " - True Positive (TP): the number of correct positive predictions\n", - " - True Negative (TN): the number of correct negative predictions\n", - " - False Positive (FP): the number of incorrect positive predictions\n", - " - False Negative (FN): the number of incorrect negative predictions\n", - "\n", - " The confusion matrix can be used to assess the holistic performance of a classification model by showing the accuracy, precision, recall, and F1 score of the model on a single figure.\n", - " \"\"\"\n", - " y_true = dataset.y\n", - " y_pred = dataset.y_pred(model)\n", - "\n", - " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", - "\n", - " cm_display = metrics.ConfusionMatrixDisplay(\n", - " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", - " )\n", - " cm_display.plot()\n", - "\n", - " plt.close() # close the plot to avoid displaying it\n", - "\n", - " return cm_display.figure_ # return the figure object itself" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Thats our custom test defined and ready to go... Let's take a look at whats going on here:\n", - "\n", - "- The function `confusion_matrix` takes two arguments `dataset` and `model`. This is a VMDataset and VMModel object respectively.\n", - "- The function docstring provides a description of what the test does. This will be displayed along with the result in this notebook as well as in the ValidMind Platform.\n", - "- The function body calculates the confusion matrix using the `sklearn.metrics.confusion_matrix` function and then plots it using `sklearn.metric.ConfusionMatrixDisplay`.\n", - "- The function then returns the `ConfusionMatrixDisplay.figure_` object - this is important as the ValidMind Library expects the output of the custom test to be a plot or a table.\n", - "- The `@vm.test` decorator is doing the work of creating a wrapper around the function that will allow it to be run by the ValidMind Library. It also registers the test so it can be found by the ID `my_custom_tests.ConfusionMatrix` (see the section below on how test IDs work in ValidMind and why this format is important)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Run the Custom Test\n", - "\n", - "Now that we have defined and registered our custom test, lets see how we can run it and properly use it in the ValidMind Platform." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Setup the Model and Dataset\n", - "\n", - "First let's setup a an example model and dataset to run our custom metic against. Since this is a Confusion Matrix, we will use the Customer Churn dataset that ValidMind provides and train a simple XGBoost model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "from validmind.datasets.classification import customer_churn\n", - "\n", - "raw_df = customer_churn.load_data()\n", - "train_df, validation_df, test_df = customer_churn.preprocess(raw_df)\n", - "\n", - "x_train = train_df.drop(customer_churn.target_column, axis=1)\n", - "y_train = train_df[customer_churn.target_column]\n", - "x_val = validation_df.drop(customer_churn.target_column, axis=1)\n", - "y_val = validation_df[customer_churn.target_column]\n", - "\n", - "model = xgb.XGBClassifier(early_stopping_rounds=10)\n", - "model.set_params(\n", - " eval_metric=[\"error\", \"logloss\", \"auc\"],\n", - ")\n", - "model.fit(\n", - " x_train,\n", - " y_train,\n", - " eval_set=[(x_val, y_val)],\n", - " verbose=False,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Easy enough! Now we have a model and dataset setup and trained. One last thing to do is bring the dataset and model into the ValidMind Library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# for now, we'll just use the test dataset\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " target_column=customer_churn.target_column,\n", - " input_id=\"test_dataset\",\n", - ")\n", - "\n", - "vm_model = vm.init_model(model, input_id=\"model\")\n", - "\n", - "# link the model to the dataset\n", - "vm_test_ds.assign_predictions(model=vm_model)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Run the Custom Test\n", - "\n", - "Now that we have our model and dataset setup, we have everything we need to run our custom test. We can do this by importing the `run_test` function from the `validmind.tests` module and passing in the test ID of our custom test along with the model and dataset we want to run it against.\n", - "\n", - ">Notice how the `inputs` dictionary is used to map an `input_id` which we set above to the `model` and `dataset` keys that are expected by our custom test function. This is how the ValidMind Library knows which inputs to pass to different tests and is key when using many different datasets and models." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "\n", - "result = run_test(\n", - " \"my_custom_tests.ConfusionMatrix\",\n", - " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You'll notice that the docstring becomes a markdown description of the test. The figure is then displayed as the test result. What you see above is how it will look in the ValidMind Platform as well. Let's go ahead and log the result to see how that works." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Adding Custom Test to Model Documentation\n", - "\n", - "To do this, go to the documentation page of the model you registered above and navigate to the `Model Development` -> `Model Evaluation` section. Then hover between any existing content block to reveal the `+` button as shown in the screenshot below.\n", - "\n", - "![screenshot showing insert button for test-driven blocks](../../images/insert-test-driven-block.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Now click on the `+` button and select the `Test-Driven Block` option. This will open a dialog where you can select `My Custom Tests Confusion Matrix` from the list of available tests. You can preview the result and then click `Insert Block` to add it to the documentation.\n", - "\n", - "![screenshot showing how to insert a test-driven block](../../images/insert-test-driven-block-custom.png)\n", - "\n", - "The test should match the result you see above. It is now part of your documentation and will now be run everytime you run `vm.run_documentation_tests()` for your model. Let's do that now." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.reload()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If you preview the template, it should show the custom test in the `Model Development`->`Model Evaluation` section:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Just so we can run all of the tests in the template, let's initialize the train and raw dataset.\n", - "\n", - "(Refer to [**Quickstart for model documentation**](../../quickstart/quickstart_model_documentation.ipynb) and the ValidMind docs for more information on what we are doing here)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=raw_df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=customer_churn.target_column,\n", - " class_labels=customer_churn.class_labels,\n", - ")\n", - "\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=customer_churn.target_column,\n", - ")\n", - "vm_train_ds.assign_predictions(model=vm_model)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "To run all the tests in the template, you can use the `vm.run_documentation_tests()` and pass the inputs we initialized above and the demo config from our customer_churn module. We will have to add a section to the config for our new test to tell it which inputs it should receive. This is done by simply adding a new element in the config dictionary where the key is the ID of the test and the value is a dictionary with the following structure:\n", - "```python\n", - "{\n", - " \"inputs\": {\n", - " \"model\": \"test_dataset\",\n", - " \"dataset\": \"model\",\n", - " }\n", - "}\n", - "```" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.utils import preview_test_config\n", - "\n", - "test_config = customer_churn.get_demo_test_config()\n", - "test_config[\"my_custom_tests.ConfusionMatrix\"] = {\n", - " \"inputs\": {\n", - " \"dataset\": \"test_dataset\",\n", - " \"model\": \"model\",\n", - " }\n", - "}\n", - "preview_test_config(test_config)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "full_suite = vm.run_documentation_tests(config=test_config)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Some More Custom Tests\n", - "\n", - "Now that you understand the entire process of creating custom tests and using them in your documentation, let's create a few more to see different ways you can utilize custom tests." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: Table of Model Hyperparameters\n", - "\n", - "This custom test will display a table of the hyperparameters used in the model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.Hyperparameters\")\n", - "def hyperparameters(model):\n", - " \"\"\"The hyperparameters of a machine learning model are the settings that control the learning process.\n", - " These settings are specified before the learning process begins and can have a significant impact on the\n", - " performance of the model.\n", - "\n", - " The hyperparameters of a model can be used to tune the model to achieve the best possible performance\n", - " on a given dataset. By examining the hyperparameters of a model, you can gain insight into how the model\n", - " was trained and how it might be improved.\n", - " \"\"\"\n", - " hyperparameters = model.model.get_xgb_params() # dictionary of hyperparameters\n", - "\n", - " # turn the dictionary into a table where each row contains a hyperparameter and its value\n", - " return [{\"Hyperparam\": k, \"Value\": v} for k, v in hyperparameters.items() if v]\n", - "\n", - "\n", - "result = run_test(\"my_custom_tests.Hyperparameters\", inputs={\"model\": \"model\"})\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Since the test has been run and logged, you can add it to your documentation using the same process as above. It should look like this:\n", - "\n", - "![screenshot showing hyperparameters test](../../images/hyperparameters-custom-metric.png)\n", - "\n", - "For our simple toy model, there are aren't really any proper hyperparameters but you can see how this could be useful for more complex models that have gone through hyperparameter tuning." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: External API Call\n", - "\n", - "This custom test will make an external API call to get the current BTC price and display it as a table. This demonstrates how you might integrate external data sources into your model documentation in a programmatic way. You could, for instance, setup a pipeline that runs a test like this every day to keep your model documentation in sync with an external system." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import requests\n", - "import random\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.ExternalAPI\")\n", - "def external_api():\n", - " \"\"\"This test calls an external API to get a list of fake users. It then creates\n", - " a table with the relevant data so it can be displayed in the documentation.\n", - "\n", - " The purpose of this test is to demonstrate how to call an external API and use the\n", - " data in a test. A test like this could even be setup to run in a scheduled\n", - " pipeline to keep your documentation in-sync with an external data source.\n", - " \"\"\"\n", - " url = \"https://jsonplaceholder.typicode.com/users\"\n", - " response = requests.get(url)\n", - " data = response.json()\n", - "\n", - " # extract the time and the current BTC price in USD\n", - " return {\n", - " \"Model Owners/Stakeholders\": [\n", - " {\n", - " \"Name\": user[\"name\"],\n", - " \"Role\": random.choice([\"Owner\", \"Stakeholder\"]),\n", - " \"Email\": user[\"email\"],\n", - " \"Phone\": user[\"phone\"],\n", - " \"Slack Handle\": f\"@{user['name'].lower().replace(' ', '.')}\",\n", - " }\n", - " for user in data[:3]\n", - " ],\n", - " \"Model Developers\": [\n", - " {\n", - " \"Name\": user[\"name\"],\n", - " \"Role\": \"Developer\",\n", - " \"Email\": user[\"email\"],\n", - " }\n", - " for user in data[3:7]\n", - " ],\n", - " \"Model Validators\": [\n", - " {\n", - " \"Name\": user[\"name\"],\n", - " \"Role\": \"Validator\",\n", - " \"Email\": user[\"email\"],\n", - " }\n", - " for user in data[7:]\n", - " ],\n", - " }\n", - "\n", - "\n", - "result = run_test(\"my_custom_tests.ExternalAPI\")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Again, you can add this to your documentation to see how it looks:\n", - "\n", - "![screenshot showing BTC price metric](../../images/external-data-custom-test.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: Passing Parameters\n", - "\n", - "Custom test functions, as stated earlier, can take both inputs and params. When you define your function there is no need to distinguish between the two, the ValidMind Library will handle that for you. You simply need to add both to the function as arguments and the library will pass in the correct values.\n", - "\n", - "So for instance, if you wanted to parameterize the first custom test we created, the confusion matrix, you could do so like this:\n", - "\n", - "```python\n", - "def confusion_matrix(dataset: VMDataset, model: VMModel, my_param: str = \"Default Value\"):\n", - " pass\n", - "```\n", - "\n", - "And then when you run the test, you can pass in the parameter like this:\n", - "\n", - "```python\n", - "vm.run_test(\n", - " \"my_custom_tests.ConfusionMatrix\",\n", - " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", - " params={\"my_param\": \"My Value\"},\n", - ")\n", - "```\n", - "\n", - "Or if you are running the entire documentation template, you would update the config like this:\n", - "\n", - "```python\n", - "test_config[\"my_custom_tests.ConfusionMatrix\"] = {\n", - " \"inputs\": {\n", - " \"dataset\": \"test_dataset\",\n", - " \"model\": \"model\",\n", - " },\n", - " \"params\": {\n", - " \"my_param\": \"My Value\",\n", - " },\n", - "}\n", - "```\n", - "\n", - "Let's go ahead and create a toy test that takes a parameter and uses it in the result:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import plotly.express as px\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.ParameterExample\")\n", - "def parameter_example(\n", - " plot_title=\"Default Plot Title\", x_col=\"sepal_width\", y_col=\"sepal_length\"\n", - "):\n", - " \"\"\"This test takes two parameters and creates a scatter plot based on them.\n", - "\n", - " The purpose of this test is to demonstrate how to create a test that takes\n", - " parameters and uses them to generate a plot. This can be useful for creating\n", - " tests that are more flexible and can be used in a variety of scenarios.\n", - " \"\"\"\n", - " # return px.scatter(px.data.iris(), x=x_col, y=y_col, color=\"species\")\n", - " return px.scatter(\n", - " px.data.iris(), x=x_col, y=y_col, color=\"species\", title=plot_title\n", - " )\n", - "\n", - "\n", - "result = run_test(\n", - " \"my_custom_tests.ParameterExample\",\n", - " params={\n", - " \"plot_title\": \"My Cool Plot\",\n", - " \"x_col\": \"sepal_width\",\n", - " \"y_col\": \"sepal_length\",\n", - " },\n", - ")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Play around with this and see how you can use parameters, default values and other features to make your custom tests more flexible and useful.\n", - "\n", - "Here's how this one looks in the documentation:\n", - "![screenshot showing parameterized test](../../images/parameterized-custom-metric.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: Multiple Tables and Plots in a Single Test\n", - "\n", - "Custom test functions, as stated earlier, can return more than just one table or plot. In fact, any number of tables and plots can be returned. Let's see an example of this:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import numpy as np\n", - "import plotly.express as px\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.ComplexOutput\")\n", - "def complex_output():\n", - " \"\"\"This test demonstrates how to return many tables and figures in a single test\"\"\"\n", - " # create a couple tables\n", - " table = [{\"A\": 1, \"B\": 2}, {\"A\": 3, \"B\": 4}]\n", - " table2 = [{\"C\": 5, \"D\": 6}, {\"C\": 7, \"D\": 8}]\n", - "\n", - " # create a few figures showing some random data\n", - " fig1 = px.line(x=np.arange(10), y=np.random.rand(10), title=\"Random Line Plot\")\n", - " fig2 = px.bar(x=[\"A\", \"B\", \"C\"], y=np.random.rand(3), title=\"Random Bar Plot\")\n", - " fig3 = px.scatter(\n", - " x=np.random.rand(10), y=np.random.rand(10), title=\"Random Scatter Plot\"\n", - " )\n", - "\n", - " return (\n", - " {\n", - " \"My Cool Table\": table,\n", - " \"Another Table\": table2,\n", - " },\n", - " fig1,\n", - " fig2,\n", - " fig3,\n", - " )\n", - "\n", - "\n", - "result = run_test(\"my_custom_tests.ComplexOutput\")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Notice how you can return the tables as a dictionary where the key is the title of the table and the value is the table itself. You could also just return the tables by themselves but this way you can give them a title to more easily identify them in the result.\n", - "\n", - "![screenshot showing multiple tables and plots](../../images/multiple-tables-plots-custom-metric.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: Images\n", - "\n", - "If you are using a plotting library that isn't supported by ValidMind (i.e. not `matplotlib` or `plotly`), you can still return the image directly as a bytes-like object. This could also be used to bring any type of image into your documentation in a programmatic way. For instance, you may want to include a diagram of your model architecture or a screenshot of a dashboard that your model is integrated with. As long as you can produce the image with Python or open it from a file, you can include it in your documentation." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import io\n", - "import matplotlib.pyplot as plt\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.Image\")\n", - "def image():\n", - " \"\"\"This test demonstrates how to return an image in a test\"\"\"\n", - "\n", - " # create a simple plot\n", - " fig, ax = plt.subplots()\n", - " ax.plot([1, 2, 3, 4])\n", - " ax.set_title(\"Simple Line Plot\")\n", - "\n", - " # save the plot as a PNG image (in-memory buffer)\n", - " img_data = io.BytesIO()\n", - " fig.savefig(img_data, format=\"png\")\n", - " img_data.seek(0)\n", - "\n", - " plt.close() # close the plot to avoid displaying it\n", - "\n", - " return img_data.read()\n", - "\n", - "\n", - "result = run_test(\"my_custom_tests.Image\")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Adding this custom test to your documentation will display the image:\n", - "\n", - "![screenshot showing image custom test](../../images/image-in-custom-metric.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If you want to log an image as a test result, you can do so by passing the path to the image as a parameter to the custom test and then opening the file in the test function. Here's an example:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.MyPNGCorrelationMatrix\")\n", - "def Image(path: str):\n", - " \"\"\"Opens a png image file and logs it as a test result to ValidMind\"\"\"\n", - " if not path.endswith(\".png\"):\n", - " raise ValueError(\"Image must be a PNG file\")\n", - "\n", - " # return raw image bytes\n", - " with open(path, \"rb\") as f:\n", - " return f.read()\n", - " \n", - "run_test(\n", - " \"my_custom_tests.MyPNGCorrelationMatrix\",\n", - " params={\"path\": \"../../images/pearson-correlation-matrix.png\"},\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The image is displayed in the test result:\n", - "\n", - "![screenshot showing image from file](../../images/pearson-correlation-matrix-test-output.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Custom Test: Description\n", - "\n", - "If you want to write a custom test description for your custom test instead of it is interpreted through llm, you can do so by returning string in your test." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import pandas as pd\n", - "\n", - "@vm.test(\"my_custom_tests.MyCustomTest\")\n", - "def my_custom_test(dataset, model):\n", - " \"\"\"\n", - " This is a custom computed test that computes confusion matrix for a binary classification model and return a string as a test description.\n", - " \"\"\"\n", - " y_true = dataset.y\n", - " y_pred = dataset.y_pred(model)\n", - "\n", - " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", - "\n", - " cm_display = metrics.ConfusionMatrixDisplay(\n", - " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", - " )\n", - " cm_display.plot()\n", - "\n", - " plt.close() # close the plot to avoid displaying it\n", - "\n", - " return cm_display.figure_, \"Test Description - Confusion Matrix\", pd.DataFrame({\"Value\": [1, 2, 3]}) # return the figure object itself\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can see here test result description has been customized here. The same result description will be displayed in the UI." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = run_test(\n", - " \"my_custom_tests.MyCustomTest\",\n", - " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", - ")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Conclusion\n", - "\n", - "In this notebook, we have demonstrated how to create custom tests in ValidMind. We have shown how to define custom test functions, register them with the ValidMind Library, run them against models and datasets, and add them to model documentation templates. We have also shown how to return tables and plots from custom tests and how to use them in the ValidMind Platform. We hope this tutorial has been helpful in understanding how to create and use custom tests in ValidMind." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Next steps\n", - "\n", - "You can look at the results of this test suite right in the notebook where you ran the code, as you would expect. But there is a better way — use the ValidMind Platform to work with your model documentation.\n", - "\n", - "\n", - "\n", - "### Work with your model documentation\n", - "\n", - "1. From the **Model Inventory** in the ValidMind Platform, go to the model you registered earlier. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/working-with-model-inventory.html))\n", - "\n", - "2. Click and expand the **Model Development** section.\n", - "\n", - "What you see is the full draft of your model documentation in a more easily consumable version. From here, you can make qualitative edits to model documentation, view guidelines, collaborate with validators, and submit your model documentation for approval when it's ready. [Learn more ...](https://docs.validmind.ai/guide/model-documentation/working-with-model-documentation.html)\n", - "\n", - "\n", - "\n", - "### Discover more learning resources\n", - "\n", - "We offer many interactive notebooks to help you document models:\n", - "\n", - "- [Run tests & test suites](https://docs.validmind.ai/developer/model-testing/testing-overview.html)\n", - "- [Code samples](https://docs.validmind.ai/developer/samples-jupyter-notebooks.html)\n", - "\n", - "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Upgrade ValidMind\n", - "\n", - "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", - "\n", - "Retrieve the information for the currently installed version of ValidMind:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip show validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", - "\n", - "```bash\n", - "%pip install --upgrade validmind\n", - "```" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You may need to restart your kernel after running the upgrade package for changes to be applied." - ] - }, - { - "cell_type": "markdown", - "id": "copyright-997b933948594ddd929ee9419957dfe3", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.5" - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} diff --git a/site/notebooks/code_samples/nlp_and_llm/rag_benchmark_demo.ipynb b/site/notebooks/code_samples/nlp_and_llm/rag_benchmark_demo.ipynb deleted file mode 100644 index 35969b1eb3..0000000000 --- a/site/notebooks/code_samples/nlp_and_llm/rag_benchmark_demo.ipynb +++ /dev/null @@ -1,1857 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# RAG Model Benchmarking Demo\n", - "\n", - "In this notebook, we are going to implement a simple RAG Model for automating the process of answering RFP questions using GenAI. We will see how we can initialize an embedding model, a retrieval model and a generator model with LangChain components and use them within the ValidMind Library to run tests against them. We'll demonstrate how to set up multiple models for benchmarking at each stage of the RAG pipeline - specifically two embedding models, two retrieval models with different parameters, and two LLM models (GPT-3.5 and GPT-4o) - allowing for comparison of performance across different configurations. Finally, we will see how we can put them together in a Pipeline and run that to get e2e results and run tests against that." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About ValidMind](#toc1__) \n", - " - [Before you begin](#toc1_1__) \n", - " - [New to ValidMind?](#toc1_2__) \n", - " - [Key concepts](#toc1_3__) \n", - "- [Setting up](#toc2__) \n", - " - [Install the ValidMind Library](#toc2_1__) \n", - " - [Initialize the ValidMind Library](#toc2_2__) \n", - " - [Register sample model](#toc2_2_1__) \n", - " - [Apply documentation template](#toc2_2_2__) \n", - " - [Get your code snippet](#toc2_2_3__) \n", - "- [Read Open AI API Key](#toc3__) \n", - "- [Dataset Loader](#toc4__) \n", - "- [Data validation](#toc5__) \n", - " - [Duplicates](#toc5_1__) \n", - " - [Stop Words](#toc5_2__) \n", - " - [Punctuations](#toc5_3__) \n", - " - [Common Words](#toc5_4__) \n", - " - [Language Detection](#toc5_5__) \n", - " - [Toxicity Score](#toc5_6__) \n", - " - [Polarity and Subjectivity](#toc5_7__) \n", - " - [Sentiment](#toc5_8__) \n", - " - [Assign Predictions](#toc5_9__) \n", - " - [Run tests](#toc5_10__) \n", - " - [Generate embeddings for the Train Set](#toc5_11__) \n", - " - [Insert embeddings and questions into Vector DB](#toc5_12__) \n", - "- [Prompt Evaluation](#toc6__) \n", - "- [RAGAS evaluation](#toc7__) \n", - " - [Semantic Similarity](#toc7_1__) \n", - " - [Context Entity Recall](#toc7_2__) \n", - " - [Context Precision](#toc7_3__) \n", - " - [Context Precision Without Reference](#toc7_4__) \n", - " - [Faithfulness](#toc7_5__) \n", - " - [Response Relevancy](#toc7_6__) \n", - " - [Context Recall](#toc7_7__) \n", - " - [Answer Correctness](#toc7_8__) \n", - " - [Aspect Critic](#toc7_9__) \n", - " - [Noise Sensitivity](#toc7_10__) \n", - "- [Generation quality](#toc8__) \n", - " - [Token Disparity](#toc8_1__) \n", - " - [ROUGE Score](#toc8_2__) \n", - " - [BLEU Score](#toc8_3__) \n", - " - [BERT Score](#toc8_4__) \n", - " - [METEOR Score](#toc8_5__) \n", - "- [Bias and Toxicity](#toc9__) \n", - " - [Toxicity Score](#toc9_1__) \n", - " - [Regard Score](#toc9_2__) \n", - "- [Upgrade ValidMind](#toc10__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "- **FunctionModels**: ValidMind offers support for creating `VMModel` instances from Python functions. This enables us to support any \"model\" by simply using the provided function as the model's `predict` method.\n", - "- **PipelineModels**: ValidMind models (`VMModel` instances) of any type can be piped together to create a model pipeline. This allows model components to be created and tested/documented independently, and then combined into a single model for end-to-end testing and documentation. We use the `|` operator to pipe models together.\n", - "- **RAG**: RAG stands for Retrieval Augmented Generation and refers to a wide range of GenAI applications where some form of retrieval is used to add context to the prompt so that the LLM that generates content can refer to it when creating its output. In this notebook, we are going to implement a simple RAG setup using LangChain components." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Prerequisites\n", - "\n", - "Let's go ahead and install the `validmind` library if its not already installed... Then we can install the `qdrant-client` library for our vector store and `langchain` for everything else:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q \"validmind[llm]\"" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q qdrant-client langchain langchain-openai sentencepiece" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "
Recommended Python versions\n", - "

\n", - "Python 3.8 <= x <= 3.11
\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Gen AI RAG Template`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " api_host = \"https://api.prod.validmind.ai/api/v1/tracking\",\n", - " api_key = \"...\",\n", - " api_secret = \"...\",\n", - " model = \"...\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Read Open AI API Key\n", - "\n", - "We will need to have an OpenAI API key to be able to use their `text-embedding-3-small` and `text-embedding-3-large` models for our embeddings, `gpt-3.5-turbo` and `gpt-4o` models for our generator and `gpt-4o` model for our LLM-as-Judge tests. If you don't have an OpenAI API key, you can get one by signing up at [OpenAI](https://platform.openai.com/signup). Then you can create a `.env` file in the root of your project and the following cell will load it from there. Alternatively, you can just uncomment the line below to directly set the key (not recommended for security reasons)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# load openai api key\n", - "import os\n", - "\n", - "import dotenv\n", - "import nltk\n", - "\n", - "dotenv.load_dotenv()\n", - "nltk.download('stopwords')\n", - "nltk.download('punkt_tab')\n", - "\n", - "# os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n", - "\n", - "if not \"OPENAI_API_KEY\" in os.environ:\n", - " raise ValueError(\"OPENAI_API_KEY is not set\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Dataset Loader\n", - "\n", - "Great, now that we have all of our dependencies installed, the ValidMind Library initialized and connected to our model and our OpenAI API key setup, we can go ahead and load our datasets. We will use the synthetic `RFP` dataset included with ValidMind for this notebook. This dataset contains a variety of RFP questions and ground truth answers that we can use both as the source where our Retriever will search for similar question-answer pairs as well as our test set for evaluating the performance of our RAG model. To do this, we just have to load it and call the preprocess function to get a split of the data into train and test sets." - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the sample dataset from the library\n", - "from validmind.datasets.llm.rag import rfp\n", - "\n", - "raw_df = rfp.load_data()\n", - "train_df, test_df = rfp.preprocess(raw_df)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_train_ds = vm.init_dataset(\n", - " train_df,\n", - " text_column=\"question\",\n", - " target_column=\"ground_truth\",\n", - ")\n", - "\n", - "vm_test_ds = vm.init_dataset(\n", - " test_df,\n", - " text_column=\"question\",\n", - " target_column=\"ground_truth\",\n", - ")\n", - "\n", - "vm_test_ds.df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Data validation\n", - "\n", - "Now that we have loaded our dataset, we can go ahead and run some data validation tests right away to start assessing and documenting the quality of our data. Since we are using a text dataset, we can use ValidMind's built-in array of text data quality tests to check that things like number of duplicates, missing values, and other common text data issues are not present in our dataset. We can also run some tests to check the sentiment and toxicity of our data." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Duplicates\n", - "\n", - "First, let's check for duplicates in our dataset. We can use the `validmind.data_validation.Duplicates` test and pass our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "\n", - "run_test(\n", - " test_id=\"validmind.data_validation.Duplicates\",\n", - " inputs={\"dataset\": vm_train_ds},\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Stop Words\n", - "\n", - "Next, let's check for stop words in our dataset. We can use the `validmind.data_validation.StopWords` test and pass our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.StopWords\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Punctuations\n", - "\n", - "Next, let's check for punctuations in our dataset. We can use the `validmind.data_validation.Punctuations` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.Punctuations\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Common Words\n", - "\n", - "Next, let's check for common words in our dataset. We can use the `validmind.data_validation.CommonWord` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.CommonWords\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Language Detection\n", - "\n", - "For documentation purposes, we can detect and log the languages used in the dataset with the `validmind.data_validation.LanguageDetection` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.LanguageDetection\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Toxicity Score\n", - "\n", - "Now, let's go ahead and run the `validmind.data_validation.nlp.Toxicity` test to compute a toxicity score for our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.Toxicity\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Polarity and Subjectivity\n", - "\n", - "We can also run the `validmind.data_validation.nlp.PolarityAndSubjectivity` test to compute the polarity and subjectivity of our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.PolarityAndSubjectivity\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Sentiment\n", - "\n", - "Finally, we can run the `validmind.data_validation.nlp.Sentiment` test to plot the sentiment of our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.Sentiment\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Embedding Model\n", - "\n", - "Now that we have our dataset loaded and have run some data validation tests to assess and document the quality of our data, we can go ahead and initialize our embedding model. We will use `text-embedding-3-small` and `text-embedding-3-large` models from OpenAI for this purpose wrapped in the `OpenAIEmbeddings` class from LangChain. This model will be used to \"embed\" our questions both for inserting the question-answer pairs from the \"train\" set into the vector store and for embedding the question from inputs when making predictions with our RAG model." - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_openai import OpenAIEmbeddings\n", - "\n", - "embedding_small_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", - "\n", - "\n", - "def embed_small(input):\n", - " \"\"\"Returns a text embedding for the given text\"\"\"\n", - " return embedding_small_client.embed_query(input[\"question\"])\n", - "\n", - "\n", - "vm_embedder_small = vm.init_model(input_id=\"embedding_small_model\", predict_fn=embed_small)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "embedding_large_client = OpenAIEmbeddings(model=\"text-embedding-3-large\")\n", - "\n", - "\n", - "def embed_large(input):\n", - " \"\"\"Returns a text embedding for the given text\"\"\"\n", - " return embedding_large_client.embed_query(input[\"question\"])\n", - "\n", - "\n", - "vm_embedder_large = vm.init_model(input_id=\"embedding_large_model\", predict_fn=embed_large)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What we have done here is to initialize the `OpenAIEmbeddings` class so it uses OpenAI's `text-embedding-3-small` and `text-embedding-3-large` models. We then created an `embed` function that takes in an `input` dictionary and uses the `embed_query` method of the embedding client to compute the embeddings of the `question`. We use an `embed` function since that is how ValidMind supports any custom model. We will use this strategy for the retrieval and generator models as well but you could also use, say, a HuggingFace model directly. See the documentation for more information on which model types are directly supported - [ValidMind Documentation](https://docs.validmind.ai/validmind/validmind.html)... Finally, we use the `init_model` function from the ValidMind Library to create a `VMModel` object that can be used in ValidMind tests. This also logs the model to our model documentation and any test that uses the model will be linked to the logged model and its metadata." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Assign Predictions\n", - "\n", - "To precompute the embeddings for our test set, we can call the `assign_predictions` method of our `vm_test_ds` object we created above. This will compute the embeddings for each question in the test set and store them in the a special prediction column of the test set thats linked to our `vm_embedder` model. This will allow us to use these embeddings later when we run tests against our embedding model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(vm_embedder_small)\n", - "vm_test_ds.assign_predictions(vm_embedder_large)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Run tests\n", - "\n", - "Now that everything is setup for the embedding model, we can go ahead and run some tests to assess and document the quality of our embeddings. We will use the `validmind.model_validation.embeddings.*` tests to compute a variety of metrics against our model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisRandomNoise\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - " params={\n", - " \"probability\": 0.3,\n", - " \"mean_similarity_threshold\": 0.7,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisSynonyms\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - " params={\n", - " \"probability\": 0.3,\n", - " \"mean_similarity_threshold\": 0.7,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisTranslation\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - " params={\n", - " \"source_lang\": \"en\",\n", - " \"target_lang\": \"fr\",\n", - " \"mean_similarity_threshold\": 0.7,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.CosineSimilarityHeatmap\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.CosineSimilarityDistribution\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.PCAComponentsPairwisePlots\",\n", - " input_grid={\n", - " \"model\": [vm_embedder_small, vm_embedder_large],\n", - " \"dataset\": [vm_test_ds],\n", - " },\n", - " params={\n", - " \"n_components\": 3,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Setup Vector Store\n", - "\n", - "Great, so now that we have assessed our embedding model and verified that it is performing well, we can go ahead and use it to compute embeddings for our question-answer pairs in the \"train\" set. We will then use these embeddings to insert the question-answer pairs into a vector store. We will use an in-memory `qdrant` vector database for demo purposes but any option would work just as well here. We will use the `QdrantClient` class from LangChain to interact with the vector store. This class will allow us to insert and search for embeddings in the vector store." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Generate embeddings for the Train Set\n", - "\n", - "We can use the same `assign_predictions` method from earlier except this time we will use the `vm_train_ds` object to compute the embeddings for the question-answer pairs in the \"train\" set." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_train_ds.assign_predictions(vm_embedder_small)\n", - "print(vm_train_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Insert embeddings and questions into Vector DB\n", - "\n", - "Now that we have computed the embeddings for our question-answer pairs in the \"train\" set, we can go ahead and insert them into the vector store:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_community.vectorstores import Qdrant\n", - "from langchain_community.document_loaders import DataFrameLoader\n", - "\n", - "# load documents from dataframe\n", - "loader = DataFrameLoader(train_df, page_content_column=\"question\")\n", - "docs = loader.load()\n", - "\n", - "# setup vector datastore\n", - "qdrant = Qdrant.from_documents(\n", - " docs,\n", - " embedding_small_client,\n", - " location=\":memory:\", # Local mode with in-memory storage only\n", - " collection_name=\"rfp_rag_collection\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Retrieval Model\n", - "\n", - "Now that we have an embedding model and a vector database setup and loaded with our data, we need a Retrieval model that can search for similar question-answer pairs for a given input question. Once created, we can initialize this as a ValidMind model and `assign_predictions` to it just like our embedding model. In this example, we'll create two retrieval models with different `k` parameters (the number of documents retrieved) to benchmark and compare their performance. This approach allows us to evaluate how retrieval depth affects the overall system quality." - ] - }, - { - "cell_type": "code", - "execution_count": 26, - "metadata": {}, - "outputs": [], - "source": [ - "def retrieve(input):\n", - " contexts = []\n", - "\n", - " for result in qdrant.similarity_search_with_score(input[\"question\"], k=5):\n", - " document, score = result\n", - " context = f\"Q: {document.page_content}\\n\"\n", - " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", - "\n", - " contexts.append(context)\n", - "\n", - " return contexts\n", - "\n", - "\n", - "vm_retriever_k5 = vm.init_model(input_id=\"retrieval_k5_model\", predict_fn=retrieve)" - ] - }, - { - "cell_type": "code", - "execution_count": 27, - "metadata": {}, - "outputs": [], - "source": [ - "def retrieve(input):\n", - " contexts = []\n", - "\n", - " for result in qdrant.similarity_search_with_score(input[\"question\"], k=10):\n", - " document, score = result\n", - " context = f\"Q: {document.page_content}\\n\"\n", - " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", - "\n", - " contexts.append(context)\n", - "\n", - " return contexts\n", - "\n", - "\n", - "vm_retriever_k10 = vm.init_model(input_id=\"retrieval_k10_model\", predict_fn=retrieve)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(model=vm_retriever_k5)\n", - "vm_test_ds.assign_predictions(model=vm_retriever_k10)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds._df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Generation Model\n", - "\n", - "As the final piece of this simple RAG pipeline, we can create and initialize a generation model that will use the retrieved context to generate an answer to the input question. We will use the `gpt-3.5-turbo` and `gpt-4o` models from OpenAI. Since we have two retrieval models (with different `k` values) and want to test two different LLMs, we'll create a total of four generator models - pairing each retrieval configuration with each LLM to comprehensively evaluate how both retrieval depth and model capability affect response quality." - ] - }, - { - "cell_type": "code", - "execution_count": 30, - "metadata": {}, - "outputs": [], - "source": [ - "from openai import OpenAI\n", - "\n", - "from validmind.models import Prompt\n", - "\n", - "\n", - "system_prompt = \"\"\"\n", - "You are an expert RFP AI assistant.\n", - "You are tasked with answering new RFP questions based on existing RFP questions and answers.\n", - "You will be provided with the existing RFP questions and answer pairs that are the most relevant to the new RFP question.\n", - "After that you will be provided with a new RFP question.\n", - "You will generate an answer and respond only with the answer.\n", - "Ignore your pre-existing knowledge and answer the question based on the provided context.\n", - "\"\"\".strip()\n", - "\n", - "openai_client = OpenAI()" - ] - }, - { - "cell_type": "code", - "execution_count": 31, - "metadata": {}, - "outputs": [], - "source": [ - "def generate(input):\n", - " \n", - " response = openai_client.chat.completions.create(\n", - " model=\"gpt-3.5-turbo\",\n", - " messages=[\n", - " {\"role\": \"system\", \"content\": system_prompt},\n", - " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k5_model\"])},\n", - " {\"role\": \"user\", \"content\": input[\"question\"]},\n", - " ],\n", - " )\n", - " \n", - " return response.choices[0].message.content\n", - "\n", - "\n", - "vm_generator_k5_gpt35 = vm.init_model(\n", - " input_id=\"generation_k5_gpt35_model\",\n", - " predict_fn=generate,\n", - " prompt=Prompt(template=system_prompt),\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": 32, - "metadata": {}, - "outputs": [], - "source": [ - "def generate(input):\n", - " response = openai_client.chat.completions.create(\n", - " model=\"gpt-3.5-turbo\",\n", - " messages=[\n", - " {\"role\": \"system\", \"content\": system_prompt},\n", - " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k10_model\"])},\n", - " {\"role\": \"user\", \"content\": input[\"question\"]},\n", - " ],\n", - " )\n", - "\n", - " return response.choices[0].message.content\n", - "\n", - "\n", - "vm_generator_k10_gpt35 = vm.init_model(\n", - " input_id=\"generation_k10_gpt35_model\",\n", - " predict_fn=generate,\n", - " prompt=Prompt(template=system_prompt),\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": 33, - "metadata": {}, - "outputs": [], - "source": [ - "def generate(input):\n", - " \n", - " response = openai_client.chat.completions.create(\n", - " model=\"gpt-4o\",\n", - " messages=[\n", - " {\"role\": \"system\", \"content\": system_prompt},\n", - " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k5_model\"])},\n", - " {\"role\": \"user\", \"content\": input[\"question\"]},\n", - " ],\n", - " )\n", - " \n", - " return response.choices[0].message.content\n", - "\n", - "\n", - "vm_generator_k5_gpt4o = vm.init_model(\n", - " input_id=\"generation_k5_gpt4o_model\",\n", - " predict_fn=generate,\n", - " prompt=Prompt(template=system_prompt),\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "def generate(input):\n", - " response = openai_client.chat.completions.create(\n", - " model=\"gpt-4o\",\n", - " messages=[\n", - " {\"role\": \"system\", \"content\": system_prompt},\n", - " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k10_model\"])},\n", - " {\"role\": \"user\", \"content\": input[\"question\"]},\n", - " ],\n", - " )\n", - "\n", - " return response.choices[0].message.content\n", - "\n", - "\n", - "vm_generator_k10_gpt4o = vm.init_model(\n", - " input_id=\"generation_k10_gpt4o_model\",\n", - " predict_fn=generate,\n", - " prompt=Prompt(template=system_prompt),\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's test it out real quick:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import pandas as pd\n", - "\n", - "vm_generator_k5_gpt35.predict(\n", - " pd.DataFrame(\n", - " {\"retrieval_k5_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", - " )\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_generator_k5_gpt4o.predict(\n", - " pd.DataFrame(\n", - " {\"retrieval_k5_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", - " )\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Prompt Evaluation\n", - "\n", - "Now that we have our generator model initialized, we can run some LLM-as-Judge tests to evaluate the system prompt. This will allow us to get an initial sense of how well the prompt meets a few best practices for prompt engineering. These tests use an LLM to rate the prompt on a scale of 1-10 against the following criteria:\n", - "\n", - "- **Examplar Bias**: When using multi-shot prompting, does the prompt contain an unbiased distribution of examples?\n", - "- **Delimitation**: When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", - "- **Clarity**: How clearly the prompt states the task.\n", - "- **Conciseness**: How succinctly the prompt states the task.\n", - "- **Instruction Framing**: Whether the prompt contains negative instructions.\n", - "- **Specificity**: How specific the prompt defines the task." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Bias\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Clarity\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Conciseness\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Delimitation\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.NegativeInstruction\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Specificity\",\n", - " inputs={\n", - " \"model\": vm_generator_k5_gpt4o,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Setup RAG Pipeline Model\n", - "\n", - "Now that we have all of our individual \"component\" models setup and initialized we need some way to put them all together in a single \"pipeline\". We can use the `PipelineModel` class to do this. This ValidMind model type simply wraps any number of other ValidMind models and runs them in sequence. We can use a pipe(`|`) operator - in Python this is normally an `or` operator but we have overloaded it for easy pipeline creation - to chain together our models. We can then initialize this pipeline model and assign predictions to it just like any other model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_rag_k5_gpt35_model = vm.init_model(vm_retriever_k5 | vm_generator_k5_gpt35, input_id=\"rag_k5_gpt35_model\")\n", - "vm_rag_k10_gpt35_model = vm.init_model(vm_retriever_k10 | vm_generator_k10_gpt35, input_id=\"rag_k10_gpt35_model\")\n", - "vm_rag_k5_gpt4o_model = vm.init_model(vm_retriever_k5 | vm_generator_k5_gpt4o, input_id=\"rag_k5_gpt4o_model\")\n", - "vm_rag_k10_gpt4o_model = vm.init_model(vm_retriever_k10 | vm_generator_k10_gpt4o, input_id=\"rag_k10_gpt4o_model\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We can `assign_predictions` to the pipeline model just like we did with the individual models. This will run the pipeline on the test set and store the results in the test set for later use." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(model=vm_rag_k5_gpt35_model)\n", - "vm_test_ds.assign_predictions(model=vm_rag_k10_gpt35_model)\n", - "vm_test_ds.assign_predictions(model=vm_rag_k5_gpt4o_model)\n", - "vm_test_ds.assign_predictions(model=vm_rag_k10_gpt4o_model)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds._df.head(5)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Run tests" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## RAGAS evaluation\n", - "\n", - "Let's go ahead and run some of our new RAG tests against our model...\n", - "\n", - "> Note: these tests are still being developed and are not yet in a stable state. We are using advanced tests here that use LLM-as-Judge and other strategies to assess things like the relevancy of the retrieved context to the input question and the correctness of the generated answer when compared to the ground truth. There is more to come in this area so stay tuned!" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import warnings\n", - "\n", - "warnings.filterwarnings(\"ignore\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Semantic Similarity\n", - "\n", - "The concept of Answer Semantic Similarity pertains to the assessment of the semantic resemblance between the generated answer and the ground truth. This evaluation is based on the ground truth and the answer, with values falling within the range of 0 to 1. A higher score signifies a better alignment between the generated answer and the ground truth.\n", - "\n", - "Measuring the semantic similarity between answers can offer valuable insights into the quality of the generated response. This evaluation utilizes a cross-encoder model to calculate the semantic similarity score." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.SemanticSimilarity\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"reference_column\": [\"ground_truth\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Entity Recall\n", - "\n", - "This test gives the measure of recall of the retrieved context, based on the number of entities present in both ground_truths and contexts relative to the number of entities present in the ground_truths alone. Simply put, it is a measure of what fraction of entities are recalled from ground_truths. This test is useful in fact-based use cases like tourism help desk, historical QA, etc. This test can help evaluate the retrieval mechanism for entities, based on comparison with entities present in ground_truths, because in cases where entities matter, we need the contexts which cover them." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextEntityRecall\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"reference_column\": [\"ground_truth\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Precision\n", - "\n", - "Context Precision is a test that evaluates whether all of the ground-truth relevant items present in the contexts are ranked higher or not. Ideally all the relevant chunks must appear at the top ranks. This test is computed using the question, ground_truth and the contexts, with values ranging between 0 and 1, where higher scores indicate better precision." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextPrecision\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", - " \"reference_column\": [\"ground_truth\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Precision Without Reference\n", - "\n", - "This test evaluates whether retrieved contexts align well with the expected response for a given user input, without requiring a ground-truth reference. This test assesses the relevance of each retrieved context chunk by comparing it directly to the response." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid=[\n", - " {\"user_input_column\": \"question\",\n", - " \"retrieved_contexts_column\": \"retrieval_k5_model_prediction\",\n", - " \"response_column\": \"rag_k5_gpt4o_model_prediction\"\n", - " },\n", - " {\"user_input_column\": \"question\",\n", - " \"retrieved_contexts_column\": \"retrieval_k10_model_prediction\",\n", - " \"response_column\": \"rag_k10_gpt4o_model_prediction\"\n", - " },\n", - " ],\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Faithfulness\n", - "\n", - "This measures the factual consistency of the generated answer against the given context. It is calculated from answer and retrieved context. The answer is scaled to (0,1) range. Higher the better.\n", - "\n", - "The generated answer is regarded as faithful if all the claims that are made in the answer can be inferred from the given context. To calculate this a set of claims from the generated answer is first identified. Then each one of these claims are cross checked with given context to determine if it can be inferred from given context or not." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.Faithfulness\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Response Relevancy\n", - "\n", - "The Response Relevancy test, focuses on assessing how pertinent the generated answer is to the given prompt. A lower score is assigned to answers that are incomplete or contain redundant information and higher scores indicate better relevancy. This test is computed using the question, the context and the answer.\n", - "\n", - "The Response Relevancy is defined as the mean cosine similartiy of the original question to a number of artifical questions, which where generated (reverse engineered) based on the answer.\n", - "\n", - "Please note, that eventhough in practice the score will range between 0 and 1 most of the time, this is not mathematically guranteed, due to the nature of the cosine similarity ranging from -1 to 1.\n", - "\n", - "> Note: This is a reference free test. If you’re looking to compare ground truth answer with generated answer refer to Answer Correctness.\n", - "\n", - "An answer is deemed relevant when it directly and appropriately addresses the original question. Importantly, our assessment of answer relevance does not consider factuality but instead penalizes cases where the answer lacks completeness or contains redundant details. To calculate this score, the LLM is prompted to generate an appropriate question for the generated answer multiple times, and the mean cosine similarity between these generated questions and the original question is measured. The underlying idea is that if the generated answer accurately addresses the initial question, the LLM should be able to generate questions from the answer that align with the original question." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Recall\n", - "\n", - "Context recall measures the extent to which the retrieved context aligns with the annotated answer, treated as the ground truth. It is computed based on the ground truth and the retrieved context, and the values range between 0 and 1, with higher values indicating better performance.\n", - "\n", - "To estimate context recall from the ground truth answer, each sentence in the ground truth answer is analyzed to determine whether it can be attributed to the retrieved context or not. In an ideal scenario, all sentences in the ground truth answer should be attributable to the retrieved context." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextRecall\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", - " \"reference_column\": [\"ground_truth\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Answer Correctness\n", - "\n", - "The assessment of Answer Correctness involves gauging the accuracy of the generated answer when compared to the ground truth. This evaluation relies on the ground truth and the answer, with scores ranging from 0 to 1. A higher score indicates a closer alignment between the generated answer and the ground truth, signifying better correctness.\n", - "\n", - "Answer correctness encompasses two critical aspects: semantic similarity between the generated answer and the ground truth, as well as factual similarity. These aspects are combined using a weighted scheme to formulate the answer correctness score.\n", - "\n", - "Factual correctness quantifies the factual overlap between the generated answer and the ground truth answer. This is done using the concepts of:\n", - "\n", - "- TP (True Positive): Facts or statements that are present in both the ground truth and the generated answer.\n", - "- FP (False Positive): Facts or statements that are present in the generated answer but not in the ground truth.\n", - "- FN (False Negative): Facts or statements that are present in the ground truth but not in the generated answer." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.AnswerCorrectness\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"reference_column\": [\"ground_truth\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Aspect Critic\n", - "\n", - "This is designed to assess submissions based on predefined aspects such as harmlessness and correctness. Additionally, users have the flexibility to define their own aspects for evaluating submissions according to their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not. This evaluation is performed using the ‘answer’ as input.\n", - "\n", - "Critiques within the LLM evaluators evaluate submissions based on the provided aspect. Ragas Critiques offers a range of predefined aspects like correctness, harmfulness, etc. Users can also define their own aspects for evaluating submissions based on their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.AspectCritic\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Noise Sensitivity\n", - "\n", - "This test is designed to evaluate the robustness of the RAG pipeline model against noise in the retrieved context. It works by checking how well the \"claims\" in the generated answer match up with the \"claims\" in the ground truth answer. If the generated answer contains \"claims\" from the contexts that the ground truth answer does not contain, those claims are considered incorrect. The score for each answer is the number of incorrect claims divided by the total number of claims. This *can* be interpreted as a measure of how sensitive the LLM is to \"noise\" in the context where \"noise\" is information that is relevant but should not be included in the answer since the ground truth answer does not contain it." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.NoiseSensitivity\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " param_grid={\n", - " \"user_input_column\": [\"question\"],\n", - " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", - " \"reference_column\": [\"ground_truth\"],\n", - " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Generation quality\n", - "\n", - "In this section, we evaluate the alignment and relevance of generated responses to reference outputs within our retrieval-augmented generation (RAG) application. We use metrics that assess various quality dimensions of the generated responses, including semantic similarity, structural alignment, and phrasing overlap. Semantic similarity metrics compare embeddings of generated and reference text to capture deeper contextual alignment, while overlap and alignment measures quantify how well the phrasing and structure of generated responses match the intended outputs." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Token Disparity\n", - "\n", - "This test assesses the difference in token counts between the reference texts (ground truth) and the answers generated by the RAG model. It helps evaluate how well the model's outputs align with the expected length and level of detail in the reference texts. A significant disparity in token counts could signal issues with generation quality, such as excessive verbosity or insufficient detail. Consistently low token counts in generated answers compared to references might suggest that the model’s outputs are incomplete or overly concise, missing important contextual information." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.TokenDisparity\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### ROUGE Score\n", - "\n", - "This test evaluates the quality of answers generated by the RAG model by measuring overlaps in n-grams, word sequences, and word pairs between the model output and the reference (ground truth) text. ROUGE, short for Recall-Oriented Understudy for Gisting Evaluation, assesses both precision and recall, providing a balanced view of how well the generated response captures the reference content. ROUGE precision measures the proportion of n-grams in the generated text that match the reference, highlighting relevance and conciseness, while ROUGE recall assesses the proportion of reference n-grams present in the generated text, indicating completeness and thoroughness. \n", - "\n", - "Low precision scores might reveal that the generated text includes redundant or irrelevant information, while low recall scores suggest omissions of essential details from the reference. Consistently low ROUGE scores could indicate poor overall alignment with the ground truth, suggesting the model may be missing key content or failing to capture the intended meaning." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.RougeScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - " params={\n", - " \"metric\": \"rouge-1\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### BLEU Score\n", - "\n", - "The BLEU Score test evaluates the quality of answers generated by the RAG model by measuring n-gram overlap between the generated text and the reference (ground truth) text, with a specific focus on exact precision in phrasing. While ROUGE precision also assesses overlap, BLEU differs in two main ways: first, it applies a geometric average across multiple n-gram levels, capturing precise phrase alignment, and second, it includes a brevity penalty to prevent overly short outputs from inflating scores artificially. This added precision focus is valuable in RAG applications where strict adherence to reference language is essential, as BLEU emphasizes the match to exact phrasing. In contrast, ROUGE precision evaluates general content overlap without penalizing brevity, offering a broader sense of content alignment." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.BleuScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### BERT Score\n", - "\n", - "This test evaluates the quality of the RAG generated answers using BERT embeddings to measure precision, recall, and F1 scores based on semantic similarity, rather than exact n-gram matches as in BLEU and ROUGE. This approach captures contextual meaning, making it valuable when wording differs but the intended message closely aligns with the reference. In RAG applications, the BERT score is especially useful for ensuring that generated answers convey the reference text’s meaning, even if phrasing varies. Consistently low scores indicate a lack of semantic alignment, suggesting the model may miss or misrepresent key content. Low precision may reflect irrelevant or redundant details, while low recall can indicate omissions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.BertScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### METEOR Score\n", - "\n", - "This test evaluates the quality of the generated answers by measuring alignment with the ground truth, emphasizing both accuracy and fluency. Unlike BLEU and ROUGE, which focus on n-gram matches, METEOR combines precision, recall, synonym matching, and word order, focusing at how well the generated text conveys meaning and reads naturally. This metric is especially useful for RAG applications where sentence structure and natural flow are crucial for clear communication. Lower scores may suggest alignment issues, indicating that the answers may lack fluency or key content. Discrepancies in word order or high fragmentation penalties can reveal problems with how the model constructs sentences, potentially affecting readability." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.MeteorScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Bias and Toxicity\n", - "\n", - "In this section, we use metrics like Toxicity Score and Regard Score to evaluate both the generated responses and the ground truth. These tests helps us detect any harmful, offensive, or inappropriate language and evaluate the level of bias and neutrality enabling us to assess and mitigate potential biases in both the model's responses and the original dataset." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Toxicity Score\n", - "\n", - "This test measures the level of harmful or offensive content in the generated answers. The test uses a preloaded toxicity detection tool from Hugging Face, which identifies language that may be inappropriate, aggressive, or derogatory. High toxicity scores indicate potentially toxic content, while consistently elevated scores across multiple outputs may signal underlying issues in the model’s generation process that require attention to prevent the spread of harmful language." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ToxicityScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Regard Score\n", - "\n", - "This test evaluates the sentiment and perceived regard—categorized as positive, negative, neutral, or other—in answers generated by the RAG model. This is important for identifying any biases or sentiment tendencies in responses, ensuring that generated answers are balanced and appropriate for the context. The uses a preloaded regard evaluation tool from Hugging Face to compute scores for each response. High skewness in regard scores, especially if the generated responses consistently diverge from expected sentiments in the reference texts, may reveal biases in the model’s generation, such as overly positive or negative tones where neutrality is expected." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.RegardScore\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Conclusion\n", - "\n", - "In this notebook, we have seen how we can use LangChain and ValidMind together to build, evaluate and document a simple RAG Model as its developed. This is a great example of the interactive development experience that ValidMind is designed to support. We can quickly iterate on our model and document as we go... We have seen how ValidMind supports non-traditional \"models\" using a functional interface and how we can build pipelines of many models to support complex GenAI workflows.\n", - "\n", - "This is still a work in progress and we are actively developing new tests to support more advanced GenAI workflows. We are also keeping an eye on the most popular GenAI models and libraries to explore direct integrations. Stay tuned for more updates and new features in this area!" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Upgrade ValidMind\n", - "\n", - "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", - "\n", - "Retrieve the information for the currently installed version of ValidMind:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip show validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", - "\n", - "```bash\n", - "%pip install --upgrade validmind\n", - "```" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You may need to restart your kernel after running the upgrade package for changes to be applied." - ] - }, - { - "cell_type": "markdown", - "id": "copyright-09e315440ca84258abe1aaefaca3a3d0", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind Library", - "language": "python", - "name": "validmind" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.10.15" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/site/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb b/site/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb deleted file mode 100644 index 485c81ffdb..0000000000 --- a/site/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb +++ /dev/null @@ -1,1681 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# RAG Model Documentation Demo\n", - "\n", - "In this notebook, we are going to implement a simple RAG Model for automating the process of answering RFP questions using GenAI. We will see how we can initialize an embedding model, a retrieval model and a generator model with LangChain components and use them within the ValidMind Library to run tests against them. Finally, we will see how we can put them together in a Pipeline and run that to get e2e results and run tests against that." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About ValidMind](#toc1__) \n", - " - [Before you begin](#toc1_1__) \n", - " - [New to ValidMind?](#toc1_2__) \n", - " - [Key concepts](#toc1_3__) \n", - "- [Setting up](#toc2__) \n", - " - [Initialize the ValidMind Library](#toc2_1__) \n", - " - [Register sample model](#toc2_1_1__) \n", - " - [Apply documentation template](#toc2_1_2__) \n", - " - [Get your code snippet](#toc2_1_3__) \n", - "- [Read Open AI API Key](#toc3__) \n", - "- [Dataset Loader](#toc4__) \n", - "- [Data validation](#toc5__) \n", - " - [Duplicates](#toc5_1__) \n", - " - [Stop Words](#toc5_2__) \n", - " - [Punctuations](#toc5_3__) \n", - " - [Common Words](#toc5_4__) \n", - " - [Language Detection](#toc5_5__) \n", - " - [Toxicity Score](#toc5_6__) \n", - " - [Polarity and Subjectivity](#toc5_7__) \n", - " - [Sentiment](#toc5_8__) \n", - " - [Assign Predictions](#toc5_9__) \n", - " - [Run tests](#toc5_10__) \n", - " - [Generate embeddings for the Train Set](#toc5_11__) \n", - " - [Insert embeddings and questions into Vector DB](#toc5_12__) \n", - "- [Prompt Evaluation](#toc6__) \n", - "- [RAGAS evaluation](#toc7__) \n", - " - [Semantic Similarity](#toc7_1__) \n", - " - [Context Entity Recall](#toc7_2__) \n", - " - [Context Precision](#toc7_3__) \n", - " - [Context Precision Without Reference](#toc7_4__) \n", - " - [Faithfulness](#toc7_5__) \n", - " - [Response Relevancy](#toc7_6__) \n", - " - [Context Recall](#toc7_7__) \n", - " - [Answer Correctness](#toc7_8__) \n", - " - [Aspect Critic](#toc7_9__) \n", - " - [Noise Sensitivity](#toc7_10__) \n", - "- [Generation quality](#toc8__) \n", - " - [Token Disparity](#toc8_1__) \n", - " - [ROUGE Score](#toc8_2__) \n", - " - [BLEU Score](#toc8_3__) \n", - " - [BERT Score](#toc8_4__) \n", - " - [METEOR Score](#toc8_5__) \n", - "- [Bias and Toxicity](#toc9__) \n", - " - [Toxicity Score](#toc9_1__) \n", - " - [Regard Score](#toc9_2__) \n", - "- [Upgrade ValidMind](#toc10__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", - "\n", - "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "- **FunctionModels**: ValidMind offers support for creating `VMModel` instances from Python functions. This enables us to support any \"model\" by simply using the provided function as the model's `predict` method.\n", - "- **PipelineModels**: ValidMind models (`VMModel` instances) of any type can be piped together to create a model pipeline. This allows model components to be created and tested/documented independently, and then combined into a single model for end-to-end testing and documentation. We use the `|` operator to pipe models together.\n", - "- **RAG**: RAG stands for Retrieval Augmented Generation and refers to a wide range of GenAI applications where some form of retrieval is used to add context to the prompt so that the LLM that generates content can refer to it when creating its output. In this notebook, we are going to implement a simple RAG setup using LangChain components." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Prerequisites\n", - "\n", - "Let's go ahead and install the `validmind` library if its not already installed... Then we can install the `qdrant-client` library for our vector store and `langchain` for everything else:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q qdrant-client langchain langchain-openai sentencepiece" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Gen AI RAG Template`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " # api_host=\"...\",\n", - " # api_key=\"...\",\n", - " # api_secret=\"...\",\n", - " # model=\"...\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Read Open AI API Key\n", - "\n", - "We will need to have an OpenAI API key to be able to use their `text-embedding-3-small` model for our embeddings, `gpt-3.5-turbo` model for our generator and `gpt-4o` model for our LLM-as-Judge tests. If you don't have an OpenAI API key, you can get one by signing up at [OpenAI](https://platform.openai.com/signup). Then you can create a `.env` file in the root of your project and the following cell will load it from there. Alternatively, you can just uncomment the line below to directly set the key (not recommended for security reasons)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# load openai api key\n", - "import os\n", - "\n", - "import dotenv\n", - "import nltk\n", - "\n", - "dotenv.load_dotenv()\n", - "nltk.download('stopwords')\n", - "nltk.download('punkt_tab')\n", - "\n", - "# os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n", - "\n", - "if not \"OPENAI_API_KEY\" in os.environ:\n", - " raise ValueError(\"OPENAI_API_KEY is not set\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Dataset Loader\n", - "\n", - "Great, now that we have all of our dependencies installed, the ValidMind Library initialized and connected to our model and our OpenAI API key setup, we can go ahead and load our datasets. We will use the synthetic `RFP` dataset included with ValidMind for this notebook. This dataset contains a variety of RFP questions and ground truth answers that we can use both as the source where our Retriever will search for similar question-answer pairs as well as our test set for evaluating the performance of our RAG model. To do this, we just have to load it and call the preprocess function to get a split of the data into train and test sets." - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the sample dataset from the library\n", - "from validmind.datasets.llm.rag import rfp\n", - "\n", - "raw_df = rfp.load_data()\n", - "train_df, test_df = rfp.preprocess(raw_df)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_train_ds = vm.init_dataset(\n", - " train_df,\n", - " text_column=\"question\",\n", - " target_column=\"ground_truth\",\n", - ")\n", - "\n", - "vm_test_ds = vm.init_dataset(\n", - " test_df,\n", - " text_column=\"question\",\n", - " target_column=\"ground_truth\",\n", - ")\n", - "\n", - "vm_test_ds.df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Data validation\n", - "\n", - "Now that we have loaded our dataset, we can go ahead and run some data validation tests right away to start assessing and documenting the quality of our data. Since we are using a text dataset, we can use ValidMind's built-in array of text data quality tests to check that things like number of duplicates, missing values, and other common text data issues are not present in our dataset. We can also run some tests to check the sentiment and toxicity of our data." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Duplicates\n", - "\n", - "First, let's check for duplicates in our dataset. We can use the `validmind.data_validation.Duplicates` test and pass our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "\n", - "run_test(\n", - " test_id=\"validmind.data_validation.Duplicates\",\n", - " inputs={\"dataset\": vm_train_ds},\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Stop Words\n", - "\n", - "Next, let's check for stop words in our dataset. We can use the `validmind.data_validation.StopWords` test and pass our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.StopWords\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Punctuations\n", - "\n", - "Next, let's check for punctuations in our dataset. We can use the `validmind.data_validation.Punctuations` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.Punctuations\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Common Words\n", - "\n", - "Next, let's check for common words in our dataset. We can use the `validmind.data_validation.CommonWord` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.CommonWords\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Language Detection\n", - "\n", - "For documentation purposes, we can detect and log the languages used in the dataset with the `validmind.data_validation.LanguageDetection` test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " test_id=\"validmind.data_validation.nlp.LanguageDetection\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Toxicity Score\n", - "\n", - "Now, let's go ahead and run the `validmind.data_validation.nlp.Toxicity` test to compute a toxicity score for our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.Toxicity\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Polarity and Subjectivity\n", - "\n", - "We can also run the `validmind.data_validation.nlp.PolarityAndSubjectivity` test to compute the polarity and subjectivity of our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.PolarityAndSubjectivity\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Sentiment\n", - "\n", - "Finally, we can run the `validmind.data_validation.nlp.Sentiment` test to plot the sentiment of our dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.data_validation.nlp.Sentiment\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Embedding Model\n", - "\n", - "Now that we have our dataset loaded and have run some data validation tests to assess and document the quality of our data, we can go ahead and initialize our embedding model. We will use the `text-embedding-3-small` model from OpenAI for this purpose wrapped in the `OpenAIEmbeddings` class from LangChain. This model will be used to \"embed\" our questions both for inserting the question-answer pairs from the \"train\" set into the vector store and for embedding the question from inputs when making predictions with our RAG model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_openai import OpenAIEmbeddings\n", - "\n", - "embedding_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", - "\n", - "\n", - "def embed(input):\n", - " \"\"\"Returns a text embedding for the given text\"\"\"\n", - " return embedding_client.embed_query(input[\"question\"])\n", - "\n", - "\n", - "vm_embedder = vm.init_model(input_id=\"embedding_model\", predict_fn=embed)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What we have done here is to initialize the `OpenAIEmbeddings` class so it uses OpenAI's `text-embedding-3-small` model. We then created an `embed` function that takes in an `input` dictionary and uses the `embed_query` method of the embedding client to compute the embeddings of the `question`. We use an `embed` function since that is how ValidMind supports any custom model. We will use this strategy for the retrieval and generator models as well but you could also use, say, a HuggingFace model directly. See the documentation for more information on which model types are directly supported - [ValidMind Documentation](https://docs.validmind.ai/validmind/validmind.html)... Finally, we use the `init_model` function from the ValidMind Library to create a `VMModel` object that can be used in ValidMind tests. This also logs the model to our model documentation and any test that uses the model will be linked to the logged model and its metadata." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Assign Predictions\n", - "\n", - "To precompute the embeddings for our test set, we can call the `assign_predictions` method of our `vm_test_ds` object we created above. This will compute the embeddings for each question in the test set and store them in the a special prediction column of the test set thats linked to our `vm_embedder` model. This will allow us to use these embeddings later when we run tests against our embedding model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(vm_embedder)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Run tests\n", - "\n", - "Now that everything is setup for the embedding model, we can go ahead and run some tests to assess and document the quality of our embeddings. We will use the `validmind.model_validation.embeddings.*` tests to compute a variety of metrics against our model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisRandomNoise\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " params={\"probability\": 0.3},\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisSynonyms\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " params={\"probability\": 0.3},\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.StabilityAnalysisTranslation\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " params={\n", - " \"source_lang\": \"en\",\n", - " \"target_lang\": \"fr\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.CosineSimilarityHeatmap\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "run_test(\n", - " \"validmind.model_validation.embeddings.CosineSimilarityDistribution\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.EuclideanDistanceHeatmap\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.PCAComponentsPairwisePlots\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " params={\"n_components\": 3},\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.embeddings.TSNEComponentsPairwisePlots\",\n", - " inputs={\n", - " \"model\": vm_embedder,\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " params={\"n_components\": 3, \"perplexity\": 20},\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Setup Vector Store\n", - "\n", - "Great, so now that we have assessed our embedding model and verified that it is performing well, we can go ahead and use it to compute embeddings for our question-answer pairs in the \"train\" set. We will then use these embeddings to insert the question-answer pairs into a vector store. We will use an in-memory `qdrant` vector database for demo purposes but any option would work just as well here. We will use the `QdrantClient` class from LangChain to interact with the vector store. This class will allow us to insert and search for embeddings in the vector store." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Generate embeddings for the Train Set\n", - "\n", - "We can use the same `assign_predictions` method from earlier except this time we will use the `vm_train_ds` object to compute the embeddings for the question-answer pairs in the \"train\" set." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_train_ds.assign_predictions(vm_embedder)\n", - "print(vm_train_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Insert embeddings and questions into Vector DB\n", - "\n", - "Now that we have computed the embeddings for our question-answer pairs in the \"train\" set, we can go ahead and insert them into the vector store:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_community.vectorstores import Qdrant\n", - "from langchain_openai import OpenAIEmbeddings\n", - "from langchain_community.document_loaders import DataFrameLoader\n", - "\n", - "# load documents from dataframe\n", - "loader = DataFrameLoader(train_df, page_content_column=\"question\")\n", - "docs = loader.load()\n", - "# choose model using embedding client\n", - "embedding_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", - "\n", - "# setup vector datastore\n", - "qdrant = Qdrant.from_documents(\n", - " docs,\n", - " embedding_client,\n", - " location=\":memory:\", # Local mode with in-memory storage only\n", - " collection_name=\"rfp_rag_collection\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Retrieval Model\n", - "\n", - "Now that we have an embedding model and a vector database setup and loaded with our data, we need a Retrieval model that can search for similar question-answer pairs for a given input question. Once created, we can initialize this as a ValidMind model and `assign_predictions` to it just like our embedding model." - ] - }, - { - "cell_type": "code", - "execution_count": 27, - "metadata": {}, - "outputs": [], - "source": [ - "def retrieve(input):\n", - " contexts = []\n", - "\n", - " for result in qdrant.similarity_search_with_score(input[\"question\"]):\n", - " document, score = result\n", - " context = f\"Q: {document.page_content}\\n\"\n", - " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", - "\n", - " contexts.append(context)\n", - "\n", - " return contexts\n", - "\n", - "\n", - "vm_retriever = vm.init_model(input_id=\"retrieval_model\", predict_fn=retrieve)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(model=vm_retriever)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Generation Model\n", - "\n", - "As the final piece of this simple RAG pipeline, we can create and initialize a generation model that will use the retrieved context to generate an answer to the input question. We will use the `gpt-3.5-turbo` model from OpenAI." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from openai import OpenAI\n", - "\n", - "from validmind.models import Prompt\n", - "\n", - "\n", - "system_prompt = \"\"\"\n", - "You are an expert RFP AI assistant.\n", - "You are tasked with answering new RFP questions based on existing RFP questions and answers.\n", - "You will be provided with the existing RFP questions and answer pairs that are the most relevant to the new RFP question.\n", - "After that you will be provided with a new RFP question.\n", - "You will generate an answer and respond only with the answer.\n", - "Ignore your pre-existing knowledge and answer the question based on the provided context.\n", - "\"\"\".strip()\n", - "\n", - "openai_client = OpenAI()\n", - "\n", - "\n", - "def generate(input):\n", - " response = openai_client.chat.completions.create(\n", - " model=\"gpt-3.5-turbo\",\n", - " messages=[\n", - " {\"role\": \"system\", \"content\": system_prompt},\n", - " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_model\"])},\n", - " {\"role\": \"user\", \"content\": input[\"question\"]},\n", - " ],\n", - " )\n", - "\n", - " return response.choices[0].message.content\n", - "\n", - "\n", - "vm_generator = vm.init_model(\n", - " input_id=\"generation_model\",\n", - " predict_fn=generate,\n", - " prompt=Prompt(template=system_prompt),\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's test it out real quick:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import pandas as pd\n", - "\n", - "vm_generator.predict(\n", - " pd.DataFrame(\n", - " {\"retrieval_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", - " )\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Prompt Evaluation\n", - "\n", - "Now that we have our generator model initialized, we can run some LLM-as-Judge tests to evaluate the system prompt. This will allow us to get an initial sense of how well the prompt meets a few best practices for prompt engineering. These tests use an LLM to rate the prompt on a scale of 1-10 against the following criteria:\n", - "\n", - "- **Examplar Bias**: When using multi-shot prompting, does the prompt contain an unbiased distribution of examples?\n", - "- **Delimitation**: When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", - "- **Clarity**: How clearly the prompt states the task.\n", - "- **Conciseness**: How succinctly the prompt states the task.\n", - "- **Instruction Framing**: Whether the prompt contains negative instructions.\n", - "- **Specificity**: How specific the prompt defines the task." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Bias\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Clarity\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Conciseness\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Delimitation\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.NegativeInstruction\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.prompt_validation.Specificity\",\n", - " inputs={\n", - " \"model\": vm_generator,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Setup RAG Pipeline Model\n", - "\n", - "Now that we have all of our individual \"component\" models setup and initialized we need some way to put them all together in a single \"pipeline\". We can use the `PipelineModel` class to do this. This ValidMind model type simply wraps any number of other ValidMind models and runs them in sequence. We can use a pipe(`|`) operator - in Python this is normally an `or` operator but we have overloaded it for easy pipeline creation - to chain together our models. We can then initialize this pipeline model and assign predictions to it just like any other model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_rag_model = vm.init_model(vm_retriever | vm_generator, input_id=\"rag_model\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We can `assign_predictions` to the pipeline model just like we did with the individual models. This will run the pipeline on the test set and store the results in the test set for later use." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds.assign_predictions(model=vm_rag_model)\n", - "print(vm_test_ds)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_test_ds._df.head(5)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Run tests" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## RAGAS evaluation\n", - "\n", - "Let's go ahead and run some of our new RAG tests against our model...\n", - "\n", - "> Note: these tests are still being developed and are not yet in a stable state. We are using advanced tests here that use LLM-as-Judge and other strategies to assess things like the relevancy of the retrieved context to the input question and the correctness of the generated answer when compared to the ground truth. There is more to come in this area so stay tuned!" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import warnings\n", - "\n", - "warnings.filterwarnings(\"ignore\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Semantic Similarity\n", - "\n", - "The concept of Answer Semantic Similarity pertains to the assessment of the semantic resemblance between the generated answer and the ground truth. This evaluation is based on the ground truth and the answer, with values falling within the range of 0 to 1. A higher score signifies a better alignment between the generated answer and the ground truth.\n", - "\n", - "Measuring the semantic similarity between answers can offer valuable insights into the quality of the generated response. This evaluation utilizes a cross-encoder model to calculate the semantic similarity score." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.SemanticSimilarity\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"reference_column\": \"ground_truth\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Entity Recall\n", - "\n", - "This test gives the measure of recall of the retrieved context, based on the number of entities present in both ground_truths and contexts relative to the number of entities present in the ground_truths alone. Simply put, it is a measure of what fraction of entities are recalled from ground_truths. This test is useful in fact-based use cases like tourism help desk, historical QA, etc. This test can help evaluate the retrieval mechanism for entities, based on comparison with entities present in ground_truths, because in cases where entities matter, we need the contexts which cover them." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextEntityRecall\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"reference_column\": \"ground_truth\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Precision\n", - "\n", - "Context Precision is a test that evaluates whether all of the ground-truth relevant items present in the contexts are ranked higher or not. Ideally all the relevant chunks must appear at the top ranks. This test is computed using the question, ground_truth and the contexts, with values ranging between 0 and 1, where higher scores indicate better precision." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextPrecision\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " \"reference_column\": \"ground_truth\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Precision Without Reference\n", - "\n", - "This test evaluates whether retrieved contexts align well with the expected response for a given user input, without requiring a ground-truth reference. This test assesses the relevance of each retrieved context chunk by comparing it directly to the response." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Faithfulness\n", - "\n", - "This measures the factual consistency of the generated answer against the given context. It is calculated from answer and retrieved context. The answer is scaled to (0,1) range. Higher the better.\n", - "\n", - "The generated answer is regarded as faithful if all the claims that are made in the answer can be inferred from the given context. To calculate this a set of claims from the generated answer is first identified. Then each one of these claims are cross checked with given context to determine if it can be inferred from given context or not." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.Faithfulness\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Response Relevancy\n", - "\n", - "The Response Relevancy test, focuses on assessing how pertinent the generated answer is to the given prompt. A lower score is assigned to answers that are incomplete or contain redundant information and higher scores indicate better relevancy. This test is computed using the question, the context and the answer.\n", - "\n", - "The Response Relevancy is defined as the mean cosine similartiy of the original question to a number of artifical questions, which where generated (reverse engineered) based on the answer.\n", - "\n", - "Please note, that eventhough in practice the score will range between 0 and 1 most of the time, this is not mathematically guranteed, due to the nature of the cosine similarity ranging from -1 to 1.\n", - "\n", - "> Note: This is a reference free test. If you’re looking to compare ground truth answer with generated answer refer to Answer Correctness.\n", - "\n", - "An answer is deemed relevant when it directly and appropriately addresses the original question. Importantly, our assessment of answer relevance does not consider factuality but instead penalizes cases where the answer lacks completeness or contains redundant details. To calculate this score, the LLM is prompted to generate an appropriate question for the generated answer multiple times, and the mean cosine similarity between these generated questions and the original question is measured. The underlying idea is that if the generated answer accurately addresses the initial question, the LLM should be able to generate questions from the answer that align with the original question." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Context Recall\n", - "\n", - "Context recall measures the extent to which the retrieved context aligns with the annotated answer, treated as the ground truth. It is computed based on the ground truth and the retrieved context, and the values range between 0 and 1, with higher values indicating better performance.\n", - "\n", - "To estimate context recall from the ground truth answer, each sentence in the ground truth answer is analyzed to determine whether it can be attributed to the retrieved context or not. In an ideal scenario, all sentences in the ground truth answer should be attributable to the retrieved context." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.ContextRecall\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " \"reference_column\": \"ground_truth\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Answer Correctness\n", - "\n", - "The assessment of Answer Correctness involves gauging the accuracy of the generated answer when compared to the ground truth. This evaluation relies on the ground truth and the answer, with scores ranging from 0 to 1. A higher score indicates a closer alignment between the generated answer and the ground truth, signifying better correctness.\n", - "\n", - "Answer correctness encompasses two critical aspects: semantic similarity between the generated answer and the ground truth, as well as factual similarity. These aspects are combined using a weighted scheme to formulate the answer correctness score.\n", - "\n", - "Factual correctness quantifies the factual overlap between the generated answer and the ground truth answer. This is done using the concepts of:\n", - "\n", - "- TP (True Positive): Facts or statements that are present in both the ground truth and the generated answer.\n", - "- FP (False Positive): Facts or statements that are present in the generated answer but not in the ground truth.\n", - "- FN (False Negative): Facts or statements that are present in the ground truth but not in the generated answer." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.AnswerCorrectness\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"reference_column\": \"ground_truth\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Aspect Critic\n", - "\n", - "This is designed to assess submissions based on predefined aspects such as harmlessness and correctness. Additionally, users have the flexibility to define their own aspects for evaluating submissions according to their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not. This evaluation is performed using the ‘answer’ as input.\n", - "\n", - "Critiques within the LLM evaluators evaluate submissions based on the provided aspect. Ragas Critiques offers a range of predefined aspects like correctness, harmfulness, etc. Users can also define their own aspects for evaluating submissions based on their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.AspectCritic\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Noise Sensitivity\n", - "\n", - "This test is designed to evaluate the robustness of the RAG pipeline model against noise in the retrieved context. It works by checking how well the \"claims\" in the generated answer match up with the \"claims\" in the ground truth answer. If the generated answer contains \"claims\" from the contexts that the ground truth answer does not contain, those claims are considered incorrect. The score for each answer is the number of incorrect claims divided by the total number of claims. This *can* be interpreted as a measure of how sensitive the LLM is to \"noise\" in the context where \"noise\" is information that is relevant but should not be included in the answer since the ground truth answer does not contain it." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ragas.NoiseSensitivity\",\n", - " inputs={\"dataset\": vm_test_ds},\n", - " params={\n", - " \"user_input_column\": \"question\",\n", - " \"response_column\": \"rag_model_prediction\",\n", - " \"reference_column\": \"ground_truth\",\n", - " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Generation quality\n", - "\n", - "In this section, we evaluate the alignment and relevance of generated responses to reference outputs within our retrieval-augmented generation (RAG) application. We use metrics that assess various quality dimensions of the generated responses, including semantic similarity, structural alignment, and phrasing overlap. Semantic similarity metrics compare embeddings of generated and reference text to capture deeper contextual alignment, while overlap and alignment measures quantify how well the phrasing and structure of generated responses match the intended outputs." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Token Disparity\n", - "\n", - "This test assesses the difference in token counts between the reference texts (ground truth) and the answers generated by the RAG model. It helps evaluate how well the model's outputs align with the expected length and level of detail in the reference texts. A significant disparity in token counts could signal issues with generation quality, such as excessive verbosity or insufficient detail. Consistently low token counts in generated answers compared to references might suggest that the model’s outputs are incomplete or overly concise, missing important contextual information." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.TokenDisparity\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### ROUGE Score\n", - "\n", - "This test evaluates the quality of answers generated by the RAG model by measuring overlaps in n-grams, word sequences, and word pairs between the model output and the reference (ground truth) text. ROUGE, short for Recall-Oriented Understudy for Gisting Evaluation, assesses both precision and recall, providing a balanced view of how well the generated response captures the reference content. ROUGE precision measures the proportion of n-grams in the generated text that match the reference, highlighting relevance and conciseness, while ROUGE recall assesses the proportion of reference n-grams present in the generated text, indicating completeness and thoroughness. \n", - "\n", - "Low precision scores might reveal that the generated text includes redundant or irrelevant information, while low recall scores suggest omissions of essential details from the reference. Consistently low ROUGE scores could indicate poor overall alignment with the ground truth, suggesting the model may be missing key content or failing to capture the intended meaning." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.RougeScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - " params={\n", - " \"metric\": \"rouge-1\",\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### BLEU Score\n", - "\n", - "The BLEU Score test evaluates the quality of answers generated by the RAG model by measuring n-gram overlap between the generated text and the reference (ground truth) text, with a specific focus on exact precision in phrasing. While ROUGE precision also assesses overlap, BLEU differs in two main ways: first, it applies a geometric average across multiple n-gram levels, capturing precise phrase alignment, and second, it includes a brevity penalty to prevent overly short outputs from inflating scores artificially. This added precision focus is valuable in RAG applications where strict adherence to reference language is essential, as BLEU emphasizes the match to exact phrasing. In contrast, ROUGE precision evaluates general content overlap without penalizing brevity, offering a broader sense of content alignment." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.BleuScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### BERT Score\n", - "\n", - "This test evaluates the quality of the RAG generated answers using BERT embeddings to measure precision, recall, and F1 scores based on semantic similarity, rather than exact n-gram matches as in BLEU and ROUGE. This approach captures contextual meaning, making it valuable when wording differs but the intended message closely aligns with the reference. In RAG applications, the BERT score is especially useful for ensuring that generated answers convey the reference text’s meaning, even if phrasing varies. Consistently low scores indicate a lack of semantic alignment, suggesting the model may miss or misrepresent key content. Low precision may reflect irrelevant or redundant details, while low recall can indicate omissions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.BertScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### METEOR Score\n", - "\n", - "This test evaluates the quality of the generated answers by measuring alignment with the ground truth, emphasizing both accuracy and fluency. Unlike BLEU and ROUGE, which focus on n-gram matches, METEOR combines precision, recall, synonym matching, and word order, focusing at how well the generated text conveys meaning and reads naturally. This metric is especially useful for RAG applications where sentence structure and natural flow are crucial for clear communication. Lower scores may suggest alignment issues, indicating that the answers may lack fluency or key content. Discrepancies in word order or high fragmentation penalties can reveal problems with how the model constructs sentences, potentially affecting readability." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.MeteorScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Bias and Toxicity\n", - "\n", - "In this section, we use metrics like Toxicity Score and Regard Score to evaluate both the generated responses and the ground truth. These tests helps us detect any harmful, offensive, or inappropriate language and evaluate the level of bias and neutrality enabling us to assess and mitigate potential biases in both the model's responses and the original dataset." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Toxicity Score\n", - "\n", - "This test measures the level of harmful or offensive content in the generated answers. The test uses a preloaded toxicity detection tool from Hugging Face, which identifies language that may be inappropriate, aggressive, or derogatory. High toxicity scores indicate potentially toxic content, while consistently elevated scores across multiple outputs may signal underlying issues in the model’s generation process that require attention to prevent the spread of harmful language." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.ToxicityScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Regard Score\n", - "\n", - "This test evaluates the sentiment and perceived regard—categorized as positive, negative, neutral, or other—in answers generated by the RAG model. This is important for identifying any biases or sentiment tendencies in responses, ensuring that generated answers are balanced and appropriate for the context. The uses a preloaded regard evaluation tool from Hugging Face to compute scores for each response. High skewness in regard scores, especially if the generated responses consistently diverge from expected sentiments in the reference texts, may reveal biases in the model’s generation, such as overly positive or negative tones where neutrality is expected." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "run_test(\n", - " \"validmind.model_validation.RegardScore\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " \"model\": vm_rag_model,\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Conclusion\n", - "\n", - "In this notebook, we have seen how we can use LangChain and ValidMind together to build, evaluate and document a simple RAG Model as its developed. This is a great example of the interactive development experience that ValidMind is designed to support. We can quickly iterate on our model and document as we go... We have seen how ValidMind supports non-traditional \"models\" using a functional interface and how we can build pipelines of many models to support complex GenAI workflows.\n", - "\n", - "This is still a work in progress and we are actively developing new tests to support more advanced GenAI workflows. We are also keeping an eye on the most popular GenAI models and libraries to explore direct integrations. Stay tuned for more updates and new features in this area!" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Upgrade ValidMind\n", - "\n", - "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", - "\n", - "Retrieve the information for the currently installed version of ValidMind:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip show validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", - "\n", - "```bash\n", - "%pip install --upgrade validmind\n", - "```" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You may need to restart your kernel after running the upgrade package for changes to be applied." - ] - }, - { - "cell_type": "markdown", - "id": "copyright-397fa35a68a34dc38f5d84d797fb5331", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "validmind-py3.10", - "language": "python", - "name": "validmind-py3.10" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.10.13" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/site/notebooks/how_to/dataset_image.png b/site/notebooks/how_to/data_and_datasets/dataset_image.png similarity index 100% rename from site/notebooks/how_to/dataset_image.png rename to site/notebooks/how_to/data_and_datasets/dataset_image.png diff --git a/site/notebooks/how_to/configure_dataset_features.ipynb b/site/notebooks/how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb similarity index 98% rename from site/notebooks/how_to/configure_dataset_features.ipynb rename to site/notebooks/how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb index 30d86a1ee0..86081ab36e 100644 --- a/site/notebooks/how_to/configure_dataset_features.ipynb +++ b/site/notebooks/how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb @@ -93,7 +93,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -158,8 +158,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/load_datasets_predictions.ipynb b/site/notebooks/how_to/data_and_datasets/dataset_inputs/load_datasets_predictions.ipynb similarity index 99% rename from site/notebooks/how_to/load_datasets_predictions.ipynb rename to site/notebooks/how_to/data_and_datasets/dataset_inputs/load_datasets_predictions.ipynb index a2dacdeea0..51f0dff6ef 100644 --- a/site/notebooks/how_to/load_datasets_predictions.ipynb +++ b/site/notebooks/how_to/data_and_datasets/dataset_inputs/load_datasets_predictions.ipynb @@ -113,7 +113,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -178,8 +178,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/model_image.png b/site/notebooks/how_to/data_and_datasets/model_image.png similarity index 100% rename from site/notebooks/how_to/model_image.png rename to site/notebooks/how_to/data_and_datasets/model_image.png diff --git a/site/notebooks/how_to/data_and_datasets/use_dataset_model_objects.ipynb b/site/notebooks/how_to/data_and_datasets/use_dataset_model_objects.ipynb new file mode 100644 index 0000000000..7ea025faa9 --- /dev/null +++ b/site/notebooks/how_to/data_and_datasets/use_dataset_model_objects.ipynb @@ -0,0 +1,1003 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Introduction to ValidMind Dataset and Model Objects\n", + "\n", + "When writing custom tests, it is essential to be aware of the interfaces of the ValidMind Dataset and ValidMind Model, which are used as input arguments.\n", + "\n", + "As a model developer, writing custom tests is beneficial when the ValidMind library lacks a built-in test for your specific needs. For example, a model might require new tests to evaluate specific aspects of the model or dataset based on a particular use case.\n", + "\n", + "This interactive notebook offers a detailed understanding of ValidMind objects and their use in writing custom tests. It introduces various interfaces provided by these objects and demonstrates how they can be leveraged to implement tests effortlessly." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About ValidMind](#toc1__) \n", + " - [Before you begin](#toc1_1__) \n", + " - [New to ValidMind?](#toc1_2__) \n", + " - [Key concepts](#toc1_3__) \n", + "- [Setting up](#toc2__) \n", + " - [Install the ValidMind Library](#toc2_1__) \n", + " - [Initialize the ValidMind Library](#toc2_2__) \n", + " - [Register sample model](#toc2_2_1__) \n", + " - [Apply documentation template](#toc2_2_2__) \n", + " - [Get your code snippet](#toc2_2_3__) \n", + "- [Load the demo dataset](#toc3__) \n", + " - [Prepocess the raw dataset](#toc3_1__) \n", + "- [Train a model for testing](#toc4__) \n", + "- [Explore basic components of the ValidMind library](#toc5__) \n", + " - [VMDataset Object](#toc5_1__) \n", + " - [Initialize the ValidMind datasets](#toc5_1_1__) \n", + " - [ Interfaces of the dataset object](#toc5_1_2__) \n", + " - [Using VM Dataset object as arguments in custom tests](#toc5_2__) \n", + " - [Run the test](#toc5_2_1__) \n", + " - [Using VM Dataset object and parameters as arguments in custom tests](#toc5_3__) \n", + " - [VMModel Object](#toc5_4__) \n", + " - [Initialize ValidMind model object](#toc5_5__) \n", + " - [Assign predictions to the datasets](#toc5_6__) \n", + " - [Using VM Model and Dataset objects as arguments in Custom tests](#toc5_7__) \n", + " - [Log the test results](#toc5_8__) \n", + "- [Where to go from here](#toc6__) \n", + " - [Use cases](#toc6_1__) \n", + " - [More how-to guides and code samples](#toc6_2__) \n", + " - [Discover more learning resources](#toc6_3__) \n", + "- [Upgrade ValidMind](#toc7__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models. You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language.\n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "Here, we will focus on ValidMind dataset, ValidMind model and tests to use these objects to generate artefacts for the documentation.\n", + "\n", + "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", + "\n", + "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", + "\n", + "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", + "\n", + "- **model**: A single ValidMind model object that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", + "- **dataset**: Single ValidMind dataset object that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", + "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", + "\n", + "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", + "\n", + "**Outputs**: Tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", + "\n", + "**Dataset based Test**\n", + "\n", + "![Dataset based test architecture](./dataset_image.png)\n", + "The dataset based tests take VM dataset object(s) as inputs, test configuration as test parameters to produce `Outputs` as mentioned above.\n", + "\n", + "**Model based Test**\n", + "\n", + "![Model based test architecture](./model_image.png)\n", + "Similar to datasest based tests, the model based tests as an additional input that is VM model object. It allows to identify prediction values of a specific model in the dataset object. " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "Please note the following recommended Python versions to use:\n", + "\n", + "- Python 3.7 > x <= 3.11\n", + "\n", + "To install the library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Binary classification`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "metadata": {} + }, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%matplotlib inline\n", + "\n", + "import xgboost as xgb" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Load the demo dataset" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.datasets.classification import customer_churn as demo_dataset\n", + "\n", + "raw_df = demo_dataset.load_data()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Prepocess the raw dataset" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "train_df, validation_df, test_df = demo_dataset.preprocess(raw_df)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Train a model for testing\n", + "\n", + "We train a simple customer churn model for our test." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "x_train = train_df.drop(demo_dataset.target_column, axis=1)\n", + "y_train = train_df[demo_dataset.target_column]\n", + "x_val = validation_df.drop(demo_dataset.target_column, axis=1)\n", + "y_val = validation_df[demo_dataset.target_column]\n", + "\n", + "model = xgb.XGBClassifier(early_stopping_rounds=10)\n", + "model.set_params(\n", + " eval_metric=[\"error\", \"logloss\", \"auc\"],\n", + ")\n", + "model.fit(\n", + " x_train,\n", + " y_train,\n", + " eval_set=[(x_val, y_val)],\n", + " verbose=False,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Explore basic components of the ValidMind library\n", + "\n", + "In this section, you will learn about the basic objects of the ValidMind library that are necessary to implement both custom and built-in tests. As explained above, these objects are:\n", + "* VMDataset: [The high level APIs can be found here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset)\n", + "* VMModel: [The high level APIs can be found here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMModel)\n", + "\n", + "Let's understand these objects and their interfaces step by step: " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### VMDataset Object" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Initialize the ValidMind datasets\n", + "\n", + "You can initialize a ValidMind dataset object using the [`init_dataset`](https://docs.validmind.ai/validmind/validmind.html#init_dataset) function from the ValidMind (`vm`) module.\n", + "\n", + "The function wraps the dataset to create a ValidMind `Dataset` object so that you can write tests effectively using the common interface provided by the VM objects. This step is always necessary every time you want to connect a dataset to documentation and produce test results through ValidMind. You only need to do it one time per dataset.\n", + "\n", + "This function takes a number of arguments. Some of the arguments are:\n", + "\n", + "- `dataset` — the raw dataset that you want to provide as input to tests\n", + "- `input_id` - a unique identifier that allows tracking what inputs are used when running each individual test\n", + "- `target_column` — a required argument if tests require access to true values. This is the name of the target column in the dataset\n", + "\n", + "The detailed list of the arguments can be found [here](https://docs.validmind.ai/validmind/validmind.html#init_dataset) " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# vm_raw_dataset is now a VMDataset object that you can pass to any ValidMind test\n", + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=raw_df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=\"Exited\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Once you have a ValidMind dataset object (VMDataset), you can inspect its attributes and methods using the inspect_obj utility module. This module provides a list of available attributes and interfaces for use in tests. Understanding how to use VMDatasets is crucial for comprehending how a custom test functions." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.utils import inspect_obj\n", + "inspect_obj(vm_raw_dataset)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Interfaces of the dataset object\n", + "\n", + "**DataFrame**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.df" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Feature columns**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.feature_columns" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Target column**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.target_column" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Features values**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.x_df()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Target value**" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.y_df()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Numeric feature columns** " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.feature_columns_numeric" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Categorical feature columns** " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset.feature_columns_categorical" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Similarly, you can use all other interfaces of the [VMDataset objects](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset) " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Using VM Dataset object as arguments in custom tests\n", + "\n", + "A custom test is simply a Python function that takes two types of arguments: `inputs` and `params`. The `inputs` are ValidMind objects (`VMDataset`, `VMModel`), and the `params` are additional parameters required for the underlying computation of the test. We will discuss both types of arguments in the following sections.\n", + "\n", + "Let's start with a custom test that requires only a ValidMind dataset object. In this example, we will check the balance of classes in the target column of the dataset:\n", + "\n", + "- The custom test below requires a single argument of type `VMDataset` (dataset).\n", + "- The `my_custom_tests.ClassImbalance` is a unique test identifier that can be assigned using the `vm.test` decorator functionality. This unique test ID will be used in the platform to load test results in the documentation.\n", + "- The `dataset.target_column` and `dataset.df` attributes of the `VMDataset` object are used in the test.\n", + "\n", + "Other high-level APIs (attributes and methods) of the dataset object are listed [here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset).\n", + "\n", + "If you've gone through the [Implement custom tests notebook](../tests/custom_tests/implement_custom_tests.ipynb), you should have a good understanding of how custom tests are implemented in details. If you haven't, we recommend going through that notebook first." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.vm_models.dataset.dataset import VMDataset\n", + "import pandas as pd\n", + "\n", + "@vm.test(\"my_custom_tests.ClassImbalance\")\n", + "def class_imbalance(dataset):\n", + " # Can only run this test if we have a Dataset object\n", + " if not isinstance(dataset, VMDataset):\n", + " raise ValueError(\"ClassImbalance requires a validmind Dataset object\")\n", + "\n", + " if dataset.target_column is None:\n", + " print(\"Skipping class_imbalance test because no target column is defined\")\n", + " return\n", + "\n", + " # VMDataset object provides target_column attribute\n", + " target_column = dataset.target_column\n", + " # we can access pandas DataFrame using df attribute\n", + " imbalance_percentages = dataset.df[target_column].value_counts(\n", + " normalize=True\n", + " )\n", + " classes = list(imbalance_percentages.index) \n", + " percentages = list(imbalance_percentages.values * 100)\n", + "\n", + " return pd.DataFrame({\"Classes\":classes, \"Percentage\": percentages})" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Run the test\n", + "\n", + "Let's run the test using the `run_test` method, which is part of the `validmind.tests` module. Here, we pass the `dataset` through the `inputs`. Similarly, you can pass `datasets`, `model`, or `models` as inputs if your custom test requires them. In this example below, we run the custom test `my_custom_tests.ClassImbalance` by passing the `dataset` through the `inputs`. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "result = run_test(\n", + " test_id=\"my_custom_tests.ClassImbalance\",\n", + " inputs={\n", + " \"dataset\": vm_raw_dataset\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You can move custom tests into separate modules in a folder. It allows you to take one-off tests and move them into an organized structure that makes it easier to manage, maintain and share them. We have provided a seperate notebook with detailed explaination [here](../tests/custom_tests/integrate_external_test_providers.ipynb) " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Using VM Dataset object and parameters as arguments in custom tests\n", + "\n", + "Simlilar to `inputs`, you can pass `params` to a custom test by providing a dictionary of parameters to the `run_test()` function. The parameters will override any default parameters set in the custom test definition. Note that the `dataset` is still passed as `inputs`. \n", + "Let's modify the class imbalance test so that it provides flexibility to `normalize` the results." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.vm_models.dataset.dataset import VMDataset\n", + "import pandas as pd\n", + "\n", + "@vm.test(\"my_custom_tests.ClassImbalance\")\n", + "def class_imbalance(dataset, normalize=True):\n", + " # Can only run this test if we have a Dataset object\n", + " if not isinstance(dataset, VMDataset):\n", + " raise ValueError(\"ClassImbalance requires a validmind Dataset object\")\n", + "\n", + " if dataset.target_column is None:\n", + " print(\"Skipping class_imbalance test because no target column is defined\")\n", + " return\n", + "\n", + " # VMDataset object provides target_column attribute\n", + " target_column = dataset.target_column\n", + " # we can access pandas DataFrame using df attribute\n", + " imbalance_percentages = dataset.df[target_column].value_counts(\n", + " normalize=normalize\n", + " )\n", + " classes = list(imbalance_percentages.index) \n", + " if normalize: \n", + " result = pd.DataFrame({\"Classes\":classes, \"Percentage\": list(imbalance_percentages.values*100)})\n", + " else:\n", + " result = pd.DataFrame({\"Classes\":classes, \"Count\": list(imbalance_percentages.values)})\n", + " return result" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In this example, the `normalize` parameter is set to `False`, so the class counts will not be normalized. You can change the value to `True` if you want the counts to be normalized. The results of the test will reflect this flexibility, allowing for different outputs based on the parameter passed.\n", + "\n", + "Here, we have passed the `dataset` through the `inputs` and the `normalize` parameter using the `params`." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "result = run_test(\n", + " test_id = \"my_custom_tests.ClassImbalance\",\n", + " inputs={\"dataset\": vm_raw_dataset},\n", + " params={\"normalize\": True},\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### VMModel Object" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize ValidMind model object\n", + "\n", + "Similar to ValidMind `Dataset` object, you can initialize a ValidMind Model object using the [`init_model`](https://docs.validmind.ai/validmind/validmind.html#init_model) function from the ValidMind (`vm`) module.\n", + "\n", + "This function takes a number of arguments. Some of the arguments are:\n", + "\n", + "- `model` — the raw model that you want evaluate\n", + "- `input_id` - a unique identifier that allows tracking what inputs are used when running each individual test\n", + "\n", + "The detailed list of the arguments can be found [here](https://docs.validmind.ai/validmind/validmind.html#init_model) " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "vm_model = vm.init_model(\n", + " model=model,\n", + " input_id=\"xgb_model\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's inspect the methods and attributes of the model now:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "inspect_obj(vm_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign predictions to the datasets\n", + "\n", + "We can now use the `assign_predictions()` method from the `Dataset` object to link existing predictions to any model. If no prediction values are passed, the method will compute predictions automatically:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_train_ds = vm.init_dataset(\n", + " input_id=\"train_dataset\",\n", + " dataset=train_df,\n", + " type=\"generic\",\n", + " target_column=demo_dataset.target_column,\n", + ")\n", + "\n", + "vm_train_ds.assign_predictions(model=vm_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You can see below, the extra prediction column (`xgb_model_prediction`) for the model (`xgb_model`) has been added in the dataset." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "print(vm_train_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Using VM Model and Dataset objects as arguments in Custom tests\n", + "\n", + "We will now create a `@vm.test` wrapper that will allow you to create a reusable test. Note the following changes in the code below:\n", + "\n", + "- The function `confusion_matrix` takes two arguments `dataset` and `model`. This is a `VMDataset` and `VMModel` object respectively.\n", + " - `VMDataset` objects allow you to access the dataset's true (target) values by accessing the `.y` attribute.\n", + " - `VMDataset` objects allow you to access the predictions for a given model by accessing the `.y_pred()` method.\n", + "- The function docstring provides a description of what the test does. This will be displayed along with the result in this notebook as well as in the ValidMind Platform.\n", + "- The function body calculates the confusion matrix using the `sklearn.tests.confusion_matrix` function as we just did above.\n", + "- The function then returns the `ConfusionMatrixDisplay.figure_` object - this is important as the ValidMind Library expects the output of the custom test to be a plot or a table.\n", + "- The `@vm.test` decorator is doing the work of creating a wrapper around the function that will allow it to be run by the ValidMind Library. It also registers the test so it can be found by the ID `my_custom_tests.ConfusionMatrix` (see the section below on how test IDs work in ValidMind and why this format is important)\n", + "\n", + "Similarly, you can use the functinality provided by `VMDataset` and `VMModel` objects. You can refer our documentation page for all the avalialble APIs [here](https://docs.validmind.ai/validmind/validmind.html#init_dataset)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn import metrics\n", + "import matplotlib.pyplot as plt\n", + "@vm.test(\"my_custom_tests.ConfusionMatrix\")\n", + "def confusion_matrix(dataset, model):\n", + " \"\"\"The confusion matrix is a table that is often used to describe the performance of a classification model on a set of data for which the true values are known.\n", + "\n", + " The confusion matrix is a 2x2 table that contains 4 values:\n", + "\n", + " - True Positive (TP): the number of correct positive predictions\n", + " - True Negative (TN): the number of correct negative predictions\n", + " - False Positive (FP): the number of incorrect positive predictions\n", + " - False Negative (FN): the number of incorrect negative predictions\n", + "\n", + " The confusion matrix can be used to assess the holistic performance of a classification model by showing the accuracy, precision, recall, and F1 score of the model on a single figure.\n", + " \"\"\"\n", + " # we can retrieve traget value from dataset which is y attribute\n", + " y_true = dataset.y\n", + " # The prediction value of a specific model using y_pred method \n", + " y_pred = dataset.y_pred(model=model)\n", + "\n", + " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", + "\n", + " cm_display = metrics.ConfusionMatrixDisplay(\n", + " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", + " )\n", + " cm_display.plot()\n", + " plt.close()\n", + "\n", + " return cm_display.figure_ # return the figure object itself" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here, we run test using two inputs; `dataset` and `model`. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "result = run_test(\n", + " test_id = \"my_custom_tests.ConfusionMatrix\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " \"model\": vm_model,\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Log the test results\n", + "\n", + "You can log any test result to the ValidMind Platform with the `.log()` method of the result object. This will allow you to add the result to the documentation.\n", + "\n", + "You can now do the same for the confusion matrix results." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Where to go from here\n", + "\n", + "In this notebook you have learned the end-to-end process to document a model with the ValidMind Library, running through some very common scenarios in a typical model development setting:\n", + "\n", + "- Running out-of-the-box tests\n", + "- Documenting your model by adding evidence to model documentation\n", + "- Extending the capabilities of the ValidMind Library by implementing custom tests\n", + "- Ensuring that the documentation is complete by running all tests in the documentation template\n", + "\n", + "As a next step, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library allows you generate model documentation for any use case:\n", + "\n", + "\n", + "\n", + "### Use cases\n", + "\n", + "- [Document an application scorecard model](../../use_cases/credit_risk/application_scorecard_full_suite.ipynb)\n", + "- [Linear regression documentation demo](../../use_cases/regression/quickstart_regression_full_suite.ipynb)\n", + "- [LLM model documentation demo](../../use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb)\n", + "\n", + "\n", + "\n", + "### More how-to guides and code samples\n", + "\n", + "- [Explore available tests in detail](../tests/explore_tests/explore_tests.ipynb)\n", + "- [In-depth guide for implementing custom tests](../tests/custom_tests/implement_custom_tests.ipynb)\n", + "- [In-depth guide to external test providers](../tests/custom_tests/integrate_external_test_providers.ipynb)\n", + "- [Configuring dataset features](./dataset_inputs/configure_dataset_features.ipynb)\n", + "- [Introduction to unit and composite tests](../metrics/run_unit_metrics.ipynb)\n", + "\n", + "\n", + "\n", + "### Discover more learning resources\n", + "\n", + "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", + "\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", + "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Upgrade ValidMind\n", + "\n", + "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", + "\n", + "Retrieve the information for the currently installed version of ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", + "\n", + "```bash\n", + "%pip install --upgrade validmind\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-340a990e20194848af0efb0c965e219a", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.14" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/site/notebooks/images/add_metric_over_time_block.png b/site/notebooks/how_to/metrics/add_metric_over_time_block.png similarity index 100% rename from site/notebooks/images/add_metric_over_time_block.png rename to site/notebooks/how_to/metrics/add_metric_over_time_block.png diff --git a/site/notebooks/images/log_metric_accuracy.png b/site/notebooks/how_to/metrics/log_metric_accuracy.png similarity index 100% rename from site/notebooks/images/log_metric_accuracy.png rename to site/notebooks/how_to/metrics/log_metric_accuracy.png diff --git a/site/notebooks/images/log_metric_attention.png b/site/notebooks/how_to/metrics/log_metric_attention.png similarity index 100% rename from site/notebooks/images/log_metric_attention.png rename to site/notebooks/how_to/metrics/log_metric_attention.png diff --git a/site/notebooks/images/log_metric_auc_1.png b/site/notebooks/how_to/metrics/log_metric_auc_1.png similarity index 100% rename from site/notebooks/images/log_metric_auc_1.png rename to site/notebooks/how_to/metrics/log_metric_auc_1.png diff --git a/site/notebooks/images/log_metric_auc_2.png b/site/notebooks/how_to/metrics/log_metric_auc_2.png similarity index 100% rename from site/notebooks/images/log_metric_auc_2.png rename to site/notebooks/how_to/metrics/log_metric_auc_2.png diff --git a/site/notebooks/images/log_metric_auc_3.png b/site/notebooks/how_to/metrics/log_metric_auc_3.png similarity index 100% rename from site/notebooks/images/log_metric_auc_3.png rename to site/notebooks/how_to/metrics/log_metric_auc_3.png diff --git a/site/notebooks/images/log_metric_auc_4.png b/site/notebooks/how_to/metrics/log_metric_auc_4.png similarity index 100% rename from site/notebooks/images/log_metric_auc_4.png rename to site/notebooks/how_to/metrics/log_metric_auc_4.png diff --git a/site/notebooks/images/log_metric_f1.png b/site/notebooks/how_to/metrics/log_metric_f1.png similarity index 100% rename from site/notebooks/images/log_metric_f1.png rename to site/notebooks/how_to/metrics/log_metric_f1.png diff --git a/site/notebooks/images/log_metric_precision.png b/site/notebooks/how_to/metrics/log_metric_precision.png similarity index 100% rename from site/notebooks/images/log_metric_precision.png rename to site/notebooks/how_to/metrics/log_metric_precision.png diff --git a/site/notebooks/images/log_metric_recall.png b/site/notebooks/how_to/metrics/log_metric_recall.png similarity index 100% rename from site/notebooks/images/log_metric_recall.png rename to site/notebooks/how_to/metrics/log_metric_recall.png diff --git a/site/notebooks/images/log_metric_satisfactory.png b/site/notebooks/how_to/metrics/log_metric_satisfactory.png similarity index 100% rename from site/notebooks/images/log_metric_satisfactory.png rename to site/notebooks/how_to/metrics/log_metric_satisfactory.png diff --git a/site/notebooks/images/log_metric_satisfactory_2.png b/site/notebooks/how_to/metrics/log_metric_satisfactory_2.png similarity index 100% rename from site/notebooks/images/log_metric_satisfactory_2.png rename to site/notebooks/how_to/metrics/log_metric_satisfactory_2.png diff --git a/site/notebooks/how_to/log_metrics_over_time.ipynb b/site/notebooks/how_to/metrics/log_metrics_over_time.ipynb similarity index 97% rename from site/notebooks/how_to/log_metrics_over_time.ipynb rename to site/notebooks/how_to/metrics/log_metrics_over_time.ipynb index f3bbfbdfa7..4b5c66dc2a 100644 --- a/site/notebooks/how_to/log_metrics_over_time.ipynb +++ b/site/notebooks/how_to/metrics/log_metrics_over_time.ipynb @@ -108,7 +108,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -173,8 +173,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -547,8 +545,8 @@ "- In this example, since we've only logged a single data point, the visualization shows just one measurement.\n", "- As you continue logging metrics, the graph will populate with more points, enabling you to track trends and patterns.\n", "\n", - "![Metric Over Time block](../images/add_metric_over_time_block.png)\n", - "![AUC Score](../images/log_metric_auc_1.png)" + "![Metric Over Time block](./add_metric_over_time_block.png)\n", + "![AUC Score](./log_metric_auc_1.png)" ] }, { @@ -591,7 +589,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![AUC Score](../images/log_metric_auc_2.png)" + "![AUC Score](./log_metric_auc_2.png)" ] }, { @@ -616,7 +614,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![AUC Score](../images/log_metric_auc_3.png)" + "![AUC Score](./log_metric_auc_3.png)" ] }, { @@ -719,11 +717,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![AUC Score](../images/log_metric_auc_4.png)\n", - "![Accuracy Score](../images/log_metric_accuracy.png)\n", - "![Precision Score](../images/log_metric_precision.png)\n", - "![Recall Score](../images/log_metric_recall.png)\n", - "![F1 Score](../images/log_metric_f1.png)" + "![AUC Score](./log_metric_auc_4.png)\n", + "![Accuracy Score](./log_metric_accuracy.png)\n", + "![Precision Score](./log_metric_precision.png)\n", + "![Recall Score](./log_metric_recall.png)\n", + "![F1 Score](./log_metric_f1.png)" ] }, { @@ -769,7 +767,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![GINI Score](../images/log_metric_satisfactory.png)" + "![GINI Score](./log_metric_satisfactory.png)" ] }, { @@ -802,7 +800,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![GINI Score](../images/log_metric_attention.png)" + "![GINI Score](./log_metric_attention.png)" ] }, { @@ -842,7 +840,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "![GINI Score](../images/log_metric_satisfactory_2.png)" + "![GINI Score](./log_metric_satisfactory_2.png)" ] }, { diff --git a/site/notebooks/how_to/run_unit_metrics.ipynb b/site/notebooks/how_to/metrics/run_unit_metrics.ipynb similarity index 99% rename from site/notebooks/how_to/run_unit_metrics.ipynb rename to site/notebooks/how_to/metrics/run_unit_metrics.ipynb index 235c538a98..d9ffe11759 100644 --- a/site/notebooks/how_to/run_unit_metrics.ipynb +++ b/site/notebooks/how_to/metrics/run_unit_metrics.ipynb @@ -166,8 +166,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/assign_scores_complete_tutorial.ipynb b/site/notebooks/how_to/scoring/assign_scores_complete_tutorial.ipynb similarity index 99% rename from site/notebooks/how_to/assign_scores_complete_tutorial.ipynb rename to site/notebooks/how_to/scoring/assign_scores_complete_tutorial.ipynb index 998a78ace7..933da518ac 100644 --- a/site/notebooks/how_to/assign_scores_complete_tutorial.ipynb +++ b/site/notebooks/how_to/scoring/assign_scores_complete_tutorial.ipynb @@ -168,8 +168,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/images/external-data-custom-test.png b/site/notebooks/how_to/tests/custom_tests/external-data-custom-test.png similarity index 100% rename from site/notebooks/images/external-data-custom-test.png rename to site/notebooks/how_to/tests/custom_tests/external-data-custom-test.png diff --git a/site/notebooks/images/hyperparameters-custom-metric.png b/site/notebooks/how_to/tests/custom_tests/hyperparameters-custom-metric.png similarity index 100% rename from site/notebooks/images/hyperparameters-custom-metric.png rename to site/notebooks/how_to/tests/custom_tests/hyperparameters-custom-metric.png diff --git a/site/notebooks/images/image-in-custom-metric.png b/site/notebooks/how_to/tests/custom_tests/image-in-custom-metric.png similarity index 100% rename from site/notebooks/images/image-in-custom-metric.png rename to site/notebooks/how_to/tests/custom_tests/image-in-custom-metric.png diff --git a/site/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb b/site/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb new file mode 100644 index 0000000000..638033b4e5 --- /dev/null +++ b/site/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb @@ -0,0 +1,1093 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Implement custom tests\n", + "\n", + "Custom tests extend the functionality of ValidMind, allowing you to document any model or use case with added flexibility.\n", + "\n", + "ValidMind provides a comprehensive set of tests out-of-the-box to evaluate and document your models and datasets. We recognize there will be cases where the default tests do not support a model or dataset, or specific documentation is needed. In these cases, you can create and use your own custom code to accomplish what you need. To streamline custom code integration, we support the creation of custom test functions.\n", + "\n", + "This interactive notebook provides a step-by-step guide for implementing and registering custom tests with ValidMind, running them individually, viewing the results on the ValidMind Platform, and incorporating them into your model documentation template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About ValidMind](#toc1__) \n", + " - [Before you begin](#toc1_1__) \n", + " - [New to ValidMind?](#toc1_2__) \n", + " - [Key concepts](#toc1_3__) \n", + "- [Setting up](#toc2__) \n", + " - [Install the ValidMind Library](#toc2_1__) \n", + " - [Initialize the ValidMind Library](#toc2_2__) \n", + " - [Register sample model](#toc2_2_1__) \n", + " - [Apply documentation template](#toc2_2_2__) \n", + " - [Get your code snippet](#toc2_2_3__) \n", + "- [Implement a Custom Test](#toc3__) \n", + "- [Run the Custom Test](#toc4__) \n", + " - [Setup the Model and Dataset](#toc4_1__) \n", + " - [Run the Custom Test](#toc4_2__) \n", + "- [Adding Custom Test to Model Documentation](#toc5__) \n", + "- [Some More Custom Tests](#toc6__) \n", + " - [Custom Test: Table of Model Hyperparameters](#toc6_1__) \n", + " - [Custom Test: External API Call](#toc6_2__) \n", + " - [Custom Test: Passing Parameters](#toc6_3__) \n", + " - [Custom Test: Multiple Tables and Plots in a Single Test](#toc6_4__) \n", + " - [Custom Test: Images](#toc6_5__) \n", + " - [Custom Test: Description](#toc6_6__) \n", + "- [Conclusion](#toc7__) \n", + "- [Next steps](#toc8__) \n", + " - [Work with your model documentation](#toc8_1__) \n", + " - [Discover more learning resources](#toc8_2__) \n", + "- [Upgrade ValidMind](#toc9__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", + "\n", + "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", + "\n", + "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", + "\n", + "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", + "\n", + "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", + "\n", + " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", + " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", + " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", + "\n", + "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", + "\n", + "**Outputs**: Custom test can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", + "\n", + "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", + "\n", + "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "To install the library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Binary classification`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Implement a Custom Test\n", + "\n", + "Let's start off by creating a simple custom test that creates a Confusion Matrix for a binary classification model. We will use the `sklearn.metrics.confusion_matrix` function to calculate the confusion matrix and then display it as a heatmap using `plotly`. (This is already a built-in test in ValidMind, but we will use it as an example to demonstrate how to create custom tests.)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import matplotlib.pyplot as plt\n", + "from sklearn import metrics\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ConfusionMatrix\")\n", + "def confusion_matrix(dataset, model):\n", + " \"\"\"The confusion matrix is a table that is often used to describe the performance of a classification model on a set of data for which the true values are known.\n", + "\n", + " The confusion matrix is a 2x2 table that contains 4 values:\n", + "\n", + " - True Positive (TP): the number of correct positive predictions\n", + " - True Negative (TN): the number of correct negative predictions\n", + " - False Positive (FP): the number of incorrect positive predictions\n", + " - False Negative (FN): the number of incorrect negative predictions\n", + "\n", + " The confusion matrix can be used to assess the holistic performance of a classification model by showing the accuracy, precision, recall, and F1 score of the model on a single figure.\n", + " \"\"\"\n", + " y_true = dataset.y\n", + " y_pred = dataset.y_pred(model)\n", + "\n", + " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", + "\n", + " cm_display = metrics.ConfusionMatrixDisplay(\n", + " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", + " )\n", + " cm_display.plot()\n", + "\n", + " plt.close() # close the plot to avoid displaying it\n", + "\n", + " return cm_display.figure_ # return the figure object itself" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Thats our custom test defined and ready to go... Let's take a look at whats going on here:\n", + "\n", + "- The function `confusion_matrix` takes two arguments `dataset` and `model`. This is a VMDataset and VMModel object respectively.\n", + "- The function docstring provides a description of what the test does. This will be displayed along with the result in this notebook as well as in the ValidMind Platform.\n", + "- The function body calculates the confusion matrix using the `sklearn.metrics.confusion_matrix` function and then plots it using `sklearn.metric.ConfusionMatrixDisplay`.\n", + "- The function then returns the `ConfusionMatrixDisplay.figure_` object - this is important as the ValidMind Library expects the output of the custom test to be a plot or a table.\n", + "- The `@vm.test` decorator is doing the work of creating a wrapper around the function that will allow it to be run by the ValidMind Library. It also registers the test so it can be found by the ID `my_custom_tests.ConfusionMatrix` (see the section below on how test IDs work in ValidMind and why this format is important)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Run the Custom Test\n", + "\n", + "Now that we have defined and registered our custom test, lets see how we can run it and properly use it in the ValidMind Platform." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Setup the Model and Dataset\n", + "\n", + "First let's setup a an example model and dataset to run our custom metic against. Since this is a Confusion Matrix, we will use the Customer Churn dataset that ValidMind provides and train a simple XGBoost model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "from validmind.datasets.classification import customer_churn\n", + "\n", + "raw_df = customer_churn.load_data()\n", + "train_df, validation_df, test_df = customer_churn.preprocess(raw_df)\n", + "\n", + "x_train = train_df.drop(customer_churn.target_column, axis=1)\n", + "y_train = train_df[customer_churn.target_column]\n", + "x_val = validation_df.drop(customer_churn.target_column, axis=1)\n", + "y_val = validation_df[customer_churn.target_column]\n", + "\n", + "model = xgb.XGBClassifier(early_stopping_rounds=10)\n", + "model.set_params(\n", + " eval_metric=[\"error\", \"logloss\", \"auc\"],\n", + ")\n", + "model.fit(\n", + " x_train,\n", + " y_train,\n", + " eval_set=[(x_val, y_val)],\n", + " verbose=False,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Easy enough! Now we have a model and dataset setup and trained. One last thing to do is bring the dataset and model into the ValidMind Library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# for now, we'll just use the test dataset\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " target_column=customer_churn.target_column,\n", + " input_id=\"test_dataset\",\n", + ")\n", + "\n", + "vm_model = vm.init_model(model, input_id=\"model\")\n", + "\n", + "# link the model to the dataset\n", + "vm_test_ds.assign_predictions(model=vm_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Run the Custom Test\n", + "\n", + "Now that we have our model and dataset setup, we have everything we need to run our custom test. We can do this by importing the `run_test` function from the `validmind.tests` module and passing in the test ID of our custom test along with the model and dataset we want to run it against.\n", + "\n", + ">Notice how the `inputs` dictionary is used to map an `input_id` which we set above to the `model` and `dataset` keys that are expected by our custom test function. This is how the ValidMind Library knows which inputs to pass to different tests and is key when using many different datasets and models." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "\n", + "result = run_test(\n", + " \"my_custom_tests.ConfusionMatrix\",\n", + " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You'll notice that the docstring becomes a markdown description of the test. The figure is then displayed as the test result. What you see above is how it will look in the ValidMind Platform as well. Let's go ahead and log the result to see how that works." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Adding Custom Test to Model Documentation\n", + "\n", + "To do this, go to the documentation page of the model you registered above and navigate to the `Model Development` -> `Model Evaluation` section. Then hover between any existing content block to reveal the `+` button as shown in the screenshot below.\n", + "\n", + "![screenshot showing insert button for test-driven blocks](./insert-test-driven-block.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now click on the `+` button and select the `Test-Driven Block` option. This will open a dialog where you can select `My Custom Tests Confusion Matrix` from the list of available tests. You can preview the result and then click `Insert Block` to add it to the documentation.\n", + "\n", + "![screenshot showing how to insert a test-driven block](./insert-test-driven-block-custom.png)\n", + "\n", + "The test should match the result you see above. It is now part of your documentation and will now be run everytime you run `vm.run_documentation_tests()` for your model. Let's do that now." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.reload()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If you preview the template, it should show the custom test in the `Model Development`->`Model Evaluation` section:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Just so we can run all of the tests in the template, let's initialize the train and raw dataset.\n", + "\n", + "(Refer to [**Quickstart for model documentation**](../../../quickstart/quickstart_model_documentation.ipynb) and the ValidMind docs for more information on what we are doing here)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=raw_df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=customer_churn.target_column,\n", + " class_labels=customer_churn.class_labels,\n", + ")\n", + "\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=customer_churn.target_column,\n", + ")\n", + "vm_train_ds.assign_predictions(model=vm_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To run all the tests in the template, you can use the `vm.run_documentation_tests()` and pass the inputs we initialized above and the demo config from our customer_churn module. We will have to add a section to the config for our new test to tell it which inputs it should receive. This is done by simply adding a new element in the config dictionary where the key is the ID of the test and the value is a dictionary with the following structure:\n", + "```python\n", + "{\n", + " \"inputs\": {\n", + " \"model\": \"test_dataset\",\n", + " \"dataset\": \"model\",\n", + " }\n", + "}\n", + "```" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.utils import preview_test_config\n", + "\n", + "test_config = customer_churn.get_demo_test_config()\n", + "test_config[\"my_custom_tests.ConfusionMatrix\"] = {\n", + " \"inputs\": {\n", + " \"dataset\": \"test_dataset\",\n", + " \"model\": \"model\",\n", + " }\n", + "}\n", + "preview_test_config(test_config)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "full_suite = vm.run_documentation_tests(config=test_config)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Some More Custom Tests\n", + "\n", + "Now that you understand the entire process of creating custom tests and using them in your documentation, let's create a few more to see different ways you can utilize custom tests." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: Table of Model Hyperparameters\n", + "\n", + "This custom test will display a table of the hyperparameters used in the model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_custom_tests.Hyperparameters\")\n", + "def hyperparameters(model):\n", + " \"\"\"The hyperparameters of a machine learning model are the settings that control the learning process.\n", + " These settings are specified before the learning process begins and can have a significant impact on the\n", + " performance of the model.\n", + "\n", + " The hyperparameters of a model can be used to tune the model to achieve the best possible performance\n", + " on a given dataset. By examining the hyperparameters of a model, you can gain insight into how the model\n", + " was trained and how it might be improved.\n", + " \"\"\"\n", + " hyperparameters = model.model.get_xgb_params() # dictionary of hyperparameters\n", + "\n", + " # turn the dictionary into a table where each row contains a hyperparameter and its value\n", + " return [{\"Hyperparam\": k, \"Value\": v} for k, v in hyperparameters.items() if v]\n", + "\n", + "\n", + "result = run_test(\"my_custom_tests.Hyperparameters\", inputs={\"model\": \"model\"})\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Since the test has been run and logged, you can add it to your documentation using the same process as above. It should look like this:\n", + "\n", + "![screenshot showing hyperparameters test](./hyperparameters-custom-metric.png)\n", + "\n", + "For our simple toy model, there are aren't really any proper hyperparameters but you can see how this could be useful for more complex models that have gone through hyperparameter tuning." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: External API Call\n", + "\n", + "This custom test will make an external API call to get the current BTC price and display it as a table. This demonstrates how you might integrate external data sources into your model documentation in a programmatic way. You could, for instance, setup a pipeline that runs a test like this every day to keep your model documentation in sync with an external system." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import requests\n", + "import random\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ExternalAPI\")\n", + "def external_api():\n", + " \"\"\"This test calls an external API to get a list of fake users. It then creates\n", + " a table with the relevant data so it can be displayed in the documentation.\n", + "\n", + " The purpose of this test is to demonstrate how to call an external API and use the\n", + " data in a test. A test like this could even be setup to run in a scheduled\n", + " pipeline to keep your documentation in-sync with an external data source.\n", + " \"\"\"\n", + " url = \"https://jsonplaceholder.typicode.com/users\"\n", + " response = requests.get(url)\n", + " data = response.json()\n", + "\n", + " # extract the time and the current BTC price in USD\n", + " return {\n", + " \"Model Owners/Stakeholders\": [\n", + " {\n", + " \"Name\": user[\"name\"],\n", + " \"Role\": random.choice([\"Owner\", \"Stakeholder\"]),\n", + " \"Email\": user[\"email\"],\n", + " \"Phone\": user[\"phone\"],\n", + " \"Slack Handle\": f\"@{user['name'].lower().replace(' ', '.')}\",\n", + " }\n", + " for user in data[:3]\n", + " ],\n", + " \"Model Developers\": [\n", + " {\n", + " \"Name\": user[\"name\"],\n", + " \"Role\": \"Developer\",\n", + " \"Email\": user[\"email\"],\n", + " }\n", + " for user in data[3:7]\n", + " ],\n", + " \"Model Validators\": [\n", + " {\n", + " \"Name\": user[\"name\"],\n", + " \"Role\": \"Validator\",\n", + " \"Email\": user[\"email\"],\n", + " }\n", + " for user in data[7:]\n", + " ],\n", + " }\n", + "\n", + "\n", + "result = run_test(\"my_custom_tests.ExternalAPI\")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Again, you can add this to your documentation to see how it looks:\n", + "\n", + "![screenshot showing BTC price metric](./external-data-custom-test.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: Passing Parameters\n", + "\n", + "Custom test functions, as stated earlier, can take both inputs and params. When you define your function there is no need to distinguish between the two, the ValidMind Library will handle that for you. You simply need to add both to the function as arguments and the library will pass in the correct values.\n", + "\n", + "So for instance, if you wanted to parameterize the first custom test we created, the confusion matrix, you could do so like this:\n", + "\n", + "```python\n", + "def confusion_matrix(dataset: VMDataset, model: VMModel, my_param: str = \"Default Value\"):\n", + " pass\n", + "```\n", + "\n", + "And then when you run the test, you can pass in the parameter like this:\n", + "\n", + "```python\n", + "vm.run_test(\n", + " \"my_custom_tests.ConfusionMatrix\",\n", + " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", + " params={\"my_param\": \"My Value\"},\n", + ")\n", + "```\n", + "\n", + "Or if you are running the entire documentation template, you would update the config like this:\n", + "\n", + "```python\n", + "test_config[\"my_custom_tests.ConfusionMatrix\"] = {\n", + " \"inputs\": {\n", + " \"dataset\": \"test_dataset\",\n", + " \"model\": \"model\",\n", + " },\n", + " \"params\": {\n", + " \"my_param\": \"My Value\",\n", + " },\n", + "}\n", + "```\n", + "\n", + "Let's go ahead and create a toy test that takes a parameter and uses it in the result:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import plotly.express as px\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ParameterExample\")\n", + "def parameter_example(\n", + " plot_title=\"Default Plot Title\", x_col=\"sepal_width\", y_col=\"sepal_length\"\n", + "):\n", + " \"\"\"This test takes two parameters and creates a scatter plot based on them.\n", + "\n", + " The purpose of this test is to demonstrate how to create a test that takes\n", + " parameters and uses them to generate a plot. This can be useful for creating\n", + " tests that are more flexible and can be used in a variety of scenarios.\n", + " \"\"\"\n", + " # return px.scatter(px.data.iris(), x=x_col, y=y_col, color=\"species\")\n", + " return px.scatter(\n", + " px.data.iris(), x=x_col, y=y_col, color=\"species\", title=plot_title\n", + " )\n", + "\n", + "\n", + "result = run_test(\n", + " \"my_custom_tests.ParameterExample\",\n", + " params={\n", + " \"plot_title\": \"My Cool Plot\",\n", + " \"x_col\": \"sepal_width\",\n", + " \"y_col\": \"sepal_length\",\n", + " },\n", + ")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Play around with this and see how you can use parameters, default values and other features to make your custom tests more flexible and useful.\n", + "\n", + "Here's how this one looks in the documentation:\n", + "![screenshot showing parameterized test](./parameterized-custom-metric.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: Multiple Tables and Plots in a Single Test\n", + "\n", + "Custom test functions, as stated earlier, can return more than just one table or plot. In fact, any number of tables and plots can be returned. Let's see an example of this:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import plotly.express as px\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ComplexOutput\")\n", + "def complex_output():\n", + " \"\"\"This test demonstrates how to return many tables and figures in a single test\"\"\"\n", + " # create a couple tables\n", + " table = [{\"A\": 1, \"B\": 2}, {\"A\": 3, \"B\": 4}]\n", + " table2 = [{\"C\": 5, \"D\": 6}, {\"C\": 7, \"D\": 8}]\n", + "\n", + " # create a few figures showing some random data\n", + " fig1 = px.line(x=np.arange(10), y=np.random.rand(10), title=\"Random Line Plot\")\n", + " fig2 = px.bar(x=[\"A\", \"B\", \"C\"], y=np.random.rand(3), title=\"Random Bar Plot\")\n", + " fig3 = px.scatter(\n", + " x=np.random.rand(10), y=np.random.rand(10), title=\"Random Scatter Plot\"\n", + " )\n", + "\n", + " return (\n", + " {\n", + " \"My Cool Table\": table,\n", + " \"Another Table\": table2,\n", + " },\n", + " fig1,\n", + " fig2,\n", + " fig3,\n", + " )\n", + "\n", + "\n", + "result = run_test(\"my_custom_tests.ComplexOutput\")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Notice how you can return the tables as a dictionary where the key is the title of the table and the value is the table itself. You could also just return the tables by themselves but this way you can give them a title to more easily identify them in the result.\n", + "\n", + "![screenshot showing multiple tables and plots](./multiple-tables-plots-custom-metric.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: Images\n", + "\n", + "If you are using a plotting library that isn't supported by ValidMind (i.e. not `matplotlib` or `plotly`), you can still return the image directly as a bytes-like object. This could also be used to bring any type of image into your documentation in a programmatic way. For instance, you may want to include a diagram of your model architecture or a screenshot of a dashboard that your model is integrated with. As long as you can produce the image with Python or open it from a file, you can include it in your documentation." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import io\n", + "import matplotlib.pyplot as plt\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.Image\")\n", + "def image():\n", + " \"\"\"This test demonstrates how to return an image in a test\"\"\"\n", + "\n", + " # create a simple plot\n", + " fig, ax = plt.subplots()\n", + " ax.plot([1, 2, 3, 4])\n", + " ax.set_title(\"Simple Line Plot\")\n", + "\n", + " # save the plot as a PNG image (in-memory buffer)\n", + " img_data = io.BytesIO()\n", + " fig.savefig(img_data, format=\"png\")\n", + " img_data.seek(0)\n", + "\n", + " plt.close() # close the plot to avoid displaying it\n", + "\n", + " return img_data.read()\n", + "\n", + "\n", + "result = run_test(\"my_custom_tests.Image\")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Adding this custom test to your documentation will display the image:\n", + "\n", + "![screenshot showing image custom test](./image-in-custom-metric.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If you want to log an image as a test result, you can do so by passing the path to the image as a parameter to the custom test and then opening the file in the test function. Here's an example:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_custom_tests.MyPNGCorrelationMatrix\")\n", + "def Image(path: str):\n", + " \"\"\"Opens a png image file and logs it as a test result to ValidMind\"\"\"\n", + " if not path.endswith(\".png\"):\n", + " raise ValueError(\"Image must be a PNG file\")\n", + "\n", + " # return raw image bytes\n", + " with open(path, \"rb\") as f:\n", + " return f.read()\n", + " \n", + "run_test(\n", + " \"my_custom_tests.MyPNGCorrelationMatrix\",\n", + " params={\"path\": \"./pearson-correlation-matrix.png\"},\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The image is displayed in the test result:\n", + "\n", + "![screenshot showing image from file](./pearson-correlation-matrix-test-output.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Custom Test: Description\n", + "\n", + "If you want to write a custom test description for your custom test instead of it is interpreted through llm, you can do so by returning string in your test." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "\n", + "@vm.test(\"my_custom_tests.MyCustomTest\")\n", + "def my_custom_test(dataset, model):\n", + " \"\"\"\n", + " This is a custom computed test that computes confusion matrix for a binary classification model and return a string as a test description.\n", + " \"\"\"\n", + " y_true = dataset.y\n", + " y_pred = dataset.y_pred(model)\n", + "\n", + " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", + "\n", + " cm_display = metrics.ConfusionMatrixDisplay(\n", + " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", + " )\n", + " cm_display.plot()\n", + "\n", + " plt.close() # close the plot to avoid displaying it\n", + "\n", + " return cm_display.figure_, \"Test Description - Confusion Matrix\", pd.DataFrame({\"Value\": [1, 2, 3]}) # return the figure object itself\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You can see here test result description has been customized here. The same result description will be displayed in the UI." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.MyCustomTest\",\n", + " inputs={\"model\": \"model\", \"dataset\": \"test_dataset\"},\n", + ")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Conclusion\n", + "\n", + "In this notebook, we have demonstrated how to create custom tests in ValidMind. We have shown how to define custom test functions, register them with the ValidMind Library, run them against models and datasets, and add them to model documentation templates. We have also shown how to return tables and plots from custom tests and how to use them in the ValidMind Platform. We hope this tutorial has been helpful in understanding how to create and use custom tests in ValidMind." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Next steps\n", + "\n", + "You can look at the results of this test suite right in the notebook where you ran the code, as you would expect. But there is a better way — use the ValidMind Platform to work with your model documentation.\n", + "\n", + "\n", + "\n", + "### Work with your model documentation\n", + "\n", + "1. From the **Model Inventory** in the ValidMind Platform, go to the model you registered earlier. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/working-with-model-inventory.html))\n", + "\n", + "2. Click and expand the **Model Development** section.\n", + "\n", + "What you see is the full draft of your model documentation in a more easily consumable version. From here, you can make qualitative edits to model documentation, view guidelines, collaborate with validators, and submit your model documentation for approval when it's ready. [Learn more ...](https://docs.validmind.ai/guide/model-documentation/working-with-model-documentation.html)\n", + "\n", + "\n", + "\n", + "### Discover more learning resources\n", + "\n", + "We offer many interactive notebooks to help you document models:\n", + "\n", + "- [Run tests & test suites](https://docs.validmind.ai/developer/model-testing/testing-overview.html)\n", + "- [Code samples](https://docs.validmind.ai/developer/samples-jupyter-notebooks.html)\n", + "\n", + "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Upgrade ValidMind\n", + "\n", + "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", + "\n", + "Retrieve the information for the currently installed version of ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", + "\n", + "```bash\n", + "%pip install --upgrade validmind\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-997b933948594ddd929ee9419957dfe3", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.5" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/site/notebooks/images/insert-test-driven-block-custom.png b/site/notebooks/how_to/tests/custom_tests/insert-test-driven-block-custom.png similarity index 100% rename from site/notebooks/images/insert-test-driven-block-custom.png rename to site/notebooks/how_to/tests/custom_tests/insert-test-driven-block-custom.png diff --git a/site/notebooks/images/insert-test-driven-block-test-provider.png b/site/notebooks/how_to/tests/custom_tests/insert-test-driven-block-test-provider.png similarity index 100% rename from site/notebooks/images/insert-test-driven-block-test-provider.png rename to site/notebooks/how_to/tests/custom_tests/insert-test-driven-block-test-provider.png diff --git a/site/notebooks/images/insert-test-driven-block.png b/site/notebooks/how_to/tests/custom_tests/insert-test-driven-block.png similarity index 100% rename from site/notebooks/images/insert-test-driven-block.png rename to site/notebooks/how_to/tests/custom_tests/insert-test-driven-block.png diff --git a/site/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb b/site/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb similarity index 99% rename from site/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb rename to site/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb index ca1403fb6c..fd84b39ca6 100644 --- a/site/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb +++ b/site/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb @@ -183,8 +183,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -758,7 +756,7 @@ "\n", "Now that the result has been logged to the ValidMind Platform, you can add it to your model documentation. This will add the result where you specify but it also will add the test to the template so it gets run anytime you `run_documentation_tests()`. To do this, go to the documentation page of the model you connected to above and navigate to the `Model Development` -> `Model Evaluation` section. Then hover between any existing content block to reveal the `+` button as shown in the screenshot below.\n", "\n", - "![screenshot showing insert button for test-driven blocks](../../images/insert-test-driven-block.png)" + "![screenshot showing insert button for test-driven blocks](./insert-test-driven-block.png)" ] }, { @@ -767,7 +765,7 @@ "source": [ "Now click on the `+` button and select the `Test-Driven Block` option. This will open a dialog where you can select `My Test Provider Confusion Matrix` from the list of available tests. You can preview the result and then click `Insert Block` to add it to the documentation.\n", "\n", - "![screenshot showing how to insert a test-driven block](../../images/insert-test-driven-block-test-provider.png)\n", + "![screenshot showing how to insert a test-driven block](./insert-test-driven-block-test-provider.png)\n", "\n", "The test should match the result you see above." ] diff --git a/site/notebooks/images/multiple-tables-plots-custom-metric.png b/site/notebooks/how_to/tests/custom_tests/multiple-tables-plots-custom-metric.png similarity index 100% rename from site/notebooks/images/multiple-tables-plots-custom-metric.png rename to site/notebooks/how_to/tests/custom_tests/multiple-tables-plots-custom-metric.png diff --git a/site/notebooks/images/parameterized-custom-metric.png b/site/notebooks/how_to/tests/custom_tests/parameterized-custom-metric.png similarity index 100% rename from site/notebooks/images/parameterized-custom-metric.png rename to site/notebooks/how_to/tests/custom_tests/parameterized-custom-metric.png diff --git a/site/notebooks/images/pearson-correlation-matrix-test-output.png b/site/notebooks/how_to/tests/custom_tests/pearson-correlation-matrix-test-output.png similarity index 100% rename from site/notebooks/images/pearson-correlation-matrix-test-output.png rename to site/notebooks/how_to/tests/custom_tests/pearson-correlation-matrix-test-output.png diff --git a/site/notebooks/images/pearson-correlation-matrix.png b/site/notebooks/how_to/tests/custom_tests/pearson-correlation-matrix.png similarity index 100% rename from site/notebooks/images/pearson-correlation-matrix.png rename to site/notebooks/how_to/tests/custom_tests/pearson-correlation-matrix.png diff --git a/site/notebooks/how_to/explore_test_suites.ipynb b/site/notebooks/how_to/tests/explore_tests/explore_test_suites.ipynb similarity index 99% rename from site/notebooks/how_to/explore_test_suites.ipynb rename to site/notebooks/how_to/tests/explore_tests/explore_test_suites.ipynb index f9f0a72861..a14d064637 100644 --- a/site/notebooks/how_to/explore_test_suites.ipynb +++ b/site/notebooks/how_to/tests/explore_tests/explore_test_suites.ipynb @@ -85,7 +85,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -672,7 +672,7 @@ "\n", "
Learn more about the individual tests available in the ValidMind Library\n", "

\n", - "Check out our Explore tests notebook for more code examples and usage of key functions.
\n", + "Check out our Explore tests notebook for more code examples and usage of key functions.\n", "\n", "\n", "\n", diff --git a/site/notebooks/how_to/explore_test_suites_output.png b/site/notebooks/how_to/tests/explore_tests/explore_test_suites_output.png similarity index 100% rename from site/notebooks/how_to/explore_test_suites_output.png rename to site/notebooks/how_to/tests/explore_tests/explore_test_suites_output.png diff --git a/site/notebooks/how_to/explore_tests.ipynb b/site/notebooks/how_to/tests/explore_tests/explore_tests.ipynb similarity index 99% rename from site/notebooks/how_to/explore_tests.ipynb rename to site/notebooks/how_to/tests/explore_tests/explore_tests.ipynb index 9fc1d3292e..ef5680e8fd 100644 --- a/site/notebooks/how_to/explore_tests.ipynb +++ b/site/notebooks/how_to/tests/explore_tests/explore_tests.ipynb @@ -86,7 +86,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -4371,7 +4371,7 @@ "\n", "
Learn about the tests suites available in the ValidMind Library.\n", "

\n", - "Check out our Explore test suites notebook for more code examples and usage of key functions.
\n", + "Check out our Explore test suites notebook for more code examples and usage of key functions.\n", "\n", "\n", "\n", diff --git a/site/notebooks/how_to/test_suite_output1.png b/site/notebooks/how_to/tests/explore_tests/test_suite_output1.png similarity index 100% rename from site/notebooks/how_to/test_suite_output1.png rename to site/notebooks/how_to/tests/explore_tests/test_suite_output1.png diff --git a/site/notebooks/how_to/test_suite_output2.png b/site/notebooks/how_to/tests/explore_tests/test_suite_output2.png similarity index 100% rename from site/notebooks/how_to/test_suite_output2.png rename to site/notebooks/how_to/tests/explore_tests/test_suite_output2.png diff --git a/site/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb b/site/notebooks/how_to/tests/run_tests/1_run_dataset_based_tests.ipynb similarity index 97% rename from site/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb rename to site/notebooks/how_to/tests/run_tests/1_run_dataset_based_tests.ipynb index a901849c4b..263ee130ab 100644 --- a/site/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb +++ b/site/notebooks/how_to/tests/run_tests/1_run_dataset_based_tests.ipynb @@ -14,7 +14,7 @@ "- Initialize a ValidMind dataset \n", "- Pass the dataset to the `run_test` fuction for any test that takes a `dataset` input\n", "\n", - "**We recommended that you first complete the [Explore tests](../explore_tests.ipynb) notebook,** to understand the basics of how to find and describe all the available tests in the ValidMind Library before moving on to this advanced guide.\n", + "**We recommended that you first complete the [Explore tests](../explore_tests/explore_tests.ipynb) notebook,** to understand the basics of how to find and describe all the available tests in the ValidMind Library before moving on to this advanced guide.\n", "\n", "This interactive notebook provides a step-by-step guide for listing and filtering available tests, building a sample dataset, initializing the required ValidMind objects, running the test, and then logging the results to ValidMind. " ] @@ -105,7 +105,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -170,8 +170,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -241,7 +239,7 @@ "\n", "Before we run a test, let's find a suitable test for this demonstration. Let's assume you want to generate the *pearson correlation matrix* for a dataset. A Pearson correlation matrix is a table that shows the [Pearson correlation coefficients](https://en.wikipedia.org/wiki/Pearson_correlation_coefficient) between several variables. \n", "\n", - "In the [Explore tests](../explore_tests.ipynb) notebook, we learned how to pass a `filter` to the `list_tests` function. We'll do the same here to find the test ID for the pearson correlation matrix:" + "In the [Explore tests](../explore_tests/explore_tests.ipynb) notebook, we learned how to pass a `filter` to the `list_tests` function. We'll do the same here to find the test ID for the pearson correlation matrix:" ] }, { @@ -584,7 +582,7 @@ ], "metadata": { "kernelspec": { - "display_name": "3.10.13", + "display_name": "Python 3", "language": "python", "name": "python3" }, diff --git a/site/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb b/site/notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb similarity index 97% rename from site/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb rename to site/notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb index 5aa8d78dba..4ce26f074b 100644 --- a/site/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb +++ b/site/notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb @@ -15,7 +15,7 @@ "- Initialize a ValidMind model and assign predictions to a dataset\n", "- Run a comparison test with `run_test` function\n", "\n", - "**We recommended that you first complete the [Explore tests](../explore_tests.ipynb) and the [Run dataset based tests](./1_run_dataset_based_tests.ipynb) notebooks** to understand the basics of how to find and describe all the available tests in the ValidMind Library and how to run tests before moving on to this guide.\n", + "**We recommended that you first complete the [Explore tests](../explore_tests/explore_tests.ipynb) and the [Run dataset based tests](./1_run_dataset_based_tests.ipynb) notebooks** to understand the basics of how to find and describe all the available tests in the ValidMind Library and how to run tests before moving on to this guide.\n", "\n", "This interactive notebook provides a step-by-step guide for listing and filtering available tests, building a sample dataset, training a model, initializing the required ValidMind objects, running a comparison test, and then logging the results to ValidMind. " ] @@ -113,7 +113,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -174,8 +174,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -267,7 +265,7 @@ "\n", "Before we run a comparison test, let's find a suitable test for this demonstration. Let's assume you want to evaluate the performance results for a model.\n", "\n", - "In the [Explore tests](../explore_tests.ipynb) notebook, we learned how to pass a `filter` to the `list_tests` function. We'll do the same here to find the test ID for the confusion matrix:" + "In the [Explore tests](../explore_tests/explore_tests.ipynb) notebook, we learned how to pass a `filter` to the `list_tests` function. We'll do the same here to find the test ID for the confusion matrix:" ] }, { diff --git a/site/notebooks/how_to/customize_test_result_descriptions.ipynb b/site/notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.ipynb similarity index 99% rename from site/notebooks/how_to/customize_test_result_descriptions.ipynb rename to site/notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.ipynb index c3aca95e16..ab36da9131 100644 --- a/site/notebooks/how_to/customize_test_result_descriptions.ipynb +++ b/site/notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.ipynb @@ -121,8 +121,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/enable_pii_detection.ipynb b/site/notebooks/how_to/tests/run_tests/configure_tests/enable_pii_detection.ipynb similarity index 99% rename from site/notebooks/how_to/enable_pii_detection.ipynb rename to site/notebooks/how_to/tests/run_tests/configure_tests/enable_pii_detection.ipynb index 3d648e82cf..2eb9ae0116 100644 --- a/site/notebooks/how_to/enable_pii_detection.ipynb +++ b/site/notebooks/how_to/tests/run_tests/configure_tests/enable_pii_detection.ipynb @@ -124,7 +124,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -289,7 +289,7 @@ "source": [ "
Want to learn more about custom tests?\n", "

\n", - "Check out our extended introduction to custom tests — Implement custom tests
" + "Check out our extended introduction to custom tests — Implement custom tests" ] }, { diff --git a/site/notebooks/how_to/filter_input_columns.ipynb b/site/notebooks/how_to/tests/run_tests/configure_tests/filter_input_columns.ipynb similarity index 100% rename from site/notebooks/how_to/filter_input_columns.ipynb rename to site/notebooks/how_to/tests/run_tests/configure_tests/filter_input_columns.ipynb diff --git a/site/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb b/site/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.ipynb similarity index 98% rename from site/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb rename to site/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.ipynb index d292c4461d..74e67f61b7 100644 --- a/site/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb +++ b/site/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.ipynb @@ -105,7 +105,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -170,8 +170,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/understand_utilize_rawdata.ipynb b/site/notebooks/how_to/tests/run_tests/configure_tests/understand_utilize_rawdata.ipynb similarity index 100% rename from site/notebooks/how_to/understand_utilize_rawdata.ipynb rename to site/notebooks/how_to/tests/run_tests/configure_tests/understand_utilize_rawdata.ipynb diff --git a/site/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb b/site/notebooks/how_to/tests/run_tests/documentation_tests/document_multiple_results_for_the_same_test.ipynb similarity index 99% rename from site/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb rename to site/notebooks/how_to/tests/run_tests/documentation_tests/document_multiple_results_for_the_same_test.ipynb index 74e04e1ab9..0feb618659 100644 --- a/site/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb +++ b/site/notebooks/how_to/tests/run_tests/documentation_tests/document_multiple_results_for_the_same_test.ipynb @@ -110,7 +110,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -175,8 +175,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/run_documentation_sections.ipynb b/site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_sections.ipynb similarity index 98% rename from site/notebooks/how_to/run_documentation_sections.ipynb rename to site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_sections.ipynb index fb4412df95..0952c0c9ce 100644 --- a/site/notebooks/how_to/run_documentation_sections.ipynb +++ b/site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_sections.ipynb @@ -103,7 +103,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -168,8 +168,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/run_documentation_tests_with_config.ipynb b/site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_tests_with_config.ipynb similarity index 99% rename from site/notebooks/how_to/run_documentation_tests_with_config.ipynb rename to site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_tests_with_config.ipynb index a96df1a5b1..548c020cd8 100644 --- a/site/notebooks/how_to/run_documentation_tests_with_config.ipynb +++ b/site/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_tests_with_config.ipynb @@ -107,7 +107,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -172,8 +172,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/how_to/use_dataset_model_objects.ipynb b/site/notebooks/how_to/use_dataset_model_objects.ipynb deleted file mode 100644 index 12431963c5..0000000000 --- a/site/notebooks/how_to/use_dataset_model_objects.ipynb +++ /dev/null @@ -1,1005 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Introduction to ValidMind Dataset and Model Objects\n", - "\n", - "When writing custom tests, it is essential to be aware of the interfaces of the ValidMind Dataset and ValidMind Model, which are used as input arguments.\n", - "\n", - "As a model developer, writing custom tests is beneficial when the ValidMind library lacks a built-in test for your specific needs. For example, a model might require new tests to evaluate specific aspects of the model or dataset based on a particular use case.\n", - "\n", - "This interactive notebook offers a detailed understanding of ValidMind objects and their use in writing custom tests. It introduces various interfaces provided by these objects and demonstrates how they can be leveraged to implement tests effortlessly." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [About ValidMind](#toc1__) \n", - " - [Before you begin](#toc1_1__) \n", - " - [New to ValidMind?](#toc1_2__) \n", - " - [Key concepts](#toc1_3__) \n", - "- [Setting up](#toc2__) \n", - " - [Install the ValidMind Library](#toc2_1__) \n", - " - [Initialize the ValidMind Library](#toc2_2__) \n", - " - [Register sample model](#toc2_2_1__) \n", - " - [Apply documentation template](#toc2_2_2__) \n", - " - [Get your code snippet](#toc2_2_3__) \n", - "- [Load the demo dataset](#toc3__) \n", - " - [Prepocess the raw dataset](#toc3_1__) \n", - "- [Train a model for testing](#toc4__) \n", - "- [Explore basic components of the ValidMind library](#toc5__) \n", - " - [VMDataset Object](#toc5_1__) \n", - " - [Initialize the ValidMind datasets](#toc5_1_1__) \n", - " - [ Interfaces of the dataset object](#toc5_1_2__) \n", - " - [Using VM Dataset object as arguments in custom tests](#toc5_2__) \n", - " - [Run the test](#toc5_2_1__) \n", - " - [Using VM Dataset object and parameters as arguments in custom tests](#toc5_3__) \n", - " - [VMModel Object](#toc5_4__) \n", - " - [Initialize ValidMind model object](#toc5_5__) \n", - " - [Assign predictions to the datasets](#toc5_6__) \n", - " - [Using VM Model and Dataset objects as arguments in Custom tests](#toc5_7__) \n", - " - [Log the test results](#toc5_8__) \n", - "- [Where to go from here](#toc6__) \n", - " - [Use cases](#toc6_1__) \n", - " - [More how-to guides and code samples](#toc6_2__) \n", - " - [Discover more learning resources](#toc6_3__) \n", - "- [Upgrade ValidMind](#toc7__) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## About ValidMind\n", - "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models. You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", - "\n", - "\n", - "\n", - "### Before you begin\n", - "\n", - "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language.\n", - "\n", - "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", - "\n", - "\n", - "\n", - "### New to ValidMind?\n", - "\n", - "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", - "\n", - "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", - "

\n", - "Register with ValidMind
\n", - "\n", - "\n", - "\n", - "### Key concepts\n", - "\n", - "Here, we will focus on ValidMind dataset, ValidMind model and tests to use these objects to generate artefacts for the documentation.\n", - "\n", - "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", - "\n", - "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", - "\n", - "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", - "\n", - "- **model**: A single ValidMind model object that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", - "- **dataset**: Single ValidMind dataset object that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", - "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", - "\n", - "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", - "\n", - "**Outputs**: Tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", - "\n", - "**Dataset based Test**\n", - "\n", - "![Dataset based test architecture](./dataset_image.png)\n", - "The dataset based tests take VM dataset object(s) as inputs, test configuration as test parameters to produce `Outputs` as mentioned above.\n", - "\n", - "**Model based Test**\n", - "\n", - "![Model based test architecture](./model_image.png)\n", - "Similar to datasest based tests, the model based tests as an additional input that is VM model object. It allows to identify prediction values of a specific model in the dataset object. " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Install the ValidMind Library\n", - "\n", - "Please note the following recommended Python versions to use:\n", - "\n", - "- Python 3.7 > x <= 3.11\n", - "\n", - "To install the library:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Register sample model\n", - "\n", - "Let's first register a sample model for use with this notebook:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", - "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", - "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", - "4. Select your own name under the **MODEL OWNER** drop-down.\n", - "\n", - "5. Click **Register Model** to add the model to your inventory." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Apply documentation template\n", - "\n", - "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", - "\n", - "2. Under **TEMPLATE**, select `Binary classification`.\n", - "\n", - "3. Click **Use Template** to apply the template." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Get your code snippet\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", - "\n", - "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", - "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Load your model identifier credentials from an `.env` file\n", - "\n", - "%load_ext dotenv\n", - "%dotenv .env\n", - "\n", - "# Or replace with your code snippet\n", - "\n", - "import validmind as vm\n", - "\n", - "vm.init(\n", - " # api_host=\"...\",\n", - " # api_key=\"...\",\n", - " # api_secret=\"...\",\n", - " # model=\"...\",\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%matplotlib inline\n", - "\n", - "import xgboost as xgb" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Load the demo dataset" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.datasets.classification import customer_churn as demo_dataset\n", - "\n", - "raw_df = demo_dataset.load_data()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Prepocess the raw dataset" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "train_df, validation_df, test_df = demo_dataset.preprocess(raw_df)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Train a model for testing\n", - "\n", - "We train a simple customer churn model for our test." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "x_train = train_df.drop(demo_dataset.target_column, axis=1)\n", - "y_train = train_df[demo_dataset.target_column]\n", - "x_val = validation_df.drop(demo_dataset.target_column, axis=1)\n", - "y_val = validation_df[demo_dataset.target_column]\n", - "\n", - "model = xgb.XGBClassifier(early_stopping_rounds=10)\n", - "model.set_params(\n", - " eval_metric=[\"error\", \"logloss\", \"auc\"],\n", - ")\n", - "model.fit(\n", - " x_train,\n", - " y_train,\n", - " eval_set=[(x_val, y_val)],\n", - " verbose=False,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Explore basic components of the ValidMind library\n", - "\n", - "In this section, you will learn about the basic objects of the ValidMind library that are necessary to implement both custom and built-in tests. As explained above, these objects are:\n", - "* VMDataset: [The high level APIs can be found here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset)\n", - "* VMModel: [The high level APIs can be found here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMModel)\n", - "\n", - "Let's understand these objects and their interfaces step by step: " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### VMDataset Object" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Initialize the ValidMind datasets\n", - "\n", - "You can initialize a ValidMind dataset object using the [`init_dataset`](https://docs.validmind.ai/validmind/validmind.html#init_dataset) function from the ValidMind (`vm`) module.\n", - "\n", - "The function wraps the dataset to create a ValidMind `Dataset` object so that you can write tests effectively using the common interface provided by the VM objects. This step is always necessary every time you want to connect a dataset to documentation and produce test results through ValidMind. You only need to do it one time per dataset.\n", - "\n", - "This function takes a number of arguments. Some of the arguments are:\n", - "\n", - "- `dataset` — the raw dataset that you want to provide as input to tests\n", - "- `input_id` - a unique identifier that allows tracking what inputs are used when running each individual test\n", - "- `target_column` — a required argument if tests require access to true values. This is the name of the target column in the dataset\n", - "\n", - "The detailed list of the arguments can be found [here](https://docs.validmind.ai/validmind/validmind.html#init_dataset) " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# vm_raw_dataset is now a VMDataset object that you can pass to any ValidMind test\n", - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=raw_df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=\"Exited\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Once you have a ValidMind dataset object (VMDataset), you can inspect its attributes and methods using the inspect_obj utility module. This module provides a list of available attributes and interfaces for use in tests. Understanding how to use VMDatasets is crucial for comprehending how a custom test functions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.utils import inspect_obj\n", - "inspect_obj(vm_raw_dataset)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Interfaces of the dataset object\n", - "\n", - "**DataFrame**" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.df" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Feature columns**" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.feature_columns" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Target column**" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.target_column" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Features values**" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.x_df()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Target value**" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.y_df()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Numeric feature columns** " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.feature_columns_numeric" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Categorical feature columns** " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset.feature_columns_categorical" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Similarly, you can use all other interfaces of the [VMDataset objects](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset) " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Using VM Dataset object as arguments in custom tests\n", - "\n", - "A custom test is simply a Python function that takes two types of arguments: `inputs` and `params`. The `inputs` are ValidMind objects (`VMDataset`, `VMModel`), and the `params` are additional parameters required for the underlying computation of the test. We will discuss both types of arguments in the following sections.\n", - "\n", - "Let's start with a custom test that requires only a ValidMind dataset object. In this example, we will check the balance of classes in the target column of the dataset:\n", - "\n", - "- The custom test below requires a single argument of type `VMDataset` (dataset).\n", - "- The `my_custom_tests.ClassImbalance` is a unique test identifier that can be assigned using the `vm.test` decorator functionality. This unique test ID will be used in the platform to load test results in the documentation.\n", - "- The `dataset.target_column` and `dataset.df` attributes of the `VMDataset` object are used in the test.\n", - "\n", - "Other high-level APIs (attributes and methods) of the dataset object are listed [here](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset).\n", - "\n", - "If you've gone through the [Implement custom tests notebook](../code_samples/custom_tests/implement_custom_tests.ipynb), you should have a good understanding of how custom tests are implemented in details. If you haven't, we recommend going through that notebook first." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.vm_models.dataset.dataset import VMDataset\n", - "import pandas as pd\n", - "\n", - "@vm.test(\"my_custom_tests.ClassImbalance\")\n", - "def class_imbalance(dataset):\n", - " # Can only run this test if we have a Dataset object\n", - " if not isinstance(dataset, VMDataset):\n", - " raise ValueError(\"ClassImbalance requires a validmind Dataset object\")\n", - "\n", - " if dataset.target_column is None:\n", - " print(\"Skipping class_imbalance test because no target column is defined\")\n", - " return\n", - "\n", - " # VMDataset object provides target_column attribute\n", - " target_column = dataset.target_column\n", - " # we can access pandas DataFrame using df attribute\n", - " imbalance_percentages = dataset.df[target_column].value_counts(\n", - " normalize=True\n", - " )\n", - " classes = list(imbalance_percentages.index) \n", - " percentages = list(imbalance_percentages.values * 100)\n", - "\n", - " return pd.DataFrame({\"Classes\":classes, \"Percentage\": percentages})" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Run the test\n", - "\n", - "Let's run the test using the `run_test` method, which is part of the `validmind.tests` module. Here, we pass the `dataset` through the `inputs`. Similarly, you can pass `datasets`, `model`, or `models` as inputs if your custom test requires them. In this example below, we run the custom test `my_custom_tests.ClassImbalance` by passing the `dataset` through the `inputs`. " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "result = run_test(\n", - " test_id=\"my_custom_tests.ClassImbalance\",\n", - " inputs={\n", - " \"dataset\": vm_raw_dataset\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can move custom tests into separate modules in a folder. It allows you to take one-off tests and move them into an organized structure that makes it easier to manage, maintain and share them. We have provided a seperate notebook with detailed explaination [here](../code_samples/custom_tests/integrate_external_test_providers.ipynb) " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Using VM Dataset object and parameters as arguments in custom tests\n", - "\n", - "Simlilar to `inputs`, you can pass `params` to a custom test by providing a dictionary of parameters to the `run_test()` function. The parameters will override any default parameters set in the custom test definition. Note that the `dataset` is still passed as `inputs`. \n", - "Let's modify the class imbalance test so that it provides flexibility to `normalize` the results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.vm_models.dataset.dataset import VMDataset\n", - "import pandas as pd\n", - "\n", - "@vm.test(\"my_custom_tests.ClassImbalance\")\n", - "def class_imbalance(dataset, normalize=True):\n", - " # Can only run this test if we have a Dataset object\n", - " if not isinstance(dataset, VMDataset):\n", - " raise ValueError(\"ClassImbalance requires a validmind Dataset object\")\n", - "\n", - " if dataset.target_column is None:\n", - " print(\"Skipping class_imbalance test because no target column is defined\")\n", - " return\n", - "\n", - " # VMDataset object provides target_column attribute\n", - " target_column = dataset.target_column\n", - " # we can access pandas DataFrame using df attribute\n", - " imbalance_percentages = dataset.df[target_column].value_counts(\n", - " normalize=normalize\n", - " )\n", - " classes = list(imbalance_percentages.index) \n", - " if normalize: \n", - " result = pd.DataFrame({\"Classes\":classes, \"Percentage\": list(imbalance_percentages.values*100)})\n", - " else:\n", - " result = pd.DataFrame({\"Classes\":classes, \"Count\": list(imbalance_percentages.values)})\n", - " return result" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "In this example, the `normalize` parameter is set to `False`, so the class counts will not be normalized. You can change the value to `True` if you want the counts to be normalized. The results of the test will reflect this flexibility, allowing for different outputs based on the parameter passed.\n", - "\n", - "Here, we have passed the `dataset` through the `inputs` and the `normalize` parameter using the `params`." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "result = run_test(\n", - " test_id = \"my_custom_tests.ClassImbalance\",\n", - " inputs={\"dataset\": vm_raw_dataset},\n", - " params={\"normalize\": True},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### VMModel Object" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize ValidMind model object\n", - "\n", - "Similar to ValidMind `Dataset` object, you can initialize a ValidMind Model object using the [`init_model`](https://docs.validmind.ai/validmind/validmind.html#init_model) function from the ValidMind (`vm`) module.\n", - "\n", - "This function takes a number of arguments. Some of the arguments are:\n", - "\n", - "- `model` — the raw model that you want evaluate\n", - "- `input_id` - a unique identifier that allows tracking what inputs are used when running each individual test\n", - "\n", - "The detailed list of the arguments can be found [here](https://docs.validmind.ai/validmind/validmind.html#init_model) " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "vm_model = vm.init_model(\n", - " model=model,\n", - " input_id=\"xgb_model\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's inspect the methods and attributes of the model now:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "inspect_obj(vm_model)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Assign predictions to the datasets\n", - "\n", - "We can now use the `assign_predictions()` method from the `Dataset` object to link existing predictions to any model. If no prediction values are passed, the method will compute predictions automatically:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_train_ds = vm.init_dataset(\n", - " input_id=\"train_dataset\",\n", - " dataset=train_df,\n", - " type=\"generic\",\n", - " target_column=demo_dataset.target_column,\n", - ")\n", - "\n", - "vm_train_ds.assign_predictions(model=vm_model)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can see below, the extra prediction column (`xgb_model_prediction`) for the model (`xgb_model`) has been added in the dataset." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "print(vm_train_ds)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Using VM Model and Dataset objects as arguments in Custom tests\n", - "\n", - "We will now create a `@vm.test` wrapper that will allow you to create a reusable test. Note the following changes in the code below:\n", - "\n", - "- The function `confusion_matrix` takes two arguments `dataset` and `model`. This is a `VMDataset` and `VMModel` object respectively.\n", - " - `VMDataset` objects allow you to access the dataset's true (target) values by accessing the `.y` attribute.\n", - " - `VMDataset` objects allow you to access the predictions for a given model by accessing the `.y_pred()` method.\n", - "- The function docstring provides a description of what the test does. This will be displayed along with the result in this notebook as well as in the ValidMind Platform.\n", - "- The function body calculates the confusion matrix using the `sklearn.tests.confusion_matrix` function as we just did above.\n", - "- The function then returns the `ConfusionMatrixDisplay.figure_` object - this is important as the ValidMind Library expects the output of the custom test to be a plot or a table.\n", - "- The `@vm.test` decorator is doing the work of creating a wrapper around the function that will allow it to be run by the ValidMind Library. It also registers the test so it can be found by the ID `my_custom_tests.ConfusionMatrix` (see the section below on how test IDs work in ValidMind and why this format is important)\n", - "\n", - "Similarly, you can use the functinality provided by `VMDataset` and `VMModel` objects. You can refer our documentation page for all the avalialble APIs [here](https://docs.validmind.ai/validmind/validmind.html#init_dataset)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from sklearn import metrics\n", - "import matplotlib.pyplot as plt\n", - "@vm.test(\"my_custom_tests.ConfusionMatrix\")\n", - "def confusion_matrix(dataset, model):\n", - " \"\"\"The confusion matrix is a table that is often used to describe the performance of a classification model on a set of data for which the true values are known.\n", - "\n", - " The confusion matrix is a 2x2 table that contains 4 values:\n", - "\n", - " - True Positive (TP): the number of correct positive predictions\n", - " - True Negative (TN): the number of correct negative predictions\n", - " - False Positive (FP): the number of incorrect positive predictions\n", - " - False Negative (FN): the number of incorrect negative predictions\n", - "\n", - " The confusion matrix can be used to assess the holistic performance of a classification model by showing the accuracy, precision, recall, and F1 score of the model on a single figure.\n", - " \"\"\"\n", - " # we can retrieve traget value from dataset which is y attribute\n", - " y_true = dataset.y\n", - " # The prediction value of a specific model using y_pred method \n", - " y_pred = dataset.y_pred(model=model)\n", - "\n", - " confusion_matrix = metrics.confusion_matrix(y_true, y_pred)\n", - "\n", - " cm_display = metrics.ConfusionMatrixDisplay(\n", - " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", - " )\n", - " cm_display.plot()\n", - " plt.close()\n", - "\n", - " return cm_display.figure_ # return the figure object itself" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Here, we run test using two inputs; `dataset` and `model`. " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.tests import run_test\n", - "result = run_test(\n", - " test_id = \"my_custom_tests.ConfusionMatrix\",\n", - " inputs={\n", - " \"dataset\": vm_train_ds,\n", - " \"model\": vm_model,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Log the test results\n", - "\n", - "You can log any test result to the ValidMind Platform with the `.log()` method of the result object. This will allow you to add the result to the documentation.\n", - "\n", - "You can now do the same for the confusion matrix results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Where to go from here\n", - "\n", - "In this notebook you have learned the end-to-end process to document a model with the ValidMind Library, running through some very common scenarios in a typical model development setting:\n", - "\n", - "- Running out-of-the-box tests\n", - "- Documenting your model by adding evidence to model documentation\n", - "- Extending the capabilities of the ValidMind Library by implementing custom tests\n", - "- Ensuring that the documentation is complete by running all tests in the documentation template\n", - "\n", - "As a next step, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library allows you generate model documentation for any use case:\n", - "\n", - "\n", - "\n", - "### Use cases\n", - "\n", - "- [Document an application scorecard model](../code_samples/credit_risk/application_scorecard_full_suite.ipynb)\n", - "- [Linear regression documentation demo](../code_samples/regression/quickstart_regression_full_suite.ipynb)\n", - "- [LLM model documentation demo](../code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb)\n", - "\n", - "\n", - "\n", - "### More how-to guides and code samples\n", - "\n", - "- [Explore available tests in detail](../how_to/explore_tests.ipynb)\n", - "- [In-depth guide for implementing custom tests](../code_samples/custom_tests/implement_custom_tests.ipynb)\n", - "- [In-depth guide to external test providers](../code_samples/custom_tests/integrate_external_test_providers.ipynb)\n", - "- [Configuring dataset features](../how_to/configure_dataset_features.ipynb)\n", - "- [Introduction to unit and composite tests](../how_to/run_unit_metrics.ipynb)\n", - "\n", - "\n", - "\n", - "### Discover more learning resources\n", - "\n", - "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", - "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", - "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Upgrade ValidMind\n", - "\n", - "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", - "\n", - "Retrieve the information for the currently installed version of ValidMind:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%pip show validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", - "\n", - "```bash\n", - "%pip install --upgrade validmind\n", - "```" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You may need to restart your kernel after running the upgrade package for changes to be applied." - ] - }, - { - "cell_type": "markdown", - "id": "copyright-340a990e20194848af0efb0c965e219a", - "metadata": {}, - "source": [ - "\n", - "\n", - "\n", - "\n", - "***\n", - "\n", - "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", - "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", - "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.10.14" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/site/notebooks/images/composite-metric-in-template-preview.png b/site/notebooks/images/composite-metric-in-template-preview.png deleted file mode 100644 index e948fb901bd6c92a2bb8540bc68fb6b286a70d40..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 152975 zcmeFZWn7hAw?0Y=NP|+6(w)*NAtfOx9U>sH2gQly|OO&P`{Xx<~*JER>*m8h6f=(bsY4j4G&>ZV=Sh%Uw%pL#eyCJlL2$H8fE1Re@SoxV#xE0wS81axY z z4>r^Nnj6iG6ZvtFbYfTgELM$KJD^@Qkq957f}xWNMKnpGlZjq~fSmLE33j~)n=Wq# zas_SFGS%G*#Lj@PDWP1b>G2%sFQl<_(1i^@3`GX%5-)XKHTmnl9eQ~fOCrlwO=I}o zjOD_ayn?NiZvt7tRa7JMRybxzd&9i&ZIWXco(45PCmx-QWbxe!X(LzL{I8YuK7^~@5C@_osVcT#ll zib%p4S2|FcCws8_CKUH5xCh$n3C%z_G7fnT9ijZ(?}_<>wz6F#)LDw@ zTfij2FQ!F@YLo*Moka_8X3`yS;%A4^^KB2XW+6<>$*zMb)&i zXXaw=g?LH1Lpj5!7etrF-n5u9nXni9x}xNtn;a)|NS${+0)!Dwl5d?NkWvHCQ+}l+eMq)d=N7*y1-K ztqfzPJKs|g*jqjC8QUzC--1DW<&Nw zc*&fE+V2Zdh^XhlH{N`6ARr(+98iRF1_z>e}$Dt3F7B1zMi2h zT`&=)Twl^S6uNlUgft%}uwZwgbwx85|%XX|bj-XOE8J(@C)W4C!oots!fn;cBH&1L)oHc4Z#|ZZ#ZPlY zvqHb{wj8yV$EmRvzP+t4~rjUTg+R?TWDIJ3WHem7cNRJ@xo7hB0Y0prl-kU$YgvG++-ig8pxC$ zxmu@8Zj8`b>%F>Tb;L*f%6?ol2bvf>^4*G!ETPuFHp2DOM2YN=%26{L3+uR7yCYk} zTayz%Tqjh9(@G5se3ZW^p^a7iFx_6+V4M6^<}AHFhTehy9sdenN$y8PJj#kpCEd}-p2EI^6@e9mB^$hrtx9@Ex=8{hvmhyOsWn?AhnNITeOZ6lr+J#THO@TiBf9qS#7 z5WT<*md))F4zD8GXD?wRWG8;kuXm=qsaH{Qx8zmH-gAu-pW-|{9ev{}nkg>ZN<03a z2DP$whPJ$RtahBgGF?kuk-lKNre>pO&pg3+q7~I>U1l~^%~@VnA0N=#(OaL|ez_OFrM>DJli3w&#%7V?%vgWfFytb!bu-d2 zPCIYj+GCt{mi8;{i>O!k^N8iBY5u*6gaYCon{b^3fmxApuPV2Ms4(1~ys7n%Wl5Yv zZeR9!bFq2D8 zPmWJ6P~LTA>)WNewz@`kqzqwpb{1=PdiVRVr2EeN@kAm*t~xtk1Y7G@9o?2omgx^m zFkWFWV?4sxC%h)qp?8g-67xN~2v>k91b)ta@yzM*(b-A>1|s2v%7%)g0<%g3_sFZI zldyHhP1U}n*S3{YF9uhT3*P+b{9#yNEch|y)u1GqNC`_xdjC@Ca?1krXyU*VYILP_ zWw~jzSv2a!)5N1~dGZqWBHrQ*LpShldr|R1@zdw~RkT%ZFL2lB)_$zBtmdv!61}=x z_0~?~eXer)D03y@{qV+IPh~@q8e?*nxwMd#r{8V*CIcmob_r~HIio{R=TPHNsdDL+ zMYs&j8y!3^v8G9tUW&3}v0d3Qh{uRjxy1JtL#rqUUk=Xkpe4od=ed;rntAK|WgT^3 zS}2`I!s}*VHd40e-5iI!!En>C!(!9hC_fBXPOWZ@u7S7n$MvkI;fGy;T@*^)9M4-< z^fAg-9v7HfKi^EqnaLscr8u^MmId{XCb+P<8MZj0I=*%2iii_ubt@g!)vDh*+&UP? zw%M>zoM@z{m(tXDqWX4UG4S{($TCZq9V;9^T%@6lqme zNq0%^%Y^pQ)f%gk&$VuDeS7_0SLb}=zb5SpCJQ)cG0zmw#OD(l_q%U~uS?KK(6(^& z$cu$#L^MbJ`Mz6iK)&1Vyi$C%B}(p8U-PR9I*cDeD-!vMyhJqD#}?|?clfpkP5-X` z9hFd(lhp5m(0UiA?OBawRYRL_`)xrAPuNUS8*knq&rog3v9Q}oQxCL*n1q;&v(d=b z%FSBh&GWc&c=u7WPaMs)xI7Fob#yz{L z*5qV_o!L8e#%jjwsQyeW2#$LQ6e6zm*t_E|E)@sXASJ147W_zob-E$de_`{_fH zg_qj+bbtMKyaQv?=pjE;bUNQQt4ju6319Pz<_AIl*!At3+t zIT8Xwh&2Mrzt2$t@9>`w;03?tU+>7#!3b#JF9Ps#&p`UGv(fc3kpJr#@sLm^)cm*gIQ6UayTL!ZRNX2l?mD2na;<@E4-OGr9wC|DV>Hy03MgDhr!K>^Mwc zK+G&S-0d9T_dyVK7Y2uR7OzdI-R*4corT@S9{zQPFgS*P%=wV|uTx&zh&|MOszxmh zak8N1=iuVtdMJLEnwnbF>BURoXEJjCz8w4|_R#9}YX@OYPB%9<4mVy7h?6BJw~&w! zCl?PV4-Y#ygWcJ~{P=de_zVO%`5nym;T3B|MyF^oGqNBA$H)Quf_kKuzz3tpI`p_LQzh5 z-~VGQ{x#5leF{ce{H`eHf6tou-7meU@C29KT1Hh9yaO|X{~+;zUrhgc2giu~56LSz zPY@6!5fo&kG~E$*X5S{f7#}Y`^h|;v9QtC?|f#uxaNG;va@c!=7okN zBI-T0|M`~{GuKy{@OU|n41c7%O#kz*7l?%HJLrFMCR2uFG7^JcHO8vhpPlX>Ms)UP zjaJKe3YszX#)gL(O9-D4Wg1rQ3qrqg7_15Qq9^}8~!Ahe|CY(Rh zK>5%5xRD?eez@GjTJUEL$HEBUHt#-x3SP30n4N5l_=uzHX#QEt?mS*7^Sid;88M^& zJR9;y{l@YD;~#uQe}eKSHl}8hfudYZ!eo7({YRaVHUc9>YR}cIW?p!yeGiaWJl9On;Moz41gsQRC^OaTAFZ1z&GtziSuRQ9tLgZ*H3BR;OP_ zrq2y@eKQ%-qChM6UY_ZFWv0UcdL^ItKw>-vPv7u(w zhUM`JZqer+-Ye?K}TU3d2TX26fD`HyQXpca_2RdV4zrtj5Z&t?#=__qht<$SMx zCJ(;NH z?>w;mpl#?f9^a$*VXvfaPI15G(*9<@N8-lulh4_&u@vXQB+$Uz_-))XR7&5A#n)p` z1y(x@OyAxg%}ur!{)yYLy}V57&`oN&vg=T`S2ynq6_J8X8vBX)rTfA*kFTds2WQ&+ z?>ZRv^ZJ%GpD8*IiJklJ`Q`8IRQ7Wy4N2U3i_h%VPIIugTr5ei4kuGuR$V4|!J3W- z`?uMf4!&CFrMkXq#J2p66KNYJ1gn+6xatX&58tdq2llv;d(~-vc4VRN`Kgl67e|pD z^Yv6WmE%DXNQ$U!tHk+?^Jc2sT<^MSn@vVgMk<5e)wc1iVuG=+hs*a<%(@%ccHX$Q zeu=n;{554{7!k37gA{CC;x?&lEX`{#X)Vbz?JBj!W8%5x>b??tjpmf+cFD2Zt+nyO z&E%je?yrFYhio4GcAtG>FUGT~=)o13hFW5MKRP|Nu0YJmLc1@!wbo(u z@m4iIUr5}3vME)FlDKgfA9~9Fiw-}O;@gBm7OIw_3V1k+)y$6Iq>cG5vij?r|AAP#X z5gyk_mMaCYyS>+whHk4gWUjL=lj4l}oo)IvvUOoZkfuX9*m3%;fn9V4yR^DmPQd6X zR!^JSL%uVYPW&f)*gM`wy^J14j1WIRVAvj)SdPOi_-`#p`QPr=>BTQ!gB3dV27_?) zJ?~(m+!Gwgmt*NS)r&VG3f1mQVT_|-xu@1io7i9O z)(uU@AD!Te9!NHVCKh^tWn=Zp>>YDU-4LYpy0%fSUb!#$TF2NY?q#BBaB$DFvk1Zf z2N-rqj7+%7CHZx_cqf-w2^og^iAX%;1?l8E+12Iz?cij?bK~&-p8W(PPixP`x5P(u z!3w7Jv7K&JlRQD``6RaolUB7zV?|IqJ-CTCF8h%^?bLx+%(e+;!_Q}?gSeg4I=xFA0l}e}g zdK6a#I^FKme&|efIP+>;PmES?wly!sNwFnzXB`;1M=LhP#8!Q3*>5mGj5Y|qpCAg= zCOisQK2K|z+T-H1^_Z`BEG3kPjoeM=G{D4hn{f5>?oqh@ zq}gx%&gifBuooEK~jI7i?7_AOiU$*6Tv zNo}9B5Z7nqh+W{7?8md<9_k^N{4$x@^QkZUEy?m3Owv-Zo8cUdR#Rr-?4#9VD*6no ztCXG_nODiw!tEz7BMQ+pB{E7H);?LMx=d_VFipN`qqGC2sfwbdw5la33(jE@vRYT%lNwG2ae{v{rbfqTMm(|UA+F_%%iQgQANN7%rZe#>i znkF59p~@pv5u~}b)hs|UPg63O9Gz@97b?RRm^tvmklQ^&cGXo?s|TYr+Rxc)fWtmt zg{7sTT2I1uE;z(vOZF)3X+K+)@}8gDoQI{+=RL1}_PWn#RNh=e(^lCLqkXJp26lxp zUQH5&>>K^$#OEs31`w7*2dc>!CU*`VQCV-IL(z331MkDGeg>zn(s}1T*Gc_iUz;^x zG(&-QwZC%7?&R&fw`3f08GPIABwCi(Ejr~oZKp$FYRPp?Y03>)EIM5#Axh{(_^GGDh_-cvRI{x{~-u7#NChA9W$D)-ul91n^lqnHlNy(h)Oo0|9aI{_> zUB%Evrn*is`!O!A>{JbLg!}BrnA0NWYKm=q&O!~43=H9`?GU#Frn$eHExvdnT8t^br+fviIHBxk7kY|3Wy>UfAV`DV^inSaU7bufG9(hJj1 zz;xc{{HgPI?k8!oJm_IW1A5KEve~OkVj~+jQDK~?GURhj>%T_6ee!V}iYD)LwN5=j zj%P&tOqa~}rAMLdVVkj@XgUwOZ})cZiGoMeyzk|%R$PKRmN#R0>x@(XstgtZkCkYQ zLM;>fTryortffWXxw{{|UNsrfyG{&(dy2FQZ$7_&89|4h=2G1sKFmV%32eyrCM#`Du$R@tJskqo6HHJ>}k`;mbB@>=1U~$3A8**U$I`?Q%|cywTR<&&BCrGMB^T#yT@m^u;Eh$HJnO% zisv8TV(`@W@K4)Tt+G&%Q-z6~n(A(8G39@k@=tn!RBC5U&z!-DIGJeGe%ANk>?G*f z>6=YWyo_js2pYBe_?(eChxOe{U5bZy)>eiV4Bc8gl^#jp0xRcll3{W>oPXgCBn<%Is&Uw9n>&H z?`Vzu;K$=+k##dw`jBRqmLcI?G;--*5FyGdcl(dUT7x2YLeQqgZU#m6>0oQXd#O=y zhTX$L?Ql>Om003!o&|w=|Ij_PkyMyYo9$uy+roRI5>RQw(jJ-+^5`OTb@qqaJIq?X z8|m~Tk=uGVDkrNgWfIOFwN6wTr*zuGp$dK#!l*-5H+McXwbh#0#9OuVl)y~A?W??t zS}C|`-rki+wyjiX@yq|v$s14${3S!$I&eZq@8+*H_OfBu{oc_za9~zx1v@SCo1G|^ z8^7zT{*Q=BGC+#iBujR;vNOXVrOV+S7VT)lgmD{@Br8XAZ(+ce#WU6eD@iSyz3+h0 z2cv$1kUh(X+?R$d@_|69Vt%MJ|A*MeT44l&9AVfgN_jpUXtdeitn&Lr*}!v;}ETP-HGv8YS%2tk%B&Wa)tD(BQbQ`AYhs zY%#~7T!CGq*ONExjMH5}q9?T&OCI-YFAanc6h6e-nnNSqS-HqM^DYv{T;UL|s+ffydRw z%0TscYJI7p$}XB+G2ACqU`648CTR)4E!OZARTlr@^ZeIe*F*7X3!&WpLo$Pq7U)`k zn5-r-b4Q0JE8hh{P*a65pKf)}=R^dp!s-*M>khu{mCDsZ8JNNYSHjDIa)r z;cI`%!y$4cXOC)sAR3`txLRfJ@J6vZbciJ~;0-B^(I>KHz`3&e;%Wv28)y0io~T88 zlCxm#V1+IIkw8W6oxJMA9m$;g1awPP86)$gg+sA4>7VHJI=3f_sk)sZ>kza!m;7=K zvLUW$o&t z=l-UjPVAQ;kOE>H24(l^(lAp_AmXw4L1h}(Xu>-p16BJJ!7Eu&oR%Jcd#ts1=K{|8 z9f5ZBXmWHDo@!Z)XO-R_`l3d8u(NF~eDp+kKCBPin;)N}Yz>EjxNLW9^o ze5~oV11yWg$a=PUw<+s9Z;A1|o_R--ce}urG<<855fWHkmrN43=q3(>seV>6*Ir09_T*q-|}-_2DZ=P1*y@tl~WINT zm;|aM-FfQEy3k~Gf;?dlIld_)A3f1Id)O17FE`r1@z+!;@NCIPY$R)%4irg&ug21H z_=>@(Db2yfhZ->LX`UGQjNPB*-gZfy(K`&(rsjSw^op*GSELVV?r2S8c=pf`C@!$? zW#5P~$l>A9QtOhfPACoA03Kp+^$5&l_B$f(pJYMLcNz44%(q3-ub5|u?^*uHR~H&q zVRny3%Xg64w#h>6eSm^$&HJ58`iAWmj|rYu>7rgYl!A1`GfLo0$H-&VfT*=tI;-vC zaQS%XSy&dOL&P}4R<9_%RMwbGI4pSsPy+5zb%)ahRwLHw*$+|IjmbI!)KUX!kleRZ z3-NLsy|JPWr)f9D1z6W}`fB0)$MZ)Bo1i_xMO$@#!RVP^Q$RxmyufXjFFqCyW-w`=_?anL%%Wroa z&Xr_)`E_6SDd@qXo3`cc`Nz*D>vc*;-8~e&4_t~d5Lk?%A zq0{PDB(Q?w5&snwz2;FA${aHksh4vt zQBo2UVZHYp=f6J?akc#f2)T=??5r*_o8pfXhHkT>seb660B7@Sd_VCu9wDD1==l%H zgS6FtHNqL?A@NJwIc;MCcKLF_n~UY5g?WxIJJBR*S$u17hR)*xGdGS#%LwFIS45&? z`QO@V*cDhRBoYeIZP)8sui3-b`8`iIs(+e=ukN1;G+chNYVYojc%Wn0q1(GSrYbS8 zt{#{&t63eoJ}BhCkpr4x*+7Qgs-V{%EXjKQI7zNGYkVl#%U_bUe@wTOuEqPH9WAWQ z)4GBx-ujSNdtcXcGiOX&-|oJp!1dr^I^bW%uShcJHnVUPycV6^jAt>wxv7U17Lxe#t9{S!~I=z!s%Ka!AOoE^d{c7gCluEL}Kwn9% zI37)nf>BA?#G@;NR|Po)VG4YE`HHV^A%)oLZ-eBrCIFJ08R5+my*$42=}MzDYI#1Fp-Axg z?@|H(zfu8Bq_I(Vl4u!mofc&VS}SMN<4lMx>kYiVn_({ZcM{ zFGYVR@CHsZ-WRn9ASf@wJ(p_3bAusI)>e~zcn^7F{&*w*@zV++4|G2 zZnp9imDr$-W=#$+ev!03gaj$GTJS_MiAchn;E6~a^HL+XziR7RS`JS;K#KbG6|sF( zF5#{)sXPFTZzOd04maL#PkvIor`b_}cA;Rjr!Pf z&q!m8+-ncmvK*;mErnaZC{*XCdja@A(?fswrj_&xd%>e;J~*0Ohea>;7Hxq=OqnHU${W9l_0KptEjkdwN{} zQ7zCc2cN42Aj7}N2pV~_;ixyE1n$LUnpDEnEpLtE*;h}XKASlSYYQ)ubi@mEr9Q*} z7p*gf z$;@2lxOmR@SlWLHBRbqS`~`v1LCZv^8v)npRWc}cvuibO*#IZ$(o+)003(p1Uo~W9 zRQQrle`51DrAV&V@KUS4&FO?bWVI9BRaSBxUZ6Nr6uLMbl8`|2HE(Os%jhNUo7T8q zSWPhAt3a7vkJx=GFw2KF;Pw4#4=q4HT11|KIp9g4antp{2TBx5r7uhGSVw^iPW6kg zTkCHE#E@XC`T&GLSYC!HqVrBLhACA}MO2)Qkr4h>VI2tGg7D&lSTwOgO8;_tCJ{U73G@-6LNhKP+qfxEq z4N=T}*wlJGZhgz)%rAaOsRn8m7HtO~QuD>As`>KIf`^hWiKF?H?bnaz!+S|jENf@& z9}@-KM3cyj_8?)hskx3E#n2J6L^wpQlGG0sxVD|LbacjIyNu$(qOIC}?@WtDVq(~6 z`Y;dGB);&c@!RRU9uhk6PU13s++ulQ7>u*%m3J>rj?-d-VXKWeV&=lD`A zy#DNYUfpVH-tQoZG=L;h4yhz9feX}9Cn^sh^U5ko`MTMv(wWlCbtXN=> zslMw5rjvkC`SHzg;i6<5a&Q8Z)}tk1?hWGk(r$lAy4#vpfK#(uYV{mN+^(7O^O=TM zE(50E>FQIV)!5TthPU0gxLAh^c4vuP!?tL}Ptq6a_1KN$>D@&NV2C|o_kB{#U%i;$ zF%V8Qf|2_K_-aPKCbJ9i_2=An*so~}KH&}CqoJPet7&P~ZD2T?g*lb@Iya?gVBHna z&oh_?PP@vk`jb&pNT1l?X*Si8j;gNJ>r$WQo`zfbrH#)YH|||h_tJ`Hd$>Ab>pw~} zW*eIKuTPcRj%(pA*!Q~O_*nd7@K8W_&^6EHXlKp6K8kCpWq-(0=GuJ|!kFf6_9!W{ zX{y1k=>FC3U|tZs#3AKnEn40vSp~qa(~Y)CImV9=S`q-kdV0x%c2gL-4wX611@CL* zZ;dI49JC=5NRpBkLp$C;@u91r4^^BWGikg5y) zkq_1==CY}e@xp+>d~tVU-uD|&u{gVao08Q=ecz-$_;T+HC$4%&IjRk)%uj)cq$YVmkhD=7L*Q>qbxy@TNRU|ej!7MA6Mo#sHYLm zo_k5x{WHP-pElhg0|wmGDO2p}Z+i8&7-0i0M%lxj4g3vN^7q(d5yFKry>HL>?*Dy@ ze?H3a0;e|$yvF>ENB_H5T*`3Y&Okgh_D9N|fIU#a-pRGU<_zFCBw)>-glE@)lM_Y0GUR{{VtQGvxShlGVk?_7$PRDJXCt#P0qMTh1 za*+L_hpCxRqujtITN!}5?eUb)-(CF8(f_L3{%LUjS5)|)LGV8o-v4Sj{~@yfhn6#= zrVW&CKkNbV5Z~39_BZb!F;Tp70uL=tqrzYWQ zk{h5M@Pj-+Da6Ph;u}*AGdWhXPQk@MENePabp86S8*0Lz7>ykK&RS?tiL;rn`QR%i zbY1R&U92`>Yk9WOZku3RMPr1jq93EoLnvJ3B`$vr9l7l#VGuH`?1RM8PalJ1zv);- zc)PgjNLb=}Q*}%VjUu-7G73r7|3z?&iZJuXpOVvV^FGJNIp5QNq~8L^HVSz9bT?j5 z>&+p3nMmW;dOqhhM&UN|u@zF@e#d%TODFY5$kN&G{WuJ0uw|(hr=CZA4gg}Ad{Z8j zFl2az7uKulJ35$4#sR4KzY2;?i=qrSKv~&`cv>@|SG4^G6X_=q2^!21W8f)qG`twQ zK68;PI;LTQb6PooH^?jdl4v`53DgB+K>e6F1nFbWQ?@B>cp2r~yw|urY91vPp9knN zZEL_->~Y_Qzz(;H%5ve9-@b1hzC?&meo_N-sq z0Xd)bFN=hM`BpF|uVI^cI)0q;1P`Bo%l(HH{x6rvbpg<0z9FGvp9E|p+ZxmW_6h^P z!V9fHRbkxxXuw|Hn+c+RoJxF$&;(F@)6k{q15NUbBZP=|$eNTJ z`xAT69iUqZyNgzPOPJA$FUqm|(>#jZ_?R(_1c<6$P%4@OmEeZ(VQx?N{9{NM6U4UaQSdvHZ=rG#F1qo{mxob)GgcoX3D~E88D*z- z9@AVj%z19FcE%DLdvoKbuqm);{Ed&a%ayr^U~KAU%>qcWsx^9g3TCJ@!l|=fjI|AoYjbVQ1g|E zgl4D7Mw07;o+k@r$tudEWN=N+;FtEzZ?sW69}%2@gUM^Xz25f_Ki=Y4y(nLECuPV3 zb@k#+RZSi%{w9^3C{SU5c@FdoRbML1ul=l`_e%%~XKdVUF+_GqaF1<@j=Zpu9WA(R9kL6?`^nbLHLVfQt z@thIOg4yyGC?i{Jvw9f&;Y#E2YS&sF_L&7G$oerDGX`NB>nG%nGAi*Vf>^;_8SFsn zWSyJzQqe#(V zhv9^?*Xo`n0UzAQ61K2)eLkN~LX=^fp2Vj5Ro>esmBwOFaQXc!FUC#Wy?h<>f1M1=#MilHOr+hrIsyOcs$dGu!g zW^t;d{OethFBW>uW%fW30`*;P94lX*sxY!m{aNjWTX*bjeeQ98{}6O;1~St_*A?gc z)x)f=o6P1!-XUymcMg}c>ELnW$*2;qPWF~4P?c#vdNHsul6`#4Fa`UaI_jRi9or{hlk+pzXy&n(pB45SVW#uPQX zh`SV+Y~_+SPd}Uj8G7N`cq?A{+{cGBsGlQ&a z;AY%htxF6An6gl+LH*_08r1LOvLJkTkGFp9mH;&eDpG|9_VuQZ!j9PyCt&>jkAY&* zfHE@Yss(7(orYVlO8ugh+l8yo=l$T3)UR*8Kl#YQkACE#w%0eziJQgxvw6Q;gNmR+ zJo*DeeKwsNlVomc<%gXh_{R|2S!nFv`dv=X7k(y;)-#J%`6YnpRjOJ{+#R@rFL*A6wHJ}^y1<4i;R9H=Q|j<9qMK3O8mrOHiJ$M3fXmwafr=0d@)#lhm@|UzE>cl!Bcq`FDiT?{ zVQq(Kp~%6R=a*^LFjNbw!>YRD0yIgC3_|Znsi{i#xf4q!6qb8ubE=VJYGw|&IIf*rpUk?UcKg$d}&Q`>be!d`kiZTVv zFXn8)PT4V7D5g{>(C6vf_=L3i!BTB;eS}y~G@^MFD=~Uxmj*$3N}E3c-Fu8A z-hWYa>C7xkwz@rLi;@mqT__XfM*%DVm;FvWG~!XufEqUUW|k4Xzs2@${c>ax!!BUd zgbE5i9n2;;mHlAN69pC8u$E_JwRCvnWvZa5dIny*H6Hh2ocHxN>Q2t1Ph@0Cb`-=j zKef|FiAK+XsMCalL3Bb*B7wyG-vE!HRew(e^_uj$4V;q}ogS=$&*1jgj! z?-j7jxcK>mCi;DS@X0pVry>Li@^k0-NIj8XX4cL{r`g9eE~C3%-htD9-J+4Kea>~= zI=ikUGQ=8Lqn?F2^ac(ze>FI&Mw53zV-+8M{|UB(6y2qC*>~LwNtEfrlcz)LA_@`9 zEK|qt{U1gjN-m6u?J;%2JG@yw0-{oYFkTT(+kdf(qM+v&Dfm z3fnKn`-mWDmxxBRy&&ZYz@Wpz)BU^)@Paj5`ZI5S5JdoMIyKTWuGN`{NRJo{p;04Q zTL9k|1^0vooO^z#r7X7$4O@zC! zQ-BbwB1z4iNvHoIwym666US5+jSPbbrOKL)1#g&rSD+R8wAA()t`VfNm%$nYRkrX| z)xcX#(~OTB8Z23z+X|I!U=K!8f~i$3*l6&`Z#`WXF>pSDfISPl|2-*h?iOZZ!XmFc z9h^+&4-K$Dx}ZxQ-jdqV6{F`=yUa7nTp5%(pJ{r zQCN4c$^kR|<0=^lr0TkDimFdz`tFz{OjzYd@s!ODk)YexHUmkDu+9UMWY&`EXQj1j zwW^Y_iufp}&_TbPSE&7+CLTp?^!;v?V)12l?&7B&hi+J&J=At$YJ8A zp3h6YqhCCJj1@*f$e=7w&bh9ySJ2kB-$JJ+^C0jDD-%z@SWtbgDJo#qokEur2X*w7 zJUIc?4M4!!Uqk?YFG5LtWfsHfC;=2jKBeE(h$4EGPpKavdclOlNaze5BI?C+8l{Vc zzn+=-6{p+$GX65z#5Uh+S#4w<@n{s_UHuXjvmpR1jjlNqL>(#OEt`LC4b{0zbh40I z2&Q@(=kz^Q5I6b49gwH_o%zycGIw0epmwtNIA{F0P#rj)0twc6Fl{rfLPUeDSV{Mq zw|(f&#@2j!Hnel=E}@`izalWuf0RS; z7&bxHL;gts2L(yvZL6c2Fe&1+9ydjRk z#S3firFnl41Et-=sm{8?eqXbB=71#<-IwkJ!P!g^Thy8unQ^QAO)$$UnE;qmgm2Hb zMwvi{Vlowk9;YMhqY)9hW>XseM;0G3*{$Fkre{Mtah;L`M zigRXMrsBNv@yS{tlVfV58{;|hB~!$U4BjEpsw=*ctmuRA-(@~r?teMLaL+8+It8Lb zY%kYwHkvF~61(g7VHgp${mGYDF6w%uWLcCe3{bNNX~tKNx)*SN~p z1b1;|2Aa3sYE*D_+YJQN{RhVSBuWCS%UWQs8Ubg}zZ7=&eg*d%)M?8NjNp*WE=gZf zoXlcE^M_j6B%h4mvW0g39T|7F{39pku zhhHXH#EWV^7E%@0*Q(xNC|7m~1iM5srwlO+bO=461gxml`T!6aEf^#iIQ8-^zvo@s zLWCf(b)vIW7B?OBjo&Z=)+Qum(a0o` z0_LDxZPM<_qW2m;=B4R2bwjL!v~$LqPMKLs-e=2|BXSQatnB|n=L~PDxEJY+vK!ru z3EA`CIfH>(z3w)RK0>+ujg%U!1Co1iZ^R*uRqK-c;&3HKb!~J@?!pmaz;@32^b9L!lTWX4An9<5(SPD%`TGOFLQ_@5p z!yBD#%q*+xVK`f~>rc}Va(@QAr%QKht9eqe1n{LhhTXH%1um_*1^f413E7t(?2;dw zhET+x5AiRxcq&!H1B%Lt35A|A-%p^S8G}R5kipPi9Kh-LeFB-*@sXo0J_9+}fB_|e zv6jDwLI5Uh;=EyQ-+T3#)424@UgHU13|8soU1>;Pc821IZF`K!5^HI(n#qQ{KQ)hw zenRO%H|xF};i2D5ulFNlFLW|9)n=}LlP~ptNrARh){k!=X^)TUW}K8ffz<2R4F-?V z$*nyuTLFGV?tl;&qfao1PB65s=o*D*3T`)0kl#6(q0U=^Ep%^;#YW6JumzN{SRXGx zmtB2()btbV0d;z;_FrKLv;%hh&Kwd8tWnTy4=uXf>Tn;aqvSw2<8*Bnh*5$a8USEF zI$N$8AiR3=TVH3QhK@%_>pMBgzx2g_v(Cq`p;!P^5hxK`d9Fh)*br;%W7xpn-U3lg zDN(fA=MebbsZ#&86OtK*wxI*e=9#AWWq_V%maXo;t_ruoq9Isbdf7TOXqO`W8|AIPj9>XpDh-_{l6$i zF0lJX`VLyIqr#-5cp{FFBp{DBZC>Z@>dBwYB>uRUwM6sU$NR` z(W*oLnl9=?x#VilXb6LriO`)d`q+=2rT%m?Hp5HZT!pD+2C;DDfc9&s2+=d%;fj(W zA$TfpMPr>^?U~gDXs8r84JMi_L;f#4SyQV@na8gujOpB=L!JgyrHnp>bZ1;*+Qy;V ze2UfixB*h5C_d?L=wRD`y=`MCdXouqyJHfL;4K!><)vYM_$JG=xq_!fUiF+q33F*W zDbhP$^u2lZ9aVu_q?$h#e3{sBjDIA<*E+y2bFX_S+*W35j@>fI6ViX5k{~kzT5{kM zH3Q=5eQn)L{=BMmZiLC{Pw@SGW*5j)V82PyU%IM89&*V?nP>D23_o6;UOdM&BUBRw z;+Z0W5id4!UjV<53YWVhFr9D9-uN2VPMS;KUgHntfz51!XJ98*-4ivzy>x2fUvgOV zQgA_Zoq46q^C%PE0NW|BTP_8?aN@x_D5pYJgvV%IADmgS3+VY}=E%b*#-vG#Azoy9_x|uqT$D!BZ|1h-S^NmIRME z^(tziY~2P2CSv_uE8oC=ow375ma{7Q)%Fv(Nh*&pVPx#38m;0Mx7?XU%HipP4*N0h z?-8>9L>sH0-?jtPNIpTga5I%2JrGkuN2mX_&e~4|Q*c_hBTy&?r1%s`yp6=(1GZ?c`0R$#z zjlRmt`JD0yu5cb6o_&`FtECT>6QMME6s61IxhzKr4E88z#1a*m93-RMo2z9|Q_3dm zUJRj1YdyE{^5{BH6{RYCTnu`B3r~|9v_1Z#+Nbt-*#_HPZ#a9`C7}zUf27 zqL3OzC`o(&Rp{Brx8}^d{>V7;AkI6Ffl2I9dt}Qy;ea-~P33 zS@XWgXUb`mD1?0g*^uww-Hci?Ka=g`{ZrTuLUyN>QH@@Wvz{h zT%&gyFY6>D)N>Vk>IL#zdcdWfT8KjDN#fN{zn;SiP1_I?cUN@T*|Il zb@|S)K1eb-t0bN7bsN^n7Dnz0paU98KKG3*Jb>-ApXN#JGi3tEg@{88eA(Rw&@AK-87?x$jrx6Hw3~Mfe${$5*y)dVv>0dv?X? zs>ih}r%iXbAyK1bX$qEs<_@@I`$^K#H(8{^nXl`_A^hG|X=`_)yH1-m6PFO&pz6Y{ZSjdubby_^ z7DqkMk~Z4J9KigNI(j=x^c}iz;Ty4gnL5M<)<-mUWRR%$P1 z4*3w4175ezuZ}u`R|p@dDQlMHC?5GlBsidmRQGG9cvd6UyS(6C9dUi~R1o~~q@(rl zdw>Tb5Mr_`EpncbR2W;|Wa_^4t6PK<0YDs1l$W{KA5sTkmDc&~{85vv6ga+1vR2Ej z*HMsK{_o}Pw-Ie2eyw9ll>%=6^DW;!P301*I+O-v{B}+>cokVLUOajBfsCjHmw&`oGN^|KA?; zGr+qaGMqj0KkTqyeeyhw^PLg1%Q`!@t{yjJ|DV!f2rz)h_=>S4E%Ve{XZmZD=%1I@F(XP$1X^aYNVyh3J31f-T}aI=JWN{73KkUMwC8CLoC`_^O5P*H-DrBN4Qr=rz4s+ML>U1D^ub5;rc%(-}5@0osDm5pTz!$qZ14dgG#aL{;Lh} zV;1h|wm`up%N$bz)n8Np%Mn7QfQJzhuO$8Nu>bD?!}|Z-u*=j?eSv^D8DwMl;-(>a z7XdlyNO9`|aAQOO-mUCKq=?i_b3zaxO}rqk)hY>!o(%ROklM_!ax0T55x?;TD!)=e zQ-}khTo@-j#gAm?C48LIB*6AVKkU>F&o}$8h}IWrKh!|?0^~w9a{9uyJKh1NjN{0! zR=FoB;zx+PqSqYljBP7xD?Omrrx$?(N{5%7>y79Gc}n&;Qa5aFGBBFz1kDI{+h7;*Ph{g7iwiME_zT>7Ik*K(11? zN=bSaMI2}1f|3<@LU1U$mDH$~}28K`f6z|dEYpzK#4|v$D zftwH8GB+s?5$2)=sKoIFpTpeZ96vR`PiXz*zO-uYL=*EVB>NOS1vJGEk2t!*=7;U8 z18m>zvvu9ndtbUaCSG(+d=?=at?cfE+E@bK^U-7{%I3Pz=+c|ZF|OrMjvQJln=@t4ZG;y^5B#JUWAiOPS6iw=Z-=5P9Tx z2@_w#>@7Rg3X}NCzx6g}Giw03gdsic!+QP}Cr%fO27^66UM1NFm!kGrx2O2<2p68& znAtP!3<`cS7JrGcIim>x$K!mt3&>#8-lwqqR_TGSW&7dI$ew)2EI{!Rnth_Luiqx1eKJFTPPw;CCR^u< zI69JdP3#rVkdgNbyN`|JZ3>K>f2f>!z(Z^p$2u*vTHX(nxPmrP3Pzo;I{S`LSk1eu zV`zAiH>O*A{yO1wp=2Mp0j&YVxjK^qOS%i>i5Mo~_y^b3*CZg~LA1 zHXj*kUQjK1Nl!I4f%rR5!Q(ig`;8_C2vzc(?*qJ*+9U6z5ST-Y2d_UaYNrO!gJ9ru zMPK{gyW({`kT=?>P&ETi(h^b!{zA>b+iLh3>ubHL`B05am&Z%v-B|D67beC`AA-#d zqqELuyyTruf~S#js)6|~pWuTe9Yh*xuUsLQViWWm>x*C*+$^av0_nAg5dpO9!H+?E^t*ukrUFZX zzx$)WjC-Qp=v~;h|4=H+lhjGB5KXEPPDI7PE(#|IrRqO6Hl#jFdoLQaHwGf-d6e*K z4&-rJ@UC%Dw^Hy2o^In4SlPX|G;J%`>EZuBk6n@E0Cp|Ux zOm8}5zUdWVUy}7L%auWISE92RXRC@`fjX_2)pX4Ri@o4fy*5gsBKmZv?W+fg=@lL_^l z*PAS3xnwVg7o7CAKXvPGeBJF zwA^Zb%s?eHkK)y%c;aLEDlC~mQ-<$r!*ZEDBxYUDXFPrcQ2s+yTtWovJ*@s9_;pzP zn<~%;t-r*$YtlZ-1}69mn^ZgJSJ`-UAvpaD&I&KQ%^L?vU*MbNg=l0jS_7$AsWR<0 z%4>bHef=tX0{M`~$tao<99WqwWMRl!$AzqSWdHw<`NwPt z3D+*r2rn|hR5_6XR$`) z)ur&M>)*37QPMSBv!qud4U(Y*Eve^L)VBuYS;)?qVdw@i%Rft_eovMUf|6$}$MK>W z7|vYA$X3XCEXpcTDC&iATIGuH!{F-*iNvQBCBF7m@QbdUUPD*^eD`q8|BkQy`0x-y zqzB%m>WrTb>oIH{h_ zEAdlXcC^-SPTd&;`)EzN;6kMS6b%RWngl4kUFHWb^C_g3)-_q;q0tH zrZ)NZT6O;Mb#8)KpJxV^7&$m1loZL`@A&MZtu&avvsBMIzPSGM+qnJ&_&ql!AVhPOMyPQl4=8tpQ^StO zj0N>3YPO=j(ff&0ZOLX@f7yMi9NQy*dr_DT_U#Ps?JI!Y!m9CRc4cc=!ggS2n7$sTBn&APn69_Sbi zbMzzU#gU?WeTunW<4S!CNPE$G_@n&<$XX=CdfP#ArLv4aiTdQKfu#>+65vg4vk z&{6E@6$nAHvmD%Z9+j491|wY3bKl@>VNkZxRR{S7a_fc!wF}^u5$Q6G`OhDZ3&z8W&DNBdUjzQe&4qK@IP}&4q%s2W4?nXD2E0$V{GYK`ibt=lQ z{pxn7I%xcoGv0feSmb<>)Rp04nfU49Q1ODUT$uf7+B!gn0KOX2*vJ8}qs7%w16EC|ViRNEA65 z-M%?J>JNE-UId%qM_GEk&-H?zydCGsT29Cv-{B!Wk=K)htN$}aVF{9LECK1l280Nq zGXlH@#*Am0H0@#C@(C%27F&c_P3)2njf%9hY_jw44(M7L#s%LtE)#4S?Au!I-cc(Z z*ktNrT4CR(Z|(;fzdrXS0iIQ^Y`g=*^0VPGzTFeO*zW1>nHbXJHSm=Y!drS}Z$@+t(CIkf!D zf7S!tHEfnU7no5_=Yk4IJWrtY#A4jn*SiE_W`FuTd`eJd8f|wf-MojwxU5 z-B})-!wO`7P{*VUM>vh>>nkBy=t%L3lPWH)W>1B5QMwA|&BHJ6U8>5bG808AF;I!< zh5`$|9qxibPqDtxktC<|#PmZtsKP5PA^Jjv2vT);W*^bdo;r4)_B^|>Wr54^AOy8U z_R{0w{Yv-|3xo=kHv!60xY*p-Omh2I-allng2hdYhhBzHvu2BJqLOPuz!`R?#4tKNTl7~`=_(W57if^h|H?p#x`3EKN}FGG~`OfCjdsy+lS zxO@H>Ap{&gccl^p%(jLKT6 z@)7S&4#JQ`Ds!A%wq3Hx)ggRfSgJou+R@(cy^EABgh&;_8J2>$oJ^Nd7qB|tEETTKG|+Ty*J#khP-Y?eZX!;`{a*^c(Ddl3Ehx;Y^t;7FaMOX+w2Rrk`aN-Q+)s0 z`oyWII|hVnF6vSnvjY0x$OBo#Lb<0Oqi*2bCXWwkD$M!TbYCRgql2Y+PM~MZO{1hQ z)1$#~88k3Lq2OCfxl_5UudLZ=f%6|rr1Se>stpoh1$1K%2*_}iB%VgU<{YJ_dVP5B z=$RfZ!!iH#`i*X_s9$9?z6c|*SP;{TGZW-$`Qn*a_srwjpM_mZmt;L{DTwzerLnXU zk;StG$aG`zozO2%^8#^DQ;dtHBCuG7dAi$UZa|#GL+%ES%cwOg~_)c!V6V96B+Vm1P`a7@CyB7-ewA9G$f{f}#AE00C3ah#8 z0`mEprcLXf5YL{?6#wG^!MT0GTf2^{=gg46K`G?VHLEK_8*hX!v<~Geojm;ZXD+Wf z!fGARD|;@U?z5{SUgYBzW>e2emj?TDDt(CD}Hl9lA;tnDOZbg<1;vQvZP^;Z6#i0AB~( zMwhs@4Ou+aKgB7M&|I<}=4{0x=lH%fs8<0(?|g6ZwekbAr~&sc$xxo2NXLUzsp&Ud z+j{6PN6E~O<`_(${i&-_U5w~94*#sps48$KW|l5iZBFf_Lc>P6Pcp$Sb#|qlpBV2?Zv*Z}WZ>Sj*IK ztE|YhoLXxy9jee$TKsHV<0ezDexTzLHgGqT<32Ba51HLS&^x<{c{da-+%K$J61p-d zwxJ=E=iykoL10F;tE~1yUjUV%*Pq8Rbt@}-nWb--!b91#^@&Tb)b1`2rp7NH_Tn!e zwesKdO0~{LQ=dE1#ku1*kf9xOy?N!JdTtB~i6YbZ=4txQsCU@Lc<+8Pqw47MkI21| zrTH01PpfU$Ql`A~)M#e3)}2*$`M1Xx1xnA`Ug_ZaY*V7!MthAS{Q6Y{taUCd$i^$hLMc>GQTAdK9l+=NMRe+~Y~o(qOeM>* zxm_0g`a;$WwKUK{WP;1MGTFH@6{`E*o{y(ltirt}^!U_i6tr_V|N3D;Lnyw{`cH73 z#>T|I9OAk&IP(NAQmS`=2Ja2VaLXXY-uAJs z1WpSOZ3Zuky%BGb5}>+wU4u4w>nW3K0yh+ZkNSJ#HBfX@d*ZjcKxHx)R0&#KRt(fz zn$jcL47Y!fXNE2>d(8rf`sVp!8w)D4h~E5|j?SV@FQE2g*-%zC@$%Za#Nxd`GRWbl&r*Kalf$O#}QCYp20y|JL$gyrFeq~c;}|!+>cj+?K1p+^ee-* zetR&Ui$)sx{NZ^nwRlJtQb_Ij(Pfc@96M3`Y&o6lD19vC>x7Gx^o<$p;9~5wNw~~PtG(AcRwvp)8X56CuZ0m>yiE^Y7N7Ok@<>PXmuoqvX zA|GCTys@$|(jTL<+01dI4L*h_W90Oc!zc@^)GTT}yfHbgv@FAlkA$^Sb-8e}LXzj2 zvTcTBTzJ#%bo2XoM7WzzgRQX_wd2;@yN550$$6oK+L>q2t{@8u_(mwB|Xrj@>ca-T4B@ zh`4dCxdpAy-kh&A=bTDuU-odb2t1(N?$?OXA1~wxx$~2~b`D2100o6XIk~0;1LxYK zvlqp1gp^OH$8H!L_(5uE?t=9;qYbX-p}z#^JV#oHq=+x8g#P<&3mchPrE!qCbnI@`B179skMX}V1BhYJ zoyWy7dTaeUE$h0_<2o<=n;`2Q31+5Z_2YZe<1hAE(dZNRV>A*_5dN$URG<~P4MVUn zmQ$*1^qgJhr!V~p<6P0BtADdRf44=Vf8rRikOiCl8#DE<7klEb{abVPzptzFf#DW? z68QS3gZS4vzSi@O|F3`i`~!g;m`k5~1sDI-j{ff(Blkm7FFX79zy9qnH~u(RL~v1p z^Jla2Kg0T12y^%MFCwIS2iC=n+pVJi^Ua@q3&&`NWCrw=Q|t{&xGn-BlMKMNR5K zAi-HMe|4e19~9Mbcz*odZ%h2Y{cUD{xOCy7hQjX$WfmWdWm&WAI+*fbuMAHo#IE@d zN0R;eI(RyK?6DO9Gd6+$c%? zu}XhB4xWt+U=*zU<0gN<);1CQLGqSE!R>O$cAtd)g`LopX)099vU)ob0v4BSXMBYO-dT>k z^sf<-ktVs&9;jrGcVbDVQ)|sS?9xEUi?O3lOZPsciq@PT+A&N-DJiQNWb@nivLdZq6`^pWwTUjAxZ&2SOvjoP{ojn#~8b_x-R5XA{Ja;mA?6t2Q0xsCA zf7>p%uX1ub_aV5|rdf~@c8fxuRgdHAMp$hXLhTwWuyk9|h*GK92>)ee_Fj}J)%}2nQvzS8AwN5}+#%d7x zttfEh^y_uhiB+utO5_C4?`!r!bbJtiI_Y#_Lc#L^3yFE{NjfQAd)CyeIiK^8ei>ue z;sIqY9D!hJv>o#39%E)DBM;4mCi&N_{1>R8`CK8C$!CB5GuQuTOGxzv96cR@HIWlch{w?nmd6J5B?q;qyyvKrMw;-n1kyp^|)gyZ~+JmNzJcZq!o7 zt`w1>Fo43U*McXu>ohCVyp_L;rVSvkoPeIKvTo+I1c5I#bOJoNeplcWSz;ok*qxi; zRt8`iHhM9X3@5sgY?|&PVO%G~xkDv~V^haF3qad2n)3MKO=^au?*dk|^1alKBN?jY>&Gc!dh!2H903u`dlP}mO_qh6MPSzZ1$N2U?_^4lAZkL+iLW|*zlU9FZ|s;>ve+8& zrlnrVt98HHQv?;gT&+6tCE*fZL`EJi1Ilew;PvVkVFMPj4ekdc`Rs+a{%&St@|1xE zPw)0_oGWU{JU{iViEDI!MeyXC|H*I_1d@5!#TtH6rHLXX1Zog3M!HO_XuBCab1MVd z*_aZ?n%lYySHu@$I}u-Cja6yM9}Dc3cjuc#ji3=@2@zcw2a1HRDUA{Zll&! zjFOvBh!_QuSOvVwwGkOQc9SDDaT#lJdEB++-Ly1nAqlbiaX>c|amw|8mJ<;)rY-__ zX>XCvs8MFp3i5s@%PO@dNQTB9^RcyS#ZW}9i)#6}3L+jCIj-`)hbr z)RzS(*H6Nvqjob2ar=s;Jfdq+um@Tv%I{m^4&yxM>)uw?cV#t>3(-eXi1$S*^$Ll* zvs(gKI(CG`?JE1w)f?FKB0}s6_Um-jwNIaEPs;Oli|br0w-$@9>vqw&HBZ>#FrQ8{ z4|&(A4ilF!q#S?M1jm?e%9O&O9n?znDqyXKy%*Lg9Q^=%)4?4FP7P6tn`Cy_V}1{V zg?zWT>1gI_-7r$q6^OrVJ>#lM*xU6u+1)St`4QuZPRDE3KMj5ZI#q!9f*3e+8%{njj^1u$hXL?Q2pY7(?fe|r=(PRJM{5-+b(7A zew!8u-YDcbHqz}&^)`W=6Wc(P*;@NSL0ENm)?$vOo9R?}J^aKTqh;5J>|91xz=Lr z)Dg{HFf@Y$LmD9|cGSB{qa(nwN^|xkHplCm=xz5-)cBtRSWaPbGj`xxjN8sOl%9{cDz)yuTbbsZMQG>L zK`O z)0OV?djUW9-h4%LdkLP0Q1+1FeW=#ivkww5cxN=cl8=$Jbf&75;+D!Hq)zw-H00tB zqF-c*EAZ2(?C+;?RcGXlN@<4KY?~4reTBZo6_y)SU17UGh?!92Pl%}V<++yBRSW30 z-RgiResu_1Dyy2#V$PJ|Y;G*IrgG1%DfAryEE!JhGwS3gSHWK8t`;fvn(c>(5B4rf zT^lq8!?4novMyej!Mm4MxzOJ?Mv|0OyOmJ&aDzSQkfM6GdVO_cL4a=Wj1C^(J$bFv z3l}omld+An1e7O-@Jb7-*U)@}ol{MD>+!`ArfpbhFiqalnA8&Jtk|g>?*YF{eq=@H z6&#EQ6MIO>J6)LlGk%xy-OcRaiDAQ5ixFb@WumhHeS$kS5G4p2ojyxmY!-zv2 zZBP7&Q*>D<q+Pto(aM}uLG`-Zx~WQPIu$!xxk@i`)@fkJ9|MMP zU$I2FYk}g4o1K1QezLPQYi)&Wr1V6q-~`9eEwUvS6vqwIBj*A(J*C`?Sa#{-aniR4 zx8z_y?jRw#p0(VP^}cqg++o3OVdX9AxXnoMPk|3zCcN>Q_w0_NrVbZsF5T!ArF_8l z;TSX(U6%o1zCYydrEUEg(yH`Y;X535VeX_*PbxXzE^%qzmH)oxS@k*BCSJS|Z@2F9 zY_dR`?M($)_HKK=!rMYi3dX$M8g2~pt{6pP(|7nkOpS#;Tt++<#oY3_PE+Lq@yIM8 z-@Q+!%E>fC`6tH*4tkEz^x3w#a9^rCxBOAdOM;Zns?d-JXPRp-+{_%4j>vxU50Axg z83eXga_{GN)E=+Y${ay=CmWry47w~Mee{I=O5Wy8S8}(b6`ggupv41lIo1v?q^krF zfHyVq=8|HYhoHQ}u|bM&a<-@y_+2~20fx*c?!d<^b=pW9PujC=%2yP;K?PjGe0sm1 z2PpW_5Ez<>F#;2t(3^qvEr*HoPGQ`H^lbw#He;tV>iu}02JdLce1b6wvYVJ8JECD9 z%;%It^~VYi+i6vlo6C=1I%_;Z%kIZ}3M*y8KD2>}CQC1Pht28#IrZjoI|wtU)jFw+bHSNP%2P8=E*qR89NAANV%lBO5n-mPDaWsGP6hD zY}#GZJ~DfF5V~;`VXn7)DiuvWSTSs`?ySJQ3%ls|=OZQWDJh3jAGMjPr6jIdPNZ$! zhu+=mZNUsq%Zq}$s?}a{ZXQ%$l9Kmbn9Tiyn%qd5jU@?x8m-o80w6^2MpHSz(aeh; z_tQPfT6-YH=7u;I)W%->%-XRby~pv0a&ehrn{6gM*!+%!AnSD%*)O3lww$w`>|gGp zi4&ib7sx{z>++sfA*}@UprB&Oy%Yy-9TbKF3Nh}`A+zi9Ct2MF@3@RB;=OYdmOCLi zV zUG(RfA8r3O6!dYMU!@Ie{}pTOcCu5om+=0Z>C|3(ZOM2Ml=$V3+c*203wuR+HPO{O z?&@q-EBfnE9@=EPk}mAy3Wc((T(DzqTm*l$WNm}W6RWX^?<}{9AeUNonoPN@L`1=U zlWxSmT2?8;6<2|k6~?Um{-e&lrKS@gN*Y%x37)7uO_bRVuUS^gTbAh*0ZSOe{Oz;D zc_-n2#TWhpbySfU^chx5zOL^Gn%dmuu5G5rNMmt(3STreB;cHwQ>UzDrP%=%k+m3; zD6b;ra2q+WQ*NQqwdw0e9A3pY3ef2f!>Dz3@$U6DbCK`zn8DQfRQZgERBGFLgQ8~^ z$!slC+ez<6>SR%!$K)X}%W}=7DpIiAD|X@u|*lIy*@{XsfDLs?;m zpa4cviA>dSbIHcCtQMsPE%`uDs=?a$IFb6B?QO#3i)nyBzGZsGz*i6^%WssJ{7yR;xGZqZ^_BdWEFf#-U1Ys zPQ;wC+eaK_6XI*N|ClRt^*B;jo!GtdKZ?TrYnn1AV<~T)+jcOm_G;GT?(Y^Vw`>m| z$ZD9YtWBql2=AY(NXDe3!2L@go|tP|wRVo+h`uVj$wXm3ADa;}h0x zK?&)(-7JlQ%lDzX<4!GH;-p_NOwnlfC5 zIj{%WK569{1z;>DJNMcSzr>%fdC{DIiGdQMh~Ts3Tl&2BQl{PW@MdRkQ9Ls0XIFu} zCy?~YusX5s^jV^wp?NfukOQ7Pik`#^cUoRzGz#WnsJ50+I_Z?yNBrYJI<@>bB)jg- z`qJzpC^oVewVA(XX%!Hk!(%}4tk)jnX4@Xm@QkF+zPy(HvN!3Wt_o%S9f|=1c!2bS zsP%5=!xh8)A~_F3EP<1Dww?!IDb|RVucge+{2W8g$3d_|Jw?|0+)+8`5=ud5)DNf6 z3Vxy#l#J=C4#n8SH41RuM$@1?38Wf(Nq!#U>9)%Tfy{Y`HJ(%^U&PISqo~~0@)qO^ zJ1JavTQy|=486n68gZ~`TxY`?*qCJ5;%@4XYL+TNi{+CR z$=Z#SLrN$NxMFg-g=qMf{E_sR+hjpvGcXGKNn-yo59NN&#WBr*%}i_~cnXE&)qAX$ zFXXvzXI1X6IwI%@SLzMaR} z(zzB1I>Lj)_Fa$Og4h!|R0iB9Tw{u71&BQkzZNnu3jRPQ|K<*8)t-qJ(eKjj zGTC>=IJjbY{#&7~8+YuG6yL)FX36QHW-tLm5E)IB&$Z{h!y~wNGRrpK&Ldtz8!++ubR^6;z*1+|@_NC40G`Pqv**^4{Ky-|DbWVl&i z(7~-d?xhyLab(TBrk6C*&8)_G`I6K!E6hvBPDL|D(M>4&JjkWcxZ=935SO|4ye4mb zqF|t{=N0PNMb@4NB_H?t;*N{f)i)25{hS9Pt+7Lee9>lt^_<+GL%Dz*BFS#RT6y7K zOBq$CQhcN9Rxy@zwX|IMAm6TCB;E6}5YCOr#U-}5J*fkJ1yUxuC&}1$J$E+93CO3K zy{9BG#%VJc*~G!E?MXifv67j>)l@8%Ls<#kK9s#|#NyaJS9_84Tiw~zw#+;8k*O}# zARn~9A3oqo{_{iq0ZvQwA$dkb@*^5vV=4)L-eavbS*$)5H^-MFHTXtzvdUL+E z7G*OW&}~?`VhUmH2bw2PA@nBO22Qq$PzRkpQF-Nu1XZcob1dNtZ8L($9|c2A@+>?a zJq_lG2Yu*A%*@H(M;7bhvZYs$g!(FOvRoY zF7D<$2oN7f`*TiZ#(N14+Q_z0w3J$>&X+{%u8X{tdQ5Ihk_RshD16ZHuw-~rbr7vG zoyly@d&>f!;=T8~&byZZfLB?SBIWd2E_BBOKmo20^E=N{Mz*qoPuf&!dWRMuX4aJ1 zuI@@*Gtk=1cV?xD(g`q3@!~lz6Y0KY6Ss182v}`Sw6%_T3a&3ld?7yESJW9%4cmfv zrixC%GVRUg`~9f>=dbOmS%pht(-o=*ny1N~l_4a2)uP%RjO%kRmo0B(*)Z4^oNn9s z4bA=xKFbUyy4S(M=G+n*gjqc2HzO?gB+I27hw#QQIx-JJN+5&{Zg=CMuUoPM*H^K6 zK)3{t#;tQD+1P2Pcuj*i8o#!Fy*SVzmuTaNvy5M%MrVT_D|IU^{GnvJYi($E`SSu= z(KTIl8H6ZB0(@@tRBI4a{CcyY!nM;4hL!N3XY5(h62tYWaX_KXly%&;2{*&A%DX0M z2|FRtF$%NIGymle53Sgv$gGNw6B@C#OXw@?&}?8?{$aIv-vR}blekM&DG#7>ik%%! zHh^+*AP{XKX#O%NJ#-=o&sxz|CWWNPV8D7U64Y5Z*VCBWRby$B5;2@ zn~vM+mFStyn-|_^e$GMb1=yh0D6W3rd8Nu}(79e~HDPl3&rZN5Yg0>2g!?c}j3KuW z#Is`UF@^cb$l3Gi>M%lZY)`2kUN|rscIeGNW0SX~JZmgVY}Ig4nrZlfDrfW0u^>zt zz=9x&ifJKc=OfgIg;_McVdyz&Lehx`tfd_xOAfDvCwMa7HpSO@JSSgbb|RUSwM}YE z=do|>*p79mqOBFZi-h<_Z+msNd@c`bT+CQ97av$y%X!p~_{DCgd2T4*_<4^1A1B!L z&sQk&U6Hio=Y!})zBF#=i8A)8nn65ra_0)$4b9eYRMR9xAS1(LgHvhrXMsoDRM$i zLZr=lC&14mpCOY&H?ZB=*hqeAzF$E$MoH-`(IX+7p~^`+Wdr5G4om00s7N!Gz%2l@ zBirN7?1OnWMoP+d6iWOM-TY(QCd{9o+E&P^P4qE2?5B_Z`OPpHJaG zf5fx#I+Bci6{P)p-T|f%IL#;oXj*>-I6wV>{~XCiv&cA$|1>!I@m=~;kd4LE#j*AL zo1^lt-&G}ocnZJy<`Tt^H~r~X@=rlhi~jMg9`v7{us^=5``>~6@4%w{cg6nK6bSv- z6yW^V6#Tz41)jXN@gj`c{{cZO4TA;$+G!JF3ax^Ckr$`P&5*aZyZWC2Tl5mGyCaw|7%fO$uEASA8}EO#>`;v1i%Aq(0p|_=| z0m2-n!q#3rIi6Im5;Ul3)AD904>-i2S}=FvhCO67=vjr)!tj*%6Ri8xd=twiFqDBl}iJ@p6)PPxV(32Sr% z&1?t>c|T{_aFVjV)uHrdTdMxrbX8X45J+BT2*1yE5lSywH_BxYg{@5xGONznnZfpO z+${AtCjBp*gCo%V=NZ$#(7Wkm~Ek8;s(hN_4;)SO4~k zs?1YJqjT1aV`PivcOdbvGX3dW&}&llfIbIe`tMY*GyU{bG6O%4ca;A?_@R8S9G(dI z%6jbgbs||5ff2Wzku)`+!a&8v?)5A-M3|-1-+e2IQIgVut(DiJE!h=gRE!wm5T<>mqOh%;Tj1mu})?>iOm+8qnYP`)>>GT2H#>DM`I)M1x(-v~Oumib1L`&C-v z+TAr@D=$@6wZod->b$FI9*4~yW}L&i-WV=i4vqdGGPeNb;l_01uwWt&pcZWPz`57m z>08&trNi>nrvye{fxuUbS4eEVsalCxZ>o6cCd?y&`NP8Kc3=ri?+5-yXxVwA9 z4U!K?PpJuR%U_%jp$J>KsMVUk;0+1`t7io$8gWK^iOI4B9F_~=D}-fQ^b7eSa-N-+ zYrZ_SK&e(_hcpq!H3ItyM5Jo-g{*5L`&i8%MV~%~e-lQ*DaaIVvW2DzM3ps=J1^}7 zvd+->Zq2SaCx~soU5QxnT2fe>Nk*t=-+6Q#Um=yB$_-WEW8EHmVICvEo00w0#n8`% z$pR^lX!Qd9*66${)|>cr?ffpF>Q{1_)}g#zD>AHVj$*>#LM{#)X{y%8$-h~ozup=j zIYXRCik|*yB45@2c6(~Sy z$Y+nr5oWCV@~$RFSVmxP4$+4&$#NThPcjHyCh)cJ)I9Zv?HSah$uzR`F+&mIuJ$;t zi2Tk!PI0TOdDdVR9)78uI?C0OP7d!Za17&rrba_CRJ5F#W#4>uDww6s(XX&qbQxJ5 zL!c#vZay$_fpY9!@Smkn$HTwcZ1y!UM>{m6U_fj?c7SKueNlZj`Xifpt3uOUu0yq{ z4sl;`)Z#9fG1F#dC(-joEvg3NutPf&l)C2!uNR&#QjJ(2(Os-$12Tj^*3~Bk>ax`_ z(yhBdm?_o}`6TD>tXK#YZ@~(`NWh_;f%qKjG5xwf%putj-{pFILkWCjQ zHLC-J)YU~f$f*=Pg5DyB$C5tqj%S7#FG(Wk)T zb!UY2w*Y(SVd;_B{$?^2e+KbM*&*2a&u#plZeg30Iuc?zzgPl;JT?%3Qht|>(kb?g zLKE|LsJX}Nu7k&}!|K~;$+jJSyAd>%Iro-ottQ%f?$$y$+;%xwz>{)X>z&M3ta*QC zJtbs2wsZF*^s($0aG7_Z?*0Vn#A!;6A#wwHYo?H`1Z=nT^Vw~*%-=b+7fSO8-@B5M zH=2R$l*QD+FPyjhBG{rS(xEE>Pp=pS_%h;!d_1L3=W_y+tD~`X8rxB5Gux0qE9OhA zd3+;ifA+~-F-%+s`+IB+E10QdvwbxXp`9!M(T0w+a)<|V07Tt`WNOm7s%^>_p^=Q~ z<#Se`zQjqm$$C*etbA-@Z4mubzbC$$(sk^I5#zb-b8cQZv9PP`v&8AUg?ve&lWZr# z{w~3caLmiAL{smlv%%Woh${sdc}_91^_3hpT%Wgy@XDjYZ2M8HYC#N6vAQmxuMd!9 zg6W`%c3A)sBnyFW_1GngL4sNQ!KWU}%WAW=Mz<~`XA=Zqe7e6UZ-_xcGc<-Dxgd+e zApmQu#IWzUU3;gXCR&YVHs1S!iN67dFPderVu@YL;{GJ{;%4NjXCQ)KE6`t)_u6I} zkVtb>j?lJmX{Z?L;t_!b2HlWT{JQaWy_jT|r{_Vc;eBW}XiBnM)@)}7yoScon(Z9U zYpGd9&Oz~SvfB!>T{2$HJg7C?Bn}8sz$Ib{Uv}Z#{ot#;mB=VrleDQ76JmONyna%{ z8_VT%0dQUiyJm?m@(y)Zf9_j*$_T}a++`88#J7$YL2JS&X-S^CE3gq@f%$TS?%4T( zUGOD_UiZ*%Od6c@{zAS|?t))DvCVK519EQm=Jjg3qr10Kgt23UiU}f3_-r#WAo;0} zu;xpe*-ppktQv0c(J2SN*q3@{c$;SlvL+8~;It}T-=tcUXQVsdUw;{4vSl63b!J|~ zAZTIQlN^I~l+uOJ@49rmOC?^U+$TH!4Fer;=v|6xW4rB{=0^)9gi#3GAziqW!}sQ~ z_pm?YukG71&mlq)Q!la924=2fJBf{jM4&JP($o{ZmB_!-RF!NN66C^9qvHi6WZAY1 z`z(Jmr9dmXTlm~+luMPbgeUbyd=O6fY6R{h_V64uMXcOxIlr*Ow;d&*;8;r4Dlh}lk>flp~SaHG1LEkZZa%qCz!IB+)jfMuJa zCH_JYbx5w+u9;ih{1uzL`SSMRMICW(PyD;Yc8V^fbA)Y$?BUA?*EHznYkujE{^d0E zpfXs0MI5N_Ilz)$;K1%iu zSu?thSn{g-)SZt0>i)aabZ6BL!wC8bFEsP44+sb-hS%~;0x#7SDRF@OE}r>tr~ z-Xk|kL37d{IYA>PH^4n0UU$5EAaI%h!$>?5b}awSaMJlZ##h&l#D&@ra<;n^vzQl? zL}qjCxIz+|I{8g9Iz^-%i@B0PrMGMP`4#b*`*iB$9{dS$IX4S>h2@peN5JzM-L-ox zT|h0ZEUVWbPJZXU5zlD?7z!sGsV=mitIvyPGM=H@DH&@m3{pIM zq@OJS>!5`0%G|gN?`l8a>_vYg6&jjdcRxdsP(+Qya_kabn?3HHCJNQE;q{LpE^p1W z0;j{fT^HC|C(q;rGgvJjB#?x0exyI?*V7Svp7D^m=Yc_$VIi@MS8?E(=9_ex!6A&0 zXCFmr|E6%ZJNT`(sD0XBe$L-MIH-Pib+<72d&q)PnH=jeN0ly%CbZ6ZP#WVgzY(9N zSBS&dO@&mmug`zUt8B0SWhIFXFbB_-dpdwR? z#7K(_V3#dqb=&fT#ZSz{AN?Qp-aH)YzI`7rQK3>vf&id7c+IL=m@muV<*_bmPpTJ;SJ<^6h4)_cB(24 z$Y+q$SrP-l8-@2~NM*R6mggbHh9UqKE@H(tR&2s#IF2JV#3h2wCe4oP=KII#XKg6quDBP?C`gKh|OQoXqrPUQxUDX zUx7AlsWq(P?1PunbLLchz$f^4V1V+mxQ7B(2Z3373&8Nm=IAy$TG^$ zdD+}KuTQ0Pg0 z)HgD}N~YT=zF#;gIHy=|zp#n6=!*|B?BAb#gM84cthGk@3Y%KSXi-wyTrDS;T!X1G z&~p1At8WA%K_9swlTLY;0qYZbfmgX}=V2Dh#p^(3*C_@U8A`2jNpss(g}qp1tBq4H z!HlX=iIyirov_~?(~0MXLyV&F-b3+zoRLpZ{?0mSpngrinaAak3Px?vA5Y?+7~qpU z#V7PG^u2(HZd)UCwkB3^2*YmuSqSNTvHHIm_`2W0N%GS z#O}OQLcp3SI=V!fcJ4%+FcrBBwzVm^vV>{dYiLa&#sj)7DG+gM9+MG264PwGB)KuX zL9hdc(tG>;NHf5oLPY_m&-7&-l4F>_VFQf=M|xEZQhbvgJlx^P$^MZw@~Pj;&cp^T z+Vz4g+XT7Pzut{PB8Q?6JowDkV?<6%lprxtr~B|`loD=Cwb!{Hin>{9MZt2KX(-KP zD-Ee*D6eBAAdi};Tf7I30r{|&XYg4VU|h2i)doVWIVmwX4y3Slt*mw}KLtg_LRVl- z|6rY$$zO@BrdAD}hj!kTIIg)1K~oV<0_m`Jnx$i-sGkIHiuYa1aakMZnmnGHE*95- zA|_H_36bDOVYwfopEwG;GSQ}8^|nTHdfOCDFMs8h>}GF{>jpY@bJeHNW7;XdNo=PRH8?b#P_CbS0nG61s>q}Ehh{cK# zY;`ji@mO~|-KyM4y635}1oqo4Um)qUv;n#ux0WiGZ!bgJDpc^TA!8G=<0xp9aXSL+ zOwW9~tNxv?$M2VqYxky*&*Vi<9=8Ohfq@A-@SlW{i>!AsU2-xw*lYTHG`Q1}ig+%l zZ__l0^3DA6%r>m7FGZXANKW^{5iN~lhJE>bHDHiaBn?3XTeoHW{K?8}BrA+UV7T&g zJ&fg&^cRR<5SinO>@A82X&mkc#58FRPV>C1MDxbe+Bo;xVIG*=1Wt=T?%M-(1r7)o zrV;6!r@i~qLy3EWjEl5;7k+2!?Ne~5TEyN-LyKNDybOy!r2nY zWQp0RrXMIoAD{qc#axjm;Rb$bjPyz`8c-&85k75kfrHQ*6fQ#B*lcg8S0nm}zHyz! zF$u@tBseq1n^Bsp-FI~Q`I0aCtmcb0P<~J}hI+>9dVgY1W%hfI1?qY|*CA)9^A%>i z%Lj!bEVAOtNTp?S^68g$Qt8F7LMwu^WbUm7xUA~W6&06f8JOq*Omm+F)NKZK-UMxQ z`Uu^|s3I4xfb;c_WTZ}DV-7W|x6CLO*g^dq&R~THM!kOdhO$w5pOmOB>?H;gIMoSP z@795;siLB8$u+x(V;@{fB1-^rav6b$i)R zy-B}&nH9s=*7H4T;5YzaNAxA`nst|TEn5`y_0C&6A9c|uGc;(r{Q20Xf_ooOzqv>^ zfH)C8EUE_~4Da+cfj2u%xbXr8RYl1PV}@QUtDI%K1d*&MWwzjwb1UcqJHU=f$z)(y zsmKnw=-Qo|W}`7^9voyD&o>}_G;<2wrHYH_wVrIV85G=%C6Ia*(5e*gK!Ky&;J zbl&^Ip?&0tuOCbq?dpq|G`u)b+`+*iXY#*ScUG>G7l?=?Yy#TW582v4-#`N624p}v zo!9CGK5(pb)d~Z0Y;vL92_%N*Ah>mnUcm#(I@PCOh2($so7t2eR5A91Y#tLNCzGo| ze99Hn?T#3ao3k}6u%7M*EpQ}cGC3t+sNXx$-E$b%RETIhtpeU5YALe;0ep-eeNE!yrQF{#n^~ps{ zNJtuqWCRz)(DjR#sR~<*zwZeB6lh4gmXa4T6blEHR+O(*k#fEKg%abr9l?__MgjR^TvxaHyq%xcH^l7+ z)afA=ho6s*)yp%F&LM44+Q-=1Lh5FN>lS!5_89m9LWlmy?7)wDY@lF+JFUTca@ZRas7WkIEueaDu0o}|HEz*Yb=|M^o354*fGi(*X%ETQuSKpD zB~!;AKhJFvtaeMj#DO_3Rf3$Z=*IWETlh1CmQTufi<2P2WXv(pNLj@4_QW;&Hg5PbRDmQ|8dA~)fplM0On09wl930(^0JEIIH}K%mA2NvY z;mZgzjr2j8s=wFF^^6OZk92+}%_5SENqJF{t4Y6sP-N!$=fS$62vB^KW=KIbruvuR zm}`g*Vt&Ao2hi-+&O@#TsPhfq=hd?J-nMO#m9a6R@i3N{i%={IJ0P?M>PpiJs1_7; zk568|%;$3DJvZ@AzOn=uh4*GzyucsttuQEyQ*I5s!ro2VFpP>*uK!J?@MY4E`eqwK21PV2_|Nvj;*nEhuf@CNbms!xum!L?hyiG1;4X`@8Q@L$Xv)dM{oEoZzDA zD;KK%L5=loA@5`WiWf3?>BE`a~C%Km>w0UWpi zze%fU*7kh-K0JsRLbQWAGiKG!uk39DhNaLIJfl5KfC7Vv;FFOUd!{IG9gxp{#lV=e>GKq}Ol%c!%r%-|uLB?jZBi3*cYY!~Zf*`1e!&Sn>bI&Ii5D zC$O)Ou>qgw!nIco+EcB4qu1{r#5W^c6dUL?R&3Y_1yUpEBgc-9WmXflj`lY3Md zY0?svNH5bX26nKK;E&>B^}~9`%4v`_Y9|mPuocYKGf3$4L*=fR1NFN`k@3s|&eGG| z7VF>6Rhob!NKsKXKo(Pt0P+Z;%jwAy2gmO?cp$upv`{pJX6YK@PayN3ao zIuh}CYJ{e1OWzCy9J`}CC9iu7NoazJXe7v%?nd(jgTdurd;MSk;8oRajrTM}OElK~ z*!w#tq%Sxj)Eh*QxKbJ<_Y&H~;HSG3I0Q>@(56dB5`e=GsLpppTh|UBcb)fuSYq$D;7Qfb>#o) zAUUP}$MGBR$MH)M3#t3M#27pC_?sDKMH2(Xv3CNIiswj#=`LFi#8l`TFr096m<=*U zrr>L5+X~djh&D-wUV$S4g$ji<0Qv^Hx-nv6*zJ3A1u#=~!|X5$=Md&fS+mWuku1~{ z;HgZI&|%PUb;4LO>ilcN7njjrsh1Ie@*B8^o0Nq!Z@IV8VS*&uyChd<7KpVH3wemi zevVR3?qY;OQ}R8P*fNAWq>br^k3NSaUn|Rq7PkOgN$iGM##mihs;?Nyf{Yf)b|f~I zn~T|g*hde8)tEMS0p%)H2Y=~E;=-*8t>UgbK2xjTor69Fbsa;RK>bjtR7fJete#?D zk4&ymA2`0rfZU;OCHb4GI|yLp+}UeJtCKD02>Y+#V(Sg|Is}L>+{!5IMzC+eyehoKgt{#v<}p=M}`HZxO#HCTd7_ zHFtdOHEft--Vyl8PXXO|86oTi<<%AB0@Z;J@6EOOWL=!v&M}b}sNiW|yN?>Fz26h~ z&yYSIInXR-yjZ!h7L_R>7#`Y*vqiXx-7F_Ni-4}QOIWmJ7$J^Xoa#oXlxLqB!xOIp zYtQUUw3;K*6*Qsq6vS#o=`GugS4g+K;jAVrqM7A)-;ND@_6TMwRZrk?o!ug_+j*Z%yi3iv_ON+}AlML{F}4{b)M$&9cFHyCOpeEk*^h<Qn2^3^)6R;8qNCWkNdztjCoday zVJiozL#h|z9MgiEyXKvxBs-$!ryF{Eos%1a=F(9tb|qq4l9GWg>M~wmKI0*9JL9Ur zOE@Wd?<9-50P1dE+t!RA_*PE$0>JW3L$y3Uykh0-;{BdO3MB%6xPld7N5&NU`|CMW z%XOc*GkDs+bn0v4{mUg+5RbA$rmB7Q zvktyC`RoC(mnL24=GT!%kgA|W-{`(SkmBD*U;ZhYA0v>Pj4vZ7>)14h-W;eb-xPlD zB`&7%zP@%d0Pcg}TyP#YIbFH24o{p_LBRk}Z};#X$H$WcKA;_*q$y}m{f+OZ0#s~~ z{A8~AOfKVXN^;Ep;!)-f%TX&UAK!Gu>D-DcBi1)#U|@*<{%F)Qe!CrxS?+rU4@925 z9xuza=zzMaKz9B-aQ)Ip3&X04d4Zs(gD>9Dl2nF~NT7lTp=;is3=wk=O5=`m`mXX& zzW9)W8C)ay>tVn{e=?k+Y=_RC6m{|L33GKQm{@VRb=}y z#%j4A3g}h$wWi^A_0~_1Z7-}NykNXOQ8H*ErN5ttJLM{?*bRRMM?bsM4R056#V&-# zKThsvq0x?|P7MyO>vibcIxa^lSKA`Lklvp&UFPrt73Q_rpEFk`*jA1NM5U$uK$#N= z;bWXmk)&1O8J#F-p7Kx4a&RmA+FT$SbInawzP?;z+xC#}LqcS3Y|?db;JVuBTo=|! zS+5(0ihgp?^@Ex(f&~_}dLpJIGFDEFtx)rX2yev@9 z-FL0x6~cd>D7^pz^H@(6F4q%w8I>-G|DQS=8>mtK3sRtH28( zPKCv(%juP`KDx4iQ`bgoM_Ao5b+mzym&0@N4|loQ!L6Ea&oMT*Z`@Nnjzc_WD9`0S zJQu7;nSNLfB#E95p0V-9ZQUN|k56jT+ZsbZn-FZ;SieU2OPt%ZPIo zXdPJ0v&;2H%pS>^yH&>cEsTYHGX+&Rnpc3bwUWqQ-2d^sD(!>_c(|_A8yE&Ak4_~m zeC1vLs*|-6oOf;JHR2b_mJ{bn5uE|kz1y8CYiofa0?z@xWw5nNzD;!!X@^%a-O>f7 zLar=`ZBnPD0^`|vWaxMX$U~Y9_itBp&K)u+48F&YYs{N3Oa@{en6~JjYP+vLmQj0t z(lEL}{yI*~qT^(<=J#Ff0uRU3LZ4*SsC@rjg9J2^V00IStp*iA!Qv1(Lm?Csce0C< z(JBJVdfd=tO>mF;#+L_RQYF}34&==H7Rmj{XeZfPU}K@g(z%TIP2$TGGK~fEZmX8Z z$VEI>coz{b@NjSPvroRy$3*7nzbBi&yYYQkXfyTk4!fb-U^mM>C7h^aj;?pjCM7cZkxR}V%<^I7;qArztR zSadF0Tg*ZJQgZfm2p@y|#+*p_6B^AOmWnqp`flgtakD(Vt-!U6IFzkkmc~5>;rNDK zEgCC_uC3W9_4 zGAjeXfCW>SifFOd3B>$sZ!WrrZ?m-t(l>8{ z5?&GdIy@R5Hz~Cdos*LT5wX5-Jxp+VrTX4JRTZ|rkc#6^&Dw8nqdK0Z_-lpOhmhCF zQVfIh`mMrYu{3N+`-1H6_;U4fqB2Nt*7;FgcZ8)_A|?bF4I(C#%8z#hjT%ZhgX0-CyLArbwAuTKmX)`;{)nmBP@YU4m z`B+^oC2S@|cU3+DsGOQK#r@5+)cz?e__DuwU!ZbTZV<*r%Vvkau^WqNL*}t(aBLYcdx~^0uDJK+w=aK}U?khE8L|eyajKTs za9NnzqTv5UVer9bPUAq?CS5nFM{l^6U+>r7_3ZQG(=|0!pSJ8cbeVe4^Db9(z=y!3 z>Gy(yx`75kmriw4?YR`XN$!~IdpWL&#@>pYsi>Za?p%(hsE)CR<%hbO=aivll4mH> z<$Q|62OBLm(PpGk@Ko=KV|U08E^XROz5C)X{_&kM&`Sv~l?AM@JY zl1$IR5Go-FpN(VKHL`Dgp<$VeG&-M;4x2ebuc%`x{mOJUs73_CY<49GPw6O8oG@{F zsX?q7FqNt_T0Fky@v>?wWQ*6urXXM)#vAuKhwP!(giw~znqyqGpcZ;4pAz-2PnI{)}4+wWUpFNYVS<7tOAW zd7`l#ALtyC^TnNoE_x|Z>uJ|)DV^rh-DeB4c>5)N;hqI4dYUk&St)vYwox3%rM$XX zTXV`Fwl|>qbG4_j)%cvzaGI9OQlho-^k|D4%eu|C0j1)}gSIf%HMTw1DSa*rla80{ zY|B_SA|wGA`Z7A!*V_9oxpvYk-lJlU+39Mpe4P4Dn1XLOYT`(la*m@Fjg-^u8fq_L zCa%SzsJn*axy7%CefP!q=NCJOG7&ZSA~|BBuR{>dsVfotUfQ$Wo56W+CVrBOG%}xG zwbS{bv*PyqSKdqS6r$6-4VcNS7!q&3%#tfNlA0y@erDblaK=slUL>q%zTi z>vP)|_meDytLi(6rhC>0!19AJuGd0omHK{Qya#&rz3SO?tDr=Oo3E>z)GF(XpFi0= zkAFy`-foP~spXXNu^rm(;y?CT{PY7yl{ri+iero_QQ{U@sF*~kGZ&h(;_Hd_5hh><%D^HS2bs)cX9Qymsu3i zl#E>&RSq(VG*Ig8xD+Xw&BeKPEJgB(g&8wVR%Ak5Pqus#{20|+8~1_MF>zyETyi;` zRtJxN_P7$~juF8GW$_4<&#PqStU(+0>cr$af6P>kb_}Iae{DD)43jm)JH`w4nz2emD z#u6a8{=iAml|JjG-n*PMA`dp9mwD+6ca25LHa}HSD{-FJy6*s1N>SLr?DhE5 zF~zXlr6-s3Op3R)jMHf^dD-}>GLy2aKG<5UPF_Zk!KK)-F5`ZeCUcs|3Ejetk&m(m zZ7Aogzr39onxfN0P{KS(XCXSFL}FOsIQP1c@Z{>MH*-%>;%@)qyxso&lF&DCaDBb( zMR%6Ws^l`YHTb1N4UTnhj5Mk*?ke`f8<9DZXW8wYD zWo|5`F}^Wj(h^L+-AHBB8Bf^j=aY~_tUhg#C_#}e;m0Pq2){|WLSp$01QNgUm%wNl zqzYm8gSLgln@sQFns%{1>g1@$sI$+WZdnpnZ^ zxFlx_7hn>B@&PsCmudz{-&$o{ zmZ>$g6SN=s$ZGE-u1S<0tg7U9AumFRY)2ZGzy@A#G85WLT6l}5oz<^)W2stjJxso) zw}nPw^=pB@VG0bENUT-0rQ)4Qg2O~pysT-)?2G0SjNj$K`kHSx&Eo3!cb<*4+fnK5 zt1Y_Xlp1$NF;;Ts^%geGBbSfx;svxq<~57*vE=@dmZZVFjwk*(YG-WM)`mN_ncIrF z&>tR;LO-989k!UBgx<8>OYL2Go%7+`*uW>M1XFjkRSVt4^;`B@32;4~tH{=(Ik`J| z`mUcgW?&ni(|Z{^Tdk-)M=!r^n;j&&v5X3hkCiEJn|ncGCJ;@9?R8RP9ZVN{~ z9)5ID%{3#zQ*cuKj~yyJG06RiOEL0hr#q;s1g}3b;Z#7Wq%FzmMjbX=z~V)npw*{c ztx8&wnh6Vm__(q#rJ`4!wGm>7>H&$W1#*C%R&ejG?eGw)~g->WO92_)~?bs*v~7;w2QCaxzyP1#C1bH zK6rl4b#K-u-Qg57(-eGOda3<2hTbbun?yrh6l#NL_z0L%>8Arqy>!QKosWkZHeFmA z46y1_H^U^j6HZ|7O!p?C`)7ur1;EJ7+OZz}*mzm|G~-4$Jd&l$o!Z5Kifo;8*T@&j zQ+)at^`pCzb*_B-2zYIuxZ^KRwZiVpOy+qRk=(DGd zaSM!{rGAeKvoNlyvM5Zl)ON*Ol)@dooe()6nNt6BBZ^KQJG8|N_7fq;@>*gDPb((Y zOqM2V@8^b@&((w0AagX>K&MV#4yyXNZ6nQVfsh>Z&JVr!z&Xp?EW)|^qm&W)*BfJ) zpb?%(-y#Qxd0AArO_wj@a+CpMY)Ma4s&ZRc|^aA*9rMW6|L2`@q>- z&MFnk+8dD@z9~}lF0sxu2T7sTlow_r*OKzmMGiL8QJv+XlUdgFaXVdsF>Tn!XS>6L zy0tlCYz=al(V1cg6CwYD

~o6MX&kH*6r?e(*)+rhW8$sG!6q1G{6&z0DqW;I;gZN`(RxjnGYAbUxU*_)etVqyB6p!^PqduhN3I(@Aw32r z`-RPKrRWEDmuOFhj+!Y=t477|EX2flf|fU1D>8%TL{+?$dsVbeQ#W5N%HdhhC7l}_ zEL!Gyxv#0{ZA%>a#CDr9Mry3_qsr)1Q?KtVtgFfWAlUF2lrjzFX9g`lTE5w1TKT-W z$qsToev&d<(t_FIgopry5!&pjMPH$t;5rc$GOaE)W>Ke4)Eu0c7HNDdLg2j@yux>8*?cQr& zO^3mit5L#Co2|D$4(PGffO8uw6yJhk?eiMTnCd$RL&8@5T9>cbb3Vx@e+U~?mgIq5+wAeUB$@&4}!tLk- z%RGX%51YV9&NMCSR;!^n9y*;LX-6N;OBGu5}SudD%|qp%C>cePgNndwv!7 z@4l#enFZP0HkQa&DqoJlK}VybBy%4_3>UM?{MuPUb>7 z&a8bG!$-GA7{?v}1H7z0Pojm~E@%kcM@~PlPm3V>^rb>Ume+vNy2=IJAr75EZyVkD zo57ra_~W|R^pBtRi}dH8Prvv-9(LCsQ6TcLzsZ|^zR|UwN==5rP~=9Tq8p20d-t~g z#(OqX^U&YiDIPp}u|t?vGg#S}<98eKuct`ky&*Jz|9lUF0r4V`qx1J(+eX3RMk&*j(4KOwshT~y%f5=SNFlby2%{_z_#-Azav{6zbkFz`>n z@1o?M!`Q%S>Ai&a!%Sv-2v3>Je+dQtb6Y^VJ_EV|=qq&Et*W9II20Gw9j;s>8&MI~otooU3Ge}rLgd+#q8(tQ zA(&aYPBFY51?m4^ndZO!Q&?jBlVKasjz5O( z1l;(_Dmv8T1Q}1G-tx6LX!^?M%O)x;>=G&pwRsJ79P_c+@ch=is;+GMY3w2S48f{Q4SVjZm5sV3{@h(8|hb z>9gl&@V)SAw1hF#bERQ<+T8tbzx1Ckv7VFgqvIrgTmWbn2F(+X?n2F_X0EYp7U(g@ z=()dUVN&`*FwTU;fzr%-4+lUdyFF#8bi*OGL_6KE^1)Iyf}5mtCE6n<)kd%~5u z+8rGewNU$gq5P{kZ<~8mxP1#NHD5;3!oreShrUO(C5TdQ$)eD_h3zV{CBjI%y&1}Q zj&Y6!7#!$<#ZMSZnYks^wy)e6r^Xbl>?i;{7k?5L>Vbe!{*-e=VrA-jT-m zcZj9fT_BRbTL5Z$bv(1s_h3#bdNwPtJKPRMKTWZDbs_9&y{gAEVEA;}CLFMuDx}p6 zwzX4uyXMWxJ7-kD_GBMs2P_3Uv2;Y#PK(NZdB-W+4=Ny@0P6F60PDN%vEyf&#cW6` z<;$0F7obb0eh`$njBjBRJigcn|8~*Y)9og_PgR0BP|fTZaAyG)kTvmbia=>c4cNyV z!1U<4@2mn%F-x48OtDV5S$4^oj_P6QpzkDY0NZE5DMbtci#`J^xPxN&3SBlcoYNtQ zn5{jalXYhS5$ePJr*tp&NlrgOPv&Yxp6y93<}T${jAVjF4H=mzs2N#x!Oml~!?&4b zHOYLVsJ~6^m=zJ~|LTz_Y>Ddkl@C^$?V4Y!q15u{ z&U=t(DRaaJMBXPW^5F@M-*_*%^*(@wStY|D&ct;+dcXOd2OiAm?EsJ}EcJ9&uowZt znQ33zSLtjbVetOh`2-43lxT2Iv#=&>U|cG48!F15eFQ4DL(k|Zxpi zUW|2lo{dGC0@g^R$GUMD$38UbGhAODW`H(|_nrDX9#gJx;Fw|oL5}} zdqe~~D|1Jiz*MS2#Pi7&Fc+bwIk8b)#ZYFqVdKTslF$Nm=}~UN_T44wuNhg^tPbma_gf2`woL5N<5C7 zmPjeJc#LM+fBMd)g~;_btk^JHt^Kp?8jlVXCA^qpR0v~|5+XukKU#jhRH$k1HQqow zxwvAe9ZHs!Kfx89T@bPd?pn@$%Byp%E(aI&Yka*@mk~Py{@{=WRa8rIoEP+S;T%B= zLBUik1M53te;9GOx$~gGpIpZsM@=5^u5qtWnqTe)PlAFl=Dn@P3Mqk#EzOezleQ|q zbN%6LP@DwRJ0Hrn!00!R&d358@hRlA?w<5)jeidckKK#npB`;NkoN;|1+G6Y*nfnu z|M-ibifn%z1hmn2_}p0hxL?pDKYPuretS<=>rTaBQ#Vv(a!&xfz*sh*b+BPsF}`mnUrPSko4e`IiI z8CRRo9x>YyVyKS&4n(DSUoi2fO?;F*hBgR2wBqE zGTM@;_QAq&Stzv>0wRQm!~6w;$~PN)b2HM(yX#05LV)DeWx?EKJ!l2np~UgHL9s;s z`XO7Q)U0S;_Fkpm9+JY_;gIya{^O98B_>)E?~JvlYrL=GVxW29Wv6(DBDVq4e43sg zYez$~_pFqKY-POq4e1~X(CVirhbH!s;b`=ZJLSjRcR3{9gOcQhquG{u;%R&bW9HeY z8yis1s}51uIWUiM#1JTM=UN=0<2~*cbm=ZIjXc~L7819qlhtow1$=uMnO@#7Iop!; z`iXN%PMtZ*WTV!fEo#zi8gU}gSb7tC&as_tKT-=>D6(Wsy6YBl7OR&N_D~Gkuipv3 zB*j`O=;`12UKY=fXjHsWY+{RjHlKko(Lj2|pc#Ci)-@|_6h6BA!*r=Zys z^VVI5!r3(NucMv5*X4yEwTly7$(_yM)%fz>W1bP4wAX=ITzaCgfSj!h!!Hjiku#|I z4tkg}WnPt3rR*3f_@u6pCjrNp@azDq1Xg*$CTc66Q<+~8h-olSAjIOXIBv6|BL+rO zU&eo*H%_u3xgT7Bq#3Ob93FcxQr}B#3sF(#_ z%}6gO35=5S`8XH#1&=Ylpj|q zG7%FCM?W^fzG4LzX+_6K@fV*xK+$xzj7>&6Wvh~bv+<-o2M?H>Vq>lyBoHr^TdE~q zAzkLuCVfFXK#4+a$BNOXW*A2i;tBDq!<8#nklx$l(meMxK*HjN9wJ(>+BzOYfV87AGEVH_ zJr?TxiZ1!RluVPpTgppSZnC4{3#g{E2VF#v1&S^uhO4)Xozs*v3e_;Gvn1b}eYdZ+ z*|jF5GA6*1A*tDmIqN~f#V`RV_K#lpw$>+1r-=>m1vd<#k&owr^m&D4e}P1@D>sgd zK=Ey|LL5{~#eN%*TkJ#Tb??u4O^XW`-Y*%Lu5|^moLWFHJOG)O$6~*uKri}{6ypF7 zMYfzvIf`Vd|NhyL?yrXE<&E{W7$_9quFR@4QY4PYagAib*6Mt_u{1kaU`l5K+;j=M z&(BWl*R+!~FHVmyY#Y6l!;m(js~Phdtex;eaan!wx>rS7_^dEW#UnSukMPk2uYyNT z9VJ7~P_bqO)00Y`6H{_$fgPErMz~W84ysHHgECwW%0~GGiH~aJQCUogE7e`7RrWM= zKWKfKpD1Ta`tr`3>NMgqo4P>A@V1Oeq_tosUqJ<61^c{V$FbEm)_4>u&*v#fGF@-U zI9jCQmhDPfxU65jP3L=GwzNLdsK`%yF>c&;wL4>nnj1@A!d6x~{gc6lhD$S_^P z+}*|_<&z)RdlsAUETBpP&5oOIxeH_7)iz;cln}PW`P<;<^^Oxr-XVMBNA$yqZMlF= z*yQ2HRDo5a!mMV{;4|~(i2`F;$(2RLokE|py%q_!#zg5#C-u8hLG6+8A1AllI-%UR zG3XBin1{eLMw=t%ynm9eAo`0PF=*Y+Y&K@2RK@s@7_T0ZJ-0Eey zVSmtQW}5ug9=3t$7EQa?-VELEwNSfm$>o&O@U?DwV!4x39kW``tVwIoEWY>wz31gf zqR*Ye%P9NgwZeGW5bM4eDIcD-t6OM+?s%6)5F+_&`FVWdXPM$VnQkjDUBcfP2|c1| zVL+K5)oV?ba}Y(9?;+gSefh{%vaB8}Sqe9c4@)~EaYE(zpXU-`Z!^M~ce&WHjt{xO zbhzM))rx-3?9<;}{U4q}{r0;y^Uu3K#+}9F>ESy<-M7Hpxmv(^)~EEo9X&0n=Jmd; zOOTo!-cTtb!Gpu4d8(}3Qz+3(d##?d_>Sv~T58MV(uEca{>)6^MMxFznK>Twpm5V8 z2KwaUOEW8}y;*K7n#wZ8TIr=KvOPux!zOq2;N)qv#N8p|h@3l{k*YGH%|r{)e@yeY z6iU)+x+Pc5>VRucQ~nIq-#nhrAa|risB+rzhS24a0gCSqnMpgfwfxSZ>~VPC!OMeH z{+RhZdC3>oV-C)<8Oj!qYJeE9`han?o&b#tD?N>`?V2s#CtF?#*d!T^~$?Of{ z^2}Jhr$o&MmSA}^psBgAkfl6=^mhv=-V_0;v^y2M3Gwy7i<(`+mVGZL7EUhoxv{9# zz(K2L!Spb*m7#EOCGqZKbXW_=C1#JiyUu@l^Z>=2^&(MC&8ekJdBhk}${hMKpcVrg zo?j1mrhcAE?m{|cb9I|G3025llDetcT~mQVXe;JIJMozkSmvM$-@=cIfxG&`y@f6| zNFdWDhC`u!_t0;V+5Zs)YLcQ~<5_>psN{RukE?hPMA{I0md-#9BJ_Z}eT~x3lPu)} z!JWX!zQKp}z)`CPG=v@h0;M_PXZ!sjT#mHJF;)~1_jbY68yHu@V!$rcU?Y{v1g#@k z*TU_MhrkJ|2Grm{oY^}s*Dc$O7~eY!ZO)ocV3y20aef9NF15`-_g&Nnkb)oCsW}qL zD&mi($!*oDOq})tzp}tJ@EI)TkDh}Tq6oZz>i;I2!0e$e0bg|&nt<+?x&@X+?r=OM z55n#9W8gpcHkN6paG1)VT(PpdnLT1dZ{6{Vzw6~qbv>W(++WL2l7{!Dt>~TuT6K)T z*d6SGkiQCg)+&l#y1L~o3t@!kaFG8ZECNv@b3)PQEzHjEBIZ};oGsgC>%3WXqJGCQ z9%=Zo^Tq#Z=T|L*EzmPkkyJ0;K?E4l1JtOk&0KI5+7{9w$=AZ|;wQx9Ebl7X$O%HC zmL>K%P|hTqbXw(U=#HPbtt%Te0hxV?pW zIh?@0#mD5D2R!8w8S^QN9>U9lb7! zOeQiU(FVr&cSyupkM_!4d=*jyN9=fIjoJ;S9V!>!%LA#drwHA4V@B$80&>Z-jyn+U zBy6Y2VtSN!2URgrA|a`%S5v__AIYFOVt@L3e&XSGV0l{pfZd^7gP1rmEWDPvhj0qV z{u?e%d=5QC7x!ssHNV<;7jD;VS5jd805+VvvrE1mbc`e=>H-I~k-d=9J?>ZsA0HYH z(Nj3eyt#B&KkHfS4I&?BEQG)Dx=)xnG;=#!dv}bG0htlQ8hXO8)@Il_C;(CZ-Yw!M zFr1pWi4hn<)0dLGMMgiZtw$) z`Gm0rt-N`QqP*6LV}(HAoVe_}KQJp;Z-t_TPnOz|HJ=M~JIxN20C8Z=mSFAnxZuMG z?sO@L#(gJU@X|eKiSZb=W_=}7v0cMH(w8cT;6@4)2XCzq4(@ZPR5;njV3w-fb;!GaAsE|T zes0H+G}itpV5%dHf^k}IEcS-Ky+PNxkl>B^_>bU?!n2hR;2RF4BMmmUfxqlA+>F8! zi>!c)D+s??gIlbVX4lz0*~|wJ$Gpr^u5*SavXwdmfcs9ERBM@z6m3J>()-r3JMxM_ zzh=8)FJcx0%Prl__>ZH?jEg-v8hMzi@vt=n9DlXu)v;KtNqoFUOpuw(notfZ*F18u z_{`ScQe6N#-lkOza_&&R=@F`o)dwr4g95u#oa3(!?3Xq3DnZ;urBDwQxX^3qi<_!y} z^{qEo3-hjrkerNH9o6K!q<`bcUKcZ_UL};rd)49>c^%TW6?Drx4xP`Pvgh-wvFW!q zpR^O=&(vwDO%UAncn@7^`-ToYjSHBp@}wMQc@SsygS zp&_G%74dQg^~xNwJE>Z=wD;cVXrdWU>Bk@0j&0uB{6nNSQNe767M;cc38x2udOi~0 zd9c$|DdCXTkt=aA2yGib=*|L4m+Oer;$TJ32}+5gJB!Sd9kiNnYURgr?uJOCf6K$y zz5F9hqTBN${lzi6v&?>_H^&OOLFO$A!j*j*Gg3i);u`J#mS;{KWmYV5hLgM7lwOvy z0Ik@K-cWncn$!49i3uoT%@S{I_TXF_Ymx}Q5Sl#;h**BFG@(>n!NFYdl^I6y)jmG8 zn!N1F1p!T_!#7^P4&zyysWmCgy6c|F^gK_ik&qELEbj4y$-DIy|97^-qRiUEm)|6# zEgIxEoKB^yt<9H1Mlye1?8<+Emj6`|pd=!&xw_w9u(|G*&D4?Gv1)<9rZ285%ffsd z<8~&8ek^>r@1Ls%|BlwUPBFqs()|2?_@vzgELC;G?`z*@2*^W(dH&COtLF9-=E^UC zeaou{N_zeR)XKk|d;gkC+>k?nLy4auM4;R6Dh+>*ovZUPWaGa(_Gi|%=VaGkqx$co z^K(vn=K0(z7xMc-2QlV{)($!^%L@#iOH^toI*Kb zqPpjHzazgZkpKMDN3^>Mr@V9=%(p!dh&kB!A5O9#pGcgh={Y@(DO5xgMi#4|GyYZ^ z`{!AuTf_8&D3AW%fq2bybKiNIpL^%uR?t6Qa9uxf%DgLAM?4GBRmEiW0ME&7OzjQ5 z@D;n?)RO=DLuk|D3(}+>4^$s#orSIFN2vmi_IoZraHr ze-U&3^|JeyMdtx4g`P2Z6Z9os2_QP+PejvNS^h5BjHCbZ zQ+|HMzdpr{TmLkL=)b4%ueH7V?dUIK&-n&qJPG4YZsCKTI^%`Vh(R z40y~Kt%W1HEXZ3rQ8-ievU2NtRu#)eA3uCJ(|m|U6PwWW6@1`X5EH(c2F$u3;AX}MU!JJ*zS_M&9pEQsdlkv4zT028txD(76v z7=E$3U}8+J_ae4P3pN6g(HN6ekiV~CJD z9LivQNc%bGa3D^7v8kK|$B@V14e>5axCfh5S~tCPS->2dZ`vZeZ)fC76$DyAs^GI{PpWem;LAV$=#HHNihxQZ zQW}@iI}?J?XbyqYx4URPcHEHyMa>3jDTKSUKmp{K4HybB#3)x59A>-+sXNV1&lkhQ zmySQT^WT_%`Wk5o8?PI?F6@k$vIPpCKk;AR3-d}pPUbpcvyCechQ7nJRD3^Oe<+@9 zvM8xhP*`?hxx3l8=D3UlD#R zrXTR)W>HhdZ+VhP`8QQWO()Ic{Fl7)0E_Ql-~yFYgRea!4Ze zJk}5^hG}}qt;Ee<#uES3qMlU$){Aj){2E{3P0t!>^&tc)Tfz3(-X9&E#jZ_`&gI%a zSoM586oz51MLMA>*WLq#dChcgkM};iY4ResS~CuG>)zRj;byUoI6T$t5@o6jTBbx6 z%_4YA2MO##qdsO)Y8aen+`G_xSY^uHLgwf%)}~TE52EL%OV-NsYc;QC4!Y?%qU6`kmyr?sPguVu_X zSj=Dk(N+1C4zxfBF)JSyj-ugf{{ql|g!T$hn7GupIUJ2A1`W1c&p+2x-SZ_cDal?d zc9#bqVqxcR()OYwaB2E!kG$lc!=x=G8d1M6?XuxmYy*m49HtXJWYX1y^l@Dxy>M8& zfYoy;o?q*3dsoJzdm1R2{ylW`;w0YM4av$xN+&ufdDCf*F1&7!o9?04q02CgI<5N9rbTmZc7Jr;^GzuTq>u?*|7?765?e=$kC~-7L#M%MwHf0`C13IVo_zzE$d^BdLS!jGm#A9 zKkGDBxebh6u?cnU!+WwPIyIl{~^aWdHN+< z$|rH~S{y4bK%=fKQC7Y4aI-f{VXv)iAD9@i0RhR(bFBAmS<@`|xklQhx9O`}-0w+P zkWQ*>ggXPX8f6Wzh;R>I#fwjV*%8@S`kxys*wq>*;wx1fDOz7eb>isP_=5FgpAR4G ze)*Ay>sN?a{L{apPH*1~P129JllOKxJglA@0Xu3Jg)Ody>79A(w#|93C76J)Q4H~M z7h5icvK_?{nCX!IFt>)GC&p+OViYjfBR*9ywQS#BB+!X*!FgH^d=AL*@&p^i4t2BA z$$*%3%Hr1`CI4Cya3a5)E~OJgTKBR=u71b$0G=lr$#*1f@la$Tp4LFJ#2$3kH5EJx zN%r94SGS~z-|2zM9MIgWGX$M!4+H0V>CW)&#lcNXY%Ak0!$IPMK^oYqx2lmnRQ5&*8nER%So+GOd_hG+l>FW$n z)dtj*TFxca8DUn9)lf`JwCI>P%RR}7b!eCV$W85qgx2@+7CjO@mcn{k$(51u;?P)p ztmbJCF#3_2Owr9k>p19wSny=yEr6XS!SB;Kxh{p?ieG30MZ3Z zyAQ8$jW$yz1GZ-LtXCpu)Ms7pw`zGZvEjQn2W>1(Z!Z(3T)(P*lbDr1fpauxBAgJc zXlKT8uT(~qPSJ@G=rq5UZxwNl{n0-5|6%Vv!pZjaOPySC%8PxhS`p<$O;R*Nk&4l4P=_;t|5iA?4<8F0jtjN% zyUmQI?gRSbGB3Em3v;5?>?DdzW@>N4XX&bg^MR6B-#W&Hmu`({;A`GYejUoV5Z7|e zD&n><(WMx&`3Vq9`>RS{YA@}!UT4Jkmc#Eo)6U+Q$P^zBFk{Z8YR%C-{`v^SwAOQwQdEp zZcy(b-}k8oA2uMz&{G_Nk_Y6I1BXUdX#U&#MV_h!L2_=B8WknTfrwKLT*_g(MCbT@ zLW(0G4C&uhX$n<<2DUiivUd6V0RkzMU4yq^Ye(0quoFjN!_U}Fa$_Y91de#{sW1AO zCGC~sYAFr&PAzmgYWjT#pdwPjrK3e9A>q725|ywyIGt<2MIHV?AF+!a%~-s|rYLyq zerd25nOmii+Z1jvJNHP*PXr-dWid`T26-8)*E)os1zhIrVgY>iCtC^U{xHKJ0XU&{ z6;Mjs78f=BTV9oztk>4H;q&V7)(lnKxq(IWncYmx{s>!P0ocxi0*)Yg44NGH?>ElL z7uNso`>MkRG$ksgPUqMRl-%mS8m@~4WbXAESYdW+=^U8*?7yX&KtByzFQ2ZdU)E;P zaQbeZccV><^x$;6(OF#67cfp7^8i}oO?HbTRyx{DPzKZ&x*%lrNCc%t0;Jfu0xB-4 zf~iIJLDqq6w(_JW?D6yKKtZd%vYFfZ zenHw>>b&~;W}9$qp{n#=r^2mA=#_cV?_O+!eU2kW@&s?pq#k&SG~@%;zO%w5)zT4E z##af&ylOT1@1lC^{fzDNmsDattcJF{(rno1-|BTfzfi3S5~yXRi6k-WTeO+(O2(=B z;&)`zhz;{f`IU@Bup!`BX{-#1i&Z}evUu?pyeb91wRH;V*=d(7RES9cX0_M&iu}%^ zV=&s>O-bY{Da1`nLDLra%Z6Fy&2sleJmdnKaWdiQeH!nNi;<`8KE$VjErlQV?D~*B zw9RxTF308myZNmzRizbP6MVuzo!osRwPcQgqO0Gs)+r8Ie7)c5ZGrCy{$nOQ2j!bu z!o(F?H$jw0_nRLtm00FV(|ee)%&pkNMKRwozTUTM*GN(4m9>#il5@N*yOcloOu@?`A4Z^+ds5%U}Yo>2v%3;kCQ zFp(oJ)~608KaWyX82xeaVA0r})-O#N5eOUixRj(ly*Ke@tw0&7JET7Q>N9pt+WCs? zcAY?M$(f9kVfDqWN1F^IV{r;B5Ta5_sd^=o`I?ZD>CA6f(`sLPDeFMeOh@C^lg2yb3lU0o95p}LF zOHdEZK3>v^bC|V{OIE?N3i_R-I%XVo88&)90m`b;&V@5MgYb&Xr%dmf9&1hqiKId@ z=NO=Z@XaaOmY;sU4vS@MurLzD5Jx*>kMqYaoJr}yN0@1sD$=Ii!0Tm~9lPLfaYSpK zM!W->nOax#r|E@8tG)Njj|s{4mGl*DVe{Q2@BEUgzc!`cIc2S?*q3G{{_3s{mEw_& z$mb#gf4X`uoGH~vlU}aSm*MdaN1R8+>s~w)r(b9{ljK+%Ty7&gDu;S$arHPdSr|#vd#;~D)sEJfECY+vhyer2AmCwa1R`k zyC6oI%Ad#Bw#%~-Di;el6tPE&%d_8c9Lc&!u@wJW!K}ML#TD97t@ot@*MOnh8}qoe z3@JK^o2guhHB6J4A>7*5=c22|Biv`UV>XAeZdcFn(n(V8S&y?Kklp%g0lvYgd5gb# zdp`cFby!lXg+JLVi1V0Sv{n5I*Guj`r-Od4^YX>h7$RUp(kiGWE}54&4_i5pi)3*n z{m9sve3?(qLBBAohO)PmPR2M7s(&ux+HZ9NNn5F~9?wE8p~K;t#|!tblr;-k&%F(| zH_&yt*RHmfYkID5>tiUp)zD`cOJ0F-Ga?Jk8Q~UDvIl*(Xi=v9s~Jpg!SWhPQ9Oubl>$*KD__%U9^fmox;99 z7lZg@6!!zjCAqeixnytB-jx1KKx)*dWk6oaoH;7?NKnN#Ytr1L&!bl2fW>IF0PwN) z9#m)q5{9=6x59@3XSdCExwk_!da>ZiE{Tn_aktOb3+mRRO^2QE1_4;JyW;&YWUANM zQJ0%rz?qKdr5Y;>M6S3GZP+-)`b-440zx4fzCL0zRpAP&zso1P5^VA<&-ekTbyUpw z1%_>}hjMGA*f$=g;@+Ba1zGg6N=3v-GZTmK5#{r>t;LwVQXRK4yn#U>zBFBa zI9F?k8|P5|TiXrwd$9~k0$t<5pa$u`j|D!O8~yu6+pEbBpwpt7H_xOZo}Kk_)s%s% zm8BX${FppbB{C#0ju&&vfW4(K(U`%CR3o5>i0~iF-D=lJ9@&V~S}1Yi7swiQVT@QU zeCENdhV=}J@CAQsWR;@*N(W`iJbV0j|3sZ@S^d27mJiWlBR5^&h-a^Gq558SuHMo| zLs?SHVuaJ_iABZ0v;5`n6}`vS%-2BUgLJI~@|esrK%8$yue=+kI|~5q%LCHelSm!+ zVOtW_hnyCsS&ZL685neVo6<(elGmItUd=lVF6VCrgYvgPXGCFWuAPTj-!JBLDztoy zq&mKMvAHs}5<{t@U@O=A1nEtViRsNirH2EHiy2*#p=(^hR*yc?%)4zrmOw1nM3XcL zE(Hyisk)#`vKx#UoR>Ldl^atupZ+b`>`JLiW;@LR>YoSz$_j7&V0tfzSw8vVgUiK8 z+2QnSy`L^|FPbM-)Q2?lz{GONo6r))!60yYW(wuxQ zTviN3`nJM-28R!@Fi>VZy^oT$hC>1*oW7IniwBQ(aBIGN!y@ISk1RNtT(ug)DpdY{ zzsBgiIk~|JCRcHlnmP9k_rY#7t7i%TQX2+7)QXlUgZn_-QGKXE1Xs8PizfgnzVxex z(SUoZ$1iGXR(>UwT<)qpb7?87w)-gB`b-4yetw+W`CcqKF4TPK^ycf}8u3 zv%&;?&dhs>6sLjT7%!yrJ)MnUq)LzlNX!c%ZiFu&=KNdq3Z_;B|Dt~EjsN$}UBh>^ zoeB=@Di5DVJ#ZkrCVdgbA~o>%=>^v>#<4zxU%}O5FkxY1@bSUDj}apBBwzv%0~-C> z&rDCbLsoAqtJ>rer$ni)tM>r~xK!Gz#%438(i3d$-H-eJwWPphAvTpo>Pv&mQCKsp zU{LR+%JVWpn77r-t!|ye1Ue*gt7-`4;6){qZ2j;?R`u8RR^=VPnn@=D1eIczH;0(S zvU+)q7j3};T{8P4*qw7R_MzMiZV}2%mD$jZzLk1%R70jtBDvRJu_n7lJxZU52LMw} zzGdI~u+p&R!9*to)&faUrsYPZ@s9#Kd1|5AA*&S64)LME7#>(qshd77pHqigK$6(o zvE~j~+$&oQMNt(dR2&3ijd~<`rz4p$z0*0;8fUXAf*_=JBRH3&6WuNQu2v*|&?m8> zLS@q|82xBH<9KbnUSFwmE6~*pTVzzsuY<=Yl>J`D4d2R8M-pCu&;%$?1WM?`cXO`Q z0v(Xt`!3O?0uIqO;eay6n3xgQ0Si<9MmT5g@*C>->~X=%C6Zp>wK*t-xzoo@iWax* zzB0{vJ^1%ZL1)fct~8$>I;m7jvR&7!O!&g|c~6>InQA4>sFb8_X|(fw*Zd3Y`Xh6{ zmqOTA&Kti^0QD~Y93OEPPl!3aWMrCba#nr;X?BxV@=3w?AwX7q=ZE&G63b0eZv5b6 zEI99_94km5RKWb^61!I7f))LN2}()RhJ50?tIU_edJdTR5A6!-hAB#YhI!w!G&cZ? z3*GSGv?DN&X%o3fD}{)FOS~FCzJa4O@A(P|i(*AaY4@{iD^G$u*=b-IfF|a1>HB|y zRRAn!C>)tbj5nFR9%7DL1i%Busdl`uy3SPexh79 zTqHUcVD(!y8bbX!M%b_76x4FkEwfsC%Gbp)>-FxQe&nWQ?I3^gD2#!Y@z@* z2DzoI)U;o^`w|q2<=RZ^&&4@)OXQb7(>IZbl*}z))DD@wlA!qJ>x1j6duUt2FC*TK ztL8m<*f;Id8ws5f;|fmZUrD1~18UNLj3RfqK%>a!PlMoP^E!el|FG)_)XSGgTv9pi z%*hfka5>EE$&Y5Ed`la8+3jDj_hV4Kt8t4SaCq~c>sIyjao=GbE z8O?}4qM5uN!pa;s_bQ}>`xo7_dk1M_r(K@vG~)#``S9$x#WfjAwy9VAPNWvp7he^VWW3G=gP2l4o;7e=Fwa=n zlPBWv<;7EsRwbM~4^ydA1#nQC}fiJpVXhvIU7lN3`~q%D2C}DsE8W zV@fM6657ap{Cw=}kvpxcm9*Y7E>_U8K+TTGm>g3c+OH9KQIfK}i+S9f#+`4kn}*>> z1{y@teO7w5KqZ?}C~6a68ZV z6KJJ5k3W>pj<291)=k`Q22509OC;Y+Mhfjt+EI~7ux~B1W}jDw;ajp25S2B2$c8Ob zEiRuQ@8=)R$;XK$>$wT)ld|B_iCd8mNBH1Ng&2|EYEI#8~ zB4$}bJ{?!=bm{g~cgtfzZ^H%RbD9U))pxGSG*E~7uh{Nx$ycjOoRF3^Q}F5902#3S zv+q3Fxr9K9Jy!JIZuoj-TUjPHc@Xc_Z$O;cvbZ1gKs4Bwddy)UyCEm|NEcXeF?#88 zK4|O52p%lNTh=-@Gu4Nkh5WC-gkrwGqTDksanT)S#rGkPbmsVIQJ^bI%`V?mpmuiA zD2=pw8ky69Xe9*do&qmn4r2icds}(>H*I0*K~fBf^U*BFU$8h{{OkUg+h4GO959pk zaP>EcOLDDM4E^Y56K3z~@N2O1WZ=F0jsNvyo9NIpD^@ zIsG1_bZ@GHIc$A!W3iy*8QA5hn4#5l7R>v5c^i1EMXLu7f!)BwbG3P;Id+RGQb35* zW~k(5hT>h&i9YIHI`W}c8yc}z*c=RaYE>mXAacD{EvSz%zxU=RV4m|^Fe0MdW{rEM#*3MWG0h2)KCur0tac584jp?c;>cngFjaYk-7azxQgh z6XWKL-}V~;t>4~!Zk>)7r?fwT-CyhbJd1z3omu_wA^J-fB|7pQ_|=*2sZ#&1NKCqa zHXg08|35QeXs{hSb}*Tn>y4B@TwRyO_rFL%0XUJiZY z5&FJryqamf4Z{+KHyZ*cHtT&Gp5gNX@Z*M?%5>En>O=_^V?t=C;!93@jrjE zzRQj+AQ#_JjO!zEIT`0QWPOWoZa4h>^ZxQ*?Wxd7Z}!{b?F9d1RKQx^)Bc#f=zo6j z|LNpAxJ#<#bE`$RR%A!E9evwUXLSEx zf2p5#uUa3pB3FLMB#8HBmbrXzn(mP3_`m)(`M@SBEVfiXggf4*Ku6fz%-H8It)>2T z1~;U7@2Lev*&is+U9ytmt1gQugd|J#Y) zI{5pE?vj$IHH*|OF3eo|KOgLW{v_S%&!hQZ_0Q4#b2R@%%|B7|PeS|WuKB+_fq(9r zf69P=Dztwpw10}{-}{<>isyg&8qjh!Ctuz$3Yz2`A2f&6J6IzKm574cDFIT|Y<=p= zWM<^K#fH`Rl0wTtC-Rqg-k=AqsX5vkiF(Q;=KOt+Gx=VI9ddP8e!@$_Z#Fu!^=pbOOQ zyFZ@;>;SW%g?2#NXYq&W}Mx;UP;r7qer;I-qI* ze%cr5`kLRp&ck+2eRB+r0}BpAXFZc~W7*WFu&t+~c%b+~P(Ts$rF($v%W}9;&^=$E z<5yTPX{;wODB#(4iZHRs^0&3c6xYAo_zev!1~(%T8dH1pdXD7P3m*Ex3j`IJ94(3VHeaVdWR$Vs|ak{YPz z3e&VApUP|e_qO}@Z+8CLpUwBCTVI=Hv(9dcZ%uz{jeF1(|2Pp4;ks(pWZbqN_9)L4 zb_(}S=A7y4Dd=%3&9y<)R5W8D=KFGAEMjev?U3d+iAl%!LJsV;?^z%knlF&fk*1BY z{%0%xzfhE@>wk9UB>*%r^B8^1fKG84$-pubEqOE$Gm@|c!5@!V*%0TzH#mzs=EU(! zn$On<<90&S5fpdBLA7Lkk#U=4&)43SY|21h*m0+rM{N>rd>J^Xxp^1Zq*vts0iOKV z$MLb}$e(9X0iDHdZl{$ut!mlSS`4+etLDc_d`XbKFVGe~^!s;E1lX71 zX0c&uDM0^m9CFEK4<4AV2at4KEh(5(Tf79MB;C00lrHaf|67pee88V)VeSr{g?$M3 z%Q>K^MA!gpDP0ZAD?m|q#q|dbkkG%@U?-!Kv?3wGClweS;pk&n`dCNh+2ayxE&T3F zZvLItK44-_O^St-{d@uIbs}j~Vlymkh`M=1T6dv=L>O#I5C~mjVXbkdq5hs3Z;oXZl6sQplR%gl)&15O zO)rh^0}Qj+z%GjQP$`qwy8%pe9fRi-E2|BdQ}MQXcq(|g2F8hQzSiPQ-Nbt~9J!Ce zRLAZrcDCtmhiNd|jSyex;0w9%(0>&B*1J@W!Xxdj4E(Y&YGE|$_^e}H@UKJgW7QrA z;414{_&9ku2SX{qeEsh2>HyeCaASwBG~hP8a=(WE98Gn#O<0cbZ3-xjRX-fHAj$XM zdsyxFNTo_$dEsrGNPpu-d+M@RjshWk*vRh_r%3;fC++OWt52Na~Z{!Ex+PfN-BrS91 zMf*>_YnOj+_t4V)=zk6UpIGA`IcUj87hN&(BiA?5)B05+TwXpMwL5KvcQ=W{>%ExH z$BT&=rW#8{*cTlh8~MZ@xA$J7w}&;TOLB=lZ)D1oMt30DR2(NVT6t4qPCho>Ep__5 zo$$NO$n+6ugHqQDvb|1)=K7HhE7Y|mG z_MM%b#jTDqSdu2i$U#za2npevj7e$6y>aSHjoh1Auz~?glBP`Fz$Es_%N&#tAFf{V z$4g2G?G&WjfBJ?-(U)pAsoREtfj_#zPUqYyLZ?OA>OCIFa;KW~cWaxEZW2ryViHeL&~C22>CJ0HBB6(8&K@jaD}RkU zhJ|gEy)HTh16L?l)6q`SmGovD&^vB~FAkRm39QdoHs45Z?sXL`O#?|(n;_m9GuLN? z-|q1(zD3$MyO0A^3Ln0+@AMSj@m60ugu5?C*eU0V=sLZuZb@4r*&dzoeJBN zLx<_3M=2rP9aDLE*4{M|EpHia zBgT@l^b7V}nO^POB_)gP38D>EHYk)V0;Ra_r?|Sl!jq-0*%?}qNh+Ik4aQ2Wbm6cr zLCRh4Xr;F0E6pcb+ByaF(%J?*vs$rai`2vArJzEPFf6*PGm51Ad`mpy>jq#AUCi=M2 zy~YgduqIg!wugYpYM{Z~#u_Omc}#(cf8u3k#K*9Q?XsP%fP5Bgi_MBrakfdzh3)CP z9AO=OIWWF+j?8u=JO~^UD?zebCq7H|EjW?Y2OMio2sT?0Zk! zf4f$Ac4MW`T|;-r)q!73#%?-^eVYU#dYxwl|7IAZEAr=nW<-ilxj>1_i6eA?ez6}R zz+6Bv9*n0y7`t{qODi(NTJf@_%+w`t=W3g;wBCVdXY3rUqNuZG%faddt<2dr62J@m z3-a)exhbF!dm;SW6GViihOCiB184)@IO2M=B6)LTh&=(*cI7%l!Sh3c?w}}{UGEn4 zAiT;Ll{#H@>Px0pq?jhl*6Vi}S_w{6L_`R6?AhvyGS=TvlQ0eWK6LSi{F2Ds?Zgps zEE4+L`}Tnhq;{@nAPZs=Cz#qBr7cf`I46UacJ-^UCEGC*M4jfmrdo&wvvat@cR zz;U7zU0<0UDt~Tyy=-h%#=(pC8n`Xc!!xmo%Cd*{TB&Wf8}-Dp3V}o{2sFt{I&0f} z#2%+(uuT=sY@8p&3$MTS6TEmlk$bNdAU!A~+B2%=eFTpW9HAG&%=QePT~w709PJeO zev6^=qPdqbj8e9<&0C18GPr>mTU9(`Z;Sn#_s<9Z=LMnpd&2?pvw0a3)QTxm+@Xc^ z<6M)90xuKxTCv&W35Morn-g1Wp0UgdBiY|zpkEN7ZzN3zD+dbuHLK^hp`vQxw&vW0H^e-)R@m1o3fIA z+ehrh$tQQX-(}!3Ul9Zjl#hs=zc1E%+5yRXP1}}!-0EI+1`&FomAT>8^L4WdC5cT^ zo~P4A9Lk>i2uQetAbIodzL*V?z8*@T0+)HUP6TbGwS=nXpqFQ$jT6MZ3@&CWl}ATD zNpmKg*8H1FbmPFk2bxrSsHr`7p!_*@Dxc~3k^D|Nc)_)l5w+U`sC3ik_0ZURAIoox zusoAX1uBg$8j(SzOPO4ugEKX&{(FKh-_dL6X}*^D z0n72jxQFo&Y`7Ppt?UCk)MQ44pl4FSd=c1C>73qtH%y?SxB5-1^rw&Di6Yo%?&Gu2Bf%^XuX{DL1VPX@AoY);|7oQjM-~nl79y(~9gjRE2k+ zblIcRw8QhZCG2oSf;nX~^mvC>B(!1fk`U>AnV}=jLhAHf-JS#GMLUp&Ygk6JYWxU^ zG*bxF+;s}X#~YBZte*zCqhsy2i*`wEfaQy;d!k=GXsh?(X9q3Ms07bmtLY+<4o@o9 z9K;XKXpK0jpLAZZ~7MNZI~bAl2>f z+r*(a-8nk7R&i?Fh7>UD#y8(PPO#lvgv>^^=b)Tg1KMupCe2%Id-XMDO#RzV_BZvu zOe!8z3+}C1l+kXf8mM&Jpn#4>sCGf>$u7T&h=PbvJxMG}(Z`59% z2vc9W>%v_PI!XLEIJgDi!%?z))hOW0l83>wg-D%cXcdoY=~$9Ao+Ee9%eT7IW){V4 zWo`*M8kV}?C%G%%U{FTcA>89HU_p(Su+S(Z#^|CE%hP}7t((gAs}~M=Eh1(e5!r#jl6cG@)NE)vnZ?Sh&+#<< zvnK6x4MH}xFOv%XujLMjK{B_P3sKx?9}Uoo*F5OG;gG7XYr|M3G2FTX`RjR|`$y6i zii}x!=o6a_!{gr%`&!uXYV-~N^4A=V5RacPoqeLJux_@_uU7|Zb*iO0x)iMxvvj79 zQ-W%6JtZ~Lj)l)@oet$&lwphptoU{-5RBfo@UpRz=hW`0j`0?GyL7ToZN6E#8*eEVG1!&n)*8hv^XMW*AYsrq zH@1anB710D<-U1x-y~GD*T4^l&R5dq<#8=$-A*zC+?{Tla)3~Q z1rsbt@L!EX)j0^PoOUyoO|8FcGa4H(kmcYnJaU=^j~3ikN|mvbO&z?~O_?;~>$j8~ znu)gW_!w2Cu2U>3)VXuX^WB9Bzo}`kLZ4O#z@B8L4Gs5J?BLro0pM0 zRkq}jTKLYDjPzFvMHd{l!LvGjAZbG4eyV%9&YrOQM?L~u5#{3tGM@)=oP6~_3|kD| zUf1Cpc*)STd$mwocxV=vuN8Uz#3H-#aBT3B9of=2Q|jKc2!`Sb0LzZDg~DsF8dUvz zNC}fdUok%s2l|Vj9I5~Qgjz)ZteA07#awkQU1y9XUGp(t32MF2amINz7TgVIlR{1o z3(4J_CNri$kxsJwahcl(I37uN;b1;}1X)m4v|Di9T740wc7OE=Pu(5vI~?o0RkuOZYVdS9d>o&Hbv(Gf;JS|3bhc zqlo^qa;R#rrzTjW_yO!S&U8IKdmZG4%{j z?f;u90&k z;DI;x&!WMo5-J)b0Crj_)O%#-F`4-Zqk=tFO|szftThv?irx`k#fK1UhO-K8R<**b zimPVYdWmA79XHm!{S@Og6meqYoDj5L;cbdLwp%LmH`T?4)DUh6Vgap7_?LiD@*D4I zu-$Jv%x$;7{P~WTRq-ul@F=?E6A6lcg-Y)vm}~IY*%)F3w^iJ-;?U`@4jqN_$(i48 zh!%v5e~iw73)l15-?bP~|GT4Jf$*CP^U7Z9TTqiKf-iv47jHKwwKAjx?!5>!2Kg=0 zK~9scTb^V0`xOLRISd#kM4HJ2G65W%021rzwcqF^ zrDf2Q`c&2o7MY#%Q5nJ80M(sw$e+ora@>co43eNPRJoYhBE+V=KKh$g0ucE}dM!DX zq-IS5Nj`R&yRxde_jH!ObYVaJC+0p}JK27m{Mk076r|n0--q982rMdO=2cm0)Y1G%w-%b+upX?J`hhLJ?h{%B&Vx za1`M5RY3=ZPyFy;iCS4VBjrd`K2$dE21+ebd5smvU#AEy>;KWd!T%v^{==jFhvNJX z#rYrh`QJ1pKmX_-_W3`H0sro?{@L01XJ_M|b(;ULn@;9Owi?+Hy}-K5a=e7rQR1~( zH>YlHaVg=*);*8W8+tMe3mH_612zuU+qV`sZdUZg5bOI$-R%$&tqMeT zCF_9ulw)h2p4ueGxDd+>X;+KrrTz4O9U0O%Q>lSN#dw!~1D1?lfIr09*JP1%B6i;7~|?Mp?@v$hRA-k z5ETY}GKDzN0_al6WkEnk!ag~bx#XxpP8^4NDKm57Cpv&YquQMi*;uxZ621lzzZRgj z&%4C0S+YjSr~KN5;i;is4P5>qt9WWEDJTaA;AMwq%#4)#LfT4{V|jkZ7f(p0z!&l- ze?7Zh!d%DrV92T%^f@eOm~p{tSwg9M-_rYp=zTR<^I+k^Nh%$(&#MLc4W11adkod* zFS0LYYK~6{Zoa`_3e-mX5D1`n=>u_^)Nv`0wJG=6iUL<_j9{NG60}}22b~GXjq8}T z>o#jIp;0sWP9-AEW@f$D_p=!-?K!c2HpvB25Gtl3nIxt+)cX6jYt$8#clr$4^{t%S zC7Z?0T$N_7KHFIV)Y6zEsSiKxALqerH7|d;E=`0Wd0#-%GKbiFJsWe;PZmJ~=kASM zH%BTQe8Pah{-2WoK}v5}4+FLZQPot{#oPxGTae>i;1oYz3C94yvgYltXIf7@(g%9T z!`{drtjO~RorsPT%kZVwB!&O%_NuCXj-u=n`!7-j4Es6NRO@283)3N>_$)kIv{CJcz}@>-4+ z(fxN)7Y%U0_uj#+Un_Fa{({vVm51l8Hvyk2t59`qyI1k71lWK64)I)qm?F~)e9Pt^y+nbU}doZtDqCZG)rqy#BaSK#h4cM!8p7`pRMH5g?fIHIBg$sB3ruRzw zHQwjD{KCD6rd02Vn79B)S(4I~zJj(xkl#wcu>eCJ0bqUKKb`|j$+=0#Taed z(khZzLfOv=xi^5|wyvRp_q3{AKKH@*)#U9v;K-qAhizqaM>|nK#qGmLOY|b)(SO1o z4nH>XIS3) z`h3lA3`eMgBw;+7Ol0SHb%Z)^kS8?r_!F2N^ophmy|&t8b$Q42w6{z^@yCpxW^f&f1kwK>@RO0&KQd~!b$1T)-W1#T%4%L##|XAog* zbii=*C)YG7#$#MOVQG#dO>v$_kSgfzGfZ1@(y)FSAiVY9PO_ZJshk?Uyx$t#mWi-6 z`HBZ>uSU;e`Vv@L?6jEmq4WyQ(y(GX&oGrO<+wbZ#@TkRbfGj8{ox?`q z?I6?_ag5|fPm-6eL#S&-_EFWlMS!;Ju|4w|OL0wV`bL+aju?(yF0eTLf(S-FGI5s~ ziUAvFRK0@~2QB`hJ(AXnteZs#aa`<9bPOoS4GP>)s<)p$v+MjMd0!dtfw#}`&X$;k zBITj8m{A@D`a{55wSrGMz|ZRaCra!V@#li|1U~a+R)oxBUeW%^w}3$@bXzWiGuI+2l$5~NroaXrE^lc!rsuaHSJ=;=y#q*l=e!xq^=-<2c z-9lpt#aPS122nTYGppXmmyVbp$1E7QCv|^43O{PTsDrO#5C8Co7C>~tj!rJRbVuaz zGtX|0l?iLa7G%pEf@@p)OcyM$S5NpwiN|EsRXsVsF_;@zcX-NmI6BHsy=Y@ltQM7t zY*n1z&#dw`t~(cc#S;^*!HM*srus>5VA?9h<_QoUOe5kM#Od_V!$~u-JK}*>iT^5{ zgv7w2y-TKib{og8c3}Mkkbs)^HHn4-Gk6i;^0%}k(LXYLSbeRx9U|<`R$MIo=&(!5 zOj)=2(G*j$(TXiuv%fb5uMQ|J9T8=t;u=SKDQZ_b6O=IjW zZJDKG%t<94ytqf*SK!({SD6y7*fC`k#7!mj%195@I}62BNSAoV_(Wr>C`vaMj_%fn z&|OFTeh*r-?CEBY^9esGXC zscpOtXPjt_uZq9&iUaqeH4!n9G|Yc!vbAPXRiPtbu)tOU<~@4hSp)O5-HBvjrK$G| z$ELoYAcW(e(nm{YB~vvlKfoz;wP6mbGc7F;Oq7%f1uHlymMtpAU{#5;6_3c@HWWu@ zHEjm)Hk0C%4Alsh*reN_T&542V{G3=-&4(u?!b!~_nK8)EYyq%argVoF9HNl@Lj4K zd0Ry)Ue=*t=A#6hB^k%w)n>@1h(|3%d}f1(+KFp2k@`V` z7<00c>MQrYKz;$zY6Z_q@_YSlei@3{n{(Gnf@UP=TASPv1AtxofV*H zaXi*Of-|~0hGCb|D`8^$fKQYoZ28oQ(6t%*$IkWUt6x@soxx@AV)x31%R|yu0>(sLa%r}jP`}96J!agbAHbZCLyQA=?BdnLGWW9- zMr05N*Gzq$ji@ z-kn<2sLR0=e8EC_n^=W#ta9Q7K@b^u;{{o+A1PE*$%PBQ zPk_id!iB26+E%W9w<@9LAPOFx7e~H4kjYu_(%o2sb$Kd(n71mj;4O1Lmf_Se;dx$# zH8`_N=_RkZee<#8nX8|-&Kf$ks5efbp)nuXJ3ds_|2GAKkhz7O(@++@c6s2`=yfpfF5C<+PPxH=kLOA@noiDQVJoLD8G&VwrxuJ$IVIo2nMi;YsX|?fqx$7G5Gs;WGSJibVydoH0r=Cvd;* zKD>fZ)_G`>D->%y#mvEyQ|lPji%qZz<{u=UgN#kQx*D-MugY=O zYvJ8%!+ZHlOV?Mnb;Vhl%h)f|zo@WglhO04fK;=mw<(-f_ltg=&~EKS=Ur0HvBn^T}Lw!nvq-}JzSv?(2SSQM(9q_0{C5c~+%coiJ;ll*RxNqRwXjfA-zu4Jly zD>7JeqqORpx1tVrD(E6eCWemaf}BUNEb?*u}>=Q98bD;|pw=u%Vh%T)jXDw~VCrMu-#h%JCg0 zj3Gm&CqDN0^d}X{y=7`?u_v>z9d?*D%wjdIc=?VuP2)LT^yZ$0X2%sCW@`jKwzOM0 z>uMcr+57`W*{&kfLhtzFM6>NcKe0E$e<$Y}d1=_jWs;@S2w*wYVBnJCY5hNA2xcJI2Ej@I{I#iaPB zt341BZSUj?HI>;XG}GGNt6yF3aZ{7kfQ;r9>n4hmO43AUD00`E?pThUK2vY@CF@o|(?F|zgx1rM3Z+DtA(=mP&0iDsyddp)pb&|XO9PTCGK4vlDAK@Zz&RzZ6q=Vwi0uKqAn)32@w`^PE0>9u&3s+*>;ZV zf)FXgmjpz0BszB3T|az#x>PczC{gyI6vZVwW6I3GPy^`h$`Y1VAcDo1Z1Jx>uc$hE z9c;WL627weZShVaEfsZ`hrInQ2M4N1o<;;cSX=rW^O)5he5&m03U zD@rG1br`}}fE(oDIr)DKWn;u0?#OZ`^m-g`bVH;PRa%erHF^bRaT;C*^ z59tkdHdwOoy6UCtu$*|co5-QXp*TaD&2=1|(3UfwmXSlC-9(r%ELLk&lYvuDv@(ll z+;M)IPg2Of?#g5X#AY!iSE3W;AznT{ zLn0P3^O>klJ1rVh!epqH$wbqFBn4e4cVA&Z*(FusAvsnU_n-8uD9b-rN;<(x$*mP5 z=9-bC=$_6y&rJ)`LiT%HI?rm|Vc)qk?ZlWgzBAIN?xd0s{EKAUK~hQB04w(`Du zKFaRt@M0Cc>d&HjxwiEffqrYF{lrOuM}XAkK?wffZrjka^XZ8-2~EJq9zjyX-N{GL zP4_#3PlA1{h^gCap{6VxfRQ-O;0T@67T3hW^Y@qMs_W_;x#?=6H0!YeL@wXdzrM&j zBcSuc99lYkFlOK391aZ>M#w z52!id+cgx|s+@vn?rdbL71Y5FPoe;6HjbtW)KrJNGLN;sT@4hcw)B=NZ=y#UGPukW zS}PKE+JAt{4PYN>sdq%d?^|leT4MIqb(9Z=aL;rJqQLe>*~loHyXF7G-dl!6oxkzICgQH5f+7fl(v5)9 zrIJ#D(kV!nbi=B<0#ec-rJx`!-C)ouUBVzR#1JDf)DY+X`TuXYXLZku=RDUr=km>7 zd#MaFzxl>}-=9jd-)mnB2F80$Qu3a}ZY567dUshvgCPE{0N3@~5h9LrPY6F}Dv;m1 zMc_Z}GLZpiaNJ7i?e>$W?|Uov^E24+qTLJFA5qc92tO1^iQo27*-q7`mmIUp;@i_F zUU-U*xNn%5Wm({6fxi<`K>KAe7Vt*V{yu8|&q2%wBC`KFV4(Fre3>Ei+Qe&UJBIlT znoM3U3Q)9A0P}VZ# zDz*r_W&UO_xU$V=y4xro(AH-@IChulawKC>!L}>qIl`e=+0E!^kq+?y4b)AR+@8Q; zN{Ms-U4TxFBB+-Aza6^)7dzKM_VGT#ors12ZkepgGm2%@=#`I2j|kr5=Ri$&s>Gxb z$2P?1(7ad5Ev%yOtWIjjdxq@HTUF&WvS0in;AT+$s2NRpx#j8*<;j5U>kZk)4Hyju z+fL$!1R1n}Gl6HTDAR7!vz+W>g&M11pJ{vBa;B67%PAC9H+|iFkc$I;zq7_pgcqEM zyxqP)G1QnO#x;pX!#5vDAF$=>BV4V(#ej@UjlOc8#`I71LaRJL& z%eyQOH}Yz5K4;nHD9W&sVvQ5GH)D2a4>Iv}^IqQ(&ME8|cDhThyiRzo43t{^_abVJ zEVNWI&l*ct8Jfgy zw`|q2>vE$5X#>2mrq!NVZTh=D&v`!>&hXjx(b^=P-A~wALd*Z!R&Q6Yd15Sik9tzD zr6J}NLuE7SG`V9`O?u0Arf7%P#o^?=7*qv%YkXa5QrkI;b7i`fp9^o0%2wjo)KNU1 zS4=H+-kI(LO%!U;I*|MC1a5=u-y2Xv>zcvOssU2LHzLAcxKg#SY}8wm4~zx-@D_q6L{fJ^sIk0ux-3{&FXBGw;1Q!F%|gyPQBSaC zpR^zqMI64g+QeA0yh3Mhb()cPr9S@H$7S2@(8uUo6 zKK1blj34d|eXwI)k@9+^B6jaNW#9rTia9!jx2b&6BY?_bEp#IolSWwSkBi`O>gnU_ z6L9?Kzh1EO$Yk2T!?!EnNXKR!sGyq8R;UawSJNX}5ha)X_kU3B&On7{=H#|U<0S@Jf>ZN(GrVdoTT;7I8hxWnn%=p#sn(>p6+-c#iR%P~ zk^-A&@1Krs_7Xx3x4JavqnPqWttMuq;}nc3BIo>2&&sTq;-1M5cM_0ah<5Tr1us=_ zJnR_XSpXI^);FsqxxwDKjLbE}^VI9tH!|94&fF(m5()KU!BBMo`P^2B6dSq!aBRsG zpDAUCp@n+i#@@DM-$+(8K z+6Cca+TOMIuWoI6mCWPQx5W(HF(!^0Vt7E67kk(~xNmywd6?`>}M}B4r_|-D4M8$=RDl z6AP4a{S}>a#;9-#cA-@_D#qsf4bMHVW39m`^rk?-sJ63J@YRt4M*C;Zor23dZG~+g zTyEuXQqw-+2|j52L9+c2?|&{&;oZ9*WTFy62J6w!%)E;+6QG6E5JjePE}Pm0nZp;n z)g>qz^tX#E#~-s!#oZcJF$eN^GfKE~lq==xW^_Tkq?f~9hj^r9&H6OODptv2C3;8j zU`mTq_1Iy~n9a!j1QfWm4R1edzqi|^Mbf)s7y5+Ya&6n4YQ^H2Xl-7eF^(#bh#1bp ztb9DX)lk9BGy~ANenENZmD>^LeGr*Wzh|FM=$_c)N@D}r(R_xr(B(Dh9rT2n*wWtH zjQ|v1oNU~zud-q*S;TNGO7l8;+WWj)i}hf;HDyNh`rjD%_nTh)E~YLkDFi~dp&~T} z;~J@MsHUR27lN#j295a(5B4TK7T~V+uig;OX>KdD`e0fBH12u?T4pocjag^q$aM_~ z{}?K!n7?)_zOscRH4XA==O$tKF=K)yFx2DQdYF%JHD^xbx3>^HrRn$}{sG~-YiifMaqyY?g+H$X zjAJ^6xCrh?y*%dZ=iGeZIk(TKk`Ms}PvrVpES3cQ%M#Sm$8f$`?>|hK09=91ygnR; z1vm^Q%yG7D)D9nPd&$|y2Tm_g;_W_)RV5kjV_Zu+_G;EguUZf8W;gpiKl3#_^FfMa zs!ny9-a}eFR4Il#cA3HV0&5U812!P=Thn6?29Ip~Nr5(b!rkiPbI8ju&JT6_N=?;pW0}yNfHY35sRLb{g_{l^!c> z#3r+Oq&Nq9g3QmAGk0t82)fpk>lvAKg7q!D2>IMA1-JN_6T&(9iMj$G#*wi-_U!e8 zt$EaST$$U6ndWhy$+x_NyifTcdXaPQmNg<9MnG~7_c0qjh&Zy5H`v{pm;L9$Mz4#( z=Z*JGLzBMPt!zU+c50Z+ngZK^at6&|tF*G-=5Qr8;p!1Y_dTCtf3JcvrUN}LL>GJ% znfj+LH(Vosb@uH~I}tU9LH){vpJo)>Y`T&ez}r}@MwQElk6S6q^;&4NM_djGwJ>HCr0= zAdjfVWZ!+HRv7CYGuK3(JHE#opwgEo_H&D~%|vp!ThiA(%*T&^TIYG(KxCa!51YA! zuFGpuA9RTwcOtk@>p!@h@me^YjB#Q7r#|^S`NS+%tZqDG`2?Mz*n$j*RrU4gHHdPe zOO7(--rFh=!xwDIV)>aV4z4RtztYw*Kf}DJnmF8Pn^MXy?pfR%Na|LDY1Ai*fvzi} z3w?B+>~rJWdR2nS=;MYx1X@<9igt(ft@Mw1KQfJtI(|*6`uY85%XiIIpI6ebD{h#+ zxw_i2OlPNQBFc%sR%m^!_U>bn&x6&zl=d>!B6EA2yK;KVVVh?>xGs>WPfVG{0sP)a zJrzmj8aAvelp>jxfvy&O&Ad{w{HsrV4`+x_=*1)I_HS&fKdj19aW;!*#5`v_Vrhu- zuMT5AAdo+TxQlpte42o|f!}$uym?<}OdX@C5VI*4^AXOL`)!r^`NC3tR4Mic?WqIY zCC@jnA`rYDYT(EgzO^072_x+-X@O1xF0!P8y;=|D8r1{rUHpCO1YME)ZM4sr<;`oT zl}B}S*HD(Wbifvda&w}B6F&hW)UdjCs%WoillqTGS~4Hf`Z4vp4VkC4)vQklap5I zyE(zQ$c8yM(F6DJlWPq|5!$V@XF`f{I`URfGDa zdpmwIfOW`zAH!%!xbF_@bEERvWeK}urz2WwdsW@Sc$>9(XTqWS0RM%nNw18AJHj)U zwh+x|`PDk%LF%Li=P3gl08L|vE9ALiB9iL1j0}=>QY*J?%-(SKsS*^_mcL_5yBkcB zkM>8s@;Rh7>EG9ga^akd?r^Ahl#0$dUN|71kiT8XlIu8Z_RQSu0#kA3F&V;;c-adr zexe1+&@InI3lLK7KL@fr+?uJlD?&T5U1!R*3fJj?FD(k^Ydrj=n_TZT9**#$4SZiz z`IvjX06bxm;|D|`6~IKhHn_}Bw!=z5$ODIaEjPNl zu$hY`iaF#H*TnVMX4VX&U!EsH^WPHOo45TTkaWrWkYvQ@Cz9cl!x8I7`t%>h5ml1I zv12I8m?kfm%meH<_U5nWS+}UHP1uidteB$6%9{oHsC@)pleP4$ecA|SuUSvttIOl* zl3E~h>necabp3?z=Qo?xr{|RRUfIrPM(fzi-i>?t6x(Dk?(j-%3Obyb>&lBC-WbK& z&TK*zLPi(pR9}wgIeskL6n^3OPb>CG{=39pLKsq?CC$-#+BsF|%22sw9SvJ4>=Q26 z2`Fkk>@uIr!!EL@xn*B-ZYzk`*q8$PKcR}CX6+$l?~V`{d(?YJ)EF<9NW|FkexSiC z9uu{V;Hdok&aH;UR0PARKDsywLA;4xlqe%K*M%)}Dxu`vJO#A(?KrUqwT2>_XXjdQ z(?$xXuk{Z9~1-M(|(OKihf&HxtWPvv{Fuh`wgW^X(-(tq(=y{)7Ulr@P^M z(aqK&4TLgnPE+{n1!w964fPv$W8Eg4_Ntzb4P?ES3K8l$VKj0otA8&@^mYTJySeIq zkYlSkKTL-HyE^sF>pyz8(mgiM5=q*1R+5L}0`yLaIxS!m6v%1~^MPMONH*PQE=T!? z=h+oG`=Yzy6R6>VSBNbGCe(o_#CcAmo04@cr|+jtKP-oRY=#w1CVyq5}pQ_#t4+sxX_28A+uigs;B3o%OnJuxco$iH85M=j1QaJ@i}BI_cCc zQR(qdt1Fk65k{I71;QTV#?QO8aj}pRHe8bR^Hk+lYST{zdi@f-AtQ1i* zN8H#(8tIz@-vnx;ZYw@se}8eyWU#@=#}8X#hT`t^RI0~5xU7kLJo7QSWl*7c874IQ z{CBG<4+iCpTrVM}(9$Qrz_+@YydH%M(qXXaAVo={v9DffZwE6!Eavu67-qvvFO9@3 zoHVY}kD&HqByixVW24w~tBwx2f?{s*PO06>4}z&;wpmK0q=eP2_MuZ~TVo1|v1@MH zzDz>iXC>}Q?wYV1WPemm@yJ_ALaedaUwg|c!hYD-JA5^|&B$}LMSe+yq$Z%*`y4Jr zGrSpF+%^@0;zgGm4I4@Yyx;YD^ne=}db_6^UTi2l-C^lFgX&xwzIc!xUcd(WQ}P*& zHPpRE&5UF!<1x2$$3;vM&Ys7P3yBU_ZJ&QS_t?mdfY+qjGZ2+~=QUfM?V}Ab**@Cn zpoWd=#$i?t?S}jg=R)eK@`+-0aW0ro;P;y zGV<+myig@H_RB$Mv-5G9r3ebCbC)OKQDq#_Y3J@?F-L?u$tN;9I07~S8WCP?`0nu6 zBi1$!aW)VLrm?cCR=EMb3485Z4VPa(`(kUSd8#{nYUFZhG9o!IQ3>rfcwi&w_hAR| z(Q<)xYyaLjRsEi`a=q80L(-tLP3ar)N3+i?OCWUB$))uDc>>Xea&vJxJ`rZr7hRwY zNTN?n4KDIoVlJP5Hk- z1$d3Cm6WTxgm`>3dcWP$cV%@@YUhqwY{Js0^I~RZkBgu|kHQgy3Pe3$P+ic>O@a2W zb{#a1c0%aC6%}9@Mq!Lvo$gFZUd-!QoWJc6I5ddv&$2y7`9a~xZxu&`$M%mn?sI8N z&O2`xpD-Gg9!JFFjM!)Pc~;L$Qg)}^CmaDDTQ1l>RE*?j3E>K|1*3Xz=NF%l7KvZE1U!kM1|Cd(3wy3_FvAU}}8Lg7``XA4+VBe*F@Ix7?!4 zVIu`dZB4DWca6S-P8+fhjbP&Fl;Zu!F1EjOxlT+1EACaLz5ZyJgS z@zm|brrwt-P~B|?I(oSv2dibIHBfXq{@&64#oiJ1CFkNJI{(CBt$F*%+ZTbS`BDB5 zG7@w0+3@ulP{@6-kXoSVc*1Muy6Hi?ZgSspWHvT}2Q{{hyW|qvy;+lMhCD8J?xQw> zP&26osg7J4RTlTAI4TABnlj8&&~01KPk6nGYsd6RiZ?s*`k`aJ%IOcBs^vl1*GOchLvU`zZYfqh;<^EQ>1xo)3 z&7=V|V|>P?Z2kbda~Xp&d=Q)Z19qzBht%b5doR5@xu2|CTf(UZr(JVpqm1vURhqcY zb1zs^*#WHiFLiH;Q?T1cPlSk7$Bmv#jT=o(j~hK;>SmI(c>DhBf`AE+>ZJTn-P-n! zT$#68HPlo0mc(`&*c$J-$0e}G;xgDW>sDgYX&)$9bRkJmT~;Dj<|3lM?!iv}T^ioZ zvS>f3B(}3^@Epi8^Rqb?5wvY=Ap0Mqc${&B0LvXtTZ-3z2{m2 zJ_D^bxf?UmV&wtk3FA=;JF6Zw>mCKe0(CCQe=mtoUii?~Zi>Y74&UfBJmxLkg5=0Z zI-jo0=l@rU-J9b7%9s&fT%#YEC_mzAek2L~ND}%H^7J?Q+Se5&aV5h1norfra>FXZ<1!yAG{>3RGLUY7HDLTA)YsLA-#h+xMong= zjZYuNP4qQN{k0nEg1Eu$Lw~IC;2B;)0lqw;XM5m~l!h2{Q&X*M!OIF~gYp+w?dk)f z$KE9A*)|Y}#nRZ(*=FWeyK8}!=t_gY8sNsem#tsFC796LBkypP8Zt`>>O%Bgh2U%U z(l}bz`QtLtcbt#meH!SUF{7mtee86j(2W1V@XX}LTAv)Q9yd5Zp|$9)|r65`{SI>-0u zrg|j;%ye7t+0A8_q)vlxQrU3jzr0PlKc5-VH1f>s6dAN?yC_57eD)RFzQ|tC%59EN zjaNFvd~(Wr`>&c5x=w~lcX_RkMDK#MT2=Oh99y2mgvU%`Cr#85fxYdO%sH@$BAEr* zpq(IC;ddbhLtOm`f%1##I~RVc)^{dXL(~Kvsh-Fxm)`2{aZllEP2$%?t3y-g@AHm- zO3_G@2zX>v9HiK;5KrPy3K{20pwrt+P!z%S=qTXM@$k%p*m~B88m|%QcuEK`{?5W? zmdqD`1ZJ@m)St6_fdnJCQ~S_t8bB(C{0Y$wgCg6dy0c0(+i1=Py8Tv=)H@4UWp*%U zek08+*k9$E+<)clxT9%Xh3m2|8HoZFlvt1ev2}notkqdil9D;h7j`Sk9XaoKvwVBG zrC~nASOigJq`@+?l=rAVjoJE;V3yY@Me&j~r!+b9c1Did%yZv9JPI75UeE^fCvR=z za(zS=z>5=?cb!OpZUH=janr6O3tJ$R)x=hSg;y6?hiQ9(7D!*0-f{Bb`bP()$Cgy^ z@rS^m3X4Yn-&FnyDoeJ{#oE(lY1qd#V!mJ>U@T(9_@J!Shk^T17BO@mx zV)mTM<>taqk0{ip9X+H~(En}6{C@5Kr?M3%{rxOzq%{f4wRW_((Q-+aY@N9@N?JfRne8DJ5aUK; z@~lyt?uaL>G6l=*BDKewC-p#mJaxU-Qt>fAthsL9VXIrz0pz4KuGGVzPUv%dca*WG zVZ04|XUiRda-@z&#H)qVi5`6oh#2=Jd&$TA-H$s{&wZ(GAW_UikUgIHl5=@UgX1n7 zDmf-EHSC+{Gd0;kr#e0PzSokDLgIi#>oYqrI7>y44Wr;B@JeM4J6S&rF%x#+s+#ti zHM`~I^(l#DNq(t!&70FnNpPfc2Br&zDIh?{R`I(W9Vz2{Sq9wU%8TE7Dr3}A&w^Yx znEIGIlRv2^Bd5x|VqqhmaYc(kOCtFlVm8N>hPc6+7xk)Tu1ng#1BX5pAH6%jol&7X zBMkW;Mp1hM;Emi>QnRyg*|IO|bQ6*ByQ}e({zNc;#C9F{neB-)qzbKqv?rz7L}v~e z9qp8%;ug_7i#2<@g&@`JtE-4U=I;?iPEV@m`l($0quj?^qe!&F%ajmNL7(}}p#-y2 z{_Jk%VKV#5KEM09fsSi3Kw=~m#4WcOwLHNfTORQi&#@?;i!LyicrIh_2A8USWJtB& zivUDUh1>l82_Syq=fYDbsOtpjwI)yTH1#8QfOQ5jyCR(3t8fAd==1JCK!F?xnhG!- zOH+9bRB`M_X-r~VL6@6Q4C(2Poswr&(X^XjB}pX%)^^%jU=W_^1(`Q{Pr+`k`m)#7 zDrUv0UpJa&^CGdpB5WE9{np2JkXNLRW!&OXPL^a-CvJ(FWH)i!0FRC^F|}MSxdt_D zYgH^Oi4rD3*S4I5y_{%$o2(@xqQF$%34UNxg`w%A>qHQ<=Q63+y%zS8$q-mR<+bzl z6gH)I+z~VXshAW0b7S|{KWY5q5_PMUUoJvcz?>r3kqR|yQb3|f6__++3=%C@bLWkL z+VmU`>$TP=Cwb@{ZH)~jC<9_I!4Z4*#VeP+o%P%R6a?rnN1blsF{mw~^8EZZL+A8k zdRa&`;8T@7xA_8+B`$ewyr*ubZtzNAf}2u6U?PwQW9f>z zu3MXB*Luy~&Xz+Zm{aa@4|9{iT_7*VS>@;sE(V!{%*eg2I_(JrKUwMXPPfb93{#=I zBoQXrTKL3N&(8~-I(BN@1wu9`TQu2Tzd2xpnP0yHp9^psk@Mh}iJJ|TSw?2-yOg#Z zpwi_kP^fLC*#xG$I}e zk&<0dzr=aNI_bkBSDHm^$UvtT)}-)N?Dx+ygU5eq?<&8Paj=C`m68(s;HvN(WD%e- zF>?iUN*fzM&H~&I->nty7Pjy&+Y?GQb^Pc?Jzpvqs%N`Xx2S$}pUdvWBc27btKCYu zqqgOTpV?2@Q@RmZm(~4RCYCi-G}9U-Z!3a0L~%iM+7~KlkC+ zMm?~OjTe1;l4v_8Hzq=IVNvtwB}+!^!MPGUx)kgZD1d{|CnZh@g_`pkF*YO(iD{#? z4dp*2_jK|_|!e)x=)rrT5*+c%$Le?^m7KMuJ&NxC2QVOQ&I9 zmZmwR73D^iH}-pj`Z*FofeX8Cu&UnQMPlcS9_%5f=oV1AVgl5S*TI#1%fZ_L3bj-u z=WUCOM;9TJw)(A6g;F#_YTLGEfJ zNgERG;7Syg2mrjr(P(?#0De;z683h86tb9&l9}{-sniY)R7RU}NYLM4jlb>vm*Nyl zpOv$2B})oeFf3KW)@XYQTcdOoT9_!f%?mtd?evZ6AIHSf5K}BW!zua#@%7#`+Hqef z>Q4nTrL-+i$oWb|E^^T*)JNxGYt>}wUExkkI$27vAuTu)pvg<+I(FZ}hwbb#V)a_g zUG*V8Q0R9r0OwjOQY5eL*8(B=2(wfDi+Ydf*M945(Ob)RjWgu+PZQgj?Zr3k#=_ad zUObTebcDA5ywnjYSA*S|$nSr+qCY;I$X$l@RQPc0aqudoQwVRec+e!yb2Bj)*=*$y ztJJ{88ul|{uxmcm8oOy!bf^hQpb@y2~|1BkvE9`|IuAvtbv5eu6tO5NJgWGIG=}o%VI#AEx zuAG1}$UQ2=eHw+T6{!uQnvQo+^fH6*kQe2-lCig3C${?h#7mhQd8hvNg?)XFKhysG z+)nnGl>c0Y9Bml5pazKz%>(t|VELki=vGzNy`tTzaf62VG)-)>M*#Xul+sHu#4f|% z#BGXT-IO4JD+OZ&<&lvfI^q^w<0hHe#Wx_`>uMDi36I&rsMl!uB+kdh(2paoSX)rL zs&kqX%U*DSh?y9<+9D(WaOA*h-9;Fsv*078AL*CZND~k!(1(4$I@{e1M}|7}qT;j0 zBQ$nKV7|~yo!Ku_$Ygp9^8a?Rju2cJBrESPDgcWAZa6P=s-}L9fRj(C5#-lNCte~3 zP3F*w<%;bAPlKF?z;G>e249KX8RifrkSQvq(f~b4M4y&rw&wur-+TYB!AOfgE(G^R z`9%ea1mUu%sRMb;PEXNj_st^}Ns%@3`i9^u0Jy?^@r0r?1i`rv;()Hyz z{k?LFt88xhz?)uU7xR5OxRMW+4KC&2xx|7(M`?37F?Os9n>Qwe2rtW&L+63CXmO+v zmv%9C%JI9`Rbs{;WykTF{ZNa)XJAqM`u%@hy(8P;+#f~SWlF6dKl9@mX#H3R^go^f zG;KfD!H;KvH19vw!H;L~f5$qwyac%fHxLTBJ0w9<6l>TWI3y)$fS?RanLVV^G)J&Fquc#9L#_wmDLb4^I z9?x&q#c%w6`h5s_nQ{ys*ll{PZ5R^B@{Ks&ddg)arDYtc+Jyv$71k2Vt)7k%a4>NP z1faVh5tp?(Zz@({8tZ004V_RV;>1}v;ynCPl2nsU1X90MSJZT!dhOzXGVY?*5UuaM zSqO7fb?izNXZ6Ooc@up!<@z@N#!N~`Djr)S#$G!1z#OJE@vl9*4V|fML1|$Life@?$YOg{G z${TOxQUOt;X&`oH$p+BGD0l&8th10FrHEJ^bpuW)HZ@SWac9`NL;-xjB+ECa;<_OL zvU?Tqw<-%0x@cq~yJDxb5g!>~{cagH@l@>-Kcob2$4aVL;8a&zd% zd*hCL#oi9y=-RmhEfW|v%tB%Kt#Clb27 z&Iy3$qqhjDc@=DJ`J6(6+@BPFPSua2(<=Nxo<>p`7 zxT!c>y~C|ZfJgMWY(pC;h|CaWVcw%##2Ho7*kT%}l3WM8(#X-RzLHcxLHTwNXSEOh zq@T{nDQD%?pX>FxqLIX>1e#4l(g2zI*l1$*t)f|w;3#sfNYfD`h^m*ng`;y^r=jE4 z4Y_ZRvy|xn+`=g^yrH_>X;>dfZVscFoW;*iEpudHx`DBB@pyaqY!8n>L<`3%%#@5N z>a`!G;KxJ6c3lxXyTV0ja~g45o{cdImNiXuGjShm(c^VYuDg3qV|4f~345~x!!2o! z?z^*3W>Nr85o_Z^d2X&&=aLkq>8o5{Kg?LJ=(6cA8sbvbRGhAUUcqX4Q=y$acVyE& z+$++IOOqM>KliJD{7I_HA2WwF=4P+)GV8jp%=I5N*Aqfdb-_Ab~=m^I~+Ha&mq{e3Sui)GG5+Bu$=Nl8CQOQx4Q>T zlM(=-qJG({t}kK@BMNgQaah>~Ky5{2l!sYfLU6APVPqV9dM8wbMu5&lKc=-07H)w4 zwqb8@Odai_*pM^K*VMca@bU}dWey`nvRjf7LU3|*gA(#q0?E!M;z&JvCD%=12uA9e z5>NF36JU-~^4|?+U*s9uK|)6BjxobFk#h$SrVDXNeWBzzPNG^S^NXJ-*yQfTBM7lN zgI(oE#31^1DhRVtDxM)xy3T`J_W~ntDGJa_j&~?|FNb5O4ClQer7(T*PBL#5gb{s# z$+zH)6{^R5@}*V;rL9@yuOvK?llT$4hVz+dMN64{#4#|>-QfW0tECUR@$QwCL8~m0}#fOxy zkT#tRo=p|GP-)R z%~lQnNTJY=?jeaH|MMH-n-K4-smsHL0*tNKdJf$M_?qn5y*1lCH3~-$iAy5!j6++1 zg^6mJCOBar(3MVA^mxCe|Q_jT)Xi27oe8mN;D*TVH=1yC?P{}N_( z8HfQE@{Sb04g%<_O)o7mE=kuMF-#4ax?kqa6^3O`5unGN`Y5-uXqdR+T(u3|00N&+ z6q^HRk`>>gsX1PrcUO_>;Yfdw2NKP#>O(yhVd`}2_?}Fm8DT2_S7+>BZf@~k-b4^0 zM_ZGi(EiEiCGGjguj?B<`MFTpmcq$7jKsw0!Q{am7XVk=F`g;)+az1-YEv_Ye|{ zg^wxeV|RVVPstp^vW+plvZFJ2{o>iy0p+B;y4y@E-oQiia5ZT zocoX?H#0m!xxW{PGc65Oj$t|(MhSOAH4`R(!)TBphz3po!OXoBYeb3Oe{5HzxvE<| zp90fmLp4V8mzWgA2R9ZrVBlHK;RE%jdcXhfT)CF+(Nv#b*p)9NvrOH z!P2XBNeTzC27v_4OD0ZKo&{?k9Xc`=_dx8L-p<@0Rh zWqn40mqTKJdTQD!DNFb8IZU-$_|)*CxnjKxbzDqPui8V55{!*A@Il^)?=!|Z!pKU| z!^F3#vfgD>>f5_V+`#6)b!GQ$M|lc9ss)>rQCSuzJ2->xoSdGG-o z{`XW5KX2K0k4Ug#2^NCUfdzrG3&Cgy!Is~8t}e+uZp5!x61g{6&IPQQn}_16*F021Il+?8MD%&|gbIjn<7qD?tw;NlqH*;eVu06vw2Vi^o_UeO*nigl zv>i;z8`sa`5tL}Kz3J$|Pw~0kSOoPm(>862;_7cYdcv66DQfN25V=VxQ`p=bn=N-H zz(!g8-ca=S;TIVG0c7`JG zjEq`3o=bJOvIKIKdL6!;ZcUq)qEt;(dAXD;4)1s{G4nove2wcf^nYizZpQMTTEEay zhVk+6T}z{(u&dET%-Fdm#w7U!ap%WqkuPU^`>`|VPlb?!mv;SxNTF8oMWRXL(}FxS zF92P%lOMGAf1)^6Do85l=J;SLoyS8G)5^`2>Is(gCwyWAc_eS+g|6 zV}mOr<$P&NXuAe+DiyL0 zsTWTmSd!rU>-QPP3W6sd=@fd401%KtfAifK059fMHT`k^QAD$?L}BG~`~U-X1A{^X zHw}!ujFv>Zdh*%xSXXJBXq2}91QNN>Ws#ZPmpNo15TbZM4nwE)0xXu(Il`lfHa|{_ zMho(2_TVQJ_`h5p?Fp@27ZLM%6*lJZXm)akymnt`s4lR>WoQodN*0k5F~3SKuM_9_ zu;I=QzIy#?tRggoodJ5cLpBA_gI=~1Ud4`$u7bhe!AM$}kqU`c)3p0olpcDx@shvu zm$QxtYzK1nJ6(v$GF~!3taZXx^bT?Y&*^`h)?aT3J2^7#LVN||u$-qDcD0Dpq_`2^ z+>g++M5m$t%W0m}F%J}w%Z6!H-^6@eq0TPModMFZf&!2c$$r)wZG%de+u0?Zl#l1@L za!MUQqawqbc@lCteEK|5>~4_Cpn>gZNEt!kb<;Sv6x95wGR|PJ-HkzvwO~qJ9BO+i zLxfB2hRw7qq-gYrq)4)v6R-Enj?H+Ccv9z8D)z)SY{xHCVynwTVmp)jYc8*S@5!0} zx3(l_7wjBh(54u?{IK1uQse@h8ho11Y<+s0j>gZtAu@axFOl6GGOF!9dl@?8!7HB^f#vg%9f+BdVRHswYg6I zv>Wfkn5$y|zT0b=zl_WvCtHNR>NJlsrT50CFN{45+eMW&*PHERr_Mg+PM`~NgtXh# zc6%}>Cp5$aQ`7;O<|g%=cLDT&w=3%7I%ztHP90v^iR$-{y@o)Uw~IMV%x37KZ?BUV z&HjlGGA)%(i~nBjGrb`9|YD~ z=+_L=E(em@1u5_jz7Q1~os6YJ2rjFL^-F6WnE43@$TD`)I1#I=Fm}>H;CIrwTh;x( z-qsi(KZXP$BnKyATX$U+G=7FnE`0R!BfcKUgZ*4v>>Wc(~ZDydu~_HY5i#@ z4aRT9$mCZ&Ydlmb+WmGgrBF4_0oz~T+OR<%$aou@(q`vvS{8rf#){UEe0=+z1AAz6cH0%Q$iQsfr%mSVJ?FJ(gzmLHPzp1yn-JTf-erL0D5bc(0 zRbuQ+7Np|Qlt4L0F~nwj#-E+IU=m-tyET_N@t;+ls7c#Az17km(@BpB+EWBcUPh~{ zkvEHQ%NCx=RNN5O7{vQZ2YbWAfjId3Tgfe)xr*+XZov8^At zh6O^O5cjjSIp8bwfr>yAb=-dq;SwZfi)>Og@A}h4UlN5Ax;~;~r<1ianFo>@6g3<< z3y9@vCq>)*Nr817ZIK2|jElTT$wzxRT~W@3u3UvOA47rI&n+#@5?%>|B5)A~H;AB% zB}DRUE4fhcI_ah^HNoqe$#u84TPDwa^_0F|feL?_PGrG!;)6uMux(8h5rg#2O|iWd zvCL-n&#V7c{YJ7Gov`O}Vzk!GPJ)Z>Ck%6k@k%g=~xq@M^MDXIZ3UJpJc z5vnxcJ{mVab0x}X*d=^1J;*y=3GCJvf}D%#LIxs?#O&~5ox#Yd?37?EUL#*v3e{-l zi$`X7pxugDLZgclK+jVlsijplc+utQ^c#hZN3Vo3g_`>L2^u>4%Dna~u7>_FeVC*` zrM4Q1(ah1a=mxn$X!3Y4yTTh|59d|{<+^&N>#jgQLDw@mCz&rD*z$_jMC3nRBsx}N zt0-NllO`t|YySQ?UwmXZ^Z3wNeNVdt{#)&9U6+x^*dsEyN2hh!^AhWKK-x?6f9Y$q z|I^nTa^i%J%p%I4&?aOZg=lmI8_!P%okgDAscBCj<7CeH#j+Q}tV+`@_3^6;b(O2F zDxt$?-Y#>W+<{PuEK$;{8_NSoL@s6k9jz~F;u8@Iyl0>;#TUU0sDHsfmZ`bJ?gY?M z3u#Sin6T_a@ur{sc9ulL?lj{uIB(zp(fHFdUXj=V@?LvZ5Z`F2jnRAnTEw1WnT>9f zil&S{7;U)o_Yu3cTCT`Baf#|?=EC(_*9Sp7u+_$-{BU%GCUFikAB>Q)OM$A*d zPJe1z!a(#$Z%Vsof~yGlPt3V5hn>t5GimYLqAeY*5|*UoBw(cz6nYZEDr{ z?F~vE#YqQ5B)k>BwDn5eP+W(VSi5;L+H-{L2-=E9c@4-jZV7a!&@|+XYty?J_?{bV zOFumt47;fZkw9uZ_a`+OlI0`X_UgQxLONzpK;tXtATh=lF%x?v2ch}>m+h3?NCQ1{ zm3d8cnN7+4^BsnGP;+<9?NGiJ$QaEavJ^#&wmFJ95a-yQ)wJ*a~WU`uGdeA(y^&QMC&=UwGvQ{@K3&n_4f#YX|-31P7+W(>hy%8xX8v*t=3T2Fbcwa{YeB-1ya zL$`%YP`lp!%o#6ul&06RCIjKVUIr$N>UzbS13#~_@MobbJD-dahQ#o6Jj^yc1%Z3baO*btH^TJ3d3xs!P&_F#q46}6hXPIH&Ua}X8{!5U(yq#az*#tul=dkfZQaa+v zb$%|-roKG@mAd`s@M|ATKjCYfrvoUV^cRVMQ&{n~Z!56=?;iYbzBzb89%?cLWF5Tk z1A5DwS4Uf8xM&l8$4`n%?9)V1EIe!lmCU5 z4j4@o?|{KWWsCCS@2Fjb5C7%hSp$ybQ?e$~Z=q0SbOFkcfd$IpLNWKhDI!%8Co&00@cm3yMajTbP4nE4hm0WiL)BasS*_XHVh?I?=KuDD+V?1!&gGS|M*V++QE})84SGey?XAF6`Z&P z$}U9z_MKd~RQTq#?1_h$4F-TIb}V_3^54D=aRak2kJtRoYfvzgNsI{m^$q;vWc|&D z1Me9`j&=R=;|uxmg?wud!b<(IL;io-Ar6xy>;zXG{udUbE#a3VyXfK`@K_vFwT%CL zvo|7~u1yZdga7_dKN3GN<0^U4`|n3#U?nuhq1bDDUymsN=~uL}L)-2krzP@FfBDtH z{^ozvTYV$u2ta7%zeup)?UCu;2bzU(4(~3tXs01$U?ydbv3zfg{hv>tDm4b;J0$=! z>+Mz4{n=ph7oM!*_iwjDs{=6nrtG6-3S(NvGXC|~++w=WY4{TtqXRR%yt2xpe|`D5 zFE3y7}vaQ?A0__C<}PkuOC4Xiw%-~TdQu%0(K5Jy7#VpA z->KRe+5-HHz0N0Zr(1027Q)HPTt6YQ`o%A#J2`;kDes75w`{L+0{^&A>aq~`_-WeV zZEqv@N(v0$=!C~gEs4_YwE7k9#JId2kSQyGUN!fN+zjEhTIf^wBwKgZ50JtP7gNJK z#ATN7P`Jgtw>NjsJu;!$>tSG#cgMK5(L%crRvxiTVa+a7afdt;!N|6xo+2ur?WI*< zRTi;m?^1B3ho7PO$Fmn|XF9E83Ip>8_6Yq+Pl#>bLzrguZzL|z9C_S8IkR%{L55w> z1{RUnA%+-Q?z1nu?OWKR6G2(ME?{Y20iZ=JqO$J&O&#!qEwA-BrnBgkHDC_71k-WW zdVvc_p6V4u-acd$B4QsIq(`ec@#H#DddHw`%?@V0YE(j;`@xk7>kij?o-PnC^XP># z3-uN5?63gHnc0olMPNRkv&;%bVAoYv!IyYjK&KGwHm%&`kQ*XoJerHj&F9Ej}G_u0Yh+}cs<#>xm3BV^&{VbO`0Vu!iit)ST@43=*c z7^t~vgyRisz`hl!D=4F+J>@!TnOmhoCtQ?==oM$dc0p3rN!Vvx;PfyAh3nJ*Qili5 zmtCU58?s4F()FGy++59~{jD9hCwYBX&5xE@bv1DatuA5a^-8-?8e%Y%VVhE}N9PV~ zBavSG2I^W_;(!ZC)z$G6J`I&iLALRpO|$H8wW4!NXr!RZ%&4H>$;3w zssD@!J>8KA8x#GS6I`@!PI!NQ=m=!sRyDc@ra&&%Jzw}V`wg9QJm`LB5hz1ammUZ2;h$Mw8QRjWi>k8LPSA{IEl$ zG`kCk9iSf+8C<7>Jiw?T9)Y@;txanJjuXev_xE|60+uk% z7|C9m=Qj2Ft)eQ zKzN0OyvXiLPA~ITZ+}uF{?zUFK%cKITt~?Bk)9_fJYrsIJpDL%;cb(*^}<=Ez%iFY zhb9scC9QkYuZZwgb&RvbC3#TZzD(JT{b5kJtkhHB-Y%E%@|nt*SJG7nBHTVl*v6_!M#Z^gx2)U7Rrs3!*~Vx4At6$soEPDw*P;aP+}@Ff+?@&a zz4ta;JlD40@r2PVFH^Y{{|at95mz4)ir3v9$g`3>WLXWqEoxT_&8ETHDMoMUvHG!| zN$#;I&d=cOz$AH1x~wi&Vxm78%?maLgi?7RB?LoUD3g$Bl?Y2?;ha0S>Lve>>TRQ5(}{3vA2uZt(xsorfqL!p zA@%mOtK`~_WeSL z&$iJOBXK%WdFOybsr;Pnv1S6#uU6BRoktHOnn?xt*Ncb0F|Sq&Y-8Bp;p$y$e~60+ zZ$07V;+{Z)D!8Nr7n?hW@+l*RU>rrXLVeHO`s+1Z=TXxed?`&XW%{2hqOZ@hCgB~p zfxAGkSwV z?7jdURI5a9*DB^KW6VxjCp^9z!2RUREe(CRZe|g-H4{UenH|t*ds?V#yNy)v>glbg8u=`k$mOT`n$7L-Y*Hsi7g_ zcda_y+8r`>y_)3oRJ?-4CW^#@bzMLG#$12((8U$f{>^Bg&1f#7yvy>d4pv-`#$Bz@ zxlJh@DeV3FK3-l`gg5J4o^MCG9eU!k9G>(7aky&bX0=m7-b|(@eMW%GuHz9#+KWt! z0(<2m-ndGjUj@gGR_SZw5C7D%6bBqWAR0$54>Vdg<#wLUi+Bqz3fc>ge_2cPSk&Z9 zKPlAm%qbUpty-y<|l%Mq448Sd2~8 zxSYBic2=SO`g%;TcgVEVu4uzCxg~j4k_z{rbu%LYla*uAOCn#GuG)Usd)wt=QF)I| z-E!HZM*Ok(Xs~}PKfaVTvyN4)Hp#U_d)@H>^0O*+i>Z<*RmW?PS!8ULg7m*VDv&4E z)N6Zq3w*pw%cZupPp-#?8;_GG)EK#~J=}09opr@{N7c1bx>+u>E%y7MT@3O&9x>pE zxcjz50z86>UNd1YMiKefHZ5wt&e!zTL;Qc_KyTqtRN+DXM21QLs6fliID5L#U!4$g zGM@ILJrtPfEwyp5b!(L2f#4!@J&kdr)c-ufizz_*vn;HYdXhQ`d-C)5blOM_F^kx* zEV_UPiV}EJypItd0O5zpL00F8nnR;dr*$LlhT}7nsHpo%o@$5Q(2~WKY&e;1-zfB8 zC1I!8N3H*iPzEyu>$6Pr^J0b@G3><{>qr4uD- z4=5w~P`ZxEXxe6RW%9rr&JQ?-sJNzu5W0X@LYQNjlx%HM2;;m995{4!FS60RdU~(| zZQSz4yJk)LjSyzW$-BA_YF3>iX0&eOAP2hR#PlFN^vOwW^2whH;`b?^-^M2q1~=qB zSWz2w7*(&X&*#b8mbWs#$NeB*AU2ha8jD(vDtTevovK=aFDeqcNmg?=I9vX7BWiHt zu$Q7i`cG$kM&XuHFCa%qh`3Qgt9p5^Jf{#G;c7!ErYN$e59=gcznD|8sMV2p4-71N zCz6xazX*i-C)^);OS>1nL=D(*6Bx=jh^!95f93h}+G4m;6eBRIYsYO}uSfom_TDor zs;pZVl^{rqAgE{w0+KUQNRq6e3Pq3{ zt2lG@-skK7&Xw){bD#U1drtqT=Et;ZtvSaWbIdWu`@U4sAyHh2X0?;9Soh?}UP5vH5!o8MmEqwe6*6Whk>yX}%2akr>Y zd)KzOTwT3(!WE2cW2mTIR_>;R##@{csphkL2j-jwg^Jso< zG9S<$TKdEMpR*%DJ1XL1FUB=#wOA}XjP4VII4S*4YMcX^Oo*T{JNvoo!=K_*1Wpt& zs3IAvZ+;Yt$Y#t@{t1b~;%HCFlw^xez3J=xLW;Hp-bIK`1fYI;x9Os99s-b=!G15H z_(9{nWzOhGqWqCr4}luA-bVJ+qx(yM4@|2zWRzp8{`_EYHxf?77jo?S?F{aYZ$eN) zr@L$^YH0y{4u*8;dj(fOU#$7xlV)57NM}rg1@MN8lH;nfv+-|;RtRXTb1W5)$q@1 zxBJQcPJ?XTFV`UDzI{O}60UN!9ry6bib?N5lv4?7*^dagAr0oTTZD%qr6!Nhb9QeR z=;be1`lxbrsxjSeBEGn3G2_0*jb}*a2?0~p0;_+b=C}wcKj#&>)=LGjT$da?*mZpy zEqmRG18mQb29pyt!7w94-R_g3dCj!*5n2LT8%1E%NhQ1c6fV%<67@9SwYYq5;Tw6_h9YwUG(^>n~hsE`ZraAm(7esEMRL)Dn)7o7V>J$MUG!S(X8x9@0~L!&?h|s53m0Y#Yq-JjIxGl?=$<>4LVSC9b^YG1kGA-#3_1%ZxqF|(jcD^IW#JGE z^0D-Ok@P|!(PZm$L^9O;e{TawL0Z**Q&=WP#lAvCt5yi#(I4AAUVp=^^M>%yh~m8# z9etUMwco}|B~A92^Zsc{vrZJ_QbyU!Z&_MpMYw0LA0G-o<`UOnn%<$@Sg7X~!by3O zB5hBfD868K`Z(YDap`+yF$z?FzUuLZA~VfyKWTw==v>TmEa9` zlu!EKVa6*?B+sD}^>M3Tzf~gR|Cg~ck)faqg%=f!YCy^Kv^wI7);#~s1DgiTrYzc(J=1X%KQ0?Zw+_-!p zGPkIM?FwK1(k6U{LY}4R%^Hu1BT1^Xga|p!^Kp=A$u@YfWb8}jxj|w%4};l$FgeGe zOk=?GqG@1f(_*_Os^8Do2w&2O-SqayIzNMhB~ILDPMtWI_4i)uhAv!4gH8$6sbS1t zl$f_c=He-)bNOi!rv1=ol7OHj)8a=lxIggp6>--S3TNLdKJX~}?sdg8uwl@=4R0xpu>_R&D3CJDNgzHIT4Ve>!do^*|=BbqlL zvT?w+G93PcW@+A89@*~xX!b;3SEvl2oD>|01Fs8m#=QIJAXNky7dOl*SEZK2y^jtP z-rkz8R%Zg1HZ~W>!~RH#EdXUA{8=1|^uy0z3Cd!bHWrCAs!G5xPQP#a-7RV^7-_(CI&#q4Z+e>3(7HatP=g}+Cdxiq^Xg2t8(VOU>HSZtcuhLiEKBaT1v8xUN4aiNa_#nY*`

0Je-Se zhMamcU_P^4xHsh_%`Raiu}7t9{+tC&YXw{C^i5$Ce#PA~m&>BJr{WGcw6}Kp~eI*GwOS-W>zY_fHCh zsVkh?!(tI8nBC!x@gZB{1#D2&@p&%hD_29b8D_7~6*hCw@$kw}J5Cn!Xnd%`HVFo~ zu*A46v-Jt|(s8p!cE)%$CYY%P?*r@E+O0CEZgtpsWU1IHk-)1zQ&OWJxmF%kY#O;g zx<8O z+tJSdah+!-tWpL6O(ZrTM7bh{vo4!Q^&D*dNduTl2mMpb9N_` zH2@CRI_^wqEE0t}fcK^>7grdZM(vMdZvEJ`G}EzJhL3`*=5!E9n7!&WRRAe0ExI32 zBKOL~>~!}5wxyo0aKVEzn>6A?-FdO;vuslKjbA4+iJ{t*M1`bsJh;jb$QE7jzMvrO z`LaltWM-G0ram%c#=(+omF)^EojLsrRGV&PnbVX2Vt8FDK!yXlNM)PP8BusseGFb8 zeGYqn!5#ABD6>t7KU(fv+ph9pAVr~wjJ_HJt|jDsq$!J(849G{J*=1d;HtE50b^nF z*qzqPJGMK4*9kA*CTFC3MrUL4ihJ)8?`LUPP5|qcQRN3n3U)B(CJ()=r5#;w3UQ5e zkqmwT$l#9tMKjUA_vD%*7Q`&E!!JrcmPPYsNQFP**tP}0N6qqHQF__ZVzjc5{`12y z8S0N|WV6R%NIcAol30a1_SPsf9)xi`?G5hU6oni9o^L7^IdQWK2c$SIlF!|4QL;2q zDH&%B2CI(%?C%gM1r*KCZVgft??&pnCrRl(0a&Y*9w`vdw?2TKBtcI=8SG(TYiqb8 zkKIpKaA)tTns!8WgkAYq)wz2#v7=2h{Bg-GNj1CX1xLFYR+p#k?MjSU?~eAhay@D2f#@ z0AQ^Av}~*+HKYc0M)0pt3kttk0=M@4x1XmpC(Ba2{`{3eM@oVn*NG?KKK}mm00bih z7+Z1s7P<+jKvMyUdq-C3oEQ?Kud`>yaDhK?Gc(X$I&>=knL+Olk_mC3_!4D?8z+^pwTh*<^CElj^-*k~ZzT zmpsCa6M(?bySp$@UVn&@oa|2y+>Jte!u_%d+s>=SE+DUL|Ba`0;GvD{9Oai}xB9&p z$x5vxCS0pRMWmQ0&=qhRJQfNwQ|*>eWeRU9;@{Nk)&+FEg3%l7Q!&NSO-oe7=<5^g zcpnz>M>;)?U=^z*{;m0`^tV$duz(&{QxZFJ-R-&`MQ`4Sb5uUb2}W5cKwgCR`9s8w z`obtBFhRsyw+>ylU%0SFoa)w_|HXJ_i4*wdztIT3!~;L zU4tJJ3VRS7t^+1cRVEy%Q|VeERrVxF9+Ky>9}lRP-iL0s4A9PfqUu*gL6_djCbvg_ z4t%(-uSF(xOy|T|22PqMKdTc|Q)(VQ&2U|}0ZDFih_QFlsLU1?8@tBzLcl+ls^3WI z$I)eX1Aur~sXY|d-?*`)*l+6gG`j`_EhnJfBTx&34!_hRK@)e9qu-~RF-!^>C_ zuuoVzmDB&eabg#E(^t`3({1ZZ(h1jfPh5~OTYgQAlr}YmV+L4Z&9&GjOq=QQYjsD%LL|P>c5QMrf9@M{F2hipbCk~7o{_VxsBXc_?q427VYdPJ40n^h+FO->qPr_a zNr*uDbArJ1zl?U6H)%8Qr9&LR`K3CGwbYH!{XB^qu1R+`2p)zPW&+c74R&PKiL`IE z6AI^?6+*^D`gL?{f9W{IGM@^ouA}$ss2rV$Qrt~CF^b6D1!BhUH%_S^Zt1w^MbBA< zZfXP{+AfEtgJLR1V9GHWME3p8omi5k_P;8u;JC}v-m?b*$Cz~GmJ=D$C?GWf3iJoJ z=T7%drl;56cWi0Z??VKn_W-OG{q+i_+Ti-m|JreJaJn`bP;T{DkMQRQIen?Fx2)+G zu0II7uS#}@O-iv)LP|Z7A@AW;^EjbZo1SOsnKx)Z-42(F6_MHud;X5d%sB3&T;vTk zu|E|!zpgvBKQ&5fo;qG_6*^8T-Zu?yoOD^+jYIP;>G<+2+F+ReM40KOKej@-pxhjTe$70*H4= zs-`SGK*6uni}Fy+@>M=H3l(z@U28TWesO|YJtlGC9^Qj!bqHq7KQhZK|MgT9Y`bdO zlxMnKtxS85mR3=?dI^Jz%*`9 zRg?T^vz1HCV(DzZ8Ya2(T3&DZNdJcE>1*Wfp|BH(;~YGAJ21M+mmNue8(r&me z36D3f)NCTn@5{;FJjQ>d{*-2z7t zHEzdP#o{4(4!+MqE{mxrsXihk;jeIbhj^h(y*)+ICvij)JPi1mNJUsz)olD3-n3@2I7Vv~`>OkziYge<2ZYDI^-X zdo;7__bSq>mQn5?I4(g|jR*z)w0URvd0B${>cK!!!Xf*EEAq`9zGDTK5aa2p0XJbB zWkd>#Nt{?HhWsST>*oX^YvpOj4u!Uoiqd+&`^;lFGZmn7XD{*(v~4e3@!EX7k$A6% z4raEMK&8?5DEt=7u=>%PGI@$%5#<%yfD^H7WyLX?Be#3|_5|_rLI5~7Ru{3HAP3`D zL9SkEe;Hv5*xkz!?rOuJG+&L$ofbRfL4}m4w!_uLDaR7aP9t2nb~zZeEt3wKn4_B? zi*H+Uqf+l3SPb6U7&A;TbrWrtbqjvGSgC$%TS+#iB41vR3J#+`XDM{zKhZ4&F3|YB?;8QD6AX0C|J9DIBpQSawD^ zupJ*?M`4j~dkAQv7T(l+cNN|fBM|Q!l^Thf(I&9H;u*?%gCOd4FrQg{=?Uj>xa2Ic zMs$O(YAb5SCGShkg4oOPBCcTsF*)LeQD-#g!qTO~QJM+)0!29j(<=(EtqBJ zRSLu(^o`Wp9Oly{7Gsuu;cmQc3ZfrcQ(I%-9N);O!_>V6$%wl)PMx|{Ul~)&PQe9Q`jjPFOVBAHk*GfnP$1+u{= zx@Q!R6eN(OdR|=ll(U5$z(|+Cu9F|AcBJC9n}{RP z%rjHk;^WUADYYaRzhi;P7fpg9>=+eLCi?0CJ90_vt}fyQM6(eqNCaKYl{q2=34% zL}2d%=;=xz)Qb=YsSZpswZ9Fwq}YuVn(QDm2p3%d8upeMUUCG93Zs0AEoo#K1^)jE zB>(M?hv1QJcokEB!Pm+0%{H=@@-kMe0GdzZ*(L>B1z$Py9YYE6uW>43@e!y z0$)^$Lvg6VpXBLcinX!@vMHtNBJ)Xad%UR!r~pH{y5h0g%OqlWyH044J$h&n^cf=Ac{e9L_?S_8^g9^CkP0xJo6m#o!0OJijQ+1t z&Htk<&{6@%i^9(OI`|-P{4F}9zy}wXbLg^N!ULj`8;{!gf1OMJyK~LZ2)t~pT{ZaN zt?2(Rcu@YD1uwpt3FJJbnQWc-e22df)>#vHaUty$PB8gKw`% z7<-;mF^>P^i~cs1e;Bq`5G!QzYvB?XKm2Y{FOSsVj+a@*TP$3$xYH@*c(l0(w+J%L zZOU0HKbs5mUq5C@<)v*z1--Q8Z(mx%@HRN-IHY*1F8tbU|I-?%A=~-1O7-`OoH0uc z_kjb``uO4v+3$ynfZzZ2U{vUguljqgygpn8JLz)Q54ztEb%K1lZ;{>aRs{$5;%%@j zx8r)He?N2^{5}t<$$L|^|cL_z`9>- z7YqE>xC^Af$mq;MJ+|Kty%P^M*XYTr;_ux+SVJ7X4n&M4FSV%l+v)P^;wZ2hFU$H; z|H6v`(5;*dL>18g_HiT(sc|9Iut~^`mi#x|XuLk~IO#U`umAG!`?ufwCJ%wKnDpOfYIKO|Z8t536A9MP}!SU~3?WM>Kwokvn zr;C3(?6Veakr52l@9X|;WOl*Q8Tj$|IwHM)ZId(qxWf&({BQ^A-!CeZGaA`Z{DJ%X zw{n7ZdQ8dozrXCSruhF=Q?(Ah2+_ZeL27^I)O!WKd;oj#gCuw7VRk3UhYIvOXlEmh z18P^rdQ~X5^DwMLbQ@O^+5w+-V&=XwF~{Ai5cTk z&Fo;lyS-xz!~+?8=@Vid!7J84?d< zhrbA)f*zA0kr^CPGi8KfBqV0J29)@wYn=u@Q7Z;$5h zi~%JKR907?|4QW$E<_N&9Twu6D!R(kO4UyV6hPPZ@eL9HeI59}A%mxFFMJ}WG;QwJ zchQFDU~Bv+HSH_)0Mvn2E1( z3CP=t^+GBAFZzR!RN*3c>iO*PQNy{5^SOKY7ExP8F}0S>Z*V{fTYDRT+3NJ%fVQu4 zVp;~3$0Iiq7yh?OYnc?Ur*3inBt88}+69nJy_$0sbHNrQB0n4Qf9TX5Cw1=ej)Qo0(+vZ{rRi)AOnaPE4iU&h#3H1Z zoc#C;b_it_cFqStlMWrnH}Gf9AcjV!^P?G`73`*tVDqb#IO?!C#Q)bFSg0&`#4>549Q1O@$a-=j4bum{RUrJwV6 zMsf%#o!A8;vCGKyyl17{Sw?%EVqZ+yWGNyQ)>PB?V;OqrBxN9;=qHKzYA1hkpWKiC zG!NAuG_c8K&(t-|km7qeQU zId$U{oHSx~{3e#>b};70D2BQ;>j;sIBGg@b%8b*fqaE=jOu`DXzCOVniAL<$jNi4! z*pcc^<)Gf%PA^i^Fdx|EB8iAXYe4h1&Jd&<-t`rb`vq6%$7FTFKYToEv7)7+#Y&S; zmUd6#8<|x;-BBL*8hVCvxmI1dW`wk)4aa@o(ZjKs)`Y;vV83d3a%}_ysxYribUY#N zX$_%k%EnC#>;T@*!W}>woSHq&-tm*ygl)yPiliToA~B!*WkX}=Oj6H4ln}3NS+>q*Z6B#=$fJ~x2)`s14 zE1`N2xLoA;WC769*$61sLxi7G383@9WSIZX^O0mXf)fa(_{ORL zpbGfGA!GOCzyqy2{jwWS{;$7nt~t45E$)JHnd)b#R_FjM&Q0BS0e9|JOnm*(m->!j z{!lvDRkgn12rGItZzwXnY@*8EA|i`xMY2QoxV@qq0(FhMd~cz&Sxrz2*%{tJ0(wmL z04uPSI)azKI4Lc-v#1j;B*_{;dW_ZQ;*bex<{OZHd!0RNecIaAp3@#I1E?9ESHv|} zBQWW1$D9K3>MFg{jwf|R7w3N%7!ert8_3t%iHps_P-e%*@Nwf%LB`=L`OYc=^&#UP zMOLR1XiP#}s&&L(J$$k+*j=gFUZZk3J${V!!4@7reWnOaRy*pRM`IYz6 z`t?ebH2`fg!5Fv785#Gj)??hyjXH~6Sy{5e6+#DUh@X5XjfXHZR+C%GxM&{?<_G`Q znP=@saMtQS81kaf=CdkjhE6-ae77}=@z3Hbsv$24FY*3BdNqmejeKtRoMX_2IS*Q1 zN8{`liFu#EV!0TXPk7(DCmC!QAQMKw+mZ;`MQ?YsrruS2=+fBZJ$vuBhAIRG~;{9OsWF8+_%G#HGrSMKgt`cOR;~^-lN#XJ8hWN{54oO;wc(>dJ zcCzMPf$oGRFHd^Gy$UqW!4qF##GI&TD6X-Z zI_!3g?w?*Op19(utBUyv3AkJ;U=_yzr3Hrwhg^Bg#K{~uudE-t1!3wm`Q-|vZ%I9N zN15V2m3!5}fMGr!T3Q#sSR`hWo1`8wWR-v|?@dw{C!@q}WY$GDc#=0f*{_tHi2jaH z*$;TOFaM5ni-X;y7D=>gcM8-3tReAX+m$_DFXsgty?2z*l)5)NUpym9H&g{-{Eki; z^&O5cM_vRA`>IITdyoG3sR8SvQ|z%oyOs(;T!%Z+WtrrLk~HEblY%?mQAF8oo|?E+ z9Xr=xH&7C?dAC^)>j5;kB;yMnC%N%!bZ&WBy{W_&klD5&XR!s?8%Mk<1YL@PhZ?Ej ziwt5aQ#V}Ml|bU%Wce(MGUa6H?ue4PQ8AmD4~BtpO( ziayJ2b~3Ef2JF5L<2}|!3e3$REeRlv5_C_#JVD8dL_pCrR5W)u$C{UB92Z!lnh+k( zjJPN#>*HC|6>m<^(V#vQW$0O?cUtTo01-7>X3VzfP7wvN*uYH1AKaq z!%Ke!v}##?@w=U3g(k7JsC(`JHr{c&3eMIuYll33x((nT^^^Q=dH7%72ih^sAjFK!!KjdVycg-OWbn>)Y)U!yh`GT0U^wR zKYN~82(m6AIRoD_m4q>%jT<}5%@5HvS9jgrS1rn$U)znu7iQ3vn_^Qg3n!mZg)7Zt z{T?r0S!QlG3~NXVw_qY!=8j4V=1sqcsL$iPrqx& zdOYk!5mB-X7rAnTq1(4OvO2YqyER`ng?fMQ@++0l$5^I$n+J01-$mxGo^f9)AO=@o z=Ufq%9=dpu)n}v@N&u@cHT3m5`s#+s7h?&txgln`{1(krh9t5I!>BzlY z8(~g|m&+xdz6>Z^}2TBL%oed9Po&gXdXUJ@m+^urmxK7eWD5rRWW zJ;o9_;}c0tYtrEvi63BmIWH2F=ih9y64AEPDzAivv(j~VTGkqfszlit)4GFD&wQWt zlsH>nyeEZt#uSM!!ay%22d6&)r}2y9TI!Ga-lA~Yw@^C*-t?FAI!)}Om(|EG+l9DH ze8wa-{!~DOtC_N8q2gI(;Uh_E8E;XxMCn<@ajCU^EsV9tt-y;pgo=%41pZM{0W_rI zAVl3{o4p0lUimX@6#o>RNUt5EMdQrtADFpw#m5qz; za5*}M%)USg_YS;NsY4KBthR*TI)4ch34rCfor~cdln{CIu?y}cTaR=FXPRBYIP`I* z^)Wu&*xBa@o3#;^AiNg4r@&ga_NS-E%94bYpkaSCM#I2pjNP7V^tmlfv>Ks_o8|EU zod*bHhVwecH?;*6PqR?9NvJII6@I1DaWhm_pQg9ivB$mu4PfYS z-a!~no%JK3Xqv54xZqYb~FaB7@&34()6_>a=90*L@)9YG+_;H*n( zkV6;)Z;NS6Bfe+)YWV@pkxIN9oy&fP^bPl>>U>tb*F<7(fsE5o4M5OY6Y9$ieZZ$< zpuNqAKHcIzEuswM^8RVB^An%Rn6`uq!iZNlZ}D4*dLjg@ugt5;iOR&C^TfwZjjUR9 z5tR^O>q)T=Xcb+5-KCzi6}dVUjUQic6|_v7~KJ{HQl14&0nPET*@V`6G8C1Y^`p?`Ptj zTfynVHkgdLf;zk6B$Hb!K4A?|)?&?o!OX(Y?e=L5wDaZe=<+HId6ON`P;%&8kcZ76 z(~*NXbZN@N{e%pj^f)X;YEF}A{U=JJ1gmc@j@J+L`<*n>oJdl>{SQ~Nq%ahV-^~#4 z@EsR??cvc|qJnHZ4W`OeYntbE!jOkfMzVw(;M~} z+1uUN3!0_hHeTsTB)?j6Xh;lw|TU8pjjIF*m!(fgG{by$@A1xfEub zyZIiqi9V^~r3x;8R%2llW?{sa?5DHwjQnLm*GKw*jGLUSVS|PKc3hDlyeEZvKpx2s zQ!26}P~@dWhSSEvIi=cV?kI|X!FxoW6S@4Lug;_@n%p?N!Dnapbq7#hnShWL>~I}o zG_yTzgJ3sNj*P^84xgxdD$n|kGx@1ZC5r2=`wTMIlU8k@*U+r0D}hwI{9o%rs#?a% zCwzsJZn-{A(%UU1&E5!hRZ>(hud?#lg*^Jrp_ap1l0R<_h(VI)5i?&SL9H}sJVpG< z!`%QVwX(MJl!wD3z~ZZu*cCNjwUv98lU1dMLDL68MIC2JMrxUX99ivRE@4~o%Q)JA z)fk*l_;{(tp%vW@q0{iRN9lVqtO&ZX4@6k7T^W03!No%pbVx6iJZigp(Wz5#p%Ue0 zT@3X9N@27|>b{_Ky)%$U0~v9pf0Sg7r_ki65Z;SK zDq!DGoX06SIR}(>?!+_U}JZVQpKA6hAguc3usd5ya+)c|vfgo?3#Fbf)xt}@G z^Mf`n=Sm<2rfP@RE->z1YLY+GV{W{34EnD=DdO%7hKmYCjSBs!B9FNY?vSs01{O>of;7?tdh8-ZyWH&8X_QN z!nH$FqLR!@@(s#viI@%hna2bY=yUkr%5I-KcbQFAN>U>p-vxlQl-BO@If*eoXem%r zmWOvcq2~gfH3#Yz#J<&m9;+3LY2a-C^15G;7n@BViDj%fta;=YtEY-D2&tt@bQHsf z)rzVjms;*x3)xteo1xwxjUA!VB)up~zYFd!>WE%h*~`y}WJCC_?ADz@)9odF?Xp>~ z#i;Wuk1^T+sv`G`HAr|p$;Th21Z_KHu;iP3Kee-;63nhhcpK}>5csIANajnzIN_y@ zl;8?ih$1xk@iGfa&EJ65&;tiAjJQV zHFZ8FkBq!GU~`NjH`~N)c+?94MTHI%dp7}bhH)9h6utT%44n#Yry{dtA!bnTr1dn_ z7vqV_Q_Mzv%BAX{^`r1WcED!gULA?uhHmTi?L7;sDF3l9V3YPFO9Yr7p)&3A7l1FV z3hF=Q@3VVphBsZBViQ`QSotV|XX!Hy#!UE@g3v>vk{=tq-P1`_Mqx;z&qvMR-$I_$ z3V|xw07&X_?A_(77Q3%qf88I3gB!01dDG^Ob65^Z3L#^Ua||}lFA8}3FG78%dSwCO zH^q;kTv!*g*ruPW)sebLiTED0UqJ&Uq`tHW-rl3!gy1vpOq3ew!@MpE#VgoifjiOi7kXYwMpaBxDeJ1IM>_TdXdx&czXzcIxn^?UDn;Fd!Gri zJNBA6J54+t`vjVcZPE+cE|uuOz}jJd0;$tVVW~XEVVW3Xf*81bax)~kvbb7|;_`h7 zaOd&F*BH^V+Lb!V(zvW3IL4+}PNi;$g0t3|SAe0j?UKMmPp-VmS6Sx7f!a|qbMa2E zqfxewM(7xO0hn#T)t@42HtJj`q&QI9Ex4GZBpccQb#_8Iv!^i<0r%v)jHtI`nf6_A z&BegQ?#*rjV-Fo+Rz{o53*U(^?KeWpJ30q=7ZeY$7`q>zeEC7vkbXQlhBz1`Ri-V` zyCoT-xw3A)QBaR)7m!>U%;i$=dm|or0AJDOkxiCX;xRvk*{{puaXu8yNY5I7;i42| zHfucyIzYcK9-;F$qJ7<1!Bt({JpKhz+MOwIIvxW3YL_#*K8_vVkV-oosOO^3U zfoG5`jdJ5-Q9t# z-tmZiY`dYu)D*Wi?L**+^#XLGfxm0R+K=sUWEdxlOY)gieN^?L7_GD3=OAS(#@CsD z`#&?p^VYs~fE=ehW%3F5IUIPJY&t#cai0&AtEygQ8FlD$*9=BBe8RsR6>iC$kzgyQ zZ7I%stp%j;r}zWD`d=HoFWMliB%Xzb)6#AhMw^8%AH)XEbA}(cF=_Y`yQpjN6JIu! zmrc`_ZIDOvus8WGlV9qyb(J*VLA*_qLePDl@7NkS`yv4>jp}jq@tszx$X>d7JbDr2`% zKEpMOd?tl%i_8s5XBFf+X+gm%QCFy+e|~Kf+wcSE+fIu>Nu=o_?#G!KltQ4rE#A2L z#^Vl&r?+B5*B})Yrlnj$`)wD@-cNRe!;ZZ!>`791V;u2S=C;%UKJT9}6-d`ibD)^I zL4n^JNSiH-3$^Wepd}CP1-;F(;Ljm^T%2IXWO&>APgcw_QIVJz?(RALcj9o*m~-zH z&Z~9+u9rR&a~7Mu82be~6}^WhC7BsYH}19}uO1Lx@O$>ow6E+n%cyL4+MKMivzicQ zQmptCD6&jNn)4J%zT*Y#jxp;upbM)@uJ#jBm5?jvLfzEh6$z57;qA0|&xzE8K-<4Z zQRALRE3lK6cFWC1Qul)P5-Lo14t?cRiBY{mAF87cA_9q!Y*F*6ROV$k1B4lB|Ku9hG+1tx#W))UK^Ce+4@lllq-O z<2Qcth!jSYr}3A!9qLJaiun;%>P33+O9T7L?&7l_>BjafUe~q#AOGeeQa`jzVXy@*}RZ>(de|==Irk;cl-V!R+(T&-C-Bv8( zW(SlYUv5-igS_?ubC#7S@^tY=YP49On zE_vVzs&5fH1G%ZG^P*+qM`u-=ISN1`Pl(TYYNE?fh5*HTpg~AD@R&uC#G4=sy*9Lc z3?=_bq8WOgPchm^lNPsE5znpxv{;v*_gA#uzHtuAT9o8E@sAAS22piRQl3KCc~a?z&>hMO4%|T@|_0tMEm=LtcL?d=K^{ z8jae0v#pSs=CS1#ZF)cWXW<$mh`b~+eAyR%DeO!i2vYRTMI|NZr*>Y>FJp=IO{+qQ zSEfN5Dtr1|n{J&2*+9YK^%+n6ma8&9+k~e!gRNhFR#4vR@fCaXt=6iH{ydYLQXllQDFNEyh23=Q@7M$6E8e~i(0HjnA9^;m{^iU4kzqIJ=pd6jLnek0us7g2rWkPdx%lUQ7EVT=B z&*R2x&p`Sz<0!jJfX${&OFi0h5v`Y66>^+V1tiPH%^h26_l&ol{=AaZW1YnM2KWERM6Ugqb+L` z>epE`wK>C`{rymeeo%w1s!_GZ?}ze(tU>YNX#APIH~;f%wdlY|`GFCxsNW8a1=pNr oOZUD1LR|7c8;SmRno=yyLgEDX9c~e;bKpN&X(g$=`-boS2N)vZJ^%m! diff --git a/site/notebooks/images/high-pearson-correlation-block.png b/site/notebooks/images/high-pearson-correlation-block.png deleted file mode 100644 index dbe44392d146dd9f414762f12a226978d1e73462..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 280332 zcmeFZby$?$);NqP3Me2TAR>aKw9*~YCEX?6Gr-Ud3P?+LcQ*q=iAZ;M$H2hQFf`xz zoOqx2_nz-u&-wq%b=};1?t8Db_g-u5ReKG=ik~F0FkfM!prBw$ONl9?pkOqkprCz1 ze~f%mV`iIzg7QqnQdCq?T2z!=(E(^?X=93lA{88`@kCRlmoQx?`nxdtb75hbZTaU> zc)~L1OunZSGw9O3iXWdnBH*hoT+vpn^2O34P}=jI!Q~4zI(SNt*$2_AWB@7)PkU{8 zoOpqtZTG2{!$~0X{g#I)a4HSPAqftY9^6e?BmZ0C_2_8gT!8NbCgF#UD_xtL!CFd6 zt0?j9=LegM^1eND*4V6NKIiw-*}X&?M<~MCE011PMO$J32ZtJpE};5`yh>n@ej{yt!^rR~08_SCa z!p6dTF)=ayLELCYBo>r`n)$nVi68e$D~n zk${=tVJ!47Gtbw^m7K*Iq7~%vAKQ}$GCWG78G!y0DSdMEQAGc9>6lpz6_>=$(=vsa zN-dc?ehmr)!^xRqGQqGitZnGIUF=n&yP@t%u;Y)+o;Pg} z{G#t(!??5)p+7xhxA#^}rTEnRK8EB~-nS^<%En!rKN{f(o)~Dfyz5j^9_4 z&2B-XV9u~|QTl4#`$RJF(jw7IGn})Vy^LGY^UF%C>sdUVZ=^ql3=1ls&M1{WcwdRH zNjVil8nc=;c_!8hdEC<3v)=2^`_t!w#Iw3uZi{D<_l(<1>2&jHe(%z#ZyG~7Gj6{5 z2Ak$#DEVhQG5&7cXXdo#L?|Z9M4h|4^0hZje%tll(5?FizKCEyKR?Rl#u0p!Jy6j6 z{h`ZHAchcWF-oe?K#UCLi`UFe&sTK(AG;(AzJI*+|UG6U(`Nll1Rw39{ zuVCAu1+5%$<7D^`cM+}c9+J8WEcqLB@Er->kod{Gr<%d%$9Vmd`y;O*eb!e+kr@eY zG46*5WujHmD5(i*ZT!VmeK`sflzillz4&S;|G~szO$*92tQMoY#yD($yr{?5M6NmC^vQtW~XEPc`I?R!C33&Lb{L%4yUD z6>Omm0+^!XcLEMqMEXFBYqfSF0dgR{Wh%N>kU znX~il#gXgW;SrpP;W2~q<;?pZC|s3n-#8PkD%)4WL=NoWtNI;o9kM|OWZu)yIcnL$ zKBO8pbXHDQF89p5n~}1qo!YKds;DZj%COHrl38Ex5}qA^JfYX-k*2c`15U6fa-@WI@)0t|SZq_>1Jlj0{^OsR+r(CFZXj8n2 zY`$sWrd2;E{-CV146lq(lS9)|kEe*3&kBMs;2nOg+bHTVpdluU{n! z*M+n0(ZiYNbIt+hAFt2O;gsl<{1V?JR3jTUGq-*?yt6BG+IF09dgj9DU}NXWGiGFR z+_OnJ9DcL40gHv#PVOwdD^tqK36bCkcZb_R2SDYHt%AwCycVaaqv1IQjWbQ?x||IQ z&UsA{PI@bIdsmZ#i}Gvg(5ZF6YWkY~rgv!3V3-lJSvrE;^}=J|h#Ph^IzGs@3~GZK z=ws_^>GQQfx-}zl!)AH6s=!HH(;hW0K{y?>o#rtLK@~+T)8IKoR7NA&GB9PCEuM)X zrY`hsY`NE9&8O#HU_M45)Qc76DLVE$EQ_ZTm|-}&L>KRS7~csLau%}8aJWhw_CEh8 zrP*`*UFS6=cBD$j&Qm9xG=e$2d_qE!dgeW)2E7KVW=lq1PRG5FDg-4Xr5G+&6nQj$ zBgL^utkXnnA?4_=?ZWN8?Sc)XidM3VwwtyK z#N{gFH2v9fSV)-uTb+ha7xZjYAmgp;(w1?4h*UDO=~obiyZkn!R@hat?*GI?BVT)SXi z-hlm9%TRF@$aYBFOgzXYZ$6YV%M;3BePUzc4BOk{*#d>Q6N4Io^6FyFa39scmH!;{loyx-{C(pvHq^(#_<}fXzsaFg>NV=&EYR&3UyL4ZT9U zqM3%|YHbL_SpFiptB%k*q3pJr*?2Kr=DtrrEl@4Ac&P$f(FSP;iY`jzgqxBG+2R2p zEj+FuuR1lSPQtZ#e=wD7UfuG!!tvn5i|!h*k2AmQ>FbucpSjsyEP5)McP4dvdes^j z78uc$+jZWN6_k?FoQzFs&tH98xHH;k%-bFRrr41wgBd z`@+sH(Y?~nXP_(~0j8s;gLAuob)CftuD|Nv zPavsq)wQ%D8~45`KA2RQMlDZk8EW+2k@yvUo^r$mTu!ztS?9cWMx^cWRqPWk+0QTpqyQ;^w43)KuMSzKAOw+!xaf>&rZs?sWGH8-^Rv0BkHZ(d zYH16heJ4l5P5R#qJolL{RZfPB6F6CL?lut`*2@j8i%4=ae;Mty2-%lhZH?h5aM z4jh)eybZ2I?mHqP2O^<@)ZXsrZZM*;1P%k+Hcqdn*UmoV=AVTJ;_NotP+yi>Xum?K z06!F#NA<-XHjh0DImu`y4`C~L?RWo`z+BAD{$VEeU7IB0XIqL6%HASoq=8}v=L6q& z^eFakP)G&b3$fEDH=r_FZ-0V*16$v@i?58k0X{Np()3QCYA%A>#c`Gmax z{(MJXzhVA%e;6Hz@&x$}4|%y}qW+^dMsw!Ff3(s5fD};?m6k?!QZaTgH3fhyfQ}8d zt~baB7fhG`X=SQIB>ovoRZT}tc{yHVpe?hZ3DC%t+11t#*&79g-<22H zv^8}!BzLv70f2a21t|XL!HaDFe#}Ba{zn%_YXJ&Pc|~$jpo1wnC-Zye_Y{Jd0K2SyMy3BV_9I6UKJ5DN zA5#T&#~;NX>{ykuP4yQA?SIxp?z;Z+FU#jY!2BFVP}+4MGxCmj|ABP>4f=Vr{jLU+C@8@M8SO0*AiD{rI1V{;Kql>mgYY&+Lq!_VurV{DU~Zx8cM8SknHx z4gU?v{<{tTr6T$7HvD&*>%ZIZpBm}U((_+y_-}0Ff34xawCDe|hX2li_^(s?XEiKL zkAI=Ju?*Z;_VIpydog%(-G6h<6?gp$_S@Z0rF+Dhh$YZ{(Qp2Z$S`HB>Q>;ftea?N0Zt^}GJRi0`UuS7Gjzn3sC}RFsCuxr2 z=OgsN2#hoAi)D)cd-b_KsKi5%&TmTqIW`ji6{f*GOkUK#=+JUMMAsYwbPxRlSE}G6 zBn3T^?cGHWov&VN*kJq{z4m8^8yRF?=-mli=I>Bi%8g${WPAMfy!aoGcTsp;OxU-}s*YI~uz{BHWnIzj)H76)W=Fz6wL$+7+4f zxM~!i6SnIVubFTDMO=PtUw*XFB9gIgPX=Z*>wgDQKP+WebSYi?Gm3SBR+!X|Y8?J- zxc|(<|HrcbND^6Dj;Tn(x<&j67x@?5W&-KSGSS2M;tMUFzeux{0rJCx-Zb;B!0Nc) zOZfysC1xd*pJYPH#D9jycLhE$Gi*bRWZ@OQFj7bljO(BOc^Cg1O4)V&hUg+gMfp7i zd}ILsDTb8TitX;^M8R#fQ)6RYK`<_wIz*K%Lrub{9=xJ*$dH9YjaX)ZQB__7|onE2kE0oy0}a*$dx{EE-$`3 z*^EkGT!j0aopC8&-pw7)LJc0l1n08q9N}Yv@6+%o+iuUB*lp(K>WX2L^s-;B>hw(M zI7X{Z){l40GiV`@XT+>Mc?r4{2?pLtx=xb>%FPyrbo!;07CaC9VyOEuW4pphPHXncxvgo}f;cQu% zTS`*t?sJ2tLFr|!@hvw6zsEN;LvA^U`3$YZ^!gG%JZY=MQ?Qa3)J>hzuN+ z#QM8R#z?Nu`xB4$?_A&upB$ilxD{$+fY%r{?9*{cAi{>J_wE;;8#m)qbWNqxzHMni zVI>u@c%30qqZDXvQ_>WNQf>yWs{sz<;nOKTTs`t!i|-m8mZ*~eE!4r*IlPa=UU676 zwB21<>nh46y;m-8y((^~mCyNCd?#o69dHw#qk8=|Cf_}g9FHD;pH0|$u_oJVkegZp znCv4kf#oWdS4+{f-rqgxjRH56zyvmBdYiRk^h~{;Jd2qH5jFNX8XZ`(^wmx4y?q&7lsAD2_5=KD~)g$kpn{ zE>}Ojs~CPDc18K#{LW}+99USgp2Afcy*FFk&=<`cm1>@*187xBW-n%8O6G{I8Cuh# z#XG0&SC3k!75tWKrASIpGF7UtW~EwT^r$}J$sjVT>HVmB^Bv)8z9~Vjaj~+KC!)>vs;*0)Z`ndfV%NR-8#wSsjOK&XawWfeG zuP5%V>iQWagENWk=`VI@k!os_u(=K}E9Q=Xy(-`c1lFrI7CY=+1RuuXCFwML_D`Tqd0aSm<#L<8(1GX8)Yj8^jREV_uCfs* z{eh~vb`O&Hy}#;$1}B|yJDxu>IS&?F^5KGNyshpLu0_40wNe>k7)Mqk%Z?GLd z-&dZz40DN-QJ0xyu#FG2XzNthap>)wK>UZ>;fsURGlGh1O0lo}!RFJp(VXx_W-IV= zytMwpO<1b!SZ$taYJ@+RX~N~u{O08ugSder-t%t+ew30+F^}c7C1ahr7I=XTcf49h zr~FCDj}u z^ueg5+rXZMOIh7c+NJr7VlM|L8dAB^`NQT~tNG5!aq6e$QoU>9dZweF$X%b(Xq10+ z8iiW$VG{N*lq58%NX(zS{1;A_of*DVs6QBE@kx)z6YNIkFwA9Tol|GW`?r_S4QY+G zRD5WXoK%7?Qa3tunmTYE3~a6=BKwbnYc?X6y<=cZB+~UajcE-pgM_OwZqEBFs!ax7 zT&|=ca%~2SmGjNwOu$L$B!37p3&$fM#4@K#F8afIfu>GTk>%E4GL=^KLefW_w94dK zX4i4{?`X$R3?qlJg=V4KQ_$t!3Cl)?&mBf`1)AO)TLwMey0bLNw~wM_R=UZ2qFFaf zdXZXtv&kZUz(qa{V(v#g8ty7r^mwk%xt0a|MP4cFbSDOIA>)M&s?^ACi1@G6Mg2*nE*#|4RR28rK zLU&6Ye=fNlB}*If`DN&>|EE5YAS)&c`6t<{Y*HxtDYr23~5}XJgs{4hR8q z$%t06?fAP^of;#<&&jy|$~pOrFJrgT9RM zeJy@0dqj_r-de!Gb_!;8e5IbqWB2v-Le>0SH3{n?g+P7;KNdJzUPc~OXSy^&O$Dm0 z*W>CeL0tK{Sl9D1uV^+;RgVj4vVUf1`*SVwdlj7{m9y-*jg|^Efs!_Uvzs1wc`Hn2 z7VKf&L%Dj;b|0T}KpMkyLfF5S$Dn&Qv1mJtG!Stby8HCzsHmgQx6fS-w2n`UGsQuh z%{mZg01>NKcK$YeuBD>66&jvtI!R~txh-8}^95D7usdP%PJI5k7rABay3YLYs_k~B z;8>wZ+INP#dGyQ#MmI1e7r|rp68qi{E0u;!IIRKV>22s>a@_IZqDJyx1~MB)e=l1g=wVc1liZBfBJuX zt>+%7mePm$nW4 zq|q?@o8H+a4mQG`;pKbs1U;`jRa<2i8u(-{w0r2(t9(?iIL_5Q9ir-0NukCZS3A@j=fXc&m9ZdM;#Ea&<5t=|<`M+@u z&jVv`p;D4;mzqJHdLehufdId#@SCr&1G>$8s|t+6f_Osg7J}0|((Z*PPnpmB^-Ewx zEOE_N(_ByD5u@ALCRR3fq}b`Bv8ENKD<+}a_%Mk>GnJXkIY5zGMUj9d7DFrxZB2fp zOI#-i$rb*ce9w7H@x!f!WABF6 zbE|1d+*qwtfWxuSiHkl3FY3b-=kGwl>yLm2xcbfqQ?VI|)_WMq&U9tCTEXeVxLh+5 zL{oF+(}Jj^CIeS-!iYqRGl}6k-A>HQ^F4zYwEeQ_b4@z=r%O6AxpGBo z%C%A=!F5-BJpN#Xa25S5BLt+L8knwoya($nXKlkEplWHo#&X*oF=SXAf*N>QDy!(z zf|uMw7XZu?kTRs_8nv7q9AETs@$Mn8_X(LN73}=%i*&dfoI@kU%Q-EU^Ysa|Kkc2Q z9$=7dS8vy9qO!b@{4gg{C~DGrs`t7(BIVTE;1z9jjJo!qeGdLaM|r?q%Ef9BnYh7u zi`RwT=(P-i4+!Rz5i!9x1l6=IU zb?jC&agK7<{+!aNPce4!$=tt$X3*N|aDnqVZet)kmfE>2tvAZ7>2XSc_t*4DqnxWf zDdW(N27MAAIw_?U)2E8aVat}boT?Yq(7IlbFlQU9Rq`Tc&rB#`ocML!{YjHfIr6pF z@Z2)YH*sC57_*sBtquoMS()?&o-&C0>o=aewlRPUw#)T4M?1EHh?xF? zFV2M{w}sbN+a3oxVP}a`FXEBZuZbJdRQW|iIEuM+nkAZsh)5Rnyjbm?x;jgC4(Aba zx{YI4u*MQCyC0c9c&X2#UE^~N=SpJriYT#K_3wH!X9toHU02P&mZ#)FJkZl$9B|(A z)LY`1-fs0`slq+d9R7hdbwTE>Y;!A>J{Inz>k0=p!XjxABt zi%7JNBz0M`PieLhql*$_ET8PfZSFL>bq{wnE4$V=Bm=o;jPDLFwE+9fX$VRMWw$xB z#u=tvGi9p3@?k)aQl^}=x>-1c@Xi*PZ1!nQ2SB~E#B37PPyJyQm^-8rJ%!Lh!N^pP zeV(oKRB1rtCe1FlM$n+e$KhM(ln{%@`F@SErZ-{{;SiwjF(6~;Vx$$kct7%sAi;q7$rAbjx0%Kiw4NVLM(cW*Sp|D{`Y~VI>I#pMolIj}zbK)8i6n{Henwk3OmLRF}p+iZrQWk4ms!%aovMK~*#2&VIU7 z8%L1bSOVKVE+AQo`8uhF=v%(}wCie~Mk!PV3V8z`nd3EsvwEeMsMlo+?C{9}GCWCc zD-;jPbie>`x>6elbXtB3{6*h-zb}1KX1)ljp-kC03?msl8Jq{US(m9-s9e3?Lptya zEZ`*vP-f5N9`U=dRjE?aCGTsv%WNws;kBTrSVW?EamK)rLr~+BC5IDvhi-V%05Itz zt`IFRmx@XtaOt>uk8`?zcz~?Zn`4*{!g6+(%mQTI<(bf0elxwU7OBIx#iVJ(GdDMW z#a9S`C!gnXFzV!g-3D?Y^L4*N$3N$5{Mb^2wQ-wmH5+s}lYOGZtQwnV{kfcGUsZmV znJEVtfVJWkQt7S`_P5u%%Z>P($JXVTP0!fgJ`X|h=kjUSd8cn~h163o=1z*;5 zhHhDV8Y^4v-tbu!zM%g;td&}Nw1ITeI}y``ELYhR*e!|R;DK${0=sX&= zJj|_+0hp4>?B&3`@(IFoC7845~j%IM4C$uzS zz6Q#S=}*rUG113t+x%$|JMu!VE40@HS zk=i7X=_Cw_NugTU5-# zN$=MXG#)D|ul)tcf?4u8S*dYF1Y$A1H7Ru0-uLR&Zd7)&T*xe^W>@wL=AsEFdyBJR z(2H=-GHMaJyBO!4y1!V*ImoU)I7|5QmHlc=8*E;M%Ut}DU3J?{r}NZSc)6PiSG&+% znxL6(yda?>>L?~(>myhZZsmr|8swTIB)tNF{pzJr%!}?pI{X&tDvbDWFX0NN8Yp9Q z_%QHDzOH~_lIU!*L@hpkkR-s9*w%%aDGk_b?>Y{*oU9N19sRUeI*i`m(H=C{F}#O! z;q_ZPvHOEH%a;%IkACJ)#*aWGhwnsw>ZJswThpJv|QpVEuW~ zPB_?CI&D_Kwn)um1`wMK2WYY8rz(@%8Rwky)992wlshlhPuQsJJAh z`(z|jYE)`%SugI(Ij<3%NA&7t)b0!OS<|!Nb2z$(Zg+@R6dpqBqoc=SFcyPavxb=* z@vaV&y4NvhIYNCAnA7zM(*vlcp+m67$%Q=-#V^?j(`o@t)0RT%fNi4k*S7#nHLhfb zgBs0RWC-+)$9f2Lcb&+q9G!sP&bt)kLCIQzTO3lNi{=d_Yy*fa$`q#awwpZc1vd0C1=g5r3o70rEjp~ioSex;? zDsqV{JWYeg9-;=%Ic&z}95N1vo^q7!Q=CBMuo&L)x*e14@O>^E(5YlsWoz9eovVv| zqsNEL0NniSAFCDjnC5pHzWl{)c^S}J>a!`1=r@u_%;ij@P=3NBq`s`n`q}DMLrV+e zy%_7C&y~L}$J?h)0ed-FzZwCXOBG;Lph!k8_7U{!tYHFmyvN<2DWtLYAp%+qZtYA( zGhP|D9D=5_-kXQSGU4la+b>2!gwnFGRq3-&0@E|#e9kz<2W8Myb(iW(v$o#}$C?=q z{hhnHc>EN-LL(k0cl*{JFm%Umtzk9ZwND|FV4r)!bb011TDCO5+&;2)pOKz7n7y1G zTg;uYI40GAO8oGA5WFlMs!rMw@}y-~@$KfbEWva=hc8(vKmv~%)M>>J2_w(0 zF_yH!HWX}T!#E7DYuXDn-6xOHvZ1;S-?sSGt2(QxYs5)kiw0ntsV3%iyX9t4z-fcd zBc*Wux*)@)P-==2XCOy51eUibiXfI_nAOXVyC9Dz_ zu6f=&TMu#D2_lB=5n9I5D@{PaLD%1@;)h{K?M6*Q74FCC=h>j zl>+a5d;3xFi!~pta5ht#p&yNR%(=SjU-+7Nlyi;K`{Cs10^Bu1Y1~z_ z3DE&YlOXfSE*t2#*sO)PDu=VdksBlBEyrh2gB~s7p=h1zrMF4drkrqU6J6JXmHFe< zrURYO_;P(M*7URwBLlr{3Vl-?yJePo_j+h2kGI?;@k&bQxDh~~MQuh8KYjH+z-?Sc>2fi_px=#1mpt6#s;&RN^CtXs(6*viBg2zM zh`1w-VX$d>mvxaU2?TvE&qHGA31r>(ay(7+v>6-mw!JFxYV0%N?bj$y)U_MEr?90N zb%zKX>@5P}_@qM1%}-l?p7H>DDID^88MleLgDDJ3K8oZI(Z?1JHI+C@M7F#+}MQM0FCpr$K*F$4r4E0kj_-iTj{7g z&oEz--|_BA7Xmw#yY2E77}tdlaT*ImtT0W9iPk1ep#hW4D%BThr12~+Y>VO1o2gJ* zSI|(TiLQaFt*f(EwTkBsB$~nXM3#65fp_6+Z(F1WF51*2=;h?ywSW!8DGcKiw7ZHL zObAZ3L>4}$Nmeur$#5M}`1QQptD<9`Z5Z(VJPo45S+^R0GNb~Ii#9M10@TlpyyiqlDMH8eqvSf{Nf@};+S*uY?>T}M|Hzv+iK zf;M>qayC2%p7L}zd&$(TKYvJ}0k1yZy#+C;6{P0|>O)nxv!U^%bWEgtTqm@g-BQvZ;ms;~X0TjEU8wu?^s>~UjGACqdx<@tGf>BTq=Jo-SqEyVQU`!~ zL+jpcI)`nidmPjL>7uu4$rEvPHH0BdjIXN9{8Fu#KyydX10N|c48y3r=m_Nw@^C45i3`nGZK7@2H@Z0&r-?4{=_XP`c)c+N_hHl>m^!N#`L^wm%s5LBPj ztudRd2G1@GTqb!icOt9aYdxIuj8AZOh21zM zTivF7NS4s_6wMr8ZvVsf!ZZY%Kr;e9CfRy(F`l{4Fz>8zUU%N)Vpb4R*CH-5K4i>U zWjxY&7;&a!(&2Au749ZGmx>)x#G|Y?9kg-Lbz%n)IEYcYs<@Erb?ePmf8(WWaoVl~zN(}Mc-j+^ z2l{fAIN`qmv2#^PE`A5^$yzK2eD(NHFvNyDxs10SkiA4irhP{m_{-KqOU1T&dr5R> z+R+KX;u{?1=@e1qfu8t$+XQ;P;Cnh@@WBFga(*u{L}7X6*5P}J3>7kHnuE;d5?mwG zBxNsR>N`>(kQYp9O3g*|Y-aqjm_+pCTCB@&iBf%9Jr75_kFRzm^zH}2L(E#ASm@oD9r4h$Od6G#4p{x;Vs*M&KNjipl zR=0xZXC7WRgY69K*ctlzV-Vu0w&6F_HfvFMP;vh@?iW!;+@>|b+aO}Yh3w47lRNo} ze$5X3OkDk3%bvzVmz43sKXjDP(2}Rr^+KAX5?ADBEsGbCA&Lfeas_>0N=(hv$2qAHI*N8VtamY6ErcFE^D-E#B71bM=~U?P3dH-!435TXt(| zzPMo%q3A{G9`ACgIU)k|;uXbVGN-bMgqC^zMcj>VdEIBn)CF%$43@5n zHk%)$l*`u)OPjB5uYM_mPgLh}HwWNF6uzs=ZIoCBe0B9nezLQ^wd-$7GnqAArT~@= z=!F=xM5zyP!4pD2lVx4Rut#|jioxQ$&XK*KU9QIB07%!bU2W2$6(;>%g*|=yMW<=0 z4qd#qm=VU4iJjiii3(cNtnDRK=JVPGywhL9>im5`rFQuH_ zj9$lif5;!^4$gb4@RUwIG_HJGhPWug^BMsFec-uT`p{T${3Uyse`&;{5aJ8F-2j0G z7f-rUTFYA*GSaODXt@(AluEL8A3pOM;`RKTP@$t*Xx&;5zTL!XEy_M~)5!2ht1?*) zBUfqwc^j=);Wh8LJ8Ux0(cjdFR-81|1M6wa4EbkvpO6|PSFEs~Ho`(#T5G%D_or?{ zuZR5J8Rbvf_eZ#%_5wW47sR_}T@wX}U&26Aem8EV(71}WySdYA%GtB4^dFucVnU@o zDYOa59P#>5sbKFyg#jzVHXJwzOs14v1|xA+r!4ahnM)AN1>u3_`l-yfy$HvxXM%I~ zq2CnYJ{I1Ru_cGTH-NhtDAp}$4q5T`iJv$7(OmhW$S@m3?p>5!=#+$FQ!i8D2k>ne4Ui*7W4QmDL&l! zU!JL5;Y-=J>u^U!U`DX3UaS*Y>K&e!xRH(2sI!&q!j#Xpw?#x!vY1Vu1}c1SBlXci z{sLBKq^+Q2J&}}naJl9U@rYM1YLwvPuys5t3H44V!0atL{_# z5w}|U%<|J=d^dMmSz6p-x;pq4kAZEdjn!_HZm{csroVUKK?f5NtF`ffk2%T1a1uaM zr$ZeGbESHt_pNQv8HRgzQGXF%YJCm`3m5*Sf)I}9w{jy^f;e$} z$bE*@xcr3pCsFXO_6gv$x}@So8lv&CDF?HjUDamN-Hv75(>a2p?A{icx6$|7ce9>z z<{J-24mU|ZevcW)Li&7@iI}Yu9X(PD8`yStBbymfb0-QZmvIAO=YX$ zS2$6UO#o~I(G6_Vy4L9|dNs%5WYoNP0#;2UDzTlYUn4+YS`(idFV85Xjy3*bKAI^_ z@P4yLXF<(dp;0-0KDy?;a#lx^WQ5G649??Vq6s(S@Age2s*yt@u>}I|X9xnK%RKfo zStScFLUGV)tir`bDJzwfo<5n^xcnJ0XCE`>Gh`)?0f3q|`b1ed$?6qj!J+iHm=#3I zJS>A-FEH}e0b{k8$ipi^fO7VvV^PVO_*l;%!On$K#V z%P^TcG@a0ZB_|_Xk@y*2EFR5LaqZWc!mF4toe+?GHQH3xXEt;Aq55-_4pkOQpIU5Q z(SGVG?dP#ecTYMp0!>~~gUou1+roD$=3%b{rC<}qtBf*Z4H;4tx>xm79B=ZgGeX1E zr(?lo+MI*}cd1!BB7BZ8Tq<>&(bBBbgYgcn$ybPc_Bw{b?Gl%Y?$X;rob)Ahmlg}t zF&a)=-DM4%Z1~<%je_dCRV=4fk5Uj{LXXPuYTdBy3nzA!w@GYf@?>PO+7kz^N?9f2 z0ru1TEythXT|@b$N7p5)-$fp#qsyq2vi#iTw7{i$vDk7Fkh!i#qAX&!e?zSY50X_W zC7^7ch**kdA5e-bRX?p78dc1SpU76j2K6kW}{6A1sZEn0qCksYuV=~(+! zzi%14VHvy<~&)#r|dc1 zuD#W9U&;-&CDFk2^bUcT6?s=`(7sYhD<0=qu9kMQqBOS<&0lFhifc ze_E;b^QZ?SZVgE9-1Xht5)H+I>_o7fl{>+XBuTGZ{b_lPQceBwnruPq_Bd$wBFd}W z>c|LKA*M1hq*FPOUK>astK6rrpq2|%M!pUeI14OM7Dxm_G$@2;@cdiE;eW4ZVfkE@%B*`$DaLeqE1^tU{={9O%qn^oeSV08EC+U+dxO-)9LS_Sr{W2ASAG!>dA>&$KBtr7vsAU;0^WZZ^ zX=nPOOPZaM0rJczVeuZC-SOpTg@=IN#3BbfD4?Vhc|^&+@3@F=8kJ~L$ZIDrmV3hR2@rOVH5PnBV@ z{2cnYrp-f+bZ4H!=}M(rD16JS38I|LOZ08Sb+PdXO^eq#xo##;Lm!Yv;+{Txl!E&N zpxvyjuYH!P=6JN3ATw;4v+2{N+DlxuBw;>Z^-A)PkRdv_HfKoV5;MVRQW^ZriB8%o zmEGq?!!U>~n$mT5(){~?!9{HikL=zDW7te)f|N70ld6(P==3Bdoj+It3)V9t;Gb^*TnCMCG+`r|xi`sQIl!PqgYo?*b0RbtkECj%@*KLBYOP^V#t3=r zx3{TJBNvwvk5o)15~Y!+8P7ILv)Rv>3{Mc;tkXoH&$vLaK3;z8N<#EAE? zU5oZxB9*&AZD+e+&D({nn+T$c72CK*{i2wp){DCJ_E!j<>^IPkA_0XU*@UwAa)E{E zc14OlSIi}|A${`OCjy&$$#KpOQ>Pej#mTdj*Q|$z)Qh>JCqg3Ph5vhmA-%%8!eI$Vz@5g}f+{+Q}~E^-%H zc(uNCBi2d!?Itbw5^lwWXToDs_)kvWhSpL?*!&` zDZ0}t;PKjL3qt9qt>=5mdD+A6-dV%wlU7qW4lHAkB0AS`+w%F$#{Nb5>)bU8C$c)x z&MSGHTCV-#Ce&e~J~Mc6r-;9!2l3+l6@T__Wyf%f9BJCb;v}&*!q>Q4;hW%iK!^;% zK^(D47sTxT^kg`iQlUyX!^x~6HlaMAL++RIVLswf3m5l#8rZp4{Qf#a?S;bmId2TT zYp8j~>E+pZshjS#+(Fs8#1p>O*;|)Z(v++z=KJ#JE!^qDFlVx{nicu9G1XtY zVWnEE{l%sy{^^K4Uf6jl1y2P&T&5<>`;l>wxA5`jG#aERzCPLzri?G&p|7EJG$3GO z9?S4PZ&b@=ZevQ}`duMAr-mx}2>9Sd*uN-E{*La4Ebj;axr_rODO^EQvL-4q0(^A- zT=E7y)+;^2?xR6u*5-}&JbYOcb%?wEO6B{i`s_7^wwdG8S8mQjM%M9|4XQnx&j>n; zWK#nn9Znhi6z>G`hps0~YF^roeVYTQYsVuit>$p#{fe;7i4twhyjNdH7H^k~7y7l% zU*vokcsbq(F4R2aH#W)nN$ms#Nh}BH^y6$9Pn5{R`e@A}ux`pGv?Nu^tYn7dX4zyn z1M8UDlYhq=FjziCiOVuK`Sg-Qjy-~8Z+>~%`%>*O@rRU<Eb^4``fI=y9UVR#i51hP`WjH|l zO-=+NA6#ZMXTeTEYWf$g)n+6M2Wv(x;Uc8pL>~BqUkZoiVmP5A>t1!K)N+dP%U0wk zhhc;TboB6%go`UYZG(4TpSHblqIc&p(PjPZVU3=81NYsi508hC=cF1+8kaKr31q<- z4tu5@SUCjE@=1c-(@vk#o=&v?h)B?1qQOdmusn|9b}mXY4Q7qgW${-duXVpfZQA*v zI4&Vj~MkjQHp3&d;}Hf2*TRYlWi^7eltge%&KiRc+9T2Ga~jKJ}7s4b)nWt&JF z+D%+f6|%X|WGE}ickeeE>v1#QRxvg?p3o{^+llFx+LYPUGhGTc80>7a3qEG1<&+Q4 zEs{)IrTuY)`7G@We$*V$*YHTbGnAlSjOF|$7kjcU+hh2yyz@r6xsjd<^l`ZNwOdG2 z>YyrjqiN1HTGO)C&1GhhS=}9G z9j9WGKTH8LDl8XtF0@|hg2rFpT~zOGD`pU-8rcyfT8HTI{m%U53D*~JFcsyqI78Uy zPEth)6idF(5$h-9|Giu@Rjt&cU`ez?yzsQro#`Bs7--!MOvmhJw}p^ic@pk>-=$Pl zdRy)~Ynjzg(^*Yzs=Ytm+AnMUNaSQaw!5fZ11wQ5C#PqBMx?I0WUo-VsShd-=n6r0-RRZ_OyvmbjengE_^=L(V@i8`A%%7EP-Qtd+9!WKhWonXn*8<>)Oh>9CJHt5UWr zKADYSfbyF6Zs>nuGMv=f0&37dQaREroH^g4(fVMi{@EE>cThCLggnLY^LCTm!Xlvs zCbrEpbm+5n^_WJK+ATroSygCWalEl)G__K4(SmTXbBaOR!EAKwvx#!-KGw$B^_a38v z^jKr9)zwustLA*>^USI-Nqc%;0yU~2sB2yd6&9HQjr4hvNxHru0mM4=PmI-x&J?WE z5;yS6$SRiY`bVYATV(LqdH))R7Zlk)zV32g1G7~U_q&>l>EKkk_j#@~`&rI#tGar@ zMf2IYUAHm7k|?gtL44<}1(;~1+%`QS4PDLPY^NFycry+!G zSM}y8glu|mM6WLsF%)0&QMa@^^E;is1vC$(wLcn18a2RxCq(;cN{yHbbW%;nbXSw| z2}^6TXjRnmEiI-qR0ZB>(#+Ap4e{NFM{c!52}Mtol(>|v)>s(_u$8ubX+9}Y$fihB z*Osyk4>mzp7JHNgp%7CWbhb^F1=v$=7f;*~=D=JkGLg8+fP9I;LaP4qq*>R}K$`l}#y4W7SJ=Xi{hMlw2H3^d8h(>UGN^pTbd-HITcoaI?VU6N#Rr z*ZFMO=()ow3a-U1^U*rIvk9i59vI24Y-LGnH0~Jl;cqQ$^iO(%K8ADSe%<#O86yGc zEtH9+(9tvP#kZN!9@n4z(NnsDeYta}uxG~hkJ@Y2ceKt6Hr|!&C`vZ1KV@cktzPqy zP{4OJP@HYQ=$tQ2mCUR;^b zju^j(Gl;g}ut^kWFa`5_8?HLYhv8jc8Hn|Oyq*}sx@;~#A7#MK&AguYh@!MRU9im> zQ(st#{fZ+$LNG`~t*B`u17r&oDjaIeZ6BHs_WO$6-ZITkU#Wl*0I3}g-R#!>*`S^< zcjtmN6{ie20(k8yOQ-x%5) zU*#avE6$41R~}v$qP9wsDZ}XUg(Y08H13g8oW*y}5uHGygz-!e0)4;a{;+IHvQ&}H zqY|9Oe0eGjtN_iFM%f|gQd+M_*)!}TqP$$=y$d5zO8Y&@o;>S!MCvraG!3yG)0zp1d^RuadBIkuIgKDey} z=am5D0nqZt2P<&gY<<$6ss<0l_4EvItqT}NYeE*C4nldV4AG50aU#;u{( zYq%*Z1?=|ijW3o%u3)<1`J8<@vctfBg2|xMei68_%4WfEXi1~*dOz>|x-2by*3?AV zvXYyNylw20g5Q-Mp+yUgU=LEsk+lCYzh^~7v2=KHuY@| zdZf6y=1ri&@TA@%|3SNOPZtZwbb8cN=Y;x+szIf?^X%vXifAc(Mayg-WXQSLTdw7; zXW9Z6c1|ubX9J}N?)O{D=i0zAakZmp4;QW7UQ57+iJvh$CpN_qn|!o(5j_-kj)pk9 z9e*X*jQ@NeAZZt4WJ>Gfv@#Vv!T=*(Z&8eCmcv^Mb0xavpV@enAt|Zw(7pz}YN4}q z-Qn-L%R1W7+Wl~L>H&Ri^E}6HlSyx~)T?sfnqXA*2nNaaV_G7e2&fXf=?bfcCHqum zT$8OX({rt_4skq#O9&%5R0}?>uT|#MtvDYpd^?5yX8s+qw`q+Trrqpp8mkE;1mN+H zW)VK`lRvb`t_`O!JCDemU)fK0HfC)YGqz+61qNM?OVQvuuRncU^P#wkHm9Hu&{I$* zioVwGOmD6z-i_emj>-gIN4$UwFBmo;WE(lHeZ)ff#QGTs6jb9l}#(gJ)`)(-W?q+6B?)-#={1(daP6#(Fo&BH+9 za-+PsU#TbAHnK6YgUC8P<6F|j%0^l8H1NGWVw`po%%%NY!T_=_Ca~3KOrC~Y0Yh)- zS9gI9AYgCOxw4GfbdNU&2w&PjYXY~=XLD(HM1)N081pGdvRY)Xu^g$^)c7ojT)HsxSE5ZF@=_@ZW{tjpgCi5_@!x!eN2vU01$E18>TZ1d`4KbRcnUr?TCX2^ z4Nfg0EnR$k^9r)tZ#oYm3_b4Z$DQY_67Wtk#8aI=C~m==Ru6G4)q2+I=OC{F@@~qn z;h117Q_?9a>lfy>+Vy?icYGV+H2yASJ8urbC8qVGV?p2qidL5pTQ#2)v4@hQ8i!ph zou-H-d)~C9y(?#h_Sd}|Y3Z~RVU46minnI3H_2fqV7dNgNl`&h#rErUt|5;R4YpUQ z@jwSr1kQOBJuf%j;#Dt@nElS{A)_vT2wv@~oWfJ|=QgSY4=b^@->tHc{NnHA|IR`k zb;ip8O|jafbB_$sCm6ac*-YVp zLaXWK%*3DpkS8eE;b1=}Y1WB;f%*FWsXJxlSFXi4Y7@oMqzT}kiKgBC&83LB_SvL% z-a=#t9vhy-MXz&!shJTTf<2vV8mJQP{G0np(2*9+uD@bi?2>on1h6M~LDtS8g7;TO zYZ;)uBfj(f6dN9V%B;Pb>iiH=Q(GP{_3^obZ;8B!j{>`?y^A!*IJsiFs* zt3_Q&A-c&kCdz4e0g&*kz6Vgt)@d-H$Y6jIeO-7iYSRj2<3yT^C zYedGHbD1yEV0D$ETiaJQq0O=IdO0IGlvIL4Wz+Qi$xXs8>qSWh^Ah9H$u~O4^CQ0| zYgdpiN19jq%$VG5?`37iPJUv6_`SC(G^kQ$HjX|}0YjSQGk#ZbBNG2>iOv%nKAD^J zo8(X_r4Ikwgj%;^F^m)OZ(JQfMk=QoDTK+DhIkz zQk|-OZIM{!&^S_1GyJ}=OG_S+>vo(|a8{a2)M6cH$h6m3+h`I_s*?b)Mu*ttk6%#+@a+fL`}NCE zxe1)UfC>=WMpNAtY+zYUpE-*d{7u%?7_W@qKCE1jGO|}>Ce=L3p|FfV?QiiMsC;z}%5HPYN+Xc901BzTz*3zUAY5&` zs|RE~h9^Ipa|jyY$6#xbOx!A45RfEyuMo=uB#9p7JFPqwZcg1S(en0Q@PlS&gMb+m zWHO~RM(?ZlkaU91`L5of^&`fobCLSp?giXhozvykYeu`(jMt}B%y8e$FET=P%G?w! zzMpfq+QF8}%0YOoSD9s_<%TbQ6Pdi2CmjoIbx?Fbo0^u59|bVN1>5FKix$qB@s>V$ zG0a|fG4*S=_;PpaqQOqFG2!;*U7SMp9c&4$$9XN;M;263%WNEYO%F)UZcDF{jl5^gAB8sk=Gyj9J}LLy*!_AMtF(K?zQ#&ERbAxm zzV1`@5|E^b{angwnYQ{+pKtw!%*7UAsFX+hvV|%aDah4q#7+}@F)3*N+H@L?uXUxt zpnEu(s^iA^bVk^9+eqMe^GyoQ(Kri7N`I8bQZVry$44?)D*Qb!wDI!p`h)|@3W^-E z9S2*yehjf4zTB8Lq-XY=(B(A^!G+m1^l7Lc5|x>$3Q^i}>arp^TtlGjT5!Etb^8vX zbQjVM--i#6E^aS=Q-KGdXta=3twDu-EVO_o_mBCOQ=F+l(i{IvIA6RrK%VMKDrhuv zH=zWh07OP(`H;C?i21-SUcR+FrB4Z|U7f>(gnSCKJ=^_IRfqhMI^l0!?X%g82*O(P zAy#4&@9N)r8hJzk$Bo;bx5&3IVCKoh;E^!xp5>3-lzxc2wd06S))Rh0#$HCgcV~QF z$%tR+rRT$UD1|nli|+7g&Ou1-1Ns^L+%ROzMavqxZb~v!WtqcTf$CUSB_+SWnC*S7 zm*c?9?PTH=rnq7~zpDD`3)4Cql&AvVz z81pxG>DisxO4~_2bPh}vnBsq(cG~M&kaVJSlK^mCy}VjWs;Ay;esZMYErmC=JMMpGIA9uc`%c+rrkdY4vKg&r1vM~*e6q-i7t z1h!cdwxwL}R}PaCEi;LYy`@O`9Cr5tteWlHu!@((G>d<6WJpd3Y}&T(J&44t4}MBX zWY$tV-}9NkYvP}uOq@&8pC?Nj(VR#t4I7Rgp_rFllRjfr`)UN11U=T25sw11%|)Vb z?YlP`v+Yc`w0s!VnwioDLg)gBoHc2rEIPt*_rw|V6#X0M%DpxQP{$)gXBRYo7LV3d zwKhu{FSZZSd)Mvqho;FpFLEi+7YRJw58JA3Zi_x;cAxiKW-B=sH}c{t&qW`ujQD;u zRv&97A;yok@w{Y|7mj27fN?=4lRkOV1trcpyWnDPLs#6j~9}Da?c5{K+(%K%V5jU6pV2(Hf0w-(glVo-VB~MY( znl@~hkBYRLyevM6N@qG3vIR_Cs!*;fX#d%4^@kBvNPT@R@c<(t`2KhOLM0N|dvvlrV{`+g30biW zIAtQcLcmR0?4(Jg(BuegF@j>MT{}!}@@`z9R8kO0Wl0?~5RZ+1=#68IdR;~8R9F|RynU_ydJ|I4>M2`*! z7_2y;V`u5A{ubGJ?cv8#)^eZonV_1nD_?L`uwp$y3zd2!S1WxyN{JEg0rJqrX>g(R zG^Njnoti2CS>})Y-tR1XR%2W#TJ+fUT%RzrYskE^1YG{%7A50H@wE(%JQG{ufzJxK1Zv#?5R^7UvT6aKnv$6Nkt{A@cJ1b-$ zJWT}*=!LL$?x26${Cdq0JE&S29<;=H4-^H6ryt6cvttG-zf<|L_R61=+c|6Fat)}L4N#{wV`^CHN+%UbfJsn@bOSJ*)bmsx{goBR z(SA9OXWU+6SJZkq4$VoN&|lmoA{n&=wRr7l9d2G=W=(n(3{$7&Hutq3!<9WecFSsf z`q`&X;ueh~EuIznWiN}pr**GIx40S&+I`iuUCdNT9g^;zhLf)d$%^Gr#etY39e%yk+?b-pJI;q5A@geHF1 zO_(-N&R6q5B9P<>X?=o2$F2NtdVBFZ4$Y6o&H7pC>9IU4fM2uN*M4RY4lZszZniBz zEQIGa6Vy6^zH6WeMk&tL%5HHMy*8Dqi60O_8ECn|B8@YUW>MM+W=JqDeh#bj5;siw zN1WDdR817v+qI%}Q@G>HLtiP4&|6!FlT|_bB<~;=*WMo1T3ueuN-^cu@%v$CymR|~ zPGQSz$=uy_Nx*hV<#QWOz^2zZWcM&fLaD=e>Vo`Q=qf8aML$#$w(2 z|#8NY(zKbC`m4=e!18|3Hx3ft`GWLTVdZZb%#| ztOjQE!|A+czVWaN$_thJdLG}ORtS*&mR(`OYBaEri!@5ol-c$ zeY0pJpY9qLmay#~8gt#L2bXD73d+uYI+7wDn3vwJ*zctU3K;EN;6^m-?ay8Xlw~t@ zik1XG61O#8+{jHNm3o~4yJ*?`onlo0Ux(#k+Em%#C1>u#4@(E<@=#B^ z$ry<&Bgnv%OYssCD$MZf*Z)%?4e;EkG5r(qT9vj%zsr-6fLS=xljeYYs0!I+UEl_5 zA=z)@?SS$3yGor2tI{xqB_5bfO<3nIktCJY9bwHq&oHZeRrG=r#yL05vAObhYV^Tw5 zPa~>5t1e|AV88o5J6fp^niA4EfYe1Hyp+%zqBmM^KSid0avS|dB>K!9`Xl4Uo=$Xs z*R%iK@FhYFbV_DUj-f(o*GJ!H_-;9`q_P>Ht8F!#@c`;A?HBmgj6e;kx_KZxOMZf~ zg!R!MoHJ)VPG)R5*|m})$NUSK=2#apn;SrR@FA(-DBM@&gXowTJE&Y^J-p&8B=?An1b?pSc4)Ga z!kh!c{#x?5D5gK1uXW2#66DUHi;1WK)LK-Han@uP$t-NnmP+GlEC*F;KxGV%?*MOe zB+T4b8DG+x(!-gqyIO3m|c_;`cWpBadA9uKBW?86U+r zTX0*A&(*v4-WKACYi5g_U^%D5&ObPKkcz*j;t>c=a9K2<_4* zrJ};e3`!jOrEfVCu=p<{1A!PSO#;wKjue|*(2$6hchtVLx~(_zTF-4(GTdN(r^Ok~ zH>t#>(;i>>eo=IBL=|`7>_XClVDLgW>zNS=J;En zQip#N-%0PV_0l0oY_W=G+)pPJUutn8hQr zIzwWK%J)+3n232zP+5V9gIR&7S*pgLVw*(mNCl`w#F{&$#0OhWwLLH)wAH;3-}-W9 zo1JVZpyGqPmSGuFmN4d%6zCV!@y)va!*Kj;CDGI?(nrZA^74qb*@G|H+Zq$2rQ$4x z?i7~w2ljN^&G*gAVwBsh*kac`P9sxCs+^$lmnHhMcimP0QFBc-Qnnmf7FI(*b5}}w^GXEK_>wF@8G9L%8m|S)Axd)x;2Q2Z&KnEDvBWYpAX^g#2qM+`@57(j#bEr%0NkhX+u~V~(saI0v(Qg>1Ju~NJ zaKish;6tvzQAcWq5L#h(XAE)rSj zfl5`bL{hBW>D=aEOfpeu{l}yC!R7})hG6BX8az@l{;Y<1sdZWtZwdFSbSC*Q<2Q88B-%lUp1iG+`HMd8Ut(r5uC z#k;RUz0BbN7^T}j+=wSAWoA50(}5*C-cVb<^lg?$L1Xs15chV(`zn3{6(FQGni`P+ zqVn1wJe1+fO2k%6B`w7`KS{#Fcj_ul9`2Axt#`Z?Dgi#NoY?GG0$#mlcwUYBo30rC zjsF$h0}iWs5pc4V=MH9v?HA|v2uAs&iao&x&A$pIUJ7Zt=RGE{R3sIXpnSFSy*n5@ zp_%1sHK^RyQSxgV3v{E=8N;P%RlTs_w#N+_@i=#Hd&*YfvUFZ2vQYD>{=-n(Y}OZ> z=TU%8(LaUFWVkt-U#HokSAr>a>5e|jnConA=ZmAVojzjQtG|4UZmZ2?s2dID(w?28 zI>d9Db_Z>zcq3sh#@G}@k9VKV>rtm!;*wqaQg3bSutBGBqqpM05+S@?N&fKa-WH)g zMmeX!9IhJwR6a4*PrI(9fV=JjrZyyN6ghgp( z)8Z=%vs@bW@yqsTuG+em%Li8=kLW{R>QZNa-^ZrP)1_mdSrb7=Xh6m1bPHlY5wg$$ z_MRSxsHj~X&%3CAYY>7*T&PWN7d_=2e;^I*Ri*uTxTR259MlodQa>fsOIWj$tS1^r zM4v9!jr8|1n2}dXpC{Wvi9&{u5yW<$@536lKmY0J4p(K~KH{MgIT;r7QZ7<_k+*xz(A0lDGg1erZaYe>Sd9D$c^u&JxcFZjg}s0XyoJAas%w|}tDFVDIN zCTcM)>=T(&7`PeQIjiv7^f`nePPuGRl6_Tg^qSjNzSz`o35+;f7-#98V*Wsb3Xj^48<$nPgzWFKsTJEMo;6r zr!!J6Kz5ixGd$Szdp>cIE^+3STypx^b$`&LruY_gmPRh*F58}VpQ^ABlvr*jbLNIm zy1KN!((ZP?Vo7#2-?5dWAb#T9<5 z>3sK%FLaBmQtrb^69nFU?Y@v(SzQJX+>n$XeH>SWOS#FRrm6pIXU7$mlObD&yAi_m z(j9|QiOO86cKKOA*kEbk$-11|5r&{ai<@0lCaNP{P{QcJqL=wWK<{glI|PBEHK`iDGXO$Y}Lt?8?HH>Q5!Yq{+$kqL%0*i7>HGkST9;h+WnC7p{ zS>P$TmC6I3s;Drk{WWWjT_Eh4KP5z&d4`Cgq^-9V*=+FSMsrl`6?8qI*C-n0O{g^8 zX+gyER-|DBRh)nRWxLz3Ki-1CdDE=xCH`uUq+|J;|?)vHN=N!=9R~kwLWiftG5>K|Lajg9e zsh8`=Snd*y77Ho3(c<1;2!+eoUQ*ecNNl}TW(mU96%U1f+ zRsJx+BF2cr?O>tCdGLQgq9e$x?lXG9t{{ zOIW|Eff`g^ zXh2m zwr(tuqLK#LVTE?5D8mY{B{*!M@RobXjFSTPK9)L4yS_jBX!|w<2eZUWo~32%b`XTV ziqohlE^ew~86FSYzigbKBC&pQU_`P?$GSfMBDa!Mi@`n+wwm7~m)FLd+0+}w(V?sw) ziT+Bo#&e39wgJUHn=nm_7|n`(n#xzf5{6G{nAPD%nvb~~jrL}m4DQth)-oR-yiy?Y zVNhl&rVj8fht^)-);eHJlQjx&4ksuGfmgiKO35#%vugUgnkLARo%kdVpJF*adSeOU zp(ZbO^z%rntP3E}#GEKpVzC^}Ob~DuOEG8BHqe<`)1oYSL8aokWampAz{e;<@=d1S zdzi7bOmLwrsJCdNKfA%&las;zPV?;LW>+XmMl* z-B~7N^34==6Kvta1CFbcr4j^3NAa#*-(J&k+4Wp6Wu{Pl(xwvWoTUP{k%KapgX5#0 zrS0RN(8KIkqyr}*$`=~K7qL$l_&?=+k%g5#!EKdtdWa)x_5ZZr{7hI~e@)*83h+w# zQrGGLxD=>tb~^KkO94BM@?@UJ1a4UldNY4+WDYGlDX%@{ucQ%9$0&9tA0;kWGpM$y zP)xrsNw+1}4*5n^BD9}p%Lp+n$6mRPeDG3s6zgl4p7gO9Pb{f=E<)YVmnR%Iur(BUG>fkw}`+VIM zqX#_osqPeAYhh&{SLL?Rg%|3*n)s`ok0P$y<{@{mFOzYuI>I2EA&;}T?*bqW0l$@i zP)DeK9mLf%v+*6mCA$a6lE7zH7y3WhO>ZjspV$}d23?pH)YYND%xSp$S{#(mjDsx4 zbrf_4FR_c<3qpJv#{*yQ=O5Wiz5Tt8txuLrsGqmX{4B85ZG+RP%Nj9?*)6h9QC6^4 z@p+Hcvo6r(qQg{URJK@SdG`gqG|_hQj&SA1XoKk?*UUuKpH5<~jla10c(JC5Ngl3* zgsXsula-A;Q{|>{sGfFyIyveug?qS9Uw)Wl-50fe=&x4ZUyA2Td)R8VM)&99emepE z-j6aWVj$z!2r>*$VJwTGKBW9X{Rr>Be;Vu2dJ31H70`G}?kzP-Vt1dRNI(J#BZk)G zg9uAdwGTr)pQ{n~+46Y5DHVBGLRtSB`uVH%Z~(`L#nlaV2;?h$o<_|^#H zis@)GuU~dJE$TC~eX!dvC7KT4n_T(cU{s^t)V6==n;>J7|_gsrARs2$725RYNOrh@A}T;v00hKi^Ch8mtB8>79oTgb`|H;@C(l6O@*^f*eQ z_v6mV_3x$4ys)I@F`xP`St3*t*9O~0Z|AqupS0LoicJcub^<96=e;0ho))z;IU1fy z9{bTOElS^OG#}v%^@aC~vh`#7F*DH`heQSekL;=T1NgYj`pN~y+cmIb(hDFAt8e0z zK$!W?lSqcT4GvdOMY!p!01U{pskb{$Nek)(c?nkIp$ANLd-Ip+4K1&7C0%zd5bWgl zgC`B%dngHo4iC97qMOW^kz9dVaHBxCIO=DY64y5|OI;sXf(#BVRIV)MGneG}Cp$|< zn2dcZg3C3#0#zR$u$86w9VCxTB;qSQCf8uZ#HJ9WjY{t*4E_kc>Wr2=TVn;o9I*BB#{xEw zo7=!7stcUvxr-Ht>)@^3(nN4pX6MWZ6DY4_T8R4~3 z5#wn|7jyGHTko#MxZ|&QhO4nw&Q#52A3KQr#i84sYg8XF|j}$wszX5G5}h6YEhjwps<;+b6#@ zmFw?c`A)aUFj`+m)P5c@_RX1a?KH@lpPdf;TJF=LJzh7)5`AC}?2ok_P8+52N_tut zKEtB>lHRfY*j$QlYm`TYFCx4jB(123mz#TQ(MDf-yv?}c39C^h;Z{k?V=^D z#%;}2R*tsroIIOKF^LrwS(I{igtfF7O3|~n<0-0Dg5nqm*;|#Via#zkNM;q^wb~jL z#i=mFB*`%AxajwI{2_7^e+q!meh^VUa@o*AU^uUVp= z5+fzp^$9Fo**|A%wh6l08+$DLJ~go+JA8U8B#H|AygEcbRM_OSE74z3^D36vAWpo>_FdhK{$NKe*b|$DZBN(vma=f}!Eq0@-{sv) z#xx}UbJxyPu)>A!Yag&>Gs8`@@`NMa^@l*{&o0v}1w{*x_^#^DRTvAo&c z@mwR@{yOcMa|<-sHw2GKRRTm)U$lS(li^pm;9UZCq3&20WjZpgd$+WsV9=UrP8_qA zy#KeqI93_FfhP_o&WulGm#i9k9s4G+>jQE7fYWtM@4A7C9toRv9{{TBS4{q29>Nwq z;+8-{k2zV&$c8S3_L$e;Co4#9_#F)tyBfx7}FZm>%9x8*&dJ zj-Gp@iLXdO)mX)RfKWfpsf2!ek*(NDh1K07aR9@pcTSj2`iQY?g@g;{0dlWJnPe3w z0TeM&ja2MP8%Z+**Raob7kkjL;VASAI{vCoYwlm)I%3`l`33NmudR=4aoBAaFijK~ zsEnS*Ic^6gnDEn)2SAQD1_$3IJtkHTyr*TKx$(-Hofwd=QFiBTD;UQRYNEL@L!Q>) zpAbOKYKwE%;k%JQU(#2ZzVM_{D81*jo^C)OBOb?v&7i~X8(rjpb2G|}@&hZV4!YaC zT6Zt8$7c-ISJJ66S&fZ7BsE$5DUyAiL!$R0 zs{OAoME4l5zKw2$J_^uyk1pH7?FKruk5D?gS^%{~mzIe3GsB(Anm zeM>q|#*Z?g`k(gfvSrsoxFuUOOXW=B*p2jerrHve73%Eoo=%tI1XgGqpaRg58g&KT zxX@El9F<&W5(@A8$9V&uAX{5EoC30AV%ahDqns(VU)E$UlGxH%zYmLGPO zUB-fNwOTv5{WLGDB>o%xTh;(8&6Z9BG5xNCE{cfDMO2hBYfrDac*IXAhJNo4IVmM+k1Nn$AbC+mwnL{mLXK8&eu3nplt4R=gK z!h8A%9k#m7-z9=(jb3pi9(-IXNiCiDE)(HEZZbbFB!gfw&=;(CH5m;z))ZK^84L| z{8VWY@&U3x(jJc*>@?L`IEX{Q`9n&YiY9TRV{QW3dx|nM4c?>!r_;kn_ARrGJ2E7% z+=J#@N0_w;7d8XXGWhrzMN@eVqFF|^CIwvXyO2|6ICMtHk3S5P9xv2_3RF99I|9b6FG)*R;v|D=Ma0iErY)jp+(Uev487oKAR@Cl=(O% z3H>5PC-bwEnK^RgNWYqy=&vnw_4`bnes)mWWSU)N#_UG$^;to(VAbPlJ}J-jZ0!30 z8~FF?gQcSM)KkGP!*^XJs)>-xp+Y$6Y(Z6v$j!MevBbitDx-_HcGY>9Nj36Y^>bxC z6A29dR}zX`N=>IUN)uEt zWUKVLU22`RRo}@)_dJ<7d`kwS5|e}a-c@9%O_OG!$i=^aUJw|X#Y`77SGAr7h@b9o z06XCm2yTA=RauR3v@^wp0zZ4FK=sGemb)yL^MG!js*rSO&w$3F>l2zdzW#JagxI7d zvbuuAciX5(J}tJDP!Hubci$+jtyM~MAD@GaX1)@)($gJ)yTsg6cLo<_+DLj%8|#%& z@BK}$jGRWN;!66wM6*D)&{H`nlg^8`>10IM2_a&>z9Ns!50vEMv zWXm-e8N<6ZmW%^)2VRF4NmRZ+rY)`)=i%?JKMeOz`joYg+DYhHZS8A59U1+Z`dUu; zOt_beN9ADQ`f7iDbGNPy;t^+o{piZlnP~8L1Y9ZNRp0A#@9lARvV2wOz4}C5Cmhwb#BKP)K)gryCcoGM0&c2Mwb-eYnW-Ic~B+3AvZ001IWQO>uLU*-R-$#q}jS4X6_vvCw+N ze!Br3W<+D8bcEDFc>b=W^Av2cSxy@|Cv^HULgu0p559!U)nUYIn{OsO@sCNxk$ zg3R>6-D%39<7cv8(;fKu3+M-|nsXHmT4lnl8Vt=vrA`L^_bq0KFXYAxjylvn zK0=HAta{j&Al`hw&5D&wZZ^bxv~6U`fB%6OIBJi> zHRUI6=$cXdf*G50yO!h)Dfy5pO*+`0pu{9%Yc;EXLBShHVAQF!k6N!t9(I|V5N}u! z{PQgNV=@tE8p-~)O>X~K-w=^}L-%)kQCa-ghs0HB|I6kMxdDqZP?TV=+1ZwAA^@PCVG9oG<B4uI}MKB@uBZt1GRz(0Y~y?E8A~=* zp0L#FNcQAOWDn#Knmep1e?LZzh5AB1L*~WUFwa)rG8+~ax7#0Mk3aStaM4i;=)2(O z|Hx~sa{9lO8!VnHYAs;{A{FrJ)`;ByU^#zCge*g89M`gaX!%AK1fN+b{C<*}23OXl z<%PWAHW(5X07>^n)&1wY_Yq#B7V7tneVM5lGKhTOtnzQTQI;ABPcH98p_YMtWeeFm zK$9Y_%*6XYBxe55$j?Zt$o^!E8+~G8{gI_L@&A}-oJe@|BUOnOdijr00E_taTo>&R z*yf*SJ)V zdJ4^F7M>FZ+r@vs^C;r8yjdE*$s~UZi3>q12_AX;KWX!S-`2zF&oa2P|82ZAGkopN z$?YHiTNLd@vk}nBGA;FM!+i_Zg5`Y+)h*F~VbDJ#u_9Hz9Glmp_iZWOG}3-}D7-H^ zNk_(!`Fs2Wknl=s5=tI4nBokcZfqy(Mag9U8vZM1c{p}C_Me{j|8)*hmjWv!(Nt(? z(myQye;Snk>G~_Am+o9I_+R`h^YB046=w5$2RvIb-T33w`oCVp1Og6BG^~HAfB*BP zd1=2_Yae%Afq!9c{%6yvA4z@=;FAoH{C|4bznB5oi8QGwwWI%^M(sP!zlZ!k2lAi) z^6w%4x0U}F3;t(J_up3j&-Bv&dC$KW_J1$z|JBR*C%*rCFaPKFvEO%6zro}8SgHAV zb@gRW)b)+8$XRGcv%8+}%?U@-@ib@=MlcXB8=^U|5sk@2f3a^}@UG!hU3l8g$?G~1 zEgppobb|ImT}$@=qwKAN;_QPh;RKi9mf#XRSaA2?6Wk@ZGq`)uU?Es=cLsvHLkR9n zaM!`z+2_4?<=g%Ct=d~P#UE2O56{T2yU*#Scu9e&frZcVrWfq3Yv6{>^AN^twEU?RZ@41 zEah7`xx`ZpxDm61fQa?CSwIbWVb=2x0rOvaoF%?mTC(u1H6jEHz2|iZDaM%~6V{|l znPE){uLQ4)52?aYey8?Ql}r7=Xs~Km!Qff@dj0+XGr0dxJRtV?p546PA@%wy=ojkp zAwH$A&vDSBYE@6>wR8quPy~)F;<_H1z#HPO|2m2QFif8UTx(oLib|A6NJb0S-%m8N z*7;@9?*rysyB&tE00xT~>N63H@7NQJ(Fgs)lTG;8{Qh(D`@`+u#c;sf?`N4PH`@P~{}JRV z|K%h|vSr5yi`2b7#}Tl2JQJ#Piv92SOe_b%!oS^!?!zkH-T{uz$@UQ9_Mo$GtTX~E zy4(Bjx`n(-xlSs8_-&gtFm3nUzwIg%X3IHF=Yoj0j_45m2 z$9v1;P$X7y?OuCN<$;n(+8(dG*p+`k7-Kq5m!nA^blN59P0i zKVH`S#+ge^I0izvw7Bn1H`;%Q+P3VKy|qoqK`r>4DFGb!jGRR)FexI~o%-K7njlY! zK)?*ZIY(RRpEnPX1(=V{>c0bojB3mWKiH?u5xMStl$lrQAl}4-utrA@h7}ki4}~^tWUG z$`%D#y$>8SYRabk{tZ89Y_;dk1$Or3McY7Kae!+{OPTI}FcxoKHx*w}Tvd|_I7{eq zRWKKRBjSqUdxIC$EAx41{-RUvx+0YpMwZ&6ZnK1MO3=DamZw>d+;h!pOwDnb|8kWA>Q+qaGa6T@YbuFb+Yxbk9xu$>gbzA6lu+My@U*{g7<)Cl+QsuMd zl0KXe9dAm2@A5+;L96+}0=$aB(=Npb+ZEH&yID80ZCt`2^}z)BA^bsH&3xawuwY&a z$EJvXMSgjqf$g=ttOo??>ZK(V0>^QgeEH>R%rb<7sZX&v0+Qz)D2+2Np%Xp8Ebo8cq;i``FVx-BBCxd?%A!^(4;vcpbhqnyNIE#7bDt=oFhN zL=zsDF{8wwc?*)uW{uHvxuy`Jwj4*?y^d!CW4e9#*pjqq+v5#BuP0AVV$;*-&qcHh zZLo*FzVp8THJ4lYvk(t?rWe26t?H~C-hudB9!CKt2K!S**IG&1EJioIJa?YgDXLwz?j0YMxYfbIulSJQeyK3FHmyx)alR{+`c2hc_?u&n&y1&xoac zEp(#m`ZiC&tL9!(RV|Y99dF93xWUnF3W^*5^B!|Ar@o;Q1#=tANNDE`i6~`b@n>rz z{K9bh#x#lZGwp-;I zt1uhMrJ2h9mtY$>DdA-#f6C%#cL%IU!9U&WO+#D5Juo66$r>7fHKyz|+zWwbg*hvaTH?B}E;iXi6-h>w3<8IKFO7v*jNfh!H(?nJ&nnX|FN$mW+JpRXKR zJC@ns2f4gIAf*^4J%0ukHFAmwUsM2#K9CyY7~B)agd14FlkYv#c`QqvJ)@e7U$5Mx zyZ^cBd?7?zp>|R zvOz~sr0bAM`B!;6fka3!FKL~Fy-@<`yaU$l?U6c&H$QPYCh~UX4tzj*;COqhQ)3wr zt%o~=l59}&fN|sQ=u7-CDb!r1KdM@!$clKrgI{1J=zUn%MTHJ>x1KKep&65%ASC~H z7a!X{`uI| zeirTTG_a#7O;pLr^*#LcwS9F~slxU!5H4R>Y6MUBqOn^W6Nh7F7A~VVhW5+^xMCewLZKogd^95a>L=ey=)-U}W`BOeDYv=G70! zHldaKp^fC`Bbk|Bx|HTsiwF)d>|ipukAu3|+@GtdFdCxN#eUk#63~#}hc@ddWBzbJ zrp2W1p_M={7@G4){2KwwlRLEl-z+YFLK`DOTRAlA;Ut4@l%3897^mPu2ey_%SGSfs zHo0bqV8TT!6biI+88l$EI>rTomLk@)7^zHQ6#tb_|96Oj_mi|jH3g%^tTUMAt!qc?$U}e7-OA=Y zw%xolaOK6#%r-P}+A=#&TiX`F6~}!+%=l9Lv&<p3zXLYyF^o&?U+0FvOh(Hk%ts zIFY)&{&wtVtKwY6YTk4YSA~(nF^k#=yg&As+K)rdsC4}Mfc9f{iAil9RX3EXF@E{z&YLZYklGKW< z6)P>L|L4jvCR5>nr_h} zcPGcjX3&zTn1#RjZMRxnDC2cFFJC+DCD29Br7&*zzc$(b!JCG7>>xHPDSfn+iU$mC zNU6aoQg~I+5I5y_w&F>g<>pmB+E=PxN-r^b(x*tqx5;nu(%!%=#Mt+b1WLN7- z(%&27S_>sNM8inCOD3BNT*^r5bM&em@9(9zebG5emS3Pe8nq2$n-$1ppf_x zSdN8>+DI}+u&oH~*!5X3sI-_x-WnvvnorLurXSa@P@AASdfz8^{geI3W&U=P)sHkj zYi&2#@$i)pDc%sv-OnPMzgrghJ6W~4>Z^M#cUKF|3G8Zg!VaF4L^K+ND)-?F|<+;p< zq(2`s#VV}hIxCG*G*2+6sb#<>Tn{`!ioGnqZYBnU&s8W~#c0wxb!CoOom|w?8ekl{ zju7E1ZE3H^Ni_1&+_h2o$bslD2;kZswiJz3hh#w)%amA%;W-2If`vBVl<2Ef-O?H) zZrA4q^hp+V@I9^~ZJjih>OVXmGydoAU{~H3udV*=OU$9L`(!(gIEYp$Lss>@0<~7v zOTMb?KtUoXle#pVW={0LG|4vEEqE!$_DW)YTA>qZTYoy99Jq(Zwb9do0onXdoosk? zij%cz+_vw*byhZn12ALUEAh|wTrYlK2&zan^Uv|y?gD$I{}rS=)}r#Ejz3FsZ>a*w8D}3YH1_e|8R1xm$*L@mX6m)pHiitB!&Q%P4EJ0+2+e#fCD*|Uf{Sw<~NBamDQT|M+;PkYY5TI9l2 zg-xN)c6|8I0hiyw>W16G_(tf-Ws6{OLk|4eUL@x^Y#nbQ_rvhfI}L#eeF5%bHF1!+ zm~ZV(m+{^9gcDaAw8w68*K-RxAZi8f4c|A5SAb- z2SO5V5>G5iOo_=avk(VDuQ$-j7nP!MgR;yPBzIe!hRv&T=UABNxoLZ+JJT$KYN|j# zmY(}y-+s1j(BFwH7B}G$@091Z4@ch7DjOvvs{ysL{O) zp0I2Em*mOKm4GA0heXjva#4?GUcVRlJW}fTPj7E}{`OhZ_QA@Qh;d{^4_ztHir0>A zM4x9$ZO1cODFe!Vw2Ky}U=Jv&uNUFp#FqLj(Qaq)h&R8dc056ll(KMp8Sop|`)P@k z#UY!L!TCTI8?0V@g^Wy2QtnGyU(4$xdG{4M3A4f9a9C? zCsSZ-YIc(x;Gs^Yd$Ij$;9UylZTfbsG44;QRW&cBdGgI97SV$;HTEa426f!9nobu_erOhw^IZK&XHFkzwlbm1QyQ~Pb4O6rO)3wicI;Xs}7QW*DU(ClF4Z(c8UJ4@D}Cyxx1 z|HIRd#|l2`U*LidGT*K-08fd!!!mCTYG{Y z9WOxkx|&y{4aKAH`e)$saeHT($Ktf^FCd^B9X$V(%ii^xX^vMxoy~+}42E=OKQ~<# zD?S}Ew=Yg7bqmsy zT0J3VSfa2B9ZO`N# zO{fu9-ynKQVfSfLb*21dMx$Kj!$~3{iGNaN_&Dr*LC7m*Df#yke5L}la(JCH#q1^? z-MhU!pug!POf{%liTwNGbnzSW>6}JX_0zCb`d0U&8oNX`NCJi0u#j z2tRx`O43tSEyjGh@%V}qgu0?gBj|p>6h*`~dEyd9G@6vb?60A6K8#(K_M=?_S# z(5bbucy8c$=wZjTZ%79J<}6ULR!VB(r23~Ky4$O<|F@`jmKCVsX%S-OG|2$#FoB(h zfk;p=zhh#-aq;Qjp+(dQT#*&`lg4Q@8&T3p+$$Ib^E#&0e+ea|&>sF&d;x$Vl&K#6 zW*Z$un&riI*3Ks;-J_)HMELvwtxaHPf1Zh)%eE&BnUK{U=(SNmZXn@pka z{KokLdVSk*&k7+*|4O!y%mj`FMub2X#6Vh}a1@*g88|XO6!p}$VAE#|cycpft2+DR zr~7uhI=A-?pNoO21XBc3o7;v-tw?Y2-k4qM+mVCkxxM}c)#KwlOrOY9a{|hG&-%(l zCY1?TOZSIYSkretaSNiVBLax3C)9{+eQGoh4NkBxT*JQ%J}2?uJ+}ALXS2qGdF^I1 z0hjo;g9cqYuNH#qlMp#IR4Vw&VS}=K!36~HUYk zLC(?$oVIHjwvZ5;)I>_=L|`Dz`L#N0SVXR{-ae+25C1xLn?ciUxNoy_dj}F_NBt*& zg+a^W`J*%k58U)ziA`Y7U2iw^ zVNcwN0)`qr35~6BQ^6-l_Qp-&Sr(tLt!g!wx{3tN5mUsaMnecOV($0q&&cE);xxBi`v?9ezkYlJMpq0c|?I-7Qb z40A@Q{zl{uii^IYgiGfLNOcF*J6QfcVb+tFtxntI;DD{Qq!%G5c@otiWqDw)cTT%R zXOUTVr)#vK6)#*`s}1LHOmBJ6Hr0e1+)^AkJy{Qkyc0CfVikGYeK<$Hj7a^>zx z-%#|u5z3vo|CbyuVzN_hfMDFQShg%=;L1ic?_~XYd+x^e;Ujn&K2eZEWTE;~Q>EAA zT+rHatJ31~QU&CpM}`-^g$H_qICp13dXr&p%kxJf32E6s%ajb|mD7n6ZfOrc(MeL! zZ^@e^)y~4D)g-uOo+TIy&`${a@Rt~7Lh&I@W2MxkkrkGnN$={Vwy2J?G_uQ#Lkf3O zc3GOB5KsE9PT79*ne(IhEcxBVw+sFf-t)8KZM^UL5A3#H5I`JlI0qIN%?;luuP_&pe0u){go0P;uqIC6tJ>3~Tc@6Vd?LHVX3`eZA0M+$OL-<4~U+`+>_7HV+w3vF@0dD%you%5>3EwEzpc~>5W2pSNWjA z=hCra+LjC<{#rSKSw=4as0X3Izn;`9|2^(~dvSu^HA0B5-ZXhbIpjv*zpfLg!1W zKN8VpYe3&^v&S2|`#p+Hwl6zs6EiMiL^}+9&JiT_@vt*7*y89gOOm&I=-2f-#)8c< zSpkKk(2?NR@8CY<+RF}sI$sqdUd>6>I=;4fSj;{!n?%|76M|5K&2W7%q0R5o+$!dv zY3=GXN-LM9Ett0b$GcUJyW{Seepv;>X`#g5{_+l8CX;iQRUV1fq5CSzMiXbX7_Wk88sw+IPDXNY|l((mXk*`-lAyjW$E@>zIuRaC3$ zi8ep+FAj|QG^48>^1X+jAxZu66_NPzM zojYy_xhPbE#(R^YJUAUK$MffV8`967uwVQY;L1Oq|{L{9?25&f^~0&muXs zE@{qOh{4Jz!J{0(Ygo{a89Qrq-(Xm=me`NcWHW|FRhkIchZ0f|4~%e&J6f?0Um|Ya z(nuh~KoBE$yr1dLYnq{TT`2nbj=mw&+d)Qa+8(m4&KsUI8>**B*V%p@s+?Wbe@aPT zomB*WC0Vc~+6(j6?|x@pf+HJe--&#Hh#5fmL5JC>@)i`87Jg~9?F7VkR8LL5*hrlG z&7VEh*b2z

clYDm#_*A@Ka!dQFfo-}sD=1PR+j?;F@ZG!x8 zEQ|tbTdc|E<7@XiH{&iQ1kO<@7-vid*4o6X|HPOf8Ygy1h{0hp$)j`Go2#wWpT2f< z2|l<_eatL>e1E{JCV#aqUHD^u`-(~0tk?evljDS5i&!1{?RB(}Ax}!1=62_*z-v3) zb0)xZgu;(YBkWC#m%lN+PR$b4O(JlPZwwQa(i^OhKk_yCJx(G3;^Rg=;fmnK%l#>v zGrc0k)B-@8yS3DOQKQEH+d!*FeW@`VgNz|gM0eiahm#}Vd3UN_k4-po^{GFxR^Mh@ z@qYUAqW(tn5SCWrWPW(65*#@av!`S?WYt?nvr?a>{~KZad~LD!q3sQ@o7<-M_rcdn z(b^zc3cFCAF);mPL=7_ z3+7L>L;06(u$aXAz}RNvO^cQ^iz6=g(KnkBYVuA?w-j(t-4C@A@*sL`}oBSRn54;I@8|Th=z_Q z!6lISVK@CTkv-+Kmu+3z>%TdSqE!N&&8o5+bI*V;B%h~+$2O$=WF~9s9~h;bbTfdbNZK5KZAy5m=6^!L2jxg;CANy-gl0x97HgRb3rE%<*{!8Lzq* z7v+VbtY-IgSahfG8bq50FqAiJ`HTqF81tyC*ej#35$qT1jYQN-RjBX-Wf4qpZ0#Pc z8$p?9XIxS({fP=m{4$Q?_sRw*UxId3(k-ssd`7N=<|Zq^W9Tzj$KNMOKb=|}&vxj~ zR~)f#@Y2NAu-3-u4f9Ae0WtCy+tJp2i?j2SmVU4;QXXT0=qE1qys3v{#@vBX3Q`w+ zil#UdAD^|Kl66?|L?K2`0=zExR`ZSMfUkpkh@J-V(h$Lrr#=Y-$-_V^?v!{>Hw0*A z9JbEDxuA^VU)eS^DY8*7Tt)Xni#An zL)*szjnj$6B0pV+KenzCx(Qj*EBdKZ5bl`;`q`k2y5msdV@Df=_luiS#80akVRISv zCbBVo&cmf%@9gT+ojN}>twiHVJTn-o817@`=K1v(Y7YC?D&s-^0tZYx*@0&l>Ld2vN_i%MT={7+9luCsiMQl2klX z|0-M%uW5%WqL)9EYfDztt3ZrRpgMrWO%S@KQ2Ehb+XeS#o*1;~F&0?)(cAd?#W-OUtrCNc1 zJI{`OJz{Qe6; zTk6K3J*+g= ztUy2^L7>o8ZG)x!@2KO)Sf`&n$8r{_8f%CgG6QpLUVKft(>`4#arLzkqSaZ9yotF^ z=UokciMPW{AAP`6q_ll}IsLZSDtw{!EZVV{T&@N4@R?s^)6v8+1WqO+!9BW2r1lHU zxhCu&jiL;(Fh?zM;R(5Y76bdvcnaI_Onv?u*@0v>{v!PTvmlZY5jS$m?+0%fp&Lw! zyF#R=-V7?=mZkq`SOtS@T^}VWH<{P4xg8aw^I!Ejf#A@r;5tu!{iXmdE zgRF^S9MO(b$Yuq)!}tf{f8c;pLB_%!!2lF)sxf~s9L0xQW+FH_`-^ytNA?gsd(11# zrUOFVsK6=t`DRQ_RxZj1t&+p#SZ_Eu2T4LBWI5h77`Ipaa_2|{KD@CxN(<&1)3gJz z{tgexFCEU1E_Z7YSo$h++;z5G5-)BhfR9)3+RG^0!&7FjY=!d>jwRse7h0As69?VW??>rh!Mp+6#{5&K z646L1jQaV9<4l`?^HJsjYCH(&u-a0+s}$7_zYcuWw%CPL??c^ZV6eb-IfQNpp`gt9 z025@(AYI?v!*2@TG>qkWb9f)7KpN#XDonnK{VIjY5If)DHBO$FWrRXE3DbJt9K=~1 zp zNtT%>Cu!j>?^nP$jw!!eCD@(z$C(cV9>fAh)XJp9gUV~wwAE(rUF2gw2R9=1Y7GH& zR%ROMOq((Sm{Hy450%cQL{k2m*15ZjanVBEEBfumRsRG5HyeejECYk#3Yx#72k}&r-Z~Ag@mf$W z7>|VzupDQhewATTm&BlNfm?w#-xdDOl({|&Y@rI zSk{fY9jR%us+~`9i`caTBn6OK+$BDM(>nfMBmQrk7HVW*{RJBoM^TW7wH=3xR{c|Uxa+=!$I=h(-L_N-1G95}VggkS=54LOoznPn>*E=hCN)won$ zkv5KTyatunb04$(CHRYlHjXnkAZ=fEX_O8FCn;dA)v~n6IG9Uyhi6YkWFNhDOoT}) zxs(jO7a(6rsoxgpNewCEfhrX(beh8X=%c5<3)7a*r#SKWYV)78DMK4ulaoXR*eB!P z$s!ElnUe`ReQIU_P4-mmRL#Y=D&RXG^)Tt8X`J#%6%wrkoJN!@7ULA)NZ)rV)BIE0 z#I#(#h(&3Ux12a^AE?q4Fm+uyJsf8zv9Famzh*8Zi};R)L(Lt*@cn$IK>n-jwfVMc zh%J=dF&)P5`U6CBZPo95Zg<2~sR0~#fS*Tb2UW|V)~)FxRiFLo0P`E#f)-P~dXrue z@4sp1moyfJHU9&W4hL2zBGeD9m8uOWO+M-yVY7+9M_(sL@LlN<@;Icn=G$5O+v|p5 zU3?ae#sycJ4_H`1zh7f=6b0zsQ=iXLJi_|p%kMDf^de#KeEYR`UI8%v?GbHo53 zqZa^|MvXo#lT_8iOi5`~SoOCY$5H5_f4-$2UvJm-5#EI`j-TPY3aO!sR&W$ZB6t4M zNEmz5dPUMWEK}Ptkq($zjgs)H-6)5x#FksrtUx{iVi4R^8ZyqME9@psr#2kN=Ob>j zALQZ8oT`Ii%p2!FUH3%|Tlc(NK%Dk9hAIDe^&RxvKGpI#?4e*B#N&1+oRy#90!0Bcvc7y!*!Dj3Iu3##ST+m~% z)T-)2mF_cjcAfJxWZx|xo9Gs|4N>*|(jPP#F6A&xY@|meW9Hvz$^m`Z+VA*Q{YoTRqWWZrMag#?miDmF_ znlh^NJeeXF+mHMcWfmJx5{aH~zOkBQ;nov;&=!l!=5talgLh4{%x>SRt%vcvvoxvo@R2% z@zCRch=M?zu31#y$?cNld+}%pt+wdkg1X>?`S^&m`!zp#e(`wCnd!yQ7?839*T@78 zs`H=T`7j#=KwL)a>AuI_Dvd8&4GJ7g;0&%SE>dYn){O# zuXD!)g;x{Gnfw;i;xv8?P+Z~X=`5ZDOGZ^&KS|r-IBW`P47ID^iOgQC*gw^5-9a^m zbU&pRjVr_ldo4Gh4EqwVaYMS)H%7(5wtVd~cL>2*)eN$07b0 zQfIzDT}0@*MNLp446))nuxIOC7(6Veamta;j!>`J9wtnI8lJj}{Ypm1u`KQ|y{Zs1 z_!7L13QZ6{QNrUJ$!g)<~MkZdw@Vg!E=8ci{WW_G12@%}HCN9NqzC)y|zURtPGG;9FK@V_8Z!*CW{&yT3JRbV^1Amg0eC zA&hfwdFD$@o|2i=XB{R4Z`0kw6;8t}S+m+f#_iAt+>|xa{&v(Gr;OUYLzzcN$o!{Q zwnWMTM_3ntn`gH%?y;(;NR}2O$vR)egUCj38D3U=1n8g{y!8I+&D8*~8(Zqef{bP) z{L8x@bDy`n+#}&p83|wW{77J2$7)SRg`gzWRnGOa{ciy>99T*Vm(#4mbpc|p!};qd zsD~Qc`Bcd@2T)C1|IrXgxPj2Oo3ccF#9PWPmr-C|wyN2t9o$N6Q#ult2wFop?W^ZZ zJbmunpPJ$Z2d8%vZ?AsIL+{c%Rw_%eY!nYw8Fiehu2Pa`ndH$mP6-dV!B6d9cd@77 zX<|HL?HZBa=Y>V>P0f~1;PNZ@=;`#)tU^DBz)G+g4#t_^H@F#!DG1B)WIv-6swuT639u0mu#=GuMsf`%aeLf@!?C_CUXRk3xbDooy#cMYd( zVoeeCcp*55m?F1t+GRF?)X<5&Xka>rZ zRmw_m!e(#8QZbEv(n`VlyQ=NMqZ=9e%b#+LwX=*%>*+fS$%h4`9KJDbeUy>M=a+Q zG(IWNZp%MQwJclswgCwtTh2B#6EwPX{ys8oN z(O*bln`vJ+GjTAD$%lgV^i`5{iDDyHq(p*daJEwe6G)F!B39(}5sblgVL%hfTcD5z zqy8CW!kb9I3hcOv(@>Xi-sN(kJ7T{^p8Pl_YTNbXn1x#`0R(g1(gi{hgl?trto&_ z-JNfnZG`+>O#mH1bjmA91zfdPeDh1W>=*vj%hy^!wiewuqFr`I*3(Y7obNB+HM<|B z5srxS5m|*$d+N2gHrLrh~a^pfgzNx!!qfAPN%vF6T zJyb{`4A?kC_Sc$h0M3YC_x)d^0f0>Ye|rQ);Gu+{en^xLGwlQcMVR?dT=X2@*4~*M z4z+E|Z6Wicz~1|RN@<(m#aK2Rqy`yn!G^V42GGC4s z1=QCgHdxjs5Q^!V{&q)BXlWy|0xQ(J&!rZo$tCGZxH6!wHX6ls-arpU0MY;S=!PQbx=EydU@J&OMB)GW2lE27$ND4b z&7=_4B(^{}SPcq#ZsXtXE3I|DytS5T>=WxV+euwl4**|$FKakyJ?9A(`&j?FH_cSO&3}3UP*=EWX_>s5zdX%vqaEP{u#<92gonYKk5Xq+6SHP{ifr;belE>( zF2-f*jbWSb97jaqiC2A#!`Pkfl-n<|_Ql6&9+%17noh1*bz;<<%3!c3IMqk zIz^=|Iq0iGBHneZuxz~lC3Ov6NBE}bD znXAr4`fx?(DVZkf`Ycf})9J#wtmCHLe1EBtX^-L)p2ORnj`K^5k|*yoN#kWn3*|6H z6GvAP@}yEBX2>fh2i^sqT!?T-Zt zwuH@Le8t87*GXAnd`)%E1`>(wtr5%wmO0S!q!dxA!j;IQ@r`ZR)BUOK)Ztb2x0UEyQkd9M zQ8-UfdRFOkj$K4Tg##Ek$9M9f?QXmN+k5<4TJU3*j$)iayRVm?*R19Y3vI$SON1** zX=$iDd!HZ%aU8mbh`HJTi%%AMvFh-bDciS2N{dZ3n73_loeT?l2=sLi86H6%%GO6t zMzw;>_sbJC5{t+Amz)bJj#y*yo{dufAlJy&!UKo=b-v1-OzKU0dB@e}X5UV)@Gh3y zq}^=nHlK|rgnj#U(xoC8$~KRC@1)ZV#Dq#QWb^0G%WKb1TsF#jl!P8=O+>xhzqHr? z@U=;Aq=nNQuA9=EXSX14@SC_{yU7CsdIhhdIGz^oe5+hdwz${;5KH{Vh>@PHDXK1G zVGC@y-RdlxPJi-4B-=ij;yP=t{K3VuTvMR9&Nr_>QKw#O$IpX(Z>CI#v)~6u4D)Td zu6FW+ytRPe{fB+gFA*!BV@c-=4TZ1lnB7vAQrW7U38Up8=;Fjux)F}(E76PZ5X>aQ zFQE^oH}v)kbpZ9u(Yg8riQ3LZSve572tHhFG_si~VUk38(W;V55KUh1l~E{DFWq`L zncKTD2D|~&!w*9~`fRpvBlQ7j-XlnyMsRLh(`4`FAEzyI;)$0$D) zSET*_mNx&tSu`j_@;@qAs}R#pAXD!B-JkKO0M%ZaU4<7EIs2yq6BzP^m-+d6>Ac+} zA^dxn;*|BtRrqe@9m^j8VJ+089MA&6xJv%ADtF7iM+BpH-Y;yoz@6!Q!yi=P9uOm; zG?YsaNP2T{9i~U0QqMAnZv;ZL*JQQD7Ne?gP@MbN!{}LyhM%QoQVuY)PG!jhLIqV4 zUDdx2){KoN3wlG@3$?=O?)es>kw<_kpwl*m!nHtR=9MvBh~hIGS;`)zbdD)MxYEv$ zm!pyl+fn?_HjBi%HR)+v(!RaO!b0s+fOSM0gec_#0#^wrVn4aQEomhQq2ipaPX-1E zj@D9W76Zua_Ehhl;zZqC)R^PZhb{>}nMEA6*87dtr|oOIkD{0$K!uk8w@AY(O+(-t zsxWErmoLcJZ@47qZm2YEFZQh1WA8n%`* zn0b6^baT9{)o-y<){dyQ@$XIAi5aG7I_~_-akGD;^b2%eePhPp&3M-JAkPi8kPZJ< zR|2ZGJz@KJ-aGHDk`vtMkr>EA_vCDyS+rQH&pCb{edZ>r|H}*_?{%NKG9u?`u9|j_ z`|Zxfv$kVg`$VIt@;PkhFLe^l)NL58i081=5tplR%d{{-Mpm+`FYOeqqLF(W4AYC%}M^({7&d@hJ3Muf3Pf2GyMG` zsm>-Phdsh=a8w`h5D z=XihHfxwVII;8c~vG|;1KG2e*S0H z`Q9Wta4dZdW25H*-7;yeUVi8dJwU#3pX95%ha{r>&LG2bHZz|3vZR66pNaVMNf|X- zmmS3#Z;cUT{Gy7NrxT6&2EXt`7B$`9aTG`L%ht)JP1e(dUiHR!!puv-jpV}jLmAB% zi2I5uzV9x4W?D`hqmh9w`ZtY;DeT@bj6@ra@}{!0ejDRhm#1-$W6lA7pT;qMpA%ss zkqhr`{+q+Y<3k@?_v^-56w1I?aCq@zI6;=Pj`$p_1%bRt12;1{P$6=UJqMmw$JF^N z9~xGj$?*JUO;o$my=?_H3XB_MzP;26YeKB?w~`WnI3#XkygduTwQ1`(z3fe}P&+&E zCT)HxB}%3E{cPv$s<(7l-Ah+GeQP$Ob*v#jR5#A5_jdW`HKRV^m594*r-5CZ*fpk- zV#|GzPmEy__jGg&Z-Xn13}eP8mDgZzf;_{>k}aP-r+}^7@zwXEXs@PWg=W(}n}!yd zS;EeW%{3o7AMevKbX!;2;S_}nf$dVPd-M)v0Z&hj{3c!O`Cn2rl8uHA>#pvd=0`lP z3G@drJNhtS?PtX9{K;{nT993@lqcw;mCu-*g114zXI&7Y?|xD&nnYPKK9XZ0qG!p- zfpEo3PLzI}sQ2s^rk=&kG0*8C^Kv;ScSi%gy_Se|w(!B$Z^MWg!NnHaiZ$dUKc>}q z`0(bgmAwwg%9LTi!`GU}NzvzDRpO7=t<{CuPs|UPD?WlmugYF~-9RVG(*k+zTMu$& zhVeV#n$`Ip3X=nQ11?r>B=AAd0I0v~AE(_GEmX^^(}ovvy?L!=?(35E0qE}droR&m z)c*GBsJ3)JYeMq-d0DvVRoL5;4v8x;X?-wGaKi-Dwa$K|IV*C-+5=77xKmgYXFycdE!c{YKB|EFQH*8WZ^lqo`Vqt_`1F`c~%=wIha>& zPcI|-)J|MKwPmzfs)d$q0Uxq?ZQ8z|(d{LFMSQB+=dJOi-M)I@mg95M)?6M{dh9$C zB-m^^vkpJz(%xz<*m{&TU~n>UfY`jJ-X}2Wq=7|V(WI|c3v(ywC*(Mc;b2};542oS2{RF9E55ZCVuUmpIyG8 z>RsjPH6l)G)t^@651sdE?~e&yaqvB$;e5Uk9X&@MH*PSp7r)pksot5=yw)xNT5d-62uGT#2h8&ciEf3x%(Ds5{Luy{Y?ysYJv z4G$8Uvuif@5-ZGJr%f?jH}op&jKcKv*;|(xr3xqAx48P7f0K?$W^qw#f6ng!3~Nnj zCj>Eys^gyWTn|u;x-TBtbv&LkPnW96JSou((92!1alQ4*pL2S6-u9S%6N(|>{keP8!Y1pVGa?T%(2s;P* zQCjU*w`oslzwI^6D8XqF$&2NpA$Q{SVtb>G^BH1(KluE{iHAYHN;c}c2bVSK(|P-@ zG(GkTw)$sVw4qZVnBV}xq<-EUx_sna8ZkCJ&Utj|0fE&3KIS`5!I8{4_c=u53sDQ8|i4DG(N`|RD$tDBFn z{kZvg)BePSw<#;WVYLUzvTBJ%plq`F=q*VX1Yg*4mnA>D#840dWv~@VGZHRlUsKw> zjIkewjH(!+`ZOlW0C77_bE#UYdrStSZNgs5yZ%>jn6D~W75Q3pan|(57#MAW|G7?K z`IeXs+?5`{T@oDoS@?Rjx{Bdy+jIQdo4fDYzwp1Y_nlEqZtJ>&0tzC!1W^&NAtKVG zDP3KHfb=FEM0yFmBqV?=!3L;wA_VCz1SttE0YO2fNev`{fJi7I2@sOdxpD7v_TA^~ zb;mh-+`sD&VAgE-Ic zdDkblfL)_L7IA+XnBDU8E!|&TXE^bw7-;!O$oG$WS=ElGRO5LBLO*&L@j_SbPQ{`k zrlPRt{NibkOUU*64&ZEC;9?2KGLLxhA5j}K-43?^n8mZhO>|BpQgmn1zH&TwE)^mk zIkf6Q?^BhU^ZpOFl8erJR|{TSz48U;(e~TQn1_uhm)-%6L&tUg>z^#aLoW_%RCbmB z&83t-&{}EFlq+Wy*T=H?e%Q2=@a7zrnB6r0u8-?cnd%`4G;rmq4O`<3!R=gvpN+Hu z?f0$Qp#fJb64fS;n=GwR&3E8}-!ed=e09 zrjr~%GI0f%Le?jQNqc&EinLcdBStDx4@(aZ)Cc=MJYBD(aH#dGW4D06-qI~|&HE%H zUM)WW;CMIsi8Z?V9b_*|<@?;e1MTLLsE0y}V|gM!`kIL*y^QBxagk0|Rq%*?J5=DD zCE_*jgXEGx{2&=y3-rk8=U!XTMZbOW;#eIba74i#{kYubD%!wCzq`oODSDXEb=i4N zpV#;%mieeG_LwLw91poxC`s+w>_O`Kmyzedue?iE$-{jJi?A?8@g6rn?jxIBy}SFZUguoH%jo{Kd;jn#V`K$A(jv&YUwKCyM(qh`%r)puocM2^@z=*Iz_^Q?0~NQgJ;LmfMxM|IW`QT4XIyHPs&Lgu za~5;8dTTI#3&w3?PfxCxo4bsxt2`rv@QP2Kd{4_*s(nq~ynHtLwQ9Sx%(RgzV4I=c z-6u&Nt&knw-y7y%>EwAw3uw0f(SA#m4MLjR*l5aQCP(F61;E7cF_gy4GLRt0h?>{OanS}Dwi z;Mr(me3S#QXO+zh3`Dm~=zK4D%XwUGb3j!0c9=o4LUq3Iq!&wHlTYe?U5+Jy1E7Wx zQ!HDx%Iy|!ZW-if%0I;$Z-9naUI=AiUnTI)y1pt&XuQJai`3%7V(9lA6Z3YwFCbYd z(+N~c4(c(Hl+C&~|7v)kZHz1b8f+|=b2)T}PPXVS%=L@)U2iDm zL0Qp=nk{%)tSb7-Y6D6uFpL|$W~S)}agVq^kpAMAqtrmoIf$SY(1pl zf=Lc(a(XBvUm8MjQaNB5muwJqa4L=Uf}98yb_VG=_q=@)B72ikQ%e-9L%*L|Z<{hz z*1=CZg!KBX%f~)xUDs*1t7OPyr-Iu=5VsCaDaCJ#gQkwgOrH^6scvh~?$h+-5!qo? z!e8!K2BTNjFAc0i~V zv~^^=$t{mm>GELoE_iY!&5aoz*Y53J5&jTP{vE%n@9Ah(!B=rEclqGqwh3a}HlzxF zB@|kvY=xXnoxNlNJr6%%EGIZ+c?Ywms85)bJ{y96Jw= zAWuFPU!7gF`WiuC74e(hDRu4x!P8(@mnu8N>CpN{(yIX>9 za|Fcf3`SBaeT(Y`&iXljo~Zj8Iue4P+CJx^o3huzKOC8>J^+X(=r+a5PbG8KxFgw~ zXfdHSphLfQx+&+zz$i(^q zuXgC#HPhSk{1lp&)cvTW^h{;wgxi=qc4`hdtIr!be3O^jxXMfWwz$R|r@R|eE}6GEZY@oc zV=?yKLWR{3ItO5B)CfJ53m!}|`Zga=+>y>rACTliZWh^qE06t_sRWh~KV)Osj!#BL zsBp)_B-Wr^omQt%ofZ zKcA4p@i-&V7T(00n+xd9+eR4atow1kuvc3Zf~Iqd)*3MtYVbMzX2%McB6OAfMeTiD~#V zMg@=dF)c`0Wk{9FLG=iHKw<4Y^A$wD>rxV6<%1E3cD=`8R=lD5q~@ph?fo^0T0;8n zHUe!xuXbz-JBRXi;Ifh)w~O<0^B3*oQS^Q2&<=$l!eBO&x97!22ZTaT8!bgeu@Or@ zx$vH1O#j1KZ#yi`Nnaa@=Gtz?!19uX)cV8iADt+SRrcTa)}O6}R5$?MUW!w9uMMP(~liP(leU|al46>p~qsqpJV(&9mq6+&WqdF)O=wl!JwYtOv zMlnJtTbsNij(>3L8`OijcPhtJFF9~L|jY0Yox+lR`jO_*H>3nPw03!7&uoz8gBC4CO+&t`f2WO$q4#ShDh}9X z?BnvNVo_`E!p{m~KLPJACC%D{QmycnPbP6&YUMn`W3ur9$M#)7W|lP5y73s^$n5c< z1-Ukb(xXi{Zm6-VF;bJ`DPrq-u&sM}R5@0J@jytp>YcJSqv&S94x$=WQ14dztswmo=U zXJ@P`SC(|Sm;wOo6w1z%MJq(xMQcl@gAS^n=sc%Yq2Q?UtPKEAOKWXd& z0(fLZ<^9xwd6WR;caUpk(e+W1%p;hM&TJN-Eke_(=P}N^G<^PzPqm{ZV&nr^x&`J< z{4U*T)wYg!SRH~4tA`Tci=Eq;CIUPU8-vu6(LvAzCB@^v{5Kp67Q z)nosvdTci2y^b$eUc*>)r#nEJC?2`Yg_V8-MBI7|qu9a6^*I1=>wLBB)hMP_@DjM9 z$=Tjo@};4|aWCsQasc?d24ja@&B~qY>^Lqw(b4eo}k(EIRe=u>Gz&$N$wb#elCbDl(()+6rY zB6F=mSVJQT!cCuFLUO>LrzM#}@&3gTsvJ+lxEWtg374JXS&S#HjVV;*nh;ty`QyrB zpG5iUE8_E>X4WI@a!(JH4LU;ymfP%xh8%@Uc<^=y&Zt_s^rp0BYuy!tZ#zwmd>rEf zRr{&+X6@zlj68Ke&nd*)tCUNb^j}*R^O1h%=jb6h4VSgydwS=;LOy@5bEd`P#uoZ6 zx0VymA&zc)F5>o39G*^8C%Dq|GYnJNu?~sjJh3vC26+dT>7S9z19;?l14DfF!AJVN z4L->oK5weWiD+HTYg)&(IJ}PXWuCU<*B|eJa*nGwdCqk_nF9+1xyCx!)62Wkg6yo@ zv8c*i@(;cN&n-%We?uX5G!nRQ56jtb8LbQXxH(_${@bgUIu-3!DXpH>;I+aL4Y+0? z-bxJ$x6Wy6c=>Th@V@?9wllZ$M!qJfRQAMqz#zxF4(sDl?g47gZ1FIar<)bh9= zUm-Sm6Z$QEePqawt77)ZJ{uMweLvZ+7&0*N@thmzxNqO@?e-SC5xNi`*~T^N;gGoi z53OnIe`hp-{$xvO*} zya%QA#VWhqRWi?J;_w>+Uj%gl=<#ZYaAiMwWmUK@KBdjnVuFPSOV{a?Hu>goU%K>K zqK5X_1p3B#FD)Hzl_00VdgbvwdD7l}syB2<&`Wjmr$^UfNlwwJJhs~Ugw@sjsoGlQ zk2y(JI0gVCDy^&My0_U1+xrWy_h=pMLicKf3{0&(bxHo)qp9^eN(+5%S2z4mh#VK> zUf#CJ-Z~GO6Wtt~4@i^$(q|HXK3eXh5O=0D_glTMYShaU;skc5vY@u|KRhYr{@>Woqs~;HgrNcXF{0zxWNIEiMc{x_J2KiLCGnrbmTT z>6g^7f}XqW_}@_mw&(vOiT~#{ESa1?ien+j)?13XJeTH4HRdJ^3?LgPQ}8x^v~gpDpfBzLO6+?7;W}A~lz4`5>pJMk?_t(h(~5 zaHaiE99`!1N8-Fo-gP|CwAxlN8_T$QnFAp6zE|VFBh-Jk@L%8A{r)#^D3Z7Loj_kJ z>)<2`E+8T7``Jq-)w7Q&8c$yA`S{>?mO)$qrBpuEBtTWiTt7Az^&8hOv;Jp!{Xaim zITa@!lkGxfGYs^f>drR16&C2`P&N@M?;Dzn@0I7*^fS7S*VU9ubaGU)+72&4@2WOk zl+iajbU4!YPKUt%PN?iog49?b7(|(+^@5!DytKAGS~p3h#NL6nJ#WZ9T@IDSE1%>z5a?`|L_t2un#sv^8W>7Tspq7US~#{jL^$j^kjp6%`102l5cqk zB&-Sjqs;mzKOTp?u!41t$&Mh3c~ zqJ4MxWM5}xa{QTD_erubFBkQ#mvVjGu;^1e=9Ab_8^>~`AM9yDuqaZKFo77-Klt%_ z^OufdSO3MHT>*_p$v3xtr+1EjD1J1cL4QC>%7opZ;EbHM;Lqg5W7f*g?0&WhFz<N}I%LS?rag7dNwv#lO0`6rD^lj)Z|`xSmRVD++s zW@8jAjFgLQ-**S!>hfy3?%xD?_%o_$AuJEc{qlMQF8Xu=J2APR@03Cxxr<{mC%%85 zUNR7N1~#B|vrX-?7-P1>4v?=+#cjKpZ^WY?@1gfyNxN^vH(-SMgDk?M=5X6yFyOiW z5B92+*~cX(UWxBl-r5`QG#VJSmX#KW5tN?BXWCU9)A&$D)GyC19G9uyEVV|=kC2?# z)^9y{g;kyBZyC{89Z@Xi(#Gu1D5Fv{uXJU;ICS%h^{4c#CJDU8Z+stP^$lDIjjN>Z zmr@iFXqArT(FozXc`z+48xrDV%Ds^ zKe2l>`>qc)ByP8`PC>KYd&2C_j#*CCK_r{f1JCkLI|~2 z{LzQ%Ya`|Lv~DbS{zI-KQBAkTqHt%gR}j4EE~nW=*)-SE>P-{PPl#OP_Cy2>qAL)g zJN&wiI~$H<&vTm>c7}{D*WxiM8!^!nG+_Spgrs<2Tf>OTUm#*oo1c|G^V_6UjBn4v5YlT+=8Iux_5_I6RskP z8)>t!|F1*>0M)=CRLKv(Yj8!x$6Ld9vWo z>m*+s7wJ>DnO696N=f%O-)v$GAUbE`J$vZgOF86yh1X#>(&JQOSJk(_rLcSAWJ7+s z?%EYuVs5{UMU@b|w{3~wAOInlo^(=DSxOmL|G3@W_Z;Xqn_y}F z#+ed)FI}$nCke9jwHxYPnp=`?Fr-nJE7wzFxrI#;ibH?~q|L(cTAob8e)r_i3h*)_5t;6~^>q`}DAg6wL`Wl_TP*@v--_K_qIq;hR}4@!&ec8$S6UJM}VsW0Mvx9ozDB`DxFZ%aGbNM+mR{-39 zDnKky!6Gnoh}uH+GFH3Ld9)W>UN{jt%WxBUd~HJ4?$Wdj4MNBuC*I4;q*u}HkmZ4d zi#?*oC^3vp@G(TZ5#p+W!9FWye;)YNOb5C& ze90*a_luwQV`hwBsn;e1;~hPAW5XN=d5;8W6S zNTL$5+cA95v~CaRD{;#EQ43}wEM>R1Gev9gag^NI@QjJbPDaj7zd)aKyTXUl#0)?B z)sjySij@ytyT$ZOHvwpZR%W_;*-EkqjCJz*$W^QB&75IOW!PWVnQAtAL-&Jw8yWQA z?7Dg*e*EIN=C`cclaJjrXudgg&pgkPEyAd8)f%Wt1;V+=6*JiuVB&;#xBptQT^OdG zuNO(RRW=Y2`=D$*@^?vz7~}Rvl~mg-rA_tkP8N+U+-|pTu(u4GwF>5Nm^^p763%B? zn{<3-OvASYo|lRDX`)wsHt3&;D$x^T^Wu*A5_pcA+#?tau0!yc_EO}wi_Yx9s1)sJ zlzvN8F|<BX0-Yw4?bq-Md4A;{|UGjOC0K)mOpYd!r=-*vwLJHBn!A8m@ee$6Gjy0W`}C zQOvnq*U#WMeC#2i+RN;tY@{ zIHz(v{ZXUzV0)Q}yoQHgZ)psye~CZ9*}zx-Sp7hEEDmd9pLItf>^$=MqcJc0uyu;z zp;jxT=mP`Fx%96}&eT>4LMF->{@5Ejd)n+PIO+M~7O7UcT0!!dhD?UDW*TPe$4L*Q ztu#$^okf()8g!2KhHNc*)_v*V%Q@q5HD)(r`KSeCBy`_gk} zBA2Fa2xzRAZ7x+RP=Dla2T#6odRoDqW-wGYfcA0%&MDJ3=L`D;E$UTY%GTyD4!JAT z-AS3dmi@riNt0`xXip1moAhx`U7><5O-7^xb{S`#2TR3~}iprR{IB+imuE#2b26FlYRoZPX`2 zhjWtY{cgZ&%!@^z+#f@Lw%veg3{zLNY_@g(hKPSnfP)GPnlaeM9Wl#~Ri^jT5#y~l z8h5KflQrzqa>|wPT5~BFaB!Q1r}eHQ{lMW@X}3-AlV|~r<+T7Wih5WMc87s=CSiEQ z6c89@!X|ZFytjbKS*OPe@K`L)P_i(|-S1dIv$W+?#;g0Z>*!h0YhNiFhFNpXGVZ5r zA74;yn|F?!XbnJF6k}Q#AMjW~DkyK?2{IQ(U#mp4RN29_L_xi-dzH3b{eaIx&6#`2 zkTa_!JV{7B$wYgb5Dmx-u?Ahhl*!<)h`<3J?}8G%n6+)-_{XPLt{76J5pAm(o9rd# z&iOEfYJ0-%oBaq2;_NI)39+}gQfT6kxr&pF*_^YfTe*{)>us|A4JTFl?nXdb1!Q|R zMf;W0tsRm*4HrW@9fqj#2R)t3Y}kM6IGdq|xYINo0HdMz-}~2Zd=cly41o76Y@(l!9tD|*gs4_qe zC`5mfI}f;Fq&Zg>)nmL5sp;0Fw7rx;KAUjaZRiuE;FQlu4hTf>w-L+UP~XWUW*=wa`$xtHqE9{&4~=B*L6GY~}bXIS>H z$|STF|BSh6W$pBlUABzL`(7ZzBA0CGFDCc(CMeFs&cH>9zBT5?&jF$IL#<#}epuw% znTS_ItTQFBe*ak@<4N+@jk~Sg70~q1LgE+aVKFoP z`XLirpL5XGsy#AL$SY&@#HHw{iF9o9z4iHXaHu^%JirOKX%#`W0_&>`T8^l_GG>+R zPX%=CZKSGqtxtkW7p8}ZGeLf~pa3V#c5Siipxoa0&7NMxY=6{LfznNl6k6k?b#d2% z59&>$?r#*RT1=TNd-GzNLMr;j`6fbOOQ%MQe3-LETdtZ|6x!&`{9I8ObzTT!oF)Hp zY*hRPZ34JwgUrK4#ceuFRwZf*n`lplR%6ag#&oQ*E0>#&U<7GgI5%O>Iz^+x1mjLN zfF*Acdc9M?biCV03G=j2ge;tNPE3`!0GI);hWU)5Lz8VB?HO;HC<-QFuai_Bmi`dY zyyqxjZO=+p*Iqa~r@L+7K57*sraHCUOlNvfltH@Qy#v}QA&i<6NZAtqpsq6>@^l;- z!5DKK?$vfKy|S4KSz&w~ENqk4#)){+cYFgPP6pfsQEv4qSx6?4ASw3R*E0L%#G^Of zpaVAguDg*Ho2Q178hujy^`9O_h=$eyo=cokHI$neb*!`w9;P+TJKvDQg+PynLNeS` z?WxMZYM(362MzNZmWxfbLpy^_2|e~-8_b7-wA8@6aNypoiH84_$maV&W;w9S0m=X9 z76@Vog6BQQHjJJ-IJ9(Ou-0l=Wv2ARe6_3He2Nw`t%aH8TYYOXyn?=Ry{>mzPv)nE zyN$>v5G^H?axDZOEy_io)ZIjC7;?`u(`hHV?&5|6q6gdzN@P70}ul z*Vye1>38s_Rl=KV8M4~UGSOuxyl%d)rQ?A2##_ghM&G(3-~5$#bNCjkgo{+ir&@cU zNXjW>$mYym9WEqOT$`d8knQU&&LAslI4{`gA_gs?P@HE%E=bq!O#wbUZ!3`S$BPP+ zKNvUzs#SNkb)BeH$Z`t(a10--{GOkEDXyZ^(sH8Hy9dF@wwo37YgDb0aJt{JVnT8Ni_&Lv~ZoyIQp%sSfM%F`FE*p!MgZEfK); zTzY^y&4v5e%`4##H)ai)+|Ok#L*sAB!hH@#8Q)qQAJ<7~+?fIJ2HJbZ%XGpXwGo4Q z8mbi(&XpOAeJ#y%uWp%Xqb|M+%)&@}Rs5>cQ;jun4^%1jgCD81Mn3jsx{Cz_0Re5R z1Vzf{ag@r#%d%3Zm+Egzj*CRoRnN4pEs-Z;ljM4<>UtkJ#cVzw7ZItrDL0HR-E>`k zsJb_(ckVHBJA!y+cu0UZ`WB?UJYbw%3OBLP+pmIDU+>neJX%=lk&~Znmhnc8`$$G~ zFr#+nj}Dsv0-E=NMT$m7_XW4h^E2wpY#dG4j#D zg}~wo_e==&p?9gR7fL^KyXa=lqlhtz_UucfKx%SoGMbc#9wVEUeDp%i3eAoQa%_ax zd{YYu=2dE0@3=5iWJnzKD(E^1qc&5TKe>FyY_BV_@FF)1SV!Bw*5Y74UmwAud%-EG8WV(sX#mxY1a+VuK3kJ+R#WqbdY$|!ba zv3s@HMDuyn)azkN-&Pmh%vR6YWI|94`gyGl(Xj(&q&&Y;d*t%Y3A=J*xYLPnN(+qd z)AQs4m({`)jmGPd$~oQpjw1{gco=PA)0NsZABvUcy|E@mEy)q9QQI0%eMFY3psi0m zviODRScqdbw#b0qH@haG9NIw)TDJfB70mXsD{~SQ;hc5_HpnQ9&7*8DFFq+M$QF7K z(vxI!>bD1JVU+w@a#w;_jz0Sc-BJwUYjNU8K~xwV)bT+tZB(;UV`f!x;8%=;INA0< z!{CDbSn>JY`rT3_$E1=taO$3}s>kfxY5?~*v=xd^WEGU>xaq7DPUn{Fw`3rX7+rFJ zz$G0KG*J5t^J50j&UFJ=g)|Jf7Gxw=BJEzb(!;M3$#_*apvnf5{Fk$uWe1fX*n#4b zX${DePZdq1Qv<$G>t|I+Uplvxj|3dUyLT);C60P97I~5cXq(|r10&a4NGG*4-b6sZ zs3k~0oL+l5kOclNsX_e)Y4P7{GCtO0Lyl@Fqg3aB4SNRPVJYt|#v2U7V0L}QHEuoxfnND>3gw=It|#aq7< zH2tkle9y5j{4^E6_s)*oU?Ic!>XUNf{oVELR=(>ecFVsK2a(hoeb@cNafvE zon^YklEgKy=nhOqkp*(B?WO|vt#enubZ%!;mK*x`X%*wg=YKuvm@IuOm0gwxE^fE zq7E&4=6LJUndE_tJ*M*zl2X4_Bkl~lH!4}X)L_f1C@dMBItV9?ivvG_1z1us{YU9S zR?NVhAR@V31~ulp-8jb=Q&|!)+YLxoDJa0LM4#y#_^!agmBN(1o7+EIOnhG39P}_ZM?B&=7n(L`A*{9d{N-1_&b+)+ zbbXVTj$xZMy}H3LVhm*GYpDkfEW(~05kc=9Ss#&jYY)Z>xQJMl2#cF z(LRD_=@e8o_7vB1DeqQaMDtZ73RVBY*P5%-h_F{WeM9-mMiHL`V?iFWgFtrIm@%a$ zj|999xV%u}R6j4Me8*w0&B--^7!(hxAe!{VEG0BvE7F3u4y}<$WW-8L2AEopZ=3x5 zE?}y4Zz$DWRX@5|1s7Rz^g7+1V?pqwRycb3`U z?4xZ>bHMc8XzAlddjo^YpY1{>`9UpWI-v)*`mnI){IAVcioa2Z-}BHsk7=zC_XdVa z#}?mCT8AAA;=g`rD!O|agWc0jca{(>O*83s1KK*n-xb_`@`z_lKD0IgsE+5S`JA=ht!Ip6pRHQ|RR4DQR91{u#4tJ4M7K%j z$}ML$Gea7mH3?GLyEtXi8RJ8=TuIKmZeiZ^IKNa*xtCR6;EHPt6FUOZ_NKU*z_&{? zD#sY5A*r&Ew2oOszk<(YGjl-pynyB;cwpPPG1n=4)l_l0qbKQXLxbkhPQ`)M0e{2| z$fS2OdhB7}RGG+Gk*a*?P|NZMXK#BO2&1;ljm>H8`fJsrm&&?rB`7epyn8UM(VF@Y zuZNzdEOOmX(4`XE)Gk3MqE79~;k)zg z5-o7^3K1We`K)y#e7oE^a8*cF$1%niUDZalgS=Mv*lG_`9^X!AXIGWqw+JEJ_^nTQ zvRAE4pcV7?M=x|hXO=!QaFw#vWh`4EOwR622dq!54A@*T;eb z@2;QxGZQu0tLJr9KZo(3H-rz{12DKM)S$Jjy>+|&3C44~6fh)e-u$DdZGTk~@B1`{6OnGhm|#QS z4Ik;^8Jttt51NdMz2i0}gSP@oCz(>Ci$8C+X(pC=N;A~Xh^43y{Vo+ICjS7g@MBYe z1>*tTnuJZgZimqmqgj>Y<1ov_!MnPH`5Nnc&GU7H?VZ;VmziY7hGq&TFV*HvHW3o1ywE z&T#*eIhKt(Y5}|cc&ibW{GzQl8go@5Uk&W$*>gaS-;?pqNAER{J0))Z#9MVG?SGE; zL6q(?oq-4D&P7D+S#&L(p1 zN*j&(CC*?RKQ&mu*>ODigst)3`_-qv#nRMv%4Q9*CP7g@Aj1vVgUoa1t~7*{^*Q4Wr$RqQ zd5#ILTmP=9>kelR`t$yy`%_~`MBY|>Nnu)sDxuBF(-MGRioDApRx>e)Qw6iMeJ{Ia zlR?jIc;6{JdnBmz)>fU`gTpk%B^minTH}YWVYZD9_XsWK*+b@y;O>T~5g^q7iV{C1 z8*+kD2mvoee^}Ker{gv(VKBGkW8wRXr1W`lMqLE#eLSSBNNZWr(h_{p4l%kejkXcZ z8LRzOin7G?VKZ7n5u$dM6R9Lx3?h@w2vt*dZ%BZxEl`8?IfY4OWh$)&8NVvfvfg|3 zwyzA?O4n=dt7ZZ)Pcs5)lEV16`XoilhUC{g8uztHxli57cLDiBkBd%r58fr%wFL%cd#>6-$+)H#gVK^W2*eQ!dV)4>RT$J@e%F zQD$TAojc<1>ySyC!cHZ(=U$eu8xqj2A3tY=4cGCP41VNCLW5MRx($4l&oEamnWTOD z@E|3RUsPRgB5*V&!0Y>sjrBe@@xENUjVX@yg`=9pEt%d2v(BoJ(QZ=2@<- zZeVzKpg~1;E^29U{#(MfcJp=g+a9-rFv8wNb&<4Lo)!}jy~4Hapn#l_VUPd83q5NK zDfJYyPG#eVSMa98E`VB<@mPNT@E&(*ZvK}tAD%$5(R{L_F3!b260Ej}SmzY)g;uXw zWc}|bD*{rq8p@vVRQb{AkIan4vb}>jy%!zCnyw>%^71no-}qG?mZxEzHdOrTI;q9% z>(HkSIt!d%jMD~gl3Ryb*~1Kq7p@bCYEByzGZ-#o_ASK@m1Z)`pUUI_&CDkFI@hWK z&ere$q2H|mAo%7QiR|sQ1ZgHjtwp0WvFYD}2+!lq^`2Ab)l9}k!ug5&D|fv$KtZE{ z%SYl3hDV~#J9rkqV!lML+I}l_O9s39#dp~ZF)QdVnR>LS&sNxG*U=qMSdMRv>H&k{ zp%0@|BiD=!ASj{&LG&Pd_9|Z(;iaL9SF@4~-q3bp(`eJZb~}$JZS5U|)D)B&)Yh>6 zq4e3V1(#g#?!*+zz*?}>~Sgc(ULsSHWs50-{jQ81W~0$7!d`sDikO@Gt*wuMy! zH&N-G^5ch6DU`{2Kdrr#zlGd~MTP>ZGWZs^cz+lF8da~!%b^Y{ry0qOqJ>H5lNz)r%-)Ek-M3NWn_q`hT(~Z`5s=OE=Fx<6n`< zr788zJ4ZLU7LSg}q!YPqHf^mK*%9ozHO#&ekVq~cqX*xFH-=R=zb2N{_l-|+_l&`+ z&lszRl=Yk(JE%PA#XE21+oVh>NCwX|nGYXDq$-K~*I_W_%`N>{Hcx&_LoI+ne4nnZ zbxh0c{)BSQQj+RawEf~nK0{DbcFS8hO!J}Tq`HRo!oy17t7@3vv|YS**OrM)JlM$C zTfIJMo0_#u?mJuE{8StH^&@7>P&-p%G|qaF4Fakqo`d+Ms^wPCcW5++zcifL8bnr+ zc|=2YqwdUlwHzUDk>Av-QwN|aU(Reme~G||D)ujY#Y)-W`0I5*EIa-unK5t z@h(AY0mTQZGcUAB4^93ax^6PR3!;K+mT~tPYX_gGs~1&T}modN^g~S}A3sxk^-@yxL({eI2gnh|nWP z&IGX%paH(Mep~OK(R~K-j!t^;o{Z*zl+Q1B7Vw#GnIY~2_MS-()R+TR4Ex_5CBwTu zrrfKF54;shsA$M7eD=56wL->}a$W?00mtK%I@R9b4C+(1-7=Q$Hx#wvT^j*dAcgJy z&3kfLr_G+5NwAa)!aD5mhA(d(v#G86R;d>DFmYtH2b&n`+zs&Es^?HFPetA4cgJ2i zs~0v-musw_d7}plWJwnIH$b+SA4yk+$08w(TBEX+^m?X$gJC7*{_N8z%l$GY#@dXL zbun49<8aFJpm}PQn)~6LU-2$%z%Z}*7%v%(Ohj`}+I*K6g<6#kF{`(QU5CI0HDJAx zK|$HfWJUv_<%O|2l@_R-#BO2;qP~E7uA>68cNt)rzePER3)&46zM`8<#!Bn^T0tvu z*#fN-W4^vk?3i}ygQ!*Ta!NIEf%x~J7CKQ#PuXsHOFrEO78=a3A87y+Zbax8n<8s_n--=((6(v|X7Hy>dp)sO!#>iU; zTg96Y@Oq>DqSbgRCy%npq{9l?QdPgBNaRp+*-#)C?X$#WGAOHJufM@uv$eT?kM z-ezWjAj-2mABapnF>?q*T(ppi;CzeM5ob}<3_mtoj}L)iTr0>ow%Z_ z8-3hNUjzhP`|=^@vYxg!zQ_K@Wk2kGar8IDtdr8B2ZA%GA%G2J(m&ObpE#=)DvOHt z{2#{NJDTnPeIGAv)hMb~t!Qhtw%WU`s-mG(?NL=F)ZQzqR#j2dp0zh6h&@siwMSG! z5PQTH5{Ym6dcEJ@@8@%V|NPE*{uw9ddE&nA>%Ok*aX%9DwYdX6&!|RQ1}B+c5|GEC z@(~*jZ8Cx_@@t116DnSY%~aJN66|@tZ9Zz&EGIm_AbA-C*Q46qX1~qGANCJFJSfUi zgDHd-x&3y#oWaQmOCu?cunoQ-lpi*@#Wn$mJToON-K=gP<$VxvNtv|FVKi~Iy@(CC zxUbSo7dTIH{aq?+H%@O-tl%~_UOt*rzzXI7JWKdi=Hl|Vnnk7>VaEnre)5@yo>`6^ zG#uqlyQWd)z3q?lAt;ctL-QEZaQU=5tKYVnbB|@!c6rqg>m_0pF62b@>XGC8|3GP_ zvbC ziL-JSckX{ld&~6fsI>XDRQQ$#{Ki;hLm)V2tGIA0(jk#NByI5&Y+2bs(suWA zVnJgCjs~h+7ID0MJ6kF_4W;&3D*orc6^uyxGfMt5hd_2u9a*o&@$+3}^@~H?tSbI5 zyawtyF`6d3p1lH3btN_g)H|KpPSS2TBmgLn;FGh}iADK4aMfq|C!g0<3@4BCk{{8- z&aOF4t|l4;PkJ>}JNDnX|Dhm3;e{R|ELm+0&;izA3v2*?2A(916b8akzK$l@dr%Ia ziPn;Tp|dZ!)mbbQcnaI4alfu{dut(X^_y?~_#MdQB@Ttg%7i;I?uevij(H|zd^Og?4z(be_}o+9f$dwO~P z{m5XUx5diD8M6IOiGj?8x-G<||9(dDti2FWeahyu!b!>bqJ#tnnPph~MKsuSew~g+ZyIZN zW(b~YYO>C^r^q6v{1V@PPQ}{R%+<)U0SMNIl{n~m^Rsf9l}gTgqg^Q?qU%DL9UrBb zMsv^Ilp6|Onb!&4S_I}nGho*8M@;E1OhzA!j+Pe!F7!?Em1FKNJk}j|Wb+Kk&A5hh zwt}+`cL&i2AaUzx-Sq3&|AOOu(fGJDoZunJgV_k0yROgX<)Oq*$&}?RC8th;#a00a z)ga|ZEGH!k!St$xtIhtwpZ>~TgBjDGD%`vzmWzvDm)UC}0AAm#J0A>b$hIdq8=rx2 zmM_0`q+8B+b4YET-MP!xkTp+u|}`(`)6Jfs|?*{AE`wlW2m?Qh;C zi)>n)Ekb_`?Piz^h{~DfOR?ah8#uPmJ-%nGIWf8~RDMk&3x0m?iPYAjPSsH~a!^8* z7dm(LNJH=I{!f$}6c!0bln(Q3KH?F%--)Q+g}#0hQ*M*w9}LuXKpsu#(Z`N`^QNQS zx-RIUpWB%8j$vz5S7_&@Y{2=04b4^igOC|&>Pgj=S2ygXN{iW5v!EQ$NMJU5W4-uQ zzsypl^vHT*2iTuDfn6{E&}OPX|5wlvw3?MDnBPP!`j;KIa_R^*4^M@bp)6~wY?n8- z8rU@^Vv1ZX>yul4+2qha8;|LMd^0pETjtD`p;r}#jLT$s2_LM<{VCTN5SQ-z$MPUS z*SvOZ0{z=*H~HB?Z#n021^v!sKutm%;KevjOM6oq?t*t~i;dXQ=&}U$yOjIi$WL$1 z+Z3i=e(86_wgZw=aM>~2oHB;HS?@aSwyD-aojMVU2d>L7HvHXFHFS`+EzuV#$ceTRtLIZxK(R=Yk#1o&1G+a&E&-to*jx1j44H zv#`7^P>uctNO7ve{w)XOTaBt*G>KI< z2A0auf^3iFviV&Y6Tk=UoQ}77RS6GMck%J#GCOEi_5|WEh?J!h4^^Fch4PV{leBs; zp}|7iLcod0ZunDyG9`c7JN9l|vB=zXc@Pm3fYIfJHx0sRkBY)cKgN!DvkTmPXL-=f zF-U)+Kj6#W;qSvVYpFCu`~J1TtkgOytJPkPSD@odDu|%)xze zet#4%r`Wc^!!u?B27WyIN$pwD4&y9IP@x=e8KVufvYBlz+yg(5YwO4Z+3tBLmx7bmNcvw2;W@*UaP2B4was~ ztR_V(zniD8Zomw!U^?+Em=7$|^b@?Muz?43<>~J64o#paU;S*iyu(uTkJ!%>XgckK z=g?8aDPLcY%HBI}tUlGSDnz%C8P$aQ%Z&-qBpK;?K1tm56DbYpGg4S$AL*ukX&Psc zZs*JMk~P6vzcahz#Z=;56V2hp2@_n)>^JO_EFYY|dPQ(;3%og_Cw7&Cw_ay*Jt3(( z!uQoQ>|2BKCP=Cy=H83DLg8(MIq;gY$)Akt+~D2Ig=ZkMxRcS*w_&N=-790ntmN+3 zpY2_~sPAGOo+lJGwAkBZpgICWqjuF7KIQ8Q3z3$~f`fia>wv6|Va$}7lZtwOo|K>G zCqy*6s=}{on~raNAv|~zP&C=T7%H1+d7nwgY-^O8H80t^yL433Ry_cukixe(IKX9vDw-2c*pgx|C($hSGlfWbh!Y{Tu>PY&RyVmkHaE)9hH;`%SV%Xqp7X(*Zh6%pzZ!~(hKA4W45{~ntgo1 zpW1a)hSfqiz+4;FU$qkbZ|PNPG9dPv}2WWnjD$ z3AJ}8}2 zTZwbbAKTN~e5g+PdkNmz-hu}!z(qG%|B2RL|Bv8|*!%}F!2f}a2C%dxgI5;h%+Z+P zuM}!Zl0vipgBj=-)38s()bfYKf8YYy=-B2xU^ZWYD=13D-ir6h9!ObhkI4C#*>G63 zSLj|X@Ury$L%PtK{!Ak@wc8)k03?BYO85M3l5X|S=i1E}eU*!mMikENT;2qByr2a! zSIN1!&Yk)y^0Fp<_v!hVZig-iJT+}2HFTF0@sW`5$4s}5Z+>G36#a#dEHCTL%c2)0ZlaSLq*ceI@9OH1ur4L66E~%^T{tFwQ z-)B$)qi14C;pYHE!H8?l=N3AV=55Ut%HZl{g@{H9Bi{}WsqR-%+esbI6BmG;kd#;h@>DkY$BlIeVq0ZFci_~uGvA=#HkFJ`5Ty@Rk zJgjSE|MAjEN5s-P>H0LzJlm7DTVWOvakxNF(1V;9c8l7m`I>{Yz(Ss?ePLlM^^03T^Z+B~Dz_)$Gzk|X_NHB;_-fatei^SJlvu_(I9 z(VSvzIE2x6_Qbd6)kSNBF>5&ppC1vfOwH?cKYKYBHO2F3*V%NiTELPCnL4GWFE`}s zH_v@$EmzWy-Rmz|-|~$nHKv8q_3{9(#K(hM0b3KrL`&-=uYrLI4E(r-$k*nrp?iu~ z+Vp#d|Lh$2ic!f}=I|GHlGNm@+8jq2B8A}*bHwIYw^&AOj|n0_0{_8WddVPNwEd`T z{5+2=-fd$1Jn+mDIhZjiUgnV${s{H*d;AcGLi*n}eq!|06uO>>Z9Gbx~ zw$H@7nPLrBJ`v(T-h_uHO}D`g)v{M6B)YDhILQ0VfJP{;rWYO?Gv74_+xig{Ig^j( zxI3ge>~+T%as9K&FnH+GYU{*RE+Qb=LU+w5quU6nfjx$THYrNzA+aY33;@-`Oq5W+ z)pET)-ytpS(d`&=qCw=SbeQwx8A0g&dQF1O2UNpS)e1oxgsiFgLJWY4saLPTRz^Is z&5H1w&hq(Nh!LT!iIku;fO1_eKb;uPe@yj=U)Kq1a&f-Z0mDkv!#c8(LtQ^tUuL}--hDbS_HpzIdv>FLHF5f1NRBzrPmWxf{ zkD#0;J89`n9jj3dB+oJ1D>qOj{4f@vYt~#TJ5v90M$r(tKOkl_Z8yJDGajV4W2vs6 zCRnNKppeALld#G`P#y`QRn_71>V}+*-;IN{zxN~E6;oD?UY}_E9)T7d{f+%rp+x;P zdaG$?hLN3n8#@uOUuXpDmpZHZAY}!gl(UYcOfL?eF@zs7Q1z<f;P+G0_$DaMPT7wN7`XMBXTtEzjgY zvI)B+-KnuXTC^p{W!8=lE&;|5yyVt&927a2RdM>02RoQ_>h?~Cq8&Zcafp_6&cx}b z-|YNBPQGr3IyjM$TAW)3zWuT99J@~i>3-@QkduxIeGa8mBWbage0crKt%c%XlWH<* zfi$nj*AF)XX4T)G4!^ZZZq3;oX!?sxH%?5YN!GFZ&{J8bpg;K`H%SVVl{ONhj#+^f$DIS;wKbY+vU(FYR`yoaXTlru7SIW*O}!RM;mk!9 ziH@`Cp_Y-B_PXzNtq?m^wOW#H6KX3_y0^$#7GJ)h?LX3?RdUGq zzq3NAEIC>MMG=hMp?^;l<5Wa7UGTG2TMRN>ps(^oN&6 z;A*(k8I4oh$DU$W?x|X#jR~<}-~7&Nk3kBW6jnjfMP1l}jpFaRSC>oR;uZ!gZHt?O zx3DIfY8HI>rq!g$8x=+jVS@(!KGMFg(kII^a{5TekA^udx}Fo|O3{}3`f;=h?>01F z?x3w+NqoNw&M{`uYzF@2Rc&2|sxI>sH#UgtTV3=f2ol2e=&neygIc&*Q`o zB?~SeL8t`m*2DKK4oye=S-_#;X0ned4bD0%#S@F#M-nuaa6;`DZi&3@b5J}a=o6G^ zor?D=?a1QKlA?3ld9eMnDf~f%QYh89R4?;*WJ`P0;b0>NE2eczg|en4I>Jbt>?bJW3sm5I(cUf}X8!M)W8E)LNF=`y~{ z=~bg0S~i~Nn~n)1ror!iBFByw`GJEQ1S@L9o z7X83Bd|$7Le`G(8gP+zc*)eq*D;$(7yz|&RSaVRcw;VL+eCq=0FFWEby&ISkZUc-# zz)o=^9f<2aEQ*{`or(xC~L+G5{)nyg|CjCGL;)HKSH9X(+HMSIa)=!U0EW_SFc zBfgG#f+JaM%`M-YtDLQzUI;k9XkrtYqw&~#3GNq zPPJfanl3PVDAztatJNi;Dl|L1>-3~$_pHWRBzbi;G$va^HR?MeSkJuvt(N4maPDRZ3N7kDAG-85IU?60xxnI?!WIQ zC-(I8(d#0w=*+JGz+1y)+b0tTc{I_qjqjPafA?L<^}eh3)LQ)6M3jBw_1Tp9;9Ixf zJfseJmcxblaVcQ2C*@B_0T#505N7l;8IUKxA`>s!xpUG_&-gUovn$t9XjDWxEtv7>WNqRD45A#|EjX z4;icK+Hb||-$>7fzOfTOcodhfD_HXMl~OfTh-}yoq?9K*}e+|T;K z8pgF`fDuDO*W)>#YUf~9432+T4CQPNE!V?#1#=pOwE74=My{-kr4FtpH$>lw_Ki3DXP&_Do*RuIbU1`CG|oR>((8i zy}+D;9;Ka>yWC0ni$k=ESLz0&)9GJm6iFNZdJGNv4zD=2W5GKI)^8WK;cNf6;uU}3st;m z7N>4q8OynnSm6~mBC)M9)Eog$DbzLQKy%RA%}h;+<{$gb3dLR=0dpjA6li{EqNlJ} zyU*HzIjzh3VG*tNw>DvdbK`JH0XMoiQDZLAK%n=S`}`=XE~9XVC`8>F@`}$XWkg|n z9yqA9s(~J#I7SU6>e=#P{EUNdVNwt|qhP1Z9KGH1dWRIyc;-h%qZLQDSC%H5U;_VU zxyPT!=tc!^*e)q>`FW)xC8sAcZDMOSyYDSgV^-23w9>I|9<2Ao7kVYq#oCm5 zsJWr0fZInOUHPb5t@Bi4+@$wy$K=)l;nY+6ph2E(K9Q0_D%pec^6EB|TPg>=XRWFVhW26j?QMS! zYcNacMOeft$!sSYjMP4#w7tK!r{r*Vh8Y<)ixWO$@UCN*y--VhnBew`BY*yomZG9L z^@CLBW7%0IPTlB(kN&S{Hywe+B`1|Kn*dtwDw9Dw>2a~aOEa_qGcK|@c>qlt&qnUq zcY7~F`vr_9?v8UsPo}=@a&yOwg|741o8k#qp{($T-eWc?VK;aM=+>IyjH0mjUFS$H zH%Jg|mw!yT8UMan&endeU=4HQ>-2xSwlRJ``ae}J(q}SZdiFhMH6tD4QmiRJn~#`2 z(};oTWSW)rllxJT7i;#;2{Rb3qR$WJVrhJ&H&x)loOYj@ubRRmGwTnvAlA z=K@M$Dkw#p6F^kQzFOYR?#t#3L}NLSN1B^CM*p;KLWX;c!`=rLb4V%Zy3Dr^8LHrc zzrB0+gjtA7i0^FcgtlwtbiSB1w~1p`aLzsU2!75eL%cR9?Dfz9@ku2t8LhO+pRlyM z7NT`!9$BNom4U&tdgrY9GN3Q5L|r|uKy&UL0+Ur<0Q|^_j(>VuHD(yx!ME};=*>AL z#8m5{1xqX#SuX92B~Z1%B^)dmeJ5&?C*Wf)b>?+NkIYqQ4`m&)VU>voCAy0~P9w*- zeXG=~^9P%{A2s<%#cr@WQQ=rJCmhs0%Zal0o!eUY&hwENbK)^>ImZ7;DA{pT%%>w5 zpXD-6_>5~bXQOvqVbvzTWKFyYNWMl4gP!f=^`Ljf&kX1RL zd=>$lEhM${DlYD>&JfGf^hED-PO(RO&kc z2QFbIe$%uZ3j2*P+zCp2j^wl)Z^YoG+`M*-JS-mdRr>vwZ?ax>xmm*hx33qNk@`B4 zOkBZH+IrlR&QvXW+RRI_mbfgx5 zg8j}7tLJ$w?{Htz$Rdr3S}h-<%gh0ON1AYlv0bQ+h9y^#t{uV8Q^>nPfQ@pi>GCUA zF?fMf34`cLczC~Dof@L8(jvI)X~!6qCREmL*gS~SPZ{3|4negq=$L~&S=$DbW#k^b zux;(s%S`A8n<>U zlc(dcI5qaP9m%8lQtSH*s0jJz|2w`^<7%y*_Cof+Ik1yV;l0&{9F1(qV3 z*+Xq8@1p_fH-hqK86xRgrm3$u&%AbYWOaV>|2jMDhOFiiBBn*xEH)7r+0N}o?jk*h z(n-`|Cqlj1yqA!hK?5d_Odhj6e}D;NKpq1#tw_Ay)hHixNbuQp`w5Vuvu`7hscP>Z zkHDiM=2|McWq3|e3#qQOIjHQlG*!|Z*s`NGz6{}p*Wl?@PD>W3OX?kiUE`(S8@y1N z`Sh)K{T&K-1b}%`J1TK>K)AC>Cl+Q5LQ=%wY6u5WW;&PyGExb>xSB zX2-f@S<%Da*Is+P{uaVwWNQQrkv+{>Lq0J-hb@^#b!KpQyY_r^k>}st&+F~qHA-fb z2~7NNA8)c{<$w=JyB#}c?>Nk#iPH4R$@s-0(C1@53LQ-R#<||44n74&N{a@e+b4WL zI0c7Bq+`0&FZMW60}qoJ!5N^bfhuyzG2hAwL0poxBLM z{A2i=fj|T5;>Dmi=ul;K;CM2z#{0fDg5Yz**F(}cy|>8})^pNf=WE&(E~ug`uiXQoh^x`D#@}KQZ5;58Uv+iYoHRcPQKY4=$4O%1 z>rQ)h^Pw&T9}v64^Ys$%2iy9eB|X-NiZ?!M@$5x#4QoEVmSlXzL;BJ}3-o;5f%4~A zkXO$`$+XoiVe&xm>A&^5Da)^N!@Gbg%vu_ddg>md$Z&E_BX%_uro^xF)Lc<;k5%J! zk_~e2k+0D3+xu&-0uFa?Me7+~BA4V0H*^Z^c>?6Nx+M48rNqw&UxE$vc(#_m9q^NW zMQIq?O`Sy9x?+L-8E%47|F<^7m1wOdr-q70D;6E*uH$=|$+6*!5@}_Ai~Uq1VL*H8w`dfn%?sGzabL)X!_D+tO03T3CpvMArf!nIS zI`u4K==u$6!_a=up$rg?dmUk8B-lPi`+Sq7${A;1wlKteNdY$#+|)%$h>_sM$)9?J z$0C8(h(%w;BWps`(jKKwNtJei{Ytu<7wclI=cvhpYmT^Um-+|Hza50WZ) zW~ff(RfnTTl@To0$jJV+^-bTwI#ntbKRAP{lf7(?N# zPxn-gnrHn8T6xtTz52m(%wq27>O9@nxa1x^;ed1jXI)3l2k`#IOm6su1Cr?SBj&V4 z2`904XQWx6ROjc$!-3tt9PsBaxf%0JzIfq;m&j;ngAhhBXsyV$@G#$1vTC;kyeYj) ztZC-(irLa#mX;;!uqe{>pl0$NY;Y6MCVu{hdvhnDy0eSZSgTe!^$IK9r_VRf{!1&S=~5nP&eFmj}i}yQ(E+YdB9=ImH1HZ&46lq?oaohm!E$QVICLKvcHw{<_BBir2*#S zY!^>E@z~U8Ny_pfStRW4i`3HoW`XL1G#?x&ZN zo-c(>jER;T0!~eUA&sr$FW3s2@|?OQ)<1FZ%hv1OH1yuTesVjtGO_W5(hL3co90WD zT|bW-fm!wXe{?jdm0Ku*WBMS~r7QCe&Hrd==bAh2nP#sibM>?{;$B4dQe`yH&VAio zsB&34KiJ_9SFCP!S^aHt)t+-_8tohKABewnB)Z}Z*kNrEt zP?}fPv-golr45}N-K7MhqpX_#djq)^Y>{j87i8wURDQ5B{KDeG5}ZjMv@YA0TOvvFa2 z_Bb4Rz$vD(H)``J>yi)pI_~Q;M`8PREHU!Cp`+RYt!a#x4n2sek%7SmU{Cgx{gk{` z^Vr$)pXToLr>x&Czbn5V9($&IZ| z5qVaXZZ9y(R&%q+HoE9*s`CE>6r9rj?>p~3@Bi5Q5}w_eM()x>?&1fr7?8(o7V(XL zPdL*%Wt7d{gT)&K%f%6&Sb8vi%Kt}#i|I5=ZbsQ{qXg>H{37SDs@wBi6Z}J1yM8N8 zk0w)DqX4eS7?0U@VEyXfk=+BGaHQyDK9%KACA2KRRasIhOiNgrq3bR^C01#}d?>vzeQ0DR}< zB4mPJDFR6OHxEs@BQh2agV3y6R0~Y?*lWzI`={n3T!}?VK#b%+$npHf-k3)lgN$%$ z2eH>IB$s>Uk(_>g)81^N2{@(pfV1s~n?xel$+I^j+ z-qY>+NuYH?!70?4AGok=f|;`N8O<}#AEzIginT-^q;mKOlRsf=%m$iDMY~@fIi`GU z-~RK->u>$~!;4RmPsGDSy{xay^0`6WQ@gI{W3vS)&U2k|IX%$8V4YvD^RCkTvF@D< ziGVz|JQ><^bf)YMZNC5eb^NbN^rGS9T^Ar++3(v(>jeFo(cn{kGp1fd(CX?-DAi&` zW08(W<&};*u6e|}R;Nys<3=c+IoNd;!#Sxif&+Eyo$G!D*2d;b*-gKyMFpBxbn;y0-y+EQ$LDA{N6S7OL|>(8Z~ zA^QQTU5^`EZgVuwq&Luga^yJ7%=i=?Y-}#X?0X0UNCVVR=CT5>RetTlmgOyf`f9P7 zywIg%lX)P#dBVm5s23&8_^G9|BwZ9H(lV+8m^NHYLuedHbuqe2_|! zRstg19&MtI=iY+>##sx9xMtl9`nu#D=#Q7A;UVT_y29g~qNe8bwZmJg61y4{FSIkS<)X658qG$QvsIMk*8Tk!80*xd<3c#TRsmvXWs}89>@V1Ri?tZj|NlRzbFWf?quS);8IFIE>-XMJ*4LnyUu=_Ny28e_^K4R zs`H9LkOs)-<%_CGrEXRGy(wJT`e(9C>WoeR|74<+K#!NhAx1pp^7zwE_LplH#;%^JoztD)O27399!lcEGFXOF?)OR$fTBkhc z5;e}QBq~jeDBrxoDE|6$I5%fb+7$GRiDvbln*Bs|i}jZ&z_ssel(p9T#)E7JPmH;* zoKbk)n$ORAN;*ip*jKSYYfzw;`%s$L1hR1Y%c;a^%zWT?d^tx;9hcX9uPNe4nD{&$ zr=krQs|CoEZbtI_ZR;y?z>85@8;i|?l3e7RaLo?Co5CLmTo-%Eb+-BteM18kfHY=BoYE(ZDFIvD{~NwPk6*@ABQ9)ue)cP>oyMRu=^?WOwFDxEmAB;TH`im4UD> zo-f=*Lx@K7p=|K->l5FVyYHqu3_nrn>o!kyo9N>Xz?~f4@D0tt%#d;LEQt0B+)}(P z_Ao$VDO~kvWi;}TmW6V^au()R`HW_!N8;7G8rK+MW?Th$Sd|R$= ziN8EcrgsvbZNG+TyI6dU&P7yzV)WHEyT(f*VM9RP9+WOz>+1XhU?3h*47+l!EvEb9 zP;j>$8!|OsLcGoW82v1@-=p=D`1jqCv-Z9*0p^<5aZOk4e?F1-IV2y@HL%z$a!YXi zfUtV3c+)Q7;L4^e^ZG+2ah<@U{o?n3)U(44+27Yp%H_4~h09G$w)jE*qN^yjM8X|_ z_J&lK0qKR&GEKdO0`rCckT&1}%`SRecBEa$-1y)xWeYp2BuGLmtof^PlU^+itNwH1 z3ym$tCQ+Scvw^Ag?;W|B4}07D%T-G;MWsBG&a>qJhKS}%PJd?1`Vip!i=e;mX^hLc zI1ZbtF?K?3bksD?4heG3_)Mz@Cc@4ew!pKfb`q^!1_)JSZtH%Z>s_UzENlugt7j|{ zz60bu-0WL^^%qcI7%mFhedwm7=u7^91Sgb_Et1)(jT$OYaqiwJU?Eq%ntk@Jz$7gnbS26pYoUz*e zu}z7mF9uT8N7r~NZu&|nDnCYT!T6Gm9}~UCf4GlRX>8^HHsayE)21sw2#%EF!MMHq z`HQ26lK;`y2gB({n}2}ADu+xao@46T+i7I6Us}PCAQiuH8~y5e|Bx(*QrBK8n#iY0 zJbk(6G!r+CXiV6uX1jZF+DC0&9crfw%;bG)&janaz_^XK{xR%o!{;9|Xhi?nci|>Q zhn@laUbo_#mmAZ5Gx0F-ZB?`-w>~&4&BHSCs|Z;HaN6b2vd`$!ZA|`HZwolw7m)n9 z?b{HiqtsrUIab_=S2%T22zNi%ERISmpv!JjZT7_Q5=7<9bd(87k8rTtVs~|uZJxvK zX>j^gr1pQv8;QR*tlXlzw@unbm-2FCU8pFYF8fTCv>}m!E%2=84NLEp{kgYj+Lg3o zmawsUM;6bYl>N^lPj(96V{h}EefyVZ1B{>?xi0$59y(#&f$&R#PX|0Q^Fu23RwzM!=ur`BV1sB3_gNaAV z-X-4fS)F*IMw8Fnj@{-Zp4*PxB8}U%2g4a+l$(;M)Tmp3YC(=?XDL?O^y*1g=j!F%zkSy69T#G>0t#Fg@oS4!4L zJCX&h^58OCnygE<*JbB$HXd=o%hxQDj9F#pYvaw8j~@K-&Bs{}7x=$pNiNiGw0v`g z>h4Z;%kZrGNS3Vjh@}7mSEY#-c|PZ>5k=6;@9Pg|rT=zzcyZ6KnEq>Hin%*glP&UN z^Ks1m#J9&&hEwpadYR2^t-)dnm60RG-H+NUMH1*qP(aL*)!Y7-{lg8?{^|Rnkzs49 zfuBzZ3DLvMko(rFD@K+{fW9LZ$$s{eSrtS@SJ8M~AH$m~OWTA`)}im)FyI^Th@u@y zC}t%oqAJ6!V`%y7t*+Qy9fd>4ynDBsvYP6n_FJxJABk&A(IWgWt%6k=E)p8rbu%=U zjPYau8BVX!k2QG}JR?aqkVOg(0h*vXPncbia{e9{Tk9C>$5^Kb~Wd*pa2b0!#E2~vDZJ^_n3G|~_ z@mcb1w3$njrT470B1)?);0u>HAYiGq*k-4?5u@NfL2ER)O3M*8E7|SzhCh>W?drmC zsvhj>PMJ&XbdJdmtV5|+OOpsbn}U0{s_e{R3HKdZKMmm%VvH zd;{IH37J(1*?6X4+3nLduR)oQD*n-{z?JrvUR?D{6bIP@$!d2k@%hiVz4iR&fhM~w zc7={m7t(mI<9~8>!>hYA%0sziO^n+&e!D?CWAt@?upj<3V5bG-qLa*QbgogNtR(fg z;0?>-4jwb7FsQzk!wixIW#(NK9cBP&H~xfE@}(jq?R zT{ce~Zt}SC-v%SQTJrlTCEj;c!-X36a3%F$g@8kGWh!U#w5XD25}aQC=T!6^Bjh!m z9_^}7QJb`hp)uGm$iS{o_R%FZ@av&Vwm{ld?8$CT<2=9R7|ZucqfqIDwpkJT-M#BJ z8a(SF?+&X>soF3<5W@{FaZSFdBJ9%N`ncQ6e~PY3j>;5DCr!@n@l=DK1H5$86&MJA zNS6?pe-YQdm*-?Em5At@d%e2Z@}!E-_CCtEE{sq>Tq12Ix&6)~#3CQ&+b<1TDmA>Sno&qCpWl+?v_hve1mgZ8l0070!wA6R~{B*;N`#m|6e&lr_<9M zq<^DRM|;kMymTJIuSt&d5pafob7VE8b`1+H-vPp|(b|lD(I1YFs^UJ~(PHMX23&Zp zD;t>F7)g7NJEk5{)2hCEqk5zmbgMC6kVe=3)<+5fBXoMiWz$Q!BAuwF0_(l%8QXk~ zhBrP+-jANJ`S>76K5AaY)e&^Tc6pQ9rM{HMYPY;^zVW%yO_<#Z2B>GGM94mweExxH zcq70qKDq9mh|K8qbqB9eDaW~YV&x$;m{OJmURsy3Du803t3i1Sj1`l2{_ya2Y**9+ zx3e(ud6pD$*7#F48^J+ z7mvGFV-~1~$Oh;X3zxpMgmR+_CEgt%Ft?8mFK;f7otfR!)4 zXY6!n*k zw+^+`yu>!BE!j2fb3Z?*VtEpn__mg0X&ih?M%7sS1RIZUQy+CiHS4~NtvjM$)Fm-( zJI&$(Z~165In$d#*4ykFX4JG_N97{c#4$71NH#^ZmFA-N+%ReRS2FjJqE}yYvY7ez zIVw}ksERz3CGRf)z z^iH>(HmL7$L|Dp6%J- z&oqn&UCe3`tmZ(oO%^GyRQ2!1KL5a;;|wWEP`J-Zpy0GBMSHJ&*t(MySIf>k<}$hF z#y)=oBjxpQw&BNkQnaXthSEYX%J5@W`l0;(0 z10>ynMedb?vxf3#H}#n@*K|5i6-cVJD@!V9T8dbx`B%);TrB0K9sdbCR^3t3C_FCR zvU3_bZno~+Z6$rMQtGdksUzuK22qWwDc3a9C?QB0WwKwYw5fV3c_9@=N+F~JP+>g` zrYCcLKuFZD36twO_O6KB5_#hIEkcbWzg0WPHj41Se1;i>(dU6zD+m6$X}IVP?OLTzVhrP5U({W z>cxX$PfjLD3B84O=5hz?^lMW&C7#ZAzkm#o6Ar#GQX-CcFZjq$KdsDgX6gd!=rpuq zsL&7>{c@&7Du{zn#qROMZ)g#83tdHm9*lvb#iO}u&tc8E#ebAc_W3eVePavy+^cI7 zCUtL5)~wRG*gBqt0%enFH>VPNScIlB4d=doYmVjWdDr!{ow&+j;xzT@J#^aCF-AZ7 z75EK*yKV0m-4zd*JzIWZ#XN2+y{3Y@2@f6UGO^Fx_1o?b>8;)1ZgrLt6iL&bu>&h; zgD-Y`>V)irO#CHA-gpboWwxv4pzSdtZYUsXUKsphO{;7 zV2!EWhptcI!PpWX(EO{Cu!)OmKi3oYx6c^*&QCrW-;!_dNs z`Oo_a;2Y^%1~-H4X}WcUzM-Vy#&eiW;YQ3usNS_)y2z&qF)|fadI-Y_FI~$O$&n+q zv#{Pr_Wt|Q>b!*k$XMy^$I^Y|rb9^_&d`vGw&`@uVdl0zjJQ2zrqjVt?^6%E%ZVE3 z*~q-Oa^FNFc0-9zO^OzpfPQ81J9I@NeRpyUk4i+bjY6M|fA@`6z2V>dnRx5&R!%nwpwmcFglya{Y0v-T zv{TYnzm=f;s5efM5$~aYiqi1P&Pqa+m(3elR-34{ zTsv9sPi^*?gJUldy*r)k!3!yBTv z@&!S1GK~yP-FiOv#apUwC{Uzfd`c3D?(5Tu+Pp{nTMJ4clgJ`uD< z3BM>nrLy+2ldWE=8#d!@$~scn^*;@zS51iIK6^PKu^LW4(-EXE{&_yMuO1#lm%Zpo zo5{{-E50T>cx2Xb0gjy_QUIO(9bz45V2zr^)E~m-%U<#;5h`A*P^DkMEdzjxEHif1 zbGD5Zh}C(P)u}j=)pc#l;p!oM@w>3Bz?(@rf$bk2*3VWs496uWdjH>vR*xf%s` ziK8*H+;#8}ZJf0U$k9t~2R3;;naYM8qM5vM_^R}rm(P$~jnl{S#Y$*kBMX2OJ23Md z|9pop#GLZAo5=P(6V!f|H^?P!#@KcxY3K)Nt5=X28Bhrp2MixbQ|~EPP;t{5Dbx`M{d=Xluv-kQ-;3lVpLx*{jZ0Y;lt?SX!+Aw z8=X7r0NdN^(7D(EifSw}h_z|Ix%^`Q@peeZO|s!QVv>HM+no@7@=TbJZMOh#jrfh+ zPDJ-tt?N8g%TGEv&vveoXX#y}{RqPK>8}u(*}5#F4K)D14`b4tRS}vC{CO#x;#+rH3pWb*1aN!ifCrI|q z%SihF8v+uXJ+1Kp7))( zXU(r~Efx#uLvzmFRlDk{tFGdL9hyXdlmYg2voERVvX=={Mgj-BRKf!=Oo<)HMU0Pr z;<2Can_CR(95#C-bJ*4>kG@;{*`gcxWJ?PEmcG~Zrw9+i&`O}=iqiAET_n0Am1(7K zE7ZQpXd;@LK*HL;rXXL%ys9=Hlw|!jiQ9_ue0TbNZ5uW=Ygiy1zhruf)GIg4 zvnKb0DZHu<%SBh~ErUKpb%i|yRh@*e)ApYjO2_@Dd-g#pPL8J+P95|41b;v1r;GKdN2cGxWM*u6v$o%{fxeCItoEADH>cSv5h``mS# z^-1fAky-0ox?bPUz5+uM;zZ-LU*GOkOJ$Qa{of15(N46eaUr>O32ShmP;T-X?AO# zZ?&qbem-Bm`y-ZU_lyfY`*jO%Lv*@X7}5?gzyq8tN7A?ZkWPapk_6M<{5HNlZW`aL z5ccDw)X3lc+^BMpi%nyRqzAiL6TQlPs{m)kVpWV*C-x2x*E`phx$KChu zCmV6PA&jpo<_ZcE2TymUqq;Qh#1*i9BQ#R|(-TLd74faLWA^;`m;d`~6aHf>{*OK@ zcB3gS7l*$=^B4bLU5_gd(79K%uA=|RK>yiT|NF1g!|~EvG|6OeQG)+h*Hb0~o4l>= zi2i@{VE^L@|JxTLWYoZ;9vqjO>uB_{pX z4fi)!{^LD-?SNwubE_@)f3(j3*S>HHee@4%eylwIyX#i~k4ZQbs385{uke5I+kdk7 z|Kb(@;{N{S+qOq8Q!-}Sum8e^{&lJU^UEgKfJawoh~BpR-@op^`fUI{uxojUgx>tm zSNoq{nP>QT*mOS+X#Hn1{vWORUw`HQzXorBK9Iz%u;>97PdJ~odQ6;*`k6qz4*kra zTP)qbU8^sT20`Au{m!IsZ@9Tiq1Jp<7MLbNKP-6zyCrvh&rBg8%cbTm_;Lh}m|y!) zzb!mu_HNkD^ek}-E{U}0>R51hH9lN1@mc0$VeH`U`vJ>}T$Wg@T`ZX5upo+rx>n^R$8o85PH3z+lM7Wv zA>dz$Im)WAl!o^6lj!O7+9j_zf>>So8k@Q795$&NRk{~@YKL>X-SJ2FgflS5OYQZp z-4cw0{%eFFmL?8~90mB7hto9phx?mzqtl=y=~ENUm^H0c?j^6=l!oIe7`xLBM*_Ut zQ{AJu+6tx)6h^);jUoPxOPW9doR8ZC3qW*{)DfN&N)gy*qj&k`y?A#$R%EQ%YJfZt zsm(9bj5$=%TzIUjvr-~C^Jh~xxMh*-fnI4TmFC}zB^nWGa&eU!5q3@I! zpQ6pjB5|r-=;hY}XuwIH#~Z3+*~f$@Jx^UGP<9lBBX%Y=$vr&9dZARRXL0`6e-uvy z2K@p3XE{kNk1Q?IfInV0*IU07%BXhdD-~i)G`Fa)xPTUrjlEN5E>Kew-LYk)1YzWY zi#0=g5O#J~{c}Fx!2R!hmm&%Zf2 z)go9LA*GDaJ@!c7yj~$`BT9>F32kea6Kqv&fcj~4BWqr{>&Mo+Q&>uqg2Vgaf;Dl$ z>+V>}e9Q)&`U-v+?P4pJxP zF3-g2f|+U;>}#zQZNWf%pTbz)@jTV`dSUbB!JSx4gv`hvpFrmLf8g$_3Q9UcR1drL zpMhrxt2L(urWe#8zQz6Wot;yCD6{Rz#a(`JG(Y7M(Ys2)Gb*-aqIM}JnA}U&dCuNr zHsid;TsTYcIT~Pl$aEI#QUC!Oz^|7q?-MUJjTrvmZ`jLSNN2Vw!iIM(piwT6oy-ZBL&b8>Xkkf{d{=1W0!H3n6|!Y; zt@KmJ0C(Qa`BRq2QSK!Z;Xq2zs!z@HhgDaV39x@jgN9E0T}AcNeRrIB4~7U77F4e0 z|K4TiVLTe88j(tCc_8OJ`NN1N5Fi-j#{_fML6TLLi&aEMl%Ate{RD3Q8 z#F7>q8y2F? z<-D6?UJ`W)*_(URXEn?#@iwH>DU+Uez0YQyhVP9Y?u3o*&!=LJYr}vk9w~q@u(mtt zIR6L_x5E{|nAf2o*s_84pN^#e$Km6vK_%kLjdR=fmZzpH2-Z&FF5_FP$XI>y6VbNX zFC~@vHPVUBvhIh>LA^)EE@A^1TP&9Gf9&Y~U#jETJEPtAGLQW@mIU1~2SAkEGmOo1 zh2{o`i5UadT;A=(J#RJs0jkeZ0uVd2zV>chOeW->$Kr?YzBCyWN;=!a9IM-9yKu?% zzldv%(JWBHdz;9tK@r+M7AS!6Uk7vsai{Xik)6;RPEzuOXqM@oi~d<-#>`z^gD zS+V|0#+~`j?G~6*+LxrXRF1<(;qY@A5>Mx!hKz^y?^*y^hU8o!q=!ih_qty9VOCWu zC}UNeaRMGCNC3lB+Q+W40Rsn%=YnJfMfdq!#VrV#r)r?4birz7RLtY>F z29x8XLBYqJaqJdnkmdV2(@S)(!^SO}$I^K#m{)lI3T(O>oaLKH+@+b2<)NZ=Op%}R z{w_87D0Cq8zV7Z;Sa^oocX?`Y-#N}WR-JDu1sGBIp@~h3pt$_T!lC71zj(ovMvz58 z;JnJyW;H^9^{xOJQ^_@9vT64ke!m3o&5BT9y}lO5pJFjmjtWZRocW@`VM|gV-FD9v z(gK^P&FuxIchbKTzL#wRP@|%XzP_$BIin@;T(9s$x!s2Ob0H*%F-CmC~%BYe|g2bNi6cf#zO zM4^En)>wj)g3RXw$vlJa&L4s)V1|a?4;w`4mSK~>>++|Ur&dE)$hW_c9+t6+^Y<)V z3cXeuXk$Ea1>%5>jx`w~mHvF{S5LPaTn_??jOU;fGqhyv&1;{@hC4vgTYR za)WYU>0_U@8u%0bbxkV`sF#`l@C6|m&}_{j?d*NEA_a4n<+Mhd>-8=w^DC-dbh;j{Ah}RifVq( z`jR^w|3>)!EXqp%21_%+8F{;D#vo+Q;j(82Oae!a9l!X=0(wV#Y&GJr1|9Zpf9)n5z)4p)u!k;AD82CGIu0*Jz z6fh~+Nz;4DY53_$ig~Wy#e%5<%xjlF%p+FdY_xlP!%g$Dfkl(s+F-_t<$Ow4HKj^* zA`I6?R2a`|4ZOxgGAj^vhWUQ>sBQ>QvI$7_CMT09M>%ldXyr3kY6 zL%QTGI?k&X>x-DjJk8woXBTXUVETGWS-mv$tsaEpu>qb}MI{t;#1=!i+)VabsLi z5=7gLGgP*4x7G=B)=zFuY)W1p7d&lq(zi>%>_$}*>*^G;!(!{XlsH{icJ}5rxd}4d zpZ>mOBmq+u@!bPF%fPNLOQS66;~l^Xu&!jP#WPE7A6>4EXrvk-NsIi?$;iJZ86sgo zc99f3PegiqSPGAM%IC!CXCv&&)yX`+=+sR-JHeP4_;RT@FK`w(7a5*1L)q1=oEF;J z#`1Q9UuMLf|86B@+Aq`*gnnmu5&$#>#RO>H2m43>wWB_!wS%9J5f=rjDx*r8^Un9N z>i!I81Yy^e`xdc-a@yp#K)l-NQMpWL-(b~(Tw%Do|6`E?sQyJR<#n=ghyT=e`A(Rp z=%GKIeI-h_5;`s)yB5Ape&Kes_Xt_O8>(x;_@Kwu!_81!l}S2XC{@H?ZwBi?n-x51 zAsf4mt_=M-;_s`3+t5C$V^J|TeP{z+v#IkunnYY7U(%z3$ZW|wNTw}D`tj@>@>!O? zDV5csELwq{b|onpkHoYW`}o1JFJ+jjqe&z32BE9_A5&ELl5{<~WtlZa1RraQw|hEqON_zJuA>f7q?Q4%wp~6L6=B?E(Ua=S5ssgmoMu%duj7^m zLQL4P-qjzLl&h$b`rR$KC+h{Q>M)k0`i6@>WvTFc_l5<2 z`9w@nRU4c#Z@<;(cL7Wqj;gVC{)yMlj6;wLJ*%K%ze_~+d)cVK({-Wh$!!NeVJPkM z%}H((;n{zHb`ws*NG{8qhlj;Yi8G?au@B7`s<8RsqTifne)_6V+D)0JCDmsX670AX z&7|QfzOO^aUq9Q87w2<26UI4++5^OZZSmJ- z=6ah_m|V8e@87N>?DUGuEg(4VjY?tMcmhppgPUQRz25NKLV~`g%OTCxBGohO?zEZ( zYu#Oxo$|t99hFFryZFBWz)oHP;aEgA!+>w6KfZgK4d)NNM@!-37g*>65pU{j4_*6*l9lAr&@CNKBGdL*(Qu)RjO5p#*Zq_Y~fHpB$#P zX^D94`ys~;r!%nqcV)_JIoGYI2eeWYlghdToeTxZ%G4?Hd@E+WKv&~z{S&b;x>~wQ z_`{{QUzupu3ncqKwg<#J3L5-nw70}&QtwGbq6UhFg)rmAW)3gPxns?qe}+o|&80n> zt~Gsq4E>TIYs!A+%Fxjc?taL5UG)zE_jV`MUK?oDsDgiCQiSKCO?;$enrxarQR6Wi z5?O2h?@=Gr6%fGl!P}9XCrETjteNj+Z#&lGpU zt%(82rJlnwAbkDj6#F}#<0X%}>r)hFFfp${y*v!hc5Ae9v3Sk_k(NS_Vf^BpSKtUP zWNXZwnS^(yv-Nm$iAUY}{3K&wU#^=l@mTvrMS8rkiqWTo84QIi{L)y!EHy!|2 zq@!{1=A|nNj|=iSCLw{^8F&{Hh?eqE|QPyDzO{nrBV| z>akIrUUO|&jb|n_y1*UV31U<;%xkvFX*DyLJVU5(HgN$zb6HijOdL0YjpE3v5!a!r5>TIIVs{jb6rE3%{|ke?s%*<1{SFH zt-!;2hu^Fy#SA;9S)hEd6VBdCEW1|z%i!Hu^`5V~vr8r!w04GBgNX#8gi(I#+1Pst zSzjInL#mxMsuu;zFkQfFH_GOam5b(^Wc({r!Y>$`c|(5cxK)dZfhBF2QM%0Wf$J?p zd83B+lnI?M)e{TCAG_?B zB;Mcbra9HkNu=Z)Pkja1UE_kazcw25*zFSTWq3k2^UGg=ViKb%T2>#v=!uLcM6Ay~8C#9j&>eTWp^L_0#< zamQl0323t4jfWjTy1li(rWvWODJz+4ynw`mvND+Te7q!*M|sl87aDqNY7ULKKE62r zt?Rxq`Km;+EU!Zv@D)|scIhQ&Q43#(vKH77lMvu0#m>wE7aakwbWEOpTxkno!QA8E zy!IMOu9I-KTMJ@fEX{ma%-p{4<+oG1ZMY&|LpaMFv1unSp#yz5b&yNF7Y(42(UHSA zJ#JLMQA^j2>msIVSq4?Eyc(oU*q4px$xkGfC1v%uB_Avs)7dljI2J;HK-m zaLf)=Y~$;V%jH01qwW33er~JbXH-k>G(aOf)qV%`a{2O0Y9`$l{JHGewRvk*)5(R* zDCvE@F)eXu7ggH=3i$F>v`)H`>_-4dYd9>3E2*Xoa|KD47)K0uC!E)d^Ggrfft<(X zz1Y26k_Hsw7h?N$l<;65c3g%?_+m|#@D`VETtb4;3M9vsArD__a0{farvP=(%PPv; zV`*Lm<7SVFnKef|H}L6xtf?Pm4Zj47xBCUVp*JWqq)^V|#mM8*O*H6pe3j^8I|zh6 z!KtMr##pIm^=IFiqTS?~e{Px86^gl21nZW<{3rrZYWGt3#&4s&Hf`w@KP*(LYg8xl z%sxYa|m%unn_2bn!=D)+gEtIr8K6t_eq!9EPyg2RvdHk4g3 zoR=OAA4%I&Da;{+b<(H_3*W$TZTs=?mkd#B=^+qvT=3Dmnu1j^?NxL?1Oi~`W>ZD; z+79EicbO>v28L06xM=!I=Oa{vCrctTPulEUaaKTRc)>5P%)^5)=l#IzNku68cmby) z&Z{g1I>Ar?@cpu!#*iJ`sj%_ws7;VS&uKvm5h_HHoWdbGaR4$v0)6f>3r+q<=(Z z(531?HTy1iyi`rvuD;r5Yt_u zwj+{W_Oj31Cu_Wsd3@s&T-QPqKEbOa!%w2MV()ONk$BpVX7c;u{NbC*X2B<6n$uV3i69fRNEwZ)!NTyP_CHJB>|jY;T1Un>c>R-% z_Z=_Ki!gHZ%9n^zeF#AlpfV4ju_p$y#(Z>G)i3nsZ~^rL%LZv|%_LpzQ7I7L#^WaS zw>%yl%8Hn7^s@dx>o+X)CQShiv`Q^_D@q&rClrz{j>eLPoy6ysCCja0+%XmcPQpGY zpNdUaXintnQ0x`P_MT=qV0nB_A|k-vKy%T`U|VV`++eb3f70|x&LywXv-Oh`e&X4H zeY88seShwr2gDsMqf4hP?4y^2F(W#ocniX{5c52;jvvYQDz(S4yqp&Z@GMBQJJ)O! zGgBCUvacJh_hwJr2hX{EPO3H>9skv5xSiZ%%-GA})J}BHAA%bRLe(`)JrqJ(z*R+) zzUlfuTeqZY!B?+gb3BBZp3D+5pAJ^weh|4O35d{M3+f*(94RQcCf#@|9hJr_B!osZ z)<7EN;*G1{^Gh|V?O9l>9!{)z7_SA#@-LIUpX()scwbHtZp8i^##47h32k}&4QCM?@w;M)-D33@4a{x`Q16Gxe+5FE~X%x)k z5u|UnZG^9#MhO#rGYJQ@g)3&@rd#Awx>po$)RyauNs8<|b8IJ;$^yNYY$gOx;w_T> zRZ{346*8;G(6{4XKf@^YE<^S0aR>$ct;;XM<(2NoCgS^mEL7obIuk zu42nO_d%?8;dpZFZj*S@56gmmWUDV!4uE#lLpvk(wrnx$%SE_dq!=gE7Z2n?%e-GcK4rrQ^V zzNBkn9V~h5Y0@tKU=MyqV1(?5wk3XAaG`yp<7%Evl+W3au^;+w?$U>5g6FAqAew3X z6kl!lao{&k?x?+~NkH`eeYX9R((EwQ52j3xO_Bz+%pWYEI~u~Xa$6r^DF6_OMf`hP z90Kyb;~K?^OH<8wqcpzyOQ!Uw(x&$wUqO&;Mk+DR@Zi%gfQA@eyyob{q4f; zY&x~IGggL8yvfxh?hB8VvFHoTJB1Uu=|Fghfd`1*GslU(x$xTz#4lzzkB7A8_%#hc3_-U}WlU>T97J=|%%-7HP&P{SXA41aZe{`%RMbno9N@$4_#E z@`nRP@wb9v<%L#z0M6>pZevC|QsyDL`Mr|jathn$Yx{`E6F2IHSJ-WkpG z`1CstH^*O8e(<0G3p5`6F;q~OtG!3nQTGuk-$Xy^=3$X9;~CTcH=5)Bqu%+Clh^kx zo*L2thMY(XnnX{}t`zJR0Meq|&!;dKuZ4BM`*Lh;{1?YSB`y*G*a_6XJ6axMrx@+P z4*Nbb6w@6x9E***=yR8a*y=6g+&Kc0cqM{&;h43&1+sh;_Em7{f?J^eC$Z>xKKlx) z3qB8vN48`0Gm#jM)3xC>pRGs$n%h9;&BLh`J8v0?8eMD$3?NpO_cj-wE^SgaGEavc zbWLvrw`Yku7~NsEkBzmw2k{39O{)EII$ZCAbRGR0!AFal#=ALkXUgm6y5J$a$}-cX z=DW72iUU4NROhf7iYSVl-hY7WUTnX;*gpMr*;cFkO`d&%jMvjMv zt2G6KDzxvsLduIChvK3)5(xlqq?qcq(>gM3Qm$(;t z+N$9vZ0rsjwW7Y9zx|nB)P@L3>4D{CvqQInxRLLjE!fXbC^@DN`VpquH*aPMde*P zCpY~HFx!zYT(F$tctb+t;3-(&z9CVx_jDRMep(EGKvhPdVC~Tcf8F`h#C&ZXULEar zy4hO--Ft0lLyFf>8s<<4h!@&Moromf8H8I^8hZka8DaFB6mdKh@w-4Ap;2cSSebS<_VjbL5NRB9n621|w zFMPyo$4t8S&$-S>h%Oo3(9U`HZm+v9zg{+C<5GVyzD@fRTu&oI7G=cEtW%QamXp70 zGjRxj;k308&C^Y(03+j(g7RLS^M3UVoU7knN|Z%ER(}pqegHDFCene4ba(xfg8auV z%3ttXyvc)jLrLScpuyte#{xllmefsJ?0l9i06nB$TmxV}ovPLpOBp^<$-`5W{%@92 z>d*)w9vu?fS-LCcyTB?PgUg~my7r07$Q%Vy!jRx-r3bW-?CSZ1Dpy)Z8 zt+3vJ$sCi^#h52GZYTSdwWAFnY|$=62uTl^yt#G0)Q0uRachwoL`D27b>Lr^i0PY; zr;i~y?ZAq97y+sO0#B_oTSy5Y>Sk|&kQ*VKKxOR$n?}hD9>Rr>Zi^O2$cG^v7FNOjSPePb&v+BiK1T(t&Zi)C`UE` zVzpk`hf;~R4Y$3MW-gRPSQ-1T4Xb$ig0O08rabkRAa-@&civAwM5i&;z$7Rsb%B3% z*OdI;IpCd)PS}q)As6n>Y%Gc{%bo~U7NJW>QS)N`j#i24*)~7-7<+_zie_gC3a_WR4V$*u>NKVq0;aTF&xa$_el$^W(%>t*@6O zD)vZ76$;#%fn*=E`%m%AZD`lRbH2?R>hY}L?MSCO7S`c97;|IhSN#R zhjNu5hHG{lM`+qC%n7yIn_3)^O0l+=kgM*K2Xh0JeT(WpBrXA(k)7~`_hjZ8Q5#b@ zs;Spp+h#3kUJtxImeS66@-t~yXO+lVzw`B~&Iu;DdHh~lCq4RZ#%|)ag*v4h?LL)e0f9`p~EZD2zo>7y;2myYhNQ4ky!!0}6KftKw&n@MflJP8d&%5I-5=zcK zjAwHdBvZ?cc0p0Pcj~(TtjwJCqi8F~v)ca>oWtD^57R8Y-Y=LNEgan>wo@dR98Ycv zZ`h0pt{^x6Vb})_W;E<&w&~jwC(El*i?IQ4fsp6$2{JFVKcrWzd zAqsNyn|=+J6}3?glTt8F4Is~Ja;8~e9;&iS={d(zkmLO|>2C^?W4Jy!6ePhSa_9}` zLK{VTE6cxCb)3X-AI7!w@^e~1go#eAzo(CL15jylbJfa%TjZilFp*@IOGlTLR^5R* zs-%(*ZgYG68RId-{!_GldRoVIor2*nJ(zxY4{cv!B{W?h!*X-0Nzd*tE1{Jywrq3T z>rrJXzv7uXn?Y--0#0t@27_0?=M$48LHk~iyyh(hymx}+PW&FI> z(!0#|O-;!_qAsW01%H|2cVS7){IJ?l=P!216(1QN4qJWUdn9&ZO}!Q)wj&PiBDp2V zjuST7_>x&BdM^67uF#VLOTQzrZ#d_w(Y*Tur%$2wz~izspq&4~hr2I)Rowh(tqovc zz1(O3lTRNQ_lmorH3;4hIEs3P#m>&y;ZV34>ct`2UIWTX)m{!I`{DE;IwD*iIYTXd z4aaA$qqO!bB#!<2qKh2e{6t_P=d~IFx`>VVhFgC$@yC^kGH>Z1l}jU*u(BEBqO@pN z%%pzFNIV$t>#YxNe*GH1Ydj(bYtD5G>yOpWRzEF>~d{Mz&bSSOh9wnU@{V34CZGM`t{d1o+LMhzyx zu0+?~CqGj2dlmLYmj>dm%}J*a`3|f^(eS{TAAYniwID9@4CzBbGV8^)uO*^F5}$N$S)Vu02HzKw}Z8Im4E^v6EwTNWfJnm#>>|y#bVq4=$EDf zHRC5I!bRi3QecTqlkam;B+RCMyjvPhP%?uUGxVcfmy-I{X0(@JH7T47lNyuGXcHqX zQW{|}(@~hQtUlw|X=Q!@BtYd^Nh~MHt0FU*z93YkT+XnlD&7?9l;2m1=(2=`ZwY3yuACp zaOcnWew?v{BKi$!mCS$3)D}UEc@A)%q*&$S+bcC90iQ7k;;=9$!m*Q8nD?tLT)+y! z0=9`b=`i1ArrxKu6{*ytEbXs2qCWvlQwai;UomyPuAN4qs=6MPcGm>$7*ga;Vk>~M z?pyb*MG@!^=FH`I?j;CGE}-!0Q0%yM(fIlZXC>rai8krNI*FOP7$nkbxcTVUZNYji z^c&hBVbga_#moIwp^$r2a(BjWhjm+*IF8J&C^B&`veH`Lz(TPyldus zP+{CbWy70eSpTQ{yb0};u89xik1|d zz|wU0c4P;g3p1)E35K%by2yhc$MxV+;$!_rDzR;$enMfWa>2ejeO2uSB1eU*Ax;)? z7Z?VNgUYtkSHfxvOs`;J+{z*23atXlpy|)32m3kQE>Q7;1OU;oML;(d<@jradN==c zdWuqU=E&mhAoo&zsnY!v*7FbkJP$kZ9#zNU5dtLA8gcE=?3T5dp-AfJ((H=1@j--`53dTRHu2hU4mrLgmZ2?Bcrx~+KU^%W*^8Fgs~gy}_KEOFtr5PN8S#Xr6q7L?i6~}1_c!!8bCoI!d(PG3 zm&d%h?>(~eMyFom2Dgb$jsr+nS-w#a#epxOe`n{`NfLgTu>C3Sx?&=u%7!2&X zHl9QppjvdAWqN2o@glr0P|)nTConl8WUnq!-0nD2Qxx&|)+WBK_4><^4NnT8gvH|OI1b_n z->LEZ`Dr5o?7&K1$4!*ED@wYv?V{Nd{8n4Ejy%dw@jhR)QZ{1Qmf$TP1nkR4us# zgb>A-yF;pIB~MpX1nq*5#+Uo-DW_aixOJEeNKUwMlm190?xxqRh95V5g58(n z+ZZW|Tq6JBBGrdnV9>fc8|rknooKqu_)D;1bBbcHmUV7f1P9CX`s|}0By{y(P7H5W zq!s-+$2~H%c_EkH`fi}%>?FEPnE#SauYIsPZGf(P6+`Y2aw-%Ez=I6s{-`4yTNbPf z&cT9W5&KDD{?xb<}!*_kmh`+n5f~@ z`!C=?@t-g%X$Q6IXDYBSpVXPEIDvDzwpdocMLm@{ZL$N1g#`SenU8T=QOYFIRo_V&z~ zk3k{@zAo)o{K*WN^SCn&*NaYZ*`++)946}6{V9*yBhPQc0vsHY1HG9iE5Q^Y>aX#$ z7KXekFElLs92JCL$*?p=1H_bF$1XRAM@^?JfV2GT{$fFsP}7R-u_p%*N9c#DDBU}+ z0m^3mA1-ZzWo{cUe^$#9hf2P3yQu4Bx9T`IAcPkI>T1M{kNwQwVUjJPK(~<}xTzIo zw_m!Of@v^`GbTnwRn5HPQad}4TuhO%67#!fSIy=-edjkh{y1O|RRsK7Cw=R= z9~WFFA8q~?SZxmfTHsLmfT#(yiGi)n?ypV z``QB-nAO4fRs9x=#Of7lV%4mD@cwT=B;*I4#!mpJ61O1{lz*VRqY7cEfy5)T&Hh6y zj-2sC+&1({-kHz_qeqAxz;4yA=zGi1KP8v+p?#1`+42|7;j|>*OJ7koje08{Q?Iie z#-^)P(fIK@5wZ~gS)1AcFsJY8dU1d_zH~UxU7CYlH+IEAy=DI+_JN<~CLikaExyyh zveq<$di~B(f#o<8p!y+ihX_EXQ=xtGG+;tq1A8{9)E{_OG5iM&s#9u)P5{_(am$>- z-}{^BlF5PLKf!Os-chDxD^M~g_@wI6hmd_|bRCxJiO(lPAV7CceTleO!GJK(zdM-B z;is1KHgant+z@Ci^LKY2Y4OS#+Wn(RU(by>$8NhQ@Tf0mG!RDpV{s*irmy(W{P#8C zL52`hpA+(*)vO9x-_Vpp?VXY0v%vVF+ zKR(zM;=9(y<8oZFrA-eSsjesHFWF8c0Xl=0W$qHPX)n+a2V1())k=xsY<@)Kh(#I9 zM;r3zU8dE}UqN1xqLR*cMo%WRzFRNNk*kT`&Jld?7AUG!AMfC~czwSwI7aS98N%il zHkY7!eA{LQGCp#P z+k-q!Y(702MBq8jYkY>mPWQx5G49mQ0`AXNY2OH8AW1c8NS5#$CWd zn>Hg;vsnLquzwTnvuN{k!vGbgq<(-#N^PP92=`>`*3pzcP~N$8(~K`BGcuH1ML75q zc)EZ6Jl#}N?b7)%QU~chE-s!# zr%oB-=NF22pE`oD7%~kc@Q?{Q!^sGsdI+5*dg2|6PEHF1E=_li&7|z6^1TVYDf~na zZmAwy(aF)eb4lxRuOj4<$j@+U=NvFaQ>p>QAe5yw#B=rAp#|eszrdU866n`RbZ)H# zQ`G%aeJO0*Gjx5*EOH2ZajeLk+V@!m*Ui4mVw`DoXnu5cb;|jarYg;%YjwrA+ES7? zrVv~VD2DvZVKsx@<~zvzUAO+EoDYUARhx7lDOI8Suk;sz+>l`D@BJkZm5EV}Oy@In z4=RvVS<^hv0fxHiezkG1ymHjUhN7#uNQnxDE|ISLZ)`)+mLgO+67J~2Zkn8=%BSj|@GCJj}Rrx%*CybS%qmxAlXK=tG2{b68cZ@fabtQFVlEs)tt)spB5& zwH&?No0oLYLYuC%w`BuO7o@B;l-;0jS3}qxJ*<@Xt|uhuf+t(a#UoHEq6C(3CblIhhXz)mdl?Sj<Nv$|;uD@~S~PJUG_bC{QT9Sb{@A`*cTL8V9`Q`M zPf9ZX-NzLy{1d4Qa=Ow1_d!p>>yYm^l|!T@F$V2~zjli3Udkwq$SS$|FXib5(<58c zy-5e-Bu&Ape}YOHwNl+Tk&799FFWZPR-mdnDaC+VNn>Zi&G`;Sv|dI$f)eP(BCb4L zd&;{V1GbNIu7q5u6p&J&VFtrtcO80GSk{&_xBZ!Z;%vgks;+bMWMb`-;g%sN1+uXIM{zYmxDsgvvnDMLgdh z8onosNW3ftSN3*+dciJBukIswMH#v!HlD(~=BioTaDn;IZJ>E92H2YVjEMBrgNF^l z5jv=&Tp~$c^hT8J;5rx6*pY5toL+6ehR8W{*$akpj>N7Q?R_Ek*3)%{3bn73C|P+K z~2&`ZmY5V%UiJ%$d(Dljf2yulIG) zeD~XX2ZjbJt$4*|ue;ZPyo_UOA6Ou`wM0`Yf8i@i9@$pr3#$UF?Zf1IGb;~B(C79P7lldaWCGb?C>foB-WH_w=`A`<$f|6tv?e5Svgw!z zY8VRguAFy)x3?b1Y?9G1LJ#MZhrU2j4T9_G)>c_)R?DN3W}SK=m7SYTX9sHhRhHB{ z-zL8Va(f1oT}(t;XNIqg8zT(^UR?e_8Ty@gO~a%=MgI3&Yzg``J(r>k+B7PPf)oWS zV-i^tY=IlPM*Qb}%E`WnqjtaNk2}sx-;+tlnp$^X?60q44o~r?h?b+~kYd;#q5d!y z+ChVC7?fFpXrlEqUucDbE|@iD5U}}MMcp`uAhJiWrFy!CRxjSZVszE~>nVu8u2Xid zY=hH}fzSMCV!i|>3r0&2Y=GfJod9ND7U95p91Dv%>EB2EEe9)VnzI(&XK(vR%p-Q;C@J<@RrwM1kC=34DA?hezM4#Oo|4%6vJL`&MQ$u&v3Q896_i zb-rzAW!d}1v)(a2Q6gizMIS7>oj<)IXxu?h`-g1$X9;Vy#0r+agd5*!GNzyQkE3TO z+d>9c=0pI5McI4C;%kAViFVKx(c=W>52UZC(h&nkJ{ZPT`fmyYMkP4Y*P zST>#fR0!k_z~Wyw2wvu1^prY$@!fRh{%l>p8{PTq$HItTtH`yX`RIY%S6C8~RhVfk zcVei>8zu+*rR(%e^<4A*6OaT)*O8^e+>4ri|#VK!q!scNo1UnRlUopHJwyunT6 z&XZl(32HC!OvB08wQAjV^Q!81FOnwYi!?jcHyr6dNM!O!;2~p<~}zca3w?>_=4ggcPIU&$oQ|YKvN3I0^dU@oI~32 zk{_GPtaMB({S2d5n`GR>vo<#wIs*W$MR=4{G~}X9}|-WC=-cN)wOL z7vWly#d(!p@Ozxu8ZSZcx+~jxm3<|R_rBpSjY7T9N!~8wiunJ>-dl%7-FExpVxY7N z5|WD2-3^L#H%JT!NDj>m2m>M_N_R*~cXtgT-8~H50yA{y@5{5_z4vp@^`8CibFSZ? z`@eyi;rg=fd-Z31)?7229FA6FeWf*gGnY?j(`}cDPzEFgO8sMyFp zHH}XIsqGkmhfxkIOUMaW%Sd04Q%^>-5qy|bjoL{WFFn#{I3tnqt)K8#!fzDe(G0 zH(^Gj&}P)P`?AK}uO%3nv`2@wPq5aJ5n%y54Q9emg4=<>08_JpMCj=)P2 znboB(gXjN31^xy*;Rl|?d3j})IK*PBtaom|JNuw=x-sMrqeU{rspW~8K*-wCu&MbH zE7a2CXw!~WC3n`?0OEH*Who?k_mwIi@O5T#>re0Qz9u|Xr2H|PYeS${VHvJ%MDBj> zAh=ZGi#=nwdq%=_m@&E61?U!cP26dU`xjoyuGf0YdfAxq#+2=+CyJTzstehZ zgk2Xnv&>bVeUiP)E3;4xzaHr%qrI>6lP99l!=<>kdB_5|RTg&1b;3yhdCsxtG^T5&Rxj~P?sz;M>%ZjK`G>AZyZ}`28P_7=fB(k*?UO#(uK=yU zRwTmw@h_T#|NQh)!vOhX$M0hOi$d2wS8Ry~Pg7g0tijT~%4& zX#5mUzy7~H`(5Cg72YNEd-L+Y25bKx-etHI-CH`(xBu;{^dAonyioPIpR=p&^P`pK zU)b2c8NI(tk?2-*F9^l{YZ{Zk*{Q#sI>yr+iCg>G{LUoopP%dBl<}W_0Q**SVOS}D zf1>~XYfT#9&T1K_amfB(ANgM*^t%;Z*H@K)`e^_1mrS>JmWe}+^ws|&lK+#a{`cSe zyuG#9W&lg*-;eXZ{rcJMoqZO^fcy0SO=bQ^-Tz13|F7@u|ET-_sQbS%>HoxT{*QP6 z|INEgaNZ|oegA65n?r>d$Yv)DEx=YYHNOh#dYIY;+U@^|@c&QemHP5FI+l;QO?^{5 z&*P6WYI;K`kq@Y^uTE~vIuwfkTmaPbI-8UE+V6>iO%}ZOVcgYklqtn1@@huP&%+zy1yy&M~{| zybfcS_q=Gt|8M~To7b+!unN`G8YoMHAkhZmUJ)IT&?S6E5Dbd=EOWO~FDv)g^l(Ig31#VW7~$;5!{YAaMulkdx>et8s?kAJn)TfDNYvbGh+Z!;o0p}U+SrIFJQ z)*cQ&-u<=Xbw-~u5DUmAQ*z>CeawEScfrbws6FM=!(xLos?t5OZ8ss$WnYQc$8x># zIxvpDr4Ems!RQ_DQ#J1op+^>uz7V-oiu<~uvA#|lUQzj=&GPN@z3h~gB&hgMcrgfS zGiB_1r0`J9m#yJo`@-qGOKrc{`|1Vlpn#qbx<(T38(qSrg2qkjdZHX(PleK5zxzyT z*KD}2Z$l;E+!F*9V7Rtp^FNzaWm21xU1h-}a zT|UQvyxilDMl{Oi>b5kM?~p~jj%W@eREhT3@s{lb&()WkMQdrZ2_;{Jki8tG^3Geh zVD2!Wd5EgkX>k_oQocR4>={>hVg}J?6FQ|DZOHZrr!b4xo4ex@sZ~8=f!Eji@2j?1 zNEyCujBarV?w~aCeI6$Tz5LvAZCAfT-SFId_Sp!mx~N5z9k|s@2@--^LkV=sG2!c) z!ILxG=4rO`kovU6u?fEXKaV`lhoEqZWe?JI-}ti`o?HzJ4Va&q& z!N&H7yU`Q$tWYCE#WWF}V}U@?2@uu%1`bvYp|R-pKwg$@O!@6GecIeu&(xWJ^L#p! zB!ym-09l>&_|snX{IK=q<&V5QXWJ!C81l+!hSi_;9|t5j^3987>OPawwNaTBH{_M7 z2oRU*p>lRVw#e3?W-9$}*V(!ADU&;<;R>DSG*~n+5@7JvWIB5R1OKaC>3Kr-TZPp{ z4MNLu02Sjid1xG4n8aNSP!d1t-LTcVXr)RQ-2X^KqAoXS_LjHX3Rw|;TkgdArCSG@ zbzFz-?=zS4e8rIoR@4+Sc)y8sVruCBDJo-Ef5p}hw@;x377Bb*F#L9Twhp@Mx20@- zl6!TW%hant`2Aqa?C@m%hKAg};a8&5NEjk#a;>FN^mv=mzDcW+2D8Ub^;5p`a!-X& zn`*$cSZ}+E_!Ubb6rp#0S#EtYN>5#{QUz8mj#_wjRXFJOjPj9}?A^*RiP}w7riO9P zHHVE@i*V}R!aED2aHmy|jJiRGNl&)-Ws<4`V!8PudIJsbx5tihd6tAxdYf=CxQ_vX7~o zXm3PXT3DHtRP```M^zgwRH_5bZW2T63qjmPngCO|o3)KOP;{D$_ zmrV--UG!Pvw9a#G`a9qJE)^y3jw?0mtoP4hnXGKkZ<)I13!@YQE)dX?^Ubbc?d^!t$MQwDAyr}1AvhT+^Q?Y2I`KKN2RMjxX&+k3U=L>4n;F+^5dY+p7t z5vH!=UY6)SPAgBJ?a34@;2zha9s`V*msh-hbH6Nub88PoXnG!RbbOdc02=)<5g; zO8Y8DHml13B1W$~p^bhi6&>-2C^WZZBRM)IW*}Jj$uRckRiP0VE0n7L=od)Qn8Iq4N0f* zMlI9&DF_@(4_m=bk*5`FHLVZNWUGKLyP{u&y$;q3olJINm63%_7%Kd1)fYUqIBKhG zG#purj}$oj$t}0v*3hdy`O`k8Vm>pxWrU0Vw`13ce7?<sd-*HjWFRLGq0pHY zUQw{F)m2~CqI8e5i;z%y45a!>s)O4WF<9|V%y*)u;i$tjcuwJBNJxLgRYf9f1j(jy z-i0L=E$#WW6GE$CXzAU#g_NXGcwSCbEV}VNB<#GP9uVRc;~H{KeX;P<$V&x7ebE;= z?mDsmdDNi)<)}xhTc;E38)-wu&ni5Ro<%iiN21?Xs(Sa|Ivy8Xwdl%t_C}_1qXgO% z?paK>+1pqlN_;B7vrLh~Y%wXGn|d3gJBa4Zq2-5pNygiw0oHD(gWg4K_HeN_ zJv}r~B8!-|gDZ3xxi)&$rzo*1dHGPRn> zP+J;yGl7d=n>BsjY4rnj4Z<&(Zpot{HYc-?@MxWD8N{bK*vRL}G^xOqjs*MB<1KJ` zw98bk>0I@(($CpgP&})BjwjA&@!%p8oRuAZv*+J%xo_)k(xFhc6u>OsiI2`u+KM?x zLN3xn#kCTtbj}WOqQzq(F^{(7C)tuRl#`UeHtv%rN-*7~t8EWG??zX1IP4gh!MZ>> zyD$9e92;)KIdU{+;Flfk=bL@CGhM)M1#p>N?0DZW?=*_*%z3TDj!u##ObD;5!K3O| zU<>_zPs(TO7AYQT)Ya!eUk7V{pI0X&#D!#aqqH+6f?#bxBCKHP%J<>*}j(09l``I^H@(_i&wYSh+ueI|H0)S4c@D8We4h7l*c zBN=}^=DZ(|eBlC6f<=GKv$Gse%N@(F4z7A1&`Q2Phucntc~8HCg1-iy=P@x<-ODI` zaUwd>Fd~B~c7EytgI^RGx}HhS3QJLz`-@y{+22IW$xx;Be%uXrRZ|p-AG8q)jdxfe z`Eu<`pfVb|gqS)CIpfPKlB)GwXpwP!esgv5(rPL#-gdIA);4`pKchCE(Lz6j$6J;X zeX^0A(rwSwPtF*R;ZA7cU*=BiSABD}uB3@6f{7_6+KMn*F6%CW0y9^<+n+oKJ67 zA|?Tsk>+!H;nQ-EhvY~2^*1lGSV`gQeyygNCiD`7*Uq7suWPId*Hw4$UIohj;Cmiz zd5=gtM`hb*n=YHt6ffly=G*>LZvG}$gLu)3jnrokK}GLk$kgPzX-kaOIcFipO4rZ*x(vzBrux#Ms%Y|Uq{cVyQH0zdXFg+ z4}w*0Vpi>4#FaBkeXe}uzsXQbdX{yC4>mV^9ZDC^;N8pUn^ye8GeC@#7z(c2=om@U zBixwDmGtSeS#msSBJ)CJc=!tVX^CO{3*%8i;7cr5W!mJYw$Jq3*kq z^*UQ65%v03l?y~>Kfp=&#y#RG*XG?PUD=@`#~GffE1ShlXWB=>{3I2fg<79pq#zB@ z5ahc~i<;mBJufsQBdcRP4Vj(+;mC(2$S|&o=BZfMgF)#j2R|dK*Yh1Ks5Pk_z@kRw z!G{RbYDAMV=e}7?L%?plP-fiTAj7tv`Nsq`q-)ejSE%^aD@*!$w*&CmQebffYN*1I zb}DmAxzyDtO1yo|m&Ns$3EoY9{WX5)Fx#ZwYFmkZ`>|Tu3f`DMsW8m2Tac7Ix42`E zvuLKkDpE6^jiRW+gTrlwvyY4=%wBUAm+XTxZ&RKB^R3RiR|1i*gM-P(^+IDKV${s} za_xBcew??@^e$UdJjyD4-nlvCb-`fcKFHyJXX??~knc%o!G(M{WqHRSLS-xVWkdVe z;wgTI#cZ_JVj*Jx{j>eNoAGgW|1C&kiAMJECn{jxHY0qNy(k2d6?e>66M+c4GMcQ6Jt=vkw39hH?57wlMFje22N7qBg@dr2wnQfZUYGeewC| zP@o`8p_fkcCV?<(j8@4%#bqq7@+b7j#Ye1Soi&ql$^<+|T~O}F{{ z9wCW=XYGv8ney4eV0({i2U`@Tvf$Z$GM7WOHMNzUNsu>lcA{5RGVa$UO+-QM+H6qDZmUO-{_1Ds#Py) zILDk0XzExQlB1`h?>yWw+{GtQY*-jbCz#89cK3Etv(T9M+8C7im><&jAf8CkmVac) z4wY6ubrVN;pHN2))AyB*p`+t)J@&U{TjqByaNdd2?1qbuzPrUcs@MwY?Dg+=tVzAF zq{3+Ng=Limtz^wQsgB#OU)qC&?cO37MQgc_lBZn+ICAFdcS_$+aoRf4^5()e+Sldj zdEdmC$#>tVD0)kWY9kday9h z%+}fb0!A#0ha8`Kj0?n{EOwe?ABqwg52~G6r8Gj=-QXKB!8pl&)bFCP#>V+oezHu4 zro4)?{LZC1LL@6!s;1u4$@3W??S zY2a$yBPZmbRTkG_u9xagN5I^pHd7?97 zZoNoxF1*d8RTR1H(}^F*8ml|w&@E-r8-kvef+e38M7f<=`V!tdxnaknjeKZmHdJx_ zcylDupOU^YN?mtmd)CCe_qFfaZuB9y%mevJ6KP#p#|g4F+UaIzX~UPx^fPJVwb#gA zv(W9?)|B#8fXV7`dP9w>TKCxxdC&00CGKiuv@BVD zSDEXA(k(;#{yX=iv&)JGpDuOtJEsX+bV~V>i&>#ngE7TB@my*u@}@1+2tPbxRw{ls z30v{mB9fE&YX6`{RBH)gOJnHa$?NI+K4ok}C*Wx>lB@g6jLXIKt)J>opJMwMLZ|uN zb{AlgM_&RKd|XXbem8L~%QY?#6i3!PqZR*lFaW=~tVTgq*)#tf=RfKqx`3QM&MGxx~}Y-vpkQ5QznP zFRV_uY+ZyluDNou;(xNE-dn%NFPG)>b|W(uL_3)4_e1StX0&#Mm^E&DOnGqQhm~%J z>mR-OVrI_?y(5*|yZY#g*x(Su%H2`u3IHi!mWm*}k1b)MVNtGhC~&Eq1jU6j)v&*& zHp|87)$>k}&5E`mIUB0cvZj8`Vqal7{4UhTWJ%Qaagnfn%Htq&|9OkrhrVjoDD!K+sr2_i+J|Ryg4z)%KDW*QQZx+N%r#0 zyQ*7wj=kMqy_GY1XVv-JrvG$c4Xi6fZ`lH+qeE#&b6QZ3a&iIefZU)jaqp)N@ynR^ zM#1{yQ|$iKR(qUQnzLv{PBb_w(LlfBH|p?+*FxR;)@u;&?W#eWDM44&>K$2u=AU=# z<2J*tL%_-k-y7KCt`+W5t`(LnDk6+o zhz~?nrMJeRVO~i*_UiHpnu=Ex`0P72-)@~o?S;Bq+dGt0dZ#t0^yfoici7<`{ioPx zghUK-CTR=<#;!w-^g&(Br;lf1)dE;`9gX|2nG0wDNHgkIEjR)?j>4JFoWfqU~EfG5dd)Xh z^k81Ev1+$(yC;+m4H7nUq`0r%cCT;Q!^i7wOw2KG-Hbob%7;IwM^X&0le%7RRwLPj z8W3_g^+v`r3@BX-Tu4LxTSgZtG>ggoy_^B{C>q-t+?%YaN^&NKR7y`vQ2&QEyh)w< zl0tfFuS3UFhZ-NG)%x(mtS(?kI_(+DC&ls`-rY8*=;4c3yW=f9>ZFu#8gE<>cR)(Yhq6?0t@@?L)@={KM0GoBU|Y= zlGekjtEWCO_dj@k@OZ)`_0PeK1_92#OcW#UOs zqwj`ykjCJoT)PA}n619XKote0Uz7Zsvv>>8qouEWIAD^^A7|F>oIL^%l)z~- ztYQg=rNfn?2!KPUdVOiEoP4h^zL(I;#9Uked90h-i;B=XnWPx+1?*kTkDJALuPzD{XUVFsUeJ=mKZW=LNPE* zX#!2M>j|_jzV4FyQ7hgxQRnE1`st<0bPMh|VzWQa$nQ5EtIf8UE3F<8kSc8}*QcM) z6e`s?w5g{+?)j^KFLWWb%eRRH=tVfKu`jMMlxFsIfU=7d<*=Xd@4A( zSm%=5T4@9wyJ$>VPb)_;&DZQ-3zPcML`l{>^zM84q>EQK=ko)cs@x(clgW8iU3_wu zWz& zkLR2Rp8txvi$8v{Mh$64S6y76^U0k7*~gGObhb+3V>0(v?Ux@)^x(gbUI)?_Dj4KEe8HyqCPD5+S#Y0nC>+f7I5YEGvwhPjUP{8+Y?4zI!y?Y zSe@a!lM8<=%!B;saKoWn3KG;rBi+0GEE1*{CW}ZSiRMr>i_0HI=3{LM?~5(nb*FNt z#>TLRbJ98)+{JnJgxd5Q9wm#wGaB2WP@9o7-)-^qBG}Tf$314bnSgL=aMQ^=ET2Ny ztj4I}TPewDuw&7K6n0#sP>|_NcGMgVh}hKg+@k(j^v;teh4Hj%=CQj(rb7h3b+Mv| zR8xHD&zVli_a+775(jfs8vH^ErFs zH9zE1@8NGh3z}c}{U~U1;r^Ym9qRl&i*jR|-Dakn|JKKKW(Asju5jTTe{m6w3}I1ne%l@- zdEPT$rFd5`I}}-Qf1ew7E$2ovfp`{r!s^pX&iApogKn^qgS+d!+JSalJ(&|~_)BoS z@zPqMJoq*_^8I{}_7N*bQE^#90?&4hZz?VPdduT*4X&sY* zEZNtiN=?RNZIf??tt3BWs~q%jQvTPgYV$^7=97mC(Pd~PJJzbZ=ZT{zeT*7ZBnCwodOGZL** zCGd%6LsGkW?S{n$LriWE%cBR0UqSu)(4O-8*<(W|>(?-4#@1E4WEYR^8bal6@NBAi zG9@oV+`4iMc_p*PLd;WaC+q2jhQaahh}vyYd~8(CZ2$h;vNZZ#tj&C`hw;O`o8$VO z15_?$?mmjfFg<85-?cso$7uRY#dRNhZkk{Ol?=W9L(^dGr)k%jbljCI*W8Si;in#j zjiR9pvunJ-AIL3~z2~tjgDxOM*i@VHXKG~P|HLIDEXE9;-aMRkPLXtySM4eHxEb?z z#`{@;H={26_?v-ay;9%>kjzC+=Xq&Sy`9rc+M;OUTq~9ebfO)s<#^QjeRPnsnCt4N z04?0EN-L70Cq`YhI~RixtsFOMc%q1RdEY+jMP#Tk$woSa84HzzZ*S%n1pmg1%552H zX4@pKcS5a@Hg6B}l_K7U)#rRR z*=>G8rF8Y%=sw*??ymxUCtG{C-A1u&_YwAcQ z;NE=Ea?wu=WWy`d>hVgYtSa*y2kJYH_>`y?noi0J((2Ui-FMw1ZV&tb+2$kPUa9L5~bg zwc`F2tBj;4EId;&PF06Y&>`FKxaaEXWa-m9!tt#CDSHaIL;-f!1e1+2?0;eb%M2bHX4l z3GH~E!dpMX*xpRDElnk~5fxqw45)H0d=|O~Q}$?B0%kV}<`RIYGDzM7aS(kiW=2l9 zT>}a#j`Q3DdYmr{k3FwpAkDf#PB2|Mm9(u|W0TA9_cuhvFRri%SBSFJFn>9TwgtM!ze zhD{B?UGO_e=5m!&&ga+!-6WLxLNLmflp^WYX(pfho*FvO2KxjB?%nlBoI|t~7kl2R zHrQ%)8lQg@$$mDZL){+353rbB(akx@%r!h%@!}@_QZeT~340u!7A(q!AFq(q7r7pz zELfZQ>((ncvh+Cz@y2ugXflrw+|TjxncYPcuupN? zTn1fj7PSFwD~KTW310KOjytOkT!dAOF1TlW;PWpV51ZcZO0XjC#+LKIZyh+%W%*x@ zfclC*#dK7U^t0a2T>Pm)9pMkHO+a_moXr0)8V-)XfABU=rmf)!*nVZY`_MVuEmir` zBc#nAMdOBMzCXp1op;di*RdugkeD|d6WOX z*37FPqRn|BSi!jC*AGvyTc7ZjSc|ou&^1^}DDaZL5gFh)h4Jl+@=?`ywxpxb`QC3_5n4=Kib!jW5&@=iqO6nHOzwYy$(|rZ5q# zk2>)x$RCcPka*9c@`rT4;Su3;Jh5ELg79=DyemUfJDo0JIO^B-AUOBF=&M!*Xk@%I zwYtx*s~U_?BL= zuqN~npAZV67nf%oIwsk8s`V`_Tgm7?4nvVNSq5PbCQ71|uIIGp##&1{j7{}33u9Q%Hsw2kHa4Z@)q5#o6YT%peB-z3@Uc{LcKa*Dz{fQ zUt5OLi4l2vypoNki+a;cVioAq$8a0p?*}s0pGwvK;EE&ieOcZec^bSaVElQWfk&m5 z+56=TgrsM#qwwOxrqU!nFH)6ZaLj&8`lfvxx+`7?<|IT|rYvn*F))07MkmtNz6}Cw6HF~;&mD{>BgPP);h_Y!n*n#9r*{rnNLp1v`2W4cWpvdO!rI? zr+p6{rYqT4bgs}EVu#~0rK@;d;;I5cT%^6*h)ipi`=)}6e4?8PITpI#ZF-2K)KImA z(_%B7Zn>r8^a^!ZoHJFuCb>s7gB2CMw(kwCxC81t+$!~4J_|7oTmp2VEpPjGKggz( znNFONU5p$0U)grF<+U1*^V zY=Tnk++M6R%aH7IY+x?i)SwJthu}PYSTBl-mczsjs#zl?sN{ckw!FPlE<+{|I?q-m8 zYMuPV2;Ye+tb0NuRv%OD;ZGh94>Pt)47Z;Kf(pDU@uH%K9C;uS+QBKXxpp5>M5sQR z;G}dbOzk;`-L!D?=h&%AtR2q9r^%no+pGx00D8R;)nn-K5vQ-p*rH5L>(7oVQ3KOS z@&fCQltyw)Ock|ZTHGN!>GVjmvEv4e@{%(AChjk8nik!}U(+#>VDDS2GcJ{m^RPBp z|M)k4E0qQ6M;b#6ozcSK5bfwv;xJ(zE7D~msMVeOIDIIvqGmdPTs1j>Tend)pmaGX zX*Qy-Fp@;&w@K^kZW=G?MJaG0$awr%2TZBIJ14?HXe=99(|s4gOKAOAyVZOIdc89O zrcOG}Sw~R{DaP#ypPokga@~K?!zU*AT#Ed+uzdLBF-g73m-I8q=g({m9=OzN5cn;< zS{6>~iSLY9dgis>gFipz*c>7w*HM7*b8LcGl4wU>Vy!X#z>X9%N1}D(@v}?zfsJ3xpuO8Ttlx_ z*YF9?Se!O~dF?8C*o(FfvM%n$S(ar~hbS~Sc85;@Sa;#v_Glx4&KF;3`<3C~+aIjf z!vScv6?qN%d57g!7tu{sSA}+DMD-2{B@+19y*Z1`Q#jJCJ=O@k-?%-L6I*|Fv(Kr( znPgqkbCU7p!75~95dQ6RfIjZgQfQK8T3=@h3zdEI+cJh_NzAU7kuPzrTp3WOmKU~u z{46YGIy7yeB`q)MBeMaX+|q_gNF(Vu!AbmeW)Gh`MriGux&*#^*t{n99(-*3FhFkq zV6qEpVhWM{>ftcSXEkHYA9!iGo}{5H+m<1Rq{$W9MhpgPPgH^sT5+y0NZg;!BhTm)bzr?b zbk~pS3E~CWw%2M4dA@K$e$x*ai}Z}ylX8Dsf`dKR?$vC)CyB|wK7LJoj!H-DV-j1? zlL^DpASX)a6AfG9yI)g_+P6F9FipRGH1X1X(d)VXv!ld5woQG@5oGnlValAZu;?p{ z#JQvE;NAd2_eGDfaz}-S?r$n{&wo^Kp_JQyWwW}A{T6@4FpOed^nXP$%_@_}6ETrb znvgtn+LX?8V&|C)DVarnFltws+$(+$gMR;H32n@U#};`S#|PsfG$@6(#K%h;Likwq z9Af)`5lQqOHrK-MyzJc2$ZG6xQ;cYdOJ)>4#XPD7dsfMH7*3u~A=d5tfl-+RT5oD> zdj<-t4wsP9_uRG)xY>-CkME8KH9GOl3@cwuc-gBswA@T*&)C2xCJ_(Xzv97FBLcYu zutyQT9d;we7)|{09&6vkphQ|w*-`zaQ@9YRFB-oB#2J33sMRSlBX6Zx!9apZ^edYCNLLX{YncJ(W5|4 zD@X6s;Z9#3Nh6is>h`Y~^Kk;{wsEGvmS%)bOPYifyU48wArAMLX__Ai3-|@^kq|=V zn90yT;5D-%3ss`0kCPeplWx$L#B&%=O)z9M?9!tgVUE9xqNLRx8C_uax02{ z2CcMw2+sd(2i=U+rlX~38qG~~XTg!tfjEr*#@O&1cpFg~I*qqusj*!}5y0|@n`qQ< zKfVz0d|TjEyWr(OGU~UZkXna@KeEPtM<}pP*cP%`1ywyx_U1Hd;lE6(JxTJ?6Hcb1 z$h_9(&t_TtE1GZxN8ZK~B2OM^L#~!2*BnjGdkQAMl-+4Z@R;__o`~CZ7stB;bmaVPOsU{Xkc9ILR+qRB;7N?2Ag_ORX zGLFR66AObCIZvDQ6bGXt*3r0ATJejp_~j+L7@@CD$7D?~B8FA%VMJwBuX`=jS&xO+etD#P0%x#(?>_Jb~rI=%1 zs(b(HH;?IB?JnM?^Yp3h{~i*E5zU2AMTuEjSIquOeZmJ|nxDEPmVO{Gkv~sx>!tL8 z@$j^hTJ|JFZ{)n=c=Igtv1xIlPa}glwk>@ybZyW);`iJ&(fkxPE2J$1I_p7^qeK6} z@52~K*PYF7{1XMeZKG8y71dD41>I(?Xc*Nt!@H?YZhZtkHNVxIlxvLb8v*Oo*<<$p zqFL4c6nOJB&kj;sLLQlm(G&R1VvO?_osj?)cR#KOM0Y`(v)(?Vow7#`BG}v`f%`cF z%@akuz3V}^o>PH4&3iAZBterP4!XubsMzC2M&|7pq!7VL=pkEOLVgVW*n01{)i0 zSC|f;%{8Ji|NUuBmQ(erz!!ou7Wr_o00Ja-BzqgBKf}cu8J*ZKiqBmB+NIH#aE+XV=Nw+$@=GNtx_Lec?gvkUqJU+46K|X4KF>R(qUVaC& zo<@djezkQsII_6pqR_-ANw74M6H^+-?%?p3qYrRA)q75BN?q^qTpM7>+z-5(p?Kma zMQdVu$?A`5t#V1@X@v`r}vPcI^H6y)s7 zA>-FgA@#$H0Ew>?srBzZ-BME^(CPwJMp7+Yh$cCg{GM=Zd;LfqByTxPYz;f%+DHCbJ%uq>E99e2X zdmAZe?V5qIi&ZuGr;8<$K7(Vk4*a&ZKpR&BOre8w(AsDPUoykFAWvBp4TRI-##wf3 zH;}5dD}OsiSdM%!0$jNl;?|X;KaH=SW-Fx3NPXlBTJi!1g35ifxhrpWL;8O$%;bbuLkhy?EVcmVg;W@rPT-{q+LV#Jf-WH6nbbJ<)BS%%|bo`3zsv^s5^5d-3VT1=s#3x|l{d zN=!ZvaEJqtr4rNA0Qh>f!_ia&?(=}xhw12fU2O!AN-Bf~t^m6=%V96MmJCq42<^?K5fg<5^?v!#1}Sl*+duttPXR- zRb9@5gO6R!f~&ld-wDl!hGhi3ZS*zeJ<~$MheLsW8RV9l>`G|=6VMt8e4a73V-&xa zv1Y@( zqm;Ypg!4Xt@wGGhp%Nl*B7IB7P%pZF7-=p6H}Vm8B*{@xEpAS2l4%45sWEGr0}8aI zG7mG&0(Lrx(5fH$#H3ScIZ8KgP)s(em7dd{O%`YNjw7H=3*SqyZxDJUGxGpRpCR4{zl18#*aIr zUuk4_tLkl%ki>WTZ2-7}fS-rJzG=`2!c?brdy~`Ws=#VA&6WOa`-0l6Uc29XQ?&ZQ zb{x+9>dWB2kYMue5@P^n=*zVscw8&fbt`71Qh|O-quo8Ja((llS39KmtE?B zII0U{jc;_PZmWRRzON{Pw%@<#`aJB-337)+t&_^i-4S50=>eDyY`#3XsE7I6eK{@y z+T2}z>(qng*$~A@Vz5x)&#w_x;HpTEokY?G5HOW8Qs!zgYi|hNGYq;i`nQ)hEg-MeZ$9ZlDO89*FUGaGB|`TE9P=Z5XTdl$mUBb2`GRsIFfz5 zaQkX(I=YhUcdQdbuFxZ3d)Ckb`1Pm#bLPSO#}S)S^BbT%pp7ZCdUw%oCasZb8Cfp_ ztzGLGJ`V(X1~1T%f)B1Il>7?yX#K6Gax$4u+qwStYULSIyKc#4FGTI(*B{=@3;syJ zG9o@3=G|#VPZc-5kbffU;oM zn;x}9PWl^XuUsht1n6LFPcC0bz8~4bT;+6}sB{2Xf; zPYUjkbII^xLm-Zei}LpDp zTMS?>h;h|Xjj$Ry_sopaD_pO~Rv=OCWHKayX48xo zyjnDEA3=$%#42Dm$59jhwRxd$-=qxLG|-~Ehch9Sds`5yMVzSlU*y$)X3C{FV=3Ov zr?}eBIHkd5v%2rWA%-NYSw_y)b#*%o#>tZr<`4YUrZkSuEaJ6QC=`}Q!_TM^*uMFg zJh?8En_{Pk$+!S|c3r<`T`5Koo%XqB(^_Gm0{1_vlFe zWTD|^QuNWF{+V6mLbVNm#G(IpB#zD(fE-Orek_2t^YOaA@ZQ%qPD^1-QvR3x-OxX7 zzwT)VwWNw;-UOd*ZrQ)H`jDV6xUz6%ekiqm|3r<%di8sv$cMdGXTg)qe${qbcDUqT zSvi!3X=2Wwi21vTPdG6@a!TMhpJlH1`dsW#TJY|jSy?30Ta(-Y(-+_po%}M?Q%Qlq zd2&GfB|l$%w|Me(6jjHksgT!7{yBg4_sZrLT~}ME;wWL|o0pe252AAaTKV(mGpw~Z zaXWHk1hQ~J?L|~69B{=jdL*hr*K;f{H%O}E4O8ShAGrD6HujJ@+SDI+a=7L5OH5)D z4%s0^$JPZZx&jVIAl9cqpTDq9?Kz$_pyqe^IwO8$0@DYe5`X1%eY>t_g+B(A1gD*Z zL#aI%nz+^?4%AC;cf9A#iYiRwKq+<~Z@uDhb34eGK5|98QEcF~Mo3iN6t0vp7mXLg z7#iOEN>#a8zJ-eUH;(ljku$9>1D38=NdDJbO2;DfJI03i(D7~iG)T7Ab|alNzy5*1 zX?yCQxWK_@ipJ9&(DP@bdlhYjO+`e?iw>`6>mgL5D^VrXa=so-qns${shH9?r$XAr zqTbXz9F6*90)KuqRS&MJH3yvyQuAjk`4WcS8}wZML*ZYFked&`;QCX zQG4w;?Yy5uSL_YzIAA5%P*to_l0e%a#9rshbF_f`x>B=dOo#D_cWOu(;Ku44%DL~* zEP2xuEc-us4;ZH$+KwvAb-(*OAmac4z26Rn(z2WQZcxWlBE9?MF{|aC`H+KfU_%VM zKDgFTU+0jTy~0^#OCgyOhaQ%($h*V@u+jo($Na7hi9v?pre-^IZ_#HXgOdSsG`ACMoizhG3+9 zAU|c|`{0Xf_0)TaW}rGJK1OXu-LpBsBw6iMFQZ;^PamfjDJ50UIQ8~)IR!6vU^c=v=~Q*>BdxUvZ9_FWGx7{4p5x69#gvVcf}_FkFby z4DpQ7iIg6nWmgFxXbp>zrb|3M8Xjn-jNKzgIEP1?{AG^Hx1Ir$ZcXQzw+87*3mV76 zRCV*QgD(i+2}?h`U&bMO&|5^}8DwG37-?lvKF_}4{Fd&vXkD_y>|NjU3Za{2)ytIW z65AjCSepCz^VV)dbkfPLSRht!*qZ*|>rFTEs&3rAxsUdjLuZyNm}cME&$>>AAB};C zeKDSD3Jg`Sb{L<~pkA8cVv3IIO}E_tcUDm=BbjR?GsL)M4rA?0AX}tOF`V%fU-(Z! z-6m5D3@PRN$#u2CSy`c+xS<+aPYy6sCVE*%@9P`4?A^t_6B}HpemBl(I{0?3Fc*YR z%A-jIL&<^dLy(z0%DoxV;(;`_HkvZL97TBJ5e2&+pJMwS**Bb3dzn4Ds<=jZ$n3&js{Yh9@o z`lO9AEJ%Hf)21n{!w?po z+@g}c>P^$dwnuy}WcQun^-^E#0Wq=h!$E+8?D=o3`BP;$3#AI-TFt!t5)ny9&U#T& zCkq$erQGwGgQO6(y|R?dZq*{!|Bt=*3~Opz*M^s(qN1PzqEr<{ktV%^ih^{dN)1Iy zs8T}4X^&&^Tyt2!R$QZwuJQ|mmisgN$FR33np8WEJ@b-t0-HwRw;)_#B1+4D_t?djI? z1uyBE4Y7qYmeotW&z!3AR3gN&#sMa6NP^hlEx#h~pK%c4h>hme1ea#}r72sML#0Ix zpv-|jGa;i26ArZah_XwuavSY>kEVR{-4lcKZ;7;++vlX!dyB6DjM_XuQuNdN&boP` zpsk+~Nd5sLe_5ogd%OsMC?#Qw6`1sXWN3zufiuDcu3mHe%Dn{9Vp}L(E@cG3q! z_O5Hx^Go9N5xn$5%HGYDHvY@2y1TT_vUG&o0-at4)RY==Q0qwJXhZWpbgKCFrqh;} zUe8(Qau?$cYjBathyZ?Cwgtp{LT2TS_G>}?C9e}zS;3#y5QC@)z5RPRFU@ZA+}YNt zs7rQE4Co@iE~I067;%MzSL?n*5QQVwE-QP=$^(LeRxLB_iN%8Bn>)nR*s`yblW{ku;#z!4S+zBop{h2%>B)3`of5Qu zD6N1a$7gEU${BqM;txrb2kRx*%%d6L-%OAWds_pb@qs*&K^u7RfoMyVb~DqA*hv2S zh-fJ%kq{dL^*9Mw3*&OU6#sRY_(`G(MYkDx5**jC(_SiYdTBhxYh=6gplLZ;<7ysd z^sCcGij?6qc}2;``_wSZWj-N2AW!V_DAl3Wkbn-_*n3y~to4JWo1!dQft|Pi91D7T z$%meFr{ZN%&6$j0S>aZd)pm$ftZ?MXsF#&5t*CPVB3yuz7phQe=tvc=6j;0nb1j?B z<@~sEp1I2b{4=lWEmIKR5ID=r=>`B^Szu*c@{rqu$Ahc z%XyxvH#I9kWg_Ml`?AzZ{#ExeM|=ae1kv|DL7_BVQ+Xr5TAx$>16;EEV$35VSfDgJ z9PHaCcrHqV%*(WTjRybnRqyf7DZ}T5Li|IyVL;0amqW)q?jO1G+A1NfXxd{TmZ#F^ zCfvWW{q7*yJ-l(642*2}c!eLe2Vmc*$P2bj&&fa@aDaG$>B=VN=gbXEZcFwcK}~Y? zdRw#4rL~9L<&#`{Us}a=2CS|pKpPg8Plcq0}Co^5IcuM54S4hTa_tS-RP4COu z<66T^OguU!iYrO-2~6{0nnQxBsc&j;0u{Z(BKxP-DC|0O;`4q5&|y%OnH9yz3Z2hL z4OA#vE(}d^2r1TTYVgA*ZZE~Zj<+{iEM585!@VF?cwjv%%i%^-bna^%v%g++2}puh zne=twUqmdng0R`WwosW z%+$wYKd*ZBWVSpoN0V{XtPf3+7&#ddI|qONXoLQ_5Ue61F`7}~(yyTJLc0>j*J`UaQe9*!P{wlK<4y8L83*-zISIs`( zDDbi#X%u}UpceB>b9tW;g$PD{S9nW}o#g*qt7Wrz_wikwIrHdVgV$p~uC>d21vgo{ zn&VAEyjJY}jIviK7rBeN`cY}qpx^>=X+yZ(au#1!R>rn)ShRF7o5*LFT!2TmDB^eU z4I`IZozh+5{Il7kDF6_l$?!W|AR-sUTwx3EInkO^sVS7+dBFQbz85rCb{VLWZHz@7-2LX0-pfEc1E*KpSmnr$=25 zZXg--G^uU(oeCDGcPWRt*Yd~&f|oKW)o~EK<4eQJRh%f@L$u(B-eu3Fng5aWr%NJ~ zpWiGrbgq3Fv|Um#xY-GW7gW9<3B5(5-zk%ta%?1n&iHf8ix28(guk3hWOO{ZYNN}i zH{_RW4FLYaOv@&ajilDrvm#z$M@Y4cE1(MGS+iwCkngh@4Kald%0hg{r*H}xMgf4hKSfe;SEm(T4l zDsx(U-sfKq(0JQ`rFX) zVTE#J76QFdM4ls?kR+}V-F`>PG!Bt|N6P#|h?C8}Fl7@}6WwT?FI%JF%0t*&eodX-_in$b9EPrE1{38r zAeR%TVtOv~9OQKDm=(10A{0gFdE#O;uu;|KZ@0}f3gBF&y24+_1o z6x}XSes}A2r~oz(n9D3>h%kC8Cd`9*=51y{MWqcsH9Twm>pUieUGf@%-gn|Rf{`Oe z+i3+AxxV!4)N6EMq5WtXBhHL2Xmf^5>K%<7=v{MElT#&R9u49Lizhyz?o}w*6x$DH zchgM~rHmp%Tl>YbsZKRr*;5yK{>*AGY&eip{X%o+@ojc#5Ktex*3DD!NpPi>pAmQ2 z@*%ppfsyOA28X2GTIf3%49>=p*;gMa$8OHK6QvxKJGNT~%C{sR$6P?%3<`{oR!WIg z$hl`1oD9Rraj{&~NC)JaW&Xv9k3n>c$w=?+XnmbJaO6FG=mP4K-a|BV6SKIX{XHM*fXDEh))L+*`8h8|5nE`e1C~H)Z_}@ zZ*lAiX7xglpmm*Xj+5`~&_=31QNwGn55b_VoKn(IzjQAOu>o%bDjaZw%8TXpB?R{} zN_e%#;*w1s$oQ15Nn-wFfji~Z_hx``L*(w7VJu;~-G9aRi+YlbE&9xX_4|0Nm;`G?F8QXSIeB6G<7X4)k>#mm zXLN_TjVQ0;{R1HfV@mmSw!~x2%G9#9d!Ax~q1S&_G=9yL*nr~$uz49Yue~gOigkWI z0Pl9_8;kOboBeu9VOuE6{I(nOV6yHqX*BNqJw3D57u^vz<*Z&62!R1p4_l_<0j4gv zp4ffma*AeHjqNNl!K+x^ux5^j0{tQ8S!q%PzbgJ<)@7+TJK}ZMfJyZ-{SJ3p5VqLv z#&8V1ABRnM<+!x|&PC7Ybz0m=w$5Yz%IQOeP@893-=D-+$)!#^h(uoJ`{guvbB7VJ z7!)n5r)SjESI25RBH%X*L)8~b1PEr^>AQ;+(S~s%8ayGZJqcw#qc(=t+@)N82L2rqybSnngA7Y+5X-vvQon}} zp``>QEy)0q70U^lE;pvEmu)v>J<~yN0)c}G0DF0?fv7i7wzSAD`0Drd9M@F656u)wqy5S9Biv zSbV(OYG!e+xi}G(lo@kzOExVlD~rM=RiAwD}dbGU=>%|JfR|8;o0oMK_$tgEn&Gvu9lC3HX<@e1xEw z+A(8La_Q@cOGgTf4coa=;?3uo%~#y z6XK&y=CLIP4?@}aCSzvr{c_?G9oI;yV_8t19KoV7i|1!SH9EX=dUn1$!Pws zJ{b09lRmD*ee?R>^T#%x=M>3ekd+WG)`SAEWsLoVr&mxty7lC!%7lH+c`nFASSqNS z!GChJ)@|SrSQ}kuiMu5T&}9q6a8~4lJW!$$Z0|-+vO2!4X}c4FIDMFAX!%TBcvfcV zVht1~sh3Yqe6&`>Gj3JR^KccW1^dcb{VZDz#%HuApj~pguqpD-5_u3;I;YvNUYT+9!!Uhw1aAPCIXd2WH3DSm?RP0}g{!1}xHL6FT>OOLIuL^kiX6cTdb^ zD{cqB@_z%vJwJu}S7F_7x^Q_JvDoQ4Sy`NiuYG+tK@0xO-VyRoB0{yjm7kouZ5smU zo3MhCl6GH+Uf>wspB(Q#kErn5+7jP2*|1iB0nA6q8Y323wRJ51xV<4CeLLMx@i1p# z0Mr79JJl!QpPWn}cVF_%2Wn+T0b>(HzOBy%L6m#A&Ige*L!)Mm?})3Xus)ZZEQ_8> z*slrl6lP;q% z>LR_9zEbZ9X{dHbxL*7r}kmZr=vEX4Ssrw=1>zgObwB55m)2N?> z5Jwhe-ue}s5}X3Yj>Q85opP7m@rAH3%*qFyfcncN4Oj7<-jjUK&c?oW{Vk}Z3k_6? z1`y>1=v}+@E~WH$S`!zu9_8cV05R!0&=U>_(H9mI0K+AZzdNcIIqf&UtY#C+wWRX|P?KxjC} zl%s6W-FaN=&rKg$IQ8Dl&TRoxJcAkaKJF{AT{m*73Untpv&MtuUXzwbY|=~D+Th<%xnt^&003($1TSz%@8)E* zx_mK&?Iu$MGgzA5SOM{HD0s+*Agew7iCd{ryBEbN7NB^^Y{eT5$w6L!3!8aiA<<}a}%dzKKT z?B`!zTgnw@0%&B-wpOXMGXciEa_9yYvO8pT1(-Y`>#4tHtZ|r1Hi>5m=<(+X*EaO( z@D+W)qo5(9i~kB;aGUy4LL9bN4xLfYdO)deGAxYEyBH{P^{y&q;>Jd$(Br}0h!7#l z{On8a2CFW1$C9VDC{V-0NhGnPgosX?FZ0k^crm?|mxnlnkpR(`&6lD6AFL@dAa$C)YH zvPC-*E1TSpE&-X1+`UN-o_TV?K}kcKKFiXjrkje-@FSG#KjMA22Gs{z4hMY5AM8cu zvIS3n;Aw6d_0}yB+>N5sOuf;@k90_L18@tU9`qe&9I6_Td%b_BotWQoRpQs3JWv3O z3BezrWvJ$GoHF@A!OEjXQ{Ihkk42jd{-d z;bXF1hm+r07vVCTA44V%ljktCB%P35JMsmaF!RqGR8*2D<_!VPj2Ay#noqA2?O;WB z?S$?Ft*0Sz(q?oebs2XH1^3NZ;(fZ5cFCU{m%bs536(wUuXlSAkWLt7Puqc}(Nh?^ z2N}7sgWoA}Mh-;_Kb2mSSA%K9{U&e7*O_YJzWa`R){I1ps(GDj`$vsWq~czTZ~HIv z^&gg+CmuYM?$2C97+MQWdh@{d=&nMxtRBc;uylKu?s#Y?cYmEAEqP{&g~5i-y<8sl zPbB!okVFCmX&st2fx*LyM1AOcHeLDiTM0=GpY(08QOUZpB+Tf6&R97r_hflVN54^P zX%C0c_ja;J>}`|vNP$Ey8VFOOq(I$8ajwxTARwP-B>Q#*;lU%?*`trp-9itW8{@o5 zP*HZTAoloR`eeuI_cB}Cy+792Ks&W3`pWr5h-*eBTjD_fi3j)k^-)0SwA^t1+}U7! zy4@v++{S%dH(Y;_`s)5rNy7rHkV}bfj?V_cT1x58iCrdqZvqmJV=}FyOT#J6#3>>YbU@O`KA*2d$0AXK&aYr^2hf?>DALa+tE9feZMoY z%&4>2L*of(vZ*{wA7U>#u^KepUJ4M&0Dae9;~DDR?70DdXmvr{c&vUGb-nVmD#1VK zv15GN6`yl@bh7@RSp$|{J@Orof139CrOi3!%&S;AYbWi1uE3Djv4YR2?j-C$%wfxR;jiI}&h3P1_F~0zKQ*j8DW_^507!PGykh--5p@{@=3{aNMaa9n$<5HZflm(kpBaCmk?Z~@C)Sv(Q8$%{R~hb%PfLDw zIz>Cj39t{T=ybX}=Egk=7GEt1UcV<#W{>6s2)N!}8!KUsFFS|a8Ql<`O0Wz$^JesB zDL_vAV=ffiN5yCE3rKE~!z}o?$+{!VrI>@%jmsF1pwB zKg4IdrBhSOhnCbzqP+`_j^iGnU)rZcT{0hHqenPIf=2tw&TPz738N zJKM8r^#$Hd$y)SXLA&wZ!=oBO@5(pB70!<4IJ2JBJ{g}j{6d!UQaX5_Bh+4@kBmKD_?Egc4pZs!n2_rK&838rO~9 z?9ev9nf3!?ph7k@?Dm}*qSEZFyTHc=w?4@F<~|Ph&bYS)idHS3!e`#6VKM2*3+56V ziS-{sH1)9nG86?CBkYx<7 z4CNK{b%<-f+I%xfi+#wEr6fHp9(w|pzf*0C6^CBdEZX6qiNzVs`|K)YY;?+%H(m<> zS0VbPT|s@j%%gg~>z2z`r>9jos4wr=&TdC{KH+SNUvWaDJp@r{%3t@F9^vhX(DfO! z^FO7IxiI3IoH{}=l{$G}pk}W93CL*JNb4iNEWBrtn{PhF7eM$D%9gy!<%0Cwi$6W& zzL^$~s)4K+wwRPySfx@`5djtU+q@$Bcj@>~6LAW#t#bHgd@n##p_)1CwR_CQUiIvy zEbQ5NrlD!oEJA4C+<_ew!Ho$)AbL1{%Lcv8xdmPEo!?P1I;WZ~e>?^9Ju5W8k2L#m zaN$?UH(kA!fCE@>VK+8%Sr(!KH&OoWbD5v&Vjs%%mPZx|547*_0Muw-?}V}U6)5+p z^&nT9t8*G-YJ3G*#u{9h$kD8!v8Q)9s+y;Klt7&)LXWKkLpm1To>D zL=5V@{S#WzEENoD%EG!9^3Br)!BLx|PsVZmadt9v>(x9^f%dkXX2Y6**@4`q*?Jv) zQMA(hkJ)g$lQ^vrtki6s1o2JJFeNMn7fCnhjI|E9@M+;kyb8JU(}20DO%`FYZ>}u@ zWxYZ4*j&rhKBZ{hA3XssJTLQ<_v!Pzew*e99LQ)Iw!?1PK`K%)RqFGq$=<)phL5j+ z$UAO5n|HfYs6V-=mY>rNuZZ>6(D2C4s;M=)D_rf*6<4I&flt&~BzN}~mFeu78WoZv z?4SpF9s97}U;4$w0}lFh%CNI;&!zBn=`Sh)v<8K<5GA?jDcMi{6*cK@MLeWZM4(&=DLj zWo`pdhGMQ6>J|2$9U`*#hY$8>4jf2A!Bd+*L<)A#@)V90q=}S|f;>lZ#)x`Sz3Vk3 zSqUNaG2vO%K{DDb@Ug$VkUE|IHmt#67&+P=lg$S)Ny7T~iHh6y>w|ZPb^8a+rtoky zdYYbR)CdfU(~p1Kp{ad}nCe?>?Wo$5oOwOMC|OLrPj}t}CYq{)+qaWB*r;A_?xo(@ zs4UFOUtiCPO!eR`TqNk@?+Ml5ryP|d!BUBvxHyb>3@srUtrxbS1>;uVAnIhKH}$&l zVnT>OmO5LNzf`Lj(`_%ME~!uB)7nWhjOWR}>Z5{J zK`v4iC_f)#+qG#$pen}BDZe`>@4cqS0UHXK_W_>`^|OoajjN+`GOF=3qOMNTHgfEU z%mTmCN^eL0>~7pQB54}Ugp@Y>?5U;V@6VH=s`XV6KsR=Sh3^V>jKt$+X)=Zo+gESb z@b?T9)!7z%uUSX3EhnpFi>Nc@^)G(Cgs&6uFPRtbAuG@OKHMEnMW^Ku3_gKH<03`@ z;X9?@J);t4m_N&*DI@B-xa@h=U?uKcPuZDya+t{U6MNOHfkL9q&&_@4I@l(nJv}rF znkFaB;^W+sxiq?Z&LXF zyiGVU`-#588h%iS9aKVc(dWgzRsGVw+u)1ty=J5xBP2XudjiL-|x z2RuA81>G#Z9myN^N?vGLQ*|CItX(s$s#K(VKBs$eU)9}kH~oN#fLomOFxHZAApT`yajmqKrj+v-QIm>!`xKg2J;UMrv^>}G;MZz zjYCNzz=CjWV^(>;a^;j!7;jh|$Pm@BrLtSuJz!q_R0-FtG+;v!J2ottAV)QW0Oyyk zr%#uA`iRO&drrx6muGxh@AIOA&-m(R!81D1rW^kGTcqjbG0j`hn2K^{kj;Aa-5sBi zC~Qj}?obHeCLQUHGaiVo=Y{yhDNyuAUgicVaC6S4XX6DSl&#~8&iYn%A(~HKtXiHN z3&MO_R@_N_nk64n&@|8#;RjU2GDHe}cBP&tM43$yrI*vJ%(@wJcCeN`*vRz(A0K&V z$GrZ5Or3yR1Z-h>{eBmZ*_#SMR*9vh=^mZ8g!637^=GxGp+*TMLgyDHHq?ZzwEe^F z#+X1v()Pp#D2Go4nB|P^-jroC=X-`t(-^Ttp(5jF?ruku+_JtC=-?ku-=+xJHt?6O6c4Ve?a468EVni(VE1k_+e zgLXNx!w~36HYRCl?$f0m1sd%{|E<2&I`@p8M&4jJZy6r(ahljfal8)?SE_q= zLiW9BDYpOs2Z||}gZ4#Fp~t4ZpDDzSwCIewdT>|GMBK!D zVM2uR!e+F)cO=h5TVn{;fn}Q)P0EhAs&^%8wjvnpwM$Hd#$+?8ZK1?BDNlONLkMzC z?Xsj4L631bNxE96^q?CggEmNr@OgcvNx&~HxmX*NflSiQDw&=8}&^+ED1p{;yd&RQpBMCx|9T&x_d z29&U{I}7sXS0fXX>vDO8!;5GA0Otzff$fvhTgLv@XQy9|CdO9eT)5Jjy`ZlB87sEP zE2ukOs!|>gsEYgy=5)t=sB?J%9B_)t9TfE8(iGi*1Ks|Xq@DOwL@4E;+Be?bwT5C` zSo#WZ3MvNPZhFrIGl4#&$@(;;xDl6oPnXXdvhBzWnT}Mi+awW((;{-PFXw%KZSGvB z?1@1&lN6Clb6Op<5A^*63T9^Oi?(G|Zc$SGn_}b?_GTv|OR==`5V4G1w_(Ad=Dwvi zl~F~T#7(J!CTx8E*o++}B!{TGSOcr^Eo@r!aQ6q{aFno>$-X^6P7#rv>F;$nAY6U> z=(29ouw4x#(0SU|_S?%qc5SxcHfe(g%naqbgFEwwVi z7ds|fVMT38zW9&HU!Czo&F(!W7@9MuQ6679ZQo>Po>&{2HD(Creqqt@?xM?{U`SIo zi}TNk!W_&)%4BsigJKc6$ZKzTAT4~YnFCYl82o=01T67>#%aIJuI6F@`Oz>D%E#(U zJ6P2>qVc7&qXNgx?zI|K&@QTPdPP;B{1I%L?f<+}hQjB2u^vlX13?pWoU|XT;!I2W zRQ>lGnTosZ79a1*Npw6~?{*>VlV$;FaNQUaITO0AKil) z2R>Bby~UAReB$^!v~p(Ychb-LZx4W)*KVh=#~XX<>J>Ssl~6bSf2vJuv@% zF4ykup6D(kP_b?$2L}Cm1s)>(eeVf$9jW4=&3{@!1a;yE{3vlhenZv~=0cQ?! z)DiDF0`>%vS*F@*7dJYj^@+lzYfT&iZ)w>slb94e$P2LXp*4K_J!lGq2o^l>bBp9V zu@dx$^ILfS3&?keb4JJ&dr=E_NGSZSh-~n3Ys(P471d4I?Oz#DCAYu^3y{!eNt-m* zc`}G5PzzYQ*7?PZ$G?)DG2yZ#xjRmn$)XQM;#P3(i3yrhO?5PUPoH@Q|J%L&*QtjU z9-pUErjuUZ_5vEkhEa|31>YP}{MU;Gcc!At4LTqE@r3S4oph?UddUi@0}Cz%(HYe^ zFlYJ2r&fg4kOoDY*eZ-*)2z0}hg-Q!7WAKm&{l~@ND-&JNHt&h5LTa*xgtFJ4P}XA zzW3F{hBEo$ilhJV`oAymKIhp{D1mCNR~vgdTLKDSC5+AW7aFx2B(RwmN$xg`x^-M^ z+cIhr$qwIcc!~Y@Q_rOe6mi`-ALJCSZ|coQP9u(eAv6P#Tw_Ww0X&;js>xfj`cRPv z9^}s&tV_hk3 zo9}&BSaqV2$hl8ElOO9NI~LIU3#t|^w=OUDdgzfS->yx_gCgd0lEu>emeMr(reFL~ z9sh?r^3SC-nF1=fIM+E+jyNSjMunD28+ydYQjJgD4nXYoz{MSLMH5r$6Y&`_jjcJ1_qxIcj|WS7FtA z*?(8u|9uO4O@VI{n=Ho|PvDxW`8^bl^s|MBz^^isivT41_R@wxiL8+$+C@!b{KPG# zWVrw8IA*ecB;*D4Zl|14!MDAV4jM)xY8x9rnP`(93Bmiy12cU#=KAbEBxDelif0yJ z4Vlpz-4l}F*`HtPEmlAJmQ2D5N9C=P=Ec{*<9jf8)pb+vt@3TV?SbOH%PdLX-X4~n?lkPRUJN2sWVwRz zn(`C|$hoSDm*E4JgWu>ZTe*F$NL6z;XquKVspaK8OSk{qRr!N~G2^{(!HZ|Up*~XT zdM9SJ!~t7nlX?jTuhwemnt|O;+pCa({;C=Z)Yg=e5_|4k$Q7LCyWc1=-zz-%J%rt{ zL>`}D_WF#);zP6L&s4-th^RM;gbuN2bA)sfO?<`;1jpQa8l)DT*|`hi1e92-0~|{G zvWKj}JJAvHLSnl;Qr^WAlNES>CMJ-N6)>4TYi)-K^OpR9`T+UqwcTurw<|2LYQ1)( zT^M`C$hkrlwYd(P)@{5Qt;UMnw+b2D!H$SX#}RTxRy}pE&p{@*sU{~QU062kGnA&QKhH6n4mCX7jyB#3L-St5 zm#$KqNUxPVYRj)_vZ2AwG%g}E>NSRJ2HO0AmozZ=)F**$cUf5}x>3AK5l6 zy1zZy>`>atO0V84jN=Kjj#tcRi&qyjl-H0Te)B7BvAh*nkr5%u-6?tSX~KIlOjx5` z)CRvjml2tKf|Q-+lV7<=T;!BBJlcj`L7q5)YuyzF64a}F;b`Js4_(tudQD@!iGdCg zb?{lcEUOxi{;(cRN#X!i@B!d_!FVO&vt`s(lRVuCvYzi-Zm08(XdGmLX*P)!#QzQ()(l(_?P1_ z*w^MuX}Gwt01*!%9qKrPyxt;d8tusatylFEp#u8?Bk@I~oUH-=!n;P!0;_S^Nn@4- zs?w4kb!qh_B=*%@#ZtGRhxu@?dulJpCJp-FaCV46qDtL|4rX)eqj`oVhdQqyR7Jr& zC)H1_0JFl;rm|O^_}Vhh!LYU7E>d>?#bsVqZNF@=Mvr?AA0L>ZvU~u~D;;y(@-{of zE7~F}utLOY_G<9cv0?i{T&CK*B6tT78v)GTC18riqfysIwju7IB8&{n^}@|PKWKDE zMPvxOx6Fv&!kRBpr zgPhk^_4?z1p|}RTh^ezfI+=i{&s9`8l~QskD*P^DP%GevqL?aYR@Mc5RdcvF&QIE! zTa8Hy=QSw`GGlR(tS^@J(qCZBnC&GRnJB|yyLac{2LMILMuk9!Ime3y1bGnD+4Alt$H6M@l=%kb9{b!BQJv4{M7B72|qCGyKxd z9Aw7qLDtym5Dx!$kI#-*i})De(WT8$1t&W#b?_@kk_jZNhBTVhvbG-HyCJ*{+Kw$7r?77SCKo5XJ4hNCXYElffS1%IVJQ(#!QDWkno!#E+)r-c~->;4^?df zYehbpGCLO4Ovm(KKL82ETQp5IJ8+w?zND4q)Cq~BiL%FyYpeYam9nA+E*tt)@ots_ z8JVW@1IAjsq7$>}mg9|ap;*lJjmySn=*s16*oWp;eXZJM#CHXNk9j2=;P{o5+aoDW zKmIV1A-qbg3l(TQjBaz%z4=W8CJUeRq#zJwOkUEn?a`LueR@vIn-lwtdWJkmy+fN& zsyPZ8P1iG|K~c6RBLzoTX~ellqQc1p^sU`?IU^;FXL0`}cKn0m_FEu7FYp#BvrE3U z<1RppV57uj7`>*fr?Z;0BcWi5icaoQvSYa=!hiE3btAys(ra;Ee9X^xshn^gtLwhkXbg# z#|B4}vFdNY-Q!8!*c8S#4=*~$vB^%=LM=j;5$*jD{NgKz*7mg~7J)S1^}qzb^_HZ* zmpyzzmIGs?@F-3t>t?NM4#Cs$;lY;^#H|y(v!7(7WW6#lj9bK5qYM&oW&MEmH5_KP zcV>m7b`3qpTDX!$`i*Ql4jnzqB1v*n7$)($)&&8+icwFSCDbBr$Ok|p#AW++Xz-E+ zP7Y|ja?HMJ^RAhlxqHgi9(Y(1&H>8Qnr)KmGz^|73B@|On)WhH2lK?Kv?5ZP_1$xaj?{5DC1kNUJka9o@zTl*OH>}jH?7Lx8{oS7w zb4Ic*463REuT)qucr!v~?^Rqr_X^-p3s?a4&~!(ElGWob8|U>eM`p$LA9tP~d>FbF zswqd|wO`!Ol#lT=IGU$n`xJ1^#H9AiFsbynQ-$vFto@)0n(WgDMJ_{zWk=9+dzeNQ zO1!ft@!dsEi=v(QFaW%Rz7Ug)ybyy#7$o{NA5ie|f&tf06WatZy)vtwO|=^nZ=$h3 z_@ThC4Mo6*pYE_FdJRT3H=krONn1X&_H}OJ#%{``QYw%b4XR|tR#ncKSboQoIL|@v z$SZ!$V+L!t*=wY&6a1^YwF{4SEI;r6ek?>lVZFl3C^-YRp^Y47EpZdJWQ?6*8;_4DfWYA z?V%qQ>Ck=e%-;sYj2xMjSciR9?Me$h!$k6jM0z$O$Ztm(qSeK~r7)R|C7R+Py?}`E zdQ4|TpVTUGbVHyiY0=@6$>J+0>0K&h6k(9;zrPcI<=eo1)N8LDE#+U1>n*oKk%UUY z>}1cRy|RhPzz|l(cvF85M9r_`8;ylQ`|OAQ?V6V%pycC@zu|$S{RQvEzYd}D+;NNE z*a&y)#N3RKn#pt!R_9F4j^;DbGP#}ZsVB3!YFYf<$KHza+-FqkXo}y&2(aR*`&Bd5 z4~hpsbWU0e5k~Ovotac*3%Wa8gv~bTvIrlP+{wOJSfjY-pO|HFDD%K9*z0M2yH%Cj zKtdos3iF8Mf^1%eV~TcSvca3BB2c>|MUPPzWMg_iNP3qK>AvUG(l`l_3WVQI_u+S9 zk+}&lMZBLI^Ig~m7@$jt6--_uFM~X8T+Q-j`L-BQylp|#+(56+;EdUDGWMI-HAv0- zx-uET5=7(Vh80!PbDj?XZnIVuL%)$PrQ^oQ_3n~Zm4J6y@uoNYD;g{WTW-=fCkJ;; z@*^jbuDFt*Lu>DSh+))(8oumQ1O$ zG!B3sb@Ae#7L*ch740s`UE=#g{Ln^$$frS5GW(?kZMG&4>W@kHI)Y`{&h|<%+QzL>y`0BM7+thbHe-nuq`By}ADW{4jP|uCP+- zG**iK+<-+Q5l}*z)Eqt{p1f*~Ns;dcEFK4^YHNyGrFiXKkZ6F zzhHmhQa*uTJffXJ0CwgirF4;<%vf(vy@JtSDy#2jOks2x#&@A4KxD#r4kl~!}vm@ib2=Xa0?6+66s6uz2k}V zAe2cD)1YKuVhe}n+;iw;#1F%EF7s-gSrr~&CKXWmX^a{xp<&pE-CT`pMU!x3m(Crg z=I|h=P-iqnOkUPE#l2hEqA!-*{GED<4g8@mF78klBL;nPwr5r0K#q%_LEqTNDYNSr z?-BjF5FoK;JC{QWfj&K*P{SbWyZKIXc8R4&Q+(OB(rzSvdj(0D1m@g@n+GE(#EMcy zyj*GX8+whOL}f-+^)F@m2U|XyNv>xbHP-m9Y}5 z2^!W52UqutIKuv<2=-?eZ1>UyFBp8k+wf?IcUq(XIK6t$8kVCeyI1$2Kpn47KLE^k zG#zn*Sr1SW=15%l(M-kt&uAdJUVvfBj^?)=JAR%A5UFn_gr+lp^yvQE`ksOSW3<%= zGE_ZMy!Kn*TSg#3I9LRkJlfs3kShRK?cD`|!dpj5NWHpW52Q^+9R`ptM?ACMc;L*S zTk-uztf_o92P_mbVsY$<4*iJ*Rp89fyQ=m_+|r-lfGs^gvgMTWCs*>%&SC2o5CelK zS>;FEQrqXimcCu~5!X24nf)yWe|4As7K6W(&cDUrufmkSt-)U^yuYo%zh^)Hbi4kx z27hG+|8|Q0((eEN(J2aiON*ii2~Fgrb1Lr12zC4H+FbujtMt3e+u(Gt`k~wj&*BIEe~H>PV73Gl#U;F zmXCP1U39^-^k_ag0O(shE_?P%#Tv50-%pJdoD0;6gTEOu4;`_p0xV^jCNm1hHTUpK z#oIHet3!p9+c}s1O0NEStL`w(sc_baKKivBU-#Lm9wT02v_sz|_Gg~!pFHvI#e^4i zE?-9^Vq*#%6W^XjLCKRpwgIHs!~E1g)p!5-3`fBZ-C zjUyeVnfqDy;Ae-w8=>R8fj($)&(7F*zijxQ|K`s=34*Lngua!qlQ_F`1wMb}U;pFim=%FYs>y_JRnO9_?`RS5&wD^HCAwkaPZ~f^e|A!G42?K)2aHJuyyxRPu1^=JTfzaG;v=3O`DLvVv zNyquaYUgCpWiOcEw}m4f&6VGeru^@t`Tyi-iUw*&Xq@*xwipX~c|E&>uehInrUbe{ zx+?EG3NUhnS{3|>%>T!A0>1NE54gDHpC|wX%@lY;A#{HIbQl+!F5CIXwS)g3p_`ZVI(&L z26U!AaU!SegIOT=7D7(o6~{Ugk0>cjK>o)`t54QMa#gS6*_fdMy^p>(f+mjq52T^VN4leq;k#>UW%hXQ!{;u{b+J@MHi5x zXR)wV6d-;q^yU>Q`KAiaBaTRD7yuE$gg^5)w(JnKpll7WOm8fXH)w0h@|k+=2lppS zbxIs!qnmZK)fr2%`y3LU;s5q>Z?i`;)oGnp+JS@06JOY~rBnJEfJaHokvB^%VrlqL zPu&T)n#!@R*i^`_s?7E;1;%s$DqZp1s!w;~rvhEX``c-i6B`T#HMPK$iwg&SYmF1Z zKAJt!)@j}|M@nMXCIE;qSh0`(SyW+J8+C?X#dJog2bT-^BUL*oD}_2)ko9U%b9T7*NKvpo@d=Wbl7mMTN!JaKpz@( zv9MwS7zT0&VZBl{`{6xTQf{S5bvI*<2p=6vc5{_Va5_diP7JBAI2~BwJ)0`-vzhSi zc{4D3CIQ9W1(=I?Z^5oCX^`jbxhmN6gYp4WwvBdaO&25#-Yd20IpSE+P*B=2 z9A>_~3ZLwhUJ2ujl(`EOxMnBT+_2KpaR%0j6Np&B+t*gbW|mA6|I+r^ngHVAk(|#u ze>Sk9Gh!v~TC0-PqC@KRmhMU4b2?h#wIhdjf&;qw%n#tC7c{!gD%Nr{I;=Z2v#V#H zG9I!9pjQ;(+`f$>J!jRM|fVX5iXa-nU#T>!~F^Rs(oRN9;;jXv&Q9qy>40m z)$bv}zDucNvj9Gz$FNI`6bZ-@&Uz=LFQ5LU~JGCQrjW#u(OFq&6v6 z)3KN7HDz2v2h^8Y#Fm1#XBOd>omDeoE;Non16t8m`oZ39xt*z=1fPXGi{1GgcO#?#Xid5I<^;OX7ebS2!a$#i#)w7NGjF!C=TWJDDU0{x84ewT3isezBO4_M z&n?5xSopDgku0!HoDLQc!v!@)lZnyJ6F5`oR^Jtg)dH_y`_xIJZN1u>Y@vJz+a+}K zieUY@<&fwV%)x%;+S>_;V)i3ZFNyyneCsHo$(3Fvzq7G z{glq3p)XOZb)g0l_v9rGF@0Z8!YL%YI8EdI})-&>xKxTwOCMgaJo_%yx6@mF;Yn~$ z3=*6_WCEe6@|LYW?C8Y1M|Vu@%z|tY7{BJL0n)Pjt09c75T@528J=T0T{R>e6LMMV z^K+YOMJxd6_6ZM^+p8L?!a(E&O#Qdwci+~2A9b7W%jCnrf*;J>O7HFOU^gq(J z^Nzi38NQGC`H)8y)N(4tQz(4CA8Cje>*+W5Ac^yM|9ETD{21uRdUQUj3KZW*QM*-m zBDg&Ui_0E{%^{|yIbR(oc8wJecr2EUcdUSr%EmlFgi7Yru+}9lopwf)LPQnI(^jMzEqt7)~IVL=1d}LZK!uEf0zZ$HK(-KOF-H}Bp{2mW_No&&#H6Tw}khu%9p zXy90BOL5TWI|5znzaHAIpH!Cp{vKTaG!IYz$=ZlBy@1F17n=F%DOmgh@VrV}@>zJ7FAKL0~3Tt+5nNe|ty}qfwi{2|+;|Jfu?$e8?$+TR>dvMSR zmLspI+GdHRc#VZ#p(CvVH!4b`>=M^IxH1|>v}~-;wh9w2_Oig>E{>(t--Lp*%?wlz z%EeC-3)T`H3va=)Vq%ZLt9>bvK@8eH`!-EvtKUlsLjqDF&c?OkRxx~T6-^v@&A?5C zrBi`oB7vL*4CH7bg4LsD9QoOvK;I?^4o&4a6P(p8oEXN4!y4)K=x(LB-T~w@pXQo% zqJ3tl`+%t|pdQt1Eq`Sy?ItXO#~kXX;fg;Vl-R>PhgWd+u-?9pEn|wj0bJC5_gc#Y z7Ci$6)Zkb^ClOEm&ZGTArV4h>0V*C6mLo{NhQgyva(_tR-NtR4 zrqVp`CVLFcH16%i&av61h!<5L8d>z4ikrFghlRLTZjlp|=?OM-(D9tC)DDjb*~N;L z5hV(n(qnsDdNHC0VPTtZrRysb(DkYt4lhnW1~n32zDHN!2{vIMm2R`@EzIQROD-SE zb!-5P{`cG(y~+E?+Y?h-s(S?n3dqlIzXLm~g?n3of4E<@bn!eOejuEy?OOmu3eg=Z;XMJo1GBshw>g;)OCr2(W}F-@CJ_; zKkumE)U4CrLjApVf#`s}IdlAaa^>iMO2iuc58xFaley|g%W-=^J)o@6cGPGp83Ef} zfM@(((RMkgqBV{=6N6YDd3ePtBOv<89p~uFkUM}g$spVWjJf)+*14~Dt27SH%;c;d`vkRzl~_A3hN3O*^eL-an)0)zK}Xt6&s;wHaD-~(Ty~N zVIgzAB?^B5dpQWu)xQv+3&5OTZ@lJWF9?$14Y9Aaq@bns>zOkmf^kgyxyx`b6e;1r zi1TiNsC#6GtL&YO8E`^b*^&b+ioKPBhu^AY84hcrT>8m{Z_*=WGAkq_wl=EBf(cNv zD@9)9SshO!gfH7|ccq!-t48C|g?v!4qE@bRvMIk$!W!}iS1ab9wW=z8GzbN?&YL7y zO>xUjNV~d|b@Ron1Hqp4{Oxy{Fg5Lp$u=3sLgVC3u?S|$_fn0C5Sfj^MR%N`uo++- z&vWp3ZH_6X!B>0;@PjEYP%yn|=J8(tc>T)C#0f^(`Q?hmp}Dnob{AaAN)mZjh8~hw z;}OE`A;2TMyR1~orx_;Os;5;aM7OXy(HTIxe8EEAeny7}~(= z?SobaN6dEPl)2c&jk)D3jA68F$iEJ9z4C!#J~N0+)+?WxvlCqlq#>b=Y*E&6dx($w zLB`xfe#IQX8Qqmxkk8DK6|Vg8iB%=1ty4`q!v8j?hxbEx$71x&LM6=tt}ql=ZwwfF zza1N2xxp2uHkGh+^9*)lwx6HJ|Hqfzr&UoY7}%o`K{b}7%y6hZ9!6LgUuizjlw0rX z)wmC8tuVrN>u3n~Ih>iVY0X7X#*18{9a0b!qC+}J^PppkoE@GDm)TyjH*gHB=UT@C5E8kM~P|zVz=;t9V z_<2a5$R}RVm0cdN(@u=l$V;fw+o@>gscQBhz>Nh`m3JOr@h?X`k(v2iJoaV}lHxu9 z6Ms0sas=M1E(-;fO^yhNZzwPRG{4nH^R?vP)s9#QvI^k6hgX`L5*-UPheg3Q<(hbk z9VbP~PgYcb8X$}^+#QnrIS2PSS$pt;pOYl^ss#`q0KegE+&E zk&I5j%IXNM{K)fP>94$xm@WCrNN9dR^mKIDiI8M`bA?f{q^0kJ8vU;4m7rQi>Gctt z^x%llI)YWS8Y```2ciCZ@z88uBj}8AvtpT@z;R<9HMa;tk9nD!F>-OvX$0wj~QVl8HPVhWaTgX)?XKJRq|? znWuI>sZhdxj&643p-QG0(eA6-80fQXJ%A}ux7dW8N-VPx;Muz;rM#!Z2 z@tVnq8_z4OanQTI@K9SRlkc)>oWlu{$B^@s*YwAbu(*m1R1yS*x8a9*T1|kiB8r7t z$OUXN5z)Qxh<;T~YdM zua;M=_`DWzcpNACJD zc$A662EZV9XLEfcn!g<&E1~Xx(%5!PZv$j^b*1rvSBgns@_O`Z5lFx(EvIEq) z{d18`b%cyK6V@Ba&n>$rUNUq>k@sLNV;(B{x6WX7*DxukurXwF6|)l)Q$yDe9c^F= zXaF@;8Fcs~RmVHJcD~3+_q;tv5|TQom)hv-edFmdnGg^*LEr0*LY?*vFu<-%au)x1 zrz*kVmEo!!&VH^_ZtV%}f{EZvT+0pGKZbA(TwSNt5KTW*6Cvg(T!q`)&BVRGzY%5T zGSgT*Xc@+eR@HBX%3kown{`)#j4TAPnT4xKkjlQ5Oz0Umm>*yOoZ&Yyl@`Evj1C?- z0g9w*pN(6JT>rXlny0Y3MUI zXG|FyT~B^qUokO|1PQ*F*@h%tvkx}IzkeEv+1I3@>mU!s)&Qf zQLDCQ^9tWkVO(yA&(F1MlD)f03_zvYjEj-A!x zKu*gaTfMrZ-|&(_=;^S!MQbCg>8A>LATn({BWcD4nB<1e<~5)^^|%VaY200Z-ow%1 zzye`fDWxzAP&2nM{)N@r=Q=8|%6FQBT3dnhU>y6Xr5`)g?ikPStf zE2R_AMS#2t7{%whlDsEFS8lK?@_27`(_-f8SNyn{@6tu!C*kf)-kvf&o0a0><1CB( zE?r(T zyR*6bfQi2e2>P_JW`$6Ij{nm>^Ab?Td z>7hS|;XZ}UV9=j)f9ena^>TCuWMNJ23FNzHMk;eW#>$?m8Yv*<_DJ(Nl~K&74FGJ7 zmmrvG{SEEu(TyABIbm0t!c>JSZ?c-$eYyUi=cXa>q4kyn27GFp zSrL_!iTp_}<@i7e4;|6vSRvQdM_COI*i46;Sd10=66%a~3ysgSC%LrpjeiW`wJZKY zL-wOJXH$K4LY>5FSL;J~J?@+MF1R-m|5F5k{W#4nBgTneA|LE~M(-h^>(kv=juYLK zN`17pfynNv$L^ZCV5-dWcTr0KmeO8R55?ZQtVnSbvSvTZp_k4dEazyuK0{K37$w-6sR* zjD6?c4@qGmz~;1^MXrrMd&s@{&k;eQ62}It?u9%+_fk9IIgQN#m6leQ-*`VQt3qat z%1XhR2PkoQH)x6g`=aaCD56KwplhLg;ygfG+K$3Qm{ZZ5u&u1?xSQl`ux?m0!Y?2$ zKVjNhHz1GCej;xKO)a)}LtBYjeKTDFT^nnH#Ne~NX=0V9a}_d%`40>Tu?%;SNn`0z z3b+1xuMh^*R|5W20{$``QP0G_G2SA0hLC$%Hdc8QGq%bP)bf<^Xbt}ql4y$LRI+Q+ z5~#}aD=ZJLr++Qiv7q+?b&H^g*E07?Z-7ebTrXPVl32O%6SBn*3JXtCgNMVYW7C z&9;qT@TIjQQSzH!Cf*a|aiS-wC6$wiiMUYOE@3Qy+Onh}f0i}dlWk$BZ5rLa5tJSr zOU$#6+5pd!Fy~6r27zQ6Q#w9@eWJ(wv9Qqe+?y&F%DS9p z^r5)q&mX@@`}14&U0?_}6RTo7tu}~k(zx?$d;+_(Kd0$#`e(2i1Rm~>k``zV#VQ~JEs%>HL-FW$$XP(a6*k$EEp~XVMz|_%?ZlF6>RR%Y^pq>3@BVhj z+Rt7)5C+IQv1{q9UTqf*D2HaTpSQq`6I|M4UO?XIsKfjst%v}^+r(}*9m(bRz%SB@ zh=3R&IHe-E94lJ1_-jFQS_0f!X+;o#lLOsEho9KHkx}Uu1m!zrQsFE|feyWO#p`zRFsnc=zzI z$wZ!r+@T0K=<@^B*s}BT!Z4}w>H39cn঎&2B(_kZtX?tFvMS0W(op{4?52m{+ zDew!w@lS;#kG7B>J(m*C)SSXjtPLCcvUoXWOy3D&;gx7?jp48QzE^SXm-fprFgQ4P znwSMgVP>hc^KpcBKKO^Li;RQ5oKg9O6!TO((CBnE_!zq~;257a`LY;o{Y25pA$R0; z@GlWKpPB|LAgBX_( zYv-#y`NiH@{O?ly|MbE7QtMs>8V@u)plt~|xLC&vo>Gp$Szb!~$`|-k4O9!|ao!b+I z{+vvTt6oDkT@uF4$#^zEu#lJVQg?a2y_y*>Q!S!bXc&#!+rg>7R#N62^dX&Si)((L z6Ng%a=RY54SP^kqI>+fadAQ{8sYM{oat>Cpz3J9xp%l{aT1 zO}y-YOu5^7%5OMXNh0s5#GjdCCHw`UlKUHJWgeT~`9!>`prV2U3qsp#qRxZ`hTb=? zlo!~>HDtYsU+(3U6>cz=6}wJfN+0VhZbmzQUe!GU7o0o-j|rfJZv~v}yEn3s0RCl0 z9f7w9UJdyEPSIfj>Pemr^UelCrZ zgdyvF6kCqvQanh8H}X_14e9l6%<}lzX%4%Rk0RaiQLljwN5ql;>5Tp_pzOhXx&h|v zUn}j5AY$U;8PU^MFD##C@2VyGI>2B;WogmsnbBg2^5gaFa$C{X3yQFgOG&ynIsees z0MOgd)3L$ym6Mc)bL@^4mpU4GrFDZGy?5BSJf;{R;X@AHMW*J;+p20 zcpJbIou>J3=2fmH{JvP9L)M6*0KhDG9t&|Kpy->qUGJYJT0b-n={x+i>#|n3j#jTs zkB9Z-x=xrS*7=j+?^>lvECE zOAgdk2A!KVRFn`(Hf8QrzDwmQ;nGg)m^Zg#zb_BG6kgVAjO*&+X3GY=r{F{i`{C5J?lXk!i^d*lc)0%dum7A^*d5n5IgIJg*9$BIK5P3MN>TS;p|e+ z|FkdNXceyB#cg%Fqx1FS_Flj5Se7XIdp6a)f|2vE9sQ%on;yGL!HwlQ7HA0jhjcpe z_6nu;Ek+D)_68OI$g%Aizmf9%$t5n&hlYfXZ_-w!^wzJeZpnx*qfWJ@+X$uSR)KT6 z3jz<~%o&nqpwUTeOo6?r57AtD^PAM{H<1ioduhWvVumu{f%0%(&UD5>NkQ~RXFGad zAk}{}s6?rSqo<`WaRG|2GwxZ8PiKsRL@9i#*6y)3jGi!5MbE3;A3^G<9LlqE+CS23 z(;neT@|LWA+l1GKbbUa!!5sQ?`-50UZB$m1o?xP%^!3&T)RuVskIg!XeE?z-miY)g z;egA-QTqey5%{Br^O?qrH;?9ma$qLthwGgP+nIsxYij!W1_?sEiTMct<50qc+dDgS z{W}uGc~qHIBI}IrR1sJrdaqO#%x^CGwpsd&9*un>XoRs=ynFRr_o@xzF2;EEL~}ZRiYe*6<9JGU}~w)J8Ksr}m|zqy4z*msL`$o!+`av2X`EVxlF*+4e)7MEX=ol*}chA@ToxYqg% zzg?LSjuB2sOn%A%F3A&w%Zj$WV$}jh)sM1U+T*g4Mhg?yM1`X$^rhHqgy~m|z~IL` zIaWhCJin!>aW^apx*lXXnQHf#(US>wU{qgKfxspO0%>oMdBTFXt7* zj|DX~`X=%RJ?rT5FDgMDtB?ZwCi6=W0&=$1ZQ4NLQh&OHy3s z%@`E_R2Y=<%#v+ThW(@^Kl=yB2S^0+nWtw2BWjTQgQaIW#M8p!#D3?HXUBG0QXlbo z)Kn?Sc9G~?2)L|Mr6|g(9EDoYKihdPwD$S-W>)QbGnJv!WFbmowhHcO9)HZXvwGRZ zKDtVv`u_Z&XP(!=tEjPx3!UL(0m{BhctzlYMapk?SAs>H2xjbzL=!Rk`DOJb z8_K{FgxnMu!V_(pW|NpRUXjii3E=i--=%=~jig)qo9K&>uF(^HRg$I^n>)sq)32he z+^d!_N+v$r{VQLY6m6!CG7%KP+$M8g3gKOIdw=afa^HJuhP_Ro+RMcAE7EW;Hi`1O zU(fJk-)1v!z#9ak)F(Gm-xoQ+b)qhn8f&ZlLf{tsbL2}1Z&+SoH6LdScuelB@#dl? z2Y3=&?wr1v{8d?Vi%M%tD%3Y__RWd*1lMekq!iMNmsGfY9sJf;vJ5vyL*_Y4JMOqb z)W$<^q(!t9R9w$9_inU?Bq4qJ9en$ovl>-+5n6LbkQ=tiE^cLxI8^a$s1G91J}W7^ z({bLSSFJm@I5*U#I;M_r-+A-zQ_OZbf#~&OSmi=$#o8jZR6~kb*4uC{WRY`$fQ~9Y zg(W)U_MGPyC{`2?W5mQX(J>vYko5ew3&8DA#b)YY2p%;ph2zS=jNSDffz~a`HacF} z=cF01=~dgHb<=T~#4xq-^-`fX=L!!bWi@KwYh{Dr&@g^zd$2n}=MVV2gG+UHftP&% zjTLq729K`+>bAztNLI4H*)FBs^UkaHcMZHg(#?8{*>>C=LGmoI3j@IK>?v%-if;FZ zYbVQo)Dfn;)5qg4blkia5N8gXnJ!(Jgs^zY`i+DF|4?(A2EJs8DX!AE3SR#oD_%PZ zY$a>rn1b_38^c99`f}NvUUOX*lTtsH$4gkIw5AsuAWRV`D4~hXJUsmCiIjNLG>2Ti z(wHAS9Q#`abfoKPP9+A8?fq|}@+pYmnXIrFeWL=BJKLf4cVO&|=cJ&6NH#M+an3=- z9ZG0N3TtQ~gKQtS-(ay}!L!+%tTz`TrNxZOb8Fw|yH0W$JsyE>^y`J^WF_WyhYmff z7FYdPr_8DLnI7n$08c7(z25u#U#Gwd81Go6`ENepGBuD{que_p6JzQ@QXqM)A+Mvw z5Wq)f9Qayw3`NbTZhK@+Kmii(yNJ>TX6Q$p2P)$lfDcltIKe-OR9wH>)^J9W#6yu^ ze(C{T5Z8}Vlf?hQ!!CgZ(=H~(eRe9(bE^UWp?bZ%W;Xw~e21l$7k^GL$WeAxoxe05 zv?x4`%ZPi;O@)Dp)4{XTl{1m^Mk4x8DbW(19TxA7lbS4i`!I~rt0p5zW%CiJR!vZ< z{J;@GfQ~AB+f(9=oC_I_MP!!E+h#SgM$-I_WXZ6(cFRT=5KvJ89~Z0EcKpD<=>b75P;&E0c`)|jKV0HRdbQ?)4dGX7$uoQ zXYRU}|3--tr(Rqc14Fb3kA)#tk0@C=b(w`>z+{XP4gAVPx?z)2t|T8Z*&!Y7II{4g zwI60X2#v zzvB+rZLxj(u0O}kF!S%+#Zp}llii-DBZ3iEppHBJII>myfMJ#6u`3qdLV6`4K7<6J9+2+B>knGBEk zEOiTaNu?RwdexlSmK$mLI_5lHoJCai;^lh`R_>leQ88G}MuAq<*xl`0&Pu#~S9|ty z`u46}Ue;q?J5+YMyK#)i=1Y(h8U_UTnIMulHnj!Y31hm zK5xRa(IRh>X#qh45{O>?U{YLrk)9CGV>;L&w)g3AMot|83{>9>nkNorn4jPC91`M| zFtXVvgumUHy{iDovm>P}M*47aK9(i`^$IvdTpul?}A)Iy2bk!cK zA!0i&JyKQ>kT;^|do~H1kr!>_KVmThD?mtfv9wdd5$PB`Yeh?BI#C51CV^-puJ(Y- z%LBWa(VvnIRS;&g*(VUC2uFf3wHP#cW^MndxuH4ENs+wH+fLR7%L2 zJ`ePUEXh;Z@=92Sw4Mvb$a@m_SPd&|g3xaME7tl=y`9*<8B0r6g6CLJ>I-^uP#4wS zRnQ9sUuTpcTD^McYsP_e4zZZ@UF9j~GpZSx!q7Q30redC{cI$vA71m>yA zPJIZXQ9mI5=_WsK#lK}e!l9i|HJ)tPWLeTO)bD7-4X z(bZ9NpH2;cb}Xk6{)+l|?HDOlIDgq=-yk1ELMWj6EXQk)QI`U~Dfn&2W`5;G+&OOU zuH;c&Y|d&paZt0n+a_Fm5O|bkq}^G-B|H5g-hU^t0S4j6W(31)Hq?{-hP?6e(4n#x ztwAuM)b&WM5i74Z2EQwVGf`@~c;$F;htUzZ+S!~#Av(dwBVoeIO%_KrZbsOLETF@! zs{{;Xgn53%xZd0(4^o1Uz;hP#%>$wZ@2>+nk)(&?=f8(KWSuUY2_p8WO&BLi47D8G zVl{UAIEcV;H}yGv>H|96+YDH>_qcYY>pCC6m+xU}%}PTWia=5IS~04^_{x^34YDLT z6i4aLmy83eWf6t#_0oAMI?`lH79@J_b8B;|myB}cu8bqv>SLkwJ`M*G z9nHSd-vbde5gS^|f?^dErbUaMcPXOnVoLcHesQb0^EGY(w^C#U&vQ>Drt2IK;I)lJ z3L}3?)XX190ZB5}H7Y0JaE0>+x>Itu-Mn@U26KdB+;NMn3q$2dNetG7af3%^s{Y7Y zCntlTlzB>?corYS+dEtu3L7Su*j7|>+TUxVqIC7mE@|*ncY%(*l1_-w6g!TPc> zah!5XU88XPK=v6u^&es2yo?-yU)XM3t5~^U2SCXc&dOgiCU{ao8~uE&Bm7mX>Jxtw zfd3bM$6q-#N`S83h(^q6?*rZIpnhLnDaK!yta3|MYH}=?L=W^|q(KvO0-0fnc!C z2oi5D$US)pENe)v!ms?`E!uwA0DSD$L;H#lLvr;ZEo~m`jsD~qgnDji3Ki<)n)y1I$&s? ziby-O1^^f&?W1y^^;#p}dvG#*e{bpyd)D1wn$3Sx`ruC3^mdc&K&VST$aAjAl7<73 zw7ME%;yPGlPGsaSEi&x`*t`RaJYp?={})aWZ86G@{mze_kYgb4uiJKLbp7JI@K~-A zZD8pA-;6SKeopeV|CE_5pv?%66b*)0m@wHI@M7;{V7b$NKt{Mc7MlFZg+WkeqbguNeJU+Xyb=*W;!gG zZK;AtKOtBBuT%Ob!#OYTD0Mo;CdoG4&^E<=tglxccIk>^Z%iTQJT zRElGnzuzt8NUn={$RY76_t0ysQT&A*atg$XkVB@PO?O9V*21=F-V!7PrFv@uXZ{J& z9NBGO+-UhFh?OhEtf_hiC^lJztJhtytE>Fopz(=!A25XU_v^LPMESz|-9x~@f9>C4 zLM*)wc%E~l1#Ef`fCqS-SO8(J(SSng`DzT=yL;X8Ybe;-~oN_iT zz${wh<`O?0dB8ZJT54zdt_a>LI!0ew2s~xT|MlBA=$e^y5$C_DJj)YA1GL6VWN!Ck zNJ!%oP$CX6In0bAi}(MyP;8NrKz#o7Z>jmNucDU#7PIxqi}|1TFwyW?U-nXZ=OzLI z9v;X@C7_^5g5jbLec3(bw37f7WhLZqH?o6&}hY(xY#3wGSmEb*u`h7 z9&R+2^gQNoO0Z2i2k<*#Favr5^2F&YQA&YmIh~D1fAE4J>u16_r#pf~g$t@V`bu5- zM>PS+tl&@Dt}j$U&@F}Ck*`v~G+`X@2wP@l-KNh< zo~~4|Np=oC#nItUKCZky@AQ!5c{s0_F1leU3#zjU;|$(!lN7vL5Fn@S#IQ|2f(Hxbz%6pVW}*d-@w@ zaj+P&Yr8xVbej;q!Qsl&=D7>2H;qs(Lr667JdMu%Dry_3!P2Q@$?zzJ9Vx=>I*Ga% zdC`WwR+rQvwf9{L@S01excnO|x~nKQXB0$DvEz}lkXi(#a~2I0(^4ZZ_Bf!%NWkuw z?!^A&4|N{X`n0tstPoYZw*W8hR_1J~oJwZ<8lrf+VsB#&g4)2Wt*)7;ka86*jx$LoZ3NdF)09DStKZV`cd~)n zl?G-D^=M^tLa~}|GbK-|6*w@<-Qhn8!Qj2QjaMET9ZapygETr*(_dx4Cg zn$lh?E^aul+8{xIT`2wD^YLRri6h^c)L2slqG1ki_5-}h={y=o!4@xHW@x;Xmw&!Z zh>+3kJE!*<(N>6=o%NcS&@VAx<|`6MjHakAjqV0f(v5Nbkc9dvr`}98qH*aO_Pf>; z6Ql3#RKBYubKVtYpsgi{%;lFh-D41b-)`%Cyauc~fA_tA zu)mk+Kzc?5jO{D!#4uqjvw6Utu0ZvAPeL~!`f|=7459r6cm!}B07TyYk>k|ohY?Yl zG)91+F!D?lJei37en+)=LVvfaB~qwn>)k^_c;=nwU2nTfr!v)0@cdO20Ty+xFydP- zb%QLOnc^|jl*sDBkk-sW9QZzq14%<_qb!>9|9+}`D5T2_-0g}7yM{$Kf*e{^>8C!w zFfw{4y70hMz`Ri4SQJiq=>t-f$3T1hki>evbEOA9;~)~4e%PD9T83afkZ}-((`SkLl34e3 zkyoEI(zr{!a|lv&bzsm<9z>m0=hUuT>!`9MBOH2$`2Lc7Rw#dEVGgk*Pm%C)r0Wf5U7?8WN-{15V&?F#}pjWw*P)7Ca6qcD8U% zf<(&8$abu?0n3`R#&F%}8-pc9MO~Ldc#@~aZsUP-J_gvCdfL5#Pun=f1r0 z#@r@h>Dv()Kyf;ec60VM(Ei^j@#UnB#`{0Z;vLTUF48=2D0LHw8o+en zZtwKUl|y5a;U zMv+##sunSrvUmDJOTNYRkG&N4)?I;r$=V8dp~DuP`P@)jHJz1O$mvA|bEd}@1!9HV zvK^);(Q0M}jq}8MQU6w$@9fMC%T~TD0gtx`4gO+o-|xRv)RcKm`+XVr+4o2@%IFDR1}WvdR$8F@pzpctRQCEvOC8$Yis!x(;aFVh5!_u<~vJTBgk(1r+5E3 zQLL~I1{@q6+axT0TMT$My{ESLLC^0Yt{?Y$9lcYwHY86WnPQU(6wyvO{tj8nZvMe3 zLtP>C@2D4%pzTOKM&qElr=5C#g47{57j6KFfZ;bFkqQ#WnQHFHrUFZqU0ME`rnfU(LPE)45Ph(aPRI|6$;#DE`GQ(vS&aAUqQEEOm8+@nTW~}3s>66vw zOZsDt+3wj3+e9IzM=cE5ot&DV@{k_iEiQXQ_(B)8iai=FTTV(&1ym z3YM>HtqhlTvY^hS)I2AD>;|K9t((Gh3~BzeGXP3%+eg@g$4O2EJ=HaOw*kt#wq0On z5_NC7NP`u7OF@>dev>c%Dg`XPZCs>`&(rnk;zz}p0m=e;PJ8F{30}XaPURc1<+Hsl zG?ynd$;Agq{$Y%xuGrY>Oi$$)0bYR@5TD&zD)~u^f=qWok z_K%zY`;qoPiY8c%o5O@_*nweJ62obY8`7h7Wh#-tqN8(<_+FkGo9=DsAPg@*Nw>h9%eicTiAhWUi6K9 z8G3@N$lM9N=h3+6h&TZIM=OxNPr3p{VYuuPrdmv;F*xU7BFp^ za%~(Cg|=U?jMeirwq4j5n=5qG>r%BSX6DE<_7}>>qBo7^SoMaexAHjTw%!=Hpy1r7rpg)p~Y5 z-SVXy+7FXs`)m2NI%Y+H&)Le z^ahCD3QOmVYK6CdDY?(vozso!Ch+p(MwAlikmZ-xmA<=45pw9}h#?jl~* zJ6JV0gG!6@XcX6bV(IEF{$@~?6Q=bNBRqzmu(Y#yspbu4{9&9J75vz_YOstRJg*~{ zbgh}wA+DhR43^tx@DgwhNs=7Ek~ZSZeE3jtop z&*=1Riz9jYSR>Tmtw$KX^MdEc#tq!{YN|-wo+MUoSr4u}YE$3zAS5#EU^i>C!`xmU z8zC`tb7!eX*!YRv6NHXqHMm5g;E5-|nHLST9Dznk+eB3@RpxzO1+rmVW5Y}aAc)uH za$63}z_cA8ytx8)L-E^CG?{}|;Cskr3e)=fT_7XRuu(Z9gL!YI&i^eV<_8DAy?apr zGwS^lRYYdVxVmh^NIlb1*`Bpg<$UlxcK(d)8@l!dbp#XF_84U3z~IzPp|^>%A&gau z)cJ^VU|N@fe4_pAR?ycb6itl%cJgLSO%U`6Wq-;H@9cc7!c9v~oty@e-k^PRN~r`8 z#Tw7=M46U=L1kmWZ=W4c@T?aO+9V||$g~l{+noW2<6gVE;zD@E$V-OO=EZ7?S#fWN ztc1*8`eyzaMg8+&Qa9C+!oWY-??fd`pE<~5PXpG+uQH$7-Qw%={z5mkSBTyr_Z+DF zxpQ#(;REm%M${qWviPTg*0E-&zcsg{V~u=r?hVeb7SAqzzMv7QDqJn#DEp&n4)QUG zm4miL*~N7PT9H0ifQHeUitB7$7qt%pCVW7)vjxEh^khE^AtO^kWQm>taIbT=NNRR3 z?PV6CPAO7^^!kkq1?l{?ma||N<$c_SjRzgp2aI7fUq2~@z4cz227FF}KPBt72K|PZ zZjVHE(&jXa{Y0MSqf-;ALe*CR9Fyr~|IQB%wuF|zq7Qqc$X4!&;(vh8v}k6Z-BO4 zp~|ygS3fjbT|opq2sV(f&?#K3J=4=W6oPm#6K`sNlfK?1$C*KZhy|)!Hk1~ zmF2JEoetSGIga+_HDwOw#8trhY7p8xc-sykV#;<*ved&GjT{SEIAa+~(iiT$7t>GVr`}9x$}E zT>xi7`uKdgZ90XP{ENGTldM2Cj<}Qa&qBF6FNV*xEt$3Lng7}lyn9^0dyf*|-Oy&s zMEg~S4;eub9%LxREB)*XRqJsziAYGKw1ZQyp)QoR=CFUSxtQmOAhKp4&vuEXxG&iU z+SQK1GTW$lP}Pg9k!>P@?lu2vz?UGlMnsVAh?ewJa;fD;L0;S}hit`Uq9bJF?W@QA z5PI1dj!C~RFm4C1-3@j5g&)c)vY^gcpgOV9T2hhTmRPUXHdtUQVJ_|Mp8A@M9`wT2Y8|V5n7=EtvTyaZghWVDh zsk{{7eA$XQQhU?PezARWtTqC816+Z&NY8&caYeAoNT9Ss#&Ouc8A)qp$1-{- zlc%vUH)cC!sNvIq>o_Cr&eY>SZ*RV-qva)vTpk#tn))w+B`e_a)8#^}ix^4*Rkzyi zT;0*us(g2Vo(eZlR-?Zc0MP1_^FP188)$Qm{Xg|#4Gg;yZ1DUeUDCYd0vCXh^1zF_ zexmAx4-)tp1D#mmg8zUF5NG`jv2$PyQt)#Og1Fu_9Ek4MpJSz29X%26*IAi{e7yqI z#v9CZr`z(4h-HFAq6{5yTmuiFQ%adv0qc*mncMRoE$R)GJF>kxVSX-*1@$#qXQGhZ zi=V9JY6;~|I`TK4UEE9S94_Bs5gN`lkTgJBnS*qzCtqrq=zTvFq8E?jy<|M***%B7 zjjM~r-7X>RJf|;(SnK-E7yHhaI35Hi_G2X6rn~B0O2rBBu%a(J&SQ^m^yLaGUX*Y% zm`4$K@Ye#-E*G2s8EUrU2?R=M(M^KqL4K(clk8R0pyk?iS^st5?KzdQV-4-OtY9ya z7OKhP@8vb$6M&o7l%){(fP+k3gO^jI9FXix<(&V|*#hQ6TTbP=RW^G<01|R;4A>`T zyU-V9JU0;;1Zzc5q>yd^d%V>ATK_$d$lrYzE=DP=+O5x_Rd+!xC(_JS+PooYqCdFg zlc=1QVEcevBWsiQKEw8NHKsR}2r*6+#p>>VmtDg0ZxXX?ID1zNN0Ga68}X2orc;N2;SWY=bu!Ep z_kG96t^2_kX08xgLVm?4a}`#TGr5EX9vP4Gt2>4Ol$Cu)zVzUr-qBt&wy!h36}(2s zdLzZlWx%O+Z!tPBt(6P_Oyy+f$(am=8#wLRRe2_1=}x@ zZUs#Kb(~bQzcXhLP+i*$radLP8^O%6R;JGax7&E%wykX7OnHGxgw+5g_2}l!-PM9e z*f_;}>Gsrs#qE11p@e*D*_}~ZP0N=V|7q7bV?4!kF^x{JH^|v=LuestpK=CcyC(qp zJLi@RwwN8K`4Q0QhPe-z3HK;I)2f^t%A-w{HWIVyrhMiKczOM|H1%#Q>qUK|`)Bk? zEbPyr4E}|Im{#ejRBqtjWb|qeJOk+4P0JkM0zzU9Wg1L9$rovEDX*YOnD~7j?5+!c z6O-cZ-{PDu;~V(RY&Bx^{(j0nV`^(-oC!03evzMcdtQM$EmHoce{lt;ImFS;SQO;( zHGx0eCv*=0zf~3&?GIe7AX(hh82>-Ajz@sUC5!Cmkf>1ReE{I(j0azm>@$*I0G_!~ zP!vhX%x5NJUi!qY`oxsiN4YwjW2u?*kGBzp=k+|wdWiPZUuBf;Qpht2|Rjj z`=`vG!}3d7Z*3$>pz)Ei^URUSXSS_pe~f;krV6(<#(V^MfAlaI z;MUW9^ek=q17MrWlL?-wMx77C5}mY9rXWC6!EXQ`qnoH9pEe`tn;QTne*)@{N)b@V zf(Ck?ht_!%=dOd-oz-6hr=!cV=5`;0#u}m{qsmTd65n|Y%C2{P9ghS^VJ^)^h5gQt zoq2elJ3())5_CJI|LJxn8L_}J^pP_@L)3v3QXG zb(=?b3nL_lu2M~AD*5`#K{u)r!YHaTr-v;4_&YDN`E~q%6!kfN_kbSwk!aQsJtjCp5sUWiC57ZXJ>X9 zP&XIGXKYiy55)y|gNx6gEKeU;c`>~?Gl^Miy6r)8XJIKXV`}CHJXSz2)Kv?_C^)Um68z;SPic#+{BtQDXu?H_TRZkqpMpf zZL55Fh9>T7dK##uGV%CXbGfb;8zx!xr8i#}p~$($k)kT67TNicSS_T2*5iP7@qrFK zyebRdI%+%;k*+>le6$&cG6b;jW`L(xG z>B2yFhHBiroZzC1#^_Q77!=xj)X*jaV5BuTG7H1Q%;um^Tkq!=E2E{>@B*!wWgZg01@)45$7XM^J~ znsu#9^=(8@h-v+l*?!STF0cDD`RA;?p zVBj2~a3v@kcx7e#Y_xNeZYg|_28_bOvO9!xDyIeLy5HS-q_qSav}}iqBZB=%X!bVTQNde325vmztKaxTaZOqm1WE)th{R zaAQg)&8$&fX1f6sSE{?RISl#pb3lWbPLJY;<7YSw3}O~v!ZrONCJmLqTppwyAjQx z2N0M;QO&O9PST1YKve{lwf%|CKUp^Lm;Ym}V`cWh}a_3cz>si@2sD40UlZ)A9U! ztJ3nWma()M_dWt&I}V(=qi7?+(whpkT1VlHD_Q0Pw#oiBo0V^Ht+kypbNTpI568k% z4}mHtHJ@s%|Nq6_dq*|7ZEvG{TR>4jMLEPI zgVMHy7#I}UrI%q0q|FS|nvB0po%wr|;Ywd~ctBZ>OwXO@=};NnDl$Jzd$!T-CsI>x z>@x<}wcV5OAH78wo{0?G^qR?0-m52@dQGt0pn^yo0IiHcLTNxN^bDW@cON7%AcjHx zF%!`K88nbxvN?j5A;AA8(DTSf4k3V_1+~!5$UhZX6v-BWU3|V z^w1(=5U454J;#}Hi^0r0eU+5OiUFD{vG6tSX!&dJG!FEuSenK8s($$UZViEU42qC8 zFkylSlr2EQB3UG?AjYhTtv-!F$2G->maK_^FWLTEgBakAIbD?Xz-6>_u_T-5SefjR z3#)-J$S36gRO3V{IJX024(uEX#jml|IgEJ*>v*i3UiyUidyM_>=s8aDC;{*NuVB}l z{up>~^_+&(u13S*zoSS_jiz*+wLOmzB3XF5ux1$qEy0yLfc-O@Mz-fWH{ z%_nbw|9y5dxs(DB$IdS}9#fk77Qq`R-ooUX1hGTKaVK26>Z=8G$Z?^S5BPj9$EHffRjmyuCDWPaAa6$AcP4;qMs-ya(~ zudT+rmaua3aNX}S3Q5iu0N3s<{p>#GN*}@Xdi*|PN}%tF=F>b5*2z^ux_{5%{`cmd z?4RuS%&|2_D;XZF2Ufo#&6GgCI6+jEtx!L#)ciNTxIae)2hs7*J&zN`=}i@eH>K!e zIf*hCZ~WZ)`=tL2{MFz@tgJ2|_n;(i5nn?CDk^QuHGlF6?p7VshJ)ziDp|&ccB={j zA?wrq>~p>(!$qD+5oN;9V4xGH_AO70GJbN5tBx);mW34nWxh>%zA98dv0v7|XoqA5 z8zruKdIcj7Kp4=h!00Kg#)TknE5tN={SRo-epoE?zk31v_vQZuI8r=@8-17+KVejD z=fkQC^zl`fqiW3V9V0;YlY*!EYyX~O{?ADR?Jg)bS8XDz9E>b2suM1`bM+|bQewkT zHU2~i$-hSbe&^AybGvTOU;W%Lp1Z0EY+$p&^e3HV{xnx}Umhq3(ETiU`6poFzxkb0 z4L+1VX&CoEUI!`^fGK)5_JKX-Qm+i?7+pY%KcTj}}LD3pG{JpJc{$DFYLc{e}$ zkar4@!-l_S54!o^srvWV2^1mx>|>tNc?HyK$*acwT%8RtI4_z~y#D_3e*ov4vVcpGrz#huTb?*ZQxg^`d3clSE%}@1_R{WelatDu>}8OX8s}q z`o+xr8OL%ozcl}UPeiUA4%gFpCw0g~&L9_|rnYkF-LYT}QloVu*u0z&&hSb${I$-_ z81N*qNd8)u+Gu6Eyu9o*BtQXGYu%HUB;iJ=!Ov|pm>te@qYUT4Jze96T^PcH-+n)F z?*5P;;*a_@~R?BZU&;OI-PBpmD{`9jP-PlbapW+nu?Vd~Q&yN9c+zD(A z<)6QZzj&Yv!M_DopE3ms#r7TvJB{$tUgW#(voRaqYc@g+48ODG2~R}5g7m>`^gw~3 z?ma6-h^6|=%J5> zxOpNF#J(nM*X8X4nf$%Y|5MFl*9Df#X}eG8R1#saBO452rgdIRSLr(vUFI=P5;)D}ga;$r7g-ZRFtH00d>pQ!#-Z3Du8ng`Z>1$Zp z21x{^QXXo9Q8YYVOc3*j$hlB&26S;kS z2Mhq|JS4-R+Imwaw2#Xejf%D;0%LGq5RR&HcINw-2ehjed%bf@uw;Ht3ujtw$& zP&v5X%Taj^szNEVZ(uiM`;91e!pIwAnv|`Q>)GF6dqXK%Q0b%NdD9P?guvI*?Uy`5 z9-O^j^yu$LcIw_~6M?|$Azb~4TqIK)53{~vRS_|H2ez?iNi!Z1_(`d@7?JBBnS+DK4!GrSUo# z3B_B7$SifGgl#PFDHi+Fel^e4NRi+6d(((9q2?KHO7ZDQ5}3FX!PpfoanI^*MDcRl zdwi!RA$R8yr%uDmD>4TaD2t7`?pqMDhY_dVZ7w;e|A4irY-*U=RJUZ*g`rYdi8?DI zpYCb8WvM6J-)7DS?P$(l`hi<|$ar;)VdCa!?E7O*;{MOZA}8j)ddYm{8CrX)Z5Hq# zoRRB-gG`KpbM|p3KtYJOc*mq^AwN~hucrXN6`v~U(_vlPb2h;Lg*$#gc+{2QW#NqN$Q|s&$=@HgsyZ4iq$`AdABslYl11GUlvGG1qH1 zBzX-2kxodG8K6wWa|_VzuFoe}MquADJJvLF*V(Z>tK(~_g?>8&v~1Q;F5bVD+Ol+a zY(g?jg9RXqvzAleTc8MK-BT?$mQgnEx%)Yurr{*B z>vs}cz`I0qu&Gsp80xWF0(a7?pHl=&fGfkjn8lM;-G*0#K3u|YNG=<;NJ|TsmW=x- zJc7=NS2&O)JS}C3y){fglW%*+muzNtN8`fQUcP@RA0uv3(|i`8SEtrPsxa5Zd}a1N zBiK+$;VohuFA++^d<(h1m*RH$1B;lBk)Yq#B%~3?Q($Zm&~GhEKmlHy)*MJQ)NVZQ z!y7-mt+jNSs`qgx57r{yI1Xt7annO|Ys8JQV|+Ja?o;p?b?8%hsuB9rCG~i5Yr`y# z3*V&9EiGI^AKZPd^m^+D)@H&Atd`emLN0ipu@I3q1-w2N(>c;ilngG~Bv~*YJVcpn zSbe~=G@M;sVZpUVwRR^4M=Gr4TNuR8X6a%2>~d!h)DFW&d{LjmtogVorRUMJJagavFb`I(NwZA}!SCZ3!0 zv+gk{^xaEsN{Q(wVrXcIv2nQ>oUlmm zWqw;sPVmx&O@3i$ipzIlj*SWrCkT;D8{1r0y>$@zwmL)`c`$EG+E9~$I3ecdONp5f zzx^Gw#yZfj*Z`0rhxp4p`qX5w_WDPNGyTjt?=#JuIZY|m%gGY2<|oM0DCD~`zJ{Aj z8P3&`wQkA;*0k(LTOC@pzF>*2jFrl|{x6j(sl_Y82EeB4Kn-how#rvGKGoPcuAyO> zfs~%T<9;o^mD$!r^>ySS8O2RQ;s;!INH?c$A>m4{BaUgkKLQ1;h_Z4gK|>A~w>)Tl z^3#jBlLVBu3h1*n2*tn4Sy{>K?ez2zpEqY%mRl2~SK2n%SSTm-Ka3Mq4yC2Cv|7vY zYN<=`(2OJ9?EEuxC4fWqkLL|}1 z)Oyaz`jFomuCKpal3pmWKyd1EetjW5fm;hEiB79(yx%O46!FZcP)gFe^&DT6-{W5z zmS`5B_UP?pf_uJJeuFB+0a*DzyoPsvJ1H<5e41~J{?uEyj(2xv*V_D6TYOobEj#t- z23-)qrgD|2g5?AoERq}NU9r$^sx09ccG^K&Z}L+rh9ok0YfI!_Tiv8*!$z^cRq(`k z^8;#6yS2*B7F-5Pu44=BK$zA26Kn_!q`^FCI$rvXEDokUJkU^l^>km>UEu`s%Au@w+0LSxdaXHqKtcMdSP z%eZu4p|Yq1vA?_V^+lspmQBn{yxP9H(=c2fFvH=4ZS`FeK|r02`uLJx^9{1!XbXh| z*Fnmug}T-OkzVlS6jFO?Zot8k9}LU&1zm(jHMOzgZp?Kg+>4M6aEClPIhJQmaO^M9 z5P|aUnSCV8#F?jA(Vky6%a;rDTpBkUh1=T&M{C#Ui=nZo=U|p><#??Rmxa`&_ja^% zC*y|Ju%@zCuoA+7B2v9ys;iXiBc)QG**)0{LP#8=aC%0ny`1QGf3jMy$b_r22&un~ zJ8YtPpF}sbTyra&F(xjqc65ryUj^!!o+JA5pTZ%k*Tl)T#{KZ9J`3_{?FXswH+QHhx$Q{w{GNn)p_o8g7fTmlsdL?0&$CdktO$7}?t!~uTy*Vt zkP!8iHpR>baON3ZOsx*JooybZ8Ygs)AW1AYC3nS4z}e8;=>qQN0J zyrG-LTf$EFyIpTq<(0Am;t7z~^RBC0i9Kb7jk_z=D-%-vRdQbS_fnBtR!JdDxoSzb z?`Yi=iAYuOdZH*7xi`~*+z?^TQc5$jQiq6TO}!oYLN-WOX5lC1@zG&9e0#azWsmsBPpkJ&C9oXBSG(O>vu3tzfM0U7-Gc{Gbk@$8I-To@qLvv2a|7 zx_G%ae9vf8nCNm`S99RSbXH>Ss2qo1wHFFhYTq%-SY)SEEfKLIXk$UCl6vxYb~C?L zpnLKLoA+3`OYr1x9nV3hW@0kiP7XgC_+}=>nxH@;h8s;KhZ9I5fT#LkH#3iu=dwx4 zNgrl*QUzJz+Hg*Ua!&g#cPtlJug*$vpeWr2%(o>BJ25mw#qBV$UF8*zY6!Vwjrbn^ zj?-}5x>n>f!(!diBmOOFck?gv0sS*-{hQwr;V9|ZM5Onx(U zoTnI?Qih-e>dAE%;Y`M*InC9%6ttTpuXm1&%OEuw5o@E8xch^8TOa(lcTYE~w;uRC z2wDNdN}p>LUeB{F6S_OL@x#@@g$}^HFU@Zrb29qA5c5Fp>3-RypLng?1mB82pDTl3 zVY?{T)Au%GbxD#AC)=K)$1PqCpTNRhQFXd!(&RwFiI3Dqzha5BqMc8Z-%EbqwMiFY z!Ixyy>kWVJDSj{PH3X`>1B;w>#8NroJ@B@4-sqX3BJ^-BXIeefOI%^werUt|9FI*d zy`{L};NEzAUBg)2!ts-g6M@&f4xgZz!p9mcYA1g)Eycl;6DcfIxuRuCij0IYT_Kx! z%_stl(?7>!dDuwsm9>$7UCnW)bkRBzTUkgw%(Ntbe{U?i!+d`KxYV+niRDQdM0(Qw z=Ia$~un)qH`_BcKXP$k3D*qgafP10?wwNl$uVeM&)24Jn_^36kdNcO^30_I-iy8iRT*$c`u5P3-eNDxV>)6!@!`PZ%&OnO&kw1iS}(h zh^J?7CbKO1eJhu*$e`X`rcO^fIAGl}8$gBl;Tt;*aTC(RZV~xTA4&IwiP|+#V?GTg zu1b)u+XsDBy0ZMWoDy>|uTebTyPWCqTYAote%nRT6LMh3=@VFK(Vduqg0TUyNTWQZ z+YaO}-JW?}n}Z*T!~8e8ioj7{w%i1Si7&+U3kMb9uGV=D)8k88-vTqFC${Qzi_K7{ z)ksB~0(6UDg`hg7Vd_euih@=t5#RS{DPfA9FRDM{eD0^7Jk=0<{}QGtMK8YKU`kF} zZrpY;+a=R?J33qZM6O=gcgArlp*1}q;He+g_+TP72ZVkGEzf%50LJFa!Uzt)_=|T< zr-i-VU)4OxNt8y&CF>Px^%baZXP{dpin3`24ZE2d;ZJw(xkl)2Edr&xv1%#RyC*vO zbJTA5c_MD*Yvs6aXUe^hmKI$O^*2p7+gg#jJ%X>-k38J3u&?6$S{rIK(4k1VZNHJ< zGYgdGavR>O=j=)I!d_b{`q+{rU|lrb$Mh9J9OtQ+=Z=1$an`FxPK~ulcY+Bom^hxw zo3^+CMLOq4A*+`sqhwerL%=W&4vi14+nLSrA=Gxpa;5X{CX3Du;-v_kv6h8%VslQke@~$iPR>jSVf>UIl^(<2*MbsqpS?NKyk>|Qk_c9_MqfzKc zCJ}rzm^|mt44VTT(MCs9{imV7(n{gaF8R0&04Ya6bTB1v*hIX?XstE*_KuHU1 zBlTX$uBze)l5`TkL%G1K5}J`*;>hE|WL;EaP?zIF^F)@hYS($U*{^n!$hxaJci-vh z3@B^ia|jP~x%8uph!Z!=!DLv2<%HW<6-kf4EaRCaXs6kRNHQmOdK@kXBIrwBQez6p z8qfpdbp}pc1H^Cg_~HIW*1JNJF>BGWH<$az=x?35_!Sf?;j#FLrO}YfI60@;n0hKX zXRu^bs@Y^^r38ufHP2@cJ~Lk-=P>C{N|Rd67#m=I6~0uL^la&wD){(B;xtynsIvI^ z$Z|%B1KR!4Y=3t67!<BCA^3ZOdn_ zQkrdZ>%4M=%9L7YQ+GV_)nfG6sJo3OwRAE=7^>Tu@Vk^JhLvnNx4bnkxSE~_(kr>X z*BCQW$1BdAAKzDgrRc-sMJgLE|zEp4Vc+$Y(T+I%WaNa!rjH<25g&ncd=+JWqY0(hf6dz1Xv?%he9V;iZr@ zp|Qo8^0`9X{qWsB^F;YG4TjA%k*2;YoVjl=ZP&m-6XQvkunKFGa?lrthY(Q_HLixG z4Vvm;W^c4+cfVy}YKx?FX^DG9$O-gOIkV#Te%&ik6nDe{Z6Ku?<4arsm_*fQq)Qo92K9 zuv5cUyw`Pf?mB+o7kOCdoitCAvv`ks4ra!W%C$@8CF)V3JH(70aHP(&90S% z#Y)9G@8kCJu#aP_m$i@4;F#4RPV~1BjZQ`_pM8&K8x7_;wj^x#=6+EubhU+`$1y18 z9N*RU>Gk>5lR1StrLyVTFw#w*_2)hJ(hRDezTI3T=T)c@oDir9wV=}MPK-N4rCr}J zs}b68D)|&cNIzAmyt2;bX>V$uZ|8k#o-t!sdDZNMkO5JWOWYXEsYWSd@XJE{LJ14s z0=bYx3}85MreTTH6>X7ex%VdTTGWc9fhqi3maWEHZ#^?VvYG&h(y`5uzEaR+#gTx+ zf@t3s1dMKOSfh&7GNz3j#EZ&lO6-%7;ZGQOO-hO1@qT_dM{%7@-(*zz6HNCiM)J!5 z25j0u%*NjsAhjFQYY{BKFz>OA8TR&_uoBUTEw^g}FW!&b9&lR^X>`AB1FOkJp(g}y z8y%+WNna}hqJNc1Gt)H3gJVv6OoQp5XJg+kc%z@yO?J93nzFpqH@+z&;sHtbg&ej- zZ7)=wo?hx57k^kyBev|gH=3sdD`p80=$r)m1wAi+xsWr*tEqLcGo#_BH|aEqEoUF~ ztD{+h$*HB}G@G$p9~2}aV4eG9>g>+Sm)Uydf-u)s*Yz*(#I7RprGh}zqpqJ=)rx^F zOigA#oo-m7t=OeX^bD!%jGp!!*&gDAHieQ*>b9;-X@_wwkVh>jsg77EFA*zD*Pe=1| zLtlpk9;;;M5f}Ykk-E|!?2Vi(c3a>}09u%|7eC3g;G%oc7Y?f3NQBFQ>ns2YX^!kw z{gBj$UEO?}v0%u1#Eu)+A0AADTO zw?)e8J?1{S#W0}O^ySHvO-V##^PU$xBxCSngfO*}&7=h3s}0KUaE7{VYZa?RLGzjl zRu{EGNVdtoLk-7`SB_CV?b(tz3!{+7cRz0Of%e_cqPQ`^lltUQ>D>VU>SoKYgr=E- zZWZoU_q7fW#A$rS$u(_^!D3Z4!l~{tG1RP6&8y=LKUVdW|%xC1H zR+zt(S>OFZthM+cuM@6iaPY}8^|#1I^ODLCr=hFK$ayfihL~-^M-5J@0+JIM1ZHf` zV19xzF{p0kaFg5u2aJ1@KZ<|L!<-R9w9F&MQIMI0Z9z0#QQL%D2;YYjGFHKQ?94km z+S_jZ%CGUDr+c3_FD;7=It+g*Xxa&^kYGtL47J|IG{h4HPa&Rl+T|7!1xA=(yNieL zp9fG;_1>%qlMG^MUY6H!?bLYY`%@(h_cn5Snb4&Q*4=78Zo=9FjV8yq1a z%i>@MY#QtZId6mMeac48p+!8Db+>`#{?L?sr7^TTwgN8kJHxG5PvF@6|*L zO+}G&f@nYmGI}Wt+Q54E8J+%pUkG+j@1vrjwQe_)?sFjYO5N4u0NPW%yQ8g`sE*rr ztZQR3>XYCGwd_?|eyYhUkoRDMmxhL;S;pBX zFhcNZq6u&*=nav{@1Exh{dLR1)f-0SK~pvWgNX#TIS@bx;iOg-mn4_=l|pAcvu*3g9I zkp_gHOg%}eevV*nbd zia@O;;5px7SLTZ!GTB3$hgkO4p5+_nhM{kbpEIQ_6fdA~P$M}{j>Ym$s96bTwY`@X z@!&}J4t2k)pUeNu{36REEIg6pEj>#eMvzE?hwQK$m7FK~qWOAf^7v=uLv=+Q7OFgk z&S8?e?BNPdQ{%>io3580N}Oi+q1P2WO8GG)MhEHhq*6IyH6MH|kea__-IE$`UmHFw z7qxVcp!pulhIz)6BZ!vr=_(G=lXDT@O>(Bh?ibP2EJnV${2X1E9V_g>R3AkF!S^Ig z=F4WwG)+Qp7&qZOS}~LUT)?LCUV9A)kGJDseFkf+s?FAW4H_`}VJfMOMvP@|wG_zi zx$OdL3yxDY2P!QL>eu|hEY)Q=i+F^0jzYH!xl9Go&NcgXo2Yy)-q9+(FFjThOMh^B zyCFl@C;5B9J1*l6YY6r-fxJ?jyg>8YVs8i~5x3@u@A3;e2L(4sJgIzOeFO9P3|}Mn zGxOI>{c43P-l5rU9<-o!*E*Y;^(HLt0jdpmspt@oAZ0bn!P3A>&aZt-C_zSLOPfnb z9Iw%w3!ymgt}>7vs+a`!Kp7bf+mw>!ieA!;VS~7SWvv=s9oqytUN`EE|I5u{Ug=5p z&)BIX3x=6~R}AH(9)NRd7yI8zZOWn1$ZNTnJnxAdCzi0HNk;b;&ZB37gbEF(Ct`Ze zWsO2UW$Dk=meHlD2}#BSOsF+hB7Lw@bxH>_SVH`$>Tdy`P~*B{cW#9^dFU*Rj&|Y& z?Tl&N3Rh~Svdp&Wll{H0ZRt^SaY_aHkp1a0Ga7ukhdj$NWGvU?W!cUR88dy5ro8*TT|GeY{r1b%liJ_I*8`$fTwud7_@J+pzh9^xmhfR4T!)*|;^BmyK-&OnO-}S9~^K7<|a72QQUkS(HH@c*+Xq*(D#~ zK0@T)GZ-JQkc#?29_Q*gi6a}HhpjPa=e2j|4?cn~zqXy$@m@Jxe3FE`Tfn5HI{0w& zk`Gmat1K<3cES0@8k$~&XX8VlTw?cvp#n3=$PDuY3{lJP2p~J z>L%V$XsnI`((1S&=Z44Ah~IK=#1{-EqG%wFC5_*|z}LSArm15zCp>~`0LCN=AV!T& zyA1a}=r8uri5H)U5+*o*O-u+mFsH8TsXA7Y<6~}aRqLZupFGH=GV^GQOykAW<7cP} zKFVC4-dl{0u3Q$b4Eb=-CP@!ffGHYc%X{s@jJmlE1~s76QUd9zan-R|G_)*5oKV)M z-J@^UovRyr`a)>Is-x8o7O2pTq}A1{Az?KXSmk&5?33WbAn-+}COyVS-`uo`bm7(8 z0=S*;P}p)#ltSSN&=ASbzYpk5<9rh9Zi8QsUlB#2CWV=#ON{G71by-vQPIdz4A`7B zCY)o*6gd!`WjoFE9_Rfd!BAk$>7#56A8()tOOH;dkdOnaxccU^kWH&0!n3{$@w2&^ zZL1#qy548>)Yu!3IWe88Z7uldL2{^8KJfR+vLI993hrQgU+Q`Ie00}EqQnM*@!?O! zpM$tSd9{G~-yG=11-QW0awK+%`#zT?H!u6Rd9J0eIo3>5_rzUm0!@i;BbNqm_&DM z)r~?02h-Imvz6kEq@}#`KgIThE*FFO6sBMfD^?ltbK|&fu`{!PoWzAe%9dJeJX_ls z@Y3GLyRCfkqU5OcHJ@4Pucujy_=lhFBM(;_!xEed;S{+?bqG*g#ebL_B3DK zo~8iP1nh*tR|e!FSt1IC^GV;U{C(UyRrn&q_)n}(s4vKg<=z?sTkPYpq)S-Ig!pqtqrWZe!=0j4oHqv>?*b z)4JE+vyx5OV0A;&(;v!-&=Ovl@nlAx?dYEQV4c_@>v~0#GSp; zJ2rCj%5m!(?0=dk|AQ*)YR^&1{OTKkZ{TWK+RlL)S8A@$_SjmHrH>GzX`sA5`ztCZ z@z2YX5NH`nD2M840JIl!#)GYn4=wmS_J)F<0CWG$6E5!znJG}AVQ!ht>rZ3l)+o-_ z@QdZgz3*sezZ}s)zn|)cFN1YI`7n zlLSFSizB+F`Uo)`I8Jk`Ns^aS{8-V8^eVtyYz;C1YF}Ll(UpY1$YLgxnuC%LofUN0 z1I67`TOQCKkVoCpY3nwwiqLx_sMcb6S%8AJ49I-B;EOP+>34!oMm#cTx!q`|%eN_v z*rqj*+p+BQh2AU+nM0@V*KIEnPKdz^KI+A}S3AGF+-yd-#b!5mwliszsd1xMfly03 z<27pPvfSuLNN{kHKe;38fnia&zQ&+ji89oia_FUdtb29qU>pI814HYeA8BBDa@@d+ zR?9CUrfQCnM=IK1TSO_W902*Sa@Rp$L*z8i?o>;ZeBrY_5X@NkJhfEHJWRCO1FhWd zJ1J#ViOM80`H?8<`8*V+$29{3Ru0ssxZJj6x8fhQt?B!sKS7$Mj*f?NsxQ0nI^Y1CXnggaYn(8*bRVr7X zaL-;*r-=`U9`?rNz7GZL)jrNp)m zXC4_|+Z#<3b|&i|Clhc1=6c0u91>o*gc9U|@aWpys1!g(l>4+$^%WG3x{JE9B$%YI znungI$rZ}eA7ko=2UaNMQS?y!m;inV=oW*UE+sal(8%StJWggPx}LHvw!fRGRc2?3 zHoYJlT&;K#yQwtSMS77c>Cq7oARlq@UG|wt_YWXZIsmXJ77W_W+Gh-wf4DTDv23gx zCHQW^k*VOeSwVQk?F(|r;u3CZi?nN#th+r0qQv~JR3QyO(${9VLbIz1%2d!RD@{}TI)_Z#LqS32K+D9O|S$)zlH{JvODbn_I zt2~NBM*()*Ak*{N#>FiCsL;8-14Qbs^AUNv8PH++F$-u4t7M%hzWFpoPYp*rX*~_0 z!>2#uvktCSHzf^UV{_Y28^cIgT&U)8vmdh!OLM=`O&vF@M&d!tL*# z(aMpf=D45NhPT)CMmNFg)qEU6&O+wJtoz0dwX;?8FZB3lvAf?^WMOcx6oTi}Kh48K zCjE>bav9n(i~GY>Q&@HR@vtqGTcPv?8yLZ>wf$5dk{H+mMclYIb}&omZBlY0)aIlL zUD41MUNdQl4gKu)bfz2>?1&udZe$Y5yD`)D2>d8z4&RphNRq1r!MbHw>5O;pv#75$ z1=>MA;F>k4P^VkC=>8tlP6zdxKdxpY|FvUqxt-BG0O#Jf?8ttTwH>7uNPy^>QK42R@OeNM+TcOk16z+(l@ z&BrpxYUqz46vC=2nB0`j6&78OPHs#am?RS|PHiL_as9lvgwOgrdx&4Iq-I0jxN8Lg zfZ!hi*{0%dhsSJT3DPB!G?z}iWmx{)Q&${T{wz6UlfCioXxAe@NtbV4?fNxu1Rkb& zUHZI3bJmCJl&w%14|Bp8if6RlkHMR3uvYG&5saW*yJ#D-oYx#Kc~Uuyq82{a9j7u> zZ1O~dau3DCmG@3)1raMX2JWjnLDgWZn{`aSL!2DJ$iodN;hg6u=zNhA(CQihb5$dD zTq&^dZ>f481aP%|Ek)r9j|naSeloGvy|XebW_Rj7A^Jsn-O9l?uC<&tqkjuvG&pcM z>7!m=_R(iAJBmbPAgU$D%if*^_(p4lmu!%`5l@k8wrk(Z+EIN156s8|P^ul*JGMm= zhoec2WR@}fqV|6Eie4dyUcL@CNd&l?ivb@qP^GX$iM1RHE*AD7sfH{uU@HL{q0vlZ zj?0^h1K=4vvyJ$>wHT&_YBT*DJ8YVsEu1grdxWyABrYRI*swc#uqU zx&QHZf$8;PR-L{$UjognRQvfOok ztue_F8o6o6wu&wN^*P*(tcnf1AGJjc)K#25H|zrrX#f+{K*F zSPxHnoC~{V0j62PI{DWvD!rN+_G+S&MzvBQb|h(uU4R(T>oGJ!dQ!HP%fx@WHF2Vl z&&Yzw+FY-x+mJZs#F08z=MW`*Jk0}p8gxn>+L+xjxaL{A(=DRlzqg4e8o;Hk9+YBU z>GJnk%lL95IQE6W)P9qPG+@u+T#zU}rUufo%f0rY9tnDK5_E+YTmlbS+42d%UoP~G zM`NhO%>;>W>Fp51RdU3L5jXluOttM~HzEnO2@us{q+Zg~rS-ucU-D|7b*gj@VZq#g4u4^-Q7}yw0WGr7<68#9+S-jvH|0SF+7)M@3WXghC^4zMVQuLZ zxuiZK-G=`rfA2y>07VU{JE{8i6n=}TmbAPOT^|et7g*lmO-f3cb+H_JJk1dK+b*DO zl@_ooLJ9l!5b)TQ3ezLLyDc4iT!Y5v%ffik)f$vf(q#ygCnB*}u#m4O4f%ODwq>~_#ot|XERWe4jRD?xWr*`HGI2RJWmd=8Jq-W8=#S9?I~Xxh3E!t59=>j9Dz8?6yv#@f)pd`)#BS=Ky?8$sc?J_X)K^ za~h*_)z|YIv&co~UKl*!N^>_d;cT$2OgxfQ4i0H$tKM5=&Yvtaircz@tT0U;$hTbB_5fVCn?76r zv@6`l!E=U@hY_|p=jZ_o z5VudSH@(9s73zEi(4T4h8ECtKw}d497q1rf8{U8_zcQLeQ5l5M`?`IYZ~7e}UznQ2 zkoPI27?p%#IyQW7z!R+|($R(CIEu@0J{wSnvB?;kG{1Hrq1=7_yFM^vQC%_h@6ife z#xQflS8Pee8n8ZPw;yS)g$rfVY-zn#xQr@$1i@rmO96qdOkRw|Whm1@5xC@iz@E5a zxpPiZC0mP2)MxTo9av`UV`K|F49ayF96)v!$K_| zE(1iRPrW;CWHSF#F0;+x9hh`B@o~QEAUH&mg@4t>cB~*tLr4=ra96a@PE#e8S#&>b zj~574@0e~eSCj`Ss(AVJGlf)E==!LK$_}9tmz=s{tf9~;{8Nok<7u9wJ3^RK)!J^3qOx`x54x$(OO;~urn6l%+@m-RQ4K$*BL`Um}WgA z=^c@Ceo02%sY2#oBo8x8eNbjLDBfjub)Fz*0{XRe^O?+|S@Ui7B9>CFJrlda8>29j7F=$~ zu4pXx1+E5my(jub=Q1mbdTs_V=g{CwpM9#V;rtYtqS=unSSpmN^^b#no^@pm-}s5^Tl)(fI)Il5I~Hv-TTTh z;)ovy*8eKcrYx12p7 zRTu^qu2^>0z0GK*a6(TrdR%%0P(8%=0rO?@DLon2ZUD()_A~Td97jCWb|d8exLlGQ zk~eKMoI*nty-}iDozFsbuXFQp3j-h}sjkLc8-qjZ-vUZziQ7wN43mM^-0WFimTDZf zrS-lxES@d_0=IK4dnhP!yB=3%YzZ)_eIY>zEnZ*sX;C{!p;o3%bjyUG_v0h&I+u#o zF0ga4<@eZl&)c6j`xeOCLqgJlVNU|9+bE+-Yn)iKfS{A&kDxq*}Aci5E zQXE&L6ME{$LP=Q9m+$xy?2)WeymOpViyPza=10m~R$1*(5_!GQX}q2h!6%rS-NIdG zEQ?=l-fWC$1QI29$&)NOa8EFpMa$NVAv_rDJ$xF z&mgqF2Sa;GiN|Hm{Y=h;x0pT(EtkQ?;we+t?XX&mFI9|_-gC~ z?3yAtV4Uaev%7(cQx_qR^Eu%-03p!&gx6#zpsU8AEOLW{(n|!uh@_n9ruBF`QUk*l~C5-Vx{> z{LXU8H7yo^;EC?=f`~K#=K8tx+QKNd_XfhGb$!N}O-?gP1&BtDnsZ&KTsYng72Nby_Pfh5rE6c9L&Q>2)y;8 zH(J`LOr!8_U669!xHv>bwRU)NZObZUE%romPQJn2&dGKa32l;RN0 zT4;SD8n$r5Q(m-!dpI*hL!{XBDP=8!EGx9;*!DvK>u0m6EUGINUHnZkhLMURO3pURVV#&*Dc>9A?v_~iVZ2~mXTo&IA#8x%@gxG zCPfX}={JESIgZUxZNWJ*d+d1~N7RT;M*^<}?cp5vDVOVh_zd-r?aW_KyD$a-ncRBK z5E?cXSa@W8MTqz)+|G_RF)#hHleP*BKF8YAHkw6~l2M@ESqs z%E7+MWr1yZ=(TSRV0SY)X$NDGL_i$s9^$Zt(MW8qj$jqIqJS$=3uW+MZ9jj{sX6vz zvkD89MyQSbz{j3!YSTk%25+>ZxPK8v=XlF+^}^L0prO3$MoOKVrvaPl0rK%xC_~i5 zM4g8tUli@KHQzb~_85YFb6nhOh2wxzAPKypS;VJyX@>5cz{)276{6xevTT=oanLK; zMqUoR2{sH3@<4LrKgcZxuTn)F@p523Z;y4+KCpL+UU7f8D`^`tWjZHN5`RKLO5h=< zKKDY!c$bZQog;$vd;TlSAzS$(pt$PVUR|0nIKhw`3%Y^^+X|nZ3+QjEZdBhSiyNp| z=<2Ni`4*`Lo6>;5A)n(;0NWJHp^^BB&HVcf<83{y^#IZKS-3!nZJUQ5vc5$uDmv&Jz6C3`{%3Qpt zwlQ(3H#-NoMx9CMS+>p3&SD_lkTV1BrzXW|^%!B{1@vP8;zzCaBh25VZ>-NLj}a{P zq*?QVpPsWZPPp-dYRvmPM_Fx$k@dl$yD^Uk?MJI@8DQV^u1d48;?uk`j@juJ>O=!I z4&oCY2Ho*P=TdWYi*Cwjgy*XQpMmIc3mG7>6OUWjy9oxYW;p0Z@0 zz@ed@8yn9%u*^-OeR#4*79j#`_hNmx)@ajC)pumD6)NM%e^e5X*fO4O{qGMSfFU`RLyRe;l>9FE{RirGfU{|dw)v(eidUIai zjA>C{L-(DY_7c^KLN}HQ**CyW0@WUqmSv%-1HJN>x$Fxb+<-Xjy&2g?WUc@Z-`3p8 zNd}DU1ywtejn~_O{b`TnlanIu-rb&huSocTN4vQ zPN>IR54I6iBp5nOH?VJd-1R!Rz;&*wzu0-Z6NOY32i3^J>){`2*lw;|b#v6FsJz{xQz(=}Qt(H3VJ?uGc?L*rk>BuX) z*wcenUyGuDp|}oV2!JJ{*6~ccnm{2#tByFJPZqsihRA59TA*@7x zXA|0~|H|ZRbQsc$sl}^I%2XJa!~-*M{G+@anA&9k(Ii$nqNR1w#G;VE) z#}o)l?VW`ewk>J~t?2QcHHgVJM*Uv?NtdS>2rt$LrUJh^CtUd_XI2~r06g!;R;~B< zjEh``Bdo>fE6X413FV;#PP3xu+47J)36Ety(CccsjFu=?Kreyr4ln`OrHj1pCao_E z{3s{<2e;K!c^jwi5^Ue^3a3KN;ZBqXC|wKiI0KyXV_=ru<3YJu%3E zEc}MK*Mvu`hz74umN{*{W}_;6>xYQ(kJl}>1d1E=rpuo52e=w5wKLK2D!dl6B>mQai{PC<$03bUf{D)!u6HEKQ{q_Ho z9TOh9B#A2h18W0X*f@ibR1#--EidB^!YV?P82USjUgBKzD`>Aw2!zQcE}u>MQ+jH{OCO%jg3CJDYIA5r zH&=sr*S0Z}3$vf0xoL%QoUPQck|T$+AuDmp{U=gY-PPZp{s)V})tAQ>V8@;=97yq5 z_Z-|vF1g2JI{GeSx8T)4aemWSJF@31o1UeJgYUKe9A)*dJl7`Prk?rvMZo0o@6<@L z?Oc)S8tzeHzcCY+b-d=N5S_Q1zGwX%H7cdFdxo;L6seRyf6A-TA5W;KNpm5g`(x6% zuXL7h$)_5~plq6`gZ13E8pAvf5WJ$7e|`uQKOV#P8iyF8JGPYKOuh>s*pE3O$u36f zP%&%TX9c!$O#k5CUfl0G#&IC&KlUn@obP0T=@sfucG}i2$ZZ4Kqm7k1+;9r zg6anJH#bZ6n#MF_4(}}RnVbs|Kx7avQ!4!Q+8Owe^GS%18ZeWdH}K%-!3_V4z4wf2 za$VYh1qBsE-69|;AR?mFO_z>hqbnf2sPr0;-a%B9Dj*;UO{GZh3B8Ddlz?BqfN-%Qjd!cD@-jYH1 zKfo9l{X?F{wLH3mdKy&h^88P`{l`yDO#3Co?ok?&_4bGNz6nkxC9?k~io&07_5ihh5|Bdku1Wi^Ie+zz&;Qq)|Eu5p*E@f8azF$m zGXfe$3PqGp$gJ)%V9>_hCTxG&_w4hDe}oAGv!*+aaLC?aQP8oJUTqFrNA+4fp6g<* zCc%w)lnI;K2u%&U-hx;Ypxboqgva{K6+^HNT3}MI33`!c(Hycu6umt#303Y$xN;?m zGf&u1bYj7aGc=uz2W8U*AB^(#eNh2cMInmYA18wLl-ExSSP9oGqm3>1(wtTXdb2rM z#Xki`*Gzq8xuq8BdtxNue^l@`Cn_!KExTv-FG~^75lby25|4_4n6z>4BDITwo zIQ*tl;xsynf2Q>;YGi{?^W3;dlR!HdSv;z}oDfsNc%<9dMcJY-e#pta%_w zqs*#af2CSB;cU2uVBUvZKXzfvj%PO1?SI_BB^2Di@T0%RIGIe zozTZdF(MKqGheXHw@z+2h~MtB=}AifeeBmu!z+xWw2H^G>yHEaGJ8Fz$<9|tR2AvX zQsp$k0;7F#u1S5vxv~ZehODiH;hVIZwVU;YMA@Y`a6z+*5&^T{TUI2UKUyX?KFF1q zUya3dZ}(vWKH+5v%uX_$Eav`JIOpb+!C4^HG>UQReoJyW9 zpD_DKWkryKXR((S_PYl8g4HaLUK@!3L%CCZ6P&I)$gPQu?^Y{ypVu*I+q3-G?U9!B zpob({|M!;lIF9yM`Fy$llA;-8d}jc(H`@y+S^~mPw;6fpeL$w)VDzccMzJwj$an+H z5mPp9Chk9P(bwQazipqsKw+ia=Je|@)B8;D!!A738bYl!mR{c>joakx@4@OTfi{iY zTi(;Q+o*b%Z%>Z-`@dm2WV~c{&PhgVB}H^AVT~&E0cYUmOL3>+Ub13KuOu1LOi_lk zxWy1}t-*E!J4HU(XcB^;I(WFVZm+=rtYg*fwKW=l$FKj$2<+|dt({!`CY<@G(zpAO z*CdkE7VP;!FYBE;Z2cW-P_{G>+&a7JBJAh-b8GOPf@!uU?21;5uUU6`<7D}myc?kM zbnZ%{D60`GpVfag2KtThEA9!ba`7U8-8LheOM)SMYaZPWNxIvZ)SV=i)RiW}GZHh7 z!I_gdK2@%&(qh1*Q4p2gQ`Td>DW;rh=lKYCDmg)qGl#VTy%wjZQc z`PJT=CfJ~)@ciK_Rsyi|)`|F5HEbdr{F2EW3Sa1|99{_CJu4MhW_#XjlrRu+(W{6!ea@Q* zny}g*45Ys!EJNBUn%BnOw$xNQd{Gltu-W$5V+T0uJSYE2rSnurf8wH|CXL?8 z+d)M0Z^_EoFc<55y3r^*)Cebs`L3QzOhfNbJGO(hF7aaBLfmCY)ri@l)WpNJextx9A}z#_bF+D6?ug% zxnstk&|vD!)iX5{R8{5f>aM`~YQZ`oqYq@;%tn8>Fi5(joiIqR5_i!xQ6%KAP~|U( z@y2x!37pYo@hJqXu#MLho$QiEdIkHabMM}VRT%?0GSYF>K$c&SVWo9_ZN9wB9V4vY zf^yjL2t6$kjlH+Hjx0C-TVGxDxlpkx;irWj)DQ#CeMN);-tLR~BGo9`^<0l*PCFa3 zzmbGw@1pnwH40+YDi~NLRn3B?`z`~LML-hJ+c3^RUexbC_b%Z+g%1#c_1X-fhOGSF zEGU%|@2EFW_rQtJuM*dq*Y}qT?!4f{+HVBZT^i&cmFcv7sauSdL`hZJOw`*tGn zkUD}NqLDdi!bBp&EbEr?&XS96fpc+H`>@d{Y_M|X|H$RY~M^JjMib|I|+3`gpA7$30$U0!BT zXso$souaGosX^Iv$2Ke9h8iyPzpb?O5EGOw*4WcpNGz1K{DJCe2m6u^!a&d=*W6LJ zNYA_h@x*DQ;?q(IS11?^mTAtbcDBTv*{QR&({%z0wWSRiZ}@@KB>el9;!&XT_+89y zw_Ll}Suk~TR5MM6XNNwCiQQscoDu?_MHm+E*#tyLRM#dFEAJ<0uwL)jjKuawl-teLW?}r;hxf0EOsMUVwpk4U16O|PnWn>r z9;;he*~dbKDRRSr>2Jsst&FxI|FA+L9kX}Dr=KT-b8CFU^ssNBwa>-IQAwhB4EODr z`1axk=bqyM>-UIKN6VvcQ{qF2!|f(^`1&@h$4o7!W!DCy<#4!!WH=eQ){U-IP$D3d zLf*~JfJDA!sNjXxy`sbCdsS1mdXM(Vwv@h?L^ViyCATE}R6CORm(LjoiJ^OTrEJXe z@;re;ahctC78f#A6aDmfBeYgYg)+bDnUE?a$+j%jTBo*pzCA5g98>QrCjhjz#6Q0A ztJbV7rV5cNKE83}{Hr&IC>1UpWX*eUKFsdz)a-0maykCq;p6jy^0h=^;k{Y?iJ_na zwycjedU3yu2t{cB^jZ6ZJek7(X)yuMCFKc`maeFVxi7gGg>Y@dddb3QJZk0B?=jMVE|FN%w_E{@v6+zoa3{(8jVtd1_gC*VsAbP5GGNw z?%V9;8~k~;FrbN}n`eMdJHB1}(C_^od74~G1li)k7JESEK}amYdDO`{`)D2OQ3^9K z#K9QY_wioCqF!$!Q+QI|l&%KqKmgFxl`#b*DgCr_z6;yBHg6?*Gccj}55_B2GVFsB zCgUxyqS?hh<{rQ*bnk4iTnvg8)r^fRRv6_4xGA(YLhGyJACkp*!qBdsv*+#(i@C}( zK98<~jSn{s@Js%kMd!Z*&;Cg9Y#)y|w-1DTy0~9i>s=2)=g0#1*~=weX*Z0&-#JGh zm6XdG?74=0zGcAFI#Ox(khh2Pvg!Rz^JP2ORx7`Pil^UOtfEYX-Ees2%uKruzHb?( zFTGbLEXg_bEim_6;PI7HOMq$b`vg|_&A9S(VPkfX_&n!{l{s=~7oemn$P{KRjUt@B zm0$h(Dx2+27qq-&T*`pejC5*hTI;y!q8C2lJ*v_j{6<$Bp?bU2VQv^2Z zJq-j+YhW{z0J`DQzY7}J2Mrd43G4peC}1wfKfm4AV>DoTslokqx)zwYLgtglC;`y|pSpXh~fX&Ci{*!+vNNTUVB?fb>5#SNjB?Pd4zQ-e90uk#Z4 zubM83+j0Ry&4V5+q@P`88-8%BJ&p%Z3Wu%`=AB(hNROFv;V(1#*~7K&hX=;mzS@!M z^@0!sdSC}!1!z^u(aMXMSR<=-X`Bb6x8kHxsh$%Q+d%-o)ih%+mMm|Dg(IwbFeLoq zGq)VImxsflMcv#XQejU(^YzZmA0>o$T`ZdvL>TO9!DD^^l5M8lG~rrI(R6IFS$XeD zyzDl~^X0&{CobW+#>hx#Mc!z`HvG_dSkcn{hViQQ4mvQPMx_xmzU1_1gKLGXhlE12tr@Kc6VuyyshY$w)bbJ`CmUJ8K0K~k=7-ZH;_24Pzg5M zz~*@_&loN%hFwi+C-1v)tGT9sb8;Y8A_|U}O>U+zYjpIJs z%M6;^z*a?Pv5KmfjmqNA$@`9Bl8hXOs&1y8S+9^UW<7wU_C5<@=MhTmG@wk%Q3-07 zzY^R%99>9?MI(3|hL5d-4xs^4=;VknLAWJloq(q&L?Lshz0Q`?SK^HK)>oZ+Z+l_UZow3dFtg0*TG)Z~+Ohm?H7bb} zI}-oXtUd4rjLX}@v)pFEgkYHjM1QLHg8RCG=N`n$Z5X$FOcWvo0~y8<%;CLXwey6+ zny`?rM6Q-w!?ozQh>9{5M|bRO^9K;dN$CyN&7=Qw_w<9;7X0Af(@YHkRNl+xfm=~gGaSj-9oKlv-B0>U zv*Wd%uPFkcD-%YYi9%dJ@Q{r`3IJj<$hx_xjB1LYdi?}oEZjjZ-p_P-yQ|$QgCR+m z-IPu93ejs4yf>$Pw}&|D>lnyCNNNNmL%Y6rp#hf*(`p~JJIWq>RqSHd%cTdl=&pbb zq92NWUQ|vux8z;?57M6JGx%`i$tl`#hPVr+d|cx^ee4&;1s`&q@8h~~q;H(-%!|wC z;g_FedFU}jasw*|5~g_r1vdCFf~X6=M={HwJ$xM$?8yXZVc|4?A+n65_J($R0d^QKimms# zRCO1%JQ6|M#l-j8+`j&<2A>?u22FcIy45dQpOd~cm(-z`l%Kan(rnKY!Yk~5+jt<+ zux!>@SXj%l_epu=CKKFU#|` zb#VF2LDMUPVsIy1WUy2%Z6_!Xnu7G3I2@TSPa-)Si&+QY1u-hO-q}~vzh1Q#-#GNq zw8mYfY*xs=f&1c}S4#L56Ja2ssda7!>oey|AaWK6Kum9nO%Rw)(7|-8zeilSp^`eUvwflWWmY z81Hv-krhF@nn=m^cNou4BC}*qUHR%;9b?sTg+hSKyc273d}Y{OgBpJ=nu=7T)O z4&V5kWUlg3G(n^Oqh1C&B^iZ$e37PZwmm`F%zmgZB`N#ez`|{LtFfoKqK-3H%kzfE z*`x{_Fc~NYys#yimZl;)Kt?xeckjX4xn4)eTWZXvfksb$fN-hyoV~Bn#JjZk))Ps7 z+L^24pDCyD@A@n*NNPr2t)%9b);H5F?DY7-WZp0QKI^rG3o25?{8Kitfbg0P5h#cRgXc&;O)c6PaemE5_R#7>Qh?ykwzy#qK zjtys++`~*=0nYdz7R2sxZPE_GdLmxS@^y^9z$O~oSD)UMCN@#9z1XW_BEhxp50?NMIUz^f_zP_cS`+88fX~Lvfg_0{Tn_Y3RWrB^HkzI4i#WLQ&hXMxzdmr zjEGRrC5So*O~YPa~foweC1r1GUhqaKbT1tMTV$~y0PqP_mv?@OGab( z%B$}VQleXVxs6fu_gvJ++ngFs)Qf1RApru5bL(Rz;mQUhaJmh+Tl`%7!@5q{8Bn* zoGl~gd;WsmmA^Mya4;h^uE+Qyd+n!48U$qYx{Qadn+9@D&_K08mQ*h-Ip(^sC@j8e ztvy`jS)Hjyv@HlmWU9m)oTxZUE0-R|p=yVGb#J_OBJEq@wR$5e^YfZ8$tZAPNV8uG z=~{)t)$T-|a*TnvuN@oZk-BEmXNadu%0_bNaiOR@%9DVr8~R zH$@REwU|;TYK(5|6AIQonJ$n?NL6n1^Hv2?hd0YB>~ilF#@RkOrPZ6AQD#q#Q=$U@^B;-B)O%o`+6EvZ?m0uxo4uiA_3vzMY?C z*NLB#8-Y9Xnab7+#AnypVI)yeJ%b-Eyno(!DPPf0n1QWL{Tfuz!P>CFGWwjtx=-Cf zjQB{yjbXlZq!h1uI+V58<*on2!*$7_RM~F_y1@C>>EGi5V~;UqDbpZIHLUo4DO{fIcg&kS{%eoHcxQMcQ`OfwcuaO-)lLEpn@P`7oV# zd5m9}Y+&&+4ZZgt_ItB$zXvDa>jPC{hN3<*v0*&=-Amd(B`A{+;hM%!9-D`yya(4- zHp?^0z3skwU7ht!uU>S|!PUv$c7HA|2wFLb@}q+aah{NMzqypg@^|4Et?)Wo2&&@A_@l(<_R z4xj3nQ(Mb}(w*l7YYM9AA_A$tDx$;i&LRui@@+qe6+J8pPU~l`Y&%}jSvDfSFt719 z?6dmfoW0$DEn4^ljt-g&wnvhXoRDRgR#<%IC$XCv6ZpL!e$*IB9-UVT$GeU?A==5~ zXEQx&g)WstIWnnS*p7G!QbBxoLqfx{t|zc)ps$)jMyk@({VbjfK@q>TMxL`wY$3S< z)PG|_z5m_ty#EB<5d{0j5 za5waH1nq1|wI?Iq_-NwnxQB9hlOuADQ$aY6D*UE2Y0a{q#m-Q&7<61?;yh{{h+4f7 zfHY?}7m&1dhD+@7bN5z^<7bnnZg!Q)KQ;-S6h9+$w^z4lmQMjf-Ct3xOqA>_L}2Xm z+O}h=a$=s`>@_b$ z&n$?#$@of=g!DpIV_`IG2XaDOLZ~tB2kxNy?5p=0{H$=@Th_y5i%%bpidyet@fy~K zD4Z`fhV~d6mri!Q6CC@x6k3lo?-iC9e*AfiabjnYG+=Y~7kTg`XNb%etdqb1Lvn$u zC`2r>L!u|D9g1}ecjAVfg02lJ2$8;qUa=a7TAgw<86lr?+ZI~R^60#jjkG=DU2bgB zbEY0e1<{OhA`-{Jn@gt~rF3@o()Nu+jbW2;;?LJBmUsY7SXyhFvJRK`BG+vXFOs$= zY^)589ZqGN^oR{A_N2!1ihdU(EtW5XLSOJGAyF6pTu(Iv#;2SeO`qYEC?tiRTHqqUtveKu zYTjAo91M$GhrcC}R$F2)zUT-fCLpBMygng-8R+v&$>TfUxB25~&Fax2j{Nzlh(<4) z&m2U!u?hA}y~IWsB7nFxPg;v)9c?Q*X#TK^Bc~Nk+7feFy@Qk1NF_%%@;UBbX!Ya7 z5tGh|%lzkw_RzNSH;vpe**%{3_snSc2m(VJQZA%c!R(PO1vkOFnUM(OvxvlEgJ+*6J6Di&0$#2_(D_tX{J*27gBN}eB`!+f)-r_&dV+{GSDuVIl z*CX$3tm!$)`XJ1EyaXWJ;20dB7NWbMdX8BKT60d?uBR&76bo}7$iM%3toreb2Gg7? z#@ow^m+4u8CgsI~AS)7f4>g!-9!DWG(E?g3&%SqrTeHdfELJ42I!6?kY`yd%0FB%H zVVz;@W}20%{IpY#7uWr~FdmIXUF(g(wnh;hAG0c#-U96C6XnBO2{~z9mgvIfQ@!f; zVm3Wnl4AaMf^xU6d*EtFf=@t=qmmT8Tt`%WU3kIV-?hKCh~5|-bkVL)?h$+#H8L`x zbYz>M<-K`;Q}+bjg-!gudi$HKvi2p0d($!Ham9D-q>10ttxby^RviKg?vM)J1vrkw zbrcK{*8nh@(l8|A&oXA|zI=qi8~lYz`!BSWl#?%HnRhF6FPrP7!CM6j&r3+VOzMEO zwFZM`yX;!s6N(o(iw$hclD!Fou6Ed)4|?)0YCvzetLwI9;p)ValT$5Mc6l(874L9W z74r1Snh7$btx^nrS5S^%Ywz>p?O|lmo#Z}Ut=5+!1SM9CdzTq5DG{iXbnq`5L1&c| z%M7Lo?JHF+k+L@3zs2)h`ntq3a_d8>QxNIZw||K#1*9GO-;W}7(Y3hqI5&>Mf$tL9h)>Buu$C+$Q_d<7D`(0hLIVum=eU|=W6AHNt%`O_~ z$LklFtmU$LRX(gA5zPpSF_2v{&s)-g+Ve2H#-cawLrCCGrH%qtoqB9Ywm5^-q{i8@ zoRk5N9-sw^F0&k_vm0%2@E+7}j6OC$C_kqZ_$#{1ak_Z}mzGOTT@ih6p&H5SRLNW2 zr+qHyJimJ1!=%QqnCfvmT^#*=O0E3-*b)o&S`q1?}6OEjyvoOTki3Lj6N>pj1b3^PHW!XPb(ESb6gx<*A;`Uc$g zl@a*!$Rq3Hp1bpb)januOdty=xU1;s$goEw~sb8UERFjJUg4r~SbaQCA19Q*lzc9h z_5F#{tJWtjtty|gK^=~mh+i5mdx5eT#YcluM=;(jsKt@)^|w~DpPiME`O_tPQDute z4DOW^{%+Uln1pRS9VCZB{Ul#1oMq4hEgW81&t;MOmm*n#m}ZhxC&m0WGkp)rhxH^! zJCjJfk9w28W*a8kNOTtFc$4)U;KR!>MLg^`B)P!EhUlByy^+>p5wPf$T}Pwlb8sk8 zD20i)SH0k1LBcogw$7oWI$d)k!su5ILy0uUyvun#)|}cVWWbTo9RiaQisPPRW~Df8 zAfd1MWu>0#Kt1vJmmW0C7y3~Smf&3ZAz?tEKZ#b$jnEc=g|nAgeCOCm->k4h-QnP8 z3MhT_tm}1dnr}loAV*Yyz)?Gl^92G1nRlE0UGWQ>S%+pSG6nKS4a>nH%@{ zMh+k7iA8kF=)7P~H4@iptzyN=Eqe=0iQQ`SUI?RHYnFT|?|$xOnaX5lVijWebL{clm$~83pxC7}!R|DXr8>y|5+|D*NQYfa5WL+johIvh8TK`649Ev{a#ml=il##imZS?VKOD!!5kqj5#mm zs*mcNA?mY3HMU3=`8p43X@wlwvNo=%h*4d8^b9l|{er=R+$MSoej+c)-SF_Fgqmo2 z5NfXG?$L77jO-1UGX271|20jgoO~l2RO&uI)+sJw|HEP3KZ;LNTfQ26O9U$1G!rSAXm|JxU( zL-apA=faPqr#k=n^3p$9trxC=Uz)UXJyWUHz1ltW7t_kWYu{ZJI(fy_z|z}o7u|RM z7nKco-T(FgCjs*mv`7@e>(A=UI{S-`@4rB!KmOb%*N{6eTQ0P~rV)v>|4BmpFFzn; z{uHI2B(h%lDr71&{&x@atAC6B{1bMqi^+NI zsgf_Nb3cF^|AUuvzUI&oZ?}cOTXYp_NorF6_S*l>6hHqhc>0e9N$yeY1U( z`;&kDSpRzTCaeBv?5r89h5p~aEm+B&f=v2Ik*95wgp!>9e1!k}sg33W&+&+3kmv~` zw*Mrx``35yPHo>MZT{<$e{0|Wy5zr}?0@lq|GMP=>X-ljaY;fCB{LuYN11t*lhOO4 z)q2oY_)%}7{?5WbS(vKw0qLihh6zG#y(R1m2j zX78iLZJCgO{wI_#!7QX(2pnCqKQE)Kg>#gYiDpqRDNW+++f=Q_= zPjenFka-=HRHst|D_20#W~t9^Q){uI4?>T3Os!4anoo5 zzka560?X_dn{>6xyadZ1U#T9Nr_i1OREnnlMs7Q8K!-9OY9q7oIgi41giISwAC7k81XCyef=F>aJK0f}G=H*`(mGZ+A> zm0A19VA`I!DV5lKXe4>wQjqumBp%PD?5ncVMEWMK!cKXNkh20xO`{C7cyZ2lDoDP5 zal%~T2uvqceuJSsNec13LOHPe%}&~8kjIN$uTDY47dX=4OP-ZZXL8PqT{4jqF8EE* z*{D&Q!*{-Ul`qUt?I0V!t}n*eukYzSAp50)BOuFMraq}$F)OQ+9?y=#sv(NlAEgp{ z-?z;Gl{L!IEwd#Hw5ROk8W0WtzCaDZG8)ubR!4G(UDHe>2x%(E7zpP!^fg53=AILK z&XRLoPU@T@xzK&F4@KuLdy%jXhp42JMUv%QE94S|3UNhOv7goCczHF`!pr5q(mEm3 z7fywE!j?6SZBJ7AJYS#^*qOzF{89p%)rgL%OZffmSH8kMr-pT&bugoAe+q zaAAGq_Rd^avgsVshZnqCne+W({ouH(##=k7i$NWgCIIiXe66F7kVcLu+6$tcI5Sf^}_T=?qCx&ECwQQ;R+7|vn-^f{qTF` zWjrczLu@UY7olrNot0XgS@IhGbo*U|6Ew5XC?I`vov8qTbU$9_xTb5a}dbD4o^1osyt_^Z-z^0nYA0qw=zqsxTxVi zd27ps8Y3R8b5OdnE1g(0^CfM|ToI!Aw^1krX8NFa>V!I6qW(C0<5J&()s_2YRe&gE zBrZ1mY%)Be68Wlx^px^S={^Y_h_8+SNSFj`*0x80u79Y9u;0Zl%D# zhAo4$72DJJqi-{Zbv@5u5XNsKSxv%!Lid-NcfN6rpLw57a>;~(;k$RhSL$m5fv6(@;r6r>;=fcy1rBc&$ z%fH|BEj>XP>jK93EOdM2&sHpe&D@yk0-1yRYQh_*IXO<1X9+tee4Nrcf#s^C=L=GOgZZXc~Dl4X9(3b()IVU)J5yb2=r%tx&>FCu!M5>~N(cku+Cg)h*h-}C& z>@A*j<7%cZtN8dqgwbj$GdJ6LLSa19Z2V2BMLAa_XL_LT4ZiyUz5V^ee3(G7ODDS~ z-a-w9{E9waYFdfaJ<&<1fDA_(ucdGRY~T`qgi^Ic zh4%*|Onf3o(9Yiqszi3!)od`x_s_$+by?HQI!KkuTe8(|4E>32P`xa*tp`ZlwG$)jjfo2<;i$?~I&IUVz*j7WI0Shh;Yu1`x}z5h3cE+ashr7+Yb5U@4^aD!5>{?mS;r$}^IpT+&1Qx2@=#@@&X=oYbVb?`K zhhSS6tc9HxG`ifakRx^VH?RvR?muJT6VL&h+Z3<58+weCz%J_rTcCC~$2%HYrb)cA zJRw)0T?juXx0!oXwL+`Zh?vz{7C;mV((pylh|Fi1OX$=Rk6F6%bY-4J6AfKbTzbLz+QTc2T5#uRAG;Am zEQdU?leobQMb^TLlfe20QSYb&>6U?at4;e5Gg>g?6a|9=V%DU4)96kkIgV8f6Tbjy z04#LR*5c@HgKSn^F>7_4R}e6kQF5?Uqow0M)!JhpdeFq4S;hBu4D{7F-RAz9h3f39 zGkjknzNx0%K+SXG{me4td%y+~0V1pV0nL66XY?!@P+y4#z*exD4Zj&jh}A9aHWX{V5AWN>j>aG`zWD;iyA8ItWc`kA zS``|{G(29Nd?-yKZLbq?LEa~NshG7E1`&QKQdd$pITRm;N3u$_qf#l!E)Gqn+27^a z)=0j!xvh!NU{maXnf`77E|GGdZ8sm=<&fX%XKZa5y-NQxCH}Q&r}Sq}!kf@evYF2H z+1Y_a&%#V6*c?|(rIrI*I!mUq43!IqVIO{xX z-U{HyUr>DHtK`#&V~}te7q2+CgFmy+r=AtWrxb$@-cbmYaX=uJC|Q zub;^l)Wd3m92zHXr#;-T75{3F3SlFJ>T(Dz-1q|EJ)$1hAS*~RV;68xFcc=t&h+P^ ztkWBV5rBni%%|&X=5DVc*fUtbCy}oo>$nFyb^l$2>7 z47w@s9~(Vxae_A2P^r;vi&X{P-J*{-y|Gk8+TO^=Y43DWC7pG(Iufre41e06y&s^K zN)3*uf$$SLjric%YGWrB6Wz zqlEb=7%5M;SZ~<>?OHz~Xt6&PN~3vLRD4Jy6tijMe8W|AQlqCwAz?`tQZMw3p6U0T z08p4@S>}enONS0V%@Bw3=W#7{c7m+_HqO^U-O_KOu`z&6%9KzS3F& zQ~X;mzYoDuNRBFBq>N6`PMOc69)s^rPig+%qDT0EMbij`t5*!)8GI$w=!CZ054=pi zpXu-&|8BUmC};la8e$^tRcDeA5a0;pIea33Pc1Rh2r+IS(5#jCgvKv6b%FG4Ds{Xft^kwa;|3yL!FB@IEVkSbwd|U>&XN zf(L$C@s4%3uX#m!wHqe(+;BNTUSTLnLjL?1;FQ;IFAeibm={305r7MCHS&5bu*Xy? zP1Z}Z!`N_FGv%t?;!`?r?0`*hos>;OuB_kO+0wGZ(5T(N&6Wp3UQy#qXP#}?sK5)7 zl{wsI%6pKJ=Zs2y%5S}+vK!R-qj40CdK&#~@pS&@aOUXpp7L)w0H5wYBP;SlyYH?$ z4r|Ae*WA~e4sOE?)g&R9Ippb_HX10?XS)b+E2K(2t%%MiN%)f`Oqxh%WU<4g8ttsK z*qn6VK3~HhRg_Ikutqn@fX0rOM~HtgZ*zO;|I^4Ws3Svwls1<~NECXdSTPqeIov$8>=-5SbNE`6o_ZR9GVSrV{F`SizP3CK0^`TcTPwSRR z>K5$QfjyDxTrZaiNtKizp8FtBuUPQ#@j!btpv-wLWB7&p?H}MhmcHs4RoYI;cTYSy zeq6kwtj>;A%>GW(0kNRARvQ@aq@Y6EG1h)_(6BnSdJ+k)0P3!ohdrmEn@Q-<`#pn} zb@4#harZ-fAQmr(e50+CUMOtTBKT#^A{Q7#kE`3UYp4}0ysQMqEw_0ap~b9>x~lkZ z*mfTgH60mhi&HeKzQLiEB$?Tz-`0>Q+S3~8hW5)%+*}?PoDpf}J4EN*2^t5lh0W3G z$-cPy?H^846j(j#yM_nZ=JB^Fnp~68x-pNzRM*8t_pEx~6DN^Ja)enNBx`NLSzg}n zv5NqQUiEuXy~WmBGq4AGGL`kS8AvtG*)Ch^EqN!%or?OcX3zd0ZCx;H;#g%~X1TT;ocx+NGbNlJ1*L`x&Kg|&2h~EubjGSy zxP5+n3zN&iZo6T2Klc^hHZKaTq8D#ZtR=qKX4lGDR=+Pp$Z zS;Oc;s7e^VFd+Yv0|_zF{iWN~PyKm0sMnHu6o?<$SmbS=3{`NYLLC=pO4m10TCMq} z6CJp&H-g0tlKqa|@v=Ex$t1l3K275~Qnvu>?=Ei;Cjt%>pO`wBaGx{iLgA>VczMZu z7mrNn(f>g}D!@dEpq#o?efz^k&nvQPZ}6&1c(<;{Vl8Stm1#b3z5hx)mj|URi&5-` zwzNgDTGK=(MGffwY+by2MEOZC&g|z#CXgGm4a>xh6BAvYga9plbv`|x=6+#=^XazF zrGjqeqzc@tc{iM8uB}&W#FjGxnxs5xcxr1u3M(3PDP0$TIG5x=w-~(7iabtHPg{SE z<9b+#J?j#pJ^6e<>6I!7rePeucg06jb!KOBahYRo1L{o~IY`8WziIgCXpt?)4bm~Z zl?L|t+10sBuSzx#^(G9v*BIAfyz`p;>{Toj-q^Y!oV@&&8ne`#bdazeH&SWeY8S)s z3tVc_q>w20M-OK8U_m{lCQk1_vvO*f!DTsLztzISWBxA;bmG=uIDZ~BWZo*+w4 z&i&+TF}>gF))45-4vNa+^S0XDi^u z|NV}YQe<#|fFY+U_Kow0chi+J(I2daUg>A~R(8BSUOh)55l$q66k^@kXL>XLM@qbV?v#67To=)#h?;=E$dSko|SqCd66*aA;p;4ojq*1*IGNO%CRLVLgozulrqvha7W@ns*)^Qe=&ExpXbX;A}E>T7IIB4 zk#=>otmCZIt{=`HJ}l^e5Yrk7yEx6OE)sBzV@RvaIzUD*s9A624pdh=C-9;{f&<3u zJvD||$cQhs79Z_lMozJ~U!wOUzz|jQ&Dz_ouU0Sh)|{H{$EQbUCmWpNuC6UN-{Q2oRe`k;M32yO2iK43kLXlFkBJy{&J_+P`2W! zEQ9@*TL*QAz}!Q(*l=(uAlv)5-nD7$6){YJ7U9I1N6<5ho>g-CMUe>U96@nzje*5V z4PpYH0*i|^6Yi)1J-)z2`S5lU+d%%1S2+cGQQjf_Ri;%uXhNz1N2CmhVAy)%N^Pd< zEFER73=sBjsU$2u1Y3xu!#O%oxC7@#9W)s6g1j|31D+wI1dQ?NVT%O5TaTJxw8gW6 ztC7_uHl2xaFQ;$Q*4^=Y6-#?*3pdLBz;t{Kz_Ks-#S9W2;*SFJ6Y%o)-rXpTzM{Qr zxYG?)B*90W<3YLjd6+Z5tWvwy*uiv<&aFC@KiGOwuBt0)*e9Fz11G^4H412tC4rt@G!xd3P~OWOnu9|#b(UaUY^ z&aaACyL|{o0Df7NMZi z9l3{bZDTL<8ocDNEA?Gj_AHw%WLrcJ>iEjMfsU#~w$JkVa+=exeRK?NScb7h2E80~B; zBmzk@5q&sz1Q6{fB!odxKq(G}Z#Kacx!@Gvl}g{o(d7U~qtt}K{3zwBt2RgA53b>= zmG{AH_>9aQT+gd>dY1kl-l1abAG}Vq(`i+EX%pTmlc}FhwT?-7(_c!)kMd(rfB3AX zz#ByN>2}0}5yMo(d+nm@9_3mS)+oftg|}lbf~EXqNFwi_-vP|+n>jQC&F9!p8p3}B zRBxe;NoEb80MPWsidRFQn0Gb}623(Aw2j?vWO?9JVIBltJD8YtE)F zHI0YN@t9H%r>3y2G`EuUiN*VBJf@=UBqDi6ihV4&-*y9Hz`pd0}?vtS7 zy<2>Qm(mJ<3OJ+22I_<6S3uKysact`*yKV?mfG3xk9yKbI_pdV4>7-^auo*)4LNe6 z8yRYWjtq;v#P7ABjen~(>g7@U_5J}zCjzIBV0%0-g1U0c!S>dU4=X&!@8dDpYnYl3P+HcMy89t_g-KG?GEfLUf&@KE*^KMCG` z>lbyXU!YFdUB zKMv~f4teL~#M(jXxRswri{^vI%ss+ZRQu%A@Lu}}LmwCqmK)U@S%KU9?)Ma3O_wP$ z1#^&I?3cJ5yWV7O{|FRwM3O*KbgJJQ=S8#+*ZeX5#k}on<@1#G)*OxP#D)d?etRUT-d`<)k^M#KY!jrXz@i6#;|i$jCfid2xm)|S$)+GkoR8^UQ)T++J#pyxXlT@N~?cFdw|lwXCFaCgKR}hI^k7%YWBuKs59t6 z4#Rz-D5$+O4{ndTrdMQS0bdB}mi4gZ_t}`icjpwDdt>|^vIPvPZn$H%fIL}C{}ysJ z`z(u~*`?H)O;MswmhYW8WHQ$k9fb$zWkqpglUaLV=tiHNUXpOCjp+MftVxy2z38Bg zsNvTFhGkbFtHW4YV^7h^Ww7vgr+R&0=2N?lHJbs$u?>G$)PWMZgme~UUH0&}yZzTl zR#cmGB{&kkO3fFtX2UoG#hb$%3J?c9It+t8tKdNLjs@uSQ4MG*yipGbFi8=Ts5O74 zDuLYc^Q5KsCV~|Ro}jmJ+4dhM<+c3&C=>tR!z`r-A>rsVGvBz=VRWgo4FZBtsc@ZY zt;P^h2fRtS{9UEQXx11G+2+VzxMS~)aONv#@>3~I>kP5+Hv+@#U|mf(S9h)2z(bDz z|6%VvgPPFxe&KBaHzMj55d{$&B3*h1u}f8YM?t{QA_NF6L_|PEL6KesAxQ5vgd);A zp+g8r2>}w27Dy;}MfZ7U?%AGupZDXPIbZzXj2go#|MjbKzPP--M?D!TtXI^ZqQuub zJW#YmK0Ou9q}}neQlJRWGucK`i*>xqKS(K}ay{0|mSf*tPlJDD0rW(buUk7I8~B(V zX>`5};8{yUwxd!VY2el2T!|}nOqZ}58L5iTmETd!4}~4){B>Wyx<)^e0S<5#kO7hg zgTHX%u_c~Z!`lalZmK$3KPVLOq5qHMZl44ah^Sb}Opn-~_!O8KPXZYl&o-cLv5;%b zhAJu?>L;&u-zzF6nm8Kk=Neunaw7#DO8&ak)H2Nnji&CRvbKTlLA@~-JTWMx8FkJ8 zJ9lJ=M+k5fr}7}EO{tnBLfT%|&GyQg=hC=jD3LhoWh^n5V>;Zc??qJzlU>zW`51w5 zpd=#1+_x9#$+$Y{59^~+Z7G_-@LivHUn6~}4^|WT9hk*c&X5Ggv@0wmJiC3FRTYcZ zi`-7nPN+FQyZr*V{^CM9jEludM8%L$WOneA|6oCT`&`K|*Cg-tTw8@P-7f^X=879F zxliHB+;NOOp9FtKCjfFKu0XY@ZsW5(rFvYn+ItCeR|;xqZ9ZM1_rbt|lYgmwb8pd3 zZtCvtw&zR>uUPAhTGY%$dQeCCdLXc`KQ0?{39j0(TY7#h*=?*`1Fbp!NR-PzTf5># z>iyMMKAvj=!-dx9Y?GQ6`8&f|cWog<=Vvce?o=tXX&QNqjH2N+RS(d~9yVp&qAQjS zhw0_B9b1mYHhV)32~!Qh>>U@9JG-#LNxyEtTd{QbEVe>OtNObOzzcJ1W0-Cim7n4> zeXwHE6^Pv3VqaC9tY^z=7qmP!hD_SttRZxQ%V-38kU+P(n8tx2U~u--(r0i^X^8xN zv2XnZw9)pT=W3s))5n50QG!tY1DJP&jD;l~@sNnD2e`%71c9Kcz41}!Ok#Kj$CI>} zaS6k=vgy_sGXVBM7&(g*Bd-HwuRA3JuV^&oGg_Dzh0M)9QaV1pDxHv_m}T~<9llYz zpZ;LqPdZkNqfnK0)?Vr6JRId%EVtZ0DS?iwX}4GO+K8eMgd{qsoXdamPH6bIa!A?n z$zd`un~5LEuvU}N-hByIIz~PH|8k3i_NZJP9W?_zwfy%TwKd6xj@f$HqzuJ?JsIiZa%TIVEJ=42__&m^7n0Ofa-k<0Ah#|EIu{ z7`{fP#m>DwTlc2(s2F4$n8}5;^6jo%)k;6mA2d;?dTfDVr15c8?c&RT9R-OZvnVIT zt)>}&wuFG&wpPi%Xp(2*hrt|;8q7t?o0NJr9Kj4Pt0cQa?v=(|g;q@F_}9qU-V zO7?b{>z*`3KwS5%Leq*+nvT4n!7#L>t6J=De$WOBK<=IDfTwxfPlo{m&*Fl8B#qHB z9!J<_etBJ!0=@Qh-aDaG5-{;AlPe5#N@_e4MAty9S=@CNxa8eE4}IL9)&u`K3VR{< zUc8f3qSC%s3wMmnwo7CH979RaX-6ky`oYy|eb06yxTO^Z-ZB_U6 zGMBE%5ZF0CDbu_FC|XXSZF$nldh1QZN}EVc<4BHpj6}A@V5>U&YM57c!cQFRCz~&G zW*HUbv%OhnGtztD(6I(;;wSOP{W&`?T*-Z%tvApQM@p6~UjqI3ecy*N8el#;{}OD> z3(w<6xPWfV>z$(oSunO&5Bs?*@r*b_hRz^aS6)2(kJ8-(mM96t!n+pj;(K-PHINh> zMAmu->*pt-{|N`AajXeqlDa=U=Hgz_(ULd~AzjX%dP>qmu9&qXy5W|f`N^vzn#g$P2f zCk!1bN*(W|}FAM5vV|JG`2`x&g0p|XUJ z)j`H1XRknO#M_QUI+~V+i4;i(#_0c^xcJxEuszg7UBBK$Fsn21@;o^6S~fNA&e zeiQRz?at~VNUy7OUlQ1%t^X;bk)G(U*<)0#twGOv)(9-c$xGb}yZjp3OeE^YYT0(! zqWTtlFdy$fM&6W5>EXQisYakHoL>ZxD|~sF5fh-emGlsAK%&C$acS}PKflc1<3)e?eWCwv(nkMlYxuXH$$xz*|6vXN z*O&79?fYL}%5Owl^fmPVnlEMjV3lH+ScRJ6|1UQ+@bH7|%7}?fy_8_7js9DQT^EZi zS^mRU@;k~N*ahk4XrR(2zyrnq=O;KI!`bM>oMq^L{n)=dvi=*9@E@*z`r7<&M8f}Z z2>WkD!vA4b{eLyJ&OqJakivY(__z+ge|1d zhiAJk(NWMEyT6JYm-9vG>D_$h#qW=B_FC+QV$jb&X4*~3>|QlA9NHIl`1EsEY1K>j z;OV_r`7$K#1wDWH{6sH~YyXR@j2HKxa<}rMB@z}vP;QY>cBU^rmhM3T0b!Cn@kG`EV$g{E!?p71~;vK5UOhs_{1Q z&vZxwh1{_{6V1^Dg(i(?{+oB-exQk?qTbm@ecSH(IM@O^^v`ei+I~D@53@LowgAFB z9_8K5urch0qk8M(}~pA$_gFP9mnG^%W^=G81JF;BM%8JM~08Pb%0{J0NT>CpJC z6@jfb2PDM|J_2itnfl(jRMb`TK2$TlxT%^zMo;_E#9z9d8zGKWz-$(GqpRGexq4PB zr#42XeTxwH#dXxOLyH1;&J%xHX_k{tY_2aF$!^Y??BiZ#Gk7%?CYCm4y}LW^JDTGX zApGfaRFID{;Dkn(Vi-_Q@ssjpb(=toib>n>buH~#1B-t9Wp{YeO&tfp3?M| zL$E(n|M0MgD3^pxgt(To{MD3ENrvr`Ga*|_d!6H_x;o|^XasXf}c4v?-L!>S+4 zpBXx8C$u*D4o6>~{Ayrpx(12NGgS+zLi9Ny=`CWMXle7cYD?#vP4h`|X01&q|LlM_ z5k)of)D2&KV+nf?KCK|D-$XY*W~ReeV$Q9uo{)LLt(id!R)=)m4(Orz|cC8Zx+9s~XJ2`M)<@%yX)}X_C;IhGDS^;nds{OuHoUU?+(PkPb_%)RH z8J_Z3+Ob+Otl+JeYd|km>m>T5q)0fLTC7R^Jtp3LUK@Am!#BJY!wA0KVQi6axZE=i z@o~8^_npg3gu#`$*`w+_hG(9rXdfJ3d&ZXd#WC+Gc9g>%+Z_kxlzIBn2lxwImN0O- z<D$iUjR{gf-6iW@HTM069OtmdjfzK)DQ{mz96C}EBxJ*1rf8Iwg4gL*$pNhq zua8zkM)Pe7Dk>#D;;Rb^yP2G22}RcRE4R#e<{i>iq*SKmp2|bJB9rNtFKCr zd-yMol-_Z|%tQ{!-|zhMpBv>F)6=xVD?m#J3aS*;+ytpMI zez6sQ`!RCx`S4pKj%(fzLHIK~k931!958XZ^OANvatB^u*GzLo_rUSQZu?=(75h#M z(Q4Sk%}r zeU!DSzrmLR;rE~yZq;xN_TJelP_^J9L&9yJpHiaDQ08wxJC>s60z78h4Z=WCg)DLx zYQl3bi($*2Tsu0=bV-AWg5Gvh%NImYi&G3@Z&xCmXVpC2b{E11)bWNDmPR#cbIhj% z1mx}|KjvW(YOq9)9e`3RTTQ+c`#@Cp%*2 z?LX&9Gc&5N%Q-0tjb&zU<$i~y-ECTCsS5D&@jE)uZ@Nw$dBKy|e{fvkq$vF6f%wo_ zKAZI7xgQR9Ju@{{qvh{v{_dy1T;7zZjq?s&TO2ofbFvdEGEJi{K%XpbMm-sbADmtS z0Q-VY>U%adiwN~&Ph6KS&XGQ*YpO-rs@FQ~sf;dFF62fmf^=+-9g7s5Z9ldrrDmle`ig{hrLk-d; z`%m#Nqf{yiBkY>Hl$(MU&k5+6jws`LHMhJ-C4nl2ZBagp&T6X3cjh~ zB9!Q*jAz_GSC?P(Nl;c;eD8MphGSLd>h>0FW%p2%}h}owTE8gU}sS#HRD$bk8p^!3Y%YBvgR{9p9f|4#&dZDZ-httYF&LB zaBy7syjdmzIvRr0&U>upeG%jczTY`?ba>Rb(BhDL%J2zRkKkF)+useV{iHFiYZK!_ zDgIuQrAfY2aQcf(o_UES{aF=|1&2`evigV2FT_ic5QVTdyNd5`1cWR%`RMj$=(D%R;o#Z7(X9eNS+H5^;QRL%_KFdAm1{FDJn{ULI*~H{f&; zde#Fg-20<~+~pdcZNweMYcEmU2DV!wd8i|Bp%@0DLDOJCrS9^05uh|FvdZ&PXJK?Q zuW2u1ru;%kU)|papU5%pS4|NQWUvFJs8JG5yP#vWy3M^MLL4pVjd1!z7W}oFjxqgf zH{B+v+h(e~hKDYG;DnafEyb8-)`sIhGQlqRl`%hvn%}Y?%s>koRrZ1mXsmE#maa~* zSbv&_IMo958Hp=f&Rx$iE1C~rS9xV3(iJa&pnC5Gy{I+$vskUlIM3^xluNc^@FTw& zHJ{kCze&k?AL%IHGB3%Q{ZiAwIK|_`bAFB^*xKk2qlI=ZBQfRU`FTa(w^S}YS?P<} z*x!Aw%RTJMyZI5)g4EN$^e1w@vX(fqox$O&%{Xrf6mHpb47J6SmHX^;l%?pr^5iMw;RbJbZd6|ed7bnX@oR-xYMvMMlUCfKm&PW3(bo#8rd1;ZO{8;| zIpaes4z)7ifDY-)Gno8UL%-^0cYc%=&I$^GKjM$j`2;<>zvLbRN*9v|M9`Ywud@o z6LzO{(2g66MjA1+vo3>R&LLSktLo&et(Q6*&?p@dQPMb% V-yl~H|RIYd?>T@#z z5P)yc@)%{hO)A5aia+Ib#9YC7v%HsGJvC%5Uz#1ZM(nLD{S%>+_ua6_=_8+&k2+w8 zVHJc#G-)+Pk#?z+Bc?`C(4s$h04_R?C_t)A7d z^QHRYm?P%aBnzDCPmL~p2r7Iji6a)W+22Z^=+`gPm+YCh-;J#vr*JD=PKq8>U70qx zKdXLb<^t>phC@;p#$RQPn7$zExl+t^ye)G5v5{k|T+pbhg1(WGg6~&N6`xU3Sqn&| zsLx$S>Bu>ZRdpoN@@f(z4w1VaA#L8o@I7`rJbGOt{j530>9ZM0;gHwu`N?`{TZozW zbXHz5t)j(2bD}I35Yjbv1MB+`81`cT3S`*T)kSj^4#kx5k~b4^MYpVzx+l zY_Q<@WoNQ%T$CnfZ^3-KUAI)WV=3WVF>(c6@7aO#Y$B~ZiFmV-Z&F4I2W>e+_!wzZ zBnA*AV+*cFYt~D3v$dOqv+RA39pxPBvR8!v6Eg5KJlLG!{3FL9^5%>AHpM#;Ke) z=^9zHTXB|>-KD=ymg^JROqZpAOI6mg&*O}uQ=#GvfkqKS+D<8R|Dc=(4;*sfr$ zSjg8JNo)k(Ye8rhRz^z$C7An^(^qj_AUcF^shZ-q)xM@ktH=oKT@p_^LuiY`ZZmJo z7I`awptkp8w48npmZryFMbKO9g%x-oqSGc5F7ogXO$G3?b7s`WMaJa}eu+>-F?yO? zTx(mJ+5x%tsn1@oNWD#eT$;lf>3zep(ht^&*JzzI|qdyb`A#!?DM_ z>7jucj4X(U<}vbkbG`{dc(wgH+P1Rz;~kdyp)5-L%x*QeoX@RF_XPp&v$Shxo#zuiv-MmjM&tbJFYy=?UZuH4Co$>O10$SX8#`BLpH#^g6Ut%RBAe-pOe^0huP|Z9Yrw zu}TkBBLwG-OsidldZg>C**YbjM2JI~A*Ziu-z1yKjnd$Z(wIBue!`GD6UmLRzc!iXbO8}Hdg1Co=(GODiZfd-=QzR9S* z#>{&rZK4>yX$j5Eb+GUF=Rn-ryJ{#l%OGh9uLDXb7tyWb_+n4Agxx7+9)G1nG2(H1 z4!Su47IFE>XyPAVfPU|bies0%(1`2FovuJAR;AW&HhM7q-X!>g@a*{7xWMx?Wo6yY zq&;~RB-eQ#=WopB{;5oIF$^Czag$ZE@t-V%Iu)nadDw3>OXGb$o>id%Tg0_j#J1c-qV>`xh_kM3#I7&xF3|G9(s?k5AcMf zq#u>#`qD?GmuKBRT_YDe4(hGFuy7oe&I><6#1b_)e!&Z(LeE6~%2?t`RawPLGD_1& zB+4dT@P{&zx&}X}0e%YhbHF-p4!Who&6x`JlYGm`0yAFj3ztCyAuMmBk_#+%a$n4>HZ+u5TphG7 z6fxM*iXSt0j@UyDcj$ zmQ`tmeba6{(?8p+riE-XHNvBkuG^hyM(QZe9g%}S6-bb83_>sNU$yR!T?ftV@Vtr0 zO4G;t&C}cUcr~&6N@K9Ruk_Vd2jh=NF^;di&c}ycTt)$7LKhcO zG&O{YOSZh;-mCus;hWd5*-1E}w&CFT*gh|0gT@-vaozacBK|4D)cE=I?%OldQme9| zzLD!acX<^Wj_qfdb(LqyB+~X1itZvrQj((P@i;h0&!$#@#v4HwVv4LwP>f^?F6_Iw z0cE%19UU`%rgGlK#c(!i&l;yByNZb44NkM!J-{8skE`FnR}G$GP%N>@8S$pZp4aTI zPS*DY&n4&3>8npDQ3A-&T=ztKWT=Jw zT-U7vZ8k-M&5ZzOly(9^(&~nlvszSMNZZOi(AewP=-_M9fm9GS84V5 zB^AX0cE-@P`1GZzWq7sVZ_>$AgqC%KY!#>2tF4~+bvBq@T^w%Ig);XEO}lUjxHd|- zofUjQoBP(3C<=kqImOpl&CBG8lTkE@i2nT~;W`&Cd06LX@4Q{60bX~%kPK<3jd0r} zk&uM&A+zqkAJS@99-$7eZ?lL8T>2^(>db1s;vnuub;xsEgYk)HMcQKcR{pX*%7J`wPPAdx^`!3#SCSou04^$?Ub57rB4-)_XCUykCb&>d%5lbt!3yR>#hexj08V|su%$|1wRcIJw`h(vgZ42yLCpUO7MnnUN!D{ zVDU-H_q(oE%{XG`(E8a+yvwM^;siGtMBv^t6s)RXN@V-cW6&1lUa%^D`s%XON&{2k zj{Std18kBIZ=V6`#=6W-=yP<%0Vv0t(Ne22?-Ig>IF9BOlAbKZ!c4U%H>K>8L3R{q zb0ahy@x;AoYHaUv`qY;^2>w(Ho=bB=Bh(h<#&Sm2wR0v7QlvmQJzwEoI!PCW{=0HT zV;`ou_l%XO-ovkQYaqu+aWtCP@^z%qA{`v7l{q0s^6Wq68^RQmI8T>7{jnm$B|jK^ zl$l**NtSmfUtW}y#5nG%zCMRu8a*~16mpu?;Cn%9A`CK;Viq4_)cfmcuvzwJxw60! z>17+6$dzpp_A3_|6`-~TyNamkDNxg*@aK(^Ynsf^>pzUs;lBukpI>Yb^m=!OWNawuwCRexF zV`!Zl3EsEbChW0%tZwtS(jkyh2if3POo6X812USbwr1nv`_&$*AsGU~>)|L( z$li3z$F^kY(pI6eRP|@ol@kC-&etiXnh}i+t6X_RlP}XYk1gSCP{~!TCHmi9VyAbG zDTQ-FtLn}|O==OqgKzVw#c8x;728{2W}mlHzC=g;iSmzLndxRVe$-AM_(<+AH3j%= zy|6v#ygM>CS`R*Hh&U#-hEJm;$-5QD1_B^M?iN*mpb@vj2ZcPd9XH?a&iZ5yk5Xqo zv+_2W-nLwlhK^jPDh_tgPx?$rj#r+nPMC4|LTMFt+gQznUe|%O*X~p!R()2T7_D>V z%Ki?+`3iMXm9X-LxRF@`ImC*NSRbCE3{1?yshkGQ)V@v@S;RQWvhE-o!^w zM4Pcc<6a$o6#@aa@nZhLKH^LZzjmpYkF_DgEl$h)S4p#$0#BWncS z1=-R!rFL*>gk+&^tD_aHa{s;+z|OdEFr;cHR|49vbSrO@SYPUBOziJ69pk-HY}52d zXbrsE-t3e7>{q2OLkr7BRUZoBt@r%OO`G^$_=qTe*DDdwK_ycI)eCFBjYo*7O+^W* z0M*~sKV-V0l`bKvQgx7~>9(YpqEpzPqhkihr^rW&zs|nnuk`%BX+q?lb^jS7=7vJZ z01P4$FeFsOQ|@!h7l`e;pR1G9D&^y1UM>=p7*+@$0`fYa<%K!3qGBogtQHH;U>n=4 z(%yLIz{v+`RehdIe`R!;CRnZf)VrQfzMegu|7R7Ck5iaW3v7q6?X3}v3TN zV-^a8d)Ri#A*IJIB&uQBBNad*Rw^ji;2 z_->Taf*FM*pJiwyQF!?yz}sRh*QhKbn%{WCSLT3rBFoO$=61073Gm*C(Zn)_uU+ei z?TZ49LcB65F!G?e_dAw#NO>0oK+PXOht4osjlJV97HLqi2-vsJqFGYT@%2IN|tG=T#|D9qyq9T&4SacddpXS$~G{ahYXQbOz@M zhbc6auvfw|xlKh}iVI`XW8)lR{rMR+Hmy2p1Mxpp zL|IPTb$6G^>Q|cjKj--K$bfMv=`w1&n3UpQBKIXjC=9GzB={r;YVRIfid@J=>$3+r zfUYfeM0Ih%VVA(MGhDyHT>jWkuGN1o?C$*Lx+ zzqL8V+S~*=NO8CTC-Q>*bMiiRd2TuTpa)aI7Rc2cb?+^<7)65Q7KlCBChPcH>m`19 zz zct%zL8p=G2IlEkBm&alTzql2XWnVV-(;K;b>H0*+F;#8BH*8NSR@mBijL>38!{wa# zlV89+N#&*jq*&>JL6=L8mtK(vpQaucnE_emxLk4(obRSkqt6#$QoSTkO)z( zXFaHW7-6oddS!Q&=OQez#ieYOYwtqz8)30A#t#^)u8oazDZ0GZMHUtowJyPPl+yL) zip}PM@AwVBeukBRFLCRnfcAsTm{k$|;`d|j3qqAw7iV*QPhEYQ#@2B7B;RM9Go#XJ z)Y(1@*w_27&m37T<2F0h0$#;k>YV4j(wfcSI6fL2S{%{iS|OF=i_+^Z=7RcGQ%~4s|_%i-QiO{IKYMxTps4#W5@{wPaIgX1LK0*4ZrTp}zoq=sk9j$^~)vEX09SNN^v`f|y*d{6=;;hm_w>G^Gz0e0X zkB+sRmG48FgM_uHMpf|k8!@}H0uLg0GgUEU@;^N`PY6aEKl<{%AgROX$9d`w7f-xLZcjR9>i_mI!edZoxem#SoD~Ka>18)BJJUyZPXZZo>e-7^Tufz<=R18wtwr^_+%Yj3#m8TEeo5`Xw=2m7N1d_-qpiRwFZRHh+ri#*1Sk=k zR?>C@@p{@eH#ww%{;Zs+5&lJ?o4qhdk^hud7cn9^MC>F1XIcx!S3&8#kKv)9Od2BFo8k|xGVECQW=gP01y-h@u$sKV`OHmy^-&j7U^&@ z{TP)v5<#3)DE*2{J6c8DBK8N*XLh|t6C>UPFwaK4VQZr|f6u%T$bAOz7zA)lcwY71 z*Q4sean;Lekg?)x)xPz?%zlT6k#40?@Nj&Sbn)4`~^BL}wU)TM5YDv|A`~_k7goganbt%Lune!>u zjeOm7=8d)W_wv6ghe{v$NyBGk(vjtOgg=~IYm$#wVF&yx-nLQ;KWj)tYD>jsgx{2; zNVFOEJQR?(3!Yv&io-WBj*sQp;|UKbYQK5&2$x*{II9)h>7GnlyL-1v_-qODj*oMf z=7~!i{Gf{cxhPYF%?6yEXq(UBp$WMXQn!~Epu38KBie$#MHFR?!%ai znkw2XUj(d`StuT6VfIuw458wxuHWQfAcQqCeGz(}HhQe>*Ma-0^tvTDY_^^N=IbWo zgyaM&X#wCk_s6}LqJ;wgnk(1>JI*4(=x$aOg0e#~kF(qNcK?=(+E!uc@E`Gk+XqVS z`OI=W#s9*k%g=c%&~3ke^N13?WH{>0&Ug z1yqTFYqyLoYv6udI$9=3bB&^shJ9Ucl+o~coq4QzZAY@$$>4*#L=*_nNo_>>rT{r_7uKF-2T(h+y z?x=w3#uBAvm$uM1;io(8JI@Uff&&A% zMJzgmG10_pKJHCszoACI?@y@Q2n)z-vhdgpjNw!9w@i)=8J^z^E>)BtXhCG?1dM!J zqCVRXYLd2WM86fx0D!*(HBUcmQG z)B{5OztEzz=ivWEfo3pP=%A{M$knTwpdM84F`K1Pr9*@(NZxcE(Z;*u+y+|@P;A35 z+LRXXs2Pl$H~mx_=aeNb`z}es3|O)SCAHUeKrs5VT-gjB|hVhHi#G7$QzzKob)XbzEw8ao&~U1J~*&8l7%rNvL+;nIFV>%|}EK#TD)H zBHI&5f58I2+nX=2dh_-I?y*|@5z4^XEg&ChADTp**<%* zeTjU0v<&A}zZss?&R?^hS3mRACbClw=O%HlIh0{bV`plx)^ev(t!`914mB5ux%+Ip=3p&21gy^q5Jwvp zsx`f1!FVH|YtrZQObHCe<_kNx`jgOOf_F53_quRU)Q+P&&%Iz`7v|w)I7@#u%f7eckn|`R}-9i zgw?;X!>=`3G>GBNh@7c!^yY^e0I{^HIB3-r()J3Z+^xs5xJPIXX6i7N|I`l2uAE|L z(SdMo`*2h2YRbIV&P^CqKkl&1`y)BGdxw6XyU=wUhn5zPalBYtyjuG%_0KAz-g%Oo zTiM-sNkod!V+DYtm?&znJDH+^NXPel-I)?VZvfSoeH3&!f6P=b?1D6zm`efd0C0zI{@`^g4R@s zPlvMtMnI&<{Pi)DO8@K++qh(cOM#y>B|M8Gx9>gAZ-n`;c>KyjsxHxGRa*D62=;6{ zk2(-48NP;!w3*j+m50pOH79bwi=fbvCX9Y!e}Cd%SpZ3tkj0V<=uh&J0%}UTFVE7J zFID^q3jPCfF_Mgokk&D7an{OnP%$U_e;uh_3$mZWmLev;_J^#nXH1^ zSR+~)d?aqG>b?)vN<1vEJd6Gat)ZEtx;$<-YQ@_yyU!J)j4>OfCeR-wj%8n&qzcBd zXttk2x;Y(bOZgf%h&_RkX9$icVasr(Y9s`TKTvy9JDRm>5oKFRDGBQ1yCA9X*?`Q2 zpddJ(_2$1vk#~SVU|b?5Nx;I{gvbv_JavZbO|h-rkTZ5MiSU@i{f=e)SE{6aU44vb z*GDV3k=cjF*F>K(O^l)VhXBU$DDR7cyK4<1F}{g#bQ==Emm|fN37KnSGT-7{^a5;e zp+f@)nltAHs=s-^B+CQ4S2}Tw)2{gdP6r6Sf%9~;PP@5E2HXCoSmIIDs2u(!X=g@xR_Ug%mMplAS zv{20sI=kpqjyc&ZzkcU>iO1bi)7uV)Lh*;8WJg>Xr<9|B+{aicc)>w<(xbAQh*=H? zlYyT|8=cJty!B~J_9(Aybto2DFaX3qLc)5LK_q3Jf~ylmN&EhB>=MvN?%!aB2k_C( z*^c`RA#B==WIC(r^5d#K=W@#Xh};Ej8drHc)m!G^Q3otepIh-c3^2su8hpYwGC4zk zW*AJO3;u5J#9^gPG0*kKss_zNK^Bk*>zdGmX=ijEP3*}#ah-El43`AEo9u7gOPm6# zAFNTtWxT&45=d%w#L(Uoa!SJVY?gXr7CLdqxWX%T}d%Fuypvw z=D}6r)p}2oJ1W)$C}s(m%K6pFDkY6S1?s%uws8EJqS*?A;ZfU8AL6h~`E1_tWBkUL|a zl0_Ox!)Rr49*l5e4FP6vbY7sw=w-un=EV8sY_q(kv0~3S!dZ=pJ}7c$n0aGUL+=)P z_ruwM?@Dq%3h)RS<}xQd8^s~i*V%r+=`9TerN=jH&>A(^J8xU~`^ApBH81!0RSx4X zbuOf}6=8NZVA;hM6{?co)sJ^WmCa<3lP5y;NM0SRYha=d`BpVM#%rnS>fI@Nq9zVU zo0a+v9L(nj#B^Sb*HB&2tk25iW}ogiErkMJh$LfN+xa;Ai%v}iq1*zo^A0c6-q4?@ z^tn-eZL|r~?Ndu4TIA?oqGUcVqj1Yi52FB|u#ZMO9U~5u0M)oD8=iB>ixa*^zm)UW z1(~X7YWylqUQ*gl1EbgwY=0ECaADfnS0*nm!$=LSVpt8z=;HV*H7+E#2 zH7;7D=(%BK^ZiG5NS|$PBA3en`ldVoy#Zgm#mOlFkZN&4~`Ct+{Ht4rf@R4$iXpq&XgYP8@+c>kMj$ z?|0v2f7<;866~G`^R^~$AKY=xZEpCBOv!$su=LJuNp~=wF>d@BH0y3vP?3c}e>0Z$o92d7!4g zdGF@aPAd`P!WWnXDXUT$#q>uVOs&LHn-Ra~bpo(;+jz1T4f2Uz`Jk0=!uyTw>T)S` zPo+y1)pDA<7`uaktziz*dKr@ULzyisW``}RZ-j2KAm@EupKA{;6s(DERtEyAM59Zf z+g`QaO7pZ1jaWe~y|+ug*KXg5IM-ZZ(J_gmfDzqN%oXk{ zsLhr=?A_V3a{*rauKf)lB;;}-nN6M>dF*wJm5){@;(|yrIU#05Wz4-e(0>gsx7r$& z^I_n~ud?xSp_M&Qu5o>ny(u?Ga5O5MUEO}S&Qb+I5{fu>&~`Q%E4Tnm-jYGPgI3~B z{N`qAy2WKB*!tqzU$w*+7%X{~8^clTndIB+kr{&4s5QY0sFg4U{mk?)d15@tvyl))fw*Re`m;=PC7&g zL_~mO0o8E@*%@$~9v8}y_gZ|?Lp)moxjz=6W>KW$J;JcJXW6@F$?x9sp)7QS=Ztvj z#}JaSTFBYqTNLlb`p{eH=g0CPS<`JF)jECFBcdO@6Z)7pvx10=Y?3~p;_gg!g;+O5 z>Q$S4tz7cmb%&Y=04#cABv)t9-13IHyAkS;nCd#OfZF7Ztr_4vXpIr60V=W5=5H<7 z+v!bxQu!9g=y@WT9E-rvv%GYHFU8GnByio{b0DCyz_H0GV=zCBGz;Y)x)~JuimB(d zCgPI@P2!8mO|d8UMDPGwnDbJsW{MtOrz#v<#AC-dc6x!5Vg)Ff-JL-M+{|m$8J?J! zOzreb{e!XC6M)j#5)xntj%3s6?;KW`>k7jX2}#HgVWG+wEaJC%g}o%a^*ys)omVT{ zE)hF)tAQXrjHS`cLZs6Q={YasS>?)P%bvJY2PhvoS)Cj?3T(C+WL0;$k-ZSdq|!35 z50QcEwnip^%1{kh2_1!XZ8YCpnK7^cH9;yU`0Xr3%# zlj=fct3|BbOL;J;=yVCtERf8^~Yw_}5m>rXT6x#3d3KG>G4Nc+}) z{QQ+#n?c)qNt%%ur6$cY^S%mlN439r$jTz9I_UCtD;{gAnibBh$|*=3kTKU_8(F>Z z+DCNNIvUY(;x-Sw^X|-KP+_y%pqap|zo#cH0d5Vtpb`rzsc!bIpZqYaZ~yX6sNGw7 zDW8_%Zc%FaZq6eogvH)_uCAT#)9}2ecBO)6ro&|ocWiO9XhZz5%;?>wQuAwsu(7vp ze-nbQ-2CooLNY`3l-0!hpOET1!->8bHcAu$?eb5l6(SMQ!wu8#4g@`EY_C{>kKt#b z)-JS)a==_^B%c;#5Loopj;*P1vo|OM>I>U9sOHyOiwpU0C6j`x_f)nC4OEIJaE+AK ztj;9)&dx6eE&ZH&lNl_nRdx15lasP8FZ5sr{}7Gnb2z^+XM*POT$yGHnfco=v2f~L z8oVnv@mx)!q5zMoL1b5r%q3vSiEuD1GF)Q`fQs{2raS3fZ1w0-@hPBA3DDE4|6AO&vKJM6|X=5a_k38B0qYx>=dGQF2RG%(w$ z1rKnMCT;;!QJm}yE91D5RB}39G3dCBfgJYuBZ-Yn=(3p5NQw%4H~K2$;A~ME#SG!N zh9Dbpo^u^2KVVV0+s$dhdgPWYVh1s(k~+4fHdGsX^c>enO~+A}3R-L{)Dm@pO|k1y zj`ixDUi96l=l<$c#y-=T3!3j+h6l5WtZBL}DV=8b9X6kS`|?A~YYb3x93j2$b-$nv zz~03vTl>GSqt17!(!t^#?2o$Z_-cTy*>uH-08V&r*mbR0MQUiyXrl5jaG9##6(GsO z%MN_V)z}%&HZ8~owk_qlRWJHmOEbqrDngDWc62XwQ_;C7-=pB`O7Ayzp4)F0_Pc$! zw@tpcXs@Yd(>B1>Te}^Cm2!Px!8uX{-+b{;5c%7e+Gcj%=3pyc21T^@QiVyRpx15J zBmHH>h8FPVWW_}5#xYKnLd_vi8tsJBghm;z8WtLz>72s;PiTjP7kc7IUL6UPu2B%+vdv$x>|r>yS*e_L@ranVm1W^h*o)^5psDQ(;%d zxGaHX=W~w$R{ga;+HPc86jZTg9K?PCyUIFR(dL8*k;g(|~trabs&`Hgb^ARa}OsS@fQIt{x-T zrk~qu=SW8BeUFXJv=P#HCDIq+f{M>kkKHb8k6n$;IX$y@_rPOgS&jNTWq9{2UPJnj z0G_Oz-Mu95zqNes5nq;mEkZb|f7hlnkyqTNV^x<6cGP8xMH-a$*nR*K_!F2nb-2E4 zx~=-@6!=~&K31$m=0l626+CHYhl=j|;9ttqjsXn@ zLUQpb>atW`Ooh#HvY9Q6rf5_s06Zv1dUs-u`eyY)(g7R19^e+l9c!zO5nDS>+TsbAN3c^ z5pE8jJ+X|vnbuZ4Kbo&=WR)M`lpT2cSK>#}&y2z5L#My+Tvlvto{mTm>5OThJQvW>H* zd;Zp3NHYPvU{=9hNHaNNEdESZ{0y`aY6Or!Ju*&ll{gEZU*lbAV7F*>EW`81aF<(1G=~puG;t$`q&4Jl zGiLDd$Nol>jLB<;MdqFwUhAIY=iDc=#23RqyjBeQwlN5-99z>m-s=-T)$roBg*u<+ zBaRhxOPqIKnTBjF7Kz-t`O3IgSrpZ2)!pQ|y~)H%tu*e99eiod;>j1MUn%cKEq31f z`tn%U8%?1&DGd&J4>!IpfiWxO#uodAOm)Yj5J<>YEf^ znA@`vnsyu?47-)%er7(Nena;;ui2$7Rs@e^#n^+|l}mv~kCZSi{4)v2FZx6`zoVDv z_hUA1@?C9wg!MAL>(4P`MZZ2VxWa54v6rM^$mK2TInTNUd{xun5$e*!Rn~^EF)n_j zQiXJ>dN4XaGHo@px{nU}Ue3J`nK3X*V;Ka|E!EhJS0(q!g>z!~~26Ieg4DvB5E=bD?G=U%W( z*PO~Qkc+q z28m>8?Nk4&y*H1Cx^LfyuartEMF^plBD7dSmJ-UojlmdF*$vrulA@AmLCC&k9s6L$ z5+VD(j(s1@*bTCdS$LJ@OPyc(>w1_^GH7s1n<%CdXnJ;f4inf$(e)y`kw#hdB2A7Rmq>T zzmgWt`OCrMUn1Cl`_PDppX)OYtk0CdCX@eUQvO_fey{%IVV?B08aVg=zGD98Q=hui zop1~4Tx0w1Xa1WR{_kx5+x7bM7yn(f|JTC+X!igAX>TsbXpiIH!R+2{ef)9!_rT!a z>@hG%)aO-wqhuRyvw5_ZQD>m?82-bmdrLn(rWD83E zBs?q8VdUweYX0A@+y6N{pH-Ri?_F2w0wAoo<9PZ%f7=q9pD&w!X_@7c#qs~-j{oza z5!pYlh@t}{i3eK##tku+&kS;}l(cZbR*Dxt`k&trhQ&1CS?k@5Zb6K~f&a-J{%bq< zHR&<#{AwbJIVL#L9xiY&-GN2Th z-^qW{Fa??q!X%f<|fL2Q`>TUqiTM&fz7bFVw%|Bn~>&=F_gW`$yvQffLYOH7~DDK!8wNebO+!iM9JE zNhpEaR&4VSGkiDjdy)yq&v0&AUD3R7_#RbbM@0FeY82_b z_u&iIY1eUhM|BlcF!r43FD_hQ;I&#*&E7NS zF^=7Cluktx%DOzJ-GpKNv$Bf6md&F(V?YLabtVBt6WhoPH9QAMC0ATW+yjSPu-S=3 z%U|!i1#Xr#<_C!+FnQP#`=CNAY=`u!yJKn)rG6Au{YczMjHE)z8aaj}P;>;1d zhCTP39{3?eKWT{_CFEI!y}9wIpZ?^KBw8s;w)M9BVpXuMm~HdZE5?MCT-R|Ow(ett z#kPWQ2^762))%IDETq+A8WKsIr|I}6<%(&qW`elHD$B40>cOMJ=s@4tFUTr&h<*dM|J72i4?y6 zHJgsC_fAW_;xldHgS`C}lLh%!8xmLG86_%_!aRe4vu^);n$A9rq@SaGEu|eOBoV7T@aQNxs*`7zXYOBIA?h8+6j}7B ziP{a=j$T|GDpt$EtTrBCe@%xG=dgbGXt)0)T~3r^ui}KHAL{;P1zT>=SF7YWT{|BZ zG$rV`A%6>B$>=7zx~XY7P*&~*I<3ZY`_u2IsdEYuha)7OP>;uyH8eu35i*=lyWRxj zy3$0T!IVn_1yU0O(f|`TzmR=g%zpQ4Zfnf|Pj3Ykwtvq`?mF~BJwxm9xlx#wp)34D z2ba$RB3(DBVwk!ZshzCxZA;ev17|4bZ%{*(ufiCBxy%f+{DJgK*V<>G?UWjA%XrLL zT-7|p&3*8|l~5UE!R5szgC0b2X}iY z82CBq21V`RrhS|uS;mSxn>d1N+zj9LspP zr2dwZi@$A~Qz0uB*B@bYM#8>RIP#`l#oR6En^RYwG$fWfP|0kJi@fd6wSn9dip$%S z>L-R^<*X?K#Cxg7Fba*G#PLmvC-;Ft)92Hq=2Phx6y6mz0j!iq==Lb@9qv(Db__qP zBExHEVAd$}F6~Hn^4)3n!~%JXOr5+h!q$UTD&(jKA9&RTxns8U!jXCxn0P9l^es4L zJDRB%K~@)ODt?{3Cz1fyCEACjgOILkWbQ%Obr*M8EFphUG8gNtTp4uYNi+>IPRU4* zVA&HV@uTMaOdE(`y1Q0aK-szYh@-hW0I=7#m;q6z*w%o#u<>IB&Mn`=8?Ym|Tb@;m zILuY4Io1yJ_z+3ztc-4?wCC@KN5sH1FO8zox{*-+DL+AC6?X56HEoo^l^Q_O#=K1) z#y#tLv3Gl|I+1c>(~k(-6gzA`e@1TmvaNN5IXLca@!jnkH(b_ppUnWyMd%%Jioyop zJ|L#J&8meqV%meWZMcv9aK?8^Kcwmz-5C?Wn47QnzrJC+dkThkTuANtz}F)+j<_%M zd%yq`AANG+c@C-_M{L-iMpNfiBn$lPT#0VZk@d<0Y=^u^Tj)w#G7oN5MO__&x?Sqn z*#3>2qF#9?-?5Ji{TfuVD!pl8x@kt*+OMhc~JVxO^UQl$=7xs0jqGbIHRkxN7?`1)boSEPx zlF%lHtd9Gmrvf-~i2NVGrY1aW9op_hQO}~0eKRd5I3M1BqcA3LMMuad4%8swy2i+4 z!#o?mWc3KqxpzXuDi}6RG6?Wm=uxW)YHRUW#2}`{wl7it0&XO`pd3${4iaj;54<@S zClrn?D8-Eod!gM3yr3S#5hG20{RycveEjtMsUk@fk|?>6H~YtR5C1Bdn|kv%F$U$9 z!C;$N7j@K!clh!&>|r~h^}D|#B7%qKvJP4IAqwf8$~r;VXRL>F%+!skOYAXpYWC=P z^YVju*M4a!#t~=Y@J56(Uw z&$Iu7^v4KX~Jk6ngz z@!AqrY+JIbHF^@|NyR{0YoJ1gyk2k>mZ$WNTgHSKCWXE5fxj=w<&PP_);U_3d7!Ci zeRoCxJFV?#8dqWBxix4ZZ!4`t1SJ>)g>^OUKic<2(^K7R{e?M5)B_1WwcK%rY~#YA@aM5EmufvV1*9>EmX8*=|(`0mRK*V zADS;*bhy=`8^90k(N*d0o8#~-c2l(!_iW8v1?8|e2S1vW{1HB*)@VAB>~58gt5t^nX!w|hKUELRYV&2fS}9+aJWN}G)zH*_ z$FL>yIA7be%=S3)`5E!vIKS_OpgB@*uj~O7fi#@T%!*um1d(7YNTd|E^dl}2lxqJ3 zBQb*G0~O<80=R=hduE&+fho#1Xq{|!I>Xk|wk@}54o{2UK-&GI$*KlrA1WPw6;;i^It8MbqlJ-ut3Y%_d&HCa z`Ha*~EIDnTBl(NGjAK(YZv=Ey)|Jv=F5;YC3M<0w_1p&J&Z@^EabD4n47j?0S9i_f@U}QVT z5?Qn2V_QnM}Pgk zAD*L4l?czsmIuBsuiVtqj?`LkCo%sGC9t!e5n4`#)aou}((&eu4Zo#}RMYy-{q zE{j7KfY)sC?8=j##v5L9RiU%OzyrwLr~N6`u0M327QTTOBY5u^WIS6d%pj3@qR?tM zVZY=w5Q;kDrs4|0Bq?sb^v*+6apu$f+;V;wPnCc>v^AhR-`UesHyb8W#0>XVUs;CP zHGR%wQr-7yxFU1WTR@>B*9ZTs;EFhh<-;6&7hL6ixX>4QqtOj=L(ql51IIoqP86^s z{T+S^9-s9C5!Qi=4+_q*?6(6#P_g*II|fBI?91PHMVDuepK9@}{bVNgjd^slCPYRF zSUFBLM+)Kj$}@!I0-PL>>Bl!~7-Z_8YB!-&%PzdRO7!P3nR(D=G-PsaxDRL~d|JnK zJ$SQ_^ASXzSS5IwVP=SG?5#oFg-#M-#^*NsP)Fy+s5h}(Dh^0U(E1@?e4IafIn@sZu!jeYkSw!I1nzM+pPhkyldQPRg%y;dj~G;B-^0CeG0^aC!H=R^4o-QI4xp(x^El-^#uXElW3sJY z?)hox<7T*0U;CN28!oyfHKr1^mM!d%7>Ln+>xz96>b!9rgxt}>2w=B6A%9zHD0atj zFkkhM`m>F9lD1;ul?}U4a}Pkgs{VpL9}}P^@+CbsP0jmJMlacfh(yj_`B>j?np>iSi`oxePVc)Wy)}!hhJEb|p8aD+;g2Ei!D_tUoL~yC|+ev(z0H&=2qpp5AslAPU4(%!!m$^xn$|7H+!G6`dS4;&K`VYbS*>RGZtV+=Sx>vM7mOA zBLcY^&+~`3qOi-qJR<${nD=%P^x4R%_@`}7$sud1+AoV-%2PL&5=Q!}+z#HX89<?a;e)|TYZpZeV8%&yZ_`ORQlHM)YegRO!hO+`g&ZFrL+zXH^e#z`jctt+yZ1#^9!QBU#;iX(3_{t}2V>(^LyP7sCQc zQ*=gY_=%~mE{wg>B`DdZ`Y+lk4e26vYpr)Wk8y}8&k9N*E-TdVc)B2zPUtuu@(`{{ zB6Y<-0Pd1f&|X}(AJD%p)DD^#7q!t)+=*(ms!$z0YuUcECuEm#WV-2o8v$-}xDJ;I zz2dYq&=)Z86j@=?XnB_(Xb`3Am8QJ*>~^CkY@BE78dN9Kfk_#pC0gl5|m?i+nBx&9;+Z@=OGh?hZOc};-s%S^vtgji)z0>pfuLy%8; zQri+d%QT5?fs~_{D)~yKe7@0a+q}|1b%<@XG8eD03Wmp)?<_pxxO}D^8SzOFUOmz$ zWF_K+t z3~fZmVT@E4UYx{zL2ATb|3=f;9dD@~8s|tg(vA<68Ft=*_=^yMfq2KO9fox{@|J zR9R!Kgx%s=G46}J;)|6Y9vA^xf(hx&lcw;V#|h_#(>6JZjIkZKLa_LH=K81&?%qEC#N5 zoSe>L)4olze}^D zG23V!jA^i*i}%sW)$mczqd6AAzZ$`9%f1 z5H8QDFBSsWx#m02m5Vn_zP@F<+3!P1z1l;~jkpZ|767=2r7aw~Zk%E|W2%SbQ!&Cd zTy}F~6w$?*B7f>=2ad;i@Bt!8;phimKIj%TYRirw8>0Iriqt5~@o;7S3BjWFR=IfQw* z*O_hNo#)ZHZtFZ}E znHJCfGHuIlm8DS{V9svHsngX*0WFz$f|(uqP^$~Vc!}BQHl>5^<~3$xc((q>gPjG6 zeFw0C1yxeDQzS(Ox98!l#Shqk_PVLO@~5J6N@Q{G$89XgOsioTZjY2^03%{DO2uct zj*2*?$aIydAFFDK6oIVHwoMWyN{0c+>mu<89)AZyW^)lc))6MCt?%SvhVUjiiKTp9 zC_%f~q7Ughn`i;ef-Z@LrRr&ABjsiGuC=YHuC`>um%Ir1Zvf(dpB=3mWY<#!2qo+# z&Ud2>Oge}n+4Wt2JWQ|Wd!VF^%QuRNN|XtqM5#q2beUOnRUX?Bd;S;O!SL-fBb_wX zK1)dl3zQ*Hw?5}OnKM`E)-m*~K;gyUd(R=;Q9=t{M(&G2qdpn^^4L93zw^g6U-DR)NSb**yMRPeKfGkrXU zA5W+$zt(?|SoSgOUf)4V70gyczFaqCxNPJhv>0Jp`-U^tX$|C$f!4UR3`-L^U*Rlw zpK^LX47sz}|4p?}x zTZ04ZTaU@2h>2<&D`UMP4^GnJnh>Tp()VdtKBMxm0x@ZcI3(6SvY8>v>9u?(C&gOW zCWWk%so7U_MJHD+`>y|8#}4mptY$`X`Extk>(w;C*nT-lWTpD%QaRO6GHcMN_p7D| z;mRVi3Z_Boo@4TSxZJK6mS^NTViqF_=YTU4;bA`tvlS-7B6luG4l1nW#G(d|0ntw> z(H=<(%R!D_LE>O<>}g{Ab0B73uI-GPgCu4fh>Ad7t}#&C0u3%?$~*KgqZ${T;D;9Q z5V4~iW#`lGOoj{QddwvR_RpwVt5zB()=(KfHVI~%FxAQ5&Ft8qELnvOwr9Tj8vc!) zmeWmJ50U_R*>}ncBJv|Qq4$*ItYS5WytEKAIeW9G3g$!ut)w^dv3Gt50?0$DEn#mV{fM-H+=9Lw z(VYfVCN7r+{~1O(?$5dBC5#hmx9KqHO3@ibmIvUWINhy^ua0-%NWSP9U691%2ZMHI z5o-Xo%`r_2DoWzr_6~Fu>k&yw6XT9pe7^a;{vSI3t0M$ps8ZeRLE#m?Z}EYdEjEhC#^?Zpy!rAguTHYKb>)q?Iiia!oQpCkX}P2M zmmDW)#7X?iqV5x?Gc|#{NBxV}+Nw+CbtTtpvWs0)6x#JDLwpN^0*pgSStq`Rm-<#* zxxw}|^LCC^pM(W#so)y*dZp9KmoxiXoeqvvP$`Ez9Z4|T8;I@IFakO@elUH$-|E;r zHzkhYC7KVKF=ou^3XAh4pRvFcJ3r1z=+B$v83=KpH9`W-;c8x@aP6_Wb!DY38c0Go_R2r$v@GLGSwDRJ!`$ z+er23vV;T2q)eF!+472r@8W}dcXayM%)i(+MKlLt@70%f zq&*h9OgXp(9}+DcelS}aHT0*P5NfX{S1#@uSea5g{g*jJJTOe3Z+wF*4G2FNVC zU~v zKYW{UQC@K(`JN9@d#olTNpYi%2p$zjuBLOQD`=c(;huB1#d-0=CA=o8WY}&xuj*y$ z7CEfr>QyaP4o*J|)B07l74^Y1=0~#q=7PM$vG1oj8di0acnioKOwufzSjKYDBa7!CQ|9>@vE}f2NAZbQk3=25Yh)UvXuS*E z@|AXisOG<}9Fo59T9RliAFUU>6d>b@0+(%R-~mXv@=)JRN0!gkkJ)M2+}5aas{5p9 zH#SX^+1<3i7P>cW3MQvcLXf%x{HWo6T8SQA5u_aR$U0gpTeIhU*zdMqm2f{e+9f(8 zc1rZP_|%DoQ6izFWzf()8%JXVAwU)1>IBSWi}+?=_nvpl)GB?%$Iz75QgL8JH`~Y> zZu{2h#^HI7P`7L4b;WpzvUe9u;CELU-MTPu%gnyQUImo7=zuItaJS1{_d7ic-+|@E z8EE-g+3z|%HfC*K7K?6639Y36qq3aExE!bWVBzzfr^)i5GzubAvYtE*1jQNs9a^CbdwaNn2GS|%}c|xRt}O}d%Om} z(R={w94#jeoOWA7ndn$3Q7sra_HJiC7TShxMfk_F?0v47K>~sEK@Bg#cWI=&*p{r& zN23F6#P2T4A!+)BUPspwX2q6^>*?DC^h)}T+{Q0?)?1&9VN(wC9zpkRkRr+eLI zuEsx0ze|B|8LXo4MRK$HSPM03y?Q-x{ZbKoS>I_LsrKyqV z#P%@^j%JCcWpEHSQrnqRhHaF0adq~lo;1|LuO8sFN8)TRJ+VVNqlgWaHhP#W}_Aw$3x9M1Z;U~Fqw5#BT zbq|J>jefTN!)PBWr3u=vgw&`V9T^#Kjd@w#4Rni&y|zs@Q>zRdxZ6-Bm!si7YyY25x;3?@o#ohX=Qt~6?YddkM)!3i#$fDaFb2qS zF~GfhF%VDa9aWQ02$<22l@R-|JSe&ol*NVXO+;R6WK3Pr%^9M#B_9;}n)<%4Dw>Dq z;Bk+Bos+64ra%MXSntiq)cnF4Zth84Rv>U*;p1<(?apaIUTeyBTVY~~(^4XmNe^)Q zDuSPTbdq$Jx>|;4>;|`6W6)V0%-Jf4n?cNyL*ar^k9lL95~DGeeH0JZ-S0|926M>u zFqrI36hix$MaqTc)Xnf&BaIGk%DQ;8$OT3>Cn4clp`AtDO#L?U)j5WFR%5-e$@gHE zEE%($o6PJ<+FuEJLQ_uFEA^fT)>hYSl}XY2B_A_pBaA6Ed^#1~QWC%&y(oc)zKxF~ z{o&VMW-F7EoE!6Ejvrq|*%+eDES>DhHLj{!MBO6z)eMx7s#DTmIpin}gtnxj*&!FQtbe${sT zvQFGTin&DbRc6I=yLX=?-0l${aUaTkew4ym0IbfDP3kh*D=)umCjUq`5S43Pc&dFX3~MfzMh!lAQXbZFC$& zZQ}QjnB5UQIMSPgec0z(F;G5s9FHulAZ{G-UU3IAq)0i@oJgx)`eC7CKZZy-)SIDt zNIO&Y811!rPXWOu_i=WzU&0@s?%J^vH_JzLL{u*z{lut`nI|m!#~d>C5->5a%`?RM zj`xaOx_r8&&BKB}s-o{YU*O_ic=r*;p^wu1!KxZ-!6+~HQD8RCSoBHLmmr2U zgG0%yfhMhrGZhzsl*;Cf$1`22?Xe0eO03wA+PrSlM3g}pjh%I@v0V(>$&tixq<&i_ znif8;s(x2M6>?sGI=tuIdjI1P`!<0dp`9eXo@U`F5rmCgd9K#P54$meAH4RXrNiny z+R)TkNHGRIr~{q6En=-?i*{jVTM)AvFIL_iz^rC049zO!*kOtrj!OyeTCwA+xL*wQ zRl)MxKIc#@U7ast21>h%dKX%;4&E^hqCc$11j>+06{J>CJNF)I5ptBH7A)9e8Cq#< z?m6D%Fikl%e4BtC$-LQiWwy9-xtiXlRl94V_pb~P??cAumZ~w|7^Ejexo=v1?#X6~ z3{>6R;*T=6oIjjMZ)QAeN1(YH={T8<4Pr!yo4x%4h!gy5y%WK1~qlr zi(p}gM9->;wn>}L9~Dzfnav3DakkG7n5h@x<>CxE&S44kNS?aZ2m3Tz^}RE^#}1F% zf+N&3bu#fY4jGVa8*N_2nRq7Dr1YTa%O%3trlN*7_N1-3>H$9kYp!mp}zoVahPbHP&u2DBAE&$lQ)j+3sRf zY1Ewk)=#YJ#8nWj^iZ5|@ide3ROw_SV?1*B$yYCN`)j{5Q zzGv|X$rU^Ygc;>T48pDA6|p?lmJ-&RXV9}1euK?~l5+HL|0TvZ_VlaEwZ3onk$3tb zc!SyT4qB8YxFn?ZLx?aGg_iiFdv((gDR3vhDtgd8BRj3Gib5N?`1-MpqYz+tt-Iz^x0vsBz+bOF6H{V2N5*`~q#LH=|M@jv_M$HD6ZzMdQzLR-?^iuL zH=lRRIwH_&ig$T)=uMAHSBun;G;^xYcnrZo_eLIQ!tM=n_QgU6+}9@-^l25xYlpnFMMBuutw ztJhuY+#zgCZTjr+=I5VVPVF-owtld))+Q5euq%ynGE&OsSbrpPt{+q!yC4@8xxGJS zo=A&}{%#)+Hp9819)1@6bgi_D6`x8-3`e=Fn0VD)o6OYWPNU}`^t;*Y!Yg0temqI4 z=XCW<$%rUXeNJPC+j)!Qdq``sQ9=D?eT?*>@3}5yt~#K)w8JU_bTPQSQ*y=u_+*Q3 z8lK_-Cpt!C-H7#>op(hlU_}KG?Ny$`h(CMD{-pujow*~0bkfL^QX8K1-q+5QS^LxWRWt!}f!y`m`0t0#H80zv~I=JLfd~jrU zUuamH)Phplq4!{cuz0`S$+81#L+-E$%V2|hNG^HHN{H#qO2uc@)%rv=6a=XOzpGwke{0G{ocax zsVZ9LaOQ{3uC%ud$NaB2rRe4yd4V&3I2jOTk8Mn=OL?E^bm}Us*v)nhB8)Zi9NtUR zSkDYC%z+hU@II&Qb9HoA?AzU>`I#KCegw)%)vM_dk&K-e`E*5FqF7CAw=ymkCNZiM zAuV8APS{GS?~BDUs#kC%7X>tgO0g ziH;DB4sRHP%uraT_Lm!o)bu33I=obwe##NnQtTx)!SR{OA1PlO5DYJw5dUGm%HV#yggybaUY4-_CLLk!beL`A866ynw@!v@ zp^a-et|Aliq;jAC zpNMRSIURWEd}Fv%r%fLESr{?WO4&i7hPAu6I~inzuPqlKj|o3;hAvg)kkbf#Wm&1g z+>&?2_SgF{x_y)b>JN`(`b?m>`J3r&*@M?Ch8$hh7u*n}Gss~o`bzWt%8%?KmZoT# zK_*Wxy)glJ+;vDR24mvLiqV;a?o%u}sCvc@Q#x!=R9g3P=hO=?50@!os;^{`lkS}A zg{&AP-zD7~-THL~bfWy5{?1E|_$kI(GTI(Boz8Op2t6zP{nasf z<**k4QjK)i-w*Ys#QPu5Pybk7_As29Fz;gD(-UJ@zh9uwFm&bxV?BQ~YH;#Gy*uj) zrzOgRl`YoAD*e0aBjVnjQd?4{rw^h%Lb2KOqb`iKM4)dgG(q@f*Ymm@I6B*T+}u~Py9pmakk+&k_*`pg9+uPIl8L0$LdH8M zDN9tJS=tr_{or|slbGl(e)ZIjRZo5-6qogZzolYwm}nCOjltu@*We7!5Em_L*`JAU z`QWDvczQsr+^|H^kiD=;p6P<9r^zHw^i5Ykw1=R_Td=BerA~y-mfqWyMq==Oe4St) zS~s}%4RZl%bYz^iilx>-h(*AlJKWRv3$1F-7BOjKW1(kd(=m^KfV6EM7d=>ZoX8i1 zW4Ic;fC<%$8*wecQiK-7Fcuj*$qB`w8LuMZ_QD}bu1@qH@FF%&6A&K!P{+pl8tJ1K zJEL^ha}!fv=dc=k4d4@%XWxn1Y+ZJhgaK9Q8&;9-<*q3yd~5QG;zCn8DGs!fwe;^P zix6&X`N9PldACgOtO^!VQi2I)le%iTFuMN|2n^O{qV4q8bgjC|`pEArdJA6{fmBl| z)*;^-u@Zpt>HbW2-9M<&*rLEpEs=rDH}8DcX>vp`8NnS1%lUVk!SnhsmX9+hAWp!r8P{sr6IGO& zo96?yzDO^=pfr!iQmquvXf_*xjOE(ZMVD&Uzi7mk_p|IW*Qfe1lw~`gv<3mk&=fSe_qa!L!GlFYMfOSDYj%vTl0tqc`Jjk_ zY$=b@dgFoOlwp~(4he2U$B8RyAbRIC+j%=6+KxW!u{3fYdKBIb2XKi?>wvlbdpQPY z)?QoR1!VElj(k1neyJr8`X6Xb!cS_?j{icRl2y+R`=R#mDiZhUiP*R|KEjC3A2AGq z_73)*Een4Xnl_0>&yzVPm({0o={Y=x zC+m-k73ufTMW}*M1WLQx&2r~ljZT2iK?3*FkSlYaY_{YykLae0-0NRf&!XLaAth+zp257f z8{+$o-ZMQ0v982Um7wAq!}v;kuU>Xz7IlKCr3nr{{?;0gKlQ@XYwHAhInm12S#M00 zziWjpNczxAJeT)X#)`Ls%DX;4&IR_@FdWZSICcTA&(yb2!Q(G-Ws!R}2G+~c9L)wx z&V18db{GHU<1tzv(vWwcvxTqenAuLC(e@D$Lr8U9@I3Z$hyqXIelfGZP{as5AF1> z6&6Gj!tL}Hb9Jq)=~Cu0%W&FEd)aVjSJC|-LHeDXRxaW?j!j~eY5nw{fo^W)flg3w zBt{##5`&nT=se)_bxDrbtrRn;qlZ9`^(|kH&3N*?2;A=QA4r&TV6N9X1if{*CPYR$t*sVo}xf1Hod|eUExLRZdpU4R2be}r;Kp@T- z{f$;U5jrI=CcV2{lQoM+?d82#Z}P!uK++-O7=lK;cd%B-viCHPD4In|m!K^w(C?5e zM!FeOATc}}&%YK6>{WSGqX?ATRT-fo_N=k6f=X0djLTwP{T6h(Bs=tTh8C{z0Hk%5 z0n#toVjbhI7(&p+da73k>@+Vh+z!rVDKH7y5NfjnNS)imWsT@t?qygw$go8}Pa#l? zr4XZy95MK1#inivRDCvl7!tE%Xlf0lHX(A%IHy|RJ-zS<`cR?wGig$9Jndbdk*Rr{ zvze29=h7jAF;yW2NNZhioDEP4#MiKa~$^R&=xvy>~P+wwl>Xs+R@o> zemA6pxmo~bIL^mGnIrqpV(9rw;g^6#fxkETZK?n6Qd1qPSZQDS5=Aqsf7~dn+7bT_ zi$+_$q&VL}G4DcNib9Tbw@)oBV2#%Aemhhz3himdjVwL4RBwG}jn6!`?pd@bZVoDJ zXza_8p8F;>qw^khg}hl>DM2-&rsBVq?aqdp?`sJ<=KbWwg4(!m?u@-KcklL&_3-Tg z-xzA~Og9&&Xs6H4wz=n%A7_!ndQ#N+5-g|{vfM_0hwEh>I@H2Yk9XoGfQUF%&O3Ka zjzjv<+lAAh&3Uwr6TW7aal0#J&AEB}d&ARTU8&KuVC@A}=rRj`mhd*q>QZ%bZ}^bb zzkmOdBNdso;CxJJ4{3DDc`$pGsSQ1kQHi{fGQ7Qzt2^o8CDGbrA@j4u=Qjmfwm*=P z5eQMBW_{i{tiRKgKP!`b+&S_y zK<_nbvhjI*y}~Fp{twmm&4GNCeRVP1K66Dbx4(6>6J^g>Eo|rY9np*12Y>AjY61i@ zG^}j2vHjjKK)Hv^QTN}UZyo!jtu&Rn@f^5wQopH@nyqn}ZB)y&@Wwu$#7*4pTW>;< z*~m2QTE*6cGI7XhBQ)o7<3ju7#*W%5pKL}lFjSkhA`+xxarJ`hnX+>YM%W>pcYS$X zs>MEJ1?@Qv=*_g%&+)TWe+9%uKI5ZW+p4dLUn+Nh^-)dY%Cxu% zp;KJd79p`f%24vLF zn1Ahm(YiNJ4Ig;1Y$@sti=SAEm%PQHOvUQIDi6CL3&BH!Xv)Lzd;RnjTW7(_YKVT6 z$m0(n6!G%&$X#1xwCFDh!&x0E3h{Cne6i@^aGT@6i65V=ZS?5jU$XwNSHND9Hu<*AqOdA12e$7#L%LF2~0$oM-& z;{T8({`{Yx1~?!SgmG~Dvj;{xNpA+Qe&^#GZ8Mfr^Q>y`B4cbC_D`p;-3H=U!8CDR ztLZ%%*nXd_aYYWPRn1H&N|&A3n)4rhVhF&b)^E*}0cv9@W8OC(#?O*V`sji_Z{1WC zOOR3%d^}jru^dP3Axx!(Ru3*=vP9sg5N@AMvWcla>I^zh$NQjtvn`Xu2AD4{C~D+MA_Z=?W;A}B+ma0 z8Kr#(jzq{7PvufZeakgVEGCUri)HI&_zAx-^nXF_|LF|=*KauVZtpJ~#&F(>V~f4= z?`HmgogRBX<{BRW=lClrSp3a(hrc5lbvD$+&*rCAB;HOy@~3&>e{}EceeKVFo45g* zn1qPGCxKW<|3vwC&=EUJU+cdmfv8ck_my+xW|q05<%WMpjO%bR{4C5tPsdG}1Z(7#e%3C`i)GE{rXBuiq-!dmyiNFX!&V G=l>7R^;$dt diff --git a/site/notebooks/images/insert-test-driven-block-correlations.png b/site/notebooks/images/insert-test-driven-block-correlations.png deleted file mode 100644 index 7ebb804881e9ad230ce94691d0f25a219bf017a1..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 84347 zcmeFaWn7fc_dkpXN{5uvf=GjOcXvvMG)s5astAH~cXvs5sD$Jq-5>}{cjvwMX z`}^$wKwMtTUf0a2d7m@q%v`&^Rg{-NMZ!mdfq_Alk`z^jfkEnlfq}(GxCgoN$})%s za-kR?BBCfIB0{R@U~3AnGJ%1Sd>a#YUsYx8f!~d%EjJ?lojcSSZ^)YA!ygDElQsuS zOTxZ>AVQiy5z~RAgG@t~Pd6cmtrJ6LDTa+4{6b8O<8x+uX)ruNtSeB&?j+@8&OM3+ ze6u}sX13A_PIG(;qiC*E&WWXVk5nd__U-OTZdV88;GPJ~9SazACzw~gy_l6BKfb&} zqp~r-JWB)Xyuy zB^#9vpE<)ZP!Pq-JkX8x?t|>9sMvd@F&IC=6$gp}nD7uq9=@=+pw7n^=~m>6$YNAk zpf++^%f}=ptc=gZyxXb5y3C&X;!z&iGDR&j)z7GMDWsYRBk+GDw14W_eRIZjU2AM2 zo#zmrE-fDgCz?*Hg*TE^{!DTp^IP~JHSRv;B_*dEJ_cV$6QRpT?1OrKq5bBB#~)OZ z$qtu3FMV^!%+&~ABH-$X^vB%Sp?<(_8B`y+)-?3ayq-_4Eussg5FC`U$ZH_`85?XY zdQa}+{j$qjR$;wX!EZZ!DdP%0$HmDz1GL{hpRDAH!G8t~q?GED8p?I`y^R|0#5+Be zzYy)c>ZqUF*IDd)R^LiGtZxJA4vyaG7i0W(IQQ&~Ll%7RjU3bNofguES6_slyuUk; zKLxaF8^EC>Zq4#*R%{X_la=eusT^d$8ss8N9i?kX2}!vNv&6^aPBC{Ve@K=sA75jV zb@hQxD8w?)`ryGG^T>>G@K9BV5}e<@ILy?|f)s^&RL!0XtcD=0wBJ2eAj31b+_ncL zeTa2;7^@ZqF_1C^dE?RDNKuXNAZozH0!Py^uwV7R;Gw!7^MWz|jzk;NAc{QjVBtaC zy@n6gyD*>INZruH+jySc6T%?Q6jYVO8hZ8aG4f|LLt)}@Mr;%@!9B$X(MWGesjE=5 zka0vP6uDI4M}=DjDKicRaq{ms+&>ceG-$bs^{&l0{@zC0)AwwBuQly32;o7mRK8kq zk}&vjd^E9R^Ll9W67*?&VfsCa4_X)*jFR98rg!@T12U^8*iwSZ7>E_H)WXsc7TLrA zG4AlBKoobNSNp;2mc+Nx~U#5YiMqmka z?r?Z-KcTuwX7#uR+Y@FwusN_fgs8pc(^tE7&U*uo@ce?ia+YEj*%w@v@a@2Me06tb zsEEVH+qD;F_cGiV`0?}u=idsv69{evlC=AhQehk-xWEYr!wQLNh)_`wk|jT$A?2lb zhZ*y>Q&ioMDnda=ct&(a!iqwV674~kB+Zk1-6_~&(v&4nVB{tJ}qo1^csa7Z7#%9y`PUT8s3w;pKn=kk&jpK zij^=#XlZZ=&{0KR#V{3^SeW=q>)n)LHB$Ak1+%m0tY}JBYC((CRs1>8kphd8>3dTI z)8aHGfu+uOC-^LEO{{SEV)&@|wQT%=AvSL|7S>vGc1!+|kLC%M{+9GZQ(4YR8y&J! zvb70K31Tc^{Y_=4mPwWp=8U6rMY9Ebg;_;QW#nU%1&%^s|I&uiIGcuT_F0yOPadj# z=zJM-?9G)Ppj8VeT)rqhaofDGRm+Y0t1?GLhmD6_C=c&_xo3%jk3x(eiIVV!`pq=T zG>XK#!*?qW=~)U{8f;}*P4V*{?L6Cl=41Qu*;_VEvmqFa{j103* zS4|zpJ{oujh&8C0E}kZ2cV$&zDSOsxb4Qy~6M^1E`>^lhfI|XB;&Z-@Qi{n)lLRwO zvq{?=8yu_Z&cey{oazYYmq)l8Xd7khq3j9l7god*g;V#YO3SUQy{*k`x+;s_H<~Dv zpH%7t<_&LpK715fc;A&V9jjeAhAE5KEB!Z*sl@DGv3^zyCQTVxqmt-vQn_q zeN=qxwP~{2al*ZD6m8NeZAN6;d|0y=>&D^HYSvg0R|z^J<-g_^Hw!!Jp$fqP{}^S>6E$2jqFUZDuukCSfNzQLV9XL~HjG$*nxuktY6H?#|-|I@7zf zzBB+fri<>fcJSpqF`*jebGBT!OEl90Od&tk)?o08Ar99|xaKEkU3$@3+c(kohE7;b zc}yyzkV<5vMbY)^qCcM@vVDd|>{enkpQq@9CtTrom-~7F-8mVbl=4PY&tFvufXK_p zc0Ogsi*?}@(c#i>>ln5eUq$Af`J7$E8OH9%*3b{@*LY~&IE?KV;tU3T!?C7CRmUwR z)LCvVG-U%gcdpNmO%zjW9=KRGz+3xEP3<(ERjcy4qE`K@Dv<^eHC&ysieYo`N=jPqaF4CJn>y>}&|J08_{Hqy zEXT3iRzcZF0bS|kVdXA(j^?H?*CJ|tsJX7$-7{@7cyhcvt-kfxZw$D2S0611t;d4Q zV#+eb(h?m&GNpQdTx#PDIwkQo|Dt1d&Rj)Hp-^+iDHSIT-}z=-w*EFDGWVEEvU}6@ z>%F(SAtRlcPxAp%!;+3#mBkb-Ibh?3?i#h)u}=LGA3itf8`Ta#^>ETa_yBhDnG5^P z<|@}-JswyEgkD8&?rHaU`(w%R?r~|b5FVrcexp;yt5N&oii>3xb)B;0rkZ=A+VwzJL>LLm?!DBTh=VTh zhnJMmlsO7Vsbgr9s44iz1lvCBCzA_~dk5us{=S8m^&2^By;#2PSEF1HIG9hR_t&R4 zP;FV~3B5LUYphwV4!<-OyBh5T^PBtRpMAcZ*rPrk9xfM4+e_D*T$y=F!tLd#1jeb{=q7u;w#fPfC4(U?lDfo0}{Tht)P3flQ z8rdN2z7TVBfDI&ljWyAbGL@5qf#k3-2r%$>9>Cm%wC+Iu!Q8=zf&boyfswvL@ONAJ z4%MH1U}0ba05EWW_R)g;z5NM?{6a?m`S)(*8<_i$cMl-HuhU`w>W$Qse)q37EI#BO zjF5_mloaHzijjkfiH)PVtyA8d<0HrgL_0}MM;I7fire2iQp!(uA-VSqKvly@Lr#|4 z$kv*{z}VK%gu%_)?v@S=uNya{X>H0#z*$O2REdBdzq1p^m`X4OFl9U zIYm+tTL%+Tc7|sR&&c?ZNJ&X~9gI!6l|{w>91eNMM`rHiWXH|O=<4dq;L5^a>tM#n z#KpzM_>7s6nVBBagWl2I#>v2q-o}yq4qZZ|NEs+B!S&k&)dp z`upb(IZfOE|8TN#{8KH6f{eF!7?~KJF+v;W1Tg)@u-iL-4EwIvAMAKE1?~M??;a$}Lne;b)5_VN<=Uc}t`n zYSm?p1t)g5iaH#I?WLX~-t#BV%B{LWa7f=zXW9VC3Kl=^>W5m%?1@#-MP=qAQ%f(( zHEoxG!;>esB0qg~FmnN&&4C{Vr}QQ`ww$<|nenF{94HwzXhizM+<}Edz#tX$gVF#} zU``xy2t|HLrJoaDH_?bMg4i;5CI%L`_1`9gdpNBxM zIN;ieO(r)M{V%KpU)&LzZ0lF&hDs5-)}&^TF~9lzw%czU`2F@DQo-NxK=3yXAZy-l z9Q?+?ZN~5$2fuOf8wY=s48QY(-|^tK9Q=)g-#GYXjkmecZyfw>g5T$j5J3FS4}Rwd|L^AqK+}b(82NuZ_pd1C*T#k< zD=(M#_HBXF^~uNxn@x@uqa_+j_6MN21v}$q>DcPKH-C5jPoq>Uok^E&d|$Zi@jp@V zZw{oXZKtQk+`5;d8sY)>h86dWp!M>D0iGiv^o-TIibE0Z#TU5TAllC@o~B-3j@5IB zN*+FvzS|LFO8-YT@*im+Q=ljS1fHW#2LAWS_&<7S$O60D9#0O{qaeiH!-f8oe14i= zy4973X*(41xN_-Dk(=G=>X9iSI{4AJzakL_A<=wpQU#5~^Yo^~!sVAR$UAUQ1sG8f zRIKsgV=;tsH<;`hKYu;dMZvIq{78Xzr}W>3{#OGS4QJAfd<({ZUHo3l)fKk^geGNl z;&p@<|0?eOYd%#i2=k#WCQc~KWyqg#u{E>ulzz~g97&gYBl#owTZX`3*q0S^NKh0< z&>?%q#N^yFXKO@J;KC@E>S3Z7ya|ywoJr23kx#~_7rPg5vZ&S5fOUS zO2|8*+)amt#ZoDynl3=K;_h}!{+T-D$|x3ux1PegK&W@2X#^2E{XRrrKEWzPCf22lSc2ARW95`6vW2=@i_rI{Y->x6c9{!NHjnjbo9MLA(1aO;>>s z0Hb4Azx-8)pRolog$KkGqZ@BAF`?c4m8OS+5X9XNWEc8X2PiluRpbMj9=$OZg=*n= z%+uAZR(r^241JG8)_}S^W~p|;zfi;p6Y?<@2Eo*)P^|M}4K8)!!2ht{me2FrY)#w% z1eG1BqB>+=8e9hIphO%vEVq>evscfHb)e%zBi}E5Qy9KE^%wF%Q(iDV5@P5Wt6-u3 z$>NQV|qemTv!G^A9Gib7?SP^|h(?)lje%u!4mDnhsN zFGwNdU_|TYr;q9lha5v3jBiZv zYI0=suNF#E*1Gv__J$}uPCplxe?Q9*=$`TcA2lPzw%1( zzaBd%tJ(oIGb!@g-4q|?<>n68>aR9kzsQQ1a|M@d^+VS8oSi1SbZ_77kd-#Usk1rl zDRc!{`5ZIf3HPgv4?w4RB9@IX1=(D%o>%qNc|FIQq*rTm)3!DOUN&B2SEFHimh)+z zMw5dt8o)IPN1)Us$$9UBgL86`{;H*#q^Tq>M;l)6ot*LM6Y$(aBAeXCT2Qg?`KH_X zbr?^XK#djEY_+wDCrjQyvW&pQh4{tw41p2(r?t`&jv5kv_j4T&AARsqTo_s&&pb=z zbz|N@E}kC$gwNFs0j+pk&PfaS&G9En`BW)^8Q=U{EnYU^7`LXF7_qmP7*-`DS zk{Ye1VpA0W-HWdr^IQ>nT#zp|f#^gqRw`RN_rB(8H={Nz?`}Zo?X;BJZxuTAqK;RS z@Sl#AAFd}!+!ClWm@RZV=!CGL8uQ&k+L*8Nac%&7N<22z6Yyef_BqwKC74vIAH}ZQ z+Hhr8zvV!Z66-L%%&*m3Bvx`iIFW~bd7>9zs~7ljU(sL+P@kjWyHtE6b5U_WC3dJ~ zQ^<1irgM62H^NohmGP|kth0LUe8Y5%aUOZuGjmwE1+V8s)RKDMa@n&K@@~B6M8|US zVqJSJWhrHRUai=7w)n``Yr&4&VNrEsIqq^!ujynQ+xKqCa?Yadp7|cfB%r?JbXrUI zGNd||1jl!yE2#BMzIUr*{=9I}wxzmB!|Rxkk^dDpeJxHJMNu#)dTjqw9C{8`9(S3q zMEP}(=0@xLofjS_hm&KyjAvJKiy*N^-TgZ9n zPq628dxLh~kL}NrG$#&(D%5Q2?O)6-Hrm&nhFa14KKD{jqiv~#bi|n-73pwwv2-tu zM}CLr_)@-iz>r~(ZUwGndD5_yNvV&En|_CO1^-n?a#pohUi^iu-q#y+7swY7$=$sh zGPd5K1MqG5xDHTfIssrirjIeV)@yLLUX59TueWb>ik130Ojm5K(t_o19m@wCYfU;X z7E(1GYjaYAV|>+@=hZE_G`$#j^}YC=w0c?(_IH}Aol>-M(#V&=QkHA{wfNL4YAlX9 zMwG$zpz8t0G=ETuJdR`0fMX6@2l(_@(=n&iv1rlLaJsl&`J{ey-_VAQMcpIsRBy)9 zzkWyh?41 zn-FMSC>QM^09WXV<`Pw*}BO)uvhMyVo=+Z*E`g zG%i}Y?991$fcc~#x#|IIrLduUkQ3V1qrACIp^Hy0_C@MQYKojR0p5f8 z;(RHJ;47DW#9#v`i*EjGIX^x5pUg*u1Ua+NbeTI5U}bz_lHuRrT)v|J*r*!vf2JIV zD(OneoO%ld(>`dR8YoE~h6WOFDQ|$j!-jFv%w24<>t;7>K;XJN%sZc1JzAHI-K$%J z?=Ss-$O}fX?0}i;N9}>om-}C2OCF@8RMU4wlYSuff2x6zp9)E^jyw#ZXU)GN#fO0r zy>%y34r(U)3-J(CZ);c&J|;5gSub>6Kw<=ho@1CVPobG&HLPuWmgl0T0pBId&x6bp--YSO|2>^QP|$_c zd5tjciBJsiqrQL7T6!v^kT}{*Qu?2=XD3LFz6JXY@ITG1$L;=3rLDN!-^~AiGzP^W zFhUN`gXVKTsQ-_x)~QLUv_UP8*nX)t6wZQpkUHlVbs&`Pe;E1y{fRPnai1Ih$25OC zRw4ricOTe7uDk@SUfGwrylQZ^jHoku zIwzM0Sse!*E)1mt&zd$1I}Ip)g(}iFw=QWIT`2ui51*-p`+95&Nkk>}`o_@nCU_^xKHu*<~^*+(W z_{1Q^^(IRz+#O&!<~V6QM=~DNsP_U9;%jj*vkX7svL12sYG!He{>_U1irSdT#v{6q z70WR;ApN}>ltn#uJI}qdUWhz8LatBScusoiLskY&P8uiBQ~5X&2IOTKj%_T%-Whjp2d7xnu*_ z^+tpO=4oE;a5u|R%tM*cd*xonRH$gIh3s@ozu1xOMQA_WY1hrQL;q=#5f0_k>kUn& zF428+TnSL>a-7kxjHmX>e9M?)kYBpkz$Xo`r$QwwIC{GsHHP^!(-GAdy^J>&7AWkN z+~+w&$E$L+*}7rgkJ%DMWO%~4JV}V~lMezLWt;B*jpBbd6d557>~Yj<{?nYq;5?jl za-2Fm@GYxYDq3&5*VQ?6*E3c6ljXR&O`yojinznq-4suVp#~Je9C$br+F9aPf29zA z17bwx#fAjGX4xIaBXfd~s)Kp{RJGZ9qXLN``x{#ELg$$2(7U>?DK+R|D|G)J^zTM` z*X_>Xy=tx>Ct1)KIC|2=SmL?*^)uW-;nlIi)q&Fz=xWxz*|c1p_@R0B^2Z7=sx&h4 zj{x{1QjN$!qNd&zv(`_9m*H#p_X~Q$%8~9ieKqstpS=e86d{SMaliRMMv8xCQE0N2 z1N_BT(5zDq#xum8qfSHFSf`;?B(yo70`bSoZ+jM0i0G^XOPa0*4x{wX4n_#ITSi0@ zPOiJUD2LY;DL*oEv&O#(BKei@>TNhC8`C`GKjw!Bsw(MUB&Uhbsx3lGV6*VG9CbYw z&{GRdBx}CWzRH5E!^8WFG)29MuL`)rL618tVk>ehYAVDZPa<7@cnAyCRJc?NK^4ew zhYmkl3I-?x{}hxgr6ABmp2D+#oTG5hC!1+P7X9jr`p|jnu?i3;`WHJuaJUMjX}x@O zsR?{P`;8OkL&8T=+CFO$43-_9=i9zVfX9<=*6K`9#Y3RWfb0%FSe*W8DI+xl_TQXl z0baOi4p0mf)o9;*V9cR!y)1Wc?(F$h&~Ez(@jbj1@aWGSh~Q#N8u@20gGn}bt%?K6 z8gD*SExF!!7iT_~`Ni@_14%)(flq(7EU7xMq;k2GHuFJF<!SAEl6wHR?vN@*t%lPf(IWG%AVZhQ4a6ZPgj&E8T`ok&#~z?5O_Fm z#P9tix&_UkPCd=6)@u_--R(=f)~M2(g4PI2_jRaMFGSR8Y{)r}$u!DOsj{L1kep-@5(`k3YV~=WyJCR)+6aG95!?sC$S0m_H?rQG)GdCu9`KD8^sf)?3p`@D=P{kdM8Uv%Lt9Ft+ z;8&B$k!*~g;@2Nm3y*?;O@3DOr&yc*5;mZ9ue$1_dtM=+_1Xzu!gWi8=2JO3_T$Nj z5DuLS@g``M)OV_@MG&I;?$$pYqG71Ku(;ImxGK`;A_?ca{&-R;S_jct%F<{P*cN!s^139BYiu*Fa50$oH~#LlJ3_gVTNaFGV^u&{V5UhdYY+hU^aaoWn4@qmF^i<4bHY`>*X4&F}?M|h1*b+hFvTaZ1ONcPT$k=%55}iP1}7%X|N1ctih6IuT!Byo>L-=Wx>XF=Os+$hANpTKxe$ zz*OYq1pc_W1kQ+1xfI#ut-ZBwR^3aGu%93u9xD?JCbM%4MS zg3a)dvm51|2k(A9j7yJ$_{;U-ncwDDTd$J1q>|+twRvy-^S1(5^8h>J?ARp^^B;Wh z#Wk!6)p7M+?T~g<7x>1o4DyW7>$dX`&)0G@j$kVvHKfBqJ;fuV!`w<*d>Xu>?;Lt{;i9}RHx(JB<~rOW_G@3 ztjR8X*J6(+J*xYE$gJC5!s(It@({&yHlo4oe}-W2Zw_oXX@E7;7VA3w-X7d)H%VujLBVFmALBnhVliO%$h zt+xfi!cDm=uALfzE_Q3LVPoBP#VW)0wqAObSWu(P9*-F?O@Y2U1gxg77o3Bvs6m3ar)c;TY7(5~cUG)v=g zu_9M@GBG@zG&E)-dZ&atGEN^~OaCc4U#mTz&U8%1Bwlo>Po{#{aJ*@yNi;L9!jBpUI9-I4|D5?%d>Rm2OnbuBEgYHhaj0@dc)kofcD~no4W`9 z<%q3!Z8Q6e)6Ie&57#Ff_2g*&Z}W%9()0Ng9n#6|4liz2wpLFW1vG0uc~H3TF-IQ+ zqIP!4J}~YVx>LfVw7%|4B~EAQ>aUKz zJ=~zoMZLN<{LHq@bqYn*V$CBcg%c1@LU11rn!*Squ!~PUSKVKn5)w+l-@yoUfZOjx z^((T9<|$fLR3(Lxn9R4s{c<0wtuO{L+F zF%t#|w$&F0>ZKz?0FLPo)1pGwlOV_=D5wAlv??(K9g z`r5qGmD?A|d=fY`29QmioBLp)BDgVNKo%``B<0*QQ{ef2D(E8=FN1-Nh-)jVHDJw> z`zH1}7J}wFL6~`JhBjd}3czVj^3un5BfPhJxY^Vx~ z-i5VYvu0hTXh47^#ZZIWO?jY}ZBK|{7XtITqI6H#1`6CD?vH>6t1yt3j4{7k^KY%vFaLKDdFV_#W==pcbA^qD6`fUntYAjct;*|Ytf7`%i6`3~zEp-G1iblv`mZa_(S7~x4tA6RnF1N{~+ zsSOURj9&Fp?^ zB4(6dmv(62Bx@B79pX%*Yuc34Y9WvD+eNX}q%!XHq2x)o%WsHmwzE=k)h~cvBL_U* zSG6|t845+CS(f9r8v8CAL#YkEvy&A>xU`^DzPzXO(;H~dE6>eFvX%IBwKS}q){gmf zn_XtVe&yxmrBeF%I{Jx}p>!%QbqtMOv~(<$bYv(D@ck3J-9b#PE6_b8lqSJA2K`(| zFhk@eVDI2yS%F5nKTU_LQYFM9P5I+%#E%t#FDn|AV214HmSY>|Jb)PzY)W}TT`ke+ zKq}w+WJqzjbTIDchk$%mvzgEHRANEjGL$)JAkp=YDM8SLElcCgl{qjQtS?wvU0V)* ztX-c}P0<${1$*k+qF+6QO|E9eVQJs0fgwO%LK_pM$5*%GLy;J{Mo^PIbN3BE`TgG$ z^zUU5BL#8O!-b67mBjl#1l#biV$Bu)!X^0+-~>rI;TEU}2tAbYi#gA?W~$Z4v%Yml zuGOT>x?QbTKiAz`bl1iqVe)8_e?t4nhlDFMk-b!cpP#y9KS1r{2I zIgx;ytV$r5Xobb}A(zKXGD)ks;>Xpt!!LRHj2>3LAb=Bo*H)xnX1=m{#_BjA%x%A4 z1!&HbPop@?-!qyAj#Rjli?NdMxJTnTA@5BxAwG!5TzQc@Eb|0xc4%TxLzwHn8 zGNR((v!w7Eb~@D4AwgL!{#Zi%8sUg}-6X*vUB#fhLn)yG<+BEDoSr{|S)CMQx#9Sb zp|OjoTDFf^a(+RgD|YAl5tT$}aRaMWo&1aORBzQ*DS~seQL-Q$WJVZW%D@LT2+7>e zZ}4l)71YagYhyYL8aM~X3dfUi`;)}ihJ@)hFf^WY?*@qxiJR?{`6EwQ^IfA&bi7KD zT_9~N?mY>`7ZMU$r-}VDfMC=k?jm$2C5OdvaM0yuhdTX&q-2g7PchZO@bk z;4+*CE9c6}(`&QmVp9w6%mR!mROsmVlOxio`)Tr%r6deGV$AW!vh8i&PHmPYki+Vd_kI;W#`=LVOnd(G}f zd>!X&EI?MnKDhIhqEI!H2pHIH47bx=%u(B^Sev;f6D?kum3Pr>GEb$K;(}g&>kt5A z#Vzr9qH8SUWms3L5t;V~!QaNPqASTjFSx#$_eAxMOFMqoym-{g;i>%Ig$k@zL|hK< zQqA|>+}sqE7{FAE;$s(mb_QGxxrFv@e=R`>wy?H(E$O{7D@qIBz8s$RHO3Ktoy~Py zyZgEIxrDTKBj($9Rx$~}4f$`SitO`@MpOo&GnnyI*Je+jYz2c|q^OsZ$fFrmC6UI|WS%8Yw#~DZuJMXk1 zN}tG)X-SvZn@UgX?$P-^hHWaFAPqC>iM(j=ihT8?Z^D3{Xn9m#f9|9s-uiyXw+Jcv zMD}cBsi@VV@;eaQuX60Ft3+6(&Ty@1x5Nq1=yz64wK;;^HQefcrq$ipeVqr$>%f8l|rFZqa<#>>Bwy4Q@rxRdR{ zvxKPCkMG_dE{2}9Rlc_|7@`Q?yi?-1c-me{E?K#x>U|v-@?N75uwu};j57h>E-;h&MfuNzJ7Zak{rq^~YHa3( z99QUTXr}Uoi11k#BpT%hA>qjUOH;v!?Sd_2BQ`SxqY;_s6Ol|4~930 z&*QIJQYob^%F%3s&jm$Sk2m0D>P8Cr$E+P<-p`%24s1Q22a4Yw;2?!Pg}aDfq}d2k z9LrnGi>}rNy&%EEfgVBy%`qObn#arGX_|%RA(?)XrsLTLQCD5f&q@zBH844wtYt?l}D<-6_(jkNkI>>nSySzmVcJGeHZJGD3Y(>eavSVOn-B!42=u<$>xhkTS@ek#y%qM~S4mLra0ZwXfYMsiClBzxJaWlya=t+0 zbKj4?R6bMee0N7R!-~4+<8Z5t)!aGb)~xTKve-XL7BH8m;F8~ z{USA5scx%~>#2rR}?Y3^~ufMJsYVGJ%9kLuH^Y4>~{!Cj)1 z&AJxgI$1ooT5zALJb^j6LJ8yNm7tEK%(C*Db6EM@ISj&cn2PZjONO6`MVHr0U@^-T;>yuL|Hfpcnc5^IVN>{Y{V+?$ zb}GBYVeX02HlFJZn&jEE>-JD9kIS&sbt?adaZ#)5y%aV})1D8n?tk;@aNO^55pA5$ zwl%;3xfRB3@wj}kPt65C(SH?odckER?kgR~j%tGRxo!SREV-qu$~x9_nS_|4(R2Qe zOGxX7ER!@P(`%ppK4-koX#o}c-j=xNXQx<{p!b^%pTHUf>ScPh4Y@sWv|?um%U?`C zasRp0yygKmHs0h^Mu~q3IlA2J1?hYxVrpJen(~nAZZ)~TFCWFwl!R5h6r5Ge*~`!E z-eWSus~gxt&W>7z-7V8ybvvL%J^`4<(q?RF7n>ztFWO}|=gai>{UzM1w= zW_K)7sww}ORyLacdL5|5^+(kv2dTQJL|+kQ$SqI=o4-LDN6r^* zUzeof@gDQSKfj<7BQ+W8ytApLS>uF=e=g%YRp*{G@IkrVLQPmR3eVJ*YwP;z{^1q+ zmHAl@n`uA>)6!F=1}A{|=^#)wTk*~?uR+Rc*GCmhWFk(t2?uk`{2+v}u~%`2jzd+N ztZ}o#WqOS^)Mh%EA2fqkLRP13QlcYXbY7WGHy&?h>AwlhPH8+&+)^8Q6Y{VqknrO0 zTn)2kvge`O_RG;D<#J)obr4v4BeN(L^cj}1GMpdnav5D~z))@}`d-m^v3gLuaa>Li zPVbjEqG{zh-uk}6a%rrHuXS%#08BRsN*ys4xxQnz%4%YM8lnC(^coA=ds!9hHp9y+ z);y~qL>2lA(R6xY&H8KKN8B&JeVC&2S>*|D@E(oe_W3eoRjUmkdxKqG$Byi__0ANeogHXwO(fagf3z`{K1QEN1#6*w_Qd-s zvPt#uxbHE5Lb~T2kLyLM&D(t}-~Ay3+h;%K3?*rEU|RVQ^iO_z?)Z z%TPO0wpvHDVSa?&|9HP|(vj7^2gM}(gHdXbIKc+HRgG0pU8`s9z(~5DoJ?Xf4(43X zd>5ilEeiHv=bI31x+M60a86B0Vny8zmP}$XuiNS1HXFtK*G@|YsYp?bo^>k2yWjVE zAnQsKs)yXT+KreQ<-5%L5@@g0S)95;Hll(qwX3q6P_-0`U)S8XT|*LkiE1Cicc@F* zNFNfziHJiikYw2DXxo-I+f$D9i+v+0LD7mr0;^|5%0a;kYU?O+W|JvaFS5i#@Y5!} zeW_WNYA!P2owxVtlW>|dnlI4AA7&I!1-#dKDjEH(7|nm_j|F<-F-%4=5$6O>1KeCo zxyvZ#UVI5gxh5WN`Q(FAi{7_URwd{C1$G}Ur;PK)JI)R(@lmnwt9ts25^6U3AfKDO zAhFPBt!XZn-(g{PGy2%AX994OzFVTGwvJV3+HKtG=6eS4&rYS?!KYK>w^XA>Jy!vL(9I}TbQmBoc4uR58@vTdeY2&`vYL|mp(0byR?x0Z!E z!HYTsGj)z9#ML$@Il5oUj~^~JH5m-!;vaQk6zZ&zZ!y%4RrBgx@sQrX^!~<9?L09+DlmZw6u;cb@bfhip0idmHAjwt*yRwpfM5u$Ws?ikV(tEMIUUc4n{TCBK?*pU?h^ zvlS(Y-N1(Pjhm0p$h?wvW8-=H@UB(=Ow40ePwT_icf%R{Lz;_x#Qi4EY&7)*SmGbU-8&4`8&`()r2UcoVt;y2H z_T>^zRwkO_@jDbAo@ftCowUND%z}Zs0)?v2U<*Fe@QBAZWZQMq<%RLc8UD$`yTQ3h$E{(M_ zSEd}|P!T1~ie&=M1^t#g+|gcB%w$@SMU^Rq+gnZB|C2LjExY+ptQuX4QAf9df!w`> zu@<2ycBjo#DY?R>=a?^QHx5xt^qXH>)wA)GQk-(NTsGVfE7sjs7TYIR+`iK0%?Yk! z=E3^v-pV!HvhXdT!={J%McXLGePcvSq9p&o*<~MR))#FWYNDVPYL$4O9fAWxf4}cSdo%e2KYcGR(<-e4U^U^%g=g;xzg_jL z$W3FjWOx=cJqvyyhOD%~E?4rde8!+)jFDwge&vx6I`KJ3m51pS$e_9~ra(F-Y?Mr` zm(7y>efI$TZJC1E6KeLUkbc@#bH5`DY^~}0AyqKiLfZnbPEx=Nz^}=5>UK5o4y34&w0Qqmux5LccNfD0p+xrcLjvK^B5!F28_55i>Qg{vJvDN;@Ik~S z#2Y|;z1V@g&a;45bf5Xr*XLD|k9#7y$iw)9WWbENxwDNYU#y8kvfo5sxqsdXN`Ee= z5=R!kxo6~CZIz+`HmVULP+3tUuix5#o&E6v6jTPxI0N8diuzH7YF+he`2=8V)jk zk`t?d>|GZi+`N#AW1tTsY|Ogk#Qa*z-Y%!w9-%!}M#@Nxxnl9zye65eVlV8~gj0Jb zIP$ zdz}`wXYUYNeLjg!kYr~FYoXPxsB^ga>>3_O+`*?clQ9~;6LNEO{uXy2A@iw{^mQs# zr$F`-(Qq1-Tyv3T%4)#&%(U|L&gh1$SjD21PGW18$>CCO9Ak`9taj~AmBanmiwvuT z$R796ny!SS_YL#O02#ae&kGTASND?C9?xGelM1>8ae9MG`5F!w{YB|$iv_?t=0+&z zN9aYZIFSe3wl_-q?m|*^tsiUQaZfb z%TIqd6oW}7&vqnIK8#3O=*@gJV1xB3KvvY-<)3^=KFQae#LQ+oV|*%W ze~b0y3>j~Ye*@E|f>UMR>tstkI)=z_R5EHRY~Gr$ays?(n@PV{y@Gr-bl?F^QQB)C zO4!(~*PA^X8&SgU^DZgo#=bo#3-%<++hkI{mJ8C++3R`3m)Tohh?o`Nwrf-gJat>0k-{b{oUTd z3h0yh7L={pwteH7N|sqGG0Oufo^(k$(m1unM(~iG8xKrBMRt#r9m&b*s66@MKqjo> zDZNVZ=lFwp0sGG~XQ_vyea9=yVyKk%ANs$p85%iXA1Rb)?(m9y@^&zHvSbB=0-Q)G|3;UcoUyIO$Y;+*L|+w+m#Vi{SR@wpCe_+#;&dl|EniTTahdXuTF$!2E@>7>cXz`grMtVkJEi}Jz4tliJsZ#W?=jYJES|OCt~uvDa%ELd5(k;cs%;028R$xEYdpuyTxzt1ShcR}bG1-WNMHL{q8;t`WZH;XXwfla`Um*H?(JL7NLZtio-0^;tnDfB)q9 z`L2aK(RMlJ$vnlup!);U;*MWT;VI_CW&F&Eq293?aGnK?2y;%Kz|~-#wAP%H>chkvX2-&M+O? zxYA3i$NjceN{!}WC}qAlAN!AN`D4O74(iv{-bR-D7S)4mUAc*>A#V3JScg#gMMr-U`~L>hDj`VO%KZvhl%w`)IlT9JsR59_!esPb$P@BUMs{#RY}(fjj+iiVOS zy5SFXFfV=n(&w(pBTAceK*7}f6iC{yC*D&sC+77ZYuX8MTON3(jFt(_CWuah?dmSiA4RCiOl(woHXaas@Mu~u!P zJcK2cN+|N@7EGD0(K4eL@D-lnHaCbF6^%3)0je~HJ7da>mR~G_hKJ#KY#nmB-v8wF zskGaL19W$RtJ-Wn>NQ%Afl++DO(K^GgqdD>MUZJTv)zvA^+6r{q4m7hRRqTOcXR^; z5!9+#P~-Wk{GGc~rICz!JrZF1RW!5N()e51?6jTnLLk&~^Y>@5yca|x&CYsxZX%N< z0TH-ce34j08e}D2;yAtI&OGvbLK|DByiTpwI!9qE`lQj7r71r~B$kt0uRmJ38t?IJ zhW&WIZ;dLPjp0HtOv+kC=Fg}2^d_U3am;c=_18yi$b5racL(4h^ZOz2u zDC2rDe&0%wxmo#oq^uHA^@B0Th68lNf_~H zoxC+;VxDXlmHCV*SSHJxVAT(mT6cL+t(?rjVQwf8VSJBrd-QIG&NI98gV5k%Qx;9) z46qr*$I;Y*q46menB(7-E*W4D?K2SKxi4&D*Jw8M*B4jMFQn+LaE+| zn+ z#ZP%9cVv3|6AUVWy5lCqDL@`{2?LR6<@#DCwUmKes9BlCygB%RCIKMr7?@suHDlD`F8mvXp%pD`)6}!CmRg=wgM__I&q!OM0i^B(w%_R zm;lU_oqQTqoeSVN^O8n@&KPft8YSbi%JSOam_NMfuW>xBqUR{WPUj5b=DEs9lR6{c zCKgA_1l#3Bq7SO<_A=mM8VrLt2ZTiEzV;P?%k^nL!Q~sOuvet<`JC=9F!gfOGMmMd zL&|r5UE6{I|6CbgYAe&DM1?3A3y*{yn<-&cqNIpE!)7a!jiK2mLmS+nD%IT>LFSN@5GbP9Bid9UQzlQ>yAhIDRs1IQqoN%aO*(_D;yoRGMBbg69c}P$2TT-pBhhs8$+ExMQk)>c7s- zzrFMI=XWk88`Q8ho(z2WwmXkOds;EFjH=TioeC3~rHa#NM0jWWO0{~w#CmrBoQvnn z)I0TbMKXeBradXAaADKj_fq-l9PU3xWV0qUE-$r_CVTbV3)3=P4usxfwZT^(1JR2( zVnslWqH6Kydk*Ae`6Ax+-fbMJ{vIv^x;#4Vj>cw zz^qq}wb#1qpavVSB}F?AXmmO5^lh4T4IxIGu4s^_V`80%FbLC?EW!u+56W<_-UAiX zq3S(o6Ixx{q&eCd1R3{LU%N_C#a$3Z6d_yj8}tXgY0`u7mHGm^l*+w|asrxurabqK zgVvj^Vo*MaoJ^qJ?9h+bs6X^g<&KHeLbCCF*ykRH0Opmp0RE;}|Gi2_??DSM_cRJ7 zOyfvU6FEVd9m=v*6PvM|z`=qYN4<1h?ttMLzuwgY>Jr6Mb7ok(ipG;g=BEfJ=Ckp;73lZ(+vL=BqZ-<+WOS7# z(`ks^<Qp1-ZtC?UlxukHC)qbAwH!3w4F@d=_Hq^0hOG;JW-&N!X9M`y$O4%pi`a z*)chF6%Xo;8eXKhdac^*XWQkKY5Bpe%4|{e6rCp1qv`ge9))~nvNFEWTdRz|B

)dq_dHb#U0M1<)6^0rlarg@<+S^RZQCYPmHwn3q`9$i!k9 z;^5ZNuUP!JW#45=>vc-wqYx?GbZj3qEx_8sQ*)s33)?UPqp<>_VmH~m6Q#3EtMR3`qJ z-6#huE(O6-ds90S6-MJ@J(5mWc}$9e!IN?nxZ5#a?*43n)fNR!RK3yUMmyH|g(l-W z?qp;l?y%JsvwZRG8(_d)dv9ug43Fc$cd)v``Jz4Lo@bgTYuwTuvXk}xLS$R9d-n;mG^N{y9kP~HH=D7nr9I6S zFZ2qTxEFaHP4F{!J78Ah&F)0GdHLwsuk09RT|83nR;$%x{rdma3t&-r_D&;z(7m@w ztF(ZXZ5Hd7b^-eAUowp_1BIK+CF$pEmu{D7J)FDj)v8Xk`}4taUnvs}q*9h7)~7Dm zh|1$7GGU3Rt9xv@@H3rE=HS+sYwF+Vum5r#J^Y22wybWSs^STu8sUoE??MVN-}(`77uU#&`O^k5K;V)B#;3qNr+NiLp1+uqdm}_DWyp-sRry z_E>dYuF6zYoJ~ndDv47P9=Fj3kwz8FO04Ha?tDzXOAf$O3U#OZH>9fg5)8`qdreKM zX&O+ll0~~D8vWqzvS&%L=;coC?+zE34Avdu4gx(3DTE-Qzwu@eeWA__jhg38_Y%h! zdfOoAXu10)_v8VG%UnUBwk)$?qz_3xTek4sh%#M>v{)yhKH$h@vYISR__OzC<+7N> zQyRfv{Y;Z;jV?lwISD+nfR^4he;ezjidAc~@kwaXaMZLgAHHVC%?jv6a{w6-S&)vP zwbCvgL;m52^M!S6`Q`$(#XOjwR6212)%{|x)?AgjZy+%=QATg(by*y1i}^C5*h<9b znokhdQNqp&cLns)K}!iYLqIguMaEmWlwqheL(I3gC>%1%9R5PjFPHbe9?ZBZ)PoXI zEDl7UVD0MK!k;jRVwkNGoF;w93HuE<@H>IMcBBJ&1K!Z#;w1>SrxWhf)BOTn9lPY2 z$d~oPvXVq*95gKr}ddb&B4-)pmDVdw#clAC40f<=pxXwTEr-GPDPjrtZvrJLGis>oMy( zk?POGdBCE>!*8s5FV%fD=IaiwkWzP-C9z8Lo=LNfyE5WWC|X~L#7i;ma?q_*2^Pf$ z=eu3gDdc@ldQr_bHvcck4zD+%rNJcg>jc(7b5mXP3+N9KjQaJzqN+6H9>Lf3vf06w z^#rnB5h!GQCjjB-cNJw`d3^kVQly~Irg^?`Bm~N4U@#A7^ooQ^j-}iB>91qt>(K0K zp1wnYOFT|(D`u|dmUb_{P}!TR{%%y#L4(_YZIXD&?z#h$+mPL6h}_eF5y|xTvLcu2 z60@(-y7+HCtC1fz-e07WRFF zZZZaXQa9g#IiM$Nc?C{l>s>wtPlqd&Aku@In31A zA7pyIIL)n#%KXv9<$esn2ID0aVlW=V6`3R;5h`Z%2CvV((@t(gTB9XEBp-vnzE8yMW{0AxxKbawI!H9P>ww}S{N)zZ*c(S@cHak!d+X2dw&x{RSfk&H6^c5No; z&doM%jp2mJT5gw~roWLkf!w72Fp1L@v*3oNMqAG+6TpvP-L(&N5ePKcUX_7X3ri1| zD_u83&d;M6eq6>MV=`#Kk4zrbEb;}F)+l1>oSaP);W_ZNb+kxOvRW)=y@17{o5*8F zB!=hUQk*9BJj+XVl^D-=Nn`C+(R;Kge+ZX1ok#V@;_aP0Jrq_gA}-dQufBdIh8k*` zJSP^zAtNk?AL)>{UgdkYA6vcseeUj=#}IKCgELxPNC`dR7v=%NUI;JWaDQC|7G(p$ zGVAXp-+vvi%9O77fI{p2I9t-DnK*;C+&iTxAUgmTR*rKZl2+4vR6l@B9$VGNtMw;z~O6j|LI(`-))iOwx&BJUU7|;s08?Ab5K$+5M{|>3Wl#h^rsen zA0CfmUWMS>^%jbQMR!9sd1JI@c4Q;j8*%^_MOiLRDjq!+tId4sA5>__6jB1S$nAD` z=P6h^b6L-!%t?tDAAA&uLTy-%yZ4xOz&R9M8Ug6+^5@(ZLJ6%2%8inz>t^au0KUliuG z+>Z9o#vS&sofnKXzY_WXFzGwe%pu{RS0DRsOkcG=bsR~QtmXb)MnKI>l_`^A&)e`V z!{$<3K%#rV{yM6XBWt7|BU7KZ=}zO3T3jvVtI5m>8Fv85nc1toVVvLlW$HsO?aIQK z+W=PrR^FR%Nd_xBOA5E6=IchAs>Z7~F|vygO;D?K&7wZwa#-uI!)r;C@a$=}DurO> z1%Ice=^cK{QHyy0U@z6|+vkAnc2kYp-sC;Uxj%eYI9oCup|2M1?0hn&+#~#wkkMo& zY^KUo+kpZ{2WRT#pXnNpF#MzIV?r>i1Hk`b4*|+&NJt7V>p}$BPCoI*53DIwk~r2U z3(-u0(jfnr7>Nk;ww3an{Zc{f%@g1DuIU!?MerNCy*edJo3h(w9o-Ny@Gqg*dMo}n8AoetVIW2g;2St@;29x`MUxz7( z!=StOVS{~!-^>(?#bQoAT+C0@M0QSIGTkrLj@kTb%zEAu*0_3dWdbtBpR!4&ZRTKu z$ww=-rHyK8swON(qBUYUyV&PTKXq;G#z)*=In&u~Uvm|zR}U{EUhHx4p5z|MU|_PD zmY6$1G8m{iiZ~Dh;Y{Ioec4{lQBHw8BSbBu4YIn!=J&?^D^){*F|_AAo)uCh1K12S zp=%5sYR&zKg7zodTq?iW^neAADPU`(;;GJR(KpUnTV zEUX?t4V`jjD(EyT(qFHY>sO~{k2Q#l@QrRHDkaWA|HtRQhg$=chGgD;OoeF~`ycbq zjoQm1WCY3vz};f^rh7tI3-48e#XvoyAvn%kW3jwo%WZnDb!C=H?e@fai@qB2z}da-3h>?Z}8;#pYh)Gq|Vj{Ke0u74KE6`g$0l1mB>hSnG+)@}P;|^%Ay^Sqi=_jJDDz?@VeL%}Zi) z4yiJ|i(x+9TuiLH^0;jkA=Tk}r1})iJCm=ExvHQGGWCN@Bym6CagR0qkS?;)m*4#k zEv}gy6^OhG%0Cq97LA{{E%!(|<<|$mjdR8;c})+gqNDoTJ&zvOvqY^)3!`KAV~DHYU}-tr&xpx)Zo`N*KL`Ck&3OPN&|D;kc_-rw$_i zOBz|lmwtvnw~c;ft9Elk>x6Z7fG3nljfHuo4Wt5$jG0ruMdOgWWgWR2A!51;B>pz! z1|I$-#MhplVak`aM{58RM1ayK)~peUPd0lCq4m~rbiOP(_JAn;GL7PyEz0Yi_vT7r zQ7IL>$_8tY);v?=N=+@R?`~a0xz~GRvL>nO=c?>p^&9xj`El}}K-g1pSuT~Vr0ot^ z@p*c@oAEbPfU1w7)s%T_Mnqk~&r!a)-m63|JYWh3@3?-#G5=9w@Lg6p5l2BYkym{+ zl{^`b*?d)`?!_vpbdqfS*9EG$`TRAh1hV(*19!_{hFnNdVE)72kEP}{0pSVrxPY%A zxLi{7dgbY1uL@T?rP+GW_Aj!v4HT?S}OrFWJRith*!#1A@gOfuL_mx!ljbf&FoyPQc$gO&iMJ-+B?D! zEeC{_YTqa-ySlm-j1cg7LN9V}hFh-)_`&7GiPrGJ>=*f!iajq;Ps4w*hDiS_>7{&F7;fhO;#= za)oN#dPKVALgzVDZzkSs42lm(1t6mBD}*kaSU0J!9DXqHDFZ}o2i0$T3q^lm=nVbBx!IOM_rXLG<0bsQTkg+6}jsyGcqaq-9;3{v~a^e^#3=WJZ}zCS2Wif zj%%aeUWv5V!^e=xn&bRA#FZB5{DAC8bVX+uIm5y&1) zkh^$M5+2GGz86UA1dDe@HwF1HSQ-m(=ag9F%TEm{UM0i1wTEsNuy`5|~)JXoHI zNTq!1Zf)biV&(u~0ypr(^UjyHj>nyYDr=7Z!mK{>{x&IeZdh}nB%yez(0XP8y5oKy zbf!zDqF+q|c!SZ0#;XYC0%bmNG>L40WT%SbM5`%nY^ZKKUI8H8a)H}awMxZ?-bUK~ zv&=iaoB;v@owx5e`{Yr~`V8ixb4e#G#ui%%R*)UUnZ6(7)3?`6Sbgp{sEYEthK)ai z>(k%tEig6A!te<^na{A>hV9qMx3iea*<9J*7tD|D&%Eb3&j`c=F`;$`lFg7A%kRF) z&z$}iaPDhkT1*{3#WwS<<5X)7Z@jAiz{B+ww@bEBvB#Ne)2Q zQS^&ynuh+%(Se46{5Y6A`T{qsqEr@f=2Wm1`h7$&URbO({5v7ksykt2@WnR~W+ExZ zo&QQ@{JRk5AtI5o;ancxHTJ^&;cfKBfzGRO=gR}tDzic$XQ$q7r9hCCoxReNw=v%6 zbtb-^PZCc9kH@`pl$e>_`RZ!4^#)#I*NQ|&pg0JfI(zZqU2Ay(ceOH+NpxSd>tOQq zfI+uz8B)aC*XT5cY%OQorv-1Ls5l(;78kt6a(f+m3~nwGXViPZVrvylmW4!ZE)tjz$Hi3NuiMUqi#G68^~Uu5FRbl z3Zm@310yp*XGliIoxGzOLZ?;#I9c3;rKepS^ZX@zGf-;^j_8Ki%szUieLd= zpuvw1qo__zr6-4#XIsNM^F?WbueYmOb<)W+l{!AR!SxNaCDvW$m|SY;POupCh$oRs zCsW{=-D8gDe;UA^t^hLGz_kbfm}9icKZsX>CJX8dD{dWaD0vbpZ_vvqOI;jTUtdY7 zG-(CV#NVo0o$075;lDQh;c8HLXF&K+B zXC>S|<*$z%ZH*7Nx)VRV-3F@XMjIQ2dmq$3zqgXyhWiftryV2XUusWOwKe|rMMaYW z9p`G<#L4=g_0;*R9li2!g#6cAY>H}o^h}GX2+mdTW*>saey7#%OFn?(=J&0p$Jh5X ze@vMn3R@2_Phu}l))l{XaU9a=kPK5NVT>tuy$$C%$Ks>aY>+LBqSEkQX%8~!ra3Gw z({m{qyspZpOZJ!g%8iEay=W0!o%V<0&kcbOIO;m=u1swf8@36C0h3( zT(@s@FTKkI>`W9`v{N6>cQ&K8_CgU?x1^IA8$L!>MR8eJ84qWbw?}AFNdEZXQFOp! zaqcNHKX8d}6G0-CF@1`XEtWicmtT-{WL~7J@05(jg&dFhE?mrFel(snSJ1(({8S%@ls`(n~=u`o1#IZI59{D>*ae z_VdoDJ2bUst&mb^w!vHD`TA0tLWympwcU8xm@21zC&83LaYj>r-U|bJFq6<7JJWYj ztw!}MTT!tjs@#Jb!jA9RHk-6tBYn&yR?}#+jSEbeA)y}A5 z5jF~qEPA`!6V*6o9h%_vCgcmg1Lsu~nqtj(%}`w5T?(Qia%Zzv?mvaTOL*xxR?fCY zNl8VMoh1?T=fjcPp;V21ODhA(l|rFlW{g?JO1V;)w2!nu9(PNnBrEG*j^y8Oa^5yV ze9wslsvoeq(30T=GaVhlJF-1%-BJC3BQ|=;bu(IayH_>2Fbv26$;{EiX>V73TC}mbqPDjwm<)#hl#CGrnMYS`ZT&AllKTTTBk0cJZYs zN4{hQ;Y)7=C#=BaXl(G@hcrPI&EI8Bw`yO z!5_6YlNxUb5Y3rQCqhVqZ(41y?GOV3pULcU-bCS%w0~5f43WHX^NvtCZV!qvn)mJR zsW>o>vnMB*XrxMd+aD%lI0!os*%XQ`OwiE>VyN3rZ8G>+C?juXD-Q~n;eo$Wcg&He zM9k83zkGhjoq5N?>9{1PiNmrZjOkxQ3IIa28Sk<<>(0;p>77v;NKlbZG^ zukk>``;oASaCqD&c;`1)Ec55-`lbst9w0Ks136JVdRy-}N#!b|Nzg{$J86JSF`n7$ z%c;v^F(1%2E{g?(IAoJ`%+#C{g72mZtXwgfwF}md2ri1}Q-~$wSp(ZDC#l*5?boJY zqSae$4ja9qSdZ(5!=iEE6xUt}W7^Adxi6;?Z?fOZkX;(GV-aGvf3zK6eE59Ap&}YZ z_P#quxN!EYLsC`rB}*H?p^>qwyHRB{6CvoE=o7cP?K)oIsF%1l*qlfSGGU+Doy4#K|Lq zr~zhfx)wBWVl&n+d0XFN{sN$1t&*nLp2vr~2VJbKZzg#Oe7%I|1X>yV)=Pp`h2(N_ zkrW3=!GgCiQvOTK6$!eP`zJtU?f%Z0!o18c1Untw^mDoKxsw@M{y9Jx-0r1q`Y`@p zKIc<3r1F8cq2W?N@7#0)fzrg>U79ZW?73RTtxC~B@Ioiooa7kzw*yuNda z@3G9Yht~_cb})#W>u~wH^riKXqUL^Vozbt0VpRHUzcm!cd7mIrcZr>{Yx?zUvu!); zykot_t)bG6yle-IdIG|OcN zS&os3j7P%;oM(te3>n;pU809_%kc{|Dwr^*Q6wu8SW(fagINRcC5m#FZ-;6gn%^_! z>nS(%kVe}nBU()jx;o2lnlJ8-=!yL*nUUT<9N$jR8wrE(1gA1k6< z(5De90hhV_P8MK$U#$_l7ZM@1=f2l&Fs{Trb-ei;3rX<&4W}MsikdXT2GtPEvvK-X zw{1MFIsT@-+ZWWDmA+C<4vY8Nlb5YpW|Na@rVFK$WM(}b}zoB>y`V0Gd8Gn503eH)KxAIohcVjlTTA1 zui&D?Y#3Z8I4OtA_ipAoF27vc8W~}^XB)4c&Pe8H%*PQu*VSd*kw+7aW%DrU{etr+ z&-Q)x_WIC_M7Dxcv!o~p!K*UEi|Ltq(HH5;_PSIy(DnMiJ z0?=u3?G6`weDVS8q?koncc5~HWMZILG?iHLkuw3^EZrN?!=XmhtIU&BCKHxnVCQV7 zWv?<)6bs`%UF<0XG;{OpZXrxJKlX2n%b}Z>nkR5q?Y$0%xzCkZw+lWg=26SD^4v8a zb|@8m=D&f%o-o1`sEO|(C64D<8P60}FFlOr zzYZ4W_x%(yvIi;nlsTTRjbJpfkp<2Jdold&+m&Xz;gFRroi!o?OYh>#2Z6u6MSQ079Xg)n8Ug>sg}FC z9s(F_U&VY_XLNhP{uP3s$Xh;~DuCJ1@SD0z-aPQ`*8SbMijq|wOC?gt~yjWR5 zqU7UEe>YTs^O%{3pt83m1^AYMI3DO*P$z?qM5+FD>#HtP_NFFnkjD_u-A5IOA#JZj zx*U$4-D{CdM&u6z$hr!vQ$?`bT~^VS7H%O0p747f@0n_&12Wa91x+9(s|aga4W5jv z`DrXTc7b2RGSj0mq|PDS|1f%*-bMscwV=aeJxS*d6mI= zcSGo(=e*k`l8E{yn;{}aE|+BqR!(-;@L98o7px{Q4n`&oA;^*FDaXvC(-qOQ8$Y0z zi4ZuWe&e`~vsalv85K(_SGJ^x1&cXGay9jaj&*0I@;R5=tw5}@jJ{gbwEbwARal|J z+{|^xK0|isD9N-syy(l^&tK=NBpHlHYLc|)Z2cAsyPfr8jmA*5iW#WJrMj5_yeZ7O z&F+SoPwbX-1hF_DxpFl!ZPMyxE6?59g0p2MY}puYcExkLk}B*8AQ()08PH{BZOgHSQ<{mY3Qmx;&~wA(V?6+w&~eCt}$ue~Y^PZ&EgHy-9U)P9;B zPAff+5X(2ypxte1+_s|Qz@Gpl5`P>Fp(?a}Wyt={l<3sHrjM)*_*RO!rx@IF11#Q;*9`Q*yrW;eH<=^!E zYJzChKX_Sh_N0p1nJWlHog}UgIDE4a_UEi6xnoNJF!cwDUV=oF>6XFo7ne`w4@2_v z-@g0Nzu7Ud=)qZQBU?^#9_%+}cDUfcTOIsY%$>u(Gz}8B-P_ctDAJW;=dO|d6{LH0 z88Rpg%Kf#6D(^Hc1oZ5I6DX6}wC0!oeVGJb4aWboA~%2>jQNb%(jyFOcjJev8R_D5 zM9L{pX6!tsmnYvyM3Y4_bfgwNV{RbJku5Yrlfik_!Mju8ZQa9L{iX1)S?V7p;fCY%E>fK8?FyzmW zEGm--r)OL5ij}n5INw|~ZqJw1H0it)6mzERJYawtdpm(e9nm<3t3ku|hNN2u2VghGsx-axp?ftSIIT{Zxue z8#>IGk}1V?fbxC^=-PB0v}${smPz1ro8yA9zTBax!KCgjvH$r?o%f3*dX*k3005xW z<}zLG`qhx7m(+;B1$jAXIPs`|mPJwUyBd)frD;Yn0l2RVUK@aw?Cgx^rk=GcYG&d( z)?jnpqGNK~;s=Q5rH2{45HPLMI+yKl3J7(#rYH?vtiU ztu0(P*S2ZbmH8|fqwvO=PWmuuCRbapH!368Bqm&9(R?pn`GxYY&hEKbhph2l^e~1?P=D2F$PQ;>dyNb~ zxno(zMH%@GeTH;4D>N3v43;H+I_Ww$rAkIv=?^{ULEPY+VZkKpdH+v!;@MR_4FH;awjN~15`2T6A$Xc43ztOGsquDeHYkoU&){^b?sBhM3H#U(lUZh7*~>2A%c=7QiQ$L9=_AI177Ki?)V z&u!kJ7+ z?8f)??pmH4y=&@cWPm0>$WS{eJZfiH1Vz$QfExg{wWCoiY{F3?=~dCHa9d4U$10irSKigbg6KuL=tNU2fn8uho>FihlQ(z z5WJ~YR+Y-igf2wzIBhF+AF4e?H(Y4R^b0k*xO2a*;tY#9>U2usNxfhaxF?Lnwo z*~6yhY)*lZJ3kPcMjRGxQC`{p{4Ur>@x4*O2!?m za-)Xbjai+GyiQ$HamAypxq_x@fEEWv>x(8vn+L8vt_~+>7*3}f!6&XH0P|w87f*>| zaJF$~MRE7)WKGt4eg>xsn?3*doLYMx;QI{s)xOzYY44rUjv{q{%R6^eFrKatKUqx$ zn0aFHV-gq)GflL*6rgr0x49};QtAz7t-*!owm#WR;WrxGKEIqkwc;evj~w3|XqFB& zp#tzCdb4S8VzyRo{w-ISvp}QV8tzSjwV?3;T=js2D-y;53rP7ujcwZXymgGIRJ%jg zp|%3$xY9)`fm5wHNI$ZAgBj5xkGP3Cz&iK+Z8cCM@<-7bgMg|`G zvqAx#g$n8~c?KY-w?{Y0VNbPfvUkS016YFv!~snvcnblKmfaBiq3|O=Z!?4kK*&(= zW*UHA&d8Hp&nF|`uABjMYZw5Mj!ROt`Zoit@~xH3-m|rCWU&~ULV}jp>HMo70OuAC z+CS|0br2!f^!{h>kSeGuxV)eLr|+g^5bx0Grmf?!X*=6R_f@(1IO2%Uib03J!t1;T&*QD zijVw+mgvXEtkEnX8zdf`R3*2mgS;xuQr7f$Wvsnyfj}p|U4gO)p2Wz}297+e8+Ahb zkPv11y3J2}GvQ$X^SN08a;1kH@BdWtyoo)~t1U*Zwvo453cAlu?&nK5EBbYSK9_-z zGqxnjCwR!xa$gwm?ElLhd;;eK`gf&ss-93zyvRk23{J&lDUbwl--}4L06iL?LD>Qc z=JDlbApxQiolV@Gk$q>N!JF?$J%GfN!%S*?wDQdHltSH$w}UDYs*P$!@p>iyB5VG~ zXTZ0G;bgao7?8x2jENb@W}(&eKKPlCGyg}XGdC4l7kRyJR{%yA;(sJo{*)x3m0PY{ zcA5~Z@$eh_XvxHmPIIEE?4x z04T`JMoZ=Zxn&ffEhyOGkQ?Z}nRv0s@feDhSVsbhZ;1WH+Hk-FQyKSxdXn7n7$UU1 zS?ot)wL+PnD@!7Syk!vRgMuM<_1cI^b>uh^1F@g6hs6EQyM)t2eNal^7jv2iAd}%i zyfVe6;rTxesiA=$u|UGaD-wIU?iWB@&iiF#_L|dy`?LD?@xIbnZMqry%ol}_mZu>r zJEVnXQ(8V{^w2cWUU<2b5bU~KxeQ{bz5*Ht-} zTlygs1b>3ZA_4&^(!9Scl(tarLJEUS1aAVXmEm>^##qiY`q|>}s`+`TRN)RnH&_tP z=RfEAXWQduCP<*sa!Qa*)=;g{>h@KzNf9bs=3KxjO3F9pM6z+x@$OdS2NWc+A2QUx z&tMCPLE^4`1c;}9h_^Qs#Ff+|@Kk^k(lRRe>n0R+*~g{x-ATr^$z?{jleZCe1_VHh z2dXeL{&{dk_B%RlRyiogzfbEK)ExfN*~kla7CUV9xWT6BByOv8Uj&+!<&WH`+^OFS zIpjtNi1}))+f#BoV;;8rf})V-RC4y9iEN7g~VS% z0~*E6xA^ici!MzR&N3P-XwTH|Yd$tkYWRBGlE)?*NtD`y=PsaWckc->#LO!3V4fc6 zd;X>SpAEM5K-tG=fF-_AdA~WR-U{(5(6q4IV z=pv>`{%@E;8>!{_I1U)-v(#omrt;VinH)MJIM(x4dwDU3u)Uj)FdY@x#q{xQA%ov6 zuh|u-UMCF${gAik*67aG_)>0AeZF)<#R3Mit}`oFIa#|c1lf}*{eMuqpLuNYE?v#W z*lszt{i-h7YQ}Pf^2m|Gx0rhtxQob<=YV|o+(SedUq8cy4ih-irx_xIEGUgRu;`}2 z_w3VhlF|H|%a*gLD!AzSuNbclM`LB93q{tJwJ5Phelz>Ol!OKXv4Z$gLZsRYcc$O( z4|LCx|2LNyUZP!pqzf4XFE ze;_ZjObF-l0xsj913mp0-xxszhmvg`#!28`Dptea?8Pb4D2@OO_Md*ozwMEY&j~#@ zQze7gwJ=l0%?W2=*Y^VkN%s0Bbyt%hN8s+R!EfdHuSFoJL9%ErX4@1Oy{g8H0`lhm zSx~n+SF$}n0@95Rhs_ZDe?4{qJ|0ATl!_<&o+`zQty5g|S;7NZc(@f=nE&qU0CqEn zBs*GXRd&UByo|)**8!A|xc|e-K_FNLeS@CUXo5`>FSe*vz>Znq@l+PYmQpWONJ6(|329HN><-miVy593jFn$X9=>(rV z@fZ;Y4td1cbeiGmXY^wDt8G@Iac8O6zg#@k*dl?R)@jaX(IkXIqJ>NvMRf|^a?iFs# zkr>D;Rn+wPlchgZ3|(vV8K!1~oIwB{ zx0vxI9?^R(t&hVb%$`r?_J7y#ulKy)_>_}g{Z6zXp)>F#BHCbdh#apCB4U$pkI?+Q zeZ}_npA);ak>y(HOn$5OgMA|TPv-lm9ZG<%*a`Z~{^s!TMW7QFN)(YgLZVksH$^8Z zO&}~3KDP(Tx^L+qJ@kpo=6+oL=L1Ot1Dy-_-TZE#XHaGI4C_d#cd zs(Vf57ylVpXxD2dgs+;fwCJg)Yy8^-`TIj?0$`Mv!#}^sJsqY8 z&7+*7T6Cr6a_uMqBf{=nteFG8a|-Nr81tzAx z;pwFl#RWM@W6-r++{6f1ws~>z`t>u=4ok!@1C5Ale!R#MF_34PP@>SJ|%-(h%1jUl01HVm7}Yb;(k@p>(a{1VdO#4KQvUH zsd~HSpDjv^0a%py>9F47)AiqqEg4_G5WQcK`$I$Fan4NC7tk~+^{@adD!T*MsL4mz z;}{?OJFWZgt$Y02De{jOf;_IGec%(L+noHUqZFdaE@ZE=B*OAH2d>L0ihGO3^NSI? zCcx{=#it+M|3(|03PF@Tprg?jgnjqnPi~A~-PH_-dSb`-c_On!0Rg4Qu#mG^ruJk4N8piQ5<%|^_rfF!)w8% zlWi10S-%3x`nj9cF71=vFLt?LtU9qA@JzZr^e ziEo5k6#JnO``S2AR|6~`R^TRzg@i`(XcI^%!U`u!%Qbhtr2gh$o|@KAGml3#6Z+vs z_~}uF^X|-mB0iV{iu)^3L@~OvqP>)%URE#np?P)W>r{|zu3PI$PK4cW4(93M{lAXt z8waFvY^?QJT`}?U>aex`u8>1_pq;bN6#K}p+|5T9izv-lzI39$=jnd@=1EG$qdHaK zDqjRW*=~Gzk9_oj-k(O1MB9<@Yb(oZoHx%)Q;qYhka%38+F%={`C14ehoL4!pX|re z-2<#q^rJPp7%J0wy4`3VLBjC8_W^i49=XWtJGJG%&`XW~)eB%^qnVnYk0?t?^D)Fp zt6RjZu`khoVw`^y8X%FaKSmDguYSxE{_SIrGKrRSWg3%HE_s!o%rD3 z)puUzGNYh>>=FK5fukTGC+X?ehWc_NSns*ml1o^~oYZP%kbxV6 zIM-p35J|CA`%^_g@NX~hzd!US69AS|W?M0>^kg|cP&C;|5qHVKG0&4jK;NSxKMuEI zCf12QYVNGVH*as>m;W#ZfnJRu*&*!lxr2moB5v2g5tQ$QQ$WLj|83oGk16GB4j|CTlboE};?=OS$ zMebfDhy4P=i1wQ6f9`7$StbI)J*jLN{8WoXV085~K!l4f)B!rey|HHTo@pRd<4I75 zG#4o>9tO#BxPsR|c7wvc5d!uO$MFdH?=ld^i-1_;bf18oz%CQPeZ@*AM2D<&s87eBLtrg8izEucp%!kuQYoV^M&l1o$_K#b z8awl+VQoANvU!YV2W)hg|H!8UVh9{y-@FdONeR|L?Zl6Ie3trxl2E63>Pth5vtqy>(Pu`xY%++=>@zad&8O zcXxMb(c&6{JH=YuwMcP7@gTvS0xeqH9YXO!keA+j&mH%jGv4?8moYLTd+*<}x#pUy zpKxzyKP%2d_#w8$VuYW!oD`ptk1aiAQH!WV%ge@ItG%)@E-u+vkL`Dr5_`)>Li8Vx z@On=%bK*DgD`Hms!a@SjnC8mhAIl64BS0USVt&t9%JerY`SWHG!ElF{63%!otqs)q z?~#7N6N*c6l8&3s^jnqsXgL7gHVT-Kq0r9fwK_0Kzb=57ru*8YP?X}+18OEpiX_V) z$IT)beT?|`bB@dtn#w1k+&gqS8pCiQS07-lp&5Y_WdNeZrVLVkCGilZTWwN{&hX}+ zR`@?@L+W$HCEUDpfzAaQllWYgbkr93!!w62nPv~l6RQMPMaZcuQ!DO5=`j|tS@Hu8-SjEsr~hKTpEUaa zJ5m~NCq7u;Zspblc_snSok2wQMBx8c;+Zcphph>`l+SNIY8_{-?A9oVF=Q_xAq3+-IsSVml#D zHuOYEapJ`Gg7M089xkVp*giq@@4)s%kl83I_1NUr63jMIcA4QXkOp(tkMHDl4?`Clv!W4rX!u+Ppu>9<|Gec-$sxIc{%s+Vy{iDy&u{)W?O>)2t!5YM{3W79DqyTV z4`R1`*EwPil5gbo3jp)_Sw_6g`zvI9i3|x`+$<9AHMU73=08UJ%bX6(@3hVeK0W$f<9>yHbxMvm;3V#`_aq+45{8{t3r)FuA#` zxhTDY!{pglk?;z6BT!!A%X18@j5efGe#C|HW{L-0tT={$T`hHJH~!Z+nb;~Kw~(>P zhjAT~<`ebgi@j;JWPT1Z73&@4{N(m!G^x3PxP{YKCnf@E%0y!y#>U_Si95Q=+3%*l zSXB{7U~kV>WhJKu=w6>l(A!OzC6;1ul^AXL5$H9=j%`65w(1C3>g{3TLV4cm(!J6p zQY(h`Q(r;-bCnyct&=SqOUZph>e=Gt;@-+f32y?+DRden0SuJ7I7%ZP_qy4-6^2rl zRu041C)r0C^*bG$Ob_sNU_cODT5NoO+-zjx&8X;i$It?+q0np0ZKfQpoANa-E&m0d zcaRnG=ANams@bz?*FV#4@JlUo!h!utTTZ9INDjD+h#i?w`c$r1QM+dZn^W_;FMuJ0 zC><9yQXCL=hmuW*{U_bh48t%4*$=FsM4QEKa*8OR;NzjtqOJrr zEC(t~A_5S48*aq2Dr(6Wf^dTU0y8mf9Er+5CTEhaP)OK-a#hE@eJc$M17f_08h>(` zH7S)&m^0&z8P~bp`ZX-J8OivbRrrfxiqd=|gWZ}Mb=kC}i2-A{oM2bS>^IN%S@1U= zVaiB*XRv|*8`;j*w7qKCD!lVqd>ICbo?|%U$Y$#D6lGIwFY^sBW?AHG#EG@?#}t5%_KyQ%W8?#xqthXSF7AQQd*xeR}mWv%#fllS2RT>w^_YiV$OA zVd1rz?;K4)zUY=ekyXKH0~g4S$hQ&Uuw#jnkYiO%GYUu$qVBR03BEVRMN}Ev0ssBQ4%JTj|z=p>O3IRNEpo zl>atKdLf;f1b{Qag>-|K+C8O5X0v!KSBi~~zoFe=tVdTm^r2kw6zv;wBVlzevNC?C zhCV$qM3ESL)nrgwtC3FP>Zo6RNy1@FTf7F-Jk{CICXTGPT76P%c3Rm@68d=wUD5e~ z12ORP>RPZ`Om4m?!WpbmHCRaY{`u3K;MYrpr-$0)$?V~nTwY$XI?*^bVAKKomxNB_< zHJRwKFBMBpl#l7VP7sX{H$5h&Qh}<{6^T#GTZ2quh(?3T!dhRhvGe_=sdii}&)2Qc z)gqWR3G=SrQ@xsUdlu2{hlC^Lr12Y8^ufo7kCbi9`trt#X79M>#HxY zUV_Tro{VWI-XenQh39vmJcquno2nzjV0I$bR2~6u@&ZpLnN@zCr zmZcf7r<9r6)58xEkKaDc=majsHQD^BamKse&aefn18yz3ynWYN7YfsKgu*k0f^mMG z)jrck7bi9QN>E3})tKp+){FSB2N|3rgshd9+meT-Js(}ueHfSl69%RQ8?>{2}r)pxzZeGVjcBz1ghhG~u>%E(7&zdU9T9 z{Y6GC&k*o;U-)I_PfrUA3Ag=BxFANxZa}xYqiI(&ul|{f@}bc}8(UAbUER|jmnb{- z(QIe_n`e$=^Db|iR?A)LP=tnJt19q{5(>B{b#LYh5%^=)Dhc} zF|=OoPwjVWX?6{1qfgB2JWCXvt4Y$WG2}-5n2KDw+FS?MO&S!MQ!e-t-lwgt4e(Cj zrFiD6N6c>M6a=N@R($*wwdNBr-#p5UG4^rH;EQFe8wTKJ)0{xKhaXOMNCcVS4ylRUU0m*+_U3myHUZZA@3$OOKdhY%mnq#9v z7!c3)cn7cZyDPA9oXh7cmpZ#%AK4$4#vDwEw772Qd4r=LGwFuUS!Gz18I;x$ngdS8MfYL)HcS+cYEuCH=hYGjgEPcXGGy&zf$L)n?o}nVHy`glzQ;P!YKaCLPH}3#e)d}JzoE8l% z2J~@OcQZ&x9+U2Z;}iGZtpP38pOmIe6CB|daWV7Be*?U#93y93jO0AXY+K^#(Jp_Ho15GEY}4laHr2k~earUN z@7h$z?bl~j(t?xT93~lzWw#yWQk;OT8ytpT-|I%`HLVfz_d;1b&e;FipP(kiD4`N& zUV6GCGx}Nl>G*!A0EsJk*7d~7m5O?0_GS5AwVXEad*F)^)UB94{(pCXen(2WN2{6g z1-NxVTXO(wh3Nrtm?l4 z{D{Ra9e0!SA|A0sD~%5JQM5?0g+WV1p*;$E_iY|#&F@L2mtd%?*=)&jrQg5f5rsEr zM4fi7q$1cYOtUgx+PoF^{}n}6(V$3tHB(Lb1?qE?!>UmrRKT{|*s_rs`<~{N{BqzB z2M(=~Q+NvWLSvbOuZR{qTnGy!qG%%~M6_gPAUO@DNm8g>r9 za}ut^tG%d2M4fF0qEYTY*%>2(^7h!SyIt+y%7Rhm(8lp2IIY1S-h)|4{$i1hCfv)a zoA|HFy3Znc^ivLMn^1MZe{f(#_0PU3PD~WRaWVm7HXf>803r9o##m#V;j@$qf~x6> zJ(Bcte)%$4i9$<0SHy{2w=7j9F{=pEAK5KN#eM{~FoOA3^5{44r_H3N*P&Ib&e1@o z(|o!VY^YwEoBrYOn^BkgM|>OMy{n-;wWXVp0hO1f;{rezx)B8p(n3m8TUR}xl*UFS z*B)od*yFaFCxaN%uSj+Em zP2y`y+g8v9Xg|ayitwuV<|;*z3%Ooj#+^!1S2xM?-EyY8#qX6IE9{}jvH1}ybN|Iu zyN9r8o`|Yw5tx_fm|r4}@YgBJg$ie$D)a`^WS)@EMq77N(sDZ>#62ny5&Gw()ijZI z{WF!EZoAlZY{{OOLe&qV#gPFzFfdH&DVn-HZiSa{i9TbYOG zNuMMF_?nSYlr@blJ;T;7xVJkan%n>g^xdG26V)f5b)1gvM~d1x-#NC+YQmn>FyK6d zp9Rjz!}d_XsWgnupprV{%D2t74@anchy;nI$Wb@#x+X`sBi!(6Ge9z^QB1D@Yj4tG zMv7Z+JP`3lf0-?p?q@t3>VVTqmRLw>XXnu-WJmgrVapIAa`yT0$s&Db8s@i$!+|`p zdyA6U3)fv0MEx@J!z%jTt^bBGIB3Vefhs)l7bx8E7{2sHTMmt4!*Hg?Y)tf26aq8Z z**g7@XDlB^Y4SWs3U)?5xQ2{&NpWonm&pT{`NxKtUgVFm#0M%&bB?CdEu1RIRn)5G zG5eOA$YeENe|AzAcaHUg1hZ9`3(>Y1QW(&Ycwxft)s)RrA@4%T!4wvY__D`kb*`wt zl7k*LXo1fy_-CtU20as1`WBPAVv~c`RGXfP{UWb_qAut2C1q*p9OW;{x45hgd{vrNK2-i! zX-OWJ`){%YPDO5iZ((>$u7opifCa$(}ccvvF`II2WMBlQfno+wTU zdsbJ<_vM@d(mqoe95k8aR z{Dy;-;SY>}28bmYFS5XoA-$4$j-i_A(W8W&N^-Q>W>%kwX=ZY1n#3?gq@p@wR$*_F z_`sL67T?V1u`+6bij;-4c_4viEv9+#L5A6PWkFvv&icQL?3^T&zD!C_gT^npoFB%S z1l0Zh9r&Qeud?o(_RgX=Yby3R)7eLc#1F&hMJO6SI+TzAD=YR9WcgD>UR1eG?OyL$ z>fYoJtmsQ{UyeEK{Y*)8;s?B6nqlx{dr-?Wep$%g=@s?ti&f4s*JNE+TmDcVRp%Ie zQ(R0<$J~?ynyln}Be@2*QS{s%hp|n)w5h35q-GB(-mvtx(RW3)G2$>uYV*txL3~Rq zkaZ|HaQA5A)r4PG}`duS2$iU|SG z{w}|tf9h}Rs}CoG*bA(<^N%5&E7opZy>Y~ZSK<-xtqB`Vc3>VDKMsyhgS`$unidd5Gb$xrh(M0m}s8y_{)jf39WV1`KsPMWd)by8@0;tkP6p zVqb})8XOz5YBwTEd(TAuKH`c}aSmMdC2ZGQw%dV?aeaW9GRNJ-j~w#51e4r(!^N)g zwy*Xm&wutqgun0iX>@sK**0TkY#_7%_~Oqvd49j8gEA3UStJCg7FFsd3&;mE+5xDO z#CACi8DvfmIe?d)EWnF%J`?NR7L}dc*n^b;4uPf&1%O)|!>B^rTM-axh|Qbsi_Xqi zQ z?d`5O@3GtGLtV2S&O>v3b|o5#=r(5{@!&`4mlr*T3oTqU8YGL2_6)&~)2Cl-#&&(L z)$;FKRSVvz+}(7Ut=VWpt1}&Wi5K-8-YM9)d7=wP#U4&s>wI_EcyQi> zMx~;xTp8FPD=$CFR@*_3EACcc5ZoeU&DkGi@pK3<0=vbml5may*B+?*q9ji0QwxHO zxq^eYI2PLH&tArbDnb0N6D6K*Z{#J9xjiQaH;*s$0yWr44h6dhbe6zyKG0_`Jn*B0 z?w4YTW0b$J|Nr*EBymE&#gXj5Vfc^X$B{^@YU78lK6=^w8Nco1l_CC=$(nhCF@JRu z=+_my=Qr_&rYIzC>p^&WM?DDcg90j_5)W1pARo7?tUk?4a2PfH2nZo`b|k{sA^F`) z$(fD5T|q|XyxfaJDeThute_l5^VW@6UnoQ5w(3^i$w*J`MGe_Z8eHOxRH5QIz$NGP z2^ot8C^vj~TqsyeqOWkwL+02|rujW6P58UiaU9pVfjyO2jMwp1u()FLAk(Xv!ZD)=r`kzh3aH46ZeotbPqV%C zWygf7!{a!&Jw$sAvaGEa82E((>Xp|W`Ew9=xBgoM1xe)Ponr)d%wlLXcbJ8SsVK-L3 zrIV%)!cR2R{o<84UL?A|kOd?KP1-#e`C_gNOaSTq4|emNNK>#~b$BUKvQRdJ{zG2w zB!5)u75XyG{puFIcLepTA;cdby@2c3oVB_`hZsq}?W51oD3Ik`%h)$Qb2&3YG6831 zxBPwxUVJ}ZMGv26xATgr%@r7X(-G`KdpPLFw0mLrIVVqvbue)2VbW7>)D77MbR#7@ z^+^QgK?HBj-raN0#{(O-q3IVf%4cNFXdd-`V;%uDr zkoO)bn65NFk7L2&No!r_)}V$U9RQ>S5y`THXq1t`ZtciD&p&6VlBsKlmIzB-D~THM zRL1$TEle`7z&v>UVd5v37;z}9>Ix@30$l37?Pg!G=czk8igUZsY+~wew8pktz&jRg zeYT2McFIrs77?iZd{_9CK0lI-Q?}kM!nI^`EOXyMTZe*f^607T#9-d^n|KJ2RdE05 zd_bWjY_%4z$H$ao&`s&l7>%}Di4rM`b-(?DGw%!my5d$_Ai9Ni8+kpT$8OlHUOH}M z?JcFbXgv98@$fPaj^FtPM{YqShE$VC3ObhFeNCvU$86_NI|@5_AqHht56LoM==h5nRuvXgKF;f<)TSLy-RnRI2s*H1& zvbw2jm@3mdTj3IIa-Axf+^z|Rq~y>umZmTJ$S^61c*@V!t}!pQoTieP2N*G`k+WgH z0_Sr)%X&I}T1|wr=#7izvF+DDpGr#e(CmU*PLhrdnK2m;H{}8f?dt zx*krbYzbFoOe#{G87Qn3TgsaQ*=U$;vgsKXwSQW?ep0jn^NxZw{~|O4Cnn1Krp-QR2tUwG%kc0-{V{V7@kpi0lZte}B)rs(=A}aAYuzk!ZrxOIs@0=YM!kj@t)uR z*`i7b<~k-pX6X_p*bGKT&HkjAhe#t?Y22DDOo%h*FB4c-`hD=lffeI;DuwXQ{R@ic zZ%Ib4k7nh5HMlRWn(p-*D2OJqr2g{&przi% zE^PhfbRi(c^j)zY&axWKH|fqYO7L{g28l}vReABQG=SZ8zFrB$<}9@o-R72WjG$m~ zWLn>hazrqlfeT%e$u)a>Qe@=6#Bo+3&^u3a7@KR-j42RE!2tv%)Q$*w4)086 zdA?Xh#J|pEzDA1i413F?*0UuB>nqZ!nxHNEbJ4qj3tSPf2j}<4Ey!pr`v{wNc-(cg z0;-V`d_@_-D9--$c&M{Wk)`piwp>15|K`fzt(aHYs)72^Ssfaxu_SwNTxVaK$Jqx( z@1jmJY36omJHCDYu}_TtUnvay3%;u_i=5?th5EY95Aq=1a0*HmOJL_BfAeXnM?P|s z8@%`djsGZx^F$$GHgVKoIoR>V_O(J#xZ?52xi^L7_L4s!U0gV1V3r*VJh$ottj8Ce z)KhTmm={4KA9Dw7LC#T#t}MF~iQKXu5^$2g=GtE5Z!EB2`H!1iAviy}TP$u{>cw zqQv8dz*OVGmFNC?kM&FbV_zZn^O6pUT{_>b(_p{?TmmRPYi~68@d2(4atvQdvXf>I z%&8Dho+bhJc|ykxn8xkidH1Ilj|I1a#|KQ?2W3wZ+p|^u+xmPxtdfQ#3<^~F;MJC=V_ z2ILe%5=o8Vu@(^;F+o0?6=+xZ5JXb)wFy&TMhx_kdL=eL>_hBxtJB?`G`zxd~ zj384Cso1GjMIy14?}d>ph&tr$aRvM*FQUKv+5seRf(9D(&CwlpIZSQDarRgA+E(1m z=;J=^%*tKC*FBMOYPp5O-fptDfXeYi#l$ah&^CxhyDu9P99~9W9FXX}&f)0fAq%*7 zoV|0hlR+mv?wxTF(y5nOt`lGpS~I3ej_gLDHB_OGQx!zTLp9dv;Ie2iQg6#@BVL@? zVD$=3>|ad}VB?Yu8(YNKZ-+8bu!)Jz+Bsbj ziw}0mMTt1zcYkM1G%2l^`Ywj%q)683hEY66$b$N^cV-ByHCjwAV|3ES(3%U=r6 z(HbNdKNM{LyfjbpTrA@c0tXwbm>c&heNaNMhB2+tcFkwn9JDCwH8%AEAuhoTEJC;K zhr6g9#v>iv=dUMmR9=7hBv?8r@kW;VcerBk=sXrhWhjh3;uE^+ER_9o_!3t?K*sBB zc!PhbjFfpmxdWkQJE5Y#mTGtMdS~-^#>8))M8=2*~=5q8J4R%)jqlS-ayOQB8jy6mp)?pPcai#>( zWHIWkae=k;xWllM({eMBIo+g+V9&BIHj|QXXK$C*9+cqX$6vzrt9J26Xw{-{PErR> zRnKk4f>M=I#9QjKZ9d=S+6UVM-{VWb`I&IOpI0i!xI#wY-`M`lgvjbOhO`)P{`H7W zyag1c@&eAf{<6`hp|Vw0kb&8Bgj)HsI^!?yHZ>LE@7`rHyOw}h_n)5*#^#-VzC)(O zBi$T(zh_tka*Cc{t&H4%P1kpZ<>G%gYfKDx6}VbG1y$V{Pkp-7*Ly94WM@OoW=g*O zUx%YA8eDblBn?7YjI?mbTdtX?F5-%*`pM8{A;j-6tZJGiBSOWceV(z}ctPEY~y5E;3_352pAWQ%RP8@bJ|3pP0ITBlY@d+F`=xKkJO%Ni5( z6xz2~ck@2Sce_~wh*3fpm@ERez9ZX(CikgR1u5zjuKE+&k|Hop*j)Ar9ZMUjzRR=d z)*?>Ir|bR31!VTj>&l=fppCQK^yObR*(up_MVBFcYIz z#(w<*&!KN*q9F9N>1Z}a3uousLTpbctxbG{0qE$T~R!$(zr!q@B8`g-@pnzc`Ow%^E_Y*8Re|FV+jU zT6?>WSVU_VCY8WUFgjL!X0DbiaL+xMtxH`M+p`N*B^rfBndANXLa1lS7aZZ%+d+QL z1{v8oliGt4`-SN$ItYV!XZg@~OmK%V*5KFLgzG+%P{-Dy)dVd&Kfg0p- zy%7YU5Vs$)%kEYFp%k)>J~*GBKlclLP57fxON?$;!s|HIw9|vZpc|dZL~wzQnzvbY zTH*55hj#9wiiF5|6CgvT$Hu->O!zqJ64x@@&2Om^r>b7pDxauskvuM>URCg7!Bh`? z&0p!3gDc6hP9ECTTM5(;8g(3k=Bopr zfN5Q1o(IaM)#}PgyKnU9+mwA<6SK`fv+8h~2VWC27bg(TSCKFk%Zs1Z%`mFf*CrvM z;f%ufAeO|&^>~m{&z@e3bB?v{q(kq3reOR&m!(db!?~K29G41ArOAA;DxSpDjI}4| zeCtx4L4MGodlvT&mXPOeH9aFVgWuU$i0SNL?-ifJc6vr~GV5%Wfj~h5g$U=-yd7@x z!w#2)r(5#)_-D4aKe{X|Zr?phXNjDWZ6!ON;OZ7Mc&uL4c)P**ubTXS$guqJJpO+p z!)lJA!)NqpqxQZf$`TFduQZ7<`x(3eb;ep*B%qQBUdjk>965X!|}W$-Cg=aA9=e%szaDYnD83G*#}W)MWLLEd)d5z8{X7Et*U zVB$See6tMwu|rg}hRW*WMuK-RFdI*d-f>M&>A{jJ7o?%Ol6EqF-N%ah>f(T%?scT* zz7zn}lV~5UwCu4=-S^W$<=X9U75kN+4?Cd7cI}EV8dQkl7Ie z6lEosbG@%kK{7zdViU4-Zv)BH&RA=l-ftf%itkHRN7lY6tMgrM6VlB7ogs*R7A?h} zcOX|iGp0;N{^lyb9XQaIfPQuHp$KzglX2T9%9gEuB6gM8bRJ>9pdFMq%#c$P_syBB zBYSxFzTOLP9YWP$As3}@J#vMMhHHoWu}9qT89I~@O}F1AaJ7w&X1x(klN1}}-r%DxEV(Ar8n5;EBBwM0^~t>F$aG>WK$Ya=uF zdylT$l6V=SmuW$CW5ax8XGSi$%^7DxH;i`m!3+g-%6cdM7Gphc`cK5?tQC|fLtDRf zK3%J8Ao&t#nyMQMehhM2JGFl(!gLN7`_-A4m`FWcswSW|Ae5DzeW=a6)M!swp*wO# zE!?3o)IY=RQjfQ^Xf<2Ow`;Z3Tyu3wGauYei_qWH#+TA)-;&KE3l=8S#%lfI|IIYM79yX#H< zMiTmx(hVQ9-(F9}Naj70Y4s8{u_oBh*LMv5H~Sw-PD>+XfsEFs_TPTHA~hThgtQh1 zT+as3dH2rDxZyQg+Q;XxFU!4&bPRH5+>3LI?mLfz+o<#cFOg|mEbHA)XBql9KBxIu z8>i1jY9}>}`p!6o$HvtT5bnhw8S)iZ+6X-Cyx3{b=H+#Y`9i0m%|yVc$_x)pk{OE; zM5~HI_=q0;a`h!F@<73iXtU$@wiqQ!E8g{kY2GFE`-orfwwfWkLk?sS6u_z+w}3uD zy3|OM0=AyYv{jO^D%Nj-!JDc|`T{0kvCjK%53Pv8d*>(9dEuct}ecI>e4~oQz#;B{^Zr7%;nL-z?Zn%_KsDTsMq@z!LN3l&qGSX-`~Cj zHOg`f+c(j@ej`<3)qBDoGaDiax5ejW4@jn9n@^W?I>)$V{{Vxr3)CH ztw1ln>D!HXx~G|mB`sTlz?d>We%&mN>p2l4_F{_+n)I=*`+A`j(@v)Fd+Agc5^|DK zQ?|*2+`aDHTeg8$C2vm}xIB3cY^Q&0v9!+n%F2R8y@P{a!WPIqaUU!vEVFJ;JVL(` zBN?xI9E~z8nR`pk)okSzPV;6oh8TJ)c2VL$>*Wr@CFJUTsCxJgS&t`)RKB`=5wRtc z42MMvAY`bptt)<%;`X6rq7Y4HOO2#s%6|g}ZkfAc*bwqz7r8$O--;jA2zMB5256{_ z1u4YNT!XP1n@T>2&@{F`u2VkEwv1Wd9*+A)e4X%h!);OcN^anv^Of@DfpKP3;M_Bo zJ%5h;B)bWlI1LUvs}dj)Tr-$CXLI}O5v%jNmW|U$S7H!%c=W=!mgRR@WQScd4u^)L>Amn5Hh7V3 z@l?6RosH~(k!XYkv*A{m% zLRj|gdU#kd<`Y07kuWBk_tPZD)^^n+ zyQ9KKb&{u{`2v1vqUgwV2h>7&F#zObt*RkQn?x0FwwQK>Xe;Fe@7E_-0O(j3p9*<= zQcS$Rm>MVETEAo)oM{A?3*soOuChw!CLhy zN$g%$U?|H1$B;BzL3`ulPZyF=vT9Yn$Qxu50NBQq(5eh9S^2EtFXU687_&04G zEu_-lz9&)7yVU-+vrH){@H9D=Ft&`shN3=>H*=ya_&&(CYKtZJ8@IjC7&Mh(Wh_BZp<*HOX+REMhSikdH^5w+RF$}AX5|iyvI)WjBiwd8n zkBVYj4}&OgIVxx0H|$^;&RF5Qb)Qre_ki~S*QU@1XkJAn5~-<*(q4pzOq#D z;=8C6N~o(Sid|TiLs@#p1hfwWh>)*i5v17NzeU}k#)V3l2{-ee=GtmXPaK?YE!NmR zOR4e*-vA4@V5wK4V0RR6-OC6Tv@e*{K7kT>M8dnszrOK*X1b1_l*Wen2!c-^$WXx8 z_HcNYYI$_OnSDLmsb zNmQvn>LXW9qI8)UbL=L=Ta5IC@PT-GDzWCgzskXXGYCJ=5DG%v^XTKrIEI2U*^Jcb zz9Id%eVtcYvl^J7c+bP7v?CvxGsTH_Vw?BhFgBp@%;NKw%E>?#KPGu0d*N{E{d6ybGg8P{-~hml*NOT^pq&hw>WT2YnXo6EZB z9cE@sg7SD8yFgQ|{2lSV0Mny6P@2gK3z3r_!Kkqfb-0*?3qp{24}h$YC*o~LSQmWD zm+tcA#E`|SI&#CQ4FJ$%Uw~V z2sJ4_bNL|f+?zBp@9Mo^4+=Sp0Yjue&f zvTpYsz(}6g-qZ`fnT2_eWZA3cVAp-klVp#}TBM+TiR#-D7A-+!6`yQYwD^i^q5>4M zVlxkq%qJcq{HT5IfF&^EaR`#g#1VY5$rAhCEn`I(SdCx=og%UoS!JLxm+#hk$%C_= z4<*lEj1Kh#n@?7zw3i9)*$$#?pAGPxR=5O1tISz?nWPsyHVW@J$!RmV{GZr68=NWZ zUzo{tu6XNCn73)KyUbPZn|E8R_r@k}TuhdY$J#6PF-tMz)nrv8f0rJP4W zmW&hL?z+s;Y&O=t{#kP|}q5+lhbm&UE@NtDXr0q6f z|2sU_IXV!}%?)4{aBVuAM@xVwdNjh!$?(3r=JlTn?K)n_U-kcQ|jziD}cT|4U_y+6UuOd&m$SKGYm0iS3f9V&KY+Jvh?~ z>b&%x&3Sc5`DcWvPL>WdWw;c-jjnRfQPOyL;f-Neg}qcS7w4u-KJ`T0Vnm}(#$%tZ zE|JgR$PY~ng0z8E<5)&FB=<)Gzk5uwm$zq+VO}gmA4IRsd&?lB4MDUfmtFf~ZmoGZ z``n&=@T}TFKW4@4m#q>we%&tImSWEVYVS-(BAz zq@3dx5}g&4{T)mtwh&F}+TN-|ne+l`vNXGM zR=zrCanfuw+VzC)y`QaPtgU7o>BY3I%^itOXxRwwQ1aghsPw&?kuRGRNV_pgRHso8oNi);J@Y zYOM-9GwUz!LM7)gRWuy~ab~!i^_$VrhcFQ#4q7NA^ip{7inA~g@|>2_RadBqJaZN3 z5O36FEv=-ek!{yJ-angbIW{kA!zd2+^b(i0=N|dPgiZR)B+OHG419x4F|*_>p#G(e zs+x#>Xx4cPvW)PKYy$iF?#StHllUz@fLnMioe>(%`pZzTog`OFFqe-oquZ>)dTZPb zvcZj>sNF;^kgMcsfY0?6euxF2h*jAm+n=VRicv0oGW~hp*sv{C>7LAyurQwLKs`)7 z7nnBi&ZIo(aD`SLK5#94O)kOKAV5Lzpi|ndB~tktH7<8d=uu0{cXF%Mvy0EUt58Sb z_E#T}Vp*p~-K;J30mmWmdS}YRR!u`AWAGE%ZT-Gq0xQU?g~m;(>g$f`ZIS zyf5%(IH{O`#9;ni*F#@tAjU2oRH*JvlLruPV1o01dLsW%$wnzDTn8|s1sIlk2b8O}T(Jn9i08Ek8xMx2h5A^zx3He3Naf9*^C?v)} zX`}76ie6P!Y$&UGrR4F?3f+b(K!ZFe`L`2UX)2Y3P5!Z_6wNZT1)M!qJ0Rvyh&!^c z(aG(6do?_?yc>Q&*-w@k)}j)XF7DV4`3z@{#b$|{3iOVDb)6(TX%_>v{IEhTJZ@fz zN%p&|b|3Ugd)E|OfbaInNq4OYk6G3yKsAk=V*avfes1#9{zK33JU$e6CxB#gA z9#K70-|#?XlXuC@_u(POO3fe$HcB|?J8_s1=&LJNZ|Av;zhUfH43MpmfbS<+vbY&IWy<>rOgI+EAw$-Vwm6*t2S- z4%lT7&g{XnUETeXJUoJWRsH(~0TL5@>!{U4?Pwb7t1<&o8Fs0lmCW}Uh!T4qSa(9k zD~`B_-LcjbKSM)Vul@7&i1Y1iXyPl8#1SXTGkd!A-aeZS?He*X;a&>c0k=87Mey!6 zqC7QL^27BOAjNXInUjpoK5RTb63q8h)#S`gUm4&h%5IMX3(waYz_$z^FVbjei$1w0 z07>AfA{+%fJvn=$XMGZS?75WFR+);GF@M*RJWAASy?HJxRHpt_EPFsv0uW{P^`e|y zY>T%i8zg3+;l-r(y=7tFR7n=i(AGc5n{+;v#cfy@1+;!1%ny zWZD6bt@4g?n$5RN;3WlhEvd(3KBP4F`h*;9z~&G(6rb1rt+V}DU zBk;azh(loPW3BDTMxVm=OvQ_scyEdIz2#Oy~x{u49I%+<&@0ef1gPzNnLl^2RyA(Jhd<68kgx4v(AQpd&Va z@}E|JpFBq=4gE+s#Bea3u-&~!xUIZ+5Yp6V?(TJC@-H!#gW36qme}j(rvEeY&4dU6 zv^k<13RH%_ja3(l4`D7VTQUH?DreP=*Z>GHQAC}kB?iqa904k{g_s35&55IP6~QbINKfXE^O zLM(J?(mRAsC_$x3uc3z~HH1hDgizjacURcGd;fbsz8@~14#_!_ncvL(W}fFf&pFib zDav%qFl6r-ne_}cg-$xBO_QvzpR9XB{Kl|h6KQM#wEZmc88RnARkZ0@*xb%7qiZKD z1il6snIYWA7VBd4HdT_Qy|1LGFV^j0-f4CN?u`q*`%>(6TQ5KXB8YZZR4@a~Dt-Zt z9rkSR-@#+jK)w~JE(7y)f6!VR&25UQ@t%>5RiuQf1I8;dnd;xvJ4dt4;T zIh1ySU5W-2ny8ostg%BhYq9&(EVEqjU3_f{1G^X%$5eCdoz>JeDIX8zHMVjTy1;8DN#v; zTt|vY@scvCogCXe4rBPlKLLg74Yvh?YvZ@D<^zj@F*LM>9E9yHm!`NipW>uA?lf1N zpNSGmFM`$H-;Z;}m3CONUrmCqa>e2cBJ^}`Ig|d{d%?6CgDmKtR~uqxI4jbK4g6eX zXvh>JQKUAAW@^PYb72s8gGvkHwf-p~Lx;$nlz{a&YJjgDl0_j!YkkP_KKr&qruAF0 zt@+X*m9XRIIt1ZC;zgzUwF2z?F+qh9!3OuAG4cm4K1&RFVaN{ovYzDE*rhO$Z% z*FUM=c$G1<_@c(=rmUhe%uMrsDG#nca<_+A?bg`%2rG z8xZCaYH}Ytg!oD%3y-Yv__=oZ@=m4T$MUW?-kU8o+hoLMx4qTe{vL5&N;PYvCLx}a zZznm5D%K`krEuf2R?-I}J>ItB^OfDm3=W+;A_48VU*dPI(){Xi+Mg6lxl0t$dHQ;a zMhL>a4#Jk8+t!ZAIz1^5M~pU$iB^@)c)IM#*K_aV;xdBBnH0b1v*+gK z65lR@rvZ9c9+lZ#P*G8dPC42S_?`}VsD|4ZKJR9@EvT=v(qo2?F1MRhMS9H@6JN?N zwjz+p2I20gm?pG(>HrQuD6Y&*d>}fMulHfx3B^fC#~emO;zmV;S zG!^nD3n+`c4v*$q!#I~(@OqQ!3Zm$D_2t{P`@WC5JqoUNS@xTc$46KUjA8OQnTptM zHEG>TL`)09xYkqc&XJTL6|%*;I(5;?wLV~MY;`v-FYoI!nAB2ayhw*S&R_4^*HK!* zw^l=ww2^OK*X~X5>*>8-DUOp4$mBPw`AjoaYs{6K>Gawq&wa<8SLqh8;=Fma(S3Fa zjmVEufC7;?q)$>9JJJv1KIig^#~+WJ9h(QW)NwA1Oq*~&oA&y0iGH8W@}IUOz4nThT<|= z)np{x(atCvsLJaTFLT1m={-6|C=a)&^mmdQn<1G~wB=mF6&e~QVSyH%X+%p37 zsgpVr*4^H&wE@FTTpDKgA5QzdRNfC^a^>!42dAdhhYh>Ra@-QF%Wl&Z2WSn4RZtO& z@EOgMEsF!ExF~hsQdT|WbLsMvz&XsrTF1V-U7JcR((SxtC!g`g&(6os#iZbw2|QBD z%T`Xh5!)dde!;=y1KI2TcbLcHR%c{hIsh;|l!UF>_hbS;`PBQR?Ljx_^s0AfF zn{cV$(zvXJ*J$TKDvR30L1tK+YjHVo5G682V&JZOAu!W>^v0ssG)_))>g;Ma)P+3d zj(ebYtLvhe1C;#LoP9+mw;GwM001SlC=heJCwu2=RH*wVVPCEL6mEKKa)j#3;W^ip zx`MOD)K|ZkXcW_;kdgQuh(lGJ)%2(|W9K{8$4}8f9zIKG2PaRvxMdt1g zJpe|wFy_re=|{PbI0ujb^Mg#ov;#5|h91d8gE#j={TCZ!pz(g@e{wPbE;A zgm|Jymav~xX1n0-){hSlZ!i{h;{$fTI+Hk>Hk%)9DA|pbhdzD!v<)S9RN*)oBcnZN zk6D%6{BC~a)A|(r<1>Cf*_SD#pT;%?ji$;@FJL}DtdiPEoto|#FFCl%{pW@H@+H&~ zxH(G|CTSy0L30`!jj$XB&I>HpZbqc|X!158$iu{-mws{>XUdtxL7$01O}K%%mDJhA zUdAyoA(z%^|Cj_1{e>LvD{|b$F`7F*hQXRHL_v*2pa>?9ltKNf`vFblNnwb~x5ko) zH}ff-fV-BgPOjo0$7pl6GFCwu<}oq7oxY^fhXHGhBhrvv`o~{9o(X4A%4)?~z62J%k}KZ%3BDX7k#YW_h=YpaTwi6tZa| zSqXb;mf-aZ)jD$9@QU!n{>`V*Fw3e{^;>e(ZX-!cc{#?D4m~n)N+qFR5IN%ooXKAn zwHkKmC?V^OsmK~jY6!X7a`*TUE&V`UxAvlhdiopI2}(A$7Z!Acy)%X&skMoQU^R+p zKE1b*yqcc$)WjD8b_H>}n>8&-Tjs2qAIQ{&@~`XUUz8ZV9>LYuOPFe)``+xmmn0QE z+lt$NQGI2Z_}cu|SS?{qt6`R4{`k%f6~cz`39-fg%#MxWfPh&Kc*WeePoCd3!IS#U zdejCpCanjkaLMqDNMsc|ldLDZv9`1VhMe#y83zRo{cBiGg>N#j%2?74gdO8f2kW6^ z_v)*WYen<=sc75oWbv13udUaw|AkH7Xq$iJt%xA1Chf=esKw#S^kR`2yI7mv8@sGhO&p9Pqn z>O{&+tA>L=Q@gzJ*G>wBKME9W@^pTa?3i2+vL6!Z#&SKLp{tLjl}m| z;C&wQwP^d7&z&UHvtMPd5_bu%Xoe9fB{L}p%~#WHH%@!wenl3`_1=?~5Wgi&WaED3 zO0}mk#=37s2 zyVcRVb-5CcQ$j*|Ohn|8`!O=!$bq%%Xb1D`yurK`vpSQvU)Wa&&kKXk52YO3@9*!< z;aDl|JfMBq-ys=^Ro*PFIutB4S^&%WeSdd8LY~fYB-Y%2lFWW?j`7>!!OH96``To= z=HPiw+)ltl`V2!-)$u$%HQdhl(c5?FW*=zz0P(D^Rpw?_+GoCoR4|Wr?}JbXozJVz z6uOl%>j@Qd-_BCW+Ds9)=^j;iF7~-17V8^XRbm9Zr{*=xZrrjUCix|h_=)%va_W-N zPN$EPTZxI4HC5s&&WF|6RoTljJBemxoHjh8%BuZmfC=(!Ja4IqgK5afi8@-O&&`nv zT@`|1oI5Ru%*rEdME3T(b#C<3b!_76FmG-y;bomEX0kL(eJQ?W(eCYFQYd%KJaD2= zi_c%;BsW(~026~i4_sLEC~2~^5ViDjQ%_6mQERL$6tX1`=#qqQ=B-L(?Z^b32~6f< z;~w9PthsiDP5#9hB4QFUYBp_>KxX49h&7cAwT9;;L+%Gho;MBKS1kL7#1#${0Sd(p z!6*@Z3U|G;zn{+(c%^ilB3s^o!aw7d4h2<(ZKChq`Y3C=!(1O%N*Bmot!khfLUU7c zXKRjPEciAl+ko=tN>{_h&xpwEvbTB+%MdhhdDZ7RVu+8_df0}&7+_{h!_&FG{_g%u zo*&UD&3MvIWO!1*EsenMc3e#Hl-mCL`dh4150ws+h{KX@m%e#IcTV;;y_^4?TITJv zRSy{M*29$&2}w#-Qx^oeVPheN#d8O7zSZ{8L5_@ISvFTCW+~A-18weeSpKXNc1Cuq zEZdC;;-@Tja0%N}s#&po7wBy~6C*k=omJu2*@LIsolsF~&Ee0%?iL>kCzMj#Q#dau zS<$)4;LEKzK-#Nq61pK+gva*?r_`^E`~_QWzUXKBQd*vnuvWeb`uK9BnT5v!Cd^&SE%p>60IZR)Ft5no>6J?n<)D8HUKx#b zy9b#!Ji{09{g#ZckTP8SeIG%h-Gyr-ax6HAtvfS#pftQxb4(zpbQ$ViI&6|B2Z}U} z1_f(8neby12YJeF4qBBKAwq1463VxV-XW!JP5t!W^hkANy)aq14wI_2;E32VS%JmD zo@859X7?p9JXV?vzz!8eKL3>YwUrpnaV+_?INn3KPdm`zGYzjBIoe=sM&)XQe85IxW? zPzi>9fXG~b6x+HbEl%q3_`)*io!}#tlXs|FBtEa-+mx6L9wAn&`-|+sLJu0jmR_=})avZQQ%K8}$MlfBw5b?nczuT#!bkiuWLFX zdW79=4X6`f@!gMw`fN+r6qozudA4i`HED>O#!@n1>=;cY7ZX}FI?TOaZkIB zr|LTQJC&FO(h#du32M#>m9go`T0C0H3#*ul?BokpL;BkSu@!>!MjH3p?-?58_0Hi9 zhrbTiyTC)O=HQccJ4MpL1)Ra|kK(o(`P<&!_r$%si|o8_kvnHk;gUJt?J*1$pq$Ps z^57vEPO$n&91?IilvK5`KMSI7jXL0pY8rfeMiBHsEvfgBcqNVe9SXj;hZE8FV==aq zuOIL9RQ@_(Y-~s|TjYmNnwU+0Mb!n0azw*$^N*5vylQTuUd!B|bhyZPr$mqF*Zn=y z^#g1{y2VPV*F!pmz1a$v}rxZ!OU*>a_zZ*_ixA zhk)Dp^G<5mMTesmjC(~Un*+u5?02$S8Ic&~3l$4rp*VC_&qEy0e-l-Uk&op|B|9nu zj$;jQ2eij3ldI~__(EGRBv2dDIw(_RcJdVoCw`>Xpn z-03lXcjc#&A{}f86QuM7zsW)}cZy+`9$2bl8w`^`Gpl(}&RX}Tjn%k}?h=mDot-@> z2rVl&(5$@dsY$3@YDcKb)tw)&CW3i|OD`Va877xS;Wl$I67#Gq;ZzAIjDx*mfNY0o zD}2Js@0S9BiRSx+7vM%*U=(Zm{ta(S?^Jmx?Crz6zYZy0cv*=@4_{4Jsyaxk>ptU; zM-M^$a2q4lu8&1T)Wcw~?hACD4CPQQQ2xL>mzQ8#!q&gq@E+H{KMLVUY-3h4W%_dRxYf6_Q=w?q3A1PBD4Gpi z*eWCtl$}I-!F{2nH#fMJ)R}$JV*K|@0|(i8gLC@`%IsbDNkrAG{WA@eR<`1a-v%>@iq?-m_g^cD z(Y4Km+YAUSnD0G!ygv)*S(MVHpEc=bE{4tA7Q?T!DHAn=Bcg(?CI?>~02nLn)ej&+ zoA|NE5peL6#Yr1Y8Gn2KzV}DTf(t4$Lm?~&Lj>a7GB~`R;`dBU;|8nS?2icrHL5a-+x>ng%8`q<09^~^?CvTby#jY9B{yw1nr>OZnj%a4)%#p5!LYD93=ms z{HgMJpaIQtnI4So$)kg%34l`e=DM_zW2e+mktzIk)#%Juqa9q0bp1le7U~K;v!v`= zfQPHnGVJNWwGas3Gi5Q2v_&ge(6!&^h9EscR^=N_jd^nY@r zlw6dp_a*=4v9x3mFTX=a9fd}=YHgqLRa9JK0)28s7H-pWp5ut2%MkF<5|X#7uf08B z+mc}@t_UY*uAP4dU9>#EHBPeWnH_+68y9U4 zX(UaIC>ixqwLA=s*P~Cmq!v@y4$2o)(5{@*^S}ty?b~~}0ukSYK{+yOIS*>Po<54Z z#cBcGeP{Y)bf3iNmCYt36N3X8G5TnXcSDj{EcOX)4FqJEJ%Ej*Q9WAi3#stj)-=V0 zPVE;x7-vib&B&Ik%Z5u9)oKQJ%NFGZIl>-%PKE`@VZXcATrA6d;8c^SRV(6o_eDk|tb7&q5Az6E^OHw6eq+{u1$8EW zZ{k7ga5Uu5nC|UC0CXXEQ8jH)mg8Yd6JJqyXlz&5%4gGY*TNfTXbA7eBwPr4C*RSl z3ha+C5o2J@F3Ar&g>`pws4k|Iw1xIaq5 z$UIC%Z^(}Xy)^tN{u?V{qnLDp8 zB0T{crqP67d`MLk!6-X^wMqt>nh~B1c6{*Ws=?!aQQo3DlD#aa(#a)vfcNsp3xWOA z)KLNvbEy>n>W_cp2iT`Ztj0%60y9Kr)jC5o)aas{emSQu%gtaj#bnh}dy6;UH&#I- z>~}iB3cMx0!9(+&^0i;DBGAoh28JsCBsbxD6p~`xZyUQM znI#%HD$^x8br1VV$@uyEuF81q(0~$9BZdDvVBF~yO_V@%-#hON$-$>|^LLnrO_fTi zTS1#KJ9D{H4RQFfifmIk@WPLv*@DNA(QWx_htCgq&Nn?ShBFHZngiaKNk89ejPvgN z^?iYc&ODEH^nq-_@?+EF55j!JBt`<-?@TL3CRQ>9l>Zn>{0pgdUK&AY4mlLI}nxqKdWg!dE^c-o?h7+12|nxg(`^en?l`aG!|F0vG-3AIa-6EJvh zTwyIC1}}VmNlOcJ<73s2q%(n)SZYG$pcGWgef+T zBdaN!3JC`?qO;VF{Ff7|_lvSQYGu<_R`ORW+tIfYb8wo(r<@ zI0MT#v|+?AOEY~PbswJCE+0x>a5RTWw>zve!2Y00_Z~;20b-?EB60hqeQ!yho@eoL z=|k!KKI-aq4II<~dIr9WbjE!!R{lhrG$y+|d+E1`O9c_vXzoG|rG`x_K@O|nT)$Xf z_R)H2?-;_>g|uF~IL`Ix5Hox9tZC^4peNySHk0l!C%*Iwu}W%y24>0Blr)HNCObFk z#?SIbIB1k5LHf)>pTSTFFI2Ais@1n5Z6QLP>}Z7 zsB0m$bhoDmUsEOU!*7}|c@v{AbWyZo>)e~)kW()i1>%!UsE-z*fLf1i-;r?vE;K#8 zdg0X9GM1aYIO2mnU8NLcy;Q*0O63>wRWt-Lj|tJ}J4NO+Mi#Az9jnoxqeRNK27I<0 zjgM#HnE%vXXx}YVOHPWbx9&E2JP*XLaDw&ks@Mm{&>!sc)ROx6t2<3xd+XMxY1ZUP zz>vl)mF7$7{nB8$ZtM>oBKA>;&n`U?@u&2H4}TIOKw4R3dL_KU-IIS0A#T}WoNpwn zSDq5WOFd8F=$FnzhXfn*<7Paq z@C#jv^vZ!FFgshNxr!J9;R??yKqE|V-`TmC z$6T$1$Q>(B4kj-Lr1IixRgV%@a`O6JGU{I;IJCP7!8`Bwyer~27>_Lq1XI*)rt~Nb z?0c^-^aNb)BJ6W@$K4mw9;$L4?5XBwu$iO*qqyjNJS(;Sgy2uzWR#}~OXeb^lTEZs z)=G23S68hVEf;QI@p+qkWcTxGg3BJJdgBR= zYFUWYXVa>du_+lqTvEAfKKLP`IfJhlLM~h!x^gdQyt=ZP{4@$D+Q_3w^b01Lb`I_D z?N8Snst+Tscp=rA1+6GWve`qmLwcynN|dFy6ef;&uw6 zQL!1rIC-k$hUfG7slgL7w`c-)kS3QXkCqvsyS_V3h;1|;0!{*o&X#2>JfM5a?7IIg zsorj!NpbLzHZy|?Cg=lS)1t&$6Euvi1`xm+^q|eBrY9BkR)eVt zaT0arOtp1oko19YVyM6!!n1 zxxCh8g4a}(#!XJV7WJtS?=TfAe#|Z~%tz=vT95PgS9T@8o5#iQIKLIGtyKmy$WFBx zezcNMg^nEnJian)`2 zono;P(ccH?A8@cy6DkD%DF>t)lhYY^0%7!3 z`QJaOVL#>k^jUS@UaS^EvCWHWY3&Ixo#VP=^`+Ax&1X9G zpB>0XAoF{_llf2EIB%2y=Uy4?AK)!#>OGqltkqQ0N(0Y>d4Fro;#p|x^VHIADxONw zzbbtERpS~diM1zYR_Bz~Vcb8E?em5b*lw@6Ltm@$T2#2fw<1%B#LVEcL}Bq)te*Wv z=MVm^Fr9+*ID|O`2z1vUqU8EhWfGs5Mi{{C&G85Z2S1QVr~(S?edqJt35712#-Z4x zEsoBH|6`uWrveG_4BRXg`5Dec#7P3Z6HNL3xM%!r#HINGn@??>>rKZ|GG5Qrs~z># z>o&%gg#Dh@RzBf8W#`|79W{5ui&V#TNFc?@ms`q&sp86}e%1gvLhOl()g5NPzqKU( ziT!35jAtdg=S|+&*ni=~UVnlUvEzoncNE8H1$vOx!EHeNdsx%;ma2u;m?7s1E6OBz zdoKqh^p&K0`r4pydRl4erl&>t35qA0o${?V4MBQZ$vDqXPzclBC7+fn47_^K&@H8@ za5SbK@chMbx`111`fd^CrTj()(}vT}`~xCOegw(g;uVUg&NCvb+Z;D zl2%jGEZmHP0;c?R0{UogFDEdjW2)DidA^vP5W^{DPXaMjjfBLy_SaF(6Qj2H0~gTa zmd$WE8p?&9aF9^F_T>-7*Fp78cF1SGe^yf)fL-h-q1Z?s%vyt z1V7SVDXtUS`-M^GvMA5*Lzchf^-mcjVM2rq-?vI7J|*|QMQaS9u9!E_6K<=}F(#Pt zg(l!N#)OVrsX4)Fkvw~Q)qiFs`DHR)5!>|V?K z+dhfKm>lWtH?PCO{)+_E_8}&`-T`@A=h9D^f$wQ@Qe?|=2-EEU$_wwlTZ@FeSh~b% zm%)8v)rrY};79m`TY!j}@J3CqoL46f`$i$+JVggddG6(C)=;kIusT#6@CLry9`}6v zf{t`$gW3+z_HTuqVBA|z`uwj`KQQ&wv^;=d!Cw26f2JDNP};{)y>4YM&gGk4pF?A# z;AnQzNhqF(I!Ut^^4{c2+{R|D1CA#a(CAu>zw(@0Y-MIz_Wjh78blRN;2f*YQS;X{ z-Go`8zZG<1t1s#PP))rX;-vlRJxl7d%}PC2cCnbyuYlH8`FJhAs|=*Aj~QEz^tF1f zRRT{n-#?vj`oO2)MP#WAw)dmN&rbufBI{X8m zX9Dk2khRl28m>Iy%qVE%fN)RfP|w|=ylWSw_o|=8$%wCL;!G4274s2L>OY(K?+#na zkuQ#PYTP>&b6e_x1TXI}L>1<&&{}m%3H7@L$!;$vhdxD|I_vLr)z&2he0aO_;>me1 zsJy%T{>E0i)q?>C)_8^D!?TwAoPg$x{36KM?K#Ys2vt%SU0>iges8lSfF{}2)AMu0$| zHQiqG1lj*1?Ed)buM-_VQ}aI^aH|n2dFkDPn~EnxjQ@1*iA8MOD5jJD{tJ`-_cPRN o{{!uRk@mm5`Tv$|ejE|)?q0|8by?A#A^fQ*YAO`nw|M#g0B0p+!vFvP diff --git a/site/notebooks/images/insert-test-driven-block-custom-class-imbalance.jpg b/site/notebooks/images/insert-test-driven-block-custom-class-imbalance.jpg deleted file mode 100644 index 6a8fe31e874d3bb1bbf04975fe64f9d5a0bb2123..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 188649 zcmeFZ1z43$*C;%DZ#qRn8U!Sz8%dEC>5%U3R-_aW2@#MGrBh;q(j`bEN;gOdD5-SE zx$)8GdEU6b>pkcH{{KJMInKUj&)RdZiM3|U%vv++-e;p{vtUui$JPn}Wo5t$0Kfof zu!{fzK``h8VB`Sx2MmBVjN(t&2*&&q2Mz$7@GtyFwgBZPK2+zA@bu^JKT-@RL5Ffs zxOjNDqhWu*FgOClg#3UnLeD;={lekq;!1=4ReKuzSM8W)X^3Ai8fg~VPaLT5d63eO zR#1Ra4Rbe3ODA_5XAc0*KI8W)NJ|^5s;kH<+>`+@h&ud6Qx_Kp_yqtQojlys<*w7} z-qxc1QwR=<`9=v zA#5`XcXLMwE`yl!(%i)j>NDDTKE1bx%Q*~1K;#B>5Q0O`Ve3EOm*=qAA8^SJ98Gm; z2&V-aOdL~FH){a!dZBz;FLP_C4!m^;=5n&Ib%N-pg(Q>qlUOzHw~aNAsrZ{o#1x*Mhh2T(2cNvYJIB9Oq>Ob3pS6I18Oa096 zFm;2VKiZn|wAVV%Kd*n&#!}&D{Rq6fHkvw6J{AB71}|IfTM&!~!5r@P8t46i`hXC< zYa@N0{{zR(Q~kXE3{bwhrIXxs2!?1ucv-p0p3{DgALi%|K|g3mq}kXioR@|8fOz9! zqbU!;5PuLY=B7XT57C0?wRDpD$rFT^rMvEV-ynQ=g@xq}83=}GM=V-8X`R!1UeCUZ z$My5Jpt2}fE)GBD29%FNXXzk!o=*qCeC}QvKg*(Md$?(y=R^HOv2yWHJ?A0RPn5fM zri!W%4D|=)DYyxw!3{tQJfUa?oPiy%?Ov^1g?|08C=H((7c0sGIKUnO*YJaK}) zHNai42Am)q_dm;B|MAokctQCmf9C#CmK|9Ac@-2So@)_!k|2;iG;*?L+;HI)pldI%aHp-|e4xqzq;tivHlCCB)yq z@dAYrs*@k(I*J%nJ5)0Y2?{x2N8y9okwTGxTIPpxrGHX*&Xr#y@^_m4qyYRi8b9$c zxiI0FmoOzUng6*Y9xERIFM5CQ`tO|lqpxOv^7~g0{gX(AVLwP zh<3y~L?@ySQ3+@f4Tvhl2Soij{Ih-8AERsiXKU&|`eP4`i|wB{zv%rn;^*?=&Q-v* z#C7~fEgqIW9*{f(X=j%^ZnoAo9<)->j$lcv;AGBmm6nH#hYx`B{pB17fGxEj`v#2U z!(Vw1rU4LZI6FJ5{VPwg0f5(n(DbkVE00+S00L|PI;zb*-Ms#&59K^Tg;oiCXr_?? zYCsQ|A$oX#0JsJufecUtDnJw521by)+CXyd2E4&N5ClR%ICu);KoUp=nII1ofj6KM z)PZ-P4RnECFa#zb=~)0PU>)p&BNz;Z4#R~J!pLAWuq!Y&7!OPcCJvK<-Gr&bZo^Dq z)-WfSC+r^V0W2IA151LX!(PE&!>VELV4bi&*cfaUwglUR9l{ZC95@l23eE`UgbTr? z;7V{!xFOsc?h3yP4~9R1C&Dw}h42daJ9rm-2tET}f$tyy0vkbsxQt*&2qB~qDhNG< z6~YbSj|fA=BQg+3L=7~~1Be;KD&hbI4TTVe7KI&>V0jcR6mt|;6n~Tmlq8fqlnRs< zlmV1kly#I-R9sX_R5nx*R0ULBRBKdk)DYBo)Ev}u)K+LNEuikAp`nqWF`)@UbMZEs z9hxs%1X?m$FNsMLi7goe)I+OeGF_28Vnu` zSqwc4Cye_T@fi6SZ!!8YzG56;$Oj+!Z*Ud ziyx0)ir<63dI9qS;|1vprWgD#ytq((VdTON0U-f5feL{mK?FfQK_|iDMf8h|7iBJ5 zTnxULd9nH893g^`o=}?5obUl*7GW#l0udS!6OjUu9Z@(@5m7hMI`KtfK4NWRAL19p z^~5tI2ogpT1ri66D3UiM!z727s4huevb^;867o|2r9Dy#Qb|%v(lF9u(jn4AGFmbj zGJCRUvP!Z~fT;DOoAiDSat3DL+u|P*GFK zQMph(r)s8Jr6!@4q_(4eM*Wt0iH49yoW_>s8BGJtGA%K!6s-eoB5e!p`en+?@|Qg? zr(f>Ad_>1Yr$u+4u9)r5y1{wxshIL^_F2(M}Ucf%fLBes9<32|v$2U#}PCd?O&UVgIE`BZt zt}Lzz?u*=t-1oU_xHoxNdCYlU@C@?e^2+iC@K*6|@Uih(@}=;N@)Pna^FQKm;y)1( z6mSzj3M>gS2$~4K6dVyE5>ge45b6*{6}~PUAY3PWAR-{*A@WA#`!)7!_Sf>SEs8RU zT8d_g&WX{9nTn;0&4|;An~0~0&q!RBFqO!V_$)~;X(5>-`BjQV%1)|K>f3d$>u%S} zukT5VNc%}Q-ay=tyAgJyTjrvSmQ0e&lq{XBwQPaxx*WfpuUw-%s=ShXwETzym4dm# zD}{AMK}A2s79|`d4W%Td&o^0by4|c%hAS&7$0|?Wx^m0uR;3D1QBa9h`J~FE>Z)3+ zhNh;f_CoE8I=A{g^>z(H4P%Xbja|*_no*h)TFhFWT20#c+6LOMwD)vw=)~yE=yK}% z>2}>FzioTFN)KI6TQ5g%M_)!iPJh9G-yp zl-zRMVeW?RRUSl;BLBgY+4G_2te3dg3$J}|ZSPVa0v{)z4|iDaJihbwuI$}xUld<+ z-==$)?*-kP^^@{T_lNnL`8Ng72LuN!1j+^G1z`o*2YtNHbwB3*_5+;*vkK~R_j68c{^`tvW zAS&f4+Eb^e!_gAac`*brcVp&am18TO(LM`%wjF02*BQ?jpPGP`;F&O!cr&r;Im7d( z&ySOAl7?Pfe^LCB^5x@~d&w5b{V9?u$W*G-@YKUJ+qBVix%7$*#*Fw(v`nweFIn1I z?b$-vuX4z9!gEe?opNXL)bm@B)cRE1V;kdM<3|&Llc~r;j_l) zi8=GRm3jC1{o{Zu4z7?Wpcd?K!jF;6H4{ zAI}5;KtlSv+A5^oi~vxQ1pp&dKfVC~s<;4X3IjMV@1OB!b${5E=ReRO8zN2h#LUs` z=c9i}{;)ZrtY7{9=Mks5t-#;sRiv2LL&K0FHzqa~LY80QIL3vJUajZCE>q4#@tlbOwOm6M#%_=-E90 z_JRPg42JA`$R-vD2cYaJWa7jEkd+DmFA{+C5=dt&2VkQT0Lfa2Kajqb^A0*uXa``j z6M*aw(9y{N04s9*r;fQl>(G|0n1suB#?8p42{KMZ&W!9d9)7?{X|fr=^^7_Nr_r%4zX`wj#0 zn=p{G2LqEwF!1dZ2GU`0uyYX(F7m;_8$mdr5QBqB12~{|f>KvFKz#rQg-_ssG8zsd zV&Q=OIUG2o!hui*9HeJKhc3Br5cCQTmI~m&>@^&`uY?1h8aT+TgM$kVaG=!&2ZNo^ z0Q^HZn*hN8|3@Ih$0uVTBO_zrVt@jd7}t3aQxN+R6n+NXAHn)Oxc?bo7YPZeD5!WS zD0qC-rPY1@r^DGN=$t6wtPv1kppjylAz*X>P5?s?z|LA=8Ia7QLJBPue)SwWib6p} zL&w0x!oC3DFa!dQf!`*Y7atdDYj6&D{F1NErBLuWX!?^n;@RjG+7f zOBDS{(4X|2jRPD63>q*50gwQmL~4?#|3$<`o*(; zD%LYVvwQ}$AhJPEM)4W2G(H185)gZ$RtwGmM0xRkFDe#BmVh&Dp>|iI_m37+M%}8vM^_Un846`A}^z1r&O1;6+51Omfa( z(j=0m$qo^t)_<(;zZ2hoHGDZIH_i0Nv9W6PV4*{nF47X?0}mSht>pZZAHsYt(EoW< zw!+GO5r_&z!ZL01#*9fwVlV~Fg;31-??UTxVXdusgy}5?!5h!q4&J)>(uWskFnm2=D8i3ve?@><81?!1y25?z{lo(UXEzaht8u z$SU#Y+pWfbEv(zgcLci3FT zs&_5F<@QHPt5yy;e%MO2R2xi{GMFkx-@Vc}v`VINr@7^>U~j-nwJfQVN*7PxU7b-^ zO6d$g8Ey(azM8le756-laUI774XenZPv-K&pVuhVzDn{eHXu7I`-h<5pAG)Qd=iL{ z`N#Q$hiE1I=$_Y{o#`lDxm5i`_)xpWpRhlktGIrG##iVBmHhPnc(nctGn5YQ2fabZ zk)|A_pK+NnzMD-Iv>61fHiwzfkUY<`8E>Mv5+jdEM(bL(gHGvccqDQ@OSQ-t)&Vj= zaf2!+<1;$-&Gw8GsYCM4jfZnlY%0=$&Az_7M6e>JFuu@f6R!YIBZ+ML+s5N|1PAd? zDDg7WX&$=m#*1GHnUoWKt+;=X*s$5I*jn!6QSV36rpAEx&JE9Bdk%d`lH^$b3|zhz zaQH=Ho8Bl`9>EmI)^qP~d3tX->V4Sh%Od7_n7oe)#sfK3K0DH(kA1n+zK14vaH#$8 zNv?U`eMg2#o&0V3b)=5+oApWUg56HD7;R9YpIg$NZ9d7L6I;;gnxja~fk*Tek#77x z9JY+JM4sSAtZ5P32(Lc4r%$pDF6ze_LdW=9$-2fYeBzf4g@(~cxzYk$K2 zuJ$c6`naIq20^boSX9WqdrO|D0JU+O=H%|@<7DAvj~US$uRBg4?()#vVxx@>&zNwDK)i9` zWXR_}B z7UgEOnNlchA_l5%7=5W57p&|rUz|{RrhvF0Bz&NMUjl!FpdHr7c_)p1R7Cb&+>lJiS2wK6yN9mbl4T1_YU{gq z^Qt!2<*7{P*J_4+`nY%>zeI{w1)<*G>TLWjN&ZZS`wU<@#hw9WiSMbW^|njn^*K2V z0u!%Ec&p5z8<6~`7me+_^xQ&f{Ia&yIwc-jafWW!mief?=Vbw0Z9c~Z2?J|qV3uz9 zO3Gp?^Su^d53OC~Bdzr>DZ{sRqYu<&vlFN`Mh6hQ?*jZ+7E_071V^?D(DuydEghrEJ%>J3rQ< zE6>}EPegRC&F;4)XY{-(Ie1AzBQxV){y4Ka*GT!w zmr|ac+1szHQqCZM+10gsa@mIRbhoj1T@VogI|W9ceb&-^AN$XnGg05K!gU+Cek7Qh z(@EZuctBiQ<*~G;H@xTUJ$qYEI$nMZ^U&jjTauiCTP<(Ak}%MEFD<%|C4Wz^U;cW2HA70E12S~KKpww@m*8aPTn-JMWZF54%3aCd z_#%(yhlak5{IG8?&vl3HDy6yqt?$JLpA#b!nL@r?=C4JUU6gUrdvqMxcLo|RZ4!OQ zQ{YG~=*pINE2{E!x^)Jg<}A}}3d)@x8+WLpr(MgEw|M%;I$ekc{$3&y3LP-yJ@b5a^d&7Q5z8mERI0hhueH zmo;xb1r&>5Oi^#Z5KOjOQ}zoA|4jw6Dz&}~srzGjjL5z(SzlYZO3BM>8?;V?$pBed4W z9W6;_yd`u78lkH{#cxPt+MB>h1A#%~tstWSS6AZVu$0f{!LprD`z$j?KKVyX+~X|) zm(M^eO@H7uWeKf{!Aiep+?vaZGk$D}X%!VA{lt}(s(G0}b2|{y-O&DX3QU+n&OXHqx*)IHk1_2|?waja+ z(Vlug+&Ay*(?oVlpK5vN zFSY6}i|s$CRYQMQt1uOm6eOu1{LM=Ro!9g8mD0g8Q2OE8wCk*gJ?u$HFW=5nu zXeMX>z$VH`PakogG~l78;kD-Y!?5u?Z!qH$5P6?QU08W%m1WV`tUcf~UdFvv;5@;2VBRL^o1q~;lzEBC#*UZdB|QEai5d`pswcjdbznLqpE^!kcgMC1#( zA-sShZ?COZPq@K#w<)`lG>?>=#U?%^Mp%;qU#*YE3gOz71iTI& z!m{lu&e_1Qly>~OoO1Ya{ztbY#zz4;hR>~@>xpH1MVx^UR8Nn8ScdJ?}+#j?A=O(H1Ca#+0i5eO}%P*gPN z3&cw?vjx43NO=&XUsM0>Nma>U_4Iu-e6l&=HO=i*oOW}rn(6=ntp%SM8soe5g*H!@ zJF+Sh=v<1{7G;3Bk&o;^m#rfs_l!0E{f1Ohq8^NC14YZyxNtgV%O0oELoMfhnVsZR z>m&68<4=>L&>HO;w(Jo}N5yzC)VucN%hEsz^)p!Kx+Zgm$bjpCjiQAp{0RrgWm0d9Yp+zSm!TE?+EmJ(lFgpn-q9@p4eA{g#kHMliECFGlcXaq zBPOE**=Tgu2law;mOmZ8QZW+OI3<35>7lcw2|sr7Vm;BtfK)}3UpiEJ#Tkfhfe1MR zhp?g(giNY}ZA5p`7u2@N-JWtel83I;wI}(8qV*rVwo-Jc6VvcA>b9C`HYuv0fjtA; zmyI47yp2vbupJMsjH%(NDS^4C$cv{80v}}08oBp4(Y;me?D;9Xv>^=v5!2iv0SAic3w)-f|dty}Tp2n`yWOM_hci4rw%A z?D-O(C<9n$k0o+!YZ7DFa!gq_;ZJ!Q6705@=lSumvIZz#V2Z{wT~oPvWS$LVd}mLv zRUY(I-ywEGN3mWfEx68xQFqP#B&qeYrFr~Ln;3ewHB<0+YZp)K!o2M+N?&VIWR9bl zL@CX!au`v%OplZ#sW|Q#h+=72$=xPO;ekLQiRsh0S7}S`w%XruJ`~=r#BI2#R&1+b zS0YDrHBYT{%a40JkbUtCY+{w40i@D#fnvVyM0wFmfo>`8G>s=Ou`1>9`KqdvREWvu zXSJnb7cpAzN_l8*kp`w&i@i38oDN~AADpj=mmyFX^_y)&cks-YYpQvE@~#Grpx3ft zoBCN}aaQHT@vRY8{HK=!_VEOwmQ)5M=AEKW_8XoeD;4_Lw-du|cwh7TBDQh-(Ar>8 zqo;@??+vqz#Wmk!XvDBU%Z6h4+al*XQ3-a@JJJIOI=(o41~*EtOGAq-D>5@ze_8ts zSXWUm%c=SmM2Qm((S7jRV_Qgt7FrxtR6i@3!F$p;0%t(rWAwZ)W$OE^4O)9;C+vQ4 zCPc}>Ha2|ZtMPZ=15*nFeAoqjpGO^P~>-F=+;d7iClTFQ+W!VZM6##+1L zqAgI9A;S5W!?1sMK=wQDUk?rcf8@#5xSpaO_@VbeXOf7s;s09Y`HviQ|E41S#{UnU zn*R^x8G1_pfQ#R&QD-|y&ba*Yntji&0z!5(an~ei&PIL<-0w;9-#barQ`P&`d*OXD zMCaQQ>>pjb|J989&Hn#h_W$3&&wm!NfAH4$En@#Sh`|5bH2$-Q{exS^zY{T&$!U9n zTN@Lj2hk5UE7p0$<5HcA?JC*nq)bR#yh^5o#Gkcz4r)A#-;{m2F^j~cx5Qf-+c9=;_7BpUy0yB$G$(~#zZrsRW-VGxxG-K9AN}}>+GcSP z(90+ouiMmb+hbYiko%DNy5PsQdcL*(ExN6|vK!fRYt3O$!oe>o{rd8*SrC;Banjaw zqyqc!1;?=Y&P990pkFuGe>ufys}8E{Htbrxi{kWOJxxrLJ?HjcOk;z5=F|jqwBu(H ze%ZTUh1yj%puzf^r`#`(xxdnUoso6hK1?KUS&3TEBy?1h2VeU&g2d7qbs~Nd33j!| zb*=-i;Dr-PUnklHxa)~uFGWids>!IoZg`*;+~b^_Ua`sDk%nFN!-GWG3L`s8l2;Nh zBmcHhbAS_v<5tN$PJ~DPtGxK3iJPNbufKmv&4s*EiiouRv-#KR-yUHE_d6kMw@;YK z(q;IuiBQ|{(B!g3{&evBv%749Dk-M{f@{5@UKNB?t+qekfh5-6+@!#^s!6?eiL7zw zb0E9g&B$T-VVT87g8hMWmy-WPej2p@=;!lq)+~d{M*(vy>d^Vm z*Mp>G`}p(946DW`P`I%i*5cRC3PD5x3>? zpGX#lGne}rO&XuncH$@Jg*UcS4yVMCq?+emlM;O4q+s?jN)q36Ns~~z(&I=w(vtz} zI5L;zz*Ul5yKgymRP<8hd^g@uG?yFtJssin(NFb5eM%K7tPVnm?=z`g{MBjGlb;HA z_bmA0zn9+KcHRmrn*F|iIH*(ga$`RUEQM#R%M+o=&dN(DzOU=uUO9N%=N9}H%tQ|? zsl4u6X0XEh?+~Mik6fJlVPAZec|TUTMOMqo zuqPv5@ht^-y1vF2zy1W!2ddD#FW4_Ql|nz+K}L(u9IB$>MZ3GY`sK;+#_MKbJ?M4s zkG)Y-W$RRS)XsSo|DD+2cAf|40j6E*&R~E*)*!|}d9B}#_H=Xz z?c)#qO5@~5J|s{lKiWKAN~rW6xf|49)<$s0{<&qFYf7NTJ@K4bq^CFKq&VT&rQCgg z;`pnk}}yV8qB=~35Z!z8aE~KkTG{MD;|D*wswOIdbiZRKw3QEB&XRf zKc>02*m=G1h3-y~oT3wn&s|)H*WZMheSztvdk? zD+;X_E4L)UlDnP$LfK#fBWE=;DV{`}rR^oIaix3%11`t6ToD80FVJwf%!O=}!< zJD{)RXV$%89%Edt&$aLZ_Vf(sO(A702l@OI$Gh#L%9>rgBB$@zBhzr|)ygH<=>~|L z+(X=>p8BzoZ{nIORtm1%wcgyQr;^23YSkM;I+LGHD3!%_Yz4713OBAwuznc!NOa^| z;^5%OdA^TDRn=3HhU=$&D3$i&$jsbIcT9e*DlDvRTDz8OdeY#)_Ms`(lY(zPD_WlT z2SggD)zZhqHTs^QbJVjd>qyDy&P74Ci^cEX210vW z{#1BgmoL%NPkH*`UZXPs68+^FNx~VoKhtL7JXkYmdmLVHqPR@6z6}O$BGvdNjNJ0}bqjinqvFXzgqciZeT(AYF z)JXTX+=sU|_we3J^ejo@jc|Uy`hD%!oNcOoQz)|Z$%l;W-rV@F ziuY|N+)IR+#zG4X(O4*?s)gELTChsY2%W}d&ORsDw$V~ysUleFJbeb@_jGVSvxUpL zFEB_xT*dEFwN2gRvXi)Rs)d&->P^jD>l4aKhaJ6Era5kX96=t{ClLpfUGd6&eKxXE zEhord^tjn20cnci@JNN ztA#|hmA>ORfny@-7BgJs;`k&*=#?X>F^2rS#92JhfqW?Xk+to1v9%xt&lo*ADXHFv z4NiDxl$1+(KnsCrjaZZ1AP7LEO@*ApEbHG^ziQz$`o5Y!$K4kG5^rotr|iW_PSuN> zgIv}RP`DPmR&Lig#Uyxf^$_ngqRL)xz>=KzH7mBQ@r$d-q;K`u>Mkg6ekxDttBP`Q za+w^N&i*8ob8NP9b{@e{nN}EeK`w7>4BPj){{VK$_`K<}C+*8@`jVC*r>F04F>EPC zd7c5Gykk>QdsC`vdS4-0UvMW4ugs&+@6NZ30V7{Q=8X%JcRlz^L^K#5TRol)s9DxL z!DyxBaV{sI4jm7cL!CqQ3wM!VebhYS%{$SBy2NNhV{hHNjdUBDT43iR6WaTf$c>x$s$Jsc{G-+gnE zbmdvR3s1{aA?ieA=8f_r>I9SWyW@U3!_UrxOO|1;lhOCJ)OMjh$O z=m|cU*%vBjh2W{8>2WP#pHHss(~#+AT-@4&=se{@-$=+In2NZfG*3T!di=DfT1tM%y-jG+Z~4ok0M|P>(MIOo zmF^VgZ`l;PPDN)ReaHUzp{b1oG@oBV8%4nKAZ%a+mHpw@V+I!V35pNgP8S{L@%#;} zYR4D8%(c3Ft#6>&-vOMV=00Tz%BU|qtEf?A^XhWeg;SCVbl+FURMoE-1ah~jK#OJ0 zhTL&b7If1Ykn99`N9IMH^d09R*rXd$HmBsjZG6?Qj@+20S`iU07jv_v?hbe}-ZnuT z*u_nBg6nN)moM~T-oNO!2woyb(R{=wWt?bX?MMpX`#Dg+@tY^EplX!LAOm5$khnqO zyN*@+9Hjmc-Fu1^SFzO|o{IgxQOdb>WKJO0PLZsyOxSn_w7Yrb+{M3}@32$Ad^%Ve zU}Le6VcAGug~YlT6KbijD$#A9S(S9nhx(2@MtH0w<@exGT$agol!yQ+ zwZ=namEIT3(@qCGdMuugPqusTWcwF+Tapc`Xu`r?rQ5#pHN&V)VY;rBDc5M^cMbBo zjQezwlEGIb)6yjKi73$IrI$pU`1tZ?TrWTI*)E3gFVZPf%{#~Wq-6!n{LJ^dIxVVH zo)e5xOR#!}s!qIc)Z_kG8aaA82VL3@AKSo{-Iix9%1hq3#Y{@5jma5rttsUj8-^#A z6$!#^(LR6OzcgI1wBX6^<|lctPd_h=cS3CA@YGZ3KM=IrBi2f(c=as-UNVu7K6j}Z zEIyS+hz*@IhT;f(oud+C;ny{-TyK_kI<>GPw5Oi)*-pbA5eXdNvq5rKSI}h~#;_*Y zj&Hxle#{x+7M*&Pd0ecaBmeMoN#AO2n>J%)qOVy^$NUrBVcibqX4)r-X*kK3)Z%xA zFV`4WlB~UMJPpV;)16#CjMx)!RbI$)tFWZKh%T8gfmN9v?V4|QDLNqE3MjUeh-grJ9v6BuhGNuo6G;wmf_Darp23MYi=!S9Q1RRd0VU+Ec zN8ZBVn6KfI1xvirM2K0qs;tWlVOsE94SllFP*Me z`%A%FbL*9$?IO;LXP|?eW^dqt9QX~rt=;t+M4gUb&OUycX@f|v{^q)-;NWTK_Q|P# zZq>F=3(ZXqCAOVx%@B3(_zu*2gNXxeo}g3@(0iIcaDp7xJJ3U-y4 zM4T4yvPkP>$*9t# z=e@si|7~j(O`P>oj+nk$?OS^8JN8&77+u;7H%GX~_x%pI4Y&gK2ve5>Siz#zAP3(^ za`f1}O7%b@n(u-Vxr$Lof<~KlXP`3m8rkM<<_CjNz<82HfK;Q0wI*Dmvn+Yl_H@5S7azq8|jXP4!rH#ywplN4fMiIYdn}cJ4Qk;8r*AiZoESRrp|$(0Lspz-7pvD2wQQk#$YQ4a<4VRV?Jd&S2B`xnT$DuMWe z7E@V;c6U8;kEAR?6?1>lK?IAC0bZB&# zgO&b7QtwONhfgvNosyZNDG_^b=9t9qr>7XINcnuJdihE3_4om)W$t2OiQQpjR-OFj zo$zmUsmrX+MDN3MDEy0gnWd)h;}e0mm+MoyDQq2ekNYEnj-{zsSyny2vml4}E(X=U z*P=i_qCW#eYW8{N6$6=01+0jDtmWx%(BfsTXP{hYi+R<}FGYLiO~lNr?LA~b^RA%N zR*m7CtL~=H@EfAqBZeyx zv?Xz7t`;PSE{oynP`NHvt2HrEn|RDuX)~|1(BkdgIg#Z37-{-QF@}WxUKLhDs6<-+ z$a~`^#Y<(}G+mCs!>h2zp0JW3BlbXBJlxaO)!jc<*p#SMmAaWJtiL3QaUeaBYn0x| z7Y8qZwQ2oBIA7%!)2BO?_~{>k)|*(}!TIWnu(|j03&OEvLW@LrVr;h#zkO?>a1iVc<1?e61yB?UTK33yLHG z<7Yvm0nvKV%O!7PPx57n7!{A6a35@FUg?by8drdhVwgLx?6)td8s3(o3Aqx|Bz@Iq z3ctieMI{@M#2-#&Uw9j{I)T1GuZ02|FR<-5lz2?$3@*0gHOWawm_+3vV--@aMG8#G zhc?O>UAw>~H>)eK)h-c!h5N#FnM(RWbBP2+u(Cq0aXREqNp6EddA#}ZUgoHFy}Mpq z>*`^ao^oM>wRV)R5M`?Afhwy0&`_D~!@cp3?^$4dBb3xa+Xdq%^S!ooU;5sYu-zBK zHCq`sAi3+S%-%AfD3AF?K-Rq*y&t&(`6_Whp6by1193oC;O?i-enn2lA*RnlIp3q3 zoU2~Q#iZbbT}Gvsm~>`QJ5(!%&}oVeTiWHA@l*cg0U??=ATVX9Vi=>a_En%FZ$iUn z;k9P5_$QL0+;O9l1dD0~$+j8!nJBfeq4mN%kLl9w7?R5gRh{|cgBul0kws&O0ieB8P zugv$Ze_W=0-c7#Yx*@t>*d<-*8cm*~4)&WQ zjOayW(=AUn;;g9eD~A?k?a=DE>1_Q0&f<$Z+sX1FacJ`{+|_p99*b%$d96W53ape~ ztX%di&zA5=7WK~vB;y~>WvmrQwPoxzHz4e|g(gpR!cxheZ{tkdQYm3u z>5VZP!kw!4+WrjCYmJ<^7@x#PkwAMM)9ES?7P#BGq9t+pYUxgV5Dh8@eH_}Fe5mX2jjmh=^vC-sT)(FO*OoK!yfXw zh;xEuZuRNKd*y1eL6gu4mQRW^_D7hf^ctJj5S!6a zis34j6&u8>7uN?+EqYw8;ctxxQunRs8ANxyDc`>8vQqhldX8A7HGb=WMdIbXH_I8K zE_nF-gN9QRx*7FdjyPV$WY8-I%^TLo56s1ayN)A)-FUM=DoTuz!ShCnH_a51#E%{L zBW?XhD`_kOV;AxxUgCYtcD8dZ8~1E+-P996&?qi!8*+1c>@O&d^*$jZSSNZix+Y0p zk{ZGQWfSvzjvwGmf?hTtk30ym$Ux#J$y~B>H#?F|bHgE@r$Bk3%jwsw_qNJ72Ew=2#;w@s07UNBza%2q|Y|?szmIJHNGwsjmQBZ3* z>#RnA;xg zJPODhL6zEttK6N9h^485acIsDLB|mWm!Q9Y%id7uwJo5F2+h_uUF$e4107{2UY{?! zvu}iGlMy_mkb4bHd#z6tYDud~_CD!O?A;^n(u?*+WacSXlIW``Eokh%lzCke`E)Wb z+PaxWpH4rpnTD-LHMocNeGN}nj$5n8g!LJZUD8V3@W4s@kiMNGnN|B;Lrk3LW$t3_ z=Z%GH=*s$C*TqLfIVp&3`byAeB`HV}bau(ZY?C7HTO<+^otwXQ%Ib$3#q|9ilUI>~ z*HUNnf;?U@^JAclys2B-l6}&622|{v7Roo&Y^yF^czE5a_*3`Kx!$$LmZ%_!yFD(>r67wif#N@pTTbC4L0f*=JpT91?{x?dg z{gj{=!Lff{O`@L-8rgh_4+FRU{%tFqot1_!_4jM8!XtJOY%<1}i!I#~6JA2%qwq53 zG^{;=5~J&m--vhw?V6}bE863on7E+-4iuZ_+(gEQp-54c5$K2O_F@E1R93LzpAr;U zFi5$Ia#w1~HJ)nq^+^(6Ep#l~N3G9aGagLU&}8Z@vaCDuaAx4KAQ^SR6O(1L`|*-D znuYP=aOLLJ`7#Yo*qji*R-K?F13P8E68&3VdM@^M)9C&ha%KWK)q&Z;qPJbbmIe1sPH@Gx+) z4ljsk?+{_Ml=c=mS|91x_AoE@Q_d~vaBS~8Gv2pVLO4x~IB}=lreRgad+qg7 z_{L4xb*SpO+WzXW|Mt@Qn1X0D38VMaSUdUz&^lyB@EEdADa{XRB?)Vq ztTy6^VYJHlW#~Bp=NP znmsl%3WY~ChsA`USxheuCp~za>fu&=KD@Zd2lmb)pAZgz(<%tj6Dvu!c2IC=0f)x&#j~}>T%Goi0MryZ2I=YfKex|mQy-u9d@fr|lhWin5vLL$xesr>9ezBz z?7E1aZJZJGjB#K#WtZaXRcdfOD8hZEm3=qaKCgAb0O|TZ`W+4Qb_iU!Wqk;#vqq>a z!g<5%z}{%HjbcLfO)MLQ$-OGlcPQpjcGWUKj>w)!>EgB|kw2xDlCn4jOKuWf%cQ_= zHy3^-Vz{_ZvgyqZ72}BdX#Y34sBblt1bWX@>3RG(C?Z8|2d+H>xMLJm#}e5#8Nred zG6E%7J&ADAt#g|#uWq*JzIZe>)&|>cB)+wEy{tC%f3WwSVNq>c+Gvpx5JUlqq5_fx zC1;8tISUjyNX|K@0u)ek5KyuR6cPoBtOUs*v7pF7a*;%GEWYJFz5AZC`+U7mKezkd z?$iF@p;)!5SZl5^#~gEv_kG8<{2=z{DTZj|EE#@x#jxW;O93HYD^cfoz}oGHWkA}E zRkcmQ(=%nuVAx3Iv)ygDLz4qSg_^d+X;LyaLq}2uq=7g%@t9D)LBXb5+aG%cl3)u% zhS-X(Zro0slie^A>~|WLpn_0~6=ZF5J0(IE8Z&Cx4*NfOdwq9&G?U8D*^`jT9JD?6zfO*a1z(ebOD~}ni7k&y&_JFJ7 z_#5ax{AS>9i61ppf1O>Ww)x!pjUF8*js3eel6@`$!UeZ3)dg6f>dk=Lek!>75Zj5?v1K`k+z#j)VN9_goGf_2 zK+ZviUuqT^F3UzpBM3j&g1<<7 zmEpA614JdWfOR;++U(uMMlGWZZ|KsRnj<1h0cU8U$jLqQ6&%a++<;7-fnrd)8lJta zJSf`sz8e?^yZ)PMgssf8o@_x8k-%#D4rV*_`xD&hGZ(r56)uf9nvI%h8qJwUo5#F| zsfrb=MuNh9?$^J{$&_Xwx?zunCQWznHL=O=0aPUYn@nJeD4QI7(@+{+2tL{UyDgZ- zwr2st4jkAxkyW=(D@K=wNPsDi#bRW;J3C`A)2+)_&m_?))MYI8;TXw(7FJ(VN6)$r zaARZ)b3<+zNFLeT1Z{o8#c;5u4^Tx73_X$?O+tXrVH@K#nG zoiIdjx4+JWWk&Vc&x=8utQY~$Ark+7%`4Losry7AHhq>HEAq||`-j`^*og)rm)}{o zaZ6EWMg?g!-8h-({o052ijmlsCMb%A=MT#pgfw@$BZ$}tRS1xCuAb~Yu56bG)&cQk zV1lbRA-*+}UW+1OZHuYf026$Ym1VWbv7{6KLdQR$-=bT{crh^umSBD8apG|W66kpt zFwW_1{^6j7JSY-d-BJAZ=&h$rpW0P|U2AF2{gGZpQ@s26;d%TOG|6$jco&2hk?U;L z{rT04zTR9Vk-`=pXXDstNvkAC*O)8L_-((vhB9Uq z-*v$6O#yyStzq5e3w%9~Cqtb2=OlP*`SuOQ=X55CE>%I4zD}CO?HVO>R@10j9M|kr znB8!02tE)QB?BbP2(_E|O7whvnG+j%L%0IRvhydvVUeB$b@GQx zm;^jpbz;2P75BNT)&a6YhR??OE^_ffxFu2aW62LTIo5nvApHI^(F)N;`qNdK6|{CU z&L{qlB{x`}XawV^0L~sIDq_FNn^M{5*!=yWN*KU6Yg4I4m~%}N8+_+l&MpNOLl!w$ z+nsk0a`Qj+klh1(L6@pn;i$)nCp)xCRZr*cMIByT%h(wUX4p~o^^=#%y1uHkj$Ilj z+Kb#=O2AA&9o<0Hd_iJ%NHVW>`krCvAx`{86#6?pB^vMqerH*r|5Cf2BWsnNH)T7W zMhlNS&K%GkF0_TDz0~B_M(i~F9K#$Uc1KJk6>`;^yg_NxpPrh1DE!XQt_}S>hHya z&GG?3x(*`(q{w`&#NHEcCl=TWGb(Z_H<+jTI_LOP?OUN#;+y3};>ofkB>H)MVqYs~ zDg%evEl7UeCMJMCwBl$Av$yv>YP3!iSJ-Pk6{18%)-(jL7xL#3`bMX7bOPakWVYtQ+5SNwr+nf+dX zfP+ibpy+V9{A80!Z-X!kMqw273kh;9Q;Z=Dy{W;A&@$-eTeLPzqmRs8c34q>frxZl zg0?z53J00(>`I$x&^=see#>$8u(ThnREE&bW+@I{|RF>}bNy*^6` zJYA)@hEnqb7+J2x(|Y}@71fY1t&5+lM67=sy{Lc=C?Ij5BPiMFyCl3mV^;RNGwa4M ztyPVAtYEy{%@^7WpvQRM)CzD_!eeZo-OE%`Vs~wlPSTu5)ik^|6G=XA_Kd&@0z?2M z+<*9Zjale%TBfX7&*~YlfeEJ5@XPMmJTU5B!q$ST<3cj?+6pN47!#6%F$_a-#ROrx zD9BU*LIuO&tDBn-y&~c_(ZT-_8^u4NIEZ2L&b*^(JLMQPS!;gbY9J$k#)CDmovXnnlQM-E7-+9 zaLPEiyDC}BN@i zqxh*V`!{`d{w*Z{_t8t}@^#k(`mfqC3}{J6^I;D(8LSDn-=yNW5ylDXIl4Pq16f55 z&{zp^=mSb7!Y_r*A}PKnl%w8y$eEhwvVkw` z>(AJskKZ1RG0h+2lt1O0$|^8p4v-MoHWp)^IdYGm&Y?Kl>d9@dwS4zfs8lw+lSB(nzL?NdxO>`+t_w=CAMhCsl1lqEriNXLGIr zq-^QmS?8i>DTyh;6#pV9@CW_)zh`a7GQw%i9-1Fs{tTA;n$hDk&67=#nSePfs50c1 z$;m6{aaD1-H+`R607(S)iZWy8Q|fO3lAr!U8)T~sbvW%hm+RR(`mwSmv~|Z z{JU2=@4n7m@S*~w_9n|{RN-l7$Li1fZQk8VoEEea5~KpaYpapV0UtcXRFQ%O-nI22 z1VRHObYS6B6aVBU#K>?>V-*JKf+1mBf=-1z?0U35ue-^oO6AsTj%#H?i&X&2)6+%< z{oZZEV&x|P6qy%*n~hwB_mnvt-X7s(L?0w8NsDy~dh*zt%+bP5nzXF}B*+H#NebJA24m#PhvH%%y$g%olb3&7IW3X$7 z7hm5-^WGgq^MV#e(BQfEAidLGK-oD@w%ueQC(OXWmXkiu&H%ugZvVsMtoOE>!^)np zI6Rakcp}nk2hbW1v^KP3cMf#6(0mZEaAsCCXr) z(l-&BM`1oV`+e_6OAhg3WVyXVN<2pc&tu~)H{aEgUwO-a%A71YPTwG1o_ zP#P#JvSrvafAsd&ejMkjIF+9@UDwIeP2mw#@7-YE)JMlUh|M^FS$jy1Q#Ek3e$2uU zA5t%OmpUt$z&>!wldowTh(7+GtmVJSPW?%V30vuJ93gvZI$E|4x*2f=A~%xJ=7;oq z)pmdbisAffYA{7mwc$Vr(An58^)i5d)xMwzdE|vGXJ|=R?yo-;O#^)yZ(cfA2At&5 zX({#KMHsyK7?6FoES=oiE;D9jR_P~nQk8GL5Pz*=?dbYY^Xpq5RrTfQAQMGUgjUFo z2WlbiJ2x#!(3ko^dgY<=lA-SACSyA98*XA%MxQN=Z1jzmvpj|_);>Jd0XWYZh*A&1 zYKquawalA)l1#Eboqk&546^Q2E zSQn9XbF{-#YCPRt9~nt{e-=>9NuRvamIcvbNYHkMW?A^X2oz9W+K2gAt=t@jqf~ox z%m}gs`<~Z5>JYDehArovbazHh zIZ^PEy|QFXu*a_4@mpfN=4XaYaRoLyNE-{2`A5W!FNN+*md#ai!Z|qi)P!w<+($pK zXZHrPppXr;`cZPGuz^>;cN5=>nv1dNjmwOc<-zMr=B2y9tQTVT`hWAFjF06J-=_n= zO4O>fJqP4Fa6UTn7(}U6W#)&=h{R`XwOq0dl=V?m!<6n#Wk2pg89@9TW3xym@8hd! zj-svd8!D@t;(8jQg|9%53WU{h!YrTF%QH-+6i-dHT4GL5#}t0Xbw~d4u>Y)CFg(6e zn^KoHb;K3O+ml|@wUK@tV9m)!I7gp+hyu)W~w>JFep~uyN$xtu?spg$Y}S0=erdyM zfM|@%xG?;Adc>D=Kx#W9i0pWTvz$56NcGzb-^-~d6Y63oImEFQiD<8JdR4)_6pp1> z_YN!6$mf)gQ~IjbAxUmKLTqvlqZQEZu}7EwSD-MEU1vk;QFOShTBYMDXfumlw!Y8W zebN3+A6+hK$n=T7wypSZ>4v-h!&36n^4@7;LV;7DL3HU{KWu{UE9DG6ZLCnKV2{mM zoC|ySc*z4qiB~IokMli@6zRk~H~gbg=NpY(xy7=I88B5Atb*Kn--D=ErD;b7A0TT^ zKj7bP$2H)=?83iH1^Nv;m9KeXt!PJY)4C>om!R{s)-TKBToZ5zCql0i=RmB&nA&v& z^ey(NWa?=s4_mK5_fr+T9ZfVKW>S?4UUYuycycnX{QSqlF)N3S13y}lXt%*!%L`Z& z1eFykxpp-}ay&iG?#WYYf((d#G7Kbw$t-~=WD|^9z!F zS2w);vBp20=9N-zV(%JT8*Ox;usUu*ls_LjE0t%-l`-Hw*7L}*GUyjTY6vgxauw}4 z1T-7QK{xt~ebGC;XBrRtivPW9??bj0l6OlS<`hFB^~37|e5vK*jTYx4IopB0ElbfI zv6*7U)egr$;sqrr(f)WiOxN9ZH%>9JzMQZzlWTEvMZUW!Gw!gU>=o}hY-}f^UKD@Y zj%7QR1q)WPI;!o-hv0V`{ysa6iu+Ot7kLZWSyS;1IEcWZ&B*Yxo~q0rxeIh13*iv$ z@q!y1JDrBC1-k;>RvXy)SEsglf7rtW^*{tvgMt?gNRG}|bk&WjZdz4AUdQ z!o87DfgUKs9V% z)Vl7Zk_R-z=j?#K^w}g(r))OK8qD)nT7`{zqTvmv0^Ss9QqVq~xfArp6)5Rc9k=r> z$JZk+HP7I^>l0X5crcXCCePE_rY!8|Y6;J%Fw*Nxj(&=jY<7dJ z&*vs`h@V3m68qFrR%x9ixUw=O3YQbnONy}u_lsTDV+K_v2H3)3E>53#$B7$!|)Z=I;CnB$$fLElgU}F}npsMQYG5{x)+0AJ`<8Bvazo{&=($n8x$v!-VYLAMl z)={h}nf?{;Nlg1~nt2|@LRlXykPBBApqAzIWOcTq47)K%`Ji9UjI*PNaFJ-@wC3RG zlI=x}JMz6k%guSDSB0~20N*~Wh7N?PNS^gOd>Uh#JZeD&*rji@xHgs;D zCdTv@P7si&a%rNxZ$lbhq=_xP^(s5bIAmY9!OG9AYmeR7H~3tokjB#Vw8L)!tcon{Eo>mCbZo8Xi0fTG7FzGo)aY2HW$LS9@x3pK+1@}=ds#<@%$CTW%$y(?Bo-#^ zZS5F%JHaCzA=ITRhakhqE-@K2ChG0CZN3fy&b*T`tOi@1+);R$txbb6F8zjinbFSWZ@UV0<0+KdK!CN z#mLfVDaCfVrR{`8vU6P^*S5jO%Lat9xd%jVQC7(Kgw{_TH

rRl~K(e+9Cy|CC(d zu%hwt)^1snO6bwle9z~2MXig{d1r_7A>1%Cl=uFe(OfI%Y|WmdGVvtC8Dz}ujO5~l zr`p-}rGBt*Bzsg<_$tsHp}B_VlS@gxW5o4Qz%f1L6=(=5NK?sg^VLym4N^20n*JUF zwDh*H_pY)%!d)eD!X}oZMN;HjoKZh`Z!KCHfiX6HFXW#;InpO{8V3ulwW*{U3`IsCd2r^bb%_9ff4ztQS5gK{buEFo0sjL zqd#APPKp`}C+1C56Y}O%3hBk$BA72&u&g2%%ikN}$~!$QGphV%%;9y0LA{-BCL|$mB2 zh+>aTMkBJ#$y}s@P)Q%?*QN?;#>zQ%61#L?ncE7+J~@X?UjXtmY(<6Ilps?K?Xr^6 za_y%p0@xy{iI1T`@5mzd4CL&{xe9(eZaOycMXAI5&4<C5Vs3cvs*%W+ zesp7c!N~wIM)w+>ea0?2l_*h#HBrJe8-&{;66o_2aL?aHoVQq#W}mzaM+bo zbeyIGvWfYYpS<&w?s9&E$tHN<`@Qt$s{aRw zm4rI$7Z@!OR2Z~jQzoDkzGe)+X)>^pfZlp&?Igc5SM4lkyUoX*`9Oi_19gPdtc2jw zVGeOt1!wcOyZh5yL2}6#B2`(*wK5=AWG(uBW17$XJpqFnWVhg;7?^FYHc(MS!J>KZ zq652Z;0vRjNu8wV$@>-J5@C4HDGUb^!KHl)+j=c`_;zk?d5dLDP}JiB zfGy?LZ`XLyu!e;;`NZE}b5*#!*JRj98-Q(^wCZs+A-Dm~V`AaK9P z$IaR2?&l(L?7j1d45VdB9Nz_DV4-2(YOer`Y>T>|Suv{5o&FUyO~Kr78(g61M`+*= z3A+6MMmkhD|E;e@>-Hm2ubrdrI}+mBqL?$W(QcssYi)Hz$^;3~%~xN+p|0 zFqTk^Rm0|46SXq1qy){V7R8{*^!u5dSSj2U+F(0B5NqkN?I=(}!{GRL;HIPheiz0+ z;$8-=vQ+p*W^ZzRtt4TDR$j`jboh}!O;xSRT6x42O=aGR>{+4{L`jgqYbm@Ere%3n z3h6ZsQ&O!ROJtAz@(T0>XuBFQB#!9VYfJny@e~_g_df)4ZNd|>TLAb%;)L9yR+K78 zHi2Bkm+Jcm^8>mWKQn-K zBYq0W=>TAwLU_;9F1^oHbs|KAtFmgKHD>y9l+n$}X(ohmk(rKBex;x5G|VpYSfz55 zjAQ0m&4C#H1)yStX8-h3kC5+f<6}nRUa>$CtvNumFrNDIDO^)(VA?kUS*}ibphq=G*!koDdbSEpHa{^(OZ>PrI7R-DvEWMBONjQ@;;;Fr zC&jUyYDWAV_(ow+D+0VMRtPkG1}9U4FT(l2$Ejz?Er{39Ak(PrE2hq+qye!8jq91! zpGFLyq#HZj77|?o1Q>}pu+ZX%kIqiOUx!Np6poS-fFpwaL=R2-i`eTVET|M%&0J)u z+?t#=>>)!uSbecVEZ@3>KuORZ6)tue!AF@EDWploGohW;K#;~!ViEt5>Gcb&%=Ty^ z?aA$hhL>$FAwx%enU3#ly()bxr1!lTlj8D0klp5skR?D|zB$u_+!%?R>m_w|XuRP; z#e6yz30gk5W){1qev`ZW``}YNcYWcMT3g#|HyV0)i0(6s9-&l|+NMM+dCPJ2V%5IR zsx3hrJu-#+Yo{fO=03|`sDP{{4Ak;%I3+I8o$+z9UG(voRR^;?s?BhNZ7EIBm_DZf zoTDYmvfc{%wuUE(rv*fJ>aasj2|(RLDj$%};_aXpcL`%NOtp(si7;o+-R$)ZueUvy&Se+2?$yz-5nR$%)jgUQz+B>scx=U>hf~oy`cxBvP zXd@OZo;G*FsBKx7$8=*6PQ+V5$>g6%H>tx$Hr#4=EZE|AT3Ucen_$DVs-`AI;hjn; zzrf)do!~l{pxF1l?f8gqEveFL=_U)SwTA4GQ_QybhGUj}VtrZA#t^ zyKlnprR?qDotI{#43KA*$<^?b_=5YXBB{)EU3-Pse5%b=smWnHcIbX#A58aLVfUYfj1SJ{Xx6hj|^tT{@+cZ|PJ z{3ZA?@8BuP2K^JXmQVVnO!;r}agF4eaDkgfvTqU%oq!2FoqwC(YdkW_Ej{$H<#IIX zK?g{$B+#qpJA5tnUaAzG9H2$N3F#oih%2zCs6}q*jWJ;nM}*>|lMS9C`9;0_^z@t< zO&<#pi2cyvgsX7>xS0G#Gw9h1WPP#*RVqFL?bzi*Z_R(W=m(km#ECqtx&nzIW4!dJ zcSbt$A4_tHOwd}>RT+!}7vYcd#c6MXdrubsOvBUK(f@XmN60t=fr@>N(ffC7&VWMa z#+1$D8P#hsW`+yav6-y9>(t5^q0*U4wmo7bFxWSJ*XcGg%t|p1+7KXW|b8cjx2Mq!#g6 z7OdMHCVSVw-%IZ>1OI~yaFgs#8v}DkkM5(q3*4iD9$%514%^@Fos0GEs0#MbiNDDq zI14=d+@GjSdA$hhdFP1mbCm9sQq-*7zDCp~WSBbJ(ENg3bxmqyCx1UbVpYsVesD!! zUGh3yipB#!YKQ=|w8ilbX_;jNfmkaLy!19@xid!ViW)jeL|zw?(1Ogu*D_(vAIoCD zOCugx?!%(i#N-*R_deu~c$_ds>gEy*h2Fyck-c^@gsVKi#|I1z(5B!*Tc^=v`?~Aa zqX%5A)R-@cXO|ZnG_4)RmHaf1dmFJV4i35$!;h6KMnUIV&mw;=nT1%f<0S`C_~xPg z5^w656Xi87itXF8y#~=u6y;^l^YX+cf*re^U_x|4gQ$^m!^vr*t=rcc`%L?KC>c-^ zx0+|bipy|a>a{kvmTceSki%^wUr@BRy^@kYWk8hn#M=RGwf!8TIN~{J&Wx1hutbvN z*SoT87u1=b`f) zGB=skv$KCCy~5q$1{;A}Ap(T#-N_s8ad*4C(~A|a6NMbEzBMnzI4W!-9Xeq6LDnb6 zp!(7AXc#m?S?%7R-Cu||A?g`sd)mEYGFQa)1J)Clo9@KYI9dz69aauKd zMDO6=$qoZHIXm~OcMo)KS(~w2V~IM66=CJAdjr{rv8KU<%`d~(qK3)v^@SQTwlJxECtG-SehR0as25lp9w6z~aaG;_W zu=0G5VHsMk5{;i?T6iGsYw#Fx|K!HRd#fMtTo~PiJ#vVA^Nd8=Ll>7;B3C&=;v{6j zU*Mj_>l_{^@5i(E$s{-AJ8WjArsrgqXL39az=)BmZ8#kpQiE6hqOD+26)X(uYOM7) z!9Tr8^jMqG&$6|{%Gah@EY((&QAsT2M@a)7IEirTn=vYhf z83EY~wVy)&)blmfp#$W#HAXK#;~6*WJ5OxK41JzFUug9HZMJ~E)Buj8U=v+ej_y=FDph z^u|8LX%GDrujP%5eI!PT#=D}@{7>BF64f5J#+k~~$CuiehEH)3;gjA5HlzckWzWlY z>Z2V4_9Ulfyt@E(`nT`;xlLwJi{`G_66Ed!VpV8ANOI@%d?Q7x6#6%ZQ$Jk`MNJA) zRE8_-jK1RS&t;sKPCY zV@B^JC(uWD_~h+V&@SaLCIFdR<83t zyeinwSvKLZ8Fxf99Ty^z;IfSqOULoh#vA>H{ zb=OBmjuY`+2OoQD?03;r061YSn+A4++N1Dy`?Fxt{aP@9CLU$~sU`QbZR>v?4dWMf z_#Z4m{%^M2e*`=HXGry5*x_HG=wG1dU!dqe1x5c~u){8__WX-BD^CbV1s8tkEGOJI z8f?kA#!}ShdF`%SJt`qGO@ZPrA<#?sj2&mkYHr$aNmMfk0Y2^*!N2+wUc)a$?tdL3 z_n&oa@!rg_v^&nbkek^)kqBsfC8(jDMfj>Fm5o0uv_y?ip*sFs{IVuybx%f1K!>_u z@sEA#>}Oib;j11~D{?vHuDGMxV!>p=Ye4N;{|fY-IM)SJa^}RacNALI7I6d8;S*yu$nyIi z6OZgE;(G5oT6Ic(Klc9#<&k#gHk3%}IyRCf>8uzxnCIJ&BQ~#;QgYgmb1;vM*vkWm z!vp>0FX1n%HHZT`r>nfUMm@V~^Cd4Ei@Zkiu0Tz3SD;(U#08O^&$YSju}d)&IchI>fD zbCg8KUw1e8_4+T=!e8USf;ITTmemFW(CYyGRNJ`v6D9y2E}w|wqxRdCGvRHZ_cnw1 z`k@e{XSZ|ddyCN7$422i<8jFH34KGEa$~~=Qrbw&TJw=*&-u+| zD*~qh8J@Uj*U+{}kY&i0r90@Z!>gne<=-#4CRO~fUdlhZS`IcjYpW0cC_}~Xe$f*9 z5;_QUv@^F67)RFZ<(H-#vc_k&R>b1bUHAORJC*smVs1J;@KhylE!uOPJ#}PWZzMO7 z?4vDKy~Em_=Vv;a1+&Jp?CE{ZRmq*h8LL2^JVs&ybz6N9UD=-8HNJ!XqzOOQbpDq9 z%H;bXO?Ob3&m~(^Megii(dw95l6!4GLk@t~d(cq4u4G#^-SX+pTWz`XI?Q0fsOD2L zN9ZQ(&~Dzy1VR8BnTG)xnse&@dPKQF8~>3#Dfj#3au@Eg0K2IP5Y#FQ4CIYA)iLd* zx6(?(^8{%=6`awdEBKxl8r0^a}%k!xKZk$*WO9St(n zhXU#9FbTsxHkQ&Zh14 zAdqJOe9tKP=Z5-U`LqAM%hA8_p(K7?F-Tskp_9qy0 zECtLYj?1ar*=gAl;g-#fvIghyGZu`u&O5^?=xy;xkQHzMIIaeC`~(hAIqx9Wb8vs1 z_-lB68O5)u@XJ&DzxR)d_`DnDuKRxKSnEK`vfuevzK0x493TMN_gid6fAuFoKCD|i zk%3&|>#(7_z)tIzc)!fACI$buCIy=GI|x+VvCWOY?j!rhz;x_)Npb>~yd2ELl*C{$ zf`3t!!~}lhGS8BonFS83uiW8xs zk{f5gOhG5ZA1PKIuj@%R#d6dn`QPO2{NiW?d01{FTMKL4vF$>E?S+Ir*|Eej$Pa`h)Y0cPx=fxsmy+)9q9*pBVi~R6m zGc_Ec-1|h>afx9p4pz-|CrZZm+iq4t)T*5D?E~yMWT}mQ|BMcFQa{Vv+Ad{-mM(z<$P{bDA+h`|~x-_@2T%CM_p35IL=T+oLm>PS-K;4Z%#g zm3$+z2`+N8z*QK-l-NmoJ`A2iALqu;F~{b09{+wm(l!f6jVn{iZ`lmEnzytGy^Xo1ZKJf2fgm!i zrz%9?O(V?>30N^Zs8#aT+$2Gi~*%z#bEG8RA0R&`V|F)++` zdd~jN{RSI{SFxlMv>@=*{Wo#)xu3Q#IEq>d41AL_+{2jEzUdp&A=lD03+C7Q4Y$q5 zV~YaUu_}SF%axrN$T_>orYKFo#+Sw{qA{8y4f@^UNV#6GsvXf5A^B#Ld zTgegtd|BY`t)fb2jqIMb`m&uWmgocv-@pSaZfPfva^H}v#-x*xVI1l}SC5J35hUp8 zIz7cd$}t;O_LlENQL-E_zCpCSq+8b397+V`Tq`Vi&roP_FJxU#G1|}|b+7%U!FfH^ zMd4p=lRb7UKrAp2$~4Nu#;^oRn!uRw*I_EsHL}%wcuT3Sa)(CYO!H%$aqg#MG%1G~ zI_TEE^KNkyL#z_e2NQFi*CA5=NE%>=kw?Anr}Y!7`ua_FuEbS;9d&;v<~fN*cOKSy zfoMV}-ka7S7+F_c?Mp1impKPLm?23T^7wa2yT#ZTT%G9qxbHeE|HU$&-URh55PHK|Q2NEnSlPdbd9V;Gu83`~VF-Icm983JG5B2# z9a=pql4su_MMk_9B0-W{S;b&s>JQ*Iu7| zHQ$SqV?|3n#KnDjofTtj*jS|q7TiIPM-pgB#)5=xP4p>aw7Lu*7m3D z=I43zD{8(Be`OfYOJ|LZ`pCD6D|FafjNRBPArOq$p(M)c_kfJ=zs7aH$9MlLeouMw zR}>G~0NL*LXENl2oBCCH7&tYeom9-{?@hOX-Z zr(=^r_;&KeJ}>G`e_^sVYRm8g_~%knr59L%{RvuhiZw1Q^sLQ5*+*38amj8*F)^Fn zPyhWdKqeKp)@zc9ct3rpW<|<|jlgW@{eXiz$!>o^D*bdLzyGim8GSicv#Os9zSTv! zn6k0KE~vb3ZPT(CTwxVy7vD?i!MpKWww@(7H_M_ntRQJ8r>Enm(mnPEQFdOaXQvDV zp0+2K>|{n-RC6O1_?+Kg#Yvd2pFWLjfZlim*O2je`425vPOjmdK^{Hd8Q#9rot;VJLcf*0 z%iHe77vuSE(rAuje@E}O^Q<8KBh|4g!|nN(47H;=DzOagH;dz4L;XE~4x7NL*1WVq za!j&cR;=gvM0Xo;%Oj?hfl6cBu>ALTfFj+EjS-aBaA}k1q#K4<@TLSMB|+9_`#r;Gl9UE9Yo(geZ(48L zRbk7{>ibMZ5TEICbQW;<$$uRA+%=?8!3O?P^rlp75TXT(VIeV2?*A|xzt%LT}V)a__)k!To)*t?{9 z4bSAXBuHbRwk0Uo2jEwrQ<`jagVIHeABQh8`wGP1egzs=fF`^H`LrYyO6e3voA2}S z_$Yc++rP5zZ>5P`Vi12@cmZHk?}-3ZB=7-)aR~ES4bXQu)fRHxFl;1Lz>dw}T_#p> z@tI9UQz^Y|U)tTC>uwf_6#n)>F`5W~d#1-7N81MY$)#1tb>+5@*X0&KMt$ zd156mS(i1Nvjq}lFsuHFzs4Aoe~iWyLxvl-8sZ~4WM^)z?e-4^XU^qk*~%fKq?*%m ziyW@Xn?myYai1%opz72TTX)--8(A9c-Pe|C#NjR5qFT~6=MhqyYD4-Ky9bU=jmsEE z<;s~|$idW(kTa&lsT0(<^Z6nz7NWi20pneN!)S zX4eOijKF0ZLz8aIvAhcHoF4VKzre$z`)=su8J;<@*NK6s4A+q4STIvEOf_;qH66LTJWMt%L^P-8cgxH>!9ULNF5S`9&Yh9d z9EO2IhR}ylzyM(34q%Xj;GZmkeB5Jyw#{wyPjb| zoUOmQ9MSGa;G3&5RQ_(}{Vu?2-Z}`BJAc}3m^{_8Z7OU5=#OOD>ML&OjOe}P#~;lI z$$CMs?F{hzUiiz%n>mhiaS6@w z6-q<^s$Hzr<9*?=qA$aMdakRl6Ja^V0a;x%dlduNYsf{oy_^A26BQh?AzFTkSsNe% z*sDGC-4$pk2IN9W{La?@!7J>#ead@Xp!0KAeZSPO{;;==;f|}K6qH>n2QWYqoFN!s zEk<2|aw0dY=TET#=TR>~{z<@}W9JREdHs8{w>vJl;7sRnriJgPcmTuE=x4Nt9Kta6 zT%rdCJ_212)ijR!>%XM1hV2u7#9cfGW>MLDAl$Uu{M)~I?{dKhNz{s_^7}_S7_c1d zqj$mah2{p?#P$lfm&5gjM??{^S-FiycxXwA`5hhR#pYAOX~5e8a}&5RlONd^{(xKk z1WmTq^)U;Sn&jNEp8PC39Mfg9;7!*+qB#8X4B!L686FKf7tcAMn~~7<`@upgen%@7 z5mFWAjdsQ43yD5hQ)!}<0nzDka{#I(RXlrbucRMBW%+}ljpo#8>zzRX_j_Og83UfR zJ1{WW^A~~Z7;{w3^^vFp|EkQl(^(yUars)vwYpKygnZvDEs3J&gC43Lbh5wV?&WM* zUTWnaZThv@P=#eul#b`08mNwBI^r8DJg3p>MxB_@sgJ16BSvwCAEA%0# zQyuxVxFc9Z8K>g*GqT7Qk8a~pjL!82qk+;UoTR=Ge2&BIuG0?PF%MrjMdTjROm+)v z==l|BqqH0s+Ev|k-gc^;4UiE8oRT$-VP?s)4V)C=dByIhb>_pS3pvzC!0QQ(I%RjJ`rU zlED$Y7}A=%2WQ%iMpMd0T-BYAU3@NN>`nyGVgTtY0LnIpoV=w_{|M2?@I#cKb=UdA zPYrzn?Aj-RL|eE^7VoC*6$tAJG-?0wCohi@yqN~QX9$fEo`SYqfsXXI9)g!TNzPWY zyERFBe4^AFvJQFeZ%PVYkL6aAJivS}P9!cX)AijI;oO(4wtouIzPP2bsBrhy1BF8s)dU*xvnhR#!ZF=YOAW5N-@lES=)_Z3h=$Asfy&1cP`6Ip5-k1DC zB&wONGiQ*Lgd8Kijr&ImOYxiw6O0)Naf%fsK7Q9_mj*&LXqN+e{b*gY?LfVsc1w;D z2xG({i82f;Sa`b~PD%wvc4vl0JK?h&iYM z9PUB%3-+OPQy^yl--(mp&r~U~JD>*kAr(|O9qnEpdG|8dfDSwTt@h_D`>I4EpukGk zGsxO2&dGcOrf*tKVB`7tGN%Wu_a8gHxx0oSqi(g)o0$qcr&N1f4>~6XC;N%PQJzkz zqzA6qu-b3NChWHp-qdU*x^EgWhkvKCvVU-1!rj!aep8<=lT1T0VPKY<`6b6>x5mmH zHTt(WIn6#1)G&&P)))N`VW(xRe=@%qB+uk(R>jZnWRAx!i z$KXub7!1Fv8X?->eeWKZ&y9a*B!h!649qlmEkm_bn{U`;`Du7gDJc3aeT4H!)xo6@ zXUS4B=PEr5HoFV{wuEmg{Jc0Dt^k``=%JjUgx(yzspeZ~++%CaoJ1xn5|XHv^TA=O z)U?UUAR@)WklZ{A%oZW{jC^IKco%K;ZHEQ3AfFRf&!DrU84@9&dF zF{NAtyYq%mv%NBRic!?b8k@wF&E$rw^{pcXt?Yv9paz<;sP{7};u#}fJKj*Nt+PK> zSd7JT`O1$o;178KIJUb5XSI7dN@hmltG*uIS#nHYnn)1&T9rncPpcKKw}#HlKoe&|As=(f}>g5G;MskL5>dY&fgtdzon}b)zv4x)UdP&O>79as=DMaoBl zsT)Y}b#H3x59-8%h1r|AB?rEOU{hr?Op$s_)zc&_VG?Qq#KI7Vv%t~WM~8Vv#JK+PeV}!#D&)aJZW7cs~7PfoZn3y5FulG-GzNNdUW9v zy5QFiFlb{(vjDZRc=Jo0)1ljA6eXrH9T(#usgm%Iewu|XoLs&0bdIU?u zH8~QVYz!|1-LLPz-{Dk>ahpfBTiycFLLUZ-VK<0sv>%2Ue@yaQ6~Lksp^n^GJ+S*S zGWqe2IL~n46$nE$POUopjN_PFw6C|g#^<6CyvIdywj^=G2x@DqSZlR37=43I*L<%y z?dww&vyl!~8EwWwdLLJ%a%eZF8nC2sJvD}QQvqPIxPu#c)OE@#?7ArFg1Q5V)T z!+jO8sov=qZpkLk`t?R$3VTqZ7mBC_E&@!Tx$LU+`Jsem+}>2tY=H;gx@1a8XR>%Ii%Q-8ZDR+^$3Ewwx8P3@7=Dr0);?Q@&V&L1+Ft4FrW%P_?N{w zp_a}=JYW}Ng`(fP{`_)kBqu(f>~(HgcKa){VHxs~y9MA`7=nIh=-$E4{P(})MgG`F zZ0J9flpy%VM?yTOKa;*qxs}wiWNY}#mJ4WIdWjFjRWbrnmAqCIBojO{b z2fh2+{H1vw|pX6kInldlJDdk0(uQv z>Udd_g(tcA9f>XNg;Up}modJNysH{H)U7ToXn2*rM)ZBueb6{VCvRASHudF!LDaa? zG^d4R=x1U)84 zbw2?r4>8(DQy0N4me?`i85~7#T*}|6%VfqvGtgbkV{I8WIRD0TMj8TM^uX zYXQOCU4jP!6+E~HcXxM5aHo*qUL;6x$a%A^zdpVD?CyJej63c)-;bhdPzCkAbIrA; zJ@c8KsRO6G%AxX}x{~Uk1Y)F5`n3_Ck*2Z$CL{{{>ejkU^XBke6fkM1;Osa7Om79! z)i7BXqP;~w##d1R1??6#X0&m!B;clOQ6CNRgPfpSN6c>b+1ywR;|VxIGo*>(Zcf#Wk90mjk|;`w;RmlH4#5vqNp0VrB`ON4AZR z>oCT|9lQI;SzVm zvsz1cB|2hi%L(8g^_LLHVV>K1luJasCtY}@`NZI=W>&H}B#R7{q4g)3R<;V5%=|;n zYWeysH_eBwka60~UCJ-50`z%EIvO*%Sj9kK-b7lCq&ocydo`jzsfyAQSHYFK9T95i zwWN2|u;+!y|5S*KN@-^EynGZ%Ou*f~qRPT308Nr|Qy=I>9`;O<%mh+UE7>d5*!jNF zw!Tc{kbbS|kEmJ_q73HxZ9WZv0S9JRbYEG#X{(*tvf40JVu@!bLfgT&PL;~HC~7Fq z4_b;kVpUc$5DeDe2F9;zab6{9L?0>H4;2z$V<1JVdRFvGO?TEDX)dQX3noXUh$NIs zg1lx2GRmw-B9gqF3AM&BB9|%pe6-%!y_n-N`=G2w)@KnL!M5iFnE~R_*Bvh=Iqk5d zWNwW-D%Yd3kz1aydS zl{wsR%*b4r*LM7Xui^n}=uAO5`Kr8vRAmN-)^^?H)3~a2VsW0eruRqdRBe`~C4z?IF`f(!eDHSkLs$Jqejv#3rQ!_?QCi+gZ zM$P;=aoQ21S!8DmFVatg?D?EYFomlvH{9Eex2u=L-_s6huU*ltB{Fm6nLLvO9bSkA zxo0_)9x3%ldwzdK7Oj>_@^DN#8@x`FeC%V16>q_J8I+iF%wd4#?-6;wH#VSN9t<0b z=1-FFh&Cgh4Xm_=kzu)s%+{~VL(=4}5_fnal8CU_VJ#)U>)s4lo9}ci+}!OtMxV+y z`T8BqtIbT5*mEl79pORn4Un>n z8*8|m>oaFa^2Da-bFkBaqPw?LCf8T|`Y}i;-hs^<;DiY0f?G&-c%5&(ij2XC{Kjx3 zUGL%jF6qgl#HD}LdbKz@h} z?eR+9Nqre76%%{|{1=y{3MB_Bp<=+{nIQ>Mt*MC`putiYX&xgb^W`IuJ(u@q8+M?L zu|J^UBdC`|YL*-8S?!m3Pu8d@^hw3Q%Kb9x<%`@{Zvirrshoi3Ve6ZBJc+2_QQC+v zn9ak3`L1VT1br`wL${5B=u=lo;{nV1bsn`K`;~F`XP6k3@7TJRhRJDkFM<2i027mo z9o~yK-kU-Xh!vAyyrls<2a+OL)4W86OQeSTHp5yY%A@bmsNij>yhWGkrH@r&*Th6 z+*EnSjUK>Q)P+=)=PLVnV3V;I*+*?JTZ(9+e12OeAMY1voNZa-P*G%cc*~PRXvAs3 zWU5jH3^WJ=w;cBV2h8h)WLxa;Fd5?+I`KFeye`2jOrVeMIO5Tw3W#ulpl#A zU7e6bxNVv=Y;oOXX22=JM)LuCX}@jxmbgxf-#NVU@oGlQI7)%kaR;Ab3^}LZF{CqBiQ}o1B^p@g^91@K0H2VMS$ z*`xoW$>UGxCg?wdD*pNVe-C5*FF`k_-3T-%kM(7nl?#P5O8y33fP&av zvRu{n!!IR80)MxE{T)Du_(%A{pRWPj9sQq=--le5f`2Rc!=xlW768!si&~u>=*sDW z{%(^+f6esQruUjKx626m#b1@_Zm=`V3!0%8H!U4 zuo+}rW5{~!Qd^crvD&DMSo4m)j z3-Cyti}hK*psFI<$kHh53tB|(^r3txsg*HSRh?2s%LcCRAIC}GZBCYgQVYX5I&j7R zcOtO~1{@X#Bv5XR(jPfA5CXFGRi{l9;H z;L^tSX3B`HXOex{?DMX4J#3k8uz<;V)wC(Z{7nRX)l(IP*g0AbUP)q6F*9X8|ISF(vUC6E%EH3K{jFK8&8HWJ@*)sF4~hy{lC4;2$;#q zZp;)+?yGig7iRPZ<7?P_EMft3@^DUg+!Z%mHeAw{N$-%WN01moaz3YemT#WDv@2n( zBZ(|KrdAcF2n^*s2SAHz%eDZh>q+s zm|SWa!SqHG5m2|k=13X&X`}EtWL_5?TMODG+s6z)8bTvy2YY<3dbG!6whCpj5P*EK)d z83Kg~lI|C_|Pd zT`1ECnWm?J%I063XLfR0;)qbBD2DUDZ3T9HvY7YeZDj5Mf>mQ~U9lH}7pjbYNil2u zx1*ntUr+L>oHsQvVFPIxgJTe}g!J|W-`S@h+-Y`J_c=kQCm&I7i}*lDy2d@jQpgO~ zYIw{Z&(56ip@S26YMQn!it)Mc7kmgde}T$X7KXXBa0^G4F0j9>oPbGY{zHvV!^S4Q z`CApEP#I6y{H}dzWcjbTU zAtk{7B#EaHv)0@*K@A*6fg!(}W+b^Lmj=w@*Cplk%*#E2;TkrVG;6-T*6TPH#&oaL zW!=h_72LaT>+b!>r)qRo#`Hh>TE%M_Lm^~QipTO~XA&+?6K+VBYvZ);RkAk9CW(SKNv-_EBNA5F_AwI+xa zvM+~I=IrF0WY*pL3Si#7=$F*TzR{`xexviwRpmlh{=x4T=wwp&T*a|E;-V$zxmh!I z;Iw72tHad~rQ!!SfMUe)o{3iqKn5V+WbGh0(`m;8 z83lx*y4_oik@@zjqRu3)KE)TKBhz)yo`a)9-S@M^-ZVC+D_zLfD_?!7Ij62IFTd0q z$5c2GIqhNq7*sSj8iE98M!wS)J;E3iay`#B*=lJ?T`w;Sie~*T=z4=u$ee`wFvpaD z5TB{_84bW%XfhKUw>Ae}nfvxGlb9_R-DH&648L{(kUKX;c#V==Eesfah~5}(vXbOd zCMdZHx1*<(Gjs3QA_a~KDBVpLzJ15=-HSWE5!8k_#PNQLSi9KBZLjNOBc^bC&>BOX z5b$M;eHr!H@uH!?@+1blQ<~x?{Z`}!IO8?hald5I2;G##*E@)!)u0n5B)k`td{;ya zfp?f^&O2do1Otgho;k>2^ebaLYAREIyp19-?uFt=a#u_(Zocgz0G>z}{WC`T!M4s0=p#?ZZTM?$9kea}>kAJuPFTIpCCEy+Jj6)QC;zCro; zec+L^AWlzVdn>!~DsHO9o-kT*;o_3Cb@EE{Qlf*J0!A%@NGU$+m%g-y&zyE_+A1iX>Ti; ztE-N7A<^w?*y_3mzB_*uz#g0)p;3c`UB-b+l!v0!xadD_dUvp`b^D_5?nLQ!O9Kij zWy$0GQX|J}yVu^ym=7cvOVS=%8czu-bp)huiaQT${hv#NnXfhHqF*ef6AsA2j~LUL z8+F548Yo*OV58G?Qx(!K4o?jFTqJ<|(v3Wm#I9fq59GzGGV zm`C<<;smI-^djKPc@v{L?J<(S;d_oZY>%kaWVh5U8c;6(u?X%?EY_RRd<39POIq?E zJ@RC+sB45o1Qa!|m!wwXSmaBDM2#J{$a}Xb?yEIeHIkGXbVPnwcHb3&6}e#PGWK$V z)m{u^;G4O0Rpx5u(#Zho8~_#&E!+9KU>p7BH*Cm%1wX*|Fd6>BX@;SnQ%Q_Gf>E>y zcXw3Qi#ToyK)$D|d%J`)8`IQmo^C`1XioZXcKa37D?Ar#}Ltan5*>iHsszj&=F@p$?jm^OI3dk?+DvFGyH#?H=u7Y*%m477StiKPpz zrRfpA8$&fjax#58I;QRKcD)@=u<~IS7cMwTm@>_JMvKLS8ds1|%G0LUrV{tL+X{=2 z0?#ZjL;Ovu%WSF0_K^+$8vT+F0~$%6fv%F4%yl@NhyEjJO8Y$u2a_dDyTaJ1TtG~j z%b=2+R%3ljlY5X$4hPB>3yA4=T+YV{}eXb>#GM1`vI$m zNtUvT7kHZh(&c3)xVcqreq@Xu)OkNUTi*5X(0k5Zt|L%5VwXd%t8$}RPX$=wn4VGt z|H1TcKSw(gbV}y4PIh_kT^zO*q$n~QW0%62^-a|-bL|IVmA)Ob);QikaHwQW`1LVo7OQMI@#bFLvB>+4PqxX1FboaC*)!KGu>n-P~$Xk~0 zv6`lhElX*K?csaLZ|9+<1zwb}H!M;#cxyhV{CQ}yI*pBlc#6nucK659A`9h=Uw*Iq zttj9@JeS3)XcYKV*RWP_<;4<}Th;&OvNnU8)fnJ=rdYvYsPZ5D`F+wqK~!GylY)eO z1QF%PG@e&sq}{{xouG#wQBj#`s*PFVJF{kxw|m*+Q!e@3to9u{(PJE&A`OhsS^=mi z&!Y-jnJx93FgW~zGBheC^Xd}_i(jK{%V1qu%v7ks+1{%^;0w!8oTqX+L|F`QKIc0x zC9$neLtmmoCK6dFX;J!a*u*rj%XcTK9Im=?abBxy7xjVlHg~05W zf&P;qr9fv%@sW-*?Q3zY*yx^m7B~yb4h!zL0bKr#z?3Y9&Rh1jYaSZI2xlLSrV9JH z1Nvj0qx2`){y3;k>VyyAxL+Wko2(Y2UG&a(!I$(J16h`Pbfi&gYTS~ZV`QbhC}?Lx z@Uw+iT_UlR!3S7Q9W~3F9p1GLj9cg{^dybONP&Szw(i;HqxdSSoC2o@+FMbu1tjJS z&Itsa)W9||EGdf6};CA$L#qOPu?o*9#n4>NK6dgDE( z^;;2{*>Ykjuj>!i{Z){S>1*cT%Mb3-{?^MtGC$g~bNTJI4$KJACQta?loK3GB0a9n z*DFW3$N4~z*{d0em7i;i1xGBLDTrGg-TD~1W--kzCod=iSL~bOylY;D6kp^80Gw)& zG`A)sP*i4e+1|?;R^@RdTB~%BloCb#tLVVO0s!=q!Cto)z^OAb5j_6IPtMTlwHY3g zOC8@~W8rnb`Jj`pPYR2>^kY8NUfPYknpDVFR1V+oDK3ZVu1kVC@E_M_k7IgiO@3r! z3ks&ZuxM;TF#A{`fVTi0>3~dfcP0-hrJNeN3b~cb15EB-+TE8Y!DF5!uoO&|(oB3g z&Ku=`R{O@lXqXQAAgL7NEZtOM-tW-FRIM`!X-G6g_AbP0+na2E2YeE}p(@I5;HA%n z`g4Q{_V~bW#b#@gp ziB}WCiXiL#q-njB@`_na8+VK)zkRgN1ZiePHY5@DEQz$I%f%jO6kP~6Gk%nhD{}oB zj`dNyh|)jJw?lq>yYu=DjLy1IM#RlE>r9$tztGIrtx@`wwZp`DAEn1vfabd=Vv{JPEjWWX8#)bA*?y&NqJ9&`&wX zC?bcf3ed$0_pz}t7PXJX7EO_C;_2&39ino3#Ry}2ohSFj{__qe#oUKnK0Iv^wwd2{ zwa7h-T+KB)W}LzImsit2TdyLkr{Q67UJu~XF@(dcH$aGMr@?YQRGH$zRnq#{!O; z;Oe?@Tg?RgBQ^9-3s9M=Q>xRCfgcmU<4v?*|AJEZ&+zHrg-@DK6f44nqo-$-arnuvucgD843B&jfUw-?>_pp5F| zB%_E4*AX%`m|~&HWkSJfK>+l!v7ZS0aIS2aL@tycWF}mEu8cxdw?uFHM4FPOPz_qk zjQ8fV88L+PGYlTw}w_qosSA2oUy6c)~vO)3DbZAY!-w|AZI-0xa6iVUGBOi$p* zTld!1^oW%UY>dGUDVzZvC_F7yS4Oe%k@r!4_Z~B|uw7at%gT9a zTPbbE>8=EhkAU^;$I{CoeEOo*FEH7VUb?X+swe@C5(b1!!#KzZXM{Ayy72IU&FjBr zcZk%T2VR8d2ouXaxB*K!;-3r7nO6UJ>{x>^Tw(b$mO=;E^+}MLQz}TaiN{a7N?YFHObZ`ohY5s8kS9T>qE=Taj3c%7K*hy+ z%mh!n@#APKI41AE+0N9dA`%3>@QBsIe#z^zkt=FZ)wpICDTw1eTSAXSGK}57l^XnV zmd4uD2|Ot{Zyy}OaW*7#`CnDK=Rmjczf7inM1G8+TCKu0Lk zD*Dq>N5wwondywpEFYShj zxoIFh~YtgUc06&Va*Slp%d*(X|8VcdOr2P+3_+?eEXwbkFN)-YqHD!pjNcairn zmGt2I;;vK^3M!kF^g;~ZT#JAU;VD<7n3mQ_PXy0g>}&01qXQFD-k3fc`tn7Ow+2m~wuK;@Fo%<_t1=Sa6Rt-@7`fR5nsx$H9%tkt4l2Ot zC_v2hzm=XEWht|sxvF2l%elzxyqG>4qv1lcp-dMUCNrp}S@f6Q;=mB6)cTZ$CJy>* z+ASYSxuXMy&rzNusu9Jd@0)d=Mso9gB6Uu?U0^DCsP_ zvV0I>TSk!NmTT1*=QaO0=pj2c>nlOAq^txRq}S$g=n5h*C8@S>qOLVO;#B;MSMLeJ z9g58)n1JQTjzxc~G2*!+!k_|u*Jkz|!v!$q%F?7!zeLuULy7KC6S&8H8wc{<73vLH z5SFtSQXx5G{Z4|BU-z-eR~#fnv3n*dD5%iN$jUH;g$5n>t=d&;GXQ&xeifWI&ZNl9 z@4-*!g5!^*33pv0ePw5)E&}-YoPe64wlG6gx)`OcugQMtOmpF6}AwTG6)rg1z+{=)pyW8uaSDm z%gQT=0YCmmr}QuAxBs-`|IUv8PKEyY_~$bCuj$*w|9KAQpZfNve*G8s*ndsm{sp)6 zpZfNvzWu+{H-(`>Q?SK=jXE^E&U^OZhU)fK(Kjb3rkjQyKquv%m30 z|KXt~&-jx{8L24eQ$C1=fXrE%@q2i1Wl4JU)Nk&_zhg*WzmBNx%KyVJ{{BW3zj6Fi zV7#THdyJzm-#^6gcX9mTIjz~Hk%p#T0KoJYBx)Rh|Jj25&z!;ksj~e`ef{^}|DW!y z0JUm`*DCT*cJL#~SM>cL|6jBJ-8c!M$I4qVD%b7Zyrp(tAJ!AWEV*wVdRsKwiiK2S zhQx7`e-eya#aPF!eDk*$s|sXaCKel~gSC*#+c^uqw`t9bdwv;DMZCGMAbI9*-&k5$ zO|7Z_g}s|{Oa1a`zYWr|f@2tmP1ea1E=r5p97^um$ZyA4XVC+n`kvgEFxd$#pSf?w z-gl4rCvAgCt$c0gjIYy7X)Ra=3F3v6@)D6;Q1+XOmo)47&YQ|Kg~bt(yHv*Q1B`3z z-8_%WOdIQ~Q#R?rFL<(v4uLW5I|W~EG`>DZCejNMFtJksc-!hGo;Rhvmlz(LE4Dg< z^eAOcd=X`t0k~sg#UB>^bwLM)DX*T(Zr)aW{xC7~_$_34oV!lqdqBXl_l&O6T@c*= zFwXzEj)3T0sGI0z(#-^F!f|ck{xLO>3ii71C+0fD)*V!N!IHIWP|}N9OWQGd#JX)s z?NyQW{o1hl9%$V$ub*D_4E^F_f3b+-=M>aSkvW540-2jzy;xl(bUn?Zut!7j-24XJ zXRb8t{4m})HsOBCK2 zTqk-w>(p2Y6j{6EOj>W6r&y0uk0*>b?Ma1rL^9}+oQn+(o_W7BmTyGI6^5>h~0yE3HmIKkGyV1trcX_0Qmur7Nrn)bhFZf<>Tb-FZyl zJ#oM3vi{QQp@8kh2szF#(DpQee=pBue*x{9xopkI$~YfEmX{4a1zl|-72zmy6-72) zD(J|X-wcXAGR-ETGAxWDtJfIT=`S3bIfysTI=&PbC8`UCeC$hAs6p>d03fG%th>6J zNpqJS7BNc4X(py7v8ux5ZHYURFeef z?Y2x}#lJxYy~u-~?hLiYn?wM>3JqgVwFj|NfNshB;WaSFhXml~I{fJWG8;Q>MQZ#F zcwVVd7){kLKto>9=S^EOWT0Q~T#IOYDto4a%SV3U_*ML{Gb*0?Grk+w8%>l$r%>K2e8?(jj7aloVRc5t}jAz2@Y-@sD{ON%AP=Arbg{;IW29gX#$oS?+G zsZEDLXcX;-n%^I!(nN1Iz6}@6iEgn21PvYErUG7=Pv%eNIZ=F7xa-!k8ltkGOt97U z%@pKoi%xtTP~r26@19At@9imS(+%!Dx~53AVDEaV=^LTha2#Qt-5X`pPghwovOHez z0|(X$?d*EFwD6cG*j(-3yH;3}YrI3>Y+4++j+mFs4x&+I52$?ztkdzyV@Eq>h9=C9NalSOGO5{*;* zY|j6hw<623lNHT=vg3vZtRx0#v4j_BF-{LW&a(H$$-74}s8F5a?l>WsnU4HshEW~7 zv>L#5^m9evLIwGarud|`-S~99@?7T@%U^_W%VP7*ssZ*P(R|2qY};%NI`oIPK=v!E z5bge5eLhJ5g)@g(OLan#Bvkrr=k+p!Sr-@FK#}@csGW?xD}SUH-6@%$TS)nVzx}lS zlP_jsLA0^%AzQA%ZNi4BB|>Grj(5$UqDoHG+Ojnb=gnPxD*$L$%v!F4X?ICAT%aqHIABUSzQJc6HxHw`2P( zOgQYe&A9x^0QvS&heN8);zwRTT)EOR-Tr6Ih09VPjv*wIjw$|otH8>lf>ejHhXve> z)=TaE*r?8YOZ`>Yg5+-z?g>8xRn=EsV_NxtZ+!UrW}d}9agVu0V^9* zvx92-Rl#GL1oPWg02g2O3-sjRo9TPqrB>Rl;xFe* zfUcAnE{u=sCTw*Yi=n7=?Z6DXw7k0n}WB=!4So@?p(6AE`S&d=;j9iP`VwPSCw zBbx+g29@gj56uEBV0uaqqi;^$-UZ>_9)ihHjYG+{Xf4}sKulP`uElNiA=qzXSkkzY z8JU}Fuy!x+D=9y(2xlqZY~d|fMT?L^di0R(sB6I2PsojOvUrf>T?I~G1M~#y&Di3K0>QeiH7gv1o$jm0kG61jSO<-9h1d7GucEmM#uKg7KWFncug! zaKQhpGSQ$!3jL1J=IgDZh$zo~57P9^J&K_XdwD^*kuN&zRCc>VFgWn8jPYKX<8IHfz~LYP+M(mWw9Gz@E(-PbnhyOA~IWiv>FX4HFVJ&fMIbW{2X zj==H07+@mV)rvY8koSQQXL7tpRI^5i>I?z7o+{x3b+9)k8sB2aK^D-Vq&{6e2GS(D z5&bTNA&Q3taV6u`C+{voRj%zzJ4hR$K+KZcMev^XoSDM7h49_w9=zs$v__e)Xen#wvMN$Qof7sEO2t# zL;8i3gp><@fml$T2bZP;!6ZLCKgni_J&yS-jERLi)>s-f<3{grv^uk(w^cc$;O}}- z4ks?~eTC$R?$|106x=-sG~0L>KeeH6B91m%c{t`TyM5$(t4OP-y9J?Q88^V?c2Ncp za+uLKTL6w6+ohrNl%DeU>yZi6h*j#THFf5#@5&1*aR~caa?`c#4ILlG>m3iLrzsZ7 zzT-mraXvkgB#!l9|2e8MrDFGU^fio%Kt!`^Pjt)@sknsjkCIwrymx_xj73}c)s7X& zXE$kpwFA!qzHsw#z3ED`TOkF&TIrdH$}Lo2>U8=v)FEs!oRCj^;WxTdO_Op7l?lk%3<@g*O8_EA=f_3ja=Oq!B(UAyQ>fOYw0MS zM0k@C5wr9;Ky+ofaBv|IOax!rH7M#uBM7fNd=1-Rveci zi;=8ll}~_uE_;j3NjfT~;3uW7DGZUm(7HFgbQ; zp;=KePi0)YdIa8;if{3vRSEVqoGvFy_Kq<#!zJ(S3inEO|G+Yjr>(DmmN3fct9mat z;kHJs=eWm%ja~iXf`KN{zExCH>*?HM#(YL4KrQe`;9H%o<5lsJuG^#lLB19 z=h&GuY~-}aaU=KDy@W_pmF%f5Ly0>OfoUVRHMPfJMKTT6(u zhx7wt%6)0BwS8jaNu4SaVk!Mn@R2kCKn@V#H;1lra*f~tJ6f(!>FGZ9%h4^IKxj<8 zc;6YW$k^;`=1b$T@Y(7J1l6yl5Jh{tw{MzHhX%f}2A*f*eAdwsXF$e_kyg;8eE{}M ze2-qWK`Scw7f56_)=GG#GGzCRU~=QjtM%aE2$TKPl?k|U{WKL`V9H}bO4VL2gt$+= zE*q9~fx)>DN9mb)h@PuvweD4=Efr&%MXBZnR*Z9p9<*4;n&m>t{p-#3K z99|Sb>km-sr^7}FYeKF5LKyoUUmY&|YixfR`_WarRDam5HfsA7WNk)Nxi(1NOY$J1X1-y{&b%( z@#rqK4Ic>PPQOc6&nS#UvMdtyAok%FifP}393|;SW$`YWHZCz1S)ope(L{KwA+?rx zBsE;nM?1>kbfcoV066f{gs+~}9t&L=oC47kQ;0{I)KkW3!&ctp`9T@k_tx^G9&f7s z8MT-f(9k&pmta}N5O9mo4@OY|PSf9|3fPsJG;Z5%kL(d{t}DP-kW{*qb0mpBE^PVv)+z*}voEWy>_Ag)&B9cREWI=p1 z``RLOVchk3$Z=R-n!iuuBX}!4v0|4qwt56})MjAP-Ts`p+#%j~xb#_-0Xx6BfhP_0 z(SD_AlVIG_>Z_%h(howU{kZ_veZ_+A0^%|bD*X|c`KZn>DRg0VO*v7jTi3zQ!MjLN z87-RbAAf|e6&lU_tm#E2$QM_Ot<>4S$QTe5xS40*R`L)3O<>&e3NZVwD&rVuNt43A>pdHxnE=_ z^*sq?&s?ELa`!dwfyw5iTqJ{51}nh~rR59t;WK_lPa0O!A2ed^9Pn2eTD{bH6QNqd zQL{T0p1BX;DgU<4v4EFNCpbxVtZ~P(Glxqfqnq7H6}{A4JKSa!Aaq0#!=5ng!Q5!P1Tq0M(rRseLUtv_GO~%bFWs)mpew? z3*8pU+R|tzeGt>^1bV;{KyMQzLtiF593T$vF;OVtSvmE^0iGB!24T7d*^Oo>KzdC< zNcnRZAPXlM{?UggohMCQwDMA(Qtt`c9zEjbp|`be`EFXgNl4dL>=OPVw!xt43zoFs z*)+gFe};flJ955cd%Z#QL44zMtv(N}zRDb(FdT{?CI5j9awXJCoMZQze}f0~1BaN2 zZ3r`ZAw#}A+ix?Z94OYy`R17@XmuU6%CfHX+J8)xabXtzymh)D3M`_|TsRTf53({v z_%v1Zwy#A_#^ekWVKzZlMf>cn!YZCDCqU5~et||gWZY7UDPzaH*$^%@kt_00hY(^! zEx!RNrhcCA`8E1}1dF}F0gtz<(cfcumsG_%_e`3$py%)Lwr~~Td?n+a!F_~Jmm$MW zTLH1}^_P5405hK1TJTMNOX61VTocIB6kR19FZ$2-%$ZmF+~l8-$EPHj9{du%SU&R@_1Zna ztMN?~eJR{a8FquiR10t#!u_VardD@Wgr@DNTqrL;TodT}YF<37z{`&ErOlqFiUqW} zM62yFl>ndh!StH$`EGwwGrTRw^H|&^-AU87EHAD1s|jQBN)A&fQ8&Eb0bt7Xjmj<2 zffQuCcppt`VqA5bhm*cZAQEing>rwUU;H&eu`#=p+NtyFN>E|XsBbr8QwRQZ1D(Ox!=Pv5gfjzI@E>r9g+8A`)rw8Cvo zv4^E;GpKgvH$N>LHZdZqdnSO6a;uzM5Y9nR-BFt`B5vI#?nvcYuFqljW_v61lpRa@ zUeEMh?;!-h;d8+<^Nrk?JsQ4q(LTV}_13&eLcd_9tytXZWb;f?+aZJ!j~7$!=DWEn zBj}Z%>RH{4UzBvs;|clTjLA_JuIp<69EEPBk>XmlOLkU_5@om}otzX4DhryOVcDrZ z@@}C#&b18Br5xFOnQ{Syvx^9MU94*{OGU2f>Jdg&cQ?obpdbW0&GGWKJRui3fXU{n zzAFrO>4qzu{v<7To>tC8jS_iXA+iy8XYhWfRgNJ*^I5Br@G|?j#Y4wpV_|3G9+Nv1 z>_Vg4kts?|N({7{yz50ToEJ$T40$|=GGgnjY0P*AZ4*IMk?fyszX;TE;eA63%&p3ToI$E%S#B0NVW0x2oXn*j8JNj6RehpfdP z-J785n5S6r1A=}8;@W8{v*pu6bqUX<(RSXZMBnCCRU7~mpYOzxr&$W)B5o~I-6Pig z2qr|Asl3&S^3X)rlg)$W@x~%a9qfq=A2g);E6zXemzy@k$yB^tj-wg^-P|)(CnhX99HNX=on>{Xhdg>-TbJ-O zsDB=LEeao}b6r0lAK@d_jLGEZXP`$@T-Vy(IMQ>q7PGsp^*(Qp_9H4<0dVFfEDwrm|x*rDVq#OAbAxaH%(yQg#N=F;6KU8rSs|@ew|(-#$9O#eKDj zkns^t*&Bl@Dz`+%#^_DKV3~(TZt-IG15BRZZYy)SVQ-Z^-J-gvf>5<{=BP4rWi!MS zL5r*i*(_~ZbgI>wA#m(J-k?V0n@zR4L&>YK?j_<%u7)7olzA(pc_0zEoD;xN_x0Mr#E84KYcc@cDZnp5zW1@BLxGn| zuGG>v7MqVdKqCltX=!^ZTVhos74em{VFUH`{c1-79V^zlz&3d*BJ z;7BtPfk-ZRtygR=ek>zHSj?}_UN866&W0#-icAvDC!hqQkK)I?z>V_ z$xkYia8BW}Xsq|?jCXs|WWfv2anWfdwC~_zpJ~+)!ykRydj~BSDfRKUznF7Z&8>+UrRp-^A1pp6m{kFjwTHCr zuX{9lE}EX^lcGOw9^)$)e!@>ZRlb=Z$5BZYfj6FS4g5=Z*;CfA<6ToDQi*IH5w)Ik zVaB>i1IwHA#<__U__0^4+IJ>jrc%JL<#){dWJ-=3BW5vFOQDbi#qJ>#9?W@IdEitU zc9_OQmSwHJwcZ0%*g(<#W<&Jlz<^ouNs=BTK-$s6?7qaW$)tv)uJPq%{$u7-%J-NBWq2z36M z3cUHEopWhlw}Yy#si^@9(O3nTd@YaB@&SWS^y>6bSN#OtU}XsQOe{(YcNp%T#M%SdZ~xyji6QkmEWOmm@2^Ric#3nTeBT77kS?l-H4$+UR0I|VykSp zz$6TB`aComx+PaTRiET?*v9HE^7XqINv#%=jQjhLtwcDPJbjO0TrLgsYSz6qyr=;6 zRiRTcOdkIt_Id(`R+5r;UArZsk>t(^2GJR&9C;L{0VAYHfB|H?0%KWDLXLWiXTrtH z*}p)9Z$#zF@7vucudND4ch9@u##2`2y8>jKs<7piee-!$n$cJpyIf$1iY9l|XW#rD zuG=M9Q}(-}$yZ|Jr4_(lI^^RfJVE@X-={2#Cxr9Z6+=m3P0Iw>|J!%cHg2vLHr9pI zN^PjG&%62d7~VoUWbT5ESYb|4Xv0~0H-Ob*V|Nu6fnbVXM46{Fc2*0QW}RRWKDI_W z$~t>2_7vAq`n;=RRdIud9rPB>QkWA~#lm&}+UI=6%{V38iOlJo;uq+5!EmMSk(~p1 zo4-3ZhA2iRz}dUy;*@Kh#LH@X*6%EaOEmn}y3+e5D&Nb@k;OC$eMb*b#6Fica6-#Y zk{!jRgEq(tZm5ly9zGM3%SR-VFbQ<^ctdOVFVTaUjU55rVD}PBm0_Gu$j-%5plz9M zvdq~^fGWC$>#3xFQCrwhB|W1|7tPMTPRqM_67;w+A6c`ni8?gIEBk~j7Y9N$j|gZ)wphnFHCORl*Xwz6c6clhklPE6~I2sbA9gD+D)u<@%`YvDUKk(cLSzI zq7sI!C;BqVti6UOT&tV@JO}8!U6XEXt#H3`LG(EX0sr>5NmBGadHN+yDQz5yNaI@e z=0-E5&O34}jn+GKCQ-u>5y;%rAl3ua_~eCem#s8Q3l&Y+Jx>uLoy*7dW>OWjcJ-QC z+QSBSBn036FZSL#D6Vbm7j2v%Atbm52th({_Ygd|TX3gwC%A;*PH=*|1a}D1NYF+b zcX!wQ7JHv_zVF;~zPj>Wz5A-({e$j;qPo{wbIvix9OEZBhbxXy3UzJZ+1J^9YNKpT zjtR65!L2$?L)!LGX`>3(JTId1L$6ccP>h=%RRcI~IG_*(GUl8cHM{`C(V;IU;=*UK z&B3fZPnnwU))yYcQ4u>D@$lrCh%OR+r<^WkSz9AYxd=iwmOoXyf&Iy`-tN;P<k9ptIFaZV@qI%^?`meush(FmaAMrugZ(t!hm(;1Xfj$=@4p9m_JAGq8rQ}y?%?OB#6Km!YtBUODf`K&beRxDR&=h;Up&ktQdN|(i29;NMS)t7?%zio( z=kyOtuy)#PYNADm89S(0n<{_Lq!2)@zmApkmYi0An?z;W?z2Tos%4@*;cBvQ0{#&^ zU+6Av7LbSPO5enia!#(U+r*mu@>6p2f3~SD@{;W!T?)qLw+9i?CB=b@D_;COn`=5S zKdf2V;X)o{r3g?Z|E_!WFSxXShgJK3`EUP)Olo{hd`&kF@bZMkhx_CIP_q7?EM)xu z`Lq7Via4Xee1jImhTJ0?^098&uZ{Kdn$(4ysj-N@zh+r}*f%WKD@?RHny){u>bN-@&l|=lwqr!~f=4eE;`q ztOI}=t0CHxB|L$xHt#NuF9{WVLbO+o>n#Xmvi4!#s;b>POu z?J12FeI?NylWghvwvB;*o0F>nHa9x{Z)DW7Vzu#_*|1vWed+jMfgOu&GLgMhB6rU0_B ze$b|EGf^VW!=zujv6>Og5zVG=Rk`oA^IF<8Us9gW3q!U6M35}bq?7SWIwZ+Hs(gl& zh7;3d#C&$F_d?%5kd;_EIOv#)H3zY*G!y4!b!67D#DuDV#OF#VdH?ssIC5tc`3ezX zVih%-)C!GV2X_`5DLw`avz^e|Ld!%CV}nEOxGNQBArU}NRGrIeHJx=+#yG^4Z5@310sn6U-=uefhdtUG50lIv6=dR0VRL z3as*+Oh*-)v}cBor;pBE*8v})rGVF>;3QhxF>hRtjo3=-YZpmM9s(gZD0XY32q-3Q zJYA=*d>~~p?YA(3``b~&GNx}e!Eu9Us3^WYaxlVAo)02+mIyNyhLq(hLcNzy_514Xmm#p7d!$W zLM_nFc4_K@m3(DffAZPeD;(pAI-;_~oEQKf%KB$#QxcPiKH@M3=<|8jKg$Rx5)!5_ z%V;ck>OS-s@0YS<=7FpgRc(&I)|6-EX_QaL9TCl`e@3%qVa5)$yZOnLYxn4#9Do5? zo#`8*g_hR(s*j%j!Y-)He1P?`Do7{0a{UA^4rftDX7u14&H>7v(r2qC=dc^4M$-;% zk+kQF6L3|s7zMr(xJY>mt>=`UYV%*C3Thpiz!zmfQw5CB19J0@3;ueMPtS)#`+iQH z-|c(}(J4(&2!3{}7Jk-cg`MJp8a9z4EGsFA@YYWQeHHvB?fs@IKQg_`J|-DD`u$e_En}a{zJK9 z>fsT=$a&e<7)nvEV|x4uXq!O*StmAB(ufA3%R?NC0WdYup~0w~;KW#=TS`oB$n z+@&*$5@raiJ|ST!)M+_-9aObZ&KB;xR?Ey6N>6x5Q5!O1*NN>Br!nT1Dite2{#}_L zelRE(`2iUEQ=OQMcMuN@Z1QyHPA5T=o1B4MI*PfLyQKrh`hp)AWaDf1pLM38jnNc}Yn9T2)+%A4E%VYi`01 z&t3ZFQ#4>O#j9= zRK#T#CH%82%+Udl7xr@HMqTZD7g)E~Lcgcwg^_T3blm*kOl8v_GLE4&tz9=B+^B2f zPOxdCSGx3y@7ELHbX1EuLb`8-y_#n#WIN?x3!xfY5hg(aR2creO7POOrF#cz5Ltu+ zd4CVu^+y$qvFx9v&gnu5KJ+MEQOgDa*Q!zVa}UjT{IPH58Zhi6YT@6HDd^0LSs>^P z17r+Krh093kuL{7DS!N&xARay8u%a##LW)9R~GM9Tv>~jS(NL_tq?h59Mq!?xgq1e z?45|lPI;xWoU44YXz1S>K8Cyso$?2Stlvf*HU*Q53cnmvS#GOu%&;>5wgd;)a8X>} zG?i)aHjZsm{*bmm3)7N;w@dZsgac#d=AVE=hRi9Bo&bgV;J3~ixevz({@nLFelugL z?~Y!#1&jpnIYx@}su?d*kLySGY~s&NWIeGQqKPL|J*`SS`vNbANwMr7-AZsU{$wuU z_I|cHft3Ejofjc6iLswJ6ig99*6$XW|bt7lbi-Fpn zFp-m@;@~?4anodf%1EhgTEaK%tn~{NYvYxxCp2;I{Fyf#>y6FUJ~codwNjt8)m*l=iL`(+ijk#OJm5Y(hlLJS0*e#x?~7C836Y-xc-~<0{L=R@jK-%e%uirk zabLfyDl!5TQEZ$&HEK562d(1*JGQumR)mQz2okAY2eKX^ym4Sf8E^E#vXZjLadCZq zWY^P370I)AUPh}d`w08qiv_!cU;MsSU@NT;R@A)uXM`a-qpe(9Tz}9&|15w#!ttHX zE?QX9o^|(jry9}^^P8Wt&Y&JQitje&MnUqAt$}M515~?3sEt`Gfi#5f9D}yB1lSg# zO9SbDZHtM#Wsx~CmNzE5x%yGR{P1TJaGIlwQK*lRL~PwHoEx;IW5Ihh0io;ns)Zd! zM0W28c#}HTY}g<>+D-f^#xadhs?!O5uEBc2_>P(E>;1k}vt1=wm zzJG?HJ(#dpX^aKQ1KEtJ9@vx}y?!;v?Ih4+x?pt+&do=082%Hhqkue#D>raE^~l6K zc}2T?Np)PvRO^Y!9oXoTcCJdu4p$&%sT4sWGJyVqyRw0~HXqvwt#i zwDR`KQ>rt*y;3j^1x;sH@Bb5^Obf{8i?6uUF71*Jbts>2-Ky~dcM7Y)Zd5aQ?TUc_ zu|%}>CoqwdsAY6)!vpd;cAq1zv!Zf-vAnrHgGJq>^3G^FzeX{Y>ekx!u;IS}b<1TQ zUvm4@@eiGPr4IJa`{0lJ0?9&UQ^0=qNvg4!s7XTyM^i5hn=R%L786N z9I@z^h&GDmpKSp!U&e705p>{tR%`wMVSi*QkLc<%bRkpdwc|OMtF>4x0_ZED9ey2B zLg%kOJ&z2{@hqI(=_5P%WtO!oKZ;gjR&H@MeuK@T`RDmV5;K&x*wuCFl2V92W|LX0 zVnz%9Bw8;J8p69_t2DBf3tKvf!_Jjleyx67-a+xFG_mik>jV(yp=7S$m1P zqA=!sEq7#x!@r<;bBrIL3gP|ccfQM! zf!;W%&0@Hjf(2$)_Ll&O!upKYzQ*Uakno3?k=7o}WPWw=?dy?+dDM~dY##tUIk-Xi z91fsw*=`9(?9vqv?2@9=ivuaPhQoE7^tayzzf(Qpb^HM2>jkdX)ZyYKwm`{O(Va(J zwZFFbuQwwf72|+SePJ-Yk?_3qj#R5l;V&vDu-R)=58<|d3x_hR$XLf(_)RHO$8sxw zDaDp>uw%em>ekZj{)OaFI{$v7gG2D=JInymkL0x2R~}pS(!!QNt}&iiz1dO&1yjYG zR(&aKAUuHVs1%&wCD^j~&Bp@dj&+0`02G^r38UxEi*%bZ*I;Gtqx^Z~yzpmG8H@y2 zWm%Y^^bwJOb?FaJ3C#M!ZHX22LV|?>RK)O%_}xtoy_@Gd?Z@rxi9DTw*Xp#|`6s?s;AA7E!>2*_0MWuts0%LAjH}iK8|Fj+ndMxtdBHc+KQl{O zcr6B`7OA{kbiq^6Sa`n?vPtI)8OaR9=!3Iqhjn{jwvhrHxExtm$%&Rg`pDd4B!jo# zoW913ZN9*Q1Hjckio(PzlMa%Z*ES&qwIE$Jb%j0=lpmyYlgYR}36b$h*c;$6oG2Z# z1T7#lciB))i&fzoVDptA%mU_5YuRM$I;E%S^RErXuCSiqv|oI{@EkuSkJ1dd&JQd< zUNOX?hnz%;2(SvF!Tx5`0npC>y_drV;~Pd6B;W<`Y!L%U(f`w`DsUSxv5SN7E98Nh zc^fKj#5F>QG_+CHu^WrzmB^*J%KAF?_1W|cTJ?%#Zsh9OZ>7e$@NicS7!>>0bsa}A zI<0chr@~uMcGH3AqsZ0Y0Fd1$;+X;m*-|!nDX-FV%CQPIG+*weaXJ!fWlshTmabOkZ2XPlVndjbrWIXHNOmof=Tu{o|;*s(@B@tbiT;rCXs$c=en9ee(L#9z&Q_eF{63d~4mCoE09( z`MMV^yp`KDK<1h~Xw%prDcojld^F5z)vSrI)cs)PQzQD8A|rtNlAEevyka{sDJdmv zFR6cH-2k890N}-fhs%n0h{7|$q8!y;zd<0tZ!^$n=Spk&40XrAG_bWWJPhB{TGRokS1Ra3A>a8jer5LE^t5FmtgVXE88|@<;CBg3ikZG48W3uM1q6SjD^Zjg59BnL#OJ`aqca6ewAbnK4;zoQMv~COoD^LdMF{Nen)!Cg&V5me$}5Qy zOID!oxDWIlQ7WWC-zv^M(PzB}Uk%x9ZJqiVWkPP%Jak$i-@vAvN8naMyEPXkWwVaQ#%{!JU^>fW41; zEnkG(xJE%u)UdBTCGAZ+TF-<@!}MC@IBZz*QU6nH`Dg=hY2Epar94EoC#>|)ZM9D4oSwXFzzPQ6%r4QlD`hec`f;0?iY|Bpp zQ8LvWf?&Q{;$}ktl?W^jZ+?u;{!OK$%KmY^1(1ic`ho2++# z?JjA8fFUkN=mp5L zTtCv!N(&+pfBPBL2bXczJH~|;Z`WBp`*2qhC$gC=G}BVB zx?>hw)L5UbFs#04+NMb+XLKoW>Hdbe@e_C#nFnFw68BB(-%&{Mf1!{-nm_Rhy2jAQ z5d{?%;((zoL9Q~5MhCMFTVx|3Sbixu$<`Iz-0M<=Q6|JwrWZt}7a2_1wB*%&l}|7Q z(IOh4tnCP69RCn9Yy8j*|Jz&d@R&RQpW*npMqO%GoF_VEQkMG$_Ej zs&|0}di40a=f3YZB9*@L$&?07w?hwV3BH8IbnqbBNsL$W%fF69mtkuhmjc-do*_ZV zm6c?Qchz?p&j&5^9mI_j>~MVOn@cUApS8rnb6j}qysvsCA86r|E^F&!F7@{R&aB8i zhNLArh|u%WOS+&C@X2yG?^R}K@0__WJE5O$?d@n2+0G3Wf+cKIJedq;6p(H=+1qch zEQOBIcJ^YU7j}4CduHm~R~iwJSz=MMP+LTK)8&R?<}eKvLJV_Ojmxz;@&p;zVi8x% z%Zh^e0|vjlwv6w$stcwAn^F`LrN}#FK*|pAs52&YFXm>AShQw7z2(OoJS_rvvA2bA zurFNd-Xxr{$WkusMs^VBF$7l*qN+6@gAThqU(|#?|k-aq0@XNo$M+3snq7btD0>njF}ep zfLZe|tnqvw%WYU{UfDeuC$yt@fOGi{a3Y#2fu;8rSiDn)No)yo$KZTf3NqYfod-%> z7GN`KA7&aUb#omL1r!S%$hYJSVk>65>fgmn(mN_(g;qNjO`^?AnjfZk&uP8|6im2e zVPad^0KN&vsQhKk0EyC3kRVWd-tq;L#rxF}-K}b2Evek&E=NthIR$_(H0hPd!T~KfOl{A@cJWTU zet`QI$gk1N{v9#CwP@bbCALQ(*V_r<8@9yIwc7g}#@==km&7Rti8FZP1eYM7n4o)t zb^pa;>EgKkw=zd9zDHSbCev;Hn;DmI%5zy(KVE9ei&M}t z)0$E@#a(22a`dR{xPhF^bjUU>+~UjnZ|qC_aIG`zch5n2D%w7{h7>rm$l7F$fIx?8 z1!mVfA|(l4c5A{0gMjJr@5i!0lEUs7X!}kiV&=wigm6cg5Fg(H_#@6w9g#O3ejj=Z zo9m8wdfr!k;ULzHnbmpMClc}7tyeH(GRMo-ymE-RQRU9}ytyq~Ppy?WFRi4PbYvJmAEBkZvOBMq@t{9nbD|yz=aZ1V;-x<=q zqu~M>!Fhl1ZcA_!g>Kb^Y95|!&DUK)Y=N#}LdWb<_HL4E+?^BM3BB(qun>zN}DA#I}2gQq8GL}oHWA^8OY5}P81mU3+wxf`F zePcHib;xw{z#t059X>IlXn1NW%d-h55L)l?Mc&>5aq9<8QA8^YgJ5p51fS2f4bgL- zMaa`H&E}0rA*4d|pggaVHw%H4+B3h$z7G?|S-wZn`<01+cl_<`qg&5Lk24;APVVLJUGa& z)92Nuj%2-w4=}3Xv+E+@UZecE?@X#Ru)x0aj@q4y829S}h0_)RkOTcg+-Uk%AZWkc z@3+m!iJWVR%pIE}? z@a&DAQ7>V?!HaVqtSi0fAS}z#X`l5sfKbV_?QX3ia*k_-mY%%L*hgIdaIdj|t7dpF z2GP%TjgD5HR)h&<#`6l!RFRnpAo*3o#U;t)oaJczGww&#go0g{OEvqs9s~P`9BPTH zs4^?=$NBN~Z)Yvf9^Ex!)g0mgxQ;C3ZL!`W=M52RywgHw zb=@rH4#CN3HSr_>#kwi~#UID$uHg?*2$sim)^o$VM{z5xEAkr--4pzKzj8sKm;M1) zSu=Q-_)b?1M#cPEyQgp=1NZaH2iFQnLqsrI_C7oggfEsu zhmxgbm};2vign=zxD&p7W*p50C2xY7xM6D4h-555xN&)%>>h1DsB-9ebaTTIrt?7h zJ6hou_V!*JrO-j@KxDCb91l;QV0zENx~2(0)yaA*RB|50@cV#=z1rw!b9`lYJC2^* zp%!qNl?KCa}m2e8Fi?>lK`$#ZyLUiys|vAdgn_Dk+acW>``4Ex(@Weu%^ zz$-h5K<4e8>0-v?QdSnGdTo8|Jd72KHUc0QH?}`3NV0V#*R96Bfm6n`ymh+|u{P26 zXF0>51TeIUE2)|X6b+#(n(`@jlD*=*K!a2s(DZ<`PCtC%I>r6)%e_Tlv(wZ1Fnl9$ zEPUN_D)TGQTlGvxAYz5G=mMxm$MeGM$qNS7Q6QvCcHCE>ZAb#nJw!Sws-1~Pa=kmK zXDrchzKz2l^zu754$}+xGyz?lyqpK|E> zOe6gPnzdv^lW7&QiE*@E!7zjp%t2&K4J;ss-mF4!c=smvH1$7mt^}i2m?fWpI^h;65-@4rU*MJ+oW}`ZADT#kAJg?@W$k+_J~8DPjZ4^_?3b6Th-M7Du{dT3 zDC14^AcpT_KI=@yK5gxH98MQ~hs|RbR>^Hc)W-|BuX6}H$>qyH7Q3ewFKXBx$_lu^vYZ0fV-}T`tQbK_6+c2o&!N5t5JZtsI@*i=^4;;@`v?~J`Pl7d^3)Y%lzGJYMCR1anA6&!r)P`=h zgwh?bY|U|=kNDV*stO(KwPxsDep&S91>ZPcoM%1bXW1&xFi6+suYWBXPTGR4c&+^? zU?a4*(jHw_CUnqlqFe|*i&AK9MRJ50>F#hTX~=-7AnbFoVuU(eB%>(0I_2frXIfQI zVBb&;c>Nhd2*_mZmSj$q+T_M!igzFgRNb{N>Pds98msG@>o}{K4*PmhY(IGzRytqf zO?PcBJYyj1A>8wI%a*l#T%|%vi0|jKPtu{k+srJ}{wx@<&S{ppzEI=Lvjuvd83fX6 z!9$TDhe^CR&^eWW=9&6fb#heIR#l~FgdUL7yu#(IHL`ck|3D#o^h}-B)0ZXG?6PNL z!pF4cXCjZ}ToS>@tC3m|TbDupXt*Zieb%jM(a!Wp*4#swLaj5J{E>o*{8SCU0MY~z z-$#|6_uA(6;bik3Dt3G7qi<&~(CIP^dX02Sd@?OTljH5xM0ly6di#kP!!K^h5{75w zxxTi{FoDy@(kfF}(xN=ZN3Cu2?_?+XDDGrzH2wfJ@c@ZcB%_{EH6wC?1$8*qz;U?> zpzZS`uRl+R(B^y}AltAJq6qxtQ)kMvWXbF4IiISQe;50L5Z_L5=irf`tn6PSZ1FY| zJ9<$|1JUZZ=^}qsYyqHDpmcgrUW=~Xtj)6;%LO-+&zg~@mRJjRs;`5^zj*jT(Ps`h zL%2UL(x2eqyK8tHv6;FYyn-qEOJv~+<~bJ4&Pxc?eDGQ2?qPVon!5yO{Q?aJaRtxiLuA$tJqZ!ut$0K&0-;JjIm1-EHrKe$i;&*OyC9 zm)N%Sc_qSr>j6|im=ol1%)-f~`K1t{u#QS9o>xVG5>9Cm%=QzPOz#LpHlBizVdXt- z_l`E!$C!98-f0<~qhN}ZB{}1CBF-~{m0<3UtmmPP_J!t`_>Snb85Z;Rj|z+<6M&M> zE<+e}vtjgOq9aV^z0S{mg5e`+kTpJ*(6=A)bG7x!*oa(6l~~l?aCh96nM`|^Nxl&x}Y@a``SV+6QodWI+j zr3(JY@z0rK!y^5dJGAu|V*HqUe(%9E#i2W)&ic*e@hhqu`PI87ckN8X( z^QKv~lTIZZR$eqP&0s*rASSbDGk5%5fh`tCe5hf}lvo#r5l@dA^nxLRbCUhWnub-P zW2RB`pk@6X5OYnDUMhhR^?5uG}KsG@e$^R1Qfb)i?Q5YH@0QV(g?M3HO+Au)Io&uB}avFQTGT8pi-iTN%H6@kJyWj<2GLkpe@o z@*P!b9zJd-{(>mhxcz!{*F&g$QAUb}-ryup%ap)1Qp4SgWgR6QOlo;a`#9ob#P*x} z2A_Uu3+BttNDch~-E11Ks>WEycvu#|0yLLA$Z~M=>QlgPQ#7R*>GMUls>;&Q>YZKkf%I}Q)_0=8 zraSRgM;nS-x0f^+Ch=fy(jbtF>TkEdBoe0OH9Q`|uj)kib69_^ijg${FsxTTMLt*Z zr=?hEK<9*1Xwh~{!Ncgx9N z#5~!kP-kB-7;SK)kDg~?dd@Yx1E*|y<4SCL!fD(Jq(KLqQ%~+@B@UkzxI4{r&|Dg1xfe5i}2Eq@!DKPp#B%kC+6Hgie%w zIM=a-)6{c30d|ovv8#TK%*C=saO{J>gmpzl0g-6^=SacKT$}QNoZp=biMP){lfgfgk4*Qpoa!${45)@fzRJx~)hb14 zx0!d7>!%)YLl%CSO$!8pL{>t?T6=%MpGPX>Y;&RQl_dUaMah{R4nnL^{zYR|0X35O zS}+U5VDz~*bW7%YAF`p+t`}NpZ_`)Bu5=LqV%3#1rLS ztSpjdY{&(<8a2^7c2X>F88_%`ybJ1gRL_$?%VW>;1kT+=um9Mabnkhe?&+8_hso&U;!LMJ@{R%$`*wMxx7%}3d=x3<}dC?*;iLtc#;lr@{s{+jYYF-K@sQjxe znO9!;nVYdj0?R;O7Y57lsxSv;Vu_=L73nwV=9!U#FYS#S_pmu!8<8njivRcfm{$3u zhxx5gATKSONYcv9VQtA15&8up0R&{zvqU^`56+XS;w{k9);0vZcMg$0hPh^a{-lJQ zG4Aq|O?=%ECWcznK|hOmJN)b!tso=|uC!rXZ8%Y}qQxkBzE74+*rQY5Jjp>kv&fw> zH7=;_2A4T%L_QFJxO5!%?2;>K-N;2y=N1q~C!-DK8e5Nvw+Sy&-svMB$eK>;Um2rv zHXsy=WDxDfmg0=zSNP$=XIISXeY{VfSYgYbK_*ERk^BpC>KE6`r-0=1GJHEa>Jorv ziExZVENn?DtN$#Lu7#^%g)nZVN(OTqnd$E&D8@2l8R!at96pmJXN{FhZuPlyKB8fo zZF_{PB`#XjWLTI`*AyqCWc1d3;R92 z>~#|e$wK{+o;RJHd(LHGqgp?;JC*Ebu%;?xa{t^~+OCe;Z=B-oZucTgzc3azn74?_t%@@QK z%t}|2aY;6){s9sS(wefisi?_(Z3lfnMPWw}ndKu5D#EL>SVsr*4glQ}b66$~9Y2ls zouDxlv8g^v0YYdDM@i!gZJ^LzI@5u{9Cm%HbZxr0x;SjyC=P5MXBcLFeV18ikJ$A= z)!Nrco&6~aZ>4;;@Id+Rf@d~tP-qlGLAkYAffjRwkqOrvJH{Un&H+UggA&L z7d=0SNkh?Ke0{j|~u@t3A~g1;!!LK@Amjn@pXr+^jCoN^g13 zIK|28&JyAKWtLzKxB8!II5==YS@d0AMcU#lMAB$cpSJGl+K3`-i1w7@P1WPey+9E7 z(w!Hwcm+jLl&mF92h{rHt>eY09E>jy&3*k3_sV!fow&}jd`u=$67#NT}+{x`gL zG=>3j&NB>ObM=e&&h=cBT!LmPhpw7is>(AAY-U@==TcNtO`&ulz{QIG!>8fD^g75p`T$p7^s$$YXO z{<}#C+#ml79wh(d*I+omG=f)crQdM6sm2K5M}VGYiBAHw;a@;+#b%=tn%5LS47Fi{ z#v*;LcVujF;8GKoU!}$7z){vrz^S$a_1`M6OsF%E{V3hd?a!mO*ADPKy|DJgIxZ@! z<=eCBhF%0Jn5^u*+bL+8}|2UNM>E#@9H=|NhE)?=#HJ)^bZpQBA7L-l!De;Xg%FwFa- za`^%hcg0=;`(bar6W~s8!@xfLO+!NgMuR~r$2%x7`+o~(GiCp4|5ZEpr zo&z`z3q{SQ#Ox+(`@a76-}dArMnaFp}x5={cIt zSTcCM>9$Pu9)8rD#(~TG@6VU7en7uli-AvO0$;iIPKl;7wn1=G```=^*u~_FVi;J0 zrj1fr{$u|A^Ecq;0ql#+qNVawbDLzYAPRl^sR}Av;l(T>z?UT~duhZ6XGD;VBoBEk z5`??%2z*`P|6o9=dR*YTT)OljQ;%ye;2EaHt`ajHDlFv< zC$j&P93Q<93O|*NF!SQHL#%o z;jzlBJww1|NnXd@p;zN%+w_MB6f8GI%HWU1N6l9-2N<%0q@6} zZ9mfrnWlJw3b zisU5o_-W~=GK#MpSkp9w5$EWHpJ$oC=o8cV7Bsr-X}Ead?Z->yHo3{fc8ewNk>KBC z|E$sO2kzpqp!agvzGHc{*8CsNrHYCJJ(Omos#1XM#&%M*C6R~DNJMwcY54Q!wAf_; z2MIa5+Va)T3bQ`waNhw5fxYnWvA4U z%Ty2-vA?k3cR*`UQKk0T;#99yNL3R%wOIjrB#{c6+{_{G~ zXu`d}jYyTZkLVn@&p`kI@JA6b+_1}=KS1@!>5zmC+Z{lm*)kMT*OsDdxZ6{;@vB%p z(jC6;Ps^Ks{Y;G}j{B=$Yr~MkJlY0=Wz`4{XAUW$nq-2*8%`e=KcYV;83U)Rn9;;4 z)Wwk=e__T*pdHOSn`x+s$U-%V>c~TZ@&T9Rvsq%B4cD>_fM|L z%g=>)8kBOMU;owhYc_@M8jC%o^tE%1G~d+%iyBAGrOf$FVmjjl_oA^+YK;66~kc&)Dp~B4%*E9eqBrZP*^BmUY`k8 z%fp49u%`!*asNk04F_lmQEB^3w_ht`D{W4mGYo1Tn8swMh!E$bU6ctpTyC-HZifpEfrDOgm6p-CTWSBpa81>u?kUQE2lo*4=Tia5?ZQ(3NU? z-Iy@<+IO!p* zr{^)!G&E1mDN;V?SAtN=5DgMyl(M3p=j%uwcbxWS7*WRJ=i9@ghn2>?M7vJBB6nx= z>p!^k<8z~A0DZdRy8T*NHWjLViY5o#U;;0p+X;)5*19}UG&b=cAROaUp#%fV3M-l_ zWA=yqWvR3H$0JhvgIi_n_w9`H9tokEb);9c=lvc1WRvWR+PXuR(X0gZh_^=L zCD;BdU}s&ym9|~uG!OgC@g|-NbQw=&ms7fMjBv8HU%qTL2YMr8a|&r+M*T^iv#R)f zW?&tWOH~il+f`?!i;I_>GgkyqMYHluU*F@C3-p^QUSoK>6*__0KD1YhQa;{@Q|oq< zL|Km_XC!kBW#rxpbmtq-w!H=5cs?VR!OD(hyN;T7wLi*D{AD2ZdW#rz!+-ENt zDCCHvubHgO9O-SEgcu3*F9TLMcvjnZXNrwUL(w*H#Bd@zexcTNM~*A(k*^c6OW+=l zga+!IO71+HQ8&~PEPcE3$B{r>Qc+5N#`W!jhCd$WlQ)tU3rUGKPc>;KZ8Y9{fBaUN z<*k}$RV?PU^$;V2%DvD#B{(^>!>Q)cjeMo@eTYU*p$^_}pCuI3ocHOnsisLXZRGjT zw#Ise9Gen=5bK)+`p!D|fbdP}1G`4WHzEOkxR)Q*BLh!TMr(lrWbxi(1JmQ6FjCZg z`1I1nkWlPc9)Ds=?P4;+%Yv&aGw7&K#stv(=^OF_J_mZkNVI)NI+2lF256@ok5q;k?86^kiGr(;6W`Ai5zD}{7XstQL>2&LuNdN z3t5Z0ypaKEAehAFkYHhbzRph4dsi;^4ufciDwapx=@WrRBv7GitRFk9FpC95 zNCVv;U>|1`1g^`&*PkO>g^ZUT)P7$)QjH&^S^ItkMnkCtRFK(cKM!hv>E5z+^f$Va zBv92^cH00)_;@L@wferDxq8dNpW3MRpJFUJP{%0?Y^nISGRJESpNhA7w^x8Y13gYt z^9klgnj^a+_(P981*nB7-;%G8iQQ_OqO!q7rX`=6uP%e^JVqT2^Cz#E7@cu=CkPLGGm7@{F`{+bu@?h z_QR&A-_eW`LV7hce{u!?HsN=u@y1I|7>-Mchw(J0N8I++%j7uI>$D39&~p+fVH1b4 zGu*MWG9ASmip`7kT@G37Of}(Xg-Pfa41aT{h|S?Q5}@-FqiSxQX0jeP4P}b2QXU48 zFy&iJW*`wh>Ak7 z)q;>{A5`EFGVp_!IGb@e!ToItDqng%71kHDiHSp}>qI`Vm_uB84%Nom#AUgw3e@^M z)yOAbZj`%s5;J;n0xoCOHBo?-6SKCZ2gQ~&!fCR9Z&N!8tH*9Hm`uC!ybnP8O^r>K zB2~b;i&esXvEI&+wN1dqtn^cRy?1eW_c3aEn}JIlS}<9PTk)P+BRabHE8<dmoDFZ96Y`3ipwluha12`S8M}1mH4ZE3e@bQ**#vNicDJ7*0;gDCpY0DtyY*an zO^BE_bLglOm5;a#!l*DbYC-VOeu*I4rfmnxJ-rttQ0X*G`Xp**V6uGTPmi@90T08f zov%*=4PSoQ+KI$u3iMegcxd-n>H~x(zvgx^PC^N($v)1B6BEMZ7uQk__u@y5Bl#9S z5E3vX_k&J)<)wGG5n@hE(!l1nGMwwMAC9LN=l51-GSKYGFqW36P5OtkG$ZJE(e)3h zvGtB^^_!`l+%Z{O;*XudW6&L864Njv##^24(;BrswwF}P)t;Hce5%(PXTac(R_y z4lnD%#2j+S39s{cyaIYdCk2SAiaZYi;6F#dWq0!>S#`1Uc7gEgTHS% zkjTQ;&kcHxc`YJy5?DFEnWfWW$q4j|qLk#pTwpSrpiUfcr{mcFTgG<~p*8z%NN4gB z)`%rWTI%45%KJH4VaWR?_@PSUDy{>CPE&HO=NQUZKU#T;pDZh&@SA*+=J4z|aB2@y zcX_zog=TMPe%Zu!o;=`rg6!3usea(1w4&h<-*|^_+j}*2j21mv-flkE3Wa{ok&=9d zG`r&&)-St)aYRl%m~`dSc{W9P&$7k$6&N2q8Oq&XsA(sygKFZ@0k^9fvt%G~pi)L{ zEqA__7>Z@^_5qyDC2iDAAVdaJzpEqvFhl62GLZ!bb7Cn<=U419IhtO*a{5~xqwz*l zspijKkh*BpSuN~a{O0EX(iI^aQ7Nk$zDk}}tf_VFZCq&|+*g{X6`uCXbm5DO&5Q=L z1D<1u%yD>zM(w2R1a>XF@8GuT*NuXFk}L);8MO)5g)>M_lqzJ6yhNt52(i%?JI(gl zJ4J7`{+ywM8}B6k&K`Nwg6i4pc;rMa!5R|PYw}}h>DzM7=gmXTEd|C^t5C81kCGza zHtRJi7Hdte-`=&9yRIzs7>d)f3Hwsbtya%>Sv!A#a^u`}XK&(7>B zBT6y^a$LCMbSl@#w)x<3B@gHq+0X#p8?Gk*la2wQEpZQ-TIJJ zR7tk|2ry2N8h)D|DAn-USbxs=s=E%uO|011Jnd3s{C3mejzVZOULu2L{t0yHj?9DC zJCpCja^jx9az|xC*EBB(h5r`h}Q{Lm$?H#FzAwZBl!4`gQVwh(_7q3oN9gn39TY_jr_pz zr2kRIl6FW=t>PIxv>rP97YpFJ)Kwt=t%T zR_xCIk#n&kJ~aPigQ~9DxMLoqE{PT%Dc}0u ztX8U0yC>iD%H7=JAo2^eQrxl89(90j)pNwQa&EBBjiQ1J>mDO}V3|bh0zdz&UCpMQ zOWoGlOHxcpwYZ+FZr)*TUWPEJspi3UhnqG1?ezWa!35dRo4lOIm&WnxI+D-sd|-X6 zpf>zf;($w&oFxxU*BFtNAr+MD$F%kfqBotEc+LK}qzvpU85LhIH-YD&gZ z=5S87Ps?cII*-f9)&G{y{7P~m&`RDxE6zQ;oY`I5%psVzHcnL%vkoHaNj;%0{goj6 z?s-Ej%Nx}~J%U;ycR4yyBl+-EbCWUlhSizDXl1Wu6p?IeiTF zmbPW@uSftFoBY^nD}LLSg|S7@k^_hHWZ^AA9^2;d1=O~+3+_BW^Xd^&rn0QcMahNB za0B5@g*za|w-)vnD99(D=s>YIYQpTHCq-WP3xiDfpf$QFZ`BBNorO(7rEzzou3CMBeg<6mTfjxpT5F z3-!44uDU16#M4k4e+#|mBwgtD@`tp)=-q6A{kqPvi}jp!zrr-<@AuuB;(uI)w9cJe zYT=f4Z(09j{hWc8@x8YJ)>O+KT6eK~nry!H0Y5Ief24z)DN-KuZmot54d%~?J6v|Z z-dBIBA#He(=~&-`&dZWfX4u89PA%Jca;00!i(LzdEYu-RdSCu>bSgVETSmX5T{leS zfN6F7ZOUvokQt3o3<>Zg)kC*?f~y+Pd%*2Q{I*s&AiYD@CEQja1G z>3(*-m;H!8=4HCSojugrB__r_+gGGut(jK%d!y?4p}{~1W7?SgrP=#q4eN3~K2{HW zgKGUa>tc{&Y2GJC(H}hDM4*H7(Gpxs6@4}C-ZBStBB>Wj+O0t4yb9JMI17p9bz6f! zedys)V3F*Xzh~fuf4!0|?MpbEO%A|z=^BM#2s`({FKItA84IgvMi( zM^vAT;QjV7q=~Jv&Us(F${QuXjWGHggE+o>>}RVli`sr;rpIoYn~ohpm1U^M4(Z>@ z3{5q*d!vBWOQgMH2CrDO5sGZ`sL;Ldv7+%Tc$32OX3J&?@Z|#U$Aye&;Z^1+He#^p z)3%ioL`?ny`9T$^ql;5mM3w+XhGmLgE#)N^TJO}JJ)_&F{^?^2gQEUglMtg_G_Wy| zrR|f#`OIcBEYTVj3o5q|p{z^cRmVdBEVq&7)mgE8RK265$J=ecRV9`eLQNbk-~oMY zr!s|dqhVp9uhd+`fLuxwOzTI@br=G(edC3NpLZ57W+16EG}+A_xUn+GxKdDE(ef_w zf)3re4;btnOe4k##449e)&0p5t!{MG#H0Y(=uWO!7ozn^Fj%c6W6Xr1dLD9lR0!am5F5Fw!D5EYTQ9h|~ zSex~GlOg6AR1}v^`bD;T{Dt2eRoRbtC-%zk_aEs@k@#LhPQ+$XjfoqRU)Kwz$o0Yx@X6)ErsJ}9;qUC!Qgd()3n5gPwmt5W+47#FAY~PmlnwbY7npoe z-s!fU90h05@x?)y%kA!g*`!B!QHttPFmc|34Wm5DSYtBQw! z-@plL6|KJJkx<81B*UTwH{(YdCd7qaC&jV?*++H>ye)b?OY+?TAF8!z&# zpU|a{u{ob0WPn}|-UMrQuxys16LmobfPyN{8oa&EA(x7&7jEE@QOQWgyyPn50pT`N zvo7RwEm~W=AD)XAFB?yl?@6AP6o}ofr+Iq<;0EGBynXQx^PJB`sqcKGP_^|;b5Br` z&Ee{g?A#<}y83p#cUBsN2YZup54S!k)U!D`EVgjX?E**n&jM~G88^Anj#+>$q5V?- zmP4WZ!zOQg+xI+b%@!+y0}Ri(KUYI$RMw6}cGJS$6p^_?p`6&GFkesd`4i+_xKl)< z*txFm9Y62&m|+KBu?4vStHrx0jal-!Q+fuC{==mAR#GnkAC$%>xR@sEg@w&+WiSBH z_~b}hZoo51qEc#&0ro15VEfrW+8-;I3Qv;wAJc-)RzuWwQCMGg&oZcwO^q+0RINiJ z%jP5_emJrbR6c$?KL8IBMWI0lZd@wI%A=B1n z$YrIqLQC%VXR(2KpFM>3BSP)5F%!L2Mec9hwoJBj*rU5@SgeePGO~a6ITZr)Gtn?) z5%|ey7GSvcKJRqM4U-wlh7ca~){v#GaeY7>$6C-2I+ZP{wy$4Au9%>x@6bLKw|Th_ zkX$Z%&Bno-;WYMwU~TTN=FcW>lHS58N?@O|F#j!-ulX^cdfvysFFpuwBEcZDx!3PmzUEa31i zcz4uG;Bkt9)RO$^>ZYai*l6ILCL#gm>rxR%OFt5mqI=sTsB<8nI73iL7IV$vaC7fk zMiB(Kth%0DmE4j@nW|Yz+Dhwa?&EfC-sipaY0h-1LW{H&@Kkv=fNo6jd;nDE{vERr zSET&QD>{|4E0P@!#;)Sl@nU#nb(L$$)l+NVk$KXDY5W?G^kCM8=F~$TKiMybOvPSe zmyxjI)WwEs(d*`Z4mc-{@Cxian!TNg`{Egy6h%Lk=CjGIDPEWF{H3W|I989oUEMQEdj-fuNv3#9>;WKI3 z5!U|;a-DJJzJRG6_^d@c@cks8g#1>6G&SD_$xHUcFS?)U)8b)>p`OT`)-w=MbeFT) z^46Ti$G82Z*csh#w++pd5niKU0ZwdsagL)~fynY)sUW0Pf=~hh)WKkWPcrCx3cJI> zydR&u4R}rOwU9^h0z6CbmN3be@{PDNsP{B)?zO{&?TYI*cYI}{5opV`@k7WOOVsR0 zhI3s8hwq5&IcOuyXyM@Hu)N?C^f49I+l$uADc{lBUzww-D!PJHgR6Gi<=4*@?CYsE zZ3BYru{HFy()%@HkUxcy+OkF76~k+N_HR;_04HrbKUb{F03DG{@7l%=`DSBDA#zvl zeD!^0IJFb&g~6!jWT=oZ_x`&NTd?-+K*(JG?D$PuU(2v*yaL~G ze<(p5()8vnvqv*0I+MXB^vQi)TdB61&TI}|&k~K_O-?NKo~4Gvp>!rHw+CM3pXU?n zjs404d>%3Szxko#`r!4D3mo-Ra}1#twz*;POib0m@RoeDi}>i>g0qTcI>aY+asKYX zmWbCa7m~@7FjIB=NL`yOX8`X3s`yNw$uY?sZ{TYBC1E2KATdA>Ea1(#+tP>4%CznIb5m&w zxM`SC`=)d$u-%jOBnu9ISL(j*q|$x65?b$C%pI@MS5!!{D0!rxAVHjIvQW(3 z^jF2=7hhJT;bt!m(rhifoF`6D1zkFkswhgu6(N2C#)y-m45jv3q->&a!cegNrgJ6v zxDK5nqW#&i2}9jdA9fuoIV=%z0hzujcPLe2)M>86(iSkl;DKKHMX6Ho45fU`@xa1} zLwTPQUwufYzx7lCEai*58M~lLTUU5rFW&!sHdX{>Ldf zCvA{+SbLgXBEfb@2RarIE1<6-fen+;vOSSI9jHRIZ8m5Xh?6y&x}1x9bRijhGU$%bMeFpHb^iHWI?9PE-R1-@}UgqSYJmXif(D0r}Jq6e|vh-CDqF!=YiIOWG|#Aqmo&vBDmZ{q(5MEx%qyZ`S7kC2(sg+W8! zR12R)e~#4Cd_!a6&L(Hf*-~3QAj?o}^eL2UaUN;reD%cjc<%AB65*xpWbgfn4<%Xy z-MEZt4=AzT-cpKvqE{fFQ%CFj{J(dl~-xdBR|cJ$dfDq0&V4+E|t1 zSY{}QMKBj@zt>({?zV}s@luavi`;G?-Eeew>(e{8?vNxS+}X*$92t9Q;RR+H@H0C( zMK00*mJwHert`2K)(pS8KiJ9)!a+?!C*%g!W5rnwtmw7zK@6eg0I)>xu#ti9SV6eX za_mJ>+>MFeJ6^tS%|TeqBbm#*k(o$J!mU#avX;_Ev;=;=CXDwddOo`)$$pdy;Y+Kx zRs=dp`HnRu?UjB`l>^TJ+bzOVSEpB}xo%VGu{iDjg>Vzz zmyv(6)W^g?{zUeT-U27>N83WwLZBNghTEDxk>eBvv2B27`W@On8zf(pkhVhtBN|5K z%7d=XU^pU_chGJZ5`pH>;BZnB=S{m$5>DS*+*-)$BXi} zmG$NXvt3U5)zX}yDX1DfL80o&c^P~uFEpTxPgHK_!Z5tdYGnu_nU@YjB;-`)&U(`a z#+WO^7+ree?%ES3RH_T=&jd7$Z|lia!}y09COxH^&kw?1zLcQ-Y1)$E98cCoSc;f+t0kH(C4?U5#z&g2V2_!oyzc zl*gJsD;u@23}z+q)4U($8U$ORZl(BH5gt|d_W3bAPvdNW8#lbZ zd*17FZxDq_NCTd-BHd?Vv19$=gCx+m)By%@)WX{XXtiyuL#pQ&k7#y_Bh*QyuGVAX zcYdyGPJuuac{4}`JyL9E#d2z-2|rlyr@^5qFUu!kS%cEZ0;TK6qWi^5__Q7+E2sh= zbRYC+N{hJ)8PE2-HOI9$#y?hm@fL%m7nF;W)@*xn`I!IV7EWowO>SC_OTGgn`4-W=px790ur^GlwEi57)9V7LMS7{; zE)3O+B9sv{)Ur3yYg}Ka0l6bmjwe1tlQSA+GaR=r%E%6Cl2Tpk!`=hacm^n@uO+k0rf!VE7C8O;_B^f-cC$ zW5DT=llC1!bmKe!Hxu1I$&^%?gYg@Ad7xF+0t+R*96paV!STS%plza4%JG(5QosF^ zGNSp8q!-FaS6i-6l};^6g7R_gSAr6*wDga*r5U0Q4I_R29Hq};02@gU)&Q@EsN2jj zG%>1fV(in07W~&+(uCN-uQucC=G=i;C*4=6lJ++jg%lvEIUci?4!eP9a$J zgSP$yrrzH&?p0a-8f8hJXj)Vq!Zi+uH8sR2IEM=}L>d6I7U;XH&M`Tn7x67E92`d# z)c-APvEN9VzPZZ|I4*CPzKi=IoAZd@tBpZ{$th0IEV=;0{UjFjt$tD%Bh$3dLsb{w zpL#jflo^zpAI7&qM)15Rh!B_#_S=YRP@riSPc#t^qQ^s0GMPt=SDEJ`T*02 zMS$=t${=y=8>6apH@-|5bMTWU&V0iBaC-YU0?bd%LUkeOkQu=bP#vq(=D5_c-_{$p z`VH`uJ*oa@!?(NOnxHGMO5kNu@g2DEZ%~-_F(^NuyQI*B!J|g;;7#<}l4LNeSP(be zM)#%0NQnbWH)ETs$qhE#K3_4=$v9B_jpzWI{G7I@6W->-$;;x`^Sh=SYQg!tyf<=W}?C+-Owj9co0 z2x<$g<7L+)$MSn+!|-(4Ya2V|85S+Spym<_s61lvZ5HeUY~P0)u+aBEQ5C;@S*>a7 z;>rGEG%NFN3z^E-Ygr(F;VMYqUhEowUXUC`Shn5(k3t^AgOSJ@Yw&_^d)tl8=JQkEVZByZdIQMIfFN$MOzdB0nVo`{e&Ftj1eTSG-sv`&$=%l zZwC6O7F+zuGr6(2)<}Cvy-Mm>fV{k|t~5HE}pq*SA#w7~j8_2j2A z!ma41t0l=ze)4r4VDS4m+`)%pkfg4MDj!!lE)EAh2^@6VwW`l(F87I6+5p*sYnqa< z@^`>er~V^g`_&5ZBZ`d+i(UcB<)%d`o)Jjo4=6CKoR$g855g&6w=DrFuQNq2&h980 zu$ng|NsW^eUN*Oza~qpkM2$IAt987wZBQCjKo7J(Knj}$(iSg|3y<|NnvEi2mT5z; z#>+)YE9CJuPp8J0q!6_!!2fn*!fT+thVQm?`r-$Un)_EL-}OuXJs?xSiIHKK>S+JNLcGJM>s9Z zm2@nfw6()dY`-SyE{fv=^QhDvmEWU>%lwZP{zM1OwMMN-QN_(NTIKi?>(l1Kg?0RbT(s)209J zULX@oGJF5*UTFFmq;&X&_oYsWeQ92M$q-&r=-cO$RLb0aRu>|=eEVOupb~JLgB%Vm zT(Xc(y2&Y6vwO&XC||nw)PB*{z#`FF^HV4{(pcUO8)?ipFq^+Wn|m_njD~KfuTw7k zv}hUJVH*AcBt<;prN{kRIB54M?(q}0Bg$c34o$@9zCn21hWq9XtXL77HIt?^uKRJc z(rUzHJ^F-{YsOMq>w0)#kb_Ci33%a^4t6fqw6W8okEH5S{J}=hz(Uk4dfTLj7&yp! z6{3dt6v}DnkS^&5wuxbp(bT!f%sB8_ak}yuTXC6h<0vhW1o}cyRogfE=be9|@jUD+ zhWkY)mEKlRTT zUzm@bON_S^kLgo z)N33+8GeLY`$R0^O^W-;nA*&zFp?DUEd|P0S_6v+1XVJ1bYslRyX-j3n>Hx~Yd|8v zGj@`40$8DT8lSCQT~!CcM!e4Gs&mul_~L)0-nc@$q_D3XMj7sB2kPDSQ+IASP4W4xmN z2K4T6KD}0Xq`Q5ZLIc-W;(s!Zje8M5ru|dSLAv+;@AbCqyVpMNpY@5G!gykj-X13o z-yis zd_f9p#dv%wIZz-`0aaw({uFkE%)iThZS+9c5R5B$kHuTg6kf0kW(HHz3=Mg|K8 z{aXu4Y&t;Rf$Re5-}M0MR(M=pNY>Hu`2$K9Ev)glPoB4(S$tQOgX!4}49JH6!|Eab zrRj2fiMi^INf#|fo(-JNWrJ|GPJgr!!`;=ULkl7=eSR7ciwpT(*WV4c#z|hCT?XjU zYY?l~Cu!ia`TgVK9l*TV9t9>rv&OP9zM0Bbra#@b)v;@mm#(t;hPVC zd;S9g3EYI>h}gQnHz9v|TeQ93j%Y^z zBujw*|JY3G@$$QTPg|JVWOJ6bPKNA){)5o=hoth)@5}uI-7M*bJ$xz!w8L^W0f^nb z-G4E3D#Fhem+3(v0hOIJ=>PEw)~$Wkt)+dz6z?kZM^_b_{eaeoY+&@EUk@9$(x%<^#%Vsd;Qkg+CCcZ9aDjrT)NU?G zC0Ed5vzzvjeVqI@C5RsND0YdQ2@tYdv^)DuD(LX_U8j;73QW0kS*EPZPU zAM=U5elwbXfBnY*P?<>$`|}mmD_jkP@(7MIzQ=O?Gn!mQP4FL!y$&orsdn~&=BM$e z*r;_sv_;_C-~($8#_-UGpDN~;dV`!ZwCt=;2EJ~@!hnJOyF+VqSZ1YNpj!y}0&jwj zI0}%``Nx%ihyB>iB%Cf3A?qgQdYJ81S)wu>E(H2EZkKujjE9zqiJ}vZ$uqz;$zgmb z<95-JM*GZJ_IUq6L720q zO}P6{22n-2FZ}$7)+IsD*zK7I*YDB$=NSCo{)LcaVrtt&Un%k&Kw+``pcCjGF7V6$ z;G;c?Qt6I)2kLp%vU6PKT3t}zf?CAFFd)>QDH1R^`}{;Fj{C6&BjI8+U#VRGGqdXN5y=JJ+Po(~|1K4VPkh?N!X^$z5Cuh85s-8{7-lK`(6KcpSkh8uN+)&23;uY*TSjwE|lGd!sw^| zHC&t}{>jnDO{8<9i$4>Kfd{Gd^~H*WlgusSA$iqHo9WQ} z^ghIMpDyjKp@_QteIB{R_?%8{$SK1Md%i-%LBY@J`>va7x)8?ZR-{9{%+Ibm>`uT^ z`B-o!RFYId$LSXIv)1@166l+JCmX9Ds@pUA{#W%IqH?K%M=6-mF zIVGi=TwuFG_EF^hB45V1|I)%gO(?NNfE#OcpEd)&U5gl1667} zVQnof?EqM6Qox$h zz^t}O&AU2G7Gr?_@O3=lt%SYVCzR z_&SsL8pCsa*)(gt>CXDiCX)Q3Dg`Dv!EvG;n9cN4xLaAf!OslGK60iU2d6!{t7kfh z(VI#+V$r%vWIA$eHTolcpJnFV=(w{>-Ar-$ zVkF_EzthD=-b{7OCqt5o*!v03@ijn}i~C}vUNc5|oePZM&pm7U3wgGcL+1*bQx;d! zJ+}TsO_MEX(7w`)Qr)qDja2){;ZhjN$Yai_$rplH3d$z-B|UZ}xNj5u9t&FGScj^J zy*(^kO3^1ZW7=0*t^i;f7kt7wbJV^okv1lWC98{)$FAM+iP*gSX&mrDq15E9?DzL8 zL!m;~|AG92eq7LtT)-n+eO=*)dkGPX3fUwkxCUm*6Oj(N`}Zxul!*@%)ABd>95x9OG&wE#x;_le``MpY zGbqPkYrPjxKPT*Sr1Y{m<4{TbTEkhN_a6DX2S344SPdsYn2+O%U8WZ>#7#K*e4FLb}6YC9Vasl&AS9>U$m z;M`eL1TpNjIn~c@_pnK&c^K26P)2m7t){<%xgBH=lGD~hxgSESO#08Vq;6a6wE-#d zfZM9;hWkq1tI%Roz-^5UYN<#bXz;DDWOH-6xe0j&!9W~h~{7u(f2sum}zRurCdt<#&=hcZmD(ya#ci93|Mccp$ zTuV7oP)wBw{R%s27I{W)PO~sps<&(f9CxY@|Z%5gzBPY%n|LfU>8QWhOMGJ zlD|o`bJ)4*2#iQI#~+!x&`0Kjz6_uDIp5S^Cf$2ImhWX~-5b2~q-3#W9zwc?_!~`XM;BK)&qegSE8A34c&+B->hJT!dbFIEoj5Z% zy>O8IWkGP&4ltQ2%^n(z#*5w{eJ~!JIKlZX%?XnPKv*#k_CD^Zs}CJKZ-x7dW&crjG02#)Y{=_T#oJ5@q0&c3*E-#2oIBVs6ASMk5j5Mzsg# zOM4;lTGcCTBZ1h)@{4mi&X4u{AGMZL3oVOJ`#I{?KhS5o?CsstUJ8rvPTp0ceFB5KDfF-j4{(A^wYCUFLKbi&zC(AoKuG7D)Mm| zec)9lJAQ;Y;~@GxwXb3228##lL>&ASxDfDw>b-u5Rhu?;+5Ko`KbuHeGn8iuJo9$e z?E(OUDji=PT&5M5rbBn^wq^7g-=lT(5;b(Sq6~}069ktVVE?qNa-NgrextfTp1S`H z^x%I=$@+f-T|-Hqu^_6S=J^h}Z5iGGsuW2)Oh5QBgn+A;XNr@U&SFP(vfIvuiBw%z z#4dM$M;C!0X0?>b1VRw`gp@_?ln?p2^QKA*FN@T0r^ShQE_m!!A&X~{a$Uu_No6KF(+15WVGgoUr7RYL7YcUuuVmML0~q&>;KX5O+802g@}y zp#@hJp8Yi6XQd$?dzVw?XJua3U#Dz8Tgm2O9NpfjV`4$gUq9H%NZz+ws0!)TMmYgk zK`s1}yb%0CXFiKnFRp-aDTwNfTlKczNtU??tCV`nf)IMLXrjuEhjgl`PvC)@(B>$C z$!W!py@1apB&{|S{ba8gb!Hj6Litgxl}nlr(iY7WdlVi34^md`bLpaUbBEaHO7bF> z=4o5>Nc*v-D_3vqQt4Zj4sU;w`v){2A1aT3>zsp(J3E^jCxd4_u`jam6D7==%SB{V zPcRa0+BkO`U1TIGK63)&ES&e9|7XQA8}N}RNasQe_t>$uGQ!>3cH1~1?m2CA8G}2j z*AdlpciTZF z$Ab2`!7rv*5iZvkly&S}6{imyxjFOzgkQsqZ~|D=Fw5Biktt^m+o*4KK_!@;@1+EV#-oz}U;DFRpU!lju{-I_s$TcNx27_aG)2}6j$u|!(k zJE5(%nl1?GWY)GBuiWwfsz!l7qcd&5sOaKo)HJuHu5KT9eut!jOh88C<L!g^CRduB^%ia5I*v|y_?~;M)9oRSEp^xwgIJ#} zz+*5+OF-`<#=H_Ln z;A>vKeBKoHNR1ZHk3!+%C(vQ1N9!}y-|u5lR)lGeV_#!PJz5P$*jRw841tdT!kdR znA`i=v5N5Klzuxfw7O+L{#$aEXj5c31}&0B|1lWw;F8V3^FAHiTO;czy(Q^+afY$S zHuQF}T9Tc1w}fL2c9VVqbStK-&nIL*T)v_4qG$*WidML;_!-HmfLZiBspePVxS@FW z^_p3|2?Y<8VoOma3QEmP_9wDJHjLP8?;8sm-<& zb@j4mcYCJ;;sPSH%$)8(Cu$KES5Jo?FRnrIO786uRXSP7oUN;-3Yy2W7zi43lc%=htqKF9hpm`R z2MnHwk!J^9x*OU36F}@3a?FJ2Ts0pG<#E(}b%~4i36~JPUnYT*3sx(zE_2kd#CAyk z?5IpvDRnzPNrFfmWpw%dI-DxY#Ii<6IqB{mF3anyTJ^OjG;@?IqJW%5BsV zFP3=QkxSANoCfeFdk4K-q}4qp`fQWe^`4sTY+Xljou5r(iY7{+jw}f}jgoLV*S^98yLVS$@1@)3qlIWjHpIZqKKb}Kfu><*Zc(PQS}#c= z%R)@YH-{O~@^V;euZtUXj`WIq0GU*#oOUW?`T&6zn*p-F6wRGNl=1f7l1uQMfxofu z;aRW8<%ii##I_SWoHN{`JjER~Mn!tzzDDD7f}{a!{*1LSPc)dWo<-lgCzXF-7|NTd zHjNPBgmFHLDP|y>RhZT=hEqTrpYS(R_+SKkAFE7qZu?O|lwNq`V?P(qCvgzTDnB%v zRcZQpUCl7JCVxU@)eZ4KH0nBIYFOkPFxjLyaBTW}F9o@Xy~~`I=34_=wE>70k}KN& zFynafd}5+<()}cGU%a_?kSI;@Q}WMpAEdnruDKHy$&%N91 zGVjxO$gH=_lP>KO6bqv}jynrgsLQGhYH~||^5j>5aXi4R)sEJk_U0WcmZy6{hMSVx zd`9(r-c6@x;@ysNQdjWm7o1#99Jd$KpZmFtjQi|b01NL8oT+IrItwWQ)HQ7q4~9=q zE#zcR84}3f);!DeV?_fSV!%>=raE0^JxB# zvvBRDXW>E|EXt*hnMa3(p|ET#L-DTEu-IJj8O|AY8eI_9!K0JVwbx>NMFl;Oy$N7N z(;oDN3YAUnxH2W;)6`;2%=vl3qg%=|v~Y16OdQe=?f-4dLb*gwU^JDf`I+Xo(9P2&+q!D@rsRYdd+&I8N?C0As9eeXP9=PWCiKEQmy#X37OS> zR-)Pd%>OUYXL}qYfYGYm)&em$O?Syw>EC9%e?Jhz?U4k+Cy-ZHAtT!h(!aSw1DMBc zSJ|+-(s2Lu%$=^QZX8=f{a%nganp!slgBIP4*RA*X|=&FJ7meh$Y%ya{5mgD9NLJ3F}!|nYCe02ST+d7>i zqX#z(l~2~Y%{wxe)$S>-lfuKMMZ!Rtb2AWcbVnC z?g;|$?o&fPKE|$NCe}54pB?5zM-Ov5qag6Gc{=aa)!BM2I{j2ygXXiBKMBVn2}okS zy7%APf*_Os9vyl!)#!&Ruy>Z&1Fz~I4ecx!PSkosRg13HLYWSK-e5!hK3Yo(Bd=VL zrvSr8vhUM^X~iN})Nth26F}K9>VFIY4ztB&CfMVkb?Y(APgB6En=Cw3(3yX;HE=57 z0Vz(JNeBgPUkKL@FQ14zbV=H0;3ZcZ;#JoV!3d)pCSYOIp?>q3{HL%%$iQI8&3Xun z9Ch{SBcAV2mo53N#n){e^Z-Ti3Yk_9Xj-%?we%lTeO#y4@hp+*KVR*6mv|O%la^6# z`WMK_XwLAde>+PaL-^@+;&DD%;`8S8!y&1UNo*=YjR8IwE))x(Q}u$Yf>V)(NPwWW zYRSHN(pAm-cx-FsqzK<}U;n(iU$yb4l|uX>((?qRe*$bw2q&r2M)?I_GNIH9V)~L@ zwoc~2074ULA5(CL_DPwCfH97E34zqwU!Y%CQ=R*(Kx^?rwl2`qot;jjDU(&oa1m{C zrE7$bH&di}ea*CO|6h|6C03y3v`uq!Ao0NKcW-Es+2Jx}QzH6J&BpZN8ok}5NY7A$ zmMUZ04$Zqt{5gyYx2_^UWPVUnW%ahe-l@iUh{Uu{(u#K^h@B34;OoBNn7*u<^~OfR zVA^eKu7HomKQIYnDmwzsIWH2 zxX^z_0mp8X0WeHQkJeuLrw1R<`U=fByH1~~&9AJrYtzL~-+AkLHmcKZ99ILt3P-Uc z2YBDMmT^9jRGwOYkiKygKkSpj>qv!lFM?`=8L^3S|yaU zkyOveOsI_-TYbJziQ&1u=x~gEE*!6;rL@MLh`x2o&?)%|zN3jaK|jN6Wq3tnK89S* z<&$Op)D8^1GA7g#(oC7ZpH3NC^85yd7{`Z{R`6xBJ&gnQ z%L(5tH`vUGsPi^BLM)ltu?Ob4^R)FvcweYH#Z)=3{j7dK3(5w`gsAn*=$pDc*6*;t z!_veVx_K-BDAK?Kn|aaki}5kjzII4BH@=LyL31J+GD90tHVoC2@Az*~5 z^)+ObmVCFtji{Db)erduh`M7 zFKNzHq`smqer!l^RHh;?s>fO>!Y*k3a&sIv)2G69J5cIF95n^Wdud>8pjkg z{oglx_debCjNPYC_x*6sxVJw5V^Cv2)w|Z3Yp%KG^LzZ9M^aKF&6kkkK9&9c9A6I~ zS)HI5RjmnjLUj51tj=1ayzf3yZ_dcv|0RyTR8HeKK|FZtr}?I8bwy@V7IHxybM$uC zn#LE0xxh*TV5 zvXXs)&IK!ZW|_9Tbdo20Aip`0%tqM_-LTEdku6shpn#923uW_{DC5fYM|<{oa?=Nr zYZ=?Fn(C`u^kHU%fvZy|@+%wll3{>{o@3~66EW`N70RArtBiVYk0A3dG`=KiG4{<*xzLv9f6!fwBY6;lta z(JsXg8>lkQ?VJoEFJGKnkdj&)O%|WWBG(y@XtS@67TYUTZY9HFUIQkZd^XN&)R^&TSrGaFiKl{~QADA*j(9 zS++=N-Wms@{$8EQh%^?E+)Gq`CC^x3E&XI7dtVS&iCIE!8midHo_0Z#aGV(L_DEPM z$;Pkae&nREjQ)*A8HuJWncn~?u0v#P3XOUyz{M^c!*b6bnuSx$iBSlls$#@LrK2s2 z`yXo3rtDEVysrjqE8R-=d9J68;cr97QAXpGvV6^21Q<|~nx-Ij@Hfq^nH*<)f-m1_ zN*lP$LAY5(>_#g23z&#j6pwP11YC~vxl~Yj@iH8?zPO~nFFYfzdf-&%X;HJe);C59 zb=Z@5fjG3&ixP94gu_<8Vs~V-sEOyM$Y<4T15!Q-c_>Ytk0W^N!jbN7Q~)H1xlr(x zbf!fs-^PP>C6Kv00LY%d zxqeJ;RVC?8E{To4ACbHWMXug|r4-GYxEGU=XH&!|QpfF856XTPnFi{`d*N z2G!?MLwT(RnB2A0Wm4rkY^^LHHcQE8jNlgrwp!_t8Z;#TnDWTQ2y8b??ly01Dnb(Y zV4qfftxkQ>_f}XJ-TZ(uTrZWhR+`HJtI(D0C;1H0=+C14hCwQ5&A>uDb1IgYNUE_M zMBW&00?Xdmo`RFS_=}qD`R_n$Q7_8?7WN=Fut+w$o}OpAQ%R+du2-PpVj5T1bd|8W zT++VvNmUV^K~Hz`WW0oaiW*8<6V#(x2|Yq3iAD@(HeQs|e0wPX2~o|!HSs$cobD>U z1d~JDGaE&jxSqnilUBIv;;gZ3d8$5>{gg(D%VrvnEgCB^Fc@O~LS)t?+Sl)vBYY4T z1=@A0A2}-vvVpr&4Gz`L-r_xqm0z;*Y@D$kzwwCMUrh*#peLbF?rzuvWTk%CekvhH zC$v#qFlXwCQ9p|-62yI_+^VAG)VyNGRl`_uGCLF0O1e)|sZ8I&$`bK%L@LAltpS*5)Zl^VWguE;=-#7+f?n~FT7>bY#R&q;h z^(+UHQ4=83!i^SA>G-E9w2MOsrntpa?nIV$JpSQk#>X!~Yeh=*S z`Ann^+bu2#Krcp#Cq;lOu5&Ar%ac?0B`TLkZPO|L{*ONeY3G&X=u3RX0z($LqlEL}C=V^86nabpc{jpz-kDA)oEA-3J+Mfh zcue1q(rJ3g7y{9K)$on>ypOdmayO(=N?AGozS0PoAU^gUcfa*k&UFf7$|+jMb*?7z zE(f;tZWW8t1pKv<5_FNqSF+uX0DZ!nxYooJFXE~(qHaThpAJ?`I{1M`S&zsNKoV$Y zC}7QPSGmLd!z5p9pFOTkwan(i&{%g_N`ti(zw2_BjPDBAgi$Cor^l|-g>bhG-n+&> z;6khvfuQ;Oam0s({e9qBykI22K9$?IECTM-L4} zx5c%o5o}f#QpJ#a$;->mizH*-VM?fp^VGsI%!HRLPQ~svb z7)Oj*E~`0yj1FworM58p^bvxFaq1lOWfEbLi1TV+29)%)0prfFY9J8t4JY2UZO7Z+ z;u-{M)&PPL^U^*WcC>`&o`@~rOlqBn0~@33rX`7; z9}Jnc!{2YZo+DvUlQ9)}uFz@_l~}(=if2P<1>~J$?g$lhqUd7z;)i?-lM(7|m!W}9 zl+22jKuGBb{gr*~5LRrf{lf60CUi=S*w-`Ng>Z3S;PWT>h}U99gxavY=II*KHopfd zO{w$;3#V)4c6WCb41S{AN<%|wm-5kyMA9nfE#!dxo$BIPIWG3jR4#gVC%E39m~=

)YITeceq#;0SRpL^vKvcU z<}-N((vh5AxLKsX=5MWrj#_ZZ^Q(8oIDF>w4p|j;l0&q%cH%dV)x!!FE?e#5@~i~f zKRv7yi{iV%AgJ?em{(E-nyIRa-ssC_<(;kYv2>yvuj=1Oo0=BCb^pSjectg&z6G;* z(mxDVGjoZcWYo%YfLso^!SSKWqvb(6H3gkp7{a9CaU~5OHF1nW9LwWQ>Rylf+ZW?; zglXP7LD>h@Lu8Af+8(Z=0CWMkp zr316hIJq>DnuZ>IX;Tws4qg2!hb3R1VJm9z0iI=V0rFuWMdc!%5XOF!=r!wqBs}HZkNE!5k;<0k9 zF%X2Ts`nO%Vpug_W(u)WeYP)xbE8=gKylotyDiLP<5_I>70slf1Y9L z@Yygu{iG^Rn|so*)x^7%f-$YQGBajUV-xAN)UUdMDO6XsRH9#Fq1-Y0M*3%$2;B8M zE!p#XynA+!d;#~5e-+~fuRREPMO5D+Y<7iV$b$UcknMy44(*(E^wg|NSJId|jG5Tz zXa}luKO`WqFF*U1vShZKnm$g@eYt|UT2{C+FQ<{s_6~U6efpw(^)5vht8TDb zX{qyosp!J6?TBwvs5Xb56}G`D{oZr0wuh8~cZeU*z)5t7iAYR!GPK>MM6lSQ97Ww7 zQG_WP5g7c}T1v)gtdF~mzGGHZ5txM|@nBB(Vdcvk3CNeu>?ZnLSm@8MNK_CB`AFPm z^NB&1f(u3Gg*tM!@LOUh(`%lU(Wm|gD2cAw-pP7*#bxn6g9To_Vguxx1WI;IWz%1Z z#8lLtCI*#j4@!E7LM!4qH5EVw$8TNhC^#dh?**Bghd|ev+GzMK10wE--pFf7PLFth zt+mc`ZI;QVeBL|=8y0H%}t2#d6c3iT@ zi*qc6=A8UQ+Uu>~1dpBQ6G^sL67RO6>$!0eawV6C8^*J?nttYGD=o#O89pGWisH*m zl1Z%%P_K~^*%n4T;C#ncRh`=*QzMC#EFVp|%SUqENhpGlQ@j1AVu{?Lk$tT%FWB|h zFUc2L=?{`lSq+!jKu{SPKj+|8(goQakJo#jLZ@%h-M*|rGXIEYcY>#lw2mC6 zTI`@v=TGvkr5IZYV`rDSHsMDkAj>%>pjH|Pu;YuZS6)dgOcw7$u4Npt-yEnv46k4y zP9+?!kDK()7}*sn;KrZshX!M##o>yiDPkOYD&!AXhsOd00q39z$B&{h|!>lY;>9FX*W;z2->d z>J7z{$~W8jCQr$3$9veBFpia3R{D}uJ5qxiX+|~(zXk4!K5G<8bgy}y_swUQ?vXfQ zr~GwSk~-&RwZztZ`$IyITFFJC#?gWF1a@{ci`Pf9>Lw%2U7}nj^^YPcHQ#4`WKQHv zvrzFq33(V4_MwM;Muu!#54r{7ac2HV4w^DK>1No~2pA=J45@kleit%2q*FvA;Ft$B zBLax%WK-lL^4cN$ou=B%oNqPC@xtn8*DwhRf zc{^grIfxrIJ`UG@ItgPm^?wNLzkVmS`r5mdjo7=)QPq3H5n6F|dJMXrgKOK_u(Vb0 zS}%fE4Ab?te~*kK2gh@mYLZmIrcCwc{q|S3eD|swfjxFzpI1YfK9Vo5d@`R?bwuYC z`^sGVKqySWY|r(iqrWC<2i_;%$5wb0B))BD3t}@xL?CJA*g4+E-!}+Id1@9kRBSk& zsGHuJ8l+}ewe~R->p6(oS(`K_Z0?H-n?y4mC08tdO{7{qeslG3HVgU&3S7bC`?I~K z4!GNT4O??*_YUfjK13Oq#!1?pTRtA>Y2u7DUyrZH)1JlHz%6Xmyf1UO+V2R9V|u$W zNp32Agq^khCy=5VfzS(0yqWc5=|Cl&Pa%Az{^G)oYajpmvyVhFqXc7|aC8hfpGQz8 zfPA?ZiVyYM?S7LQ&?4SN{+?yRr4wqx6fG(yadm&_EouuRe$wOER~3IJ=lEoIc6r#= z$GePyx@%|4kt*m|^_p=lOJg|qRcGoKCQjR^^herkx=O$JnS#{10&D5=}DEFPj*o& zTAO$nE=zrghCU)r6`|*#t!7>)4PA&>po6*I0y6?DC44)v?zBw4NV_tRIq-=s8QyW@ zB;D@`kSK{j2Hq5C)`d*nkRi4p>$FCzo;H@f+0=m8Y}e*sv2l~8&w;ma)n+0eCCX9^ zu|@4O1vL>3=0dr9(J{ZHPUfP~!_#6wV7VyyAff*IK%(0tiHm>uu}=IeSZ!Uz5R}n0 zd*>QB|I~jg=Xg8tbeS^9w!Kw%3r#O3L)gLW?=F!M4x@Hfr^XU>dpD7EzQhS3fx;!W z4Vb4UI+AUf-h)IsVSY90=8xM_pi1QHdWoR&Bh_n+Lw&_E9i{{VHyx?o>e9&EP48@D z+5*$!sMqZ=V>F_zYZBo64t|C76=A6+e*Pcp4}Wx~4UV{$porLw zP!nkl_#A7v?2fhXa}=VK0XTzXkzHZ$rOK*`{e9eFtVn-eC?*F}Y?_EAHB_*(`(nbj z8LA&Ey3l+R%vdiK8r{*K)jIpFvAGQVp&o_zM=*Dh!V~3CNpE^lW5lw=bE>1Sn=0q= z5(M9UU!B|Bhw>ypJL0&QXrDPGypj%YYpNN1WIMv~SdI|$hvxg&wQR|Yl;E0eHP#?d ztobA2j33HLY~+_4A%e3~*0{Y49?jUDr+PF2EsWKlHCq+BX@+Bb%ILi3BE{Rjw-4_2 zYj^s8d=O0H?HBZb_=6TBW#M-H?+!a% z)PPHpaJm8ep6X5Zx+&x<&8QU2mP9x*tj^+gN|l54W*=W=+%HR}rsk)5;v zVvV;gm1mow1m&%T-UauLbn@k{F;|U-oB)q=O?hHe>XN~v@Meq;`)s9bQ62g)=8J4UC| z(2%yP3}pTI8CkN}t26??ft3GTtTmlJR$C-dV96pEij#y;t;>Xl*;cOH}8tTVKT0DOH? zP7X@58M4i7(I!{5AOHBWvCXx#`AWhfExN% zsYr-Mq6#lEy+61=e|Bv7(S-e)V_azZn~h6Hy9wf_1k$t8Vb`U1&vEJ~Jc3R@enS>U zzPwta*NCT}rDZt1xzAEM_gUG*886>m zOGx^8KOkAq{Y$p$Dg3W?)2%|Cbi(e`_&^=ToPMLk0mv_98&2Qk?QJbV2}^o*{X@c$7$Ip0hA}dHSRVdc>A~@}44=Kbj}#M6 zH~`q8%z5`#I{;kNaI|kot<`ydxBkQy#ms*-V|S($>sk6=S~9L~-Q!!g!5uN%J<7>LYJ30Aruf0y@`;KX^XVcj7eTtQ7+ z4z5tW{KO{TwYdd9gQ)=WLh!JU%kDBi=gY<|uOCFWz>Q_p9Da@bHhQYORR+UPr+-rG zAG2m;3~X{cQaMS=e9X zd*c-jj2p*uoU9tv_%kP}a6j59^&QFUIAGZc zoGIIrmt`e~&(OR%4750~)NlawD5%zZJ6B|g@MNsUwhBeqcSJqz73cl-qJ5jBRA)SD z(gL&Ia5QRqx+CmuBVpR>ZKpS5s~VmtTNyZay_AD?BD|L@MgZrf?-;t}h>tQJfZ0<+ zu5pNw>Woyau8XJpF`g2BLXUU~6D03ZQKagQu-r@>Eb{=;e<`t31#YSL-z-(=*!gX> z%5f?^UqqQF85kC)uaNcff5T$YvP=YHr&LEqUJQ zyz=UUg!#4WiaKILh?FStgRWFT#`djuD$cV{{v6sGI)}E>&Y`V{M~sv)Z7GxfDDsV6LNr$O*PzFC`q}Mz_sc<2n~hDJm=;=ta@+9=lo}y)^gbxj@7?LNak1tluJE ztqK&+itq$wCf||5vRA_fht^{t2)?oR#>Z3qZ;A$cYO+Zb#E?e77EZ+e`!8o5JhG^g zD80Vld$up*;ZuiST6YC~>LvOv6L6{w%=@|p+0D&pptL?1tME<{8K_B=xO}Npt%@r( zTX&yJqwKYFjo)FJfsZ5H@3=jY-kB^RK1Ia8IJ!0um5k& zEj;eAUZcYEtrh55nEl=qaw(25i{YA?4d`6SR3`uV9v%o!d@Z;kuWZ+c(y{qBRRU^QZV z)+uEDrl9^tusw(P{%+!FyxV?GEc2De#OkZS@s8JRymea}Q@y{qxv^}= zW^?Oy1UZQMjrzE&!jQ)k$)S0w_ZC+bsczs`#RIIYJwLGof$tS5vIO9_n96hX`XB5q zu0VjZ^*Zo5xO3QvJ1yW;6uk2FM$>@)kBGaMy+J*Re;Y*d8C^>Mc?Xrb_an#0w=FX($W9W0SOwsG`DZ<&n_TLg?6uYx_oc)caziVTmBhfTkVl4272*u1=nOUjuT^g>DQ%Tk;jKn&ucIkcotybi zgxBI?>CDsF@M4RmZ0ocfV&!8q?|g@~D654+8J9ZSG0MovF}G9J#B7r*YP z_?*UcU0I@}i$5RAuryv5`W78a@|bhJwyu(oetmyCYcID$kUk`6Z{bU~?&YhuYmnp&1SRg*Km<^+MVm_931kf0r__)6g zE339JVmZ8nIoOJ}xWE_?a0ldYylA#1F>}RdoWRYsUe9OUFA4Ysg^@yiAUs-Y)F+jD zb@_Ad|2Itb>bc3PoSUphu-eG1QG<^%(1c-YjZXi#xLB`|>&Dt1!&E-3*fy?ras8vg zt&OJOaE)v^MppgI>dWl;phHW~#fwgjG1ht)NjqV|Gvq%Sr9$Tx`)3p);!N0d5iBK_ z=a6;T3C-i^)2}Gi+e`uP{>pu^`O1jC`wC4H6Y=RcIQY;-qgG44q2g<|VP742W{YWb z;;%4w02&HFMbZEH)5zVw2PAmGQk+Ztlwc-{TSK3Q?gxR`7{3`iMWa<9}e?-@+wag4E6$bq{5umwvmmn4Q?v3SG$I6&2$t}?Vb%x*BiC&9J{PQg3ANjb~O*v(7esAcF#O;M%MU>x@r$5`L z{@;HoP+OD8j-Ea3hbRC#YutSL$Iq`3{N{5i0yl7WlAB6sVyOtv0*wJI`JG-+~?y@<+)SeU)A7M(Kw2uM1jL%X86FaHqzL8ijf{dj(gnxIM$~GsJ0H^(4Vf)m^dch}(V?(S6K z(2MWy{?EleIgV&VSR+8;x{ZDG!t>f^CqX^{To3-rj9@j9tF`Cl2g06c#EDv?Q-MX zW&V^bEHu0KQzg%}D9Ob#)1*+D6?T4-duD7S&IV@ckAs#{$ky+^Zue>a+B39Rez2`> z3s_!f5$Drtmw$=WVEZk%DMv^#oaBou{xOA5HMnMgYu_sJl|PeTIRd%^IZppw)$WM) z3qp2gerUHWP-#hF1wH%+JIg;$_0d}dAjOZ!W3Ah~Dx>9g+)@k2HNOs0KN~LxcYR3Y zHjQhnv+FOOuZeYH)_^S3(g;jt8@pcrX#etYNH)X|f4W3J8X6EjN|arSBd_Xwz7%4&gipa1*Vd~WP}`t`HTV6G&UDml9RLRJ;+ z#UOPHliqM~XY98ZF`A?#eL&S*E<>1GB&9#HJ+>IHSZ{wsZ6X(cc2xzDz~wVtlB{#s z|LXH-znh(hC$^|Dm!g`hDDQUX;GRsBQ8XdB!1s#rP6Mi~X+C91t;vxZeP$vbK!Vp4 z{cjg0|Mkc(1S9=8oMq6q-H3%1YX0qp6;{dYo~-@>l0mwR}BUHX6q? zF(T+Y<-A7>Wx=o3$OxnNFUCiE(=&gx{WWf(h9ob&IuF8ZL{EfwahEF?T z)jr?ItC(qgG#gX|MuEGnzlrko7dwRmq4~TRR(f%g{xAGV$o}&Yl1B;yiL*!1BrP;L z=;nlg=!iLzr+Yzg*f$d{CJn0^^H~M+>7t`wCf%2ICbw|PR69kMVIQHm`5)D6CuF#ERP5x8-mQpx} zl_@kS4?j42t=?gGhi4UN(w@P7mCZ@wiwL4;40LIh=uLKel$-RMW7A#xCwx0@z89iu zaXx&1Z*fUbg!^Lpt&*KITSGht{Tx+@=_9a{aT`t4xB%!s3V>wZq}t7tzR0fsal~li z{>>J$)F?|6LtKaJcp#aVTxTmEf8<^ok*U{UuhY}cer3jf%C!XZTTsbN6Gql(;#jd1y%1qq(#$bgNR?Zvekf75r8D|}Hjfb!e)zQ> z_w{;pw(4sW<&ON2r8tT@X6#K}!Z-C^vTyabF{$shHJDYhBJ>)qWw9wl@{|}f(U7qQ zecp@r0iAkvCt2d6{$#<5jo~yUjKnxLaK`bjJeq%WDH`m&ud3W`o4w9$GYD>VAJ3#p z|01JEqJ5Fih;g5z+HR|lwrcN9go;VOKlvNzf0_o%NFl+GX91yoQ3S@ev1nX+gb1x4 z8PM&b_|!jP_*fO67M+*J*pj~t)W{0Ix%F>t@FY`2i3?e!KSt~a#w9!f@q{cf(EtD? zyVU4>G3L5EmOVZ_d~cZXeWq613b;!XEDW=JpsydpsPPT}51^Rnn^6OaJiE8R5#_xR z87}i_xN@$_kC5eu_u2zO`Eh5$%pbz3RNIR?Y|po_+<(ijlgjnpRRyKv62J$+B@52k zYKgWgc`~1Ri>oWX^!h9$4G5pv{%2bcwf8Uj^b1<8AvX4-WD0SYcktsyw(UtizqeaB zr#;ozDG!UR04uF#>3F&(o(OxF67;B1q{AZbaVMMRfG5dz>l}Mmd0^_^{G*s`tZ^4H zO3dA*XXNX!E;Vp6`sj8PtzW&ve%#%*XBvz5*vDR=YiEBvC#&meThGKQ>V9*rd85WD z$>>U=NwCao!T&cXw$SWlA2x&Ft)da5pQ(6ivg?Tsz}MQqPvE|Hoo#vaVn}~b5-CwH zNkycAgro_445JX;tJ~|^Mv(=23!=AQ*OOIdJ$P)3W@Eo=SM*gVU0xm-Q>v-fu2WBO z9*&fRZa2CcU#7Zq8*hCdHdLeXMY(`!8@}BXUa)CPSs(szTc^Um9Ho*iCl{}HDC!aU!0{7H zlb#p)?L0vVcDyg>n9Db1HN%QQ{Fvv9Lw`h1wLF~2A^hWDqQSvpItn%E&70hdW4TG0 zTc*(i6Vq5}5mhi`G%D@OqD#Uk0f4U36zkO7rU-jx04z^^kjKMpu1JAxp?H!1T?Cx2q=Xock$jeC2jIa;aFT-@3el*?H+SfH#eqNj#%d$aK;^b*gqit|J zp1k!%5y>FDN{u%Bj@0q2ja}`l>@-8ct?nCFW%H>#vJ=Q(*EF1aY5t}0-0-M)u+!YQ zOi!0_w_*%_uHvq>PcOsN-Kn2T{F(Mrw#0FN6ezy;w*&b_6AFC)Pv>^sMKx@h(?l8$ zG~1&%z?CAS4Y@Eit$AYh&oLR`(240OfElRs#$t^GyVi#T@P}o~w#^O-dsq>T;++YZD8vxdzli%TbR$O&gFi$xH>zl32#r#odW!@Z}T(# z-2D6WF)+sSRVwpIy(fB$)$I+x{3P!@{{C;F0<6&qCGvJnk}W6BO_UkW3=`5k*kAlXNIC(0a7{*6G$LZv0FQSQ+UPEu?M(mKX!pm!%VRTluZ9ag|CWr-gM;jAj{ctUL!bkCX_|Levoe51DQrO7c|?Lr#VB+XRRAVzoXA> zf8{(buM$w%qhs%>DMc5J?QovepFN}2WF_CtA02h}&>RFvP-MV>1KtEX958l@Gw)_j1K6VfYs@7b&m-vRz# z{4VGixu>t!dzPU|Ha+e>hx6CL(SnXTRg}^Nd`J|{_EyKHZde}G-xt}hB&mk--63or z?y9`4-OVW&>XHs+Ik~ajwKnD@oUJz!Y;12Xw0W1}l=(luuc_W)QN^X2W= zPLfH@;BlDMBZG?^PzX1zSuP4X?Ua_2adp`JUE+r)qZzF$#!N>}HFJ^NRk~QWN3ytbq99l&<4*O%DeO5FX}c=$E#pHB$@;h1;OedtP&PbostEQGiZ38PEa-o{`JNOFPl?SfIjrf|BZV zLxP*GD7FUN!h;R5FJs7B{N%%MUl?;68gN&VJ<#;0^^+vA##g`n zNtA)7RSO3pB6lZR-a-Y3!9W0tq2V-{Q{bRc6Q}#~D{P4T$CqZ+On?1A#1Cj$Rlm~< z0fMxq6C*0Xt{D5Q?nNH2{xJ*FnfF|7V9mms%6UI2a#^lOhn%#e2@k8?Eq8G+rA4MX?=502!(;>1`bmjYC^pRm zn+bb`4_Vg1NOhGc$?yZ5vH=j(IJ<}|Ej4nl$Xd}%aM7*J0~}`vMYZOtK;SEzGNRj1 z`2@_ujHTJQy+vMZ#8DpgIt0x>(hd$o+Iqg{E1XJLzcvP=<>g`-R~g$eem-xXtG5go zkR|BeP&eDRk|)2Lp&STNRD4tM0zW5Vp{#yX4$a|gL)M+K;kSgp@<@U})lCOHlD6kd zdse>CN5RQww0kFRvwF*Bjx94B2in z9rh5+>K_C*o`s4&$E|0*XIkwK`=Hj?S9qnJ!pl+fZRri;Y7)$4znZG@Y6HVPFA7gm z+rJXhJ?o;BYl-%{8nT$9#A`kwK{f?l-}nfss6qa$BB=+EQd6l<1atcm6qs8?mi`30 z44++L;vBcL(;2X4r)meTy>D7_=kN2Exb|#xaSWqS%v1WbeS`O45aC+BhqPALO30`z z(Nu#inMb`(&sMJCdteQmL4S(pu$EAnxc}mRc>(A;j62g|lf*F!y}R9&ous$M(_T+y z1owodvDSFjh3YS&oOVgMqMdNe*c!KGW|tdE`3Tf}>bMzGu6plQ2P3cf3!nwCU5P_c zuwiZ26xqY|c~AwOvBb`j7P`GsiqIdwI#>HdL9IZqUl5yX-V z#Y@;+Lflv*s83oQ;6jG&8)mjwBQJY%=>1YDCvZMfEOrIUeaV~qBYpK)WL_hTk%3Ud zYwat1IzT~Nf?fYlJli`pO?4TYW`o&a0#QUQEA?TWw{O8gi_M1at zjk|@0)OePhxj!(MYOG>q-NviSr|WbMVgc-6FIwl=)Donj>YE1~H&~IXh#Pa2kAbim z?8I9Z;%zwb$C^^OGTL@i&ULY4Ic)yW(DRaeyGX$eMjo;(vCIE(QizNh2%d-UF4D?U z*%USzHswMsEnGpz&*{|$4I1vr*hVz&UDk!fZ+=za_I(hu*-Zz`mxoLZK+ zN3}>Jp1$tyxhi*_NI*{&S+4k?*;MHV^MEvN(>E&5ojsftRt>Rg6y=HJG!MXI8K{rm zYF!Di6yD(CIOI)QrMoP2F5Y?Mh;g>}==o3LQNVHbLqMJ_kZ#J$(P()65^;V$QuzbD z>0MWUGWiWmMtxgvB!sFWusZ~FXB35XNDlZ$C(ms|ruJ(;aBij~S&A%-r`er<&e-S( zR|moA{gj`dt^68YpFl-0?KDenA9N#;BPGfM!7IwWO5L2>QL$yf;XOjGRs93!tyk1> z-F2B_JDfxLR^Nw^Z<_Eo9bQ*i742S5A7RDdAWn);-F}~`d`|?hmYDwlT8b+>JJwd0 z$pkM{RHS_#r~3?|7z5RWV%=bNjkp+I`g#+Nc8i5%2#_>ar*5J`+VQe zv~DCw5+{Um!WloIAkkZwVsnD?9YXzy-IeeAbC~_x;sUSsr=cud-O6@4x_$r+gByJM zJJ_JPG#Kx%mhgKDu3L<3 z_I+R5VnfFtM8l8_2Xe;Um;C8%0UVTrSug41xLSi?CdJrTs#+@LH@@!~*Ex5-v0|^1 zFmGb_?J13B`iBzV2P<|t_UW{RlJkr9fR3*dZZZqKmU_JUw0LQPI6ki8E`MHZU<~GL z3yi&Z|NP-C8gj6jHRngtp!bY_zC-Z+aDfAzaQn2*xj*?jM6b$`_+7jL*{dT?c1?b1 zTwIlJAm!P1YDVhL`aq3bCyIHXn-<O3I z>POHwMu+{WZbOcMz}H&kTyfdfcvIuae#%KPXy&PSQf*Z#X2q*gY~p5njS~DW23f`;kn+@|Jk?Z0ilb1q_ z2$j}I!<0j9-BwXACDw1o(o@gu3BS#u`IdweB5?$oK!c&4E@jI;5LT3nI7-q6S@lKZM!_~JtlFD z>!M)_qHBZ6L7}(beLjfsyb4Pej5Q%8R4O#nVNsoUSxWWIk$>VH<`V5MlobQYUa^dz z-h9L5^rD~0?Y44G`|}7H@QrN{y@$aZ5%v^Qkt8F99m^0}f12+kk(SllI>fhD&{X~G zi1LUpj#Fp-$*!ASrqM;vRAr^X>CXn}*^a{ys=s<^a1zna?98X^M;LF5)D0{L zlUCy}ywAj&2T1P^UyE}23-ChJz|885$IeHmr~LNXm=?q}kVQ4mmTxV{G4J!5o0^bA zypQ}(LOV3O!8o8|$9-Y6jD@149H!ID)OT;4Nlg;bz#a3H^h<{KXYKw-G`vI8Fj~wC z@Np&Rd!08Nr#Y(q*2|?tGaQmy{fk$RlU(pdck?(7Zd!wF1lzZo|5!bBE0Gm;LuN8J zX`$&LV#wCB{3TuYsUAbJBvH`HNbO7~+{14uMQ|8rr3+I!iH_iQO)1r>l8t-);=Svo z%RpJO7fjWS#u2V3DHs%ys(Sb%*E`2xe^wYafTyy1zdQ(!Vd@wqx@4e`!sM(qWtD9D ziZ8k9CfC$c%fE0&Xs)8HE!rXco5xhqk`C|iN9FRppyy^LjH2B|5?w!e&(R{q2MYK< z-o_SAp|Qf-RxQTa!>cdxCNnA0xZQJdMX@~LLYlf#s`s)n{@HSLS`k&)scU2cT6MBl zzb}ZJ!kfLKar?jxEXvKT>0C=61dPqW$4J-1hj44o8Z5ZJw3hdnRZ|*l7Vdy6seLYO zJ4Zg{dNU_2i19Yv%?}D^XZkB@6CnVvr;w_`0*$5%Chcn!S2y91J2fY}cOu z2CqlhKD3l}8ySTjuZJDJ{iV*+54o(nP%aVVzt$OEbcNM{0S05 z!n>rpEP5qDo3B};(f*P35hk2J*C4&{Ym?^$qN@+Ytv%xTwERu>kxM_`)E;$VVYThA z3pQ3Ej)~!V2vzPrfrY$}>DXN#)D%H-tAAwooibE!7Y1D@ZE&g^?FvJQF*$0E2<61? zuA+xwT1M$2X!314x_=H0YQlEEieYABVoc!pmOfjgTCR{$54m4Ch-?uuFg>NRHe?xW)WpQRZP!vLrh!6M8W{ahlv&^cXUi!T*y zQRjC_ZYKf9`F7v*nCgYcpS*`@0bZ;Rl|Q%@_<8zl{A*``zP_-yPsit@`rpRh>M|jp z$hHA0Zwpd4>&nQ7_quN=h&^6; zg1-fiW>3~a_z698?K83NeTkkiglNwAEB`{@8d7}i;kLXR`6KMyhlJRh*W*P79w>EQ zJWnDy8M?0M;S8`XR)%mTwp|_{5K~*ZdXp(=2#^%gJO%ZLS5hT>XS{5Q@tM16C@Ly( z!whZ7`w6dYOVM~OyD!6XxlV&QJWfnz4YNadDZ%U4s{uN6WnzA3=!6}C@7f|xgMU7w z@1^PU>LQ}mZO6ieg%V&5 z<bS*w6rQV?xExi)eg?E5wLn;QU^Hq^^5+muq@s8fpCk)M3Dpw!x~U854idcY>Qo zBLhP*!CP@0BC%fWr!$ASz@6{?t7328MZqZXE_@pc|FAprH@3(rxD<1RvOg=`&_h+5 z#jE2B<~yCzEaXw#c1bZI+lyKTtP{;lCmT8^x7WR$pm$7+ss`RnI_N!NRKoOsIe-FN z*-+Yo=b|SDCZM1|wn}`4=l2}`(>ndqq)dme|JMlhkUhWxc3HR#Bx)K7^hOqC%?+e8 zPcHJOvGk^BR-TDs#C|{Pm;L_bC37rGf%g$+YCo4ZdfP8&kg_5GUdV}$i}Xp7jp0Q0 z0Jw)}+;<@s!$Nl^X4WK5UG(Rx)LLEVZRL*WCJ(%V1*urg4{u)O*_!nK{bMf6!Vs`I zQL08wL2n{Yz>=Akp+B-^UkoEbArrefHod{6H!;K-+*=5#;ML2uZ0z}37p4%TbYEYk zp30Kn7hI%6;amG%dCH0UM<=&!j;I%1a%N^`6q2BJ6)Bvqnw0Yd%fhwv3P>Dc#GQ!a zinX%iH#)jV^r|tW7Z`~VHa!=FGxQ!lS>Dg#w4Gt$eg83gbak;QTWFl56s-l^ll}W!e2RKqQY@4 z(c5>i?w`tU4hdk`i62G!8=px?{zPbl=Sq?zPd`0@G$Dqpgw zvqBm4N7+TvZ#428m3tES3RP2t)!7iO<^t%{D46qHFl=Pw#Zr#?+k9x2Xe^d~;=X8i zD?E??dUYPpQDlNJ_l~jf&qM&(0|ouuh*+f%|9r=kwYz1&8T+k#i>RBMfBl=$R|Ih$ z)$F0BclD!7oQB9ZRq*ggP@zU~(Jd?Ppm8a=17Xpy1a1(VHIB}|L#1C)e*JoHxE=;s zS!$c=@;+m0FRy>91e+jpL_5ZB%$1eUi7+faVa%O;YdM}vp}s>LM`E#)?Lk^_8U9G- zW7&_|2gPj;)c}>I<0OrQ=V#rJ#8x?cbt|=9>twpthD{F$Ac%eQL1u9EWJ>_*L&_E@ zYCjdCjG#QRiFLUIiZDBtquQD(OfTF%^D4pQ+EvQZu(Gy#)}nY~m31Rh^+!sP>7vE- z+es1*#9#rKchWzBcQJLUflCn&Q`}-PGMOt;`HV(v%iDlanTkTpyvWo{qyhJZnH>x0 zr^-*{ah#4Q+vps(+#cuvPb&#Rx6w%(*%)@f8CG=kSG{q~By1hvfO!a z1HZBurrM*{F*3~rP9Vt1NLB^-YaWYa7!q3F>=f13Z>-fPoyI%NWp}OhOV>@%mP4x} zsyLpa?%iZrf+B}v~c)MX@%0?@Y!MyhxE%XsxpZ~)c+ZGtp)?SJ+ngc zk(gZ<4xkOO;+RtW`|<3;nxlYJO`F$>(kRVTBbgPjy&u^9?kB0_KDm^HqzRS?RVQ-Y`P8+ijs^2 zg=@x&0wKZbstSJ!@fOV=op7DnuU+$~`|ZR_88lvp8z=$@Ee=2Bv;eX z2(ZQlHAM99_0(gj^A%ldA>(RI-!{C9tvSUo^opiKw-cKHqh$;{<2nKb;l3dHxiVL6 zmhvw7Yk1G2`$ay)i?plqUId=8^|_p&-|!mhS{kDLfsX2|>LOWnX3BWy%P_ch#^vuZ zADK1oIUqS7_*7)vL=v6wjo>ZjW`Ebao*ne9Sv!0(5hqQc+^h9?Xa`NEx1vDPPhyYo z&GGsK?4NPfnWK0N*c>n4R<LI;sBj64E`F6&D;>?(V4pQMs z=xy<#0QxgysAt0r#9NES^8R6I`%X@8Og}N(p}f}nIg;lwhLGJ8#pIL6R=&sQO)pv0 zW~5dxZ2G!lY)XVe+;JE75jZKYjiBM!o8fp^wYdX4=#3O~-$#N1=be^0-~IY|^j?56|PWfsF1Ls(G+-lNDfw7(h_s`9IFZb{ig z;;B1?Lc_QTEFFzk5fQ}{Vfbq= zzxKcwr|@DMDq%>uk)0aI$Kb8*jfxMQC%bapFX)z`_Y~TNyT3mld9y~PZY}RancHq> zFHhs1i5*bSUbe5LwFnXWPDiT=26}%4VnyOY3W3mdNXWUga+-+jm4Nc#t zg#9VriD_B*jw8G{?ZGHoEjC%CRm3vGMRVFObClkIM;XYJlywSs%T;l^-D=6Rd)z$_=c4VVl=lF6*i^(_%$Z0A2XC~`^rUPXGP0jc_-OG z88l@J2(9wl!Cf!p#c(qy#BJI&IqQueW3?H>vMzeR=z^55k$>ZQy~jqYFnHP3yIQE0 zg%9Sww$sWfs`!0X8607KuWVAhR{vh@{96xvt{|Z0L{gAgR+6!k9PtYPa*u~&H7Q%by}C@%XliVO%eiSnt#yxzwdPzrs)1yO@^#=RObSj2D& zde@u!Tt-6@D;y#KebtE<_J|y>bto)ml^V?nmTelXW#pqm0e(f+mI*II(XX-vGS579|BG2L0;QGMT5_`)8-5tbVPwL5p}Xb!pfB)?c%KtM$aAbxV2)qMp(U z6|(&nUHR5NdBxMx#_i?lcHJNmXi2?P+AK`ZDuT+o?X|hj%vgxpoZKYk#nEqC5QY*b zUm0XY?bB&j42D+!`GvHKN+yAp!xLc5*;+3Xm{ZO4GGg})0l>7QIvz#N?XYFs{+fSA z8P_+G=mVVg*qmBaiTW#y{A>orWVD}M=sjwoOr+$P4qHCSUshfe6AhNt_{AX5*^g$o z(ccV69NrXt8VN~VttGvMP4$kG&Y~k#TW+PkM3Ao8h3RFEW=!u*N~?We^sVMHikF?~ zQzGkDrNnwCn=ez7QuvpUf4xc)n_&!uvAfFlv zv=<%M0DQLHHdp_%VXbZ+v(6q2EwXB9*e&eZ0GKtt7?-I-)30lgM~6S6CeZ%F%hDrM zvIQOw)iqJROX+(9R+gLF+SaQcD^3$ml#7wCQdiyWKIn*Go^}_7LL!_!4)Q^p+~s?~ z(qpkyR@cf0Sw-N2Zl4Lm&SA3Q*V*3~tLDxgtyhj=yFA;UGy!WMIc!R?@7TNG4b2$uQe5-F zTfo`_Hyzj_IUUM)o7>#J5#*JW!kx=H5xYYToC&Ayi$I6zqG%ZYIE<$WuwYY;(l|=~ zB|CgQPRnb9a~838x@CCo!;}R@vOD7mx9Q@T`+a%j?xD0uy5UKjzKMU|cp%=>^Bnf7i)$y__gyIq$QU<^Yae3&@9mj{b(slmbtT0 ztK8kcV2)A83y)b4;)_}~bNlWUF0?ul@KZg}nsO{L`M*+JsKGoAr1A;%{~@l@qvpf* zl-@G%duk;%Xt$&{&Xzj_K&Ex$Jv#13qzluxgbkSb#HPQ6DwSG(*E25W@i?Cg7MmZR zI&ruBBf4XyN1&0_))zui$)o?N?P1ZdB`WxiH7kyPY3ke_<`npZIJQV`mu&LALw?Ef?uXL{w~vYf`c`3Y z)BWpT4)ZQFnzX;T*w;u#TtrRL$wH@m$BOP(zHGL5DY2?G?901k_Urp;NZFQio^wyJ zZ5tU>xyH~N486amJ(YggmFspCp;>=dMhrQl)wY>#H@!t>MZm-E0FNlfe&~cakEVYmI ztj1WOA!;zv{E*gV1vh_dIRty22Kw1Sssnswl^Bn0IamGHdP3pSrI)_S=!e<+{A=0i z1Q*sLlX2X$(TRFP z2)}5-F!`r0`X0H%Zb{TLY`@!sC#mNYfpsS8EM z#~0_ktEeuyH1G$*wpWLg{fk7J=PbPrkjuhoa**BqqWS*s+qzS7CWrK-dxx6yfFu~k z9L&JuV{x^wAg$Z82(qmamD_Wh3b9MgB~`Rd>HJR4C%qIkGMdE68;5SF{CX8+Qd0X_ z{%-`XRkkM)PHUnU2Te?c6%_zz%CJBG)q?ObPD#83r&HVgJ(S!0;_I8@0F78p}1k!ntFDb^AOD73iW{*l1( zS88^y(R3yOrP8nT_XP8c1A49yQvOdIqd_K(Pk={1OZW%s=lneg235PhkC>zKBgSTQ zto;haf=CNM(<3a6H2??nsn8)Nt|SE%nEZ~L{^>ebxB(HT9!lTJ(N!zo1RwO$;aZT8>j^=dctx;~?yas+3&athP$|%CcwF-7od8CD zK@}M5`dli0e0hY|5pihBXt8mr^y88UX3>D&AT*VN-^-Nne*OQ~+@?g@? z;QAsVfut9D695PgRN8Y{0Cl|?(E6|4rUF#-5pjbc%-V_iZ3M_Wq^F7Dvzn!^*t(7*n z;6>$wU!oYE%sp_wa`(d#OfsTK?E3XtagDdy!048+iCqUdr;0dd+2z8c0F6sLF%`qV z%!EuAQt6kuzRCTIGbesX=dA)js7}mU>|o0Vq_kr9C-1)CTU)Ri^iLg@%Lf7qYBpMc znoKYo*o$M>NGzIageB*M!6y%@*SPf1U zgod|J^P{TS7b9RfsqlLPx+rG(H^i^)wg$#Z-!4w1kaVXasi1p5Njy?TZ#Mw5mcY=^ z*!^j%JgDb+LCZHb4`$W+EWiSfD>m!r=<9#E6}0$gMdFs9K>f+8_Z@cIYa46X!BN7R z5ED_7`wkMOPnmfeD*fA|*&YZK7&7<6oCyRcDcC>p?w=%Q5(vP;Pr!d#_W$9RiYV|& zm3!JR2>kn?|dk56|-J&B)=%RDJA zN&kQFVZa-3>tE*Ue}Bo|YD&N|S9?TDeD)ve8=n3dFrldb2Xpd&C-mQW2LA7a{(r7# zng3s&&?8)L%X!DXF%y>8jwQ)sTN!aT{0q8^&G*3+LT*~%R8d{Nz`#Ike#SZv(a&E8 z@85=jglQc1CsaInJFxdZH=k_ek8@i+yb5gbkkWZ$^(tGAs3qQp^N)NAOh4>on*XX`&!LFHut zF~*Zk*h`S9bOh3?NC};M@RP_E5N$rGhnO@fWnO10MhaxdZYRAuIa>Qjn+jNxI2-x`H-4)?!_{ zdmFrY#{7_l2pn;$%zF($UJN8RjCuRXBE{MbZ;4oSGL^FB<39tIo?c)2e^2(ZgkS%| z51(x^g~DIrr&ZHz4pUZ&_*W{&zI%q`({N{ zheoZk%rmF(5~>zQ+NIPa8>tzhFZ92@AOIqaNGs`((|2WdJ=iMEZluFv{J_rV!w{jd zHx2lT_69~X(0vG_I0K3?AJm9$_Z(o5yFtrM^pn;ctNdQRbRek-$_gp(}EMofzRbNvzGNJp)uRtm>+C%4a26ba6 z_y63S@f9b@16&F>6bBCU3`r)(SJ&4!ZV8twfik54}q){eomn(^JFLD7~0wu z9RKAH*+Tyx))ciO<)KgCFRn*cZ6EHOMQg*~dVq%pQ6@NA?q0dVkBAfETUzAk0*8`% zrY)TpaGr}q59&I8!@r-5?Mh?SF26b*RK3J|&X1Y}EVy(2T3e3ISc5S#1${gPfs5#J zq*o>gomekShr9ZC>C~)kU1PHI2w61uRRQKhOweY(pmVfBBDV|M`Tp0sT_-83i&w4H z>BQJa@2I`;8I*d%C>vm7<8*Qa@?|7|InJG!D$)H4FtLD+%+G3-zYDL8e0dz{Rh>r) zuVeFfjdeO}j+T?6WfRgPIvO>`O~u#Z%dR(F6?9@MH_w3%HdTlv|IDfOyos4PKQMeB z>5>CapB5dl0*E_CXuwVtvJj1Qq~Xayca{Y_-mLwW@K(Gopsa3|*W`Mq4p`yQ66W-{O!(A}zgKex=fA)gKo^#7+8y>m zi2WZKm_?+4bvlNmVzeTXJw9-o%Z3oC*YcBlRgosft~q35d!J(6_554b1tPGh(jI*% z(_kREJuz~*y5xG*zM0!s5esEP)hX_go2N0aBJiA}fISFS>AY+JR2_tvbT>5r1PSX< ziBd{SzH~jQAupMgF#HuSdxOqiN1OHo zg;m;~i(Np+DmVJd-`zQu*J2v*pJWT3jJDx=S&(!zf6}k5%>IFcuuFf`(#I#4k$#n`)2{|?kJuQ z*uz!u(5Jn3uZsL))os){+kQ9d^XH<|65WQMS@Q9UfTcY<;J+LL*evw{l-}%!ATU7l zr2{KeE;uf`f1YMAW%$g-U-Kb*Tkm_a4)=)5hz*`*VvFq%wgT4&MsTZh~><~%=1~{{Iu23bT`mm6oQkS5loD17DuD5D42I&P}AxGX-4xx$l;JhT*3#n|2x>;vX%@{swa zi0+gI_+OjYnl{{RjICjg&AwP{A!u78`%=YBtw3F7yUYqfaeUrD43p}JS^4=;#u7_b zu0h6~$1+#$^WFJ^keLd{w4iCOlw8|FI-cNy%c|18>4&uqjx-t?$Fn*rA^Uf+>OE=? zwds8W_IUV|Y}e;KGG&=9aUb7yR61iiSR>2Z7|JTK(vg8YrMl>#9@}!UtVgv6d51qz z7p>2K!UWYAHKkAPB5L?7mtHX_r9|$Zp87a8g)kScB;Tz!wcfg_#Xi>ZffYW(j)uo?DdA^tg<{ z7_L9)Vk^gjvCtodZVs#M$aOA<^&Zeuy{AY)%6-LAbLf36ZY>j*Z%SU+r>>etpboQv zAE3iz^YE!xn>`gb!$nIRZI?d+9gWQ#QFl%C2CJTuMZCf@?la=OSdvK(rcaFtC7^=` zsYRt;w-86S78e|&b#^qAE5Jx6eO&V`o9r*!zW9nbO2tEL{(fMaFQux9v&%UQv_*9g zoEvk}w99j30-AXF%FRW~Kh)DJwU5dFLRA}w^7u&RoEgcj3i#xcGzLWv=@}kZb2`YB zD%^KU3*yzU!}z$Da4M%?5HqL8g|kFvg}j-b@f>|XR+bQ+$F~x3DaCBC0$yfx1 zRMbRDnnk1FH#;Xb88mQ_i`?np!47#*mL35>ddj1 zm4pM%=YX!=WVYTtcj|2KJHO}5bZ5#Zz~S7G!Fdyq?yqCV=A)y60+)a3tIfF?-ZyDdv1?d(cAyA;bDSOVy*48%G1p|5SWj zQNaDC=!=GNC}1oxVa)ApyIqZvcwotYtGrqn+(DXF#XwdRCH&x5zCN+bLv?f(yu&Bj ztuICDiJrTR52^buuuyy#m0TUB>RjRZ_?DNbRB5R2fnY8Pkks(Xp50&t_G4+9ATR9+7>-J^GPF^U1Qj_D6nW4MZf`_ieHHDYky@PNvNd_h2NrWc>=IN)+@ z-N(HTYl8%+9@Boe(Ed7iB5>5LCDdLfyx;IOP09~^&c0C0R~C)Cw>6mL(u`m-!!!x+ z&*^^7d8q>U7^;^bi4Rd7@tj@`%{Uv0PjIg`1KX@7Z}HAkbX{4Z=Cl}cn{6~HR6OwV zaIcK$VcG;*UQqg?%NvI>@7%PkAz8@do5I9^FH<}fgZJ*^B$YkgPaKDKJ3gzP()>~$0#3^87FFeWa zjkCqN+Y;5xSawcWzx=7i3+Bq6U^Piyk&2swY|OR%G)2@$BV?uqK<^>#T)iwkUQi5NK2VyS#UPuzd@vRJ)z$3XZ*Ebbrrb`ay(KWxMVM` zV+F%7WqSrUpFYy0!w-Q~!Lj*VH$~3k507Y;M{|`t8SRp1A>Vfk&?aiVq&6(4e||bH zP9(#LQcGe{IYmSm5KQ3|%>7kd!$t0UN}~mo{mx+|CKlC8b^_S{)BXKM21^`YGAejU zP-XwmgxJMjzKh@5Mg;x6SamD1L|HZudZDS8UCQMbECH-~wEh-#Y{V~jXhdpWo!0~r zNl3bxG+*!2!!XD^*GjyY55~p?_3@fbu)6!$E{uMDRPE@{&4ovk`8HS2VA0nU*XS_h zPAqJirpcKhbZ|%8BJ}y4q|3lT$)ah~S*=>k?Hc5dT7Fu+&HUR`BY_@1(H84a4^Qmy zWYM|%?3KYWUnfo$yc9cb0w$f&EZ(MozhOL^W>U9{n+%rF<=&Pjt80;oc}AZDrudm% zm9r&gI*=F#ML*UPNL-3JKAaTHr)Chat(9TpUA~OGE&Tko;(Pa|?cpWSc=329Ti@4$ zJ}*8jnSDH>8s-46=W{ez*TdzZWeYBJ07JT7?%T6w_Tg498W}a$c(#?1!t<{fw$RyntV%mb z$Ax7zm==Eu;&ZCj`aLXon?ul0PyXYy%P9iQt~R~l7f~q_*J+neXge1xw{xqU%5TD4 z409Y*Jp_$cJQs?3mF*@)ln2IJC*m$J=Gve~`~(4#yhZW@Q#trba_Ft9mSkQ=s4GA0 zkI8_-lW6vz>>Uh{h!YzDEl=GBNFcS4oS)~(JXdq2gc2LYl`z2?M=rY(|+Snm~)4F`(^f9{PH z#_VCCI?m-^y%?P|HJ_3RdU11nWbTg@^NkWEl468kI6EgOZxi0iJ42wq_^v(?e?+3K z#dP63A6GPHvI^@>sr7;CcE1oX9_3<4qkCf$N?tqfwAu`3jqz79j>DQ`4o=AP8-EIb z5JT8Cbw=gw(+`6J%&eqoNws~C&j4f|$2AF<)y6Nsr>Bp?>a;FH{S2_GJeIw=u(HQz zm@Xy;oy>}OuA_aAGk1XB-iAG%Pl4HY#(7WZs1_vFH)MvUw??=IGXN^1X8C2Frp*K7 zE(`LlOuTLbQLBDzLokk|bP5ubVCCgISyMe$WmQ?`h=Nr%*vKyb*f$kU(1SKha~FR* z{Mz1=Cti!+NJ$dEbP%yrll zO2tRYBr?X_&1FePa;s0VGksPS|O-Z#Z535XFV0EsKFFSd}pdmpZ?wVQg>oPzB}W{D#AG49#+DLymZ zb0J;kOVr)1qtV=l*&;E2lr|pmBPML8p4wme*{Y3$`cbBQRo9FloLUUJ!?2d1>^Sf3 zpxe9lDq`{J@_mw5sCS&k(n7_VvWL#PuMY*=ZU@k?=;dtdG?^`9J1j;on8#b=NrJU> z;s`5PAoQdWZ^IHTpgVYey29fW3VcP2c)u*)GJ7?~IeTN+@Rlt8>MiWxv^nLBY8R>S zCT;mX!qJz!1=YAJwjbZui?RwEWZBX- zS3I|`$qfgyse%p*>ZPo=X^b&sz@ro~##1mqFpN~0teBXpD5MNBm{v{&Jz;Pme)a~L zS#Xz7C!kkU#|Yl;UErz9EJH8rX-$IrZ3sG;YNyDgff^-%KD+gKvDR{PdvDyn3~fL- z$=5jfRBsdMG+HKiRh)fgXU$i+jBK!%pQVa@zEqL3;gNNQ^i;nl!Fy+iu02xK>!^E2 zA24TZm?7HvZ6Js#=zO(tt;f`8!5iVpcs=mvu?h0CmSOF%K+LV(>356#l$|3rEM^Jb zV>k*jS)_Lb6siNkG?{B*q>?M<}1>5dEloZ zVa77sM{=?E-}v#_DU^IRWNqq`X2RS|#;$aXdU{ydFnVdD)8t~Zs@B(dslkpt+kI>O<^qI0VuSohDbrTqVB53U`w1^s7JpL~!29(Zp<~>o zSl?r$>qQ;jJt}@0Hi#YGi~{upxkg&_vd86E{3V_b=HsqVDS|gT!do4$Wmy}=Vzb*~ zO~s6t4_a9-LhaZ`=)&Re+~LY$mJQd1yYzab33h++<>R+}T3HlS78k~BWbrgJKuQ8S z47XkQseWU5>!ROT*O=f)ppJwrdzXWn>+h#ZwR23Jql-2W|#i#^=~ zvpH;Jlqi)p7eRo=99$Mh`D@fk)&XVQ`>kgG|sRQ%XAnZ_a zm=wHs#bUwccO5dR33H}gpYKK*D>PY#+>d5NFeqh=vm@q@O0A&(i!p}xf;3p}GIP%n zCq!*6P5W=)Mk{ar7JFl#Oq^Q|k#dpv$Is6UcT^$Gi+*WOO`S+tWRkh~2GN1B3)uXj z&_EXQhu8Vx)7)DAzOA8dO2e|(p5Hc|HOaoEA<$cWVMpiU%u0-~ERrcDQk9OgV`v)h zuaof;ED(=O zltN>>QkGobv!O*^Qk9#JK}_OUfDQ3l?poYQW6on|FgWPrpcKlkli1&Z$LPn*Y7&nR zA@Z}vlW&pewtn+Dg-)Vpr;sxrqa!6NyM3h27cYukkT66Q{VO%Dw6^7+T2=qEBJB!i z6kPH4QESW7E=hzcSrW#_E?5lcc}Qb_C!U=@W_ko2($5YFzu_7R5c0bSQjK&yb9WGn z=b-nKS&j~|)`AueWJl1rfl+UEPh_0ckVZ{ZZM$-q$HX24suI)E2APy-Hz4yV5lt7@ zWOM}mfjc23OiPb`$1vTDI&g$k36a0NHg#yq=--$hjY z8?C4$j>BIASJl^bs7gpKfmc;PuJQUPd9km!k!*Djq58T~XXuN2$R!@azq4e`;Ls#6 z9hq)>DuhENMDM3@WS^q9y8Rf@_GoV5KBzjHN9TDqtpA=LzsHiO%MIiVlo+CZKpeon zr%1vTO*~R!SHXD~<#_0Jl~I2!_<6|aAgQt3k!2>*?6m?VDNH%k8i&1su!EDP^H-b@ zMgclQg|TfN+sb;D$eLB1*CVKrKGzmKR0>x58>kZf2ck&`sJ6)2fP?q9ZeR#o?Ol{W zxF|C+#xElp$EGiTKRd5!+1?9lti>I_ktMr4lW1?%vvF`#jCc+@8+v*_AV<};X)+b) z?;6}-*8PQSpkKd$KiB@x=ObiAJ*%UVPC5iJ{7duxd^%Z|kcGhvdHq>T% z{LR7lb0%bw>z2iW-9={JyryShXYM$Q5Q+BFu`orbN)*jD=cGnp6ns~DMa&XcqfW&x zoTL0PUOlhP`hsw&S_t!V>SWO{;5{0p-`^j>uq*;ydFV`gcny?aonqR~)&z6&vhoLV z|N2N|Y?$KSd;C|Jf;WeJPO2M!+bk4>9{hSjhwH%dB^`h#oI*3y)Bh9Fe`_9Pe_eO(DGJ28{bz(4-G>xvZfT$xgYD&>x?kQT|h%c-BgN6!zZv za6hjzqL{^Uc)t^-WNPd#sWF}xOlA75oI(RFNkmEDcP@FxZyF<&O6ZSM&o1WRcKV99 z@BqpUzGR;TJyA}D#$vz9A?6ANY-~D*j_J;+mzOgK@z7G77NwOq(vqOoCR}%ui{4|@ zgz~WN(&fw$|5O!YDIxOMaoWL)4le`fKR;q#jLK|LHcA>G;C=6%k8?pUA@RVd50ic| z@yEZgnuJ&p{=%=&{G*O=+nedo*3xB`=bOJsr}0w8c@<$J+U8?d-F{TGFq2?9HDYeEedIWchR2L9D_WshA0h(t!gD*`5>~#nv3Hm~$7BkMws3dU zTnC9+jHd9ECkP2K?+9G?;s;1q=S!dV>;J7gUr^;iWR917+=n0aq4?lTThYE0$Br+? z1bZuHUHSH~#&pOHpxdIa`NV)<-Fu063HYYoa1ES#>2Hnikohc-isO<)r_)& zM7<}Q-Q6g&iY78ZyncEFb*oV$&Er_sJfTX^AC13@D$}}#7ZiY$21KkwZS!Uf`|;Wi z(}e^FgYRWl$GJ5l!&;Fq(hu-5Lz+H52`kQB5(Cgbfp*oEffeqWm&eijT3wbJSJ7mx z)YjPjz&xJs*K~sP&odo5hTE;^mwwJZsG6~JaVY9NV_$TDEG*6keZiL?@M!Js)}$x9 z1_6MbApf0ASFEeVj=2qA;7uhr7gT;hsNbK!BEX9NNWsTXzqEg`kDfZEb!C#so}{CdDR%#s;p29H9s|Eu}rFqHqYi(R1C? znrP8Ec%ad#YINJ&{TkSAO?RL)F=1$-+Yup~Lbe+;sPB$~LQFn zhVTRZ>rF?@5k2Dvo&+@rB(9-OV%vUSs*rT9z!8C$z7lg?4--31$;ZTf1HKFp6p_yA z>%s@rYRL-O)_Kz+h1%HaA>NyTkVP!1RD9(RFb2NkO=usCY6!Xx(O-Mc?>R zRbmcJW&ou+!i6}9(Md~vPZs`b zvysOg_B;?w4wN_`y`IMDWxce^@+W(_otfLj6G*RWF-@vRq?zdwKZh4zFsuTFS$PrU z#~~)6npi{&-kzbuaPsCm_A$lzT@Y?c{ekn25GLy>&2(rYjg5k3X8}ieGWSPQ?li)} zm;0B}ox@i<_PS{k#yHWVcH_aTB2$RO?R4h*j|Ql!`lstJIG>I)c@lkrib#8fB>LY+ zTEbLAq!-Sol&3<6x93n{#1N87fGG14>@Qzuj`5h3YqjWWg1q2Z+BHNcgJ_)Utpz`a<8Iywf8EuY(!2`3dSd#IAWA2A$*&A)5z}2 zq+t2T=}3g|fA#7E(7j0{8)BtoLtH2Eyq5bWL2I1jc*pz@{M?1hx@$x(A6QSPb=~t2 zC&se(Kcp@N=&zm|5O#YF34ciG_Mmdcxa#VS1#Xl$h&ipQh6Rsu*%If-My^*5FOr92 zALD=nG{f(?1)Syy_*Mxy{Chu;@H;m-RaV$9)`eD3DmR{(Tw#U78B^RG)~0I96WU$e z5gGldUomK0)96DZnz``3uo-ffW|+2bqT`%_U;c_|#xZ(MDzC1MdP^MRPoZb45bgV~PJlN`8;puytkoG8Zc^LSOb7G#UCGZmRC_TMDbrh)Tr+yt?arB1V zb-UNFF+ovR2hp{cQWq&Wx70_+;OOw+k~4dSIigd>lqiyGSGpv5GA!Y(yanPAk%+90 zU0CbqsRsAK#(Gq&gnlK1`UL0Q*{YKchLBxqksS9*`nI@0;$_Bo#r?+n5yR#eglbYI=yCSq8M0AoJ{PG46F z8tBzIsFB!hW*VJOLwWor;qP+VJY;fiGv5{T^*orUIZHk#`jzMlu z_5CR#piFTpIEqG3z@5@DrKeTOC5YCop9msm{cm?BiK%~X>qcLz35sO(I1lCk$Wva zbkPx}77is0*Kc3Pa(E1R`btK@OHn21J)POto$+|Ydk3GRZfh6cw9d_e*_`#?4Lh?p_nMDNcK8|^fydbOI;J~XcFW!pR7h{gu6EEY<$GON zg6Lh-tC>TscqG_u-gVQm4mm+wyl}&l$!+9cR2Cnd0p(5|;26IrGNv`?1yK7!Cg5Pa z*2hyRC`Dx+jgqH|47Vjf{n)rq)X`OGjJ@RSyXFq)T_e08Yl8r$tu zV+|YSvd;Dlorq%3Fk*nx-r#>|uYOKbImpHFx8urhv(NrNDDF_^F>tEr8BDzvg@4%d z`~o$B^;cTsM-MvMclHP%7LRRLZ?gUV4o3)JjfrX}NEBz2u*wR4;%Z7KDf2YmVGJ5$Ezv+}vYOxz;6> zU2fS=x8*C}{|$%u2_M1@kwX3^f+fujkWRa0E_XzSylZTPLzZ~IaIW7t%}@PtGK8KK zQemp}P1)K@e#=fbdSh$%fGZM$B+F# z%l)>T)(Zc+Z^$!@Sqyosi6 zZR=m)m0pn!#eU_Qqw1+`7ag`#+nI7HImOG)6?eD1lxXthU@y~c8iIV|J4_xsFJw$7 z^R89;xgcbeh#7j^4kR%sYchd1y~B zqPfj1)4e|5qJ=lV!wgV?t*_uZ>wGMt>;%s@^~1?2wN9nr@d;?zH=mKJ^!Qjb+Hyzv zjK<6p_QzY_F%kmp*AB4L)0TxgQ}D4ji39EoSZ<)Iyus;~|Jl?Khm;b#IS%0Hbc3SV}prHQpwrfe{x59T&f_15+YU zUz+x$%Cit{B!r~!2q=Am#|G?Q7^kNdXFbk^iG8B|vMjId{yMjxjZ-wngK2;P%nU;v zPN*#M-F;pffdpjCoV4`WQwpXj7j;LOAbEHjPh@RFiIKX`zm`|!p5i%Y?K^i63u5{& zxei7BIS-=Q^y484h06&}B27KAtN4qj@0#|5UCcQ67vys!6`*vHW!bvt%2Yq@%#Uek zyDw0lf+u+49RnHPW%A)r0$BCJs^RxYHx0#NkFcI8V#~$@$dbTGTDN)+Wiq7UIrQB*>w;<{Ar{D{!W1W zSZ)(5gjzN0@wm|_rqzjgFC1c^ARN4`*&hm!SdlLdwpw8BxTD{`P4t_T%;rrfv+5X@v?0c~+9=qa& zrZe6LLbC1zr1-ktL9`M7q|`83{ObK$kc$GWfp1gL;g#5LtnPZlbPd4=wY%I`$f{~( zNyl6pz4bbR>DM@2Kvsy9`$k7l?sr5+j`%JR_JU)3QRy1#fp!BRpTkGP_K-1{4B-BL zrMr3$Vu9NP5oCpDnN6o~*tLQ9XFb76&3+H=nfh7l{H&e-<8tp!b|k`S@$n0c&rl*; zd5qyO=d3moa5!UuA2!Zggpxv(_26@jb*C@HF4(5A7F6W9{s&M=^`vbwT_(wm+ST

_X}X<6hO4)Yv+&-KWG92zBn^a)*MDW2P3u%m0f}T~Bw1=Wh7RaW`(u ztW5P*88`X0iiu2v5nEBhD!O;KKuFMLz8GX52hUHsSfcjIIQP_IwY~es=-14W#0+kvz=7e+YRG#Zgwn` z(W_{-@zQSz|ceCb7rz(?Z<@~Xd^Jq7)W3N%j~Wc~G}oYr{TjnBV4w05 zlgz}R^#JOmViANyfxK)}!upk-r5@0+Nx z4!X0MKP#}%*k!&S=dwPJE6as#^Y<2J$`Q7+=9K zDj8_2VTqrI+(BAT+LYKkqV&Yp#ibWR8_e?(R5+;(B(mYw&i_oMNEPp7>X3r?0|IO6Yy)6Pb{duk_|@DE-$T%YK44} zsuHN8AC*)K5`;@2*0>Ybhl8HY8O`tu>rVy@@gPKg)WaFfy>*h7Yg|sek=T zrchEFhm@Of+WaTUy8rE{*OJ?S63^44Q~JRk+2`fa85Id~0o!HsT`u48juzip26OSh z03w!-TL|x~fEj5%q_#I#soRuG^z!a3fg;e*QC+aguv;u6m6K(^!rAX*J_Jte`j_0AV}ITL*l^vW^H@pAKM3rnxVnlG;Nf9+@SKkes7QeI{s z{FTRyihVJa5=xw|t;+~U21~l6mmp~jfr45Xr~c2K{Zb@r?R);}1W$j)GXuW;L%x*c@?gc1czW zKY67`aaAre+ZZ1FH906St+4|?`^Az7#NhO=-9HD20o80j`WkXnNpJJh14z>&Jk^M` z`{33-aC$GtHA>ift98zxPRLBWRkKjIXiMUsEo+>MD~HV6o$J7P9J&Sz$5)^>+Ah#K z7fx{~u3jTAbe75Z>H8M=|NPC=7On&fMuFEs;}mVOK-a7CJS17H1M$}$@pxZnd08Au zPFgZfk4R7ANPh9}Z;d++x%{`64B@k(w6;K~g0EE=?xraBnj=wgLEcNh%77x!$}0k0 z?Fv+)J`da79M)kvj;53Vpn6^yR=gcNQ7Ita&3B9rxdz{bXB?>}3(vwy$3J&h;*yk= zb+W?uZAqe*#OT?IjTT8SB{=Sxxw#l=T&~NceMZCsj7sV>cP_j(qVr$l*pvgQ$2Tl_ z=RkY=aiWlid#A3j%Il6$n5amTIUY=|0~e~PiV0;)RZ{=3rRgD_aHkw#wb%WdbLXR- z(Ua1Mpj!F1kucJ>5EJD=>bf9ie z2`6yRZvmREIIte!7|e*3WQz`Ejf*L416SWbe$%Phf5Fmvxqz`z!2ZB(j84b0ZJk(m ziEJUSh^K+W`lJ7{9oicHO+}VJLQ6>Zw1)UztUeqmvn`^Ned-_)h^RZR&8O!#GzoKf z!~^KrP81}nJ7#W$K`GLp@+(Dh%^0SJW^$%CsX?^(`Z4W|&obqu=NKDND7>sm1;|&2 z1%l}CT*&hVE=BW3EtyoUW*7fpjXauL(@XvtAInKqO*bkd?1^J!ix;nRy=<|oMc~5e z%iZ47xmvLg^G`n9kG2TKtR@?=O46*-3|HD(9jA7S!>B)LHL_&+xFv-&Z%$$*>zPVT z2839fB2{rupq-XygZpi`Jd1ezvIH*p#{%9t#IF3Jp%e+>74U3YmE6ZQEHo<6m z(CO0YjLacphfRxtVZ)EgErABG_U;1|8weQ{AL$u_g(|=k7;2HK7G(Qai(gznuT7v+ zxd-VLp>`IvPA^BQqPYIRmt~&U4XiU2NYN*wl&LCCfW3hK#>9a-G!s;{K9$NCf}H1; zhUa^sFm#q$baFS6aRWq_WzIG3;?TxlS-xGc1Y>LO9!@+l9^7o0Bb=&$5!3iVMZGll zZ&NYE;43yNU)Jfq#(bjy4Tbs>5}Y$jGt|rBVnnef`f?sD>o~i4jJ+M+O7!AO4Ihyv zN`$_AhO@rX*y!9wO4}VgEvXHE-4>OUKf^u}ppBo%XhgI7&=bZ_^^bimKA(S`6VH5C zkf;j22IWm0}7XZoi_>h&sV-!+iww!mMm^1D5WjRc3N zYy7lSR$dYq%t+kdw#qsc2l#|Qj2>DfWT_b+lVYAqfwZ)fe^`eM+(f%R7)-OB;sWS1br30QnQbwE!nD*b!3olneXwi!)3B) zHHO?iOq2KQ40|gpdc@g7m3)wQiQNqL<0pAi3U3N*Fu?K{%bu+L>Y8-ZBIQpE=Uo*J zu+qlGmTe}!0+S!K@sKz7h;z5QqS7^!1p2Q}*TZsJA45ZkjM^Sn zjC1cONpGt=R}J}!XRZN5iO)AiSby3U8Dns(~1k)u)&;oJ?hL#i>NeZza9TJetf&x@_ItZPPl zxaml&i}jCUqvWrTPEn(*U^GBo-E0GNG6C&O|6^+vKxIfJiZ4eH)SA$?DE&(C%X zu2P6pRmIH^F>~*(V>A2YUTN{2I#q8xgTk&x>TaU-?D=z4C&x2`vS4de^JZGWv+0My zf4U(@)>q@cZMpt$g!^@KYdrg*Qo7de#Hf`WOo!WvPt8~iN=fpL$Syd(IJpLtnDRrK zl3xu6m5If%%=%*oUeM{oj*$5s`(k3XhQi>7IXC3-w z{>&BE>7jAWIiAO0CX3P$FS;=Hg0RhUuEKBriT?MAR;7Hz71%V2)9V3|XJWA!xXHwV z(`Sa(jj_bWqshtoR6TwDhYtR>*SIY=R`w~in}74 zZGmA~F!S=6`P_?8i7x(Fhtssr~;)t80tkJyr4-?kFb zCVtlnj}m>+t{rIB_m4?9u#pPl!_2b|V#K7ZVdEn~dfG$R#mn~7st!gFYfBYNh|axU zeSO@imio!MxqNpnS#`k|1zthyIur`19SSsGzAM~+Ukf*s>Fy`py{_DDSt`+4Xs{zY zf(VaLIR7|S13uXp{SIGZ8Yd9y-J!VUK4?grH|_Gpw^OCGhJg(=Wl8%$r@%F+%%ON|lh&Q?v=HghSPc%#Mn3 zG4suV#Llfb)o(<>T)U6cBUU()A<;n>{iIF{k&WY>3N$2e zCEh%Ll-bJSng~P{Ij^ATzzX+imQI;D8QN{ACCK`7hS#}o+Xufk7ow?NvcoapF%zgP zWwS5whM{-KtRmh82twH8EO8T%LA=`2o>&$g>Pq%ikJD;`v>h`pZk6iY(%Zcxwdo|Z znyKisf?LBWqxPc9%8^nzeyl<>8jBJvd(50>i{+wJB_wR}4}c7^cB}25gvky1)|;WE zW?9-l>Y_UVEn#UlGYwd&3?EY_P79w z#Q8)<-l37|mKcR@(URC<{okS;}@if*lOUWHj*s!RQk)^;8Xh8^2eR&j6CH26#S_V zgK%t{I+DJ^zobz_P9RBjc z$D3C0%VsOcOGHufV}Z#K7t4?^4TgC4fx; z!c%$K5guCf-j3o9;g1d09f+tygV=35h&;5GJvkFJt7|ho^tUes^|O>?SAiaF&KBg%TfEpil!XbRPz}2HeO7B)$K#dUDyS{<+bi>z1IDf z;{WPC$AjuOw?HK>hupD6s?9}YF;G;eKcBNI={)4xLa=QHwO?=B9;3}FL zw00;(>!s?)#jKS^z^}z(o;h*gzH`2@>nD;Y2Kws6 zlvNGHV%!qG8{jMHzca_^5G&&8Xhq?i=Y`zM?e6G_(p0m%q*dFV11&CL%)S8UUF$>J z^mm7ErbTp4a#oa~@xfcoU&~7|`ny+5^Ii%;m4RkNg?|lv-BHt_-l2wqMFr9xmG()TtAJf@l8eUbozS=r3}%`Ulepk->CGix zk=u2dF^WdhNoN1!{xr+>Fmf~dRiJhHL;HzGsh6bPgCohcm27EUq`Oet1s;n{>)c>@ znexAO&}xOY>^mHxb?mfG$68>!C5Ih5x#{yq>A^;=PP5ls@vREdDGR8>&pZCAlyf)d zVGuaV3r#1Dqf>e|5Vb+&^DfXC*grM7P?gSVYc%QpiWd7*r2kCR)j_Hon4aDRcyF-?NHSo0Kxh+gQ z+a6<|lMk!piU}IK;?5k6X7KYF0qOQoi`=w4epZl@AgY(o*@^ zD*%ku*T0z&t1oB8xRw2|qA9@OWFcImvQ**>ddvwxa?ivu5fVVTF( z`AAh~G1=x3bTvTw2hHoPqEW332F28czmq~wd^0b9CE`(eT=qElPC6cPV)PR6^Ouz= zBly=8*_iD`6;olDT>lz}8MsmgXl<|#76a0#I*9h*tS3 zfPTcDh_3yF5`j3E=xTDv&oX_PZVaXhx_~RqbiMwnx{a^RTz2v^l$y}izCgmdR;n#l z;(m54jep7CSs2J4sxEnfE@%KCJRcgN3J1ELZoz@=N9|8@(jVJ2RX#5_y=5si=>8R1 z1-uJxa#PrM9|3RVgfjNANQdfMneX&QkuvVeK@AZ4R|7=jSI5Vt;)e$jYV+5Jov?mreku#vQDfcqN3ImHfWWm_m}aGlv5dD-^1go3l? z2dX!_+6KZRXI({?@S7tH43&0E*MOsJc~REKi`DQ-G`=-mR=bz?i|ma9Qs8LM?HnzbzsGeP7~d-ZTi!9 zXUX;6EsEb7b=-kuw*%bO)(Y{P0n4HtJ$ONH2Vc8o5rCktxg`7ld(G4I`V^Wi>7#pp ztd*Xk0jHW$i3KlVa_ixn55CPnpySAqA4iH|K48N{zD~?@_Ki;qo~kW}$*Ak*FJ2w< zpQIt65iN-5JFa58IG5lZ`&m3i6vm$SB^>oiD0)d>$AH`&N0S*k@Efl7SAVzACmi&@ zGWl21~=b-f*;Rt^#*K<+YhZcxR>fJ4jyNi0$`3TP3?-vrC z(~B>Fr2PtL(eJZ7g@)=rW)t2bo#G(><06bs-`EDhzAQi=xDdl#SMOFkQ!=b}eVo3o z2)Xpo!e}5cO3mN;iH=!MI`PllqP|PpeVt zhWKV;B%FwpVS|xlMT+Nlvmag=Nh=I%Cq}nqDCk%IJPgf)`P?(LlZs31_yUrAF=CWk z&6o3=R)?S9IKDMnmCt{r&-(SB`OX*Ocn~K94_F5S>p+?$M>*`S!atV19-=@yUS8RC zF7g(}86lrCZk&9hvZCU&5`64uaX1G$U!HwjbXHn+C-fZH6JkcM#yYe@%AzISMAz&k zZcltqp2S%bPolNv5EJU!`I4piL?P;)AxD`R2#VCFSwgyGfcB!gg+Dj#>O%5w%7BmA;y8snO^JL0ZfWwHuSt4a?n#qcp4|b)k);v9k2oK(4 zk~!jfA=dd3MR;HupYD4~(VfI>TW*n`YL(ExzJb;Xe!f21QOCXQa?`L<`gO414rOFn zqyUN(RHhr52j45x%g4meEpo+jMlDdFvEvGl$Yr%q&g7f-KBr z$bhV>Px8|D{%9I~QV}}K<&iklGeZcz48gX#Civ+5skp;funTAYU;9!Nv(?>C4sD$2 zV2^5-9alZz>bKzT8kL{z6olJ})WS>lz9jaiM3`^juTk@35KT6zM3o)cOrP3jgs$vCnT)D4k{Q5HG z+yjyA@DW{-dMOeZP(J%|4Xiph(FicKSGgW@wrkc&d!!npuQKudTv6aT>?ow&=v(`S|)&X2CjD zkf&)6y^SBfg%iNt;o(v7r^%L;JKQ;~I_!>D!sz(?B+KedV$M%VKuKZ{bTnm4TWcM& z`xO%?Fvc+^t!2q8lI2|RZ9{1JAkA}`0iO%ua&&C;PHVyoO7X&ZQIuYb! z3-*b?9hfKoTF>O8L0kDTah$0!|Bu=r4TNUT;*Q=VacI19jyS;{$V#iVcucc#x%Kb! zG@YHJJ$aRD5+LgPKy*BwN_HB0_xxK^&H&OoTV@it+tyggRbv6*-8Pl17*T3?^{956 ztd?l%Vr~sOtbd5iz8iq_zf3S(#q=~;A4UwzbTlf>EwrVep(5uFJAbbc*L06bIdY{0 zCDXv&no^!~7vP*w5mj^vNE{9VC;xz~Yh)xx=ux;FBPAforHdQ7OzK?83=P5Os-)d!bsK88dtZ&c z&b#1y7|p@CMVEOni;N0c9wg@PnhA6IIU{nCRjI3~KuDwz0|iD|9a62CR>DcmXg&kg zfoK0hR@}*R1#;*;29MB)$Oqs_nR8V?Eiat+EZTXGC~R0HpiBq3f? zsPTzoQ4$gxBo|lnYjIqgkx{CvVqF8=?80}1KmUM&oXPO1aD1U^=yzY!9`lQFcjk}8 zGwnwtb$Q2TCp)>h{`U)FGMp5hMef4eaWa)WnIID2!-s#^4b76<#3|%egF{W#K#KyVT>m-U z=QmFs*N1OqQ`)4jUU|?_4uij&T8K1-j@O$Tvi_O|fTLVbPlbwnPP>tOwxA)5Hj;1Q`E#ny6Q89aL>s$9 zgAc=Azr;PUp|EvTeW*_LOXG)q5gX?7IyCw`<}_HpksHd!?y|whkF+*G*x9RTnqWCj z+Q^CWTSZ^Odc^^GyxYKZcU>r>>)=dt0!9L=-sN6Y@PaG6nr*GCtq)#D7yc7I@J4X| zO@7a)iBO;4D}lKA**e4tvl z6`6v$-EhTT#ZbNcy2;LNNrO)MPAA7Oy2w_ruX`i2Ry!Iz_wE0*chzrEtzDb$l#-C{ z5Dzg(4KRRGG9w)`z#*g?=@2QUR7zSxI;9%~M3C-o80j97em9=;#`)gw`o4eQ%rE=e zd#>5@JkMThJuB{Y??XeJO;geTYS5L#tbcH^?xe;^VC=(9Ith2!G423w&NX+Z8zaEImLf-uA{;XwyZtzPu!?(ZEJ!T5lQzzKi$n#e_!)a6&uqhU)nf|%{OO8aS04<&W2C^3 zlXjrMVwK6KO-{h{9^Z(1be;($a2e=AtUet8;9Gul;-)Xhb#0rcQ${c?c2%1njo`&k zHrJ+u5|M11khW(X!{l6L;&&1?@W3yqFhF(XnS`!2Q_R6JW69Oq1oZV`j0Us!i{VHP zhrrKXht;_}kk1yZbNjZ|{LL!! zdJDKTFL7$1E7E-v6&wVM*1N@iYz`ePJSA^;!=skJc0a#sSc)DrTZ%B8ouqr~LxL9! zJ~PLx&xaer2%=nf@h-CR>)WoMZ>kV|rse-K)OL{p4G_D6#x-S?{MR>iyYf%bf!I?ZBWf8@S6 zNvhmD+0Cn~cChqDZP@swPUeIx4j>x48`@j@XFMk2`T8b=vLBID9>-kCUZP8Tg{x0~ z5(UUK=n@yC8c_t=U2j`3rNMXzJd)3xf_@r6?# zM$U}Uj6d-+1PmEWPp`zM1TKE}c;z)$Rc%!@b8-`e1zX17b|v?-drd+5KHM!g=e9{OvipQ>m1g(PUQ*(LVmL7S z*vR;WZk8xuxk*KDnQ&=k6r_K=g&qG^;g9WljR)J>TG%CO&&`x>LT7ZM8crf!cjds+ zgQ=J5?mOL&XD7 zsq2*k1hrnOuWx`x)VXKrkN1SLA$N?I{Ap_M%gvGPc6zDka!mPdtI&gq=_hJM@&)w0 z<5{9PyFKfm8o`iMvhB!XA-c5`(8nA}AiRv-O0Ib#e*lFiz(nY(IM*c5)(MBXDu3X1 zVe1xXMO82Gq^UF4YuZW_PY<;vcYS5Y)oVEt2ntfd1Fxg~Lg4)hIM@E@`Ql;2BE=Q! zE_Wg}!W0^@XW#d9Ju7sH=<0H_f`-#~@bWqY!~cYg4Cz7+B&-&zPC_prN^0Uy7Dcgc zQ%!oCFMRPUu)A#mZknv>BVxj_4AQ$DzKQ)@GV-+R-nm*pp4K&yl;W&d=p3#1gBWK+ z>6hkgZ!)iuR<_|7d4UC6)OKgHW8GG-wA|$R}9D%m%oALf%7P8=jL2OE-FmE*Y1X zVG@epT?LYe{#lnIwl|kCBwjMas?c=;`vaGev$)tD7RO?y*2^QlW28{lz4b8PVZp9Q zM*ZC_a@~EHTf&7V(1T$5N9elbayah6y#g>V455#kDk??~PUa~i(N!z(VCr29rU_&= zks%ot*5B^gEDX+oFDh9n`7u{uQY43{z3F)apLRKuA-SEL&)LKnYiPR0&+DyK!t%`X z3qd`%m=l$nHUmG_%vQf zF-s~OrgX|SW0LaNVNEYd5sA&RBOY`y`y>nZt2Cn;IYGp`3q}S`?kzzdd1{G4z1_tR z`Nb81uP}dt)8%v^o|xL~8g7KG-B&%PagEORX}(y;t*w9%kH$Cb$m zP%nn;NQMm(=Mo6A;a{oQ6+ZWfkG->Y0gA=#J#{q0W#IGRqP1pIilg@9Fi$A*2scr_ zrgJF;4UJ+}SwT)KS1nC>FSljoF;!_fGsfuW@*3=%0&Yca-@B%O4=0Tzt;+yvV!3rvYo<=ZaIQ-A{ireuPSm?QN7hM<=(tIGPwo&gj{ zaykX1`?fKxV^Na`3#1@i3+8z=PKvSft|Kg&ntbp0fS58j{R61T1bgfWt`3vWs1#0b z93B|mBDr2Amv1$!5Et&0wSQo3l=Pt~EQW^vk*LF*68QdHN61bw&3EbvsSuh0;rs*h+nu%Nu75e-l}-CL%FR$mehc|EExkIIQ!>IHA{{73U)2J(KY^v}xSy}0VW zUdyx-wB>rkh1wAWPyb1WpoyNJh)a>4uI3A&l^I4(ne`~1ep#3*@00j67sA2Xc11k5 zq!7a1(sp&<0nzR(Uqz888nA3y>HQ>0QqLc`8XseY5fEWZr=b3s6+)bX1s8$-rL>KM zuOcy@I@%4jUaSzUS%Ud8T%_KbJ8y*tMx+0zt* z%r|3MkY*lmNX=au|1*o)dH}o?;yn9NnWADym~)TfB3b|;2X+ywFB0GgrZe-h)}N^e zOFXTt8N7`agWRV0%-(OexW5pKiwH}nsYe!v|6vXOmD=LQS^}FNCkWIGIL}OIbWr&x z3bxuQ@T&s=lQR~)v!aNF2ir22`6kmJInsW%n4WNiSm|?FEo=nWkGC%R8M54)W6z1N znNw1=DK>WBbn@}OJ6MU6%APK5Y4lE!^JCzb_RY2SY{B~eixVJ=IF90ev%Mf9>up^ zbYwlz!n-z+I0xF;b~kwkUA}}g`LPw;n`;7`D2H$uoF?`9 z=TgU;5|;tSio%!vq;y{V_!=g1O1w=!K|$hab+$9t^e&iK>vF9Irwh2xsOryFHC^E9 zk6n6U%O=qjaruXKuyE*#!L1V;iqi!~SfJsfhZQp(QCL$6zs+x@#Qr9LlD8Z&&+P=<%N)Fu{3;q#R7%&4O z?vcslXR&tMcrLqe3yZ0<4+1xP_e)M-yx_w{d-EE#G|iq8p{d|88hL*2PGghrCw?AFVoRibDJz1qvlw&W7^-h_$D-ODr7&q3AI(xQ!y$(J0MEe<#{uHQ2z;BuB zxy{Vj%>W1iB;mH*RM-nDBR$1KY^D-y(x{H(z0VjsOsKu_euKOpIU$xgKcIT{#w}TD z4!gW9awPu9%{={HlcyEYeuy_BC=x}>K-UCtk_UJ|NCyTArmZyMWtd_Z+UMuX zGu{ZxYW^mS-`4kW-XcB7MoA}@Zcme8uRb7kehQrH<+_boJA$tC4*kJPpBzNh@94Iq z504S5S+Vd?JSEn-F z+EeO{Z>KB2>u!hVT|Y%xyJk&*d3dR{rkDfUFhuS|dh1BgRygiZkQF}7eR%k3Xl@Z}LcB;4$zpZ$%X}I#h-{z>R}MCz>J)dkJRh&? zC6GMcuW1PzE%=1VaR~&@6R73W#VZq%vd72+>plLQv?6+=d>X9z#G+&o2Ze9^#6M`G z27;c%wcW!|iY(fl^_X(TE<)r~(RWeAWn_v3AJTIC9Mr`1HJnDZA?Ko9lCf=TP(9VI zL3bvIHpCL4zm-9}2slqa3tJG6hj0;Ve-J_8UIk+uvwS2YGGZ?f3`sVN`9^I+Y%v>0 z$v-UCwx8+vQ@#CbJeQVF@P_et0oXZle9ePp=JhtIB9z-Eh^ga|xppY9<$mLJ)M=oo zU|^XQGp~o_P=(+2%w_mrR6w%^fA_;OLU%#ymR%C!upZ#cK|oAtk_FZIJ`i)5wV^+$ zlAKEvwSp~vlh3BSyHbur;&jGn`ezCRU3y0MXyt8US2cpiS9LBF++-pPI=?V75F=X* z_t&OfeORi_^1WQjk6C3CXS?Xe7gfD3vqh2C;VrB@u%XFpO0=~1Q*~xCU*2IZ&_la> zJ5!H9k-EFYnVAu#RH_Ym!fgDgMhGC5iJ&P0W#=xJu4`jaqW}8Mp9c~JtQE>k5(TNO zEe62aQUOWPRE(!{9^2OV4)>YGA)FFqa&fBg{S(>V>A zNgQT-Qj|b7|DG!+bDt^@6B}JSSOR8Xuw4KsAOJ`;L1gfI`xR94J8g2vZ+<+Mdza78 zb$l7IsAscy$~sA7=><1d@GlnpHJbcDJiFN+AxXzazQX3PTjJ;huo|Jd`+JwG}X5QPE>t?ypGm>!)`8}HLcXA7&j=!H<5-#36ZhRc?VkeEOFmJ1FE z6jclExYkrK{KD^r%WAWt9oN1AUyCHYUQo?#BNfFw|6tLO>a7yO+J>4{b^V@FVpu!l zk@qwe$fo9#L79HdV(kGQ@5W9Iv?sYJV>5yJQW*l(iax~5&rbGks54Z)0}wAtb*YA3jV$RA;Q6u!Yo^G zxNMYyELEnBd8KtGAUC&Cr6RY)RbhHG*7?1!6w$cz{oumSZIm-tIkqd;sM#CG|3j_a z{IgrUk7oc-+0GqR=8B6+^S^C`0&WH|88l9QhuE%HQ;jb02yR&Iae9(HE8 z%6WXj)H?gH#ITyRHuHN@!}D$Gmf>5}0}QKzwY-&^Y*!G~J>ySj3|#oDz!owBBBl~n z&kND+RHn-0HPz6Dg8XZy)siGSc%_!fz>pGyY9jbtHGmc`G?hB(rl@*@y4Pba_yAxi zRz*cU(!e6o!qOt&lC}3L`9s5SxiJ9e2eZD(@Y!t-RP|3vZ22G@;=Y65Y2ISUX6JLC z!*7QXctsM3V!BMRT1#=4yN9SZ$bEb&++$${6>Vi8j%fj8BrpzdN030_LRDG=l=4XP ze~g?m(C-x>PdS*zz_3?70z&y)K_M4V8njz5YOZ;UXzm~tlNXd2ta2EyIN;wy)-L)B zR~mUw^Fk4B*W%{mH*f1U&eZ|17dar+m@S0dXusrvArbHLdr?_G*9W_DYIXOWDLGE% z08s#U@$n|A_-OR-YxQq{KL&pAdD!5Vl@dLwZK2smS^q@-m}~XSUquH=Lipa3Dz^%-Cy5+x~&R@*pe@*IEGXHt;?~(k+V*I&mf3L~kYw}P1_`fvi gZ#ns2<)rNf*DRBJ#BTJA9u4>?E2=A$LQG!%1H`8m9RL6T diff --git a/site/notebooks/images/my_tests_directory.png b/site/notebooks/images/my_tests_directory.png deleted file mode 100644 index 47baffe80e5e0f1f05562126fc46c7150cc8ca3d..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 44368 zcmeEtRa6{b_az|^T!IF7*Wga$?(V_e-7N$S?%F_b3+|8*2yVe$g1ZxZil65D&-XMB z^Dt{ztEuX$y0@;Lv(G-e!xiNvkl=CQAs`@-q$EX^At0crAs}9`z`g>ma1pE(LO>u2 zS&E1#N{NUND>~VmS=yLFKuCrsX~1Zz^kZe~#K(eQQ9&T-9XZrEG>|kbL%;>;EUZ+3 zqA(&9CT~sgsOz>(%k*l3YOF zyfrdQh5yylOkO|s#wi3Sk63OL5 zo(0(l&B9UH2O+8l`}7i#bX~h!RP`Y-eO1W!p6fP87f-ANPpum!GBCcHUrUoo;KJjE zakdWK)a_Cc%Y+u`q(6qe*F;!H5$;$dG)jyoK?>KH&HvEG$1#L67BU+?iUb=xi@Hv% z_In1fBkp8>!A7%;UoZ`C(6$%NTwWJ^UG)S-M&My_y zF-Mj6I!90)5^qzz48?UxyD`ET;NQc&MlBOLNKL`SE3Cs&N&OoArY=s>*yL0PginSF zeMfgu9mFQ381^^_eu@&YoQysW zj6Omju+fobBM5`2Z1RS9RMs127K7hSw!3aG1)aFQ^Pt6-B?666Ek zq9?cr@LzLK;Xun{T>hR_5E!7xebllho&Yva`#>L2EDv}`WZ0K;C;GHR6u4)}5ZNH{ zUns}Anx4$dHkzgnPuWB!ovieI@9A)M!4j`*$Ku~eA`*W<4GwE)dT|u&y|#9;2UEs6 z(#`*B`*bP$BymjML$pa~hj=lKo<9B_V^eeN#bt0}v*g2D>?X`fm!1bl`(orIN4-S| zu2se{6ol)LY(0`e_3sRW@IR>zJBL?8i*oD8c;l4Lm}`jUwzlK;-1aI(XlrzzQi;UM zOGIwWP%iHeavnskZayR3&7tXp5qudjEc%Qvt5p7i`7^pE*>nU!!dmXsrD#XXtM;D0 zpZ!5RUH-rDd}?ZBx45TxF1dV_E;bPg`BN2H1Dqg1DMY5=P=Yi( z20mje>Z(rAEBAB(=2u%^UXAqPG`(P2g@F{J4uf(ehLa5%H~-?w_(J8YEHg@C0Ly2@ za>)HJ;-8@pAX>lrjX^{QsjSjBK~)CetrF_;>#aiXygdG54kaWBiDHlpk(CByUpZg~S z9{kHe2H$v31@7?zrJ2cR(fQ!;ySRjT3~6)UD+!wLg*^pl98*GTUk*(=hIIrdMRdGW znuL}}Nq%PTVo{rtni{*xrQ%0*%+e7xz5Eu9%QmZ`JqdOOc3$>|c_{Xhd9?aNE2^oZ z;`UtS5_MHUD;D;gmiO8s{BpjM4`6u>ch&9U^QrA(jnbJi12umsvI%-i`gw9~l8PX= z0JpU1GNI|uKi*k8u$x({TH#MO6^vI&&EOZ{l%(Z1YdRNgK@Zz8#3v`IgRumM%c2Xvy6ENT2d>M z7$rD$+)Ix9mNHs$KHwl3BcUVeu$9}*P8l9we%Qa+g}acxyxLhByZbachB7(&!=Q2n z{PYEa^D}E0N2=B5ud7i)#}0>U`rkai$%Gyg`OToR*Re*u%`|N4`8@S`rEivcR??SY#R1AI7_OSlQW--X^8naoDmx z*qGU3*{hp0^d1Cq_hedTn|1tXn`;~G>NYCxk&V=jY)v+iDK!1CX*KAad|XjpfmVT~ z$*$?Dv#;e`j!tkCutGb}uNHNc1TWd43pkFO=sC&qAKzqozkbh-&C46^z zbw~zF#wQ*ot{T&{3EukRMD0-Qy5lnGis(-7WaHq&{lm!QtZ$QS^yB^3#$MuK-PGf6&l?bfo_^*8+@ONOJ27|pV;i9LRK51n%DosLfov#vOIZX4($>uc%rwzu?Z#=MT2 zrQcxPzwe2E zA%Q>Amjwa=7MU8!0$~z93Pl(D7jqvyHGeTjG1Dx&hxkc9s<5PH-&w2~@B4&%Dh z^b<>^PNvFA%S$dNFA54c+?Sr}q3cp5$~JM2d#%@nV^iYD4rII7u-LM=p9NwFOB)T0 zBr+{Z+OanZCM9Sm&G#CgtDoziJp7`&;fJ&rYF^b!ch~mRc5iXNjkw50T#1T^(kIbr zlK(}^O73jDbywb3eyH_wE_}}0N!3Z&2|2P(R4Xr8_vfNz^P!QqV|bHs#CD>Z$}PEJ zysf9?a(CNzAKPDKE41}-kEDn6Cp5+MznIw#_W~``CqyT-ix%Vz*dDYD71!)pPjK3B zhFRszM>6KPBiXIbZA{$u4z{?roFlefhLN&cjR{Tlr`uK{Gb08fmq!#bS5$3O`5J$n zGJWlwOfF85l(u6y(f;g6(xKS&*>-$K%o^L5ZZk!n-kG+H`mQFm@z;7)zMtOt{+IYV z3R{__WIIi9T}o{?=jP)xg3a{Jkj)s4C_Sal`05&$%>}gt4ZUKAlGXts-AW4T}H z-|MlgQz{;67>$=cN=Rp8wGlPtX?4uBDyZ!`Zi9&9w(> zJvoR$p^?&t{~GVSG2Qmn!a#%X665@GHJO=&%tvpzVfk-aGS=hTmbRmnaIKXG$g@kD z+lq~zF-O@o?eD5wj?aB*?(B|G7FK)$Ji$lc&99{*7a|3hk`EWR9Ur1TM1TME{pwq8 zXhud`Ix>MHU(G}D?)ZQ)Pj50|5k8ZEfJ5Gi=#$iG|FX(%5h167U5C5Z;_g1fID@Tu z*GQ%yp3Bt*_jEJ*cKaw%`>N;KiP%=p2N}DRJsmwAl!v3+yIhXOhTFlT6uc@AT}vyX z3BUW&<0+LH$jYqtk!HVL@$QdT8K<1~E9nkpKRKS`!+NEuqf#SY<7E|g=eU3EL&f02)?@Kw=&{qXub;uK z(9^f*n4y?$0SZ5_E6+XbA1Ukxww)Zl2zPG&#OCinLrD&sose(JEwphVsv2K{V_itocbbcQR*s%gxQr#LU9P!omPN!Qkv^=VIu=VCPKsXC;5v zBWmhw>}2WSVrg$j{9Lc0k-e)6KPl<+i~jlhvz?|MmjAxV&iSuy0Ucy|zQgpEk(ueA zx`Cp6&sTXAEj>(av_vg!fiVN#A;8Ma!S~1W|J$8^-}qZe&3{Y2W#(Y}z36YZ{%27& zXHzE;dt2b0E&~6~*I$Kyzxh`|KBnic|27nVj`JT^fq@o)=VSV3(gfh2K0w|B^N4RL zs-OxSfmQa$UmEZqh<_oTkH8018Kv~P7C1ski3+KDyg0~$$;1$w4@E##1wpf}!oj|R z!bk)gA%i^#-;6>~(Mq!+z!ty5dRgoN84-4d^Y-0$^lxya*>IR*veeXOs8HQ~Gf zyJ=ZzS%$6V=3Y6s4{oikEiEm3CCkf|{_U6QNjd)dE%T_v5HDas5Rj+=5K#Y|f}qig zgFs4w7tQSdUi{~NG6Wa~ppd#SyA>sc^8-VVN{Ik;VW1H1y_tRPbkJ^r1umookS}xs8*)VK2@s5x?j?`t~=XTYV>e9!-+-*iy8o`1LiSrQ|9RJNqU}ZY7fYo z^k0`V1_&LQm|CE4jG97X;QIJ$5I8KFfm`Ial?q;k>vLL9G4P}L<%A{tf^*S9WIrPh zbwC)07*zri7Q@j-Z4c_V#T`fw+qd04kVKa@9S2QEG42wR&ThS)Kq{k89t^4k+TuhZ z?)Cd#=zbsIBfpBu+W38PkFe&`b34|6V%zKDt4A-TQ}Wu>)3hNeY}-ZZ8geFB%TLc#ve#STV8i#3+z?jzrj z53nR`W|eX885Ml8QLwEW7vGIXH3$bvsOEhyEd71-9H5e!45FQ7VaR$bGEJu|^>};B zB(o!c*`%dS;zSP2J8R%)5B+yi`#|q5_olWo>^u6K210mB@$B2G=Zqp5l(*_^=2*^u z#t5bhKHal!dWuaKD}9?R(->8)(&x8Lr;3J1m1mE3 zy_Fj8q0t~iW7Ooz2XEx_9@Z+IZeAH%(wDg&A=_Fbag9|bZW@vk@T za&$)&h_L?|r9emuVnd^@_i$#X>pfaYpOx~(x9TS3Wi*%8)WDh9bh=zU<=u;O<&cGJV~O%nlwMF)@5 zF7pQH-Y{i`jnOM4TocT_CU&FG0>>S9l@H$%xW;QtWy*G7Mr%?2U2cjTFMLHxjZ(snx^ZDy5v)Rvvo_Hos1m7~@rHpbwssvU`t3^3}EcxOe zX^Yj=$+KOLmzS#iHF=M?qk@XM!`>QyPlZOndat(Vv^ml2Z0@`o2$M_~(`10J6DX_> zW!xLhu1u)+hWTBZ4|$@Xu3jYM%0n$>6H|( zm!L(&OsQJg0Z#qOm%x`3@^CUOZOVk-mBE@dCgExWL6n{h+I7tQUKe4$WK6nEV?MXL zMSlDg%Ed>w#^l{OUuLU+4k0kLSUZ0UpkqXt!Zd0c^NKPmK$^iav!(F zH@T$kEL;xIRn-;UT^(9CEZDWWNQr!OC-|nws+W)km40{Bx{^Be-Dbf-$N97en?jRi z@Q3H^1rCKD=H&L(51QhaMKptMo#b7D`py=kyEEw~);x!055`$#2A$cKGnH9xWC%)b zE>#`3Rtwc?Yu=Z8Nsd+lJYQ=tG-BUO)|xA{<{YKWq7VrzX4_4bChY=a*Yb(uvE%ND z)xP7VaA}4uMg*Cg-1t^kq2+m$0mB+xU#OAB7DBscPO7$K+T(b=<7Px5!qi< z0uHMX6NPeHF{Qd;ZKvNew&ty?&Y3J18b0DMgEwA|#F)VR2}w|2LN>!<$VY*|nhPhl zlu8m!qMpS@x7Ww!7fY*MzfRpXm5Mf+BB)rnz8Zd1HdNr+B>k+`(y4xz_r7&m^|Vsw z@JxwNZ77xmTii+sS1h#UlfZX)WHdsyJ}#FXMeHo6j51KQQ4c(Cl4P1GnWDvs$^<#a zhQwN`adh>a;X{Nn*dbZbb#L-5ww{ZUrq~E~4)M3E!-Wjv7=DgzKH5R(YBcJOhbx_Z zXW`{|z1Vf%`_tYk&+z8iU-qWT*ZovEzc7;|NzN;VIad~A=&rIn&$m3!&(v?HET+rT z%mtbQ{qcL`#THo?yd=RpD^K)ftb*8ph1vOdd$5H(*e(X9<`ic61Dgesk_3R1IzIde zd*od4+4l|BxO}DXc-9^C70&1@V5to^L{)`@*nMM#*M_LuxCkEW#lyyMphv~?OnRU6 zoQAb!;E%g;3q&4gO{7{br*-a1zAc9X-5Xt5dA#Cu2DXtoz1JjUkNb-6Hj<*A_25{c z9Nt3XfF>AUL7YZ$qw8(2(RQM)9gKL?8^bEgrIvaEzEPTB=|nCL$euj0$RXRtr3Ud5 z21}t573E*b!&9Y*w)ERzBHxQR!8}Ify>9DDE&Q(L?dpx^dLt5xDAjKFfgtY%~$<$_$mBtCf7%24A(v5W;LcP&A z=E;mYdvb7GT&6|pE63VAwIfLQ+afZu&cebOByFU9LwWg6WY#U0rB*-O6ltQ~8ard_Bj@r-cZ4sV%ug>6u!dsgx^yWaI}Vymax8xP^L`6xe~n(N^1|Pm7ES6m&&^Pp@e1 zZMYKHuj=eDkaznFfQEJ^_qU8lK%7H8V%{nbV62@A)a+5w85k9KQ*< zeXp{%L=TcHvY7*2-%IxFSmuQoLcU+Z6D_WE3H$sHh;3N_*6G(V+V6lOq43a0&P3+f z0Zp`fbLKy6oK*Q|WfF5m5&35{AJhyE2SFn|eEQ0h z!@>4Dm#)XRz4(Wo*o=pr6nKHcg_$aad|MUU0^CBV2+Qvx85)?kGo97HQ`83S&i#e% zHZr~d&hPN4no+MM)%R+S>gOQ<+x_x|g1|U82?_$1`d+&ftr6+!XnhzPk%k+6aUA_= z=Gm4Z_p-a+c(!>y7;b9Joe6a8zEqu7TA%HOWc zuXt_xa2~q?Cu{~Olfrl@zPxfdM^i68Y-+2BNTN{<%KwE)p}6Qev$R!Glsy7(4Osny z+f_WfCPU#{lR%@Y_A544^N#dbO6))Y^6?H&mW)#B<{W`VLfa4G-vNQic|szEVjqu5 zcbIjBYpB0cI>E5S;X-4~ocqcUk*_5q74Bc2f+i5af+Kw?DUT@9nlsmCrQlgPtSu+R#90+m9;oi?qqdgg9Whla^;u{R>%1GNbXW$?_}_TPL+GdCTei$JdbnIhAR zOO+O_y=%*eFJnNUj|eGtMIR{-PUZJ*Xue+deu&d*@2gi1RX-p$kpMkKEqiVg@IHLi zBRPMhluK{=7PK^w^1VCUeeSG3g1G5Zp*mHGR_&+Y&KZ60IZG5nSd8Xnm5_v_uD&?V z~f=~A@`d0$BZPXNxPF1LGAh*BY7z){WMo6srlknNY(j!2z|<<%4~`9EB8o}cy- zQJtt>rRZ@i9sr>D(RL7xx==oAPp&n6r)WkVfD{LNv`))6BfQ666Bd3aqx-?f$vSrO zcjMbse>>G!bIA1U68n+vDUu)9!YbHzm;2jyziyvnJ75RyL}Kq&>3WN=Fi8-&PR@sg zUFTs@Z;j^ICo}48F}SV_O%^MWt)Xk2y+U0&H1Y_B2QC9-z07y~05w(KyTW(JFu`?5(meY!Aj!nFu_kq>CuQ~zJ^8YzKlwrvq#Ss|*-%RzU(IAU zHt(chMqDkoeoMa?x5?X4Y>nUgQT;urONp+1IK+6j>|5m5AX4vSF<@hC$Prj$x40kM zJ>fJNH#EN<6V&;L3?&Ayl}E0?+Q%?+Uoaw0Ukn?#-hRE6I>^-71(WMvjBUU-S!Xfc z6*Cbn0RIZ{O)9a8%i%4G^LjYd?QT3YJi}{VlfHLVAC)u0%>aD!YaL-!YTHQgILm!` zj+Il+MU^H8$M}2HO5)kMq)EHP}VZm z08qH#-lWOdhU72)B`g}02r1U-_b}trm=cC)o;^3}th)TY>KR1X>3eQsdh3!(bCC!u z3B52nbd+jAi|aZ}3B%>UyBpjr2u5JS-v}45W)PYwgf~YCB*l+@j*zGT2WXo(sv98< ze#35+7#DvS#3Rs4;EunwhAbJa!)*fiwV3-dwqB3eJGhSPClIXh$LJxlGBy3UQG;j!p=uAAvlf$>M^kbHHoeJ-obF6R;MB9|b( zo<4=)`Wb!egty^n?A9fCX>?~fp za0>c#0@nYnD=C5+_znLp7OU?qV(2u!mj#vcUP;lf-mh63$o^S88L10)c-vG`Sk&L6 z+u4or0QlTzv^Ak1$WG+LJfmaek~=<$tTnN-J;*czA3!Tw2`T2}dmrJ!t7BVVFlQJ_ z(6)=;n(B5qkM*%Y{xkByAKVUTlLTz}@;PpOH1K;zpxI&x(+|aFRE1f~<}h7jx}GS8 zRQ_UYrnZ5+>k+0$K;Qg$7Nf?}lfMyF z8zH&evX5cU4r^<|V<>`Y|G8$PX@f_s@+#S7h2F_L9~OgxL2-HIc^Pew<%%4#7Pf^V z;i)g@Ri0mJ2cBIm*f-8tgER#&;7Z^yl0dn%79~SKIIhX+s6pXc&b~y}G#|j*+rSnG z#OJA_0(V1#Bz8vBC=)#x@M_?u^$0q2^=hueIu+R>`7Gr;y}BK1g<(DGYOsGf5K6^3 z2I@l6HivD6+PicUxN#u3Jt=arN-q(Y-K1w-7;ADg-nU>~xY(2f zJp1NwQH?&4Sm=DU3)#~geN6ZttHygmj&tq>ch>G~b$%Rd>ajq+y3!8ug4@+pGX?O&3GwPmT^GgxYj}Hh9Rn4e(cX z8g)*=F#RDa+b4iouTjSu&fw%M%5;j5BOMH<%J%4z7n?rIXSbe~?U)oyHZ8;B!{zh% zOz?26Al?)VQW9u_#n}A0UFkMzh{0zH&2Bn?;ZDa3n>u*2ICO%GTG8%7a)Zp{7?59X|BHX@=<07f;HbUGd zRqUhc0+KC1kJlFN7F`(B0Hd(Y-G4tmszYaP!Od!G%1>6k_smKpN$cjhVp1<&rpxD{ zFftgHQi!kkUbhEGnMu)VDY;z83c~$K5e5pAGQax*aVIGAI5J;vol>z#_e1P zH{zVaMG|KOP{15^YEC^Z*(n5yfz})&p|VM*uuXc*yhUW6kWSRXAg~G@Yt9GdwO3+ySauQiePfn#>TpT?XB9UKsna>;9bSCP&hoEjc9zf#bGw^h_SB`Y_{Y?v}|408!azsWhYWduLm1gPVmp~6FUbA&Y*)c?eyH{FyZ#|s8Sf{14-vR3x?DrVmfGq*y1Fer_?mOnU1d? zjJ8M>cS49R_;WX{yf3(#6U9_m_JMe9IAe=ttdolO?^Ollc9dW%6lYDtYD5mR1?j2b z?{L5_Yv|Nofx?m-5AHyGD}H{XSFBN$y}Q)vKJMP7=R0jmmr&U3yp79X{WB<8H=Zlb zlJJv^&dvf|Roj3(p2KFjPGdCp!RK#$bP;B5M}=*ibh_?Kms^)JD-QiOp*J~Cd?U_A zQ;YDrwPx*Rh4Q}Bg2!&ka2RvT&c9A%4ry^1nM#WETJFE4yX@x7E!cP3npjqw^eZri zn@?2-Hg6`6Q~nxdqz@Dxa2d@uCzq|-`LPM>3=o+*>Sr>fWieo`+@MWTMGNH0X+DEC zk8pz`WP+Oof6k(xsVO}FdxY{FW zoeXFd7h>$=vDuOQ8pd*4T09R0z2?w5qjQ0*Mx}<*B8&&7PM&vWNob~}x83WKR+|W+ zl5oE!iR<+7?wCpwxvx!fu*M26l-cqJ`RLhNNunC}Mr?394Icktb5XX}K!$h2^r^`J zLR(*aLCIpg+IQ!7D)TU64cCd7JzoNY`A+G*N0)r>uHN1qcNz$vcA-%xmQ0g9T};%M zH8$b(@224Y`UDTqYb}qtp@bYL0$k0i>N@s%HLq?OSNu|l<1I?Z0p2Q=)BIfly!gMK zW=2R9EcvoHYKKOJuyzFQQjS&-KSiI1Zw*gX{y4|~5!ydZ$)3LVPW+^!x6MbxyR)rF zq<;HBwR#2lidU{TLb=>M@N8QBt9Yxiy-t-Me=R_^&p9hS2!mlmZ6;Oto?-Uq^yu=w zOQgN78q>jhdqaMOv(fErI;1ZQ1|5q}en~{(?^?@HO7HbzQK{&Rum@TiWqREh0%u_{ zO5x7h#f5tK=?{Q_DJ!_pIJxV#*sxw`LDvn-iQ%KMH-vDh5~C`RL_6%3(X73J`??Ymb}!VaniG6an9u{WK?%ohCQHyJV}K8y_Vu z9RZZrvb(27YgLD7P4Mv1Qmam+3&k4_8_h0S0yO3ih&Vr9s7impii{p*(IahoeppY#y68=x%P9)|@0fEPzPt+LJfnb#F&j=dCG>TP$Qp=idu%PI(XO*F1cI+*FmtQ* z++RsS_`raDVk4nyyNcYY@j9PU1b_&N8;|g2^wxfvZ@-FJllg7?#SaN{!719YDdWsbN6@ zWJlFuR7<5vW>%6*=Ul$@O!ro6}|rePz!yJ-^!^Ch-n)!OtlE4nj)=Aw3Z|U6g!?C zVXEY;Qf(V;chcOy+ap`GYzKS%$@}U<%7e!&s>MIDhR~=KrsA-gk<+v8sh=uJh!G^q zk~Bp~08ykY0LEA;P!2Nz8YkJ1nC#|5f?C1ebhV1aM7>PPm$`VotOTa!4Mih{H8`|o zaAOojd~i3sxl-shAr0Ke!TAK92W&q8bq*yY!Nx*@@9z?*`yA_BO2w=btzO$it!XyfANB`XrSQ z1sT%)3tQjgE3z=)1d-h+xG~M<>nyX;k4mn1%bmAJa4KE_nOWy^fTEn^L43pfr7uQX zAZY*1)=z!MHVFWu_3qwm=Wvelok#K3As?}?(i&277q8?8QbC8aq_1bw6jyDc z70gX5UvMBDGo_h#*FPY6 z|8sakp}8mNaxhU8vw8C!&wf_O7VaZFT7XFfX&Gine*)FrdT#Wl@%p5S0z2Q{40QjKdj`^a#H@ip zL_DWP&Gsa)Qy_Rl>uNDsZ3X1jb|*_zf||glJ*HB*$Wi+c6b4nsH*nUopL65BkWhKu zz*iclR^9a+3Mjc02S>4CUT>xtZ~!ho+V?~JgBSFKcy*Pz&O}`OWQh-Blf38bwbQ8K zB;Pki&KofiKste2xdH!g{um-Kh!UU*p!ND78Sx4B9B_O2Lr)Mx&8xQoWzGSESGa!?}CU(@(UQ}#h18OcKJJZFNMQkRrIL6O7xdOE%K{9teBW%nCS$lI8>~avHw}{ z03Bo^1u7GUN&yX>vBY~qO&j?8w%)=s4=8-8h|k9q0e`?T_DuC&c1V8`LVv81HM~`ra1O zQ$K52*q*g4Ytog!S4xcw$u&gZTH+9;#X+VF0=EIr49LXp|2}g-ezWG2X1x{W1b_Uz z@&pgL{IkMh7cms5a~62Mi&EO}Kk8J5Jd?^Tkf2aUjmjD!afHl4=R-p6LjmMX_YPZ( z|M4g;BpNNb@;zV&u8L~7#uMvB>#e2^VtnL?Wif`yWQ2@`fx)6gl4Ac|u=6KD*CPw0 zPEvJZYJTtcpM0p%|1b61c>Vv;_+1G2e>xiAfi}-`_1;g6sxN=HK7X#}KvD`t)43W` zSU?Zyfr0nGJ%EZ0siy8t!kCmx`1|tz%VuCu0b66;hEp2pHyicmX&5GOOOu@{DezxK z)c@5|fCBEBMD`B$y4(}30~| zkYb|1#t*}d%=-_8TOLqe)dj)9AGSLMDDSh7oc>3S`>%Hpy8z{L$tXl%LGO@WK*1A2 z|8C$x#MZ!Vt$b1vp!H-x_wWPb@PF@J(sP$g?R@u;11*3Ti4LYL{)dUA1B&;oWCC$S zwpO!K!JK7L*3h#Oi>}UPSDDp$OZL36^I;gsXXL3;JpKBK!s`F%lYceim%(8>1ONq_ zftE^D@%=YI{Ev1H$imV9@+)I&IHSs#QCEEj7`U@p;T#}3Yu<>LIUsOfa@Amc_az|F z{Rz-Y%Eb`!4*_z?;ld0%D?5P9Oxz9oTQK!UlD`A)J&7W{PQwu2{mHk22z@~Q22f0R zX)Uh%^YUj{IyOqQa=8-4c7Up)u|%yb*%DCP0NM%HG_gm3y&Lj>JSVCJr%VuQ7Ej(#3izY5)X%at?xUuk z_q@^g+&^577Uu-rc)eZ#lD{8>dj)*Hw=V7x48Lz{+C_NX3wdWgwp%*uPss*+~-5Awnc&}BhNfzoLX>~;Nxjb zx#&mPn)Z zqVHINq=*qAIWUl7gbUPK;8>pfS>u?r+sl2bL_kvUpH3fIl-MRPPp8q&to?GjmOb(E zD-eVk`UB({44y`jJi$bn2Gar#BgzLLL)Z5#i6{j0fm^_sF)FCzR3X7*NEhF-1FXv- zd%9I_pp2FhooXT|*P97oA0mW=9EDHO+}R#PE){*_MOyabCz7_`1m^KwL~LtAMesGK z)qrw!WWHvxStPCJ`t!1I(?nAF!T z%OstaOwpNn|2{BFED~$Z(*lr&*%Bw~{kxe0xzJ#Bz;9Sg!b06CDGN;iT^eA04Y|? zIUqi<=~aacvGWHcLm2|Tx4$T8WWXMP*6)u@)5e6b{|ghz%e^`uHDz{oDdF*bS(cNOVO}j5~Mg zS$BZiDh*hcBjWbJx^;s%3V4ak3`h!U9iN46X#h{)x>bB^BPSD^FBSqMu9kqz_P3+` zlvjIfQ|r%(EFgc$frQKUS^Z_}VME5{$2F><&k6L3l5Bw5(HU*vh*5*e=VE7EJUEo` z6B@Ou-IRztt*K!im<9{iMwN&-=?hueE4}BG{eoA?$Z$G4E3m6O)3_XsXoxU?Txs~T zc(EcTG(VbWU|D1_F_d11jSJUgrrq0Jr$}{XF0$Q#wsBs;QUHkbc$r(4+!nO9GihZ? z!B>d|k(J~8jq_G?wLt&o9}xwpm0w3{;6LvT)+|tNN>Szrpe$i*U`vCFcxLDUk;;^% z3`5adz+`kjh`5DsWN8Q`qt!T#$JH>jCzV3l;Ke!N12gIt8(zAP} zCAWQFss4`b0LMuz2z7yU0|;8@M+{1ltu+h!ia@7;@Fv~5viUn!JD{=cIEF>~vEocmQ~WV1+(M+OlHB+eK)ZP4>8ez z%a;j;A?8qw#*aHtIN100YMQf_XBJH^KA30iSrmjRctKF5MhoPqlZ1k!c#2lbMRoNy zH_mdN?#+YhLpye#Db<{qO2=J_^{_Ja6+e}x%j3-*e5ARI;h>uo!hMegb(zGXsvRFL#?z@C4R8VftKN zX??AqD_w<{%(D_G-s@97hlKnE*kx_-tzfp1MS1;j1^!FbJEIip(=G(2SrhF7Q6}Zz zeqaH*fVz0_0Nmfal9lTS;0n4E^Y;uB$#Wn>2I$K4e4hrCSC9RlS}TG&R4xz8N6xBy z39T#eGeKl!2+&BymUYV^vMPL&S~Kp=h1}LAPvBtWnw7zt;xX46;;{?!a6;1%Fo?w3 z#MgPyO9+d&f1-%}Gj-$(Eq0&tVNUgT(nn3tNgIHBH)wP{{811Y_fruXiN|U$q&v95 zc7c*>#a)G^TjIttJ^Y81tjAWG`8_Ql*=c&u*r_Tk!xJbxp-yfmNzV%?e;EYf$Z7@@ z+cYgGP&wsvvE~6Q*jnGBNHFyZO@&-oE{kUhUXm6CGQ>K`!|pZUm@E4sdpe&U?CHpN zVzh<)n`)vZsT6m5s7G7iSzO^qp1dx`B@R^)vjC;sw;>>~ud^0W1UCVmNINVj#u zCF66v?)(ko3sb!73BNmHKNrlfhsmGpSMSx$8l|SEIb%??;8q3}tS@(8N_ycto0@*g z0{oFdC(ufBf7Y%nWC+jDo?_W~*;mf(Iu)b>?~eq$(+z3#j3zj7G*Hm+1@4fM?-QRj z6`iUp3%4BZDj;&vK1fPx#Xc|LhnHHS_y7eWt|AvCG#F%+rjVpDQ8Trhf|_f&(sFh3 z4TaNoz79^`s6vX(RF{M{P(jZ?AF`>xWf?0JMjW%PT_~eHS6V4@(Nl1mMXz(;G zz!=ry09tmv514^n>k)@cfVbkXtpSlFsXJGb%L3U{rUr5}tPM4;3{&yTP}+J8CeHTE z%BFSrj9z{3LsJG?-Kk7aE_TXREcDZCy=-iyh6CyQR(WgYG9P#^!@e3}Cf3HI&u2+0 z`R~ReMVLIG+oSR?V)5IU?*4(BU>3AkE)qI*tsbjMEwZb*XD{`r*ub$79f?ESzmX_y`tGejQvN`O*$X0 z5GWh(aEqtn3t=$|#=2<%;1M&NCJi1tysT;4$&0j+GWkfHG^@p7(Xnu`JMo|_o!GQn zm|>IY9Q!jzw>Ci@9IHYSj`|f`=QT&$hx0K*rO>#f(XKKBotm6yhbklXpp`VkTJq(; zu}Un#i;tz&+M)PGyn;}-$rG?Gk?zSDLXiOnP6|1+E07^WkEUBIrD}BfG{}q=((if7 zZz-n(GRi%JNdS@L$_%h|RZh+l?`qynOJk)8*MO?6rf6HEin~8OKHOytb1V$O!{eqV z+H=Y|kJV)YeqDWJ>;fJR<+ec{ll+}zA1dD4cok_m=YE%A*BL$YqFPa4YtwL*4B$l} zL{P8iy2+Y4=3282i>Xt>Vh9DDE=Yj;j57++sa4BjC5vr@N%FV9qf97NO2Rz?Nx-VqU-ic02ISHG;>K#hL9*BaVy+pmj)*io2lMB7$4 z8NDVE7MZR29DwU3=2GL`>1Ux7${Q-+*x9@i$oW?i)9m~pvXcmnF;x^_?7^aRXV{!b z&>GLdZM3Vv>mY_gduQML!4lsyw#auphk(j1!dacz`=B~Bu2x|iQMv56f9{hou#+I8 z=wVn~2TU1qIUFarYns1@zfgk1ke5|#?`GhZl#MUWqxlJ7_}bCRavqbV*YqkUJXCEa zMz3W@{T$IiK>u4!(WpPomWoty+kHjFpbkpLa+0SpPqn-KFsatq;gkp&B0HYxwU94? z;=C4u{VH-$;9s~6>V_g0=xbi5%8|^pu+nglq!1{_q!Fl#nrPZtK87=~?`Op_f z0m!~aCA$8YXA>7A;i-Og&vB&0gb}jm>0#wbt6IX-65hr0xi>?qW^mNNPbspG$JjhN zff;^9*&9P-zaXblMgfKj_Tleac|Wf$Q#f5^u);Hv^3u27wbYuenA@luHbYJkQF8l0 z8P3@?z64XLZeZX8o}I-jWc)YuPV)~^36xsaV3!FSfDM~4MWy~h8r1{$Y*LIV1BH8H zB^FfVxqA#CB^`wVg`N8Vye!0&v5MaiMl~I_peOXV2!&F$|Wg2fz;V(iz=01;ZZB` zZ}n4hqBqVB;W55CS$j^~Hq3orTDIHlQ^UeIi^*$Hb13uM(TtW>qrz6Dx@no1bb*4G zpvcyFjt;D~AVVTZCn!rkF#r)m|he3m^Gs=23Kk1pwmG_l%CKpB9v8jgKUPVihTCT?6!^T+w8lcdLOkod7mREz( z6FMX=d&#|^fR&Re7Dmr;aad`sip~9{6M*{nC2xgZkjVxLZ#v5qOOGiv*Td%qW!?|c zl%&E8R=DMGIp#4GCCeq%J^-=ntmQ#EX};xKcvuW95gUIJ#f@sq+Sk;dUyNKomi2OYqh?1_fE^G&Wb9!${x(QrvOpwAiVm-} zHvpq^2HRHaF;>6#UZ!oV@pDH3pun+Crq@yi40z)&Ab_&6r$T^}Lm01vZS2xh1jSMW zoh2$jao4C?taP-G)4H36xB!V6EG_=lsXHKYIdq>-Dncg>8jPaVpMJU5#CtU2r8RPn zIcYdEY3;hDHx`gvm?(*;IL3}!uI^5CVo}CVumFB^@zf%zIf=suJMR%Ao`6tEZubZl;PzDRlF3YGOgVR3y@(xT9b`e?%#x9T@|0GLmVfhIx7J{#}M?qY#IUyM& zWXWzTLgI;u5>C<*6EsckR$3Ho8c(sQjqLErt%W!e&XekBvAH(V=&xTmJriQ6WRR8l z<@bNUa?3{zFzWl!Db-rFUrJ6o|Ge4&W} zV5wzld0^A+t}&nPsfr3q)E+w;i4`QmX~&W;SsNlZ8g4`0Ts>9S#bsDbnGh0683U7ou!R=ClXfR5r##=w2N%{%2U*v~j1RlIV6qKo}IMhIAbE z%&L+Az2^j}q{f_GTLS=yVL3l4{g*Ix3rM!ZW4V@scmR}-;uY|t*Jb}_jz^>gP=)jq z3lvc*XvfT(F4$3CZ2l0BH}KjU`lw_#oi- z+vb)Q|CO7vfBpTxl}S@ThDHW5(Cq)A6hQEj$qC?mQ*yyTrCoq*2^~e+{=e$adnFCg zrzF1@%M^PRu)sN4|5y1NuMhsmu~ZScA`$>vg=WRmQU1Rz5+G&!1FZscLiPHHAiZLn z6-q@I|NK6i0;m*Q;{pIg0ZB2;KcWe;H{cTB7e*1G{y>T)68PF4O3MDf8vpM+M;E}( zKo?TB|0mq{e|YDsyqFJylJB2Y{-=(=DTM!B;NQW|e=k+wGoVfW<8eg)20o{mW{&`0 z0P9yTfPZ!i+CU6&-;o_M|4g|3Csq`>1Ma2E^nK(%7Qg@db4w<32;j-r@b_E(?WeD( z@!1Pd;mgbJk^tA6|F{qwr@>}YN!A7)1|n0zW2kk#Q>w~_!lV$bN z0sdu}NB(#>{7`*G}RzpJJUf5Ng47 z82aDGOTlUZwId5Q*crj6;U<&ypD=%cIU*Ip%Kv!AgY-N6?IS8UyS$xeBxM1!l}Y!u zX-+oxO^n=M#*|c22Z&zr>gpzW4eb})6&%MOqyX2X_iSbe&<2v;-rjr1hp2+pCV`1x z-1wNK>M}OecD5Mk7D5MZQJh+KRJ!_eD25dYCA;=h;mcDGO0|In6m#rA- zE@`2Qmb)yGJxH{|U?kDg7Js-}-`*E=w$ilFrrti=5=IIW1+K=(= z_Ovj*O1snnv~lrM1%M(if@0me`43Al0i64}?x=b|{-ot{%ob4SODhzjdY$=6a^rtH z4996^gUvPq@8p>tuL2m1Eg;+I0m(KWRejhyIAD;KP7V9lY?z@Gf#Y>V_>BmljL08Y zS{+({)iRM!%+298H8;odvKN6vdx2{yq`8JT{z*;8I7{>+7HE za&Og`BS5;k9!wco=rt$d4&QRtg{b;=@57gePj`FL!oL$BOX^qLviiL~yh0Sb037wZ zYMesgeut+W-|PACOHcr=F7J}pSs)LRz=}USW)}iu3EF2(^SrZ^xfYXo_Lr`k7ogaU z^VNsh`oO#d%^_T_rxt)P)(zkuhO`fdUtS(6c2X?4mrQiE^o7vq8jPwX5!!vk{UZBV z(Xc+bc>o(jvW$L)T?)F;gIq-t`mC;dTAjbU_P0^4bpWCam7F)riRdc#x@KgNFB%85 zqpRyY?e`y-`Tyh{HgDa0ES1B1*~><) z7~VR}1kM(mGqM)6TvTq_xuF1O^XY18K}qO6q45HWAT12KojNmCVK^)oCSd?!*3SyS zL9lTh*3xV;Hxt&e?Sx)0_kOk9dwG$MPX8*3+iOjI_cEy9g>lbr z12Dhg0J4%{Q^O9YocC1@DfGb}-bUYNt*@%zo?RU~Jt!0#B-peHgOw96YyE>9xo7I^W!5L?($?N|_-j zSiRt$ZtB}3(tOQBio%3}1dH7)MHmxCg0}sjgWZq(*r@dFM&nCA5m8!t@?8f668iyi z=BySV@6ZK^FbL|Y<E+=DN$5j|J*x*PfcJf^eROn_)@{M4#*syK%0;bALm|xP)}^?xdurZ+&JjL zw{ifd;v#4TSR!@z^m~>|IfJdjWCu-3RFaN)2Ru})fJv4U;sjPX;S*Z zuhUy&(EaG){quuaabC6oML+iT`B`Y1$OF}^)Fa_R)@MOUj0C52-wLB@OH{$WL6E`_SnVprSDUsTBrsdM zBIX{)mR|(a#$Et)BKVye(JhdKL|wG1j=&|qjkOdD$48YW2?gw8wB>A5n(ZT!dj`+f zD3$0DJewmf1Ye#m?Ot57ar#a8tW8{pFNc(O+&V(P(V*BQ8z+P`+el@SMV z2GT>e)Wry^6KVT|vkP%@=iVkWlu(l~>QjDBf%4={Y-6<@-LnhjcP))~$aWu3vYFRT(HaQcwpIn*UMrc;HwCSA=sSf`}XpT)z)h(l}rVba-Dzu(8aRiV>&D} zVE9&O80UAfwLtoE*Kga6%+n+3l0JucZ41rA7Ic zWXQ4#I~wb&dW6~?oxtW*XyZ1WYCw5cut zl_0?dx1r$eB>xXJjuxB1`{d^geykAh7yy?)XN!(F90AImBrc?u821C8-S0C3Fdm) zNj}pZW&F7+F!cL1x63h{fIAEHI1!-<6YU>#)q2(8lZ-?{;~q&wO16oz<`aU3m4^0n zksxLWn7-D!Sxw6T8_z>6E^)6R?Woti;#@z7FoW59J{`@w=1`8 zJUGZ|9SWWY%8k7}Y0>U*kJ~KX8ry@nCXxLQ>Xc5UONa~ZaWSfqd^<@$wP^bD*!w+) zYzh+;Cx`AiFP9nt0NjvWMygaKw*!rS<*tvXZoBN>4}IT`%2eFCKiN$|z@qY|E?Rm;d{r~ZOX`OhihEWm`x8QdXsN@6W+Gff zv>*Ojji9h-a#P-3>VTp42q0-7QmII3+vcE1|KYH2R9fx0xX@AL*%drXY=6r8Z8_f5 z@<9!#cRo`dlVB6MgU)aU-aJ^DbcEv|i1U*yYa@_w-YI48Z!sC(=r62Gwu*YUGa+R* z6P*9X>Vpj5spiwW6YTW?L9r z=pqnl5EoQ?pUq|>`zWottv;?K>f}8vLkGG+KrXw_M@gK{}dE z(t3Yk^PxT>K7Xt>A|B5<5QpIwn3ps$V%$?UEZk4-;$wS#AJqf!N^4#qO?mk_?9!#d z?GuhaD!~n13uo{>TXP-+^a|3e!w(mJM0u>&1zKl(amjh~8PlNa;}Ca>Y^XS7d&0R$ zKi7~Tlsb!XSa4^QFt)=G0@nT?eNTWKp+=}?&ZiNWIP@CyIygBJO!_2okoXtHH;2xRheV@oJg-isK)9t8Q z6@pw*V}v)A|BQlGH^E21_M6+T@#Q2#<;W#}wmAB+=>ZEk;({yRXBLWpaTn+deqgv) zTBdzSFBU$eFq`OHi|9DI0ZF&E#bDE+O}Ngw-HAGBwQXL@J^M7h4eWSnUVVNiADcT# zxNO%qK?)Mz-|Ik2ZQZmtvpbPyG=3#2Su*4=x$zap2oUIb%xE#?UTM-9n{% zy3}nX?-NJv*X?u8fiI;LWxhoy-e$M@s`+XOf_*2s@wgy!JipgLdbkTsyB8~+Amic@ zz+)Bhen&uS6C@%2ftsG;pgnVdC25u&;p0pV0^GkU8zP`WuMi@q&2j1EYzoaVzR4p= zI;h}orJG|h9gXX$R4fVTPk5Q>eACIy6ah^LT_%r11yb~yDV~vKUW}oJN^uK@sdN=R z{W-SrMg$l}X4;7eHKy))lYMSUJXjmJkAAeU@08Pxf_2}M@LpwK*sf>!QiR%eL57i% zohYX_1e45bzEklEm&$ljS1QAQzY7qPB8#ULP4LLVVPl)!v+mqV2!(A4-u{5x?leZM zgb}~q$=h7ai-1hHoXGvIV3-iz)exO%7d3X00gImw4T|1_|MPpUjeBI#?=XlKsegAn z(7B7)c}aBAfTcC+qU)A@&JUJQ42K1=j2d;+x$QD-(laKIbA|EhsL2=_Az@N|yBGWr zhWqvw0$I9lSM!W@8(Lyqguqv&Jm$FH@@Bf5XFd{^=aznsR3Qk6km+bXU3i*6hkhOp zNDOQKeNH$M8QERC>M;FrlLMSv{|=|l4K(y6Oefi$+&#*ksEm0WUU!4;1<7@!w$J7B zgo`udHzCP%f#6CNYulo2FEP%o-=D1A`WkLPi^=^q&B55{bRNQz8jA88Sw)~Ttp_0b zK4~uW+CY~INM40E?$@FiZr{6Hb_N#6kxkcLj_@Hy!LLNHU|6XmLUi0n_J8f=QBYG- zMx{hpphvTjX>iwNf-_v2hzsiko`iPCA;vHQM9Z{G*@c{`mj5YWKPEVfS`_tZ^`qod ziAW5$%(*UQ`I~X$m1(NL+>;=3dzS%rv98;;+t#Gv$O;GrD^fM0huE8*X_!p#%77+4 z;!$BcdD0qK%oNj6BK$!&W^?MKBaARd88g!5mhaonw7y^c8%OCDrXfvW)Hab3$Kesb ztuem{AHz8iz;oPW@xC8+E$(%F$I6L8_YRQt%qu}dwoD5i&DTqZ`_k)@6UA~v+?vng zY0fONYoq3y)5>NJd?yH6&B3tX8L$cp2}L`f=$;OO5SXB4&?oWu9%Rr<&zMwPo{J5VWEfqVe{wM^%|N09 z&4X9lMSQtLqEQ>E)4cvKG^{Uw3VVw;`Fl|VNj|z2-B-rfO7YR5$(g|!sI>5T6Fd|F zULs1BJQ=RRoF@c{+$gq1n*mbd1v&{Q89|^qGeB1PYkBj}ZiJxvxtOg|aJ-(ZVUQH+ zd*Rj|YvVFlUuRm&<=tHX2wM$$pPBZ)K>S&Kd zQ17phC&dEei!)A_E*fm2qEf-WtGv}Ho?wkjw=Z~B!O228E-oR1q|fA-dhJ-1A(>ul zHO+)5L085#F5i`2h0S+@%TT0}fX+}6=gl~NtCG3a0N3lEpWFA%Pk)fr%@lM*rz1W6 zig2-#z+s`&6|O4Y_M$DC_vA_K;*QJT^nQ!wM?sbe=Z7Le#!IgaCM%lk6sMK9Ql9rM z5fSx^+vPu{5EGql;&fyV37+681Uo;&g>G z#tRUUuAvb4-@3loXDA8xS}u&%OJ|Z<1s{IEw{Y$MVWB^?N+ZXQrqNnOl1`D zCUH_EKQ+Xu9Ro0LOE_^Qt0AC;FKS|r`nxE7lRymZ{(f8df=T|#dJUjm>ba4{vXvGo z2xt;Y5=>Y^JmUx`*%mW24#kqzU4gGeXNr2-8%~-{-evG}G+YGZE?s0IWDJuHg9H#T zEX){A7gctQ-vkG!_-P;l!!+W*MHMK@!bKqL6mk!)NsolT5zXfK5nNq{bR#b%3t z4U8>H$GEmY!83O4)k;ka60kVB#qH+}AYd7fL8AxZvSGtuaOO#ECal-qA)jg4GJb2( zJQ`X;+!q`Q!%2T9%?Y3Xgv7X#2IED^*8u1) zk`;D9?w2DCZ-aysdG%YL{^fJ}X^}__ZJZs_%7-VC{YFavsP{?#0y6t3!4weCj&EBO z_iHXtijEpDyl!r}9c6A4@<{FggTY=9Dv3qf@(sXjLE>3)G5nE+O;>ej((WF`f4Q%1 ztMA`eFb6n35WBGdaF~*{Qwt{1K<9T4&tA2R4>%r%!!nnFM)o++s1a>gbkdXAq(ije|u;8fY=Dwo5&@G3fU2uSnmq{IDv|BaQ7tW;1K4*o*I- z>Jv>@3$|H2Fm3YIZoa^Pp7fFC&3Y@ELu1L=X3m)(vBo|JNu6vN!s{CsY* zW?xY=DZd@qM>GCoMm+E6{9-=75tq){Eovvkti&&GMQEktox9LG#TmB~TWzUWFf}$& zxO?Im$C4rXan<y<8}xaG#KbQ;VupxqRVyB-cTSt)E>X<&Y(Dh zVMd=?oiElW(8*}R6#MUJpL{|lGc*0YPAI=TfmDp3aLa;=iNVU=u3&c^Xc&HU!0qT> z$V|pfS_jx*p3&LGTEo2K(xezq$&gM>Z#j38W=vgojk`gh7QbC?$I$GYW>?pHt9+)C zFTk1B5F8E7!ykO{ozs0vdN$`>IIU>$EKps}fy$wxA4?!C^zqxf%dwrdEFn>yR7F&# zRv|m7i+*Sr6Ib9JGUx(TjSG87xse&OY98*mbW>;FkKibRX=`j6PPWQ7lp&zolU#dv z~8V_SrZ!yR8z!V>QT~lTj5&J&25)#NG+1WuEh{gTHg5 zSX7Rl6eOQY-kd@}x2|oS3}~i{#2|={>PFSj*00o=?@2HyqQR;$hA7k>{Yg7RQ0u!S zSZLE_%($|xc#)c0?biGR&OT~9OWp1rHYFGmE1|rXbU8oIzVVkcj!Nt=8)ueKnRNE0BZP&sEl5U_KN{`}6u@rP}LAvTA8)=eRzeI0GV{9N@&goA4^~eM)3kd1?F@*>GM~ZtQY+A#d$igSu)%3HH54qlFu{i|MN0##Dgkr&*I+4vhYA z>hpf?(i|7>$$f?PF>@twBL1uQ!_7Oz2`?k4cPFJPnHxfq;cIOLD1oq?G++Z=9J)aN z0O`?Eh9-rcJ(O5#JR!~|WfklC_Z@u`(eG}ce|-&@TQEZ~;}K`GjOSt$G+RiWNaT6| zI@>xAVlf<6vNTZaH{1{c*}iKQ+NELlRFBU-|ETwoCY}4cA1o}ay5Z({GegH2nPf-W$wCTqA75Wa9jL-GbdfD^L$mw$L{D$;gywbN#}g4qdoEed&EnR1ychsIFA%=;`5;ZxyYt zi^wDMBU#kKdjb<}Ug)F6ds=wnqMCJ(DG{+u9Cqp|G{Uz`B- z;o?yN>R7^I7++N@>`j)1r%r_S(o8_q7qZVR2S7%C|keE~X&LhKkR?r#)O2#V$eMDhj>)QLLClvdpV=Z0)q59AksoVtyE30$Hu)3uzRp)& zg{k%w_g5jc?Pn(xd>kz~nO^#_Gap4i{A+usi>u#0L!X?)PVkcRT*FUt{T|8zOwAE`<7)E;+J{&^TDi0fn|2>okQD}PVa()RQH?+8X9c}oYra3T(L28wWpj1tGR-*v$WZ!O&9_an zjQxh{6-6AC#WmM_%?9=%zr|dHWj=R(GHoQ=p{s{S_1hsPNZkE}?+JM@+Li%O_Zi1_ z7R*+nkn6c(TAG+0-|!6W^w*tDiw-lgNjGZHtim4ki@n{n_a;3n%lUCsp3E(^ zP?c053Mz~EzFVMn7wZ;K&6SmN;)$hf$@3QP!5$y(lm{OqE?Pysl zlGexXw&7KEU4O()W7Kn9g2~WaIN`jy;vawTFzbpjJnpIp+s~Bo9)xl$X|aP%vir?@ zB#Ijq5v!USCgVckoddt%Zos$(KWA}rkhD`fu!H~U7_!z<$*iCXU--D`T;4K^L8m^r zG-Mz--z8vw+bQ>rvg~sLDLc4;7}!o^2&ZZp_Bc1R;W9f<6S&7LDz6>2?wP8*S=4|T zWlaA~a3q3Rmlkoc$W?Qt5P(*!^aOv|Jp%5Q1v@vjI9ps*mZe%oLfB|FnaP$$Cp2)l zoFnQi*mUZRBaIb;?XDj@ux`H!GMq~e*_F)EAYhclY9?>~G5~O4nNuZ1alBmnMAHPS z>Y?P-T&?6w`irIW&6x+z+X;)t3x;CXMQqpEx~{IM0GScN4#6O^P`V*!1}%*b_j7JA zzuemjAS9VPqupnq`A@Ku|CPT}gNT4a7#nsgFx(skjaa1TZ zAk$ViCEvwdg-le|ugj?Q$Dc16J;(D5S08sXWtPL70ksYuVFt4-x|(I==vj=qZkIHM z0{@ zo9-HTWVz1~a(({v(VTD}#6frR1|NA1Px?sOk)HJh0&zThwD)Z7x=WXtS>!qsYcFJX zEVY>`0WXLi`@HV(i&QGnpMVLSIA}iqwUgL${G4AVDpWU9tAyZI9MbeWxuPXew; zwzV5hv`=SbRwPgeEa&^T>=HGoSv{ri)5`Q;fd#HMGmf z4P>OhrTcu?CO0=%^RJPh3IS3Rj6xm1DRCFvmJh-=#o7Acp_&U-XYZ1Q zJl#M6r$=x30GV%=q`jc-MvR4c#A%V%vr92t;{4>64ka|%q3t)sUqHZA@_84-hoNQqB zTi9y3hT^E7?m{DXYUw?6f5&W^C5s=Jphd^4|y1P{$xA0o+XI2Ls*t1=`Q zC?aB*prnd-{q{qw^hN7u%2eZ0fwbnOItO}godo-!cT_wpL8IQx1>HGM;!sQoF;c!i}= z{2~8I%eAe5kB%1)Vhj*oU`ygNa8v0R2CFkngI@55k`UtmaWbUjW|iZ2Bs$odQNX)M z?G!MO1nWC#7=V0IN(k)$ z8|m!_;>lZ7M*g9yMw*7mtuT@KAaB89TEzYy9*T+g7W7ziMs zeMITdD~Qp81%L&T07}>Zfb?~?0B-y&D{A{;DfdNo)<)FwUzmUs;J(wS{PeVl>%RFo zhi)*Zh3i`YC7rdo)}At4Pm?g93&+LjQfIEL`nCDXaPFWrdL_q2Wb#s?muB4Jx~<;r z72Avp=+PTXu#dswXtf!P8->IE12QCr;$h&%%RwPGYIxmjhPt9UPZ>)UKcR0gB-j22 zRg&?pUf3jCQq*@wjXkpW#C?U{&qaKk#VAn`#{2-&lm*RGj@iLWdp)5vLI<}Q3yAd? zup$^pU#u3HlKBA0$^~Z~^A}ZZ?GKr2cPFm>p$^oU|8X_!Ui@@pE3GkYDG zD-ouS%;G;fv?^!eRBG#n$84w+ZAqh}V=sT2UI*rXtWQma2-ubgk?rIoxt95@>f)74 z1J0f;=g8xf&o>Q&eaCXQI{|;E_TuoX6Gtl(whgy>9BY~8FtpR{o$rWl^%)idV+0+O zW3z%F8JudSG>fHql6YNxw4;XncHD{)sV=Oud7xew9!5+Sx?N5-Y-{?YVu#@?&4s9r z;5`*8J^|)2=diKTSC_SuZORWuQKFRb>(4H48rjGARYvePqSEKgr}L2)OW+pu2|4fc z2L7~?OZp82y_G}tvm`EC_eaeYtZI`_KI|MG-asr)XE};>_gOLJz0B;ZYTdERf9?#3 zawmUempWbr#M_1g3JmfU#6LN$otR8Pr%w1MJ}R1eJ*{hfu4vzT!VER3U2&&`t7 zO`sSLb{lt2KMWqloAWUREo?886OId}Q`V-ZwWSM50z5#y63h;r3}-AUI^yeQ%|1Hh ztCY=!WP(#@37GQD++mNDEE?c75eNF?pFpg8gLM*{;4|?NjHK~O(u4l+0Uy{P!fT4z zJn)g=eJvCMlnX&-@IKr{i#z-`tKOkFXc&j523X^22-^3uO@w9XohxAg|Z|+Sc$e1GxyiJlINq98Df9F zmA3Z*gG>2V?PFo{Cbb6RNYMaa_^#tlBU+S2l!>Z;p?3Pn952V%utxOQ5CX=+sp0X) zx9RE4k9>Bw%!YK?g*hrccEer9KZl&n`mf^-rTX*rQ7@=Q{rCjba{PyTVqHdVwZ0NitlZx)$ZbBomKyhWW>eqLgKe<`JW+c_u4#K zKM!}^#KAhv)E{gcpBZj@PO$~3QO<$gpzpI>cp?Jr-nE?4qs~nqww0zYn*Hd0@)~iK z&s_7h=lCoE^JtkxRAea8vUVd<6>neW$?&m)_gOc2M}Kytn{k)ya-@XQRiBYo(o976 zBIonXuOaTu)~C`Rmu6P6G)Go9cP$M{qsFC}g}>-*Jfk+9hn=c2oky!`rSTdNFx?q3 z|IK^_OF<2!Zj)1Ox#uM6(g5&xY6pl`aVr$o3^5qFFb7c?I&@xpAuh2_P}&hn;V!N_;iN@0UR`!`UpnQ2bT^{4O%=HaT2 zZ$W3|e-_T_CqH818nCww|A*w8iSxEgg|rrGXGRddG&9i~$TP4xG|h6Sus~cgf?bs} z2h5Nb!UXy^B>W5ItCS*R1E4||I78B!Zhl*uj+@zsl1OWJNDcS7-%E#SRrgaXb7hH@ z;$>02f0#pPTv&_)45wG7?=l#MeU0?C;O!~3hx%lq?hQTuJ^z3yNcO=8op}mm=pS$d z$Z~TuQO6RLRweaH(2{nc5Tc@JPFgkC$O8k_;iR1ec4C8LhRmwZ%?-E3iiv;$aY_5u z-YMD&-f+HDO5AdEIBuwlEQ=t=A{r?+H;oEu3?-GD<#M|56#(>stC8}FX|Qfs4@}|- zTS9VZkSE2v8qOr_Gu~qTij8LL1=ek~TTIhWPQj@sZ)ajLP{ZHsr98%OI zP1U5;f4(%@c9;|g2H;}82QE)VgjS>VX&)0!hPa%0e)3D3!RgH@>Dt}p0Qd28%RZW7MFm`zr<;0l$BdG?gSGOs!0r1$Eave#Wm6UD@ zNlxfZU-=I05QuNey*vJ>Tggb9dgF1HBMj?yZcG?^3w#ZL_Y>4F#^&#?vq>^B=!bGu z37-v^v-_{nl{pG^Hr58b!NQpf#BR)VHQ5jUy+$}%yf!1q%H8VF;!I2<~YBoW2& zN3wET7ZBiw>!@G*>oX$Yz2LRN5U|?{khRwVzdd;Al_m3>5G4OLlHnkI60GMP5isi{ zn?6JlodF&KgoN5vl88%%E%FBi^s^?`^DOsIrg&sa-5cWkITW2wk*pJzSrmo1JA#w? zBzQV?lew&vPi6&k0n`cVV!~oMRR3Ak|FgE60BFMfHW=n>b=?^Mp?{Hb0e7>;`t|-3 z=nt%L@S#oRc7lFR1UEy=@l(SF<;$IP!+nBQ7sqT{ceUr+;p=*UAsG<=S)0wu>w>d1U zU#U=~QBSo8I5wz^zg?ND1}M4{?s0!N)DwiESn|B9yxUkWwQ)~nvAC0!?`JG)$etRs zwzO$pcFJ4e0`1}H=&L}5ej69$0o|?np$|Sdo!77G2PK?80hS7gQ7N#!`g4I3rVx^+ z#ey{KNhcGA*}m{7Iy-D&|Zu&%aH-zNq-WrGYOn& znyU#DqOe&H%fE1>|16Ll^HgVPryl+MSgE-G$5;kraXr;-_yWGXMm1o|sp9E6S+FH6!0M}KF*l6u&#f7?yz zm_EJ?BC$w+!F)ROxYp3kWX)sZUQ6w`F;&i2EIgAVYllSmg~A+)bs~DcP1xL}HLE4W zRJpgs+G(u|6->vFRG}AC=x?d0s7!riNwh_YfQHl6x%(!IxXxgP!zus_ZbWQ3 zu+NoPo@g?HP`~tO#QVVPr-Uy-{bTj00}=(UK8e2pAgB8y9%r%PAgQ%P9oMQc4e>+? z@mm3UeZwqu2ETtyc@Y%ww=to`uIeB}EPpl7`M7sqE#R}ce)gL>@rU6sKrkQG2V?*l zhF;AkBd8F)uhvQ8TetvM&~~CW7HnAa96(P22_B)C@g8??~n=tZ`|%?kPd{XkDIYHot@n{ zw@|ip|Jl}HP%LQ~V)&&>w~Gk)UAV0G;+8I`o8MEk9RbE8-4dbt5CLCx6p)HeJj^!Y zl;z=Hybg*Etl+U!u|^eI3?L*Kohs=iydMu&t&}f6RkR6`xv1+v<)3b@3bq(1Tur=w^5NN$^)i14` zz32to1_CD`TEOE3Qy*Zw5;>*4Sc-Mrw?FC-L4lERbQ~sG7eKQafKA6YLAsv;r5xt~ z8Jc{_wV&87;`wG(m?@dkqjCZ|l)l_u=T(@&-%mon2mmxVy~H?elM>0CTU3mS9XpsA zPex^Mu_sTfPSyoKJnZ=y0d-VhP9sYx0u~WPSSS};zeZ!?{!x$__&jX zIa09c6p8s@wFd_lD@WLeu_B8M2tAQKiKEXi_&mfRdwTK3Tp1!R$Xe z!5O}W0uVY~`6rglr8g00K_R9%f17k8K8C9^7AT+Qp;T1VJ=nAQQl8H5T%>xlBBwFr zrl@`PxEl00@&UqqTwWjmNO3fO3f?-EyR^uh6r9E@HPf9Vy<3AV{s52I$a~%;w56Q! z^P;)4<>C1_gcy3pYrYm*?;$gF%&{%IkM@e+nnCwAY z)NxTejnf1Dx0rPN0y0MC5YP~Q3qD22`FuS+iGIPD2|>POQM{T#%t>a96zW&=Pi8wh z=sefQUx{1jn!Z|>H)>p6-z=b(vRDbbx><2Pc9&GyJ+~KHYgt!HYVQNIV^l;fU+kIw z{po^!SE`u%%w7)9A+KD{ANS`kJzk0@6p}>ORL%GYom6FdcF5DUwCI9#xJ zDZ!71t3)5j!WX|;za^V zZ+&9}PnNm*NI~wF$K}%-9sk#tgggPG90676OvEFm(g#FAW%>MF2NPrUzf@~d=|E1h z7w|_ZpO2u>%B5NrJm%*fq%&!9+_ zA8lwXd1$l@Xtphh*~zw^?)ru6xJ}9(tDFPgHn+3FV{yfB=_-ECT!Ztj3B3L~Ya5;e zo{utWb(}*MI8tX?>J|sRJJ4X3;|3dKyMk1zdQk$blMW#tkJ-^7vbqAogfQP3T){y2 zk4CkB)QSpZ(W21Q8!-)y!+A~+sctV29;u2EL!-;r$Z5f^#Zm|d&^ z?yFK|^6*dP8c)!4+ZXvUbvW`2$CK0DewEkodk*M|icR+wj((b_Okwt`vo-0f$|z5( zNXO8THPILZn0;=}BNo<5vRXqXT&NSR+9OEPI}zTC;V&e0c!`yM%s3~2WB<>wsJWi9 z2tqqUVFCm3rG+`UiXvla(rrt zQbiUXSWN$c*z9@N&Bh7d_UDV%BQex_!*5u9vlXil_$AFr^G7T28rXpPmg~|5)t$SQ z_ZnVSX$}r2JNVl|XPJH%ILN8errSlcD4lf`k!lK6A?o7kK_Rg;(EQk~)wc6}gSchg zu{7+Z=4s_cW z(R}rg(B@gL6%J1+9L`a}>JZlihaL%5t9}r43hKEGG{TTi$C<41r~6J@aZFCNlBymO1x~}mgsnC7ZvlJ2!lw23w?Q@QCy#5Fyq8jI`Up|O zRx-3=Oqy&n@=JS>po;<9de>H_uaA6obbqYA3x-cUUre3hbzPS}kv>cWb!pzb5fkmp zf%t|0cx5WY!PJMCNWWCU>U9=O?-Hu{Go9YWXp z3PHe29zgOOPFy4yg|VqxM3gJ|AkKCRU0fTFO~TE(h8WL95OafRFtq?5CM1q?uJHfc zY{($K#fzYWArhabSSnF?48+KL02`rtG||5G29zZ8DmuCyDshyhU;82B%mTs`p*@gr(!2 z>XG_PNOi)^Kx}npg0w@E4x=%cSQi&JN$(;q1kxqqh;1e$MD1J5=BM30Km6D2-r&)f z+%S#rvU&6{Rq9deqgiKhZV51m?wqs_pGO6#u*EKhR8%qxcWNr00m(;3Db0`^bf{-( zn$yqa0vD;=Tk)HQ_PhIWLp)GAj(?`Mb=r6mR47vsO8jD*X@F7 zMj*8)w8sdE6mF~LIlxTHAC?GwFTWP~J`#SK%I!RXEe32XgT-cE5r#`60mN6Nmnju+ zC|cGAQfk{1;qsv$>w5hIt{kocNER2ImQe$Wlv9U+MF0!*M*iix+%QRI8xA|hq&&-? zsW7s=Hqav_4-ehI5>ajKGA+N1M0m?fYM}pqM^*%GKo#IA_jel$eB;{e9E5x_y`iMD zjIdXX=7CT^CwM}-bq#&yRaU?(TaXsMh&e-5mUqQl(X+Ns*D|cvaYk-f{>X6}x4OWW z*XI@WaG-|LO*{1F4YBX53xVO$T*A$S0_oXnaqxE>G}?Nrb^4`K6{tJUfJP8`4o~37 z#XTgq%+)Uc&;P%~0Dp|ALP$b|&KPH)(7XLqrnG`Yhd4zb7y|tIb`2?09M6o-dLpRS zsL)L<0Dn;zt0;A9?y)fhEKa$iaTc?*FD~u{-6B(GwBbqQEay%%cDbp$czYEO&jqDI z@9CT^gzwv5!WvmWSJm|fjXq`v*J|S2=+MAQ{@#2+AuPLW%JM*WH4YS})QLVV4R4&7 zkV8Q@@ynP#skGGSareCdc9b)$`vAu{dvE6IqbhwM7%fZJ1o&Td_CWCI-~8tyv4f5_ zgL9p}^~?MofZrjSsz2gVszDPMVbGaXO@}h+A7jupoKIKjrw)X1bC^y0R|~FoJc8a>tfyh@~Kfg5?^{bvcD(#YGcI;-5B!L{u-f5{S;Fr(rYF=LwAu||FJml`xHew|7> z5A}YkgzE#cL>LlsyzKraoP!mVVpX840LoLkck6RE8rd0|`druNSq;;)iW07Q_;yjK zZ2icO6>r}($k$4Z1v7HpO^1poecUpiTYEDs7!w#i!>#y zxZ%*7eul)sGoa{VBD8RLJb~?=DJpeWL~DvkR0&M?+^AnMY@}*4;iImS`XoovK=kPG zAv;8=jrijPxz{US8-vi`_Zb{nh=~9Q5>TWF*Q>K^v{q~$-NiCLQCj+kk>%@AAuYQO zERPNkdu)u$a9+tPo5Ewtwz8_HYf^$YBnj*B%2CGe&tWh>*4292n3cH#acOaz$dsqW zzqm#JK_>OK-~i2l8I{-gMTin<`7U?Hj_jF~u%k#_9^~c6&o9Gn(+?W?5hH-&>XFjO zUy5#fht%~NB(a6~<;j2x71GgBEWRkpL8a<*n-Es)}~@|F6CC{%R`g z`nXc0N)0s-Iz$LkrK5oW1_c2{q$m(N(gvkPKn#WiLTKtBA|(U`6jW*yiL_8eKm?R7 zAiYYJA}H@=W}bPRXR+4%56o{jE4g=_yU)$u=kER4-vjU}H>FF1q@0eR)k;GaD|&I@ z_S=WT2M0$W8WrCo3a9)U$q-NwE3ZbPM8?GAOshgX8we)V{}eo|(mx~`^t{wbAjeko zP3SU-q`0QQRuDMz(vzJ0S`*z9EXF$HcgnBio`qA+ey$GP2`CwnKXDZ;$b-o|9UFgHI$=ZyE!qUwvl5V`;=W;QK_Y z?mL>!I9)pUMN%Xzv8(=dtY)m%Js>o!VvwX;ZBQTx-9kQ!)vDtR{$TR95744!iwU+S zIBw5FKmu*HCt&b9ZE?R_>3t_lVJ9=-j>-aY5b#M*JVumKL`J3Iu}^qn`L(W)EZI%I z&(NdnIq8748rbl9l@SZ`2L8FQTxn!DeIWa!q{xj7#1fb3sUV4odpXv z{Tr;|cF2)MPF#eH_pGo_5%+L_6xf9aDldJy@piqw+$)Uoah?R98h!vMRWnb=5<~AXAvDreZ zA&t1)`7|&f*^DQQp#8vzX;8@XD3#hcM5ah5q3Fr7$5V1VC_vkeINKY@DXx2T$Mz{mjNqKGEk&GuzfukN6DOn$v_!=@=Pi zGa|>b_lj`nuxhrA(@6O17dIMO_vz#moP}A%d;2@iJ*C$>fW#0mg1|r>dB)u3l!utt zTEnf7xhp=W0DFIOwa=^;T4G6Dq-z9$y;pJ(^a3B{FAVRl53c|)Q&rOdzvf8L1#zQE zuGYBI&8ym;*`lE-tQP=tTc0*2z4C%eI>eSwchKvh%9#D#=RAKs* zUoQIBCHbNh;7&%7i^w~6N_BbhM?@6(pk7|xTkY?Y)Nnsv9jfcq!$g0dbMNtPb~z`=N-*f5<>!jkDs0{#N4I zz%j7*b!gV`Fcg zd?vnvcOTKsC8N}Qg$u*%Z8=zZp+{isk9%cr08#~hFeXMi#uYM{DE~hjkJt=Gso>X+uV<0Nc+fvRy}`YGVA06oJ7QY+E~ z#}o;Tf1xSMD?BSdZ|m{EJ-t6DF<76e(OthV##DpQ>;^tvURb8U{y0PM-3iPl!o6Ed z9?!yeg|C;D7Xqz{6kDatm|HaFRkcgqUZahZv!i4!+sZ{igXSU*+E9_U`@Tr2j-T(S z6^e0ZBW%C3`fcs@s5Cc-2~`cWVPi5CTVz@SY=AEGxnV1tD{y$7o2U^ZRqUr5gr8{9 zu3q0bnaUZ>BvsMKJ0NG*V`FuyhqBERVZryjhW_q~^5p{_dHH7pkuY4BI zomG;Qp#Q_8^#hgcS49#XC$lNBJ8i$aG`1QHW8s!$n;aXsqcm>ItC$t_b6b!JKVuk1 z__Zz$ETe#Vj2)E2YHO?(NxW{6eTW5g1GRgOnPx=o_`PiGJOK@+NU8HF&E&W_7_tj$ zk8{s|lAXA}Ob<|ZvF?D%OJ6~juc+fkt4nt7zo(66`n}lh&W5p^vRjY`+BAwE>FiE^ zxY6G!iu36G0d+_P_E6H};!k`=8Mum~FU(b{Zs3WzYkEE^Pi1eYtfMJXSqHKW_v-Dm zGS%$>FuZ!1KQKJvoM!OcN1W7sV6e;&wLa^;EU}@=_4)g=f;b= zkz>J?s@1EXUNn7+{F}e{13oap0ZEGJI{^5nrjm_F`m}V_F6%!1{FwG{{40Alm)h0z zRFOT2`;6R+@^dhTWId^x+VUqCXhHKb%~x4KD_f?BUb6@M!se9E3wCEl~he8NR_w=0sR_beXR8t6oz^41puZmPlzb6ORT77(J8k& ze;00*V`Wbw31mpo*fK%9r55(J^gbd$K(i)3huFDLa zBGG&uT1%Ff^0Qc%!bf&oT+w}oJFCWPB3rkgWI-FJ?cT=~btXr>tYpjKx_fN%3^)nx zFC9SzcB^(_ZplQ8V%f>)AB~_wDl`x6vVZB{g0!W(saS^uJ?o$NA2{v z(tuh9??vy_>j}$`#8cUqVyW0{YSB4h<2)!~YlV6G*0CdCmmfDhC6tsy5N`H9&noIs zPFsfl=m`Bt!hf=n`>6tbI(>cpFZ*drJ$NpaJ#0wzPGfG8o@x<1$XP=5(epK$O|cT7 zl@Rr+M-Ni)uu9Pj%`;rt@1fGszrrs}bicCvAhfkzvsrrnYhH&QRh&}#UDJ=}9rhE@ z7ATT?JJCv%`K+9`WV8Pt^x8{@Ej3f!5}*~7Qa8xZvB(4oyY+%bl>VWc?ir*3is(DT zSd$F7b-CE) z=>fVyr!VSrrpjTDs5VO!`qm_}8mfz* zRcSfVjl!)`P>@YNB{3uAQfYz6unQ1K8ua&lS*x#Kk7@46Hn{CehaGCUVx{&gpwI+6 zOg)Y?M7~KmCpQ-o1^A<72L8p6@c~Gxo+RLIS z=OR-3qOBa(lSu<)*yCDnRWF>&0=>$;-$TvYqnF5!|RVP*(%(i11)~aeyEOQ zMd%IrP2~m0O#W7T`_M~@Q%2ZWG0lV5i^HLBl#O+%AAwLfVS6$D*VzKz{c+5r#%*4~ z*jq!MIebK(QgY1cyV(Z&{9Ojz8$-9mLWP>N#$J?SgE@jM>`GHTowl69Dw^rExcKBb zcx9DsXD#K(7fmM97V9r^L*L$Mkw;XX!1qk%j>Tb^nkM3|pk^ES69YXIo)^2=29+55 z;7&;Erp@1#F3<*a$O@Y!I1)=j^6jz+AHM>!DO$oNIxl8|h|^^#tlO`PRp7XhCAlm? ze(b(4Do6+I+ni5zoaO4gd8{B z=!_Prqz%c8r8T`WPm{1L=4i!9v+}AhKP`|K$}{f^<(2Jw-p!+XXDAQ@lS=1wDlM2OGNgAO6&qtPq~E*xKnX!iL);S_fM#~6k1?363g&H# zu#GiX27Z(HrB3@FWNthi?!dsxyuNOGoKQU+coY$HFx(|d=f?|$RvQsID`Rz@{=75uQ#HH~Be^2pB>0I*Yo2~F-Y zJ50k47@lOZh4XV}&^Es7AT$ifV7vu{OeM49r+)S1{~eP!ptvfkbv`1g_c+duqy@c%p#m`Yj8_03JDpGVm_aCI1r@>^3C{b;#-#Gh33 z->&G-10pA#p8)i`*}a*?*|_?KEja1?!H!DLmnNf7{FUsONM;2ocRT3B?@ zm74bPPr)ayc~rTkKKF}(2XM#E5!_HkOXz09*6UIEX9q10Z$n z#b&Z-bl`>AfoSDE}ih>oq2VPhe qxSe}B`F{VifctZX2MkB|4mZ8AR*Q-DnFA!+51AUDGkR|59Q}U*i2yYK diff --git a/site/notebooks/images/selecting-composite-metric.png b/site/notebooks/images/selecting-composite-metric.png deleted file mode 100644 index 3c741b82751ab03ad793a0e6c61148190141662e..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 232261 zcmeFZbyQSc)CWw1NGl@UA}uK$(jcJ3&>+&?-GU%WHxeTu9nu|&bV+x|&;txL4Dnr_ z=XvY<)_T{Ye|&$vv)0Vqx%ZxP@7;T!eSZ75_nGilDss5kWY|bZNVp2}(rQRZ_XUuU z&^@s*fIXAlsj)~%IO0}PQm+)Gq-b6_JD6M9njs;{hbL-bYOD7?%G8UCk+4KY%MDpX zBV+A0|pRlHXX%eEd(mZ-tQT6n!Rr}Se$$T$IVnp4-$zAQvUnMZ>|6UtrOS&S~U5Hg_W7Xv$79WJ;Uw zpTD{9r98whFJr2l;Csj}o21G=PIyJX_$wn3#64UwSuu6=shJZjZg3feQNo-}`c_ib zTrA^W8LEj91`wANrSn~zN7Dey7$~VkC%*f@Z0{4ajKP_|D;~o5%7{fIT^gqP& zl_dxXc|}N;9rl?1(Fe-akY@xA3`qB+-o`L65cf)id_YTm&nBT5t0|9OiC!pqMjI9z z-{Y10pdnI3+BP8;OlK+a| zk~tBYecvj9`7DsxL@4E+c{JCvM~}i8O}0}Wrv)dzG1KM)h4FPDe2I5|Tz-odAk1HJ zou?{Wej`E!Syew0yijXKSqgu?Yz9<2Ac3Y%pm2*sy#F>=^sAm!D;0iPrlg zJHK|OCdb_;Reoib85H`xY*NA;s~9)lUEgAxnkjdYf{bBx5DpMt5h}@!N7i$%wj9VC z7#XzMhugOtxGku9yS5;;*tR5kM|z*1I{E2cTV7jTlAo&Xo=1+8Z41T;SDLQh#CqVo zLe&d?9Gr@whdYalffI=nj3X1ei>pd{M!NkNPG&}0$|Yn!Zu64ojQ`Q&axQx=f0d~` zo^8PsXK(VBrJwB#)6#D%J99hMs|pjXnyq53TDJXoK9Y8@E!alt6zjA#RCv|H*oF7o&?f%bdw<67lKMiCl}G<3FfKE+px;`J|Qnd_j1^r^+KaIvl?{e-;v3p2|7w z(FEnm!{zzZkOm(e9#P%%=z)s%2lJ)5{h)^7A9Ef-v3}uvz%Rh^B(Nk)A=Db~Z^NZe8^zlY}JPw;H;zt-U6Jw)zbxkG0Y@#-q}E z*7un2@!f+yxPG9;=pIQe>VJ0eN#RoAU2dMKn)8d}v-O~R$PXq}wp5%Hm{l6MMyyv) z!XZrCs(q=hww1G{L+hx8A>%*C4GN6}V$-dMB*=tISxUkEt7U7g%kbmLLvOg@mFAVj zw&8a1s1J8DH^}0|{L(bZ?ChTQJM8bpCCeqsI?yWmDi2fqO@_^J2+Ky^CeJfP_!K2KT4~RcC=6Sy~5Uw$zU|GxzTUQ>i>6>~de!Nd))5{qbhQ5fNh)$iy_)_?p zfk~sIHv(slRLNX~6^HH0?ul5uaFtt9Zwb7LYRG(OksC8LorqaK3C_w6EjnGg;Rx`3-`|y&%~CJ8=+k{zphI$A7k#|=~9#k zNegR?2J#MAZaECt?pZ5Z?}$)1*VoKc!G95k(F;d?rYIH3^RtCJ^&P$M#?-^s!%+F4 za*{D10Izp*-d#{%Q#G*pWWOsw>3un$+Rl?d#64V_4ioY?Y3_!1kUSuKTE5SJmn4)M zh`q*s^DS$88>x8_8Hw*664^ZX^^>0S0B)&u9}p?Z?Y;i<58N*3U+`|)GE98n63Lcgy+QIP` zyS@J?-B((*4m}cxC_b8LE0`-QBe4M6SV(BdWJu`17BcV_Lni;vZCT`}NT`49M?pdg zvqD1q=NuK_`|cG3yzkol{f!zIii8RLA_Cr?Ur_$z?E3;=Q2%2a-4i&5B&jZ?pa6WU zn>d@9*}GUexQb?L+#R^*D6iv!g!G8Fj0;jI~m zr=8>7=}01;Lcpe-nX564r=6|6i;$-%-Ct)20o!-GIq7KrI>psSluldu6^)dGvl$IP z$1{#+bYj>vG&Ca4rshIw(z5?F2Y!jtS-QG93UP9JczAGl@NhUdTX1p-3JP*QqYDcQqZy+MD5EHJSWiEEFoFjqJgt34#|-So9R)rYYT6JD z-keocJWW;o9D8`>Fwbu;Yx6EpugugV@VW2F_E;UbT%hJ*Z@ki1MPf#~*=iMwX&EGB zGy;j={dfUd2D+>Cg*!JmuDrVtz)Y_)AEd)2=Z(Q4qzlj=mLorR%rG- zLfz_P{sZL^NYJ>V-{1-$E)~lSMHyd&U}gC{mzDbi86%^i5FBog<(HRdLKhJINR3X| zeCgiaX7NFC*k~} z=OEbZTN!MHDGthau6FY7{HR08-6cs``Jk}%`iMW85z4i{NUnjhCC%(4?+Nbmd6!NM zWlmezwH|AM%lMXHEp-SqWQdl^Fz%l`g$^?0OC=L6WLf_~rRu&1FWB8`@r?M0ad~g* zz33kp=8vxuuTc6Ww-H6p1_GVf@*!A+FRtOq97c@8nPNQ9cn$-y-b{9-f~9zJ z8iJViML)#G!Ge!@vggq%20|2mabyCogi53e2gD|@=)7#D{H3?r9ZmA9wB;;a*LJC5 zF4G8OP56n`D8rfmEB+rH9-;op2MbM0nV`v%R0hebi!(9u-*Pc4MuCAYPfJ!pBh$>< z+riQu>8<7rb5P~gpW*DdJ5x!+O>Rk#xv~+Wr0;xYoQb)DCxlFQ&p#`+EgFR^Z{fU` z_=RpgZpGH*4&~!yQ!lE2N%34{EBLV%&ZS+w@A(=HVm1^y9t&>0?OP0p=jYIGW_bTG zu`d5=P|HXWmWR84{lzvU3>qQoa~o@2LpC!;Cb*|nsJ7jCwg15Ow;)}J60+8cxec9h z#YdNv^5qia&CC|ep6MfXwr(UN^*niwvHXZw-a>VZY0i444k_)|5Psiv?UBBU+0~y> z>;X}aZSBK0#50=wX(2;K&!{W3V_zI^Gf{RNuPJJKE$-AT@fCk&1r-u?njVJ`es|s- z-*8Fg@yPPCJE(K+%l1;=zkY7DcJ=;~TD>Q_PoIN+yH^3BW7*608GqDoLBHWOLrjn% zj#U_+YxpeZ`!WHO6%2z1!nBNTX{6*LDv(Pfe}+E@4^73V7GybxihsjPD`$>liZ}aY zp)gHX{Z=P0%ir(` z?F#Lh^$pr1FU-pWhXiq9eI6aI?eOCbR<&GHMz7=b@Ouw!g7^ z7c1?;e9t@H%ru*&dA+=KwVcxRzA7(np;jYbVO+lQif-k0I(BuLJzDKd^b>v_@=&rM zQoT_hSI~boSI~cgh1m2bbz8`6%Zsvz+>cDF{4L&|?2k#wJ zAS{b_3y=eNEY-3Ejf7fToXAJ86wTD*spbsXesjX_64gpUV~qP z1*i7FY&t|UwN_?a0MF#!l* z6$;flbKUy$h@RV1h?@vQw_JwO+NLkgli}%r^W*F8i4iL`?MskcEYwnRZc7|%eTk^E zUHCpP?!TH~iEe8bODCJ~4jq;{cbRXO_DHSat)S1+Sfo@~bk9~ryxG)2i!;1Wv%!7; zvxv``eztTtUFRp_q;kZ095AH4h0D;;v1wHV`7roEQYr~T1JN<|6G zdwvVQe9Cp23;{ltJq(|kFN0q8!_#ma$atZe{2-a;*Vn;9M6@r^#HR!g^qc&pZYV@0 z@hC*tx}(4`RS*vObb*RNpB970kZQ}xp0Y<}Q!N89p~dG<=WM1N;(FE1ds+kV`rdyc z89ek)VoS}nTtOMFlrgayy+WO7u!$BlC0$Epit;-mO{rVvBz2I@^npVqlz@ijDFuq%aQEM2KV8`fLz^D*|Un+j)*{)uOzTK6@fMqIxgq4ZY?}^2i1PQ6O~?Z zGgxkV%BX9lv+i-HOmRSy%}9=1yi>ppZRf{F z+)ZfbUJH@bzmA&D4-~?p#o`l-CmZ+kzK`asf_sl1Fb#?gAw#n0yLy00<4xw+7ax22 z<)SQ~?mT0ET!E(h4c1#C4&O(!hjZ8+GV4sgr}sulUCi8MTZr?_J5bJ7+lLbh1Q7ZL zqz;F{e2X+opOSgIPYHQgGlBgsW1Gh8qlGck`&$hZLFGD}mpYCUbg~gNOMYWhIGV!X z^L1gVfS!@-Ph!XGy$>M$`+m63@kviwwyNrFbdL;KmoUdqhrDbZ#???*09~fkKm#vExcGes^G%s3i@#xZiCeuW!BzP0LlL zJDvbSbk~1)q6GypSSzf=a zv>2w-ozWR)*Q=NnE7#1axwO5yINCEoeotKU!|GZVI^*}?XEZ0DjYd+P^G1Sz?CC`5 zoJZf)`+(lgPs#gYx{SKTGQ#{22wRMJw5228xt`$0kM~NIjz?Rl&lg>MCWZ8Vr#{UBww3$;?4Lq?`yTs3RCNA$U_y(AN*2;2+9@gD1q zzNDn6eLbZpiW$@XBueuQM_eAg(zP&XKgmKC9v2s4ySB4e?pP3BA_Kl6QC!BXD z3Yot~2ugn<({w+WrN94(JrO`eGk&fe@qsP?#?J*svZ$5F(+WE6R5Pd~ygKPdMSicJ zu{5{P=pM&GU32c@u%J`nTv}}lTfM)J2)fR7%JDWAhj6`v!5xVmRT7>B9SIh*Fn%QF zpR$-ao;dR{awvh#RhYTO_&;9ZSk}+eaAJPl{K2&UvyMu-2)lazaMCpIF^ z-vm&dgjCr9Q*tC|}72g*T02m5;mQA@@F3EcP-h|s@)Sq2a7jKr%&ViY( z34=j0O!Ky*N0`6EhSEGl`{CNI0}%$MzwJvM)d2Nq0@`<$=C+D zg8J!Jm%~nc@9qBOz4IB&NL{IlCU7+W!U6G@cd(PAzgmtVSQ~1 z*V~?-P90z01U381j(l9;+u^MpJTZNPY*9Mj^k&F(1gXU3{?p|WzfZ|!Z27}ijOVvF>*>@z+;alF+Y9q?ZfEeAjH7 z!^z#VxF-{u?eKr*;WL-gR@D>O5hNx2%nA304b7r8vY*9PO^&SUw?@S|e)e3C;R66F zZHdmh?Xi4%#90-6zBi&gTjw{t=usU>NZ-hoDQT6grIKbbw%^Gkej@kh^84}Z&+kb(#vV!@CtGIjx+mTLh;&JI%X zP-Z2vSuQ_nI=$9Vt+}j&UKGAQI$L7tAiL6QaLu%IoX{)=702p~Xbf-kr5fDmK3h20 za!G&tBTTL(ybbMi#TR=5x{M5;4Rt^3hOEL!M)g&gPUAfx@b*# zL!)rrzfEyTln_@!hI|QM)s#&=gWgn=YflwY1HQf?BUT9;95?r|`#B8E7j zPfoT$GGH~*Q)b1nc|2=96tv8hUC)Ix3CWzhje&}1xF0)h?0P-44TeAuQgrg8T2oI3SIr#Z16~GKB5CJ!(eEWu5*qRcx;8$sRU~{k*{O9a&OJrg@b< zWRzQ?-|E{bl1ICMx37y3gx0U=cfzN??`Ka~)->E_1$?XluSSV-aIcStU5q>-{b$5b z0Pl1Pm!3b5KnX-_XBs31ot)WADt|>P{IdYr3Re*`|GjDZ%)+(9d_8kAj=CQu9O9`{ zxwmBfZPtjj!}|U77hXX38A2Y(g0M%a9Z9TuvjvM+bdI~e<%4>4O4p(6`yKIJ(cf5y zt=sj3H$kj`yLUz8#y@0985VlDRm?QJt9>f)1(}i+6ip@k+WQpVbi>++jxG9R;@^yE4nM^!Edk$K#;>p zPhgKSG|5ooeDpEV#%yD!rqX*5V}vHVnBQ&`tk$e;3`;PKr?v6dS`ZQ1d_3t!0=l}K z1Do~ydQBAxnJ;w8xs*NKXgxk$q3=y-3H0+`4&?l*6CZj<489Or-rL<}CFi$Kx&uM( zIdQNv$|h;4+=JOkWmX2S%G?8$CXoY$H}&QVPbyuesFJ5J-G5aSodJ_lt^%y~hC?A; zAa;E)qaR0Ts?cDe#D2GYEC5r)YjGJZ9dU9yi4(DF$NTMexGeI3>brXZ~-aj;BTgf;-4OCzfp_N{568P3O)G%#(lN38?b16a>t+Bwq{sLt*uo>eVqu zSMaNX#rTG691xi&dn1DfBzKw`5M18l__j~scyRpMXM3_H{#qo+4yZT?xTtMwp8Hld z*U#%(zT#6HNqX9(@l>m8-t9bt>z^phEqrgLLWSsOv;$PwK_D!qJNibduhpOr2V~Q( zt2YE7+uk3fL&+G8P?1^~`eiGbgT{IkfVyDJXTL|;V>tymQajd4^ZuGX=ukE7txTMx z)!}t0TK8wo7r+eL$D^`mGvHNsHXXJ;b0-0fBOU(eR1&2aad?i)Ws@=k7Ysn03Kyyu ze_k4*I(V!X~Hl4~hp}7k2eL|+K-b9XuRqny_ZMW^iZe{M{ z&j4m(vk;HkE#B+?4R_sGo-#W*h|}FI+^vaY1U)*sSFuBcUFXfkzZFS{bs%st0J9KUrrBU)WTpMWlbL3>CK3Ou z=>xr1eX9VQonF{1Dw^?Hf9p~Y*yP5KQznv36Ysi8hE678%RNK{T`}C>u=>3_x+v?&r z&p|j@^nBD#{YF;JjwQoLf{eSV)#@92M>mRd-R7eA)bB%7VnrHUNHu$me9wP=Bvr@T zyVk3|^nCT%rCm_7?_sw1H_N%!D*XoUK`QI%81Nk>^Sjw3tdLNkW}&qSmU|@>3Djuzq{vy zVIjpTkgt+By7X16OrMYSm`c?BhX@J40}Hx4Hc10f_5ryrvITC!u1G1r%g$qFwfZkL zyO7h9R2Q{oS2M(PF_wVCbiEE4pPk7PYzo{O%K<(k0CN4Xlw!ph0GN0h0Ig3|Q=g4y zR>49|wuE)73!L8!FP-^xE&z$UvJ;qD4b|-~3*~H0t^0M$%~G;aFOaYrSR?aYYV*n2 zDuyA!msU8MK*%W!rq2&GqHcOvGiF9vFme(84mdIW)cb<)+2ms zJwyC@;XSd#LZclhy!J(=oi5)4T<=EfWKN^lzKJD5T#~z}uH=I?I)={~6^%)kGC${7 zLz!NK{AjMihQs61v-CE3oxdPgnZew7u2!_YyLf3{iDqI=XB1^Wj{YoD@V8hV2d9w+ zE1QYz@Du1<$Hv5jH6IbKf-buNGK71>rl(?!X^@shi;W53vnoTYjFQ>(nC-+arWdIN zPk@wTLvstXz5-x_w#An*nYC%k8=~B-HIpxn*1|)Rm{c-3`n5{9I_8>c%eYtxg$;}6 zf&-{&mu?P={X7?n-p7d<=mJp^q(&IixC7oo{fg!`-zcm{wXM0}(G3gA&?Ob8c1iPf zFmiX{wvT&v5}&)7!>+!(pO~RmyAG>mxlxsNb!CN1llOijp+?zLd}=XnRlzoo1wRW* zi)*T1eJiuBwk=`B)(bmYKz^~o<93OfSEEqFP!NJr1EFBE*8T$!_lnQ257NX&>tFjJ zbQgYE*v;ex+s;>aii!E3P^+Eo2XxHOdF&+tkaZUF@gd6_kke#LIRFB(CtMR@pQ)cCa&R6QL zc79@2N-mF^v=gsvdTy_w+Dbh=yf}s1qFOT>xqi4pGGQN0HO--$F=Y(|!4uC5%5l@G z&tO#WC5w!W4tKjNw5wCmg>Oza4L;CJQQy>2M$^@4xuOn7vUwLWP15%IsB1p; z!Jtb}soxxaQ_pH!(%f1iz}-HHBHx6DVJEO%V zY*lZoC5($?U1dE!PqM4rhL*ZBYU~$ihSbm%PoQ86fQ17j`#>d(cPaHorL1@GT#eqh zn>UfW&&|GPGiC?L6yGs0kA}hKm=&jM@F)(yZuyB?4M|QWFvq2p5z@;fKxFH^ zCez{0mji+UQs_*}dm_*h9qu%noZ)7GI73hckir7m3@m(%SVtyH3}v`YKdsbM~7_y**`kN-HJTCp0# zpE50fc-J#Be@-rnDp_%2XO6h>RH*b&56JwPu*ZkDPMHsVW+Gd6vn^*?_T3pLWj=D( z?AA-0lM1*UG5R*S^N9vJuGh86sgq*eqbyRxqV~E>mRLe`Ia+yW&nZfwJ zKymyytzY?{XZoG-aYYO_V0_>qgF=F3+x_<&BeS6j#SPf8U;!`1KFy*vh)y*a6h4&) z4wsRYm?^p_>-PT#ecNM@*-ufMkC)6%-&6z}julGjjVz1)UNLh}%%elA! z9l=LzxY-4DlX|$(cxyb_RFWzmiZ&@>`AbYwQ$N2G5 zu+)WJE&H85&x_@u{j~DdgiU2~A!_%A0LYL*p=NowAYQBz)x zq5$&^Zcn$tD*|}a1Q6NO$2HI=Qu@%VDxVzbj}Lo=)}g0n;?(7=%O%9^ zlpWu1o|seoq9z>3vCyE@X;8g>UwcKmJ}M{Q)Zk9#}Qx4CMDm zPpdMWFM2AFm?)W4Y`1T=yGOHi=!NSczCj@G$qHVaEp5`-3NQ%GKi|*}ezv=K^W$^S zdt$fr$h6uXIq;P#he6|>KnZvsC`Rao2Dd(E6+s9dEk%KX>;Y6;!c|n)XrsGQW0M@E z)|7Xv*yuf&WG9?!R*Yx!6XU8A8R-;Wzb+Q`=avVPV_8LYU2Ca0gTBUr$n2#J6 z+kE1;AU`!>B+is)#SLX;mpc^->gMw2oR`!2X?#|eAj~@a`Wa6kN9zJ`cfQq?ZjCR! z_L#Q*3CLThr$a*{?~`h_mQFTVc#r~4o@d>L(#3aTc=cjLiV;X^M}-aAskZW*_+OHa z{c3Ik0aG`r*t$hH`SYkIj>_qR(fQ5`$sQmZV{kU|#a%Optl_5TyAy@#l>a|8%h~kh zjFCt{qb0Ey*>A#3#yhH0NXIB==p-mO%xr8)@2+y-j)%Gn@&M98dK3tLCn8v~Ya+gs zgJMr}4Vpb?umm}i01CRVQz=;*wexkvAs9oN?^(%OqcvGk4NJpjKO!l?uPLF}{&;W3 zMW@QW<*LOijr8Mq%r}gpG16h(Dp^fts}SOvOXoLs@IF3n`yU$gKM@i@wG%_?k$x(- zd5p7}l0sSf3~H19T^V0|{y@Ulel3M+%Mk{jTDCYuDTP{&?uJCU+uoO;5vwta{vVzF zrnf;-00k!ny8+-XKkCL*`BK7$?5TBejosLbx7|@@`xx7&dkrw|x8nkEi(APmHM_}y z)ZDl4>ljrJ^l!RGXdpjilp^13A?KhR$TYa^*3RTh+%#ts*H@Vh?m@}AQIjJb50`1_ zH@9^TT2Ju#w?e&frkp=qECt9ww}IV$UIh>U`@AUF=JaIr>&G;(s@9tKWk%N2PBc~$hY`EQaAy3j z?CB{`%BJvnyzs8j?1rVKsWF*q4BEML`!p_tE-u)KSj*Rw>`Hp%0o=%g(U*;QlICKXeC{k}S6V7{LE6@b}lrL5>VSM4AIpVHF>yCXb>@`xBWXAv-*zS>B5* z?o0o@K<>!$ zX(;Yroa~)`Egu22u_m+rCkvJiAKhuAGH&4ap!naa<0Js&i2kYP;UCEH|1SJnBmVzH z{X5GT{?EAn&$#|qpY(sj>Hmh){{}z)&w>2^=s<#i%0d6bt#A>~3u!8GFBVm>=y?*~ z8>+u>s>2tF+z*5ditF-3qPnKxRD8$ePumlI2X{`Ptk5lAUG(xaxIuz0$Xq^2*5{KbbzJSO}V1B;wHXEbh3(Qv_|xA|f6?@MXh zcOPJl-Br=~!Y?Tm9*QSP^Az!q{q61!K|x`%e30Vj2a0iM6@}}2rJn|x|5s1=KkmAn zJ_xA7ZFRmb`}u+DecvDzJ7dOQt$#EHf}C3&u9VVgK%QXi;oqr{K+~fcz(}VS@uu@Q z5anjkAv(rB+i|A7gUu5459)2N>Y3aJ3W!9l*B#IIM4tTyoF*VbG3Hns0Wy1+jk5MfmzL#mSP6^n9@Uf3)~eEQoSQizw^ zf{fj*JfHm!b^NbB+7tb54J|Tv-QSS|%~RlKaJG!mzwhCY{|@`tW)J*Txkxh)EGou3 zSL*rq9Y`!Q0)b!~8ddV&t27A6fP0XVO62Q8_IursT>NXX>gUEkRJ(lYQ2Euj?kq`bh>%e#@ijEGJqlkb16a>0=vQtWO`unO_c|dIc=7$sMzaI{s?*mmAJQhy} zW7(^GC)$isJ`=2zJ*%M*!MW?Bo%JDLW90;CdjCG)mq>YV*qfg7#bIbK)$}Z)*LY{; zh46fh-7ukGz;>gc#jd7@8iVrTQd`KTP0EWE}Dr` z{pK0gtDPBsXH}q+X7Wqt{Ql;U=~aJ88t>mWX_yKKMN4)t^52g+8XREJ|6eikdhyNm zkMXtlvh_;E{0?(2XU*8E|^xGEBk{{6XghK8f& zLN$iiaprjWgq~QVex8<(Y4)?Vs#N0sbe-56*cDz>BF>5&1`QlD-1&*Q;kUTeZ66yP z@qzoL5kSd0T54Yw!JPW{z8wjrTxvP;o`7DK87`ZRrg^coi^NvcRP#;@Y$sCK?C4ne zcpeKZT!(hFD&H@w$tY7PJ(S=4wrvSNXJe=ZF(!@+6p5uyvh& z)9%F%B0fb^W*vs;*ENC_(vf5`Q5gbEVbS1N9telRx_s<#GaH;zF-#aukpAN&+Z zXeZlKrpfc>G*%U#UGzR^aKEk<3AoXZV~BD9`MFOF!>3yi_#2Ic)lmdo?-(Mer}i)K zsYR2RRKbc7maiG#{`|0NgX4PhgV^ijBHUCy?T43kMW&o8179W+@EPelw@Uf--yMOPl4e1ZNfT^$cvQo3N!LO<)tMQ`hQCY0= z+?hfr_xs$@x%a_toVYbZMwOE8kAoJ0^;fATmQMxrHml7og(RB=IJy^nmh0J`vq*v4 zTqnfJR~G9kEnzFA^o-{;a7RRzqL z!SD0>vC3d2k$C3Pmr|2lcM$TeG`izgE&cgiHmm@9lXyViTnVPRZ zqevZ}YEOZt>TScX)Jk4)qzYNQPZx3-(Pzp~Ra^s-(VDJ04SF3TX|)IO6XfonhbF6j zMi450eAL63*iTFd*{WM36ZO21e^h>K0!%mGO~`7lOpjK$iH6%~jN&SeSr6tsVUOwP9!g0J(8%OP{46YsG8;JFxgy2({KpT=ln-Qb|%EarPvBEs`x)JmN((Vcxq ze!VAlZ@1*owwJEKY%QsUXrW$}EC&p7^{=x>uL8qG9z;R}{kuxY(iaQ%N;LiPDMihI zEaE;WvRJ8cym!y3G|(kdkZS9p3aSr#X)+4gQ;hx^Qy>T`-qilB-J2N77bM-JYyx zvs*mb&5$tlH^B{$Zry#|jzKvb8&;Y!8>qgq*WP$%C3M-zrH3JWH4S;a z3c}e>x(4Pq2%FWs0$tZm%A>*nRL?spn#J&}$eCKeceF&MH0bD2WKrJ$ym;W_?Dkmh zGlN)*T^Cu2sYaFB2q z#A|cqD5bvjVqDVG5y2C(5zv(y)#3RXn}l0@d5>ZE+9so9sadXC@kq{&e403>e4(CQ z{7S61gsz0x=lIr`h1qAC_)X$a2xNPlWE1Xk*du>qB2C7I=TlF zA+kLUKKXeZtdw&Zl>8-uRaA<}LJiGJC3(_(KL%&VeHV>(2m5Bo#`mWiTm9^T<($V; zeQkq$mTxsc?)Jfv*u(jeio=8C8cFkl+=AeG zo}5-=*1vL?IJsX;SOAN{SLge_MOkX>O}d|6vcvoec5fwNx^Xkcxb9@b*~H)Y0vA?V z8vR)up)&y*Cz-D!XngCUln1OAHv_LEk%DOM{xpe*B;()1vG#tnW}!Ph)*L&_W26Hx zi~-%n0&k)&2y~Re<&e1dyl79C-77A{TysNzhDTnP{>yHW^HB}?|HIyUM>V-^?W0np zND~1OsS1KfQ+h8JdQnh%Q3OKoy#x>yl`2(gkSbjWJp@oxdg#H>d+4Dfguw4b_t|Hk zviIlqyLa4s$GCquMli76^{zGNT+e)FS=APk8Kjwm0dk`Fow`9o7n^~mD4S??sOx*T zJJZWzPAP9~TTVee#tw5BEmm_)wBNcbM5 z4#(ooSEyi_Fgj;s<>ClDg86DyiIn$joI$V=!4gKS*oIluOhmbuPoq71zHkO&_FOC0 zjlSf~rOnPo9aLVCh;N7YM8$P{Oynx$7MDe5+@1ML==D05ijRvPh4Bf~&&JCIa7*Yi z*S&E9H;q_M&4;Yjq9;~aw`77pC&C%5k#$B4$YFT`#N;Fvtc?Fge~@wZ$w54jN$v>} zDjaM`ns5f)O}rb@ckxZhgR&P4Zzz{`J}d?gQu^ds87w;k!T6^ATUnCM&nvhOcg!lmp}dA=D($pDZ0p+jJ+JXWJV5szBvcNU z7go6}no6Y`qvHJtM8>7OXJ$W27a_o{khRA?}dP#vO=^S7C`O9-U4V zVz@Nxsd&c@?K$!;ZtOVk#yKe~^4^{bE{%|7cCOA?;M7n~E6B;dJvF*(iM%*iqOGcW z-p9pQ$~xM#xYBkjh=o=8)A2PGBYGnYUCUm78tjXRT%>5>Loa(==2DeML<5VQhw0G; zA)|_G6}s)Ymx6%iH6-H1W8pkM4@8^%2ffkSC^l9M^LDR?Z>Nou>#F=!|c~Ed*#n;MSwd zuZ-(lO&mw-*%M30&lay0b=7J^>C!#)RD6I_@C0Z~^Ph5{T$+xeb6y+Vk?E}!7vp(h zT=bAXUk@I>Ws`{2ZK?%8P>y-hO@6HP3xnoP1Hs&csxDhpK-H?`RULH6^G5RrrEPO| zVx1y&&Wp+SPGjGqDy`;XwCk){u;oF>)-&QT*wAL>h(%jLm*T^Qp!?i=PLoE3ucFfH zdzj47lLaz2(p_58Zvn{Dez7;_Ss_2uty97B0Ea+~|LN~^Ik7!K-+&KyyJ}V98eo|O zkKpeb7~>ucws&2O(|j!)Z16$i>d*=4u7}Y45WteA!pgYnI$;fE? zBxB-|Vn2z7=T=R%k5*l^uR3|QjG#t_!q@kj6lSgf;Bxa0dVY5Upq`W`SdVBY&<%Z9 zDDoViEv?Y0qvq(gT&+D#jn#%=hkA7&(bviDq4BLA-7qGJn2iQ?j)R7tv;Nk^sl`hu zB#yv$=Y}SD2=-XsL_G8qz9XGhF$OFR3q25Y}x_uPbVgveUiF|M0#w<-0qPw7n&f{b8ky7GRNs=zR7c7=w-Rd*(VMfq%X_}2V70qi3lI~{7*QtyHTpy{7mJK(4 zVFCo5Dt%nLW!Se`NM#mfb4zz@5kr3~w#(Kqv}>phnwaw7CF3B!!09b3As$A%XEWF4 zUR|JzBs*AS?9p^4KwNY*1-!P~f=CU%%69kiPox}G_ofU(KDiwmaxIHtNOyv^#tB+1 zMk^XQZVl7+^}Hu3x)RRH^`1W!XqHJ25Gp1a&Ln@u^Z#qqef+|STrH~_K1dDG2}afJ zK6sh}k@4f?8Nyr-`En&}HUV&3Y6I9KlWXKDS!AvN2t4l4$?^u)VZu#0_kzOcxn~oC zH$Uf$nJ)XA?`DdWyDZCcKM;-@C@}8Zudp94GqdV6qx&vR&$i~Gi7ML<7>CA!d&tGn z?FkBoq6JHvNJD0y^n0J{jN9@5uvR6v%6$1bMlxOt&34wRX=j(yc3!Y*?9g*+-D}i}dAc#9n&?S!#c<#%P`+T>X_znJ_IzM> z|Ax-_Hn~vEZ=`BcrsCj(b13Msbup#uAYl zwqMb=fTl|T-LRh>z~0}w@hDx|=UO8eP>#33%&QYF6dtV!JSE?QXm&UxHh*fma_tD z#E2F!jTb@-s3FyDRzxZh8$dc_IsnFYJ71dqJTo|%hVKM)f;!Auwz>ie7(9RXjupCL zSWbKpvQw8)<&hF`?PFkZ;zDoc#RqTHRE|asaO5rX5aNhRd*atpyBZTq*^zOjGIYnZA0FuM2FwFPd8{j zRbmU_mcG@Of>DbhonP(Giii`?>Em}x3hQ`t;J7)EM>ZzNTxweRq>HU+MlzdDI>q>= zy`oLmY>~&_lB{&WXbhEsO&sk zk^-tZAf)Z%e9VSvTpo6iizd8Xi{V7#Ft%+g&IOz{&EeW}0%EG+A^)lFMwEflX90B%dAj6s>(h4Zo{&t$gIdURgJFm~SOu{6D zhgiF^Wo3!2sQ7^d(}e(zC#mocRl~WLTj*nAHF<3b;*U+MRYV=f zZK0%z{63k+4uYgg#HZ|Z(RwZZS33gD_-b9}<6JE@W2U*71QP*u8ET2#e%OD7gZ9UN|66GE3a>t^}IQGGUxI{=|N$1@lXccxuX~+6tshz zbJ^t{R0BvsjrU*_RXg(cS@nz)TXk=q>YVgPJ&UkB==6zkpUk*laJQXM5S?I_cznrk zZ?!tOL(F8v^c5}tZ1TB{cdG%e3;I%8P?ia&a!|{er;+;HwjWOui@np)7&h_L9MCQ^ zNER`2bJw-p34DBbU&0zcISg|5Qx(ujX>3&i2FW+!~F6v)usrGcVz>1f5OL#T^~e%v-3aR-D6*!L@Lg2ku@I`xLFpUmHqQA@hud7s=Z?9KTCynpAgHK3)M?qxaiWZ8 zlqH~BxWr5i;bfRNIsg6)JIN;j`5?DLtD?W*n;0lfz9Dyx1n&z^?$4h4ykbxBP}4!Y zPXRGZNkL*-{U~f@q|r<@xu|Btf_%JDPDRuKWd+5zI}hqgNFC4k&>L7hW*w3&;w=hm zt4lQqu~He&W_Z^X$VDe$B_s*b9b?D`!P>-&Rwa&Vo^UQmcp~Q)d_yDrhpe(1O6=4a zty$oQ%IZ#I4LYLsqviz7xGJo3rJ9hcljw(yi?cGAu+AUQh``e>)`VN9(av(u*deIL(d%Mf#Lbj~Ad6|ZUXT`q&c0u;b#3_5( zpBBZInp1cP25*A1v#EF$geF&L&$Jz7bDVUJ3s1kUabyhD95x=le{rz*b%xPxD78Er zH6e+!=2N|wqCZdGG)&*F9fq0&{1hVA8*i;EuLO7}dND4h?4D1MDAfC<7FBu6_46tUQpIX{~9IF-$ zz>i5zsu4eb%0`Rh6N+;o?Cz+2r%58Ifjy3TO`qI#`!A!ge*}=hZ-60MEFN+G*lHy$ z;61WH-9kU#>AIVC!NM2TU?x!V~#NuT2psi$WR^LNk7R)CW`yrhF5idz9bEZf|`*lp|$;5u)HjseFG!e2)q787J#K zXGA65Dl4=lv10>^ErEy?Fkmn*e>8CV*^idr(v+O4;cdvVXZiUBVmoWZ_8t$BQoR-f z)L^Ry``13MPObfIy%mDfV0&%j0V>)R&PTj89`=7@<=LVeGj|O%u(l+IKhS z5kA{--c~=xpoBK12dI?0bWpmXpWKX*(u-_9w7>tiQc?8SdhihKRel&SYP(yL+tnB> zC_%gzaJ%EXh)^oX3jvkNRcf!8&Ez-d+B`@^%KlDnee^f{U@B)_^RC%?53D0;(gyy% zs~Ssemjozs;*yziu$idMd}o3txV8%9PJMv+zu^)@?4%-^cBBrNlk6+rxh&7b5fy+gm`> z!cN8i{Auz}(<1*6IEyGd)f8_1JP`h;eGNJV8YlSqMfhiD(6bi+-6q)5?m7GO()LfA z%GQ*$_s>`O8=nXj0n6A_Pb*CL*GBoje@o&#FZ0h=_#2;sfiqR&y1;3QUuo?QER+w` z&xqK+Pw4-XTdVx;6ne~WuVYOHYdLYmp(5!Vt@jIcCwPSZXJ$KJo4BIexq-T_7=Itv zedFa;dbJm)&;eMgR^%$|iUsg19eh5nXxh_O|CFu2@hK2L`Jp$a0YG|w5DTF~TK>3h zLTU7R^(Hkh-M3SquHpJcNSgL+&7o+`KFRR<{a;WN3l%&HSD>qQBCS+|{A9TVv1F5h zkXY$ItZi~p7x^!Y$`&nw@24(`jxh36-pG80AKxYcSXo9%@(gGklP1s0eKd)PcNUo z0-OM$E?fB*|GO#ptEUA&Hey?BQb5&MhNilIbO{b30ft@jI96Uf*Fuc{Qt=mbIXeFI z6Iz&Bkee;3xz{k!J>=NI430v6~)w7+oT2QH z{~I9oAA*v^eipE^$YRN-1$sqn1Q4cF38O@nNTEl!+NS+ouH&dkRwR3vYCGaE&gMlNGx8R%;3WRat3ODCaTI>lcgRh zbHKmO#;&S&egiLK)o3e^-Rm8T+Yg$4pT&O{(WV~9|Ih=H=pZ7& zZnGL*rmri>xe8s9-UMkLb&kcVgc3Z21b?}!z|d#2znTb&vqGJ2i8B401O*pyX9RZv zebA~I^YFU(rJ!Vvo<_wA2!0rp%$u_ZwYdJ*zO!=?>DkGZ89J(SF;Vl{gu&!whqbm5 z>Z#iev+eO}jx|TX4KFf&uj(4GM@mWipY^oszmfLsxASQ8PQbiM>%~_@q#7TGin_YC z#cIbTZS6?!Tdny^H9A4kW2C&+Xe~#>1+2PXpC$vpkFDN@nU-62aE|Yo)Go9vmo8rd zZqSHG@A{XEE0`35RWOzEhJSH(ONk)(3g&^M8 zB$^E7rxks}QSBe3II*|Si%bv{9G4cPqTV&&9B0lyzj?{$0QJaGaDK+7(yp}@oyzQH zplIbAk}^Z8i&*ewuwh?jJIi;r8$6PKYh-<|yEZeR=IV_#>>`s%s*Dm({1*uJbD?&@hdN z`$0iA(7bsV)*6W!tyG&&24@4s_sSpe_kpVsW9J$WHaD=v&!vTg%>x4DY$xuo8V=^z z#x#y}cs>&fH)q9$B7e4-8L1rr{vAmnF z()UeNklw?=B@w|#)7Wu?7%nYGZk^l;;dgpUs&Q9Y%6NS*7dMQ6(pmY5yEY`?Z%yI- zb46bwK7O_WZ{+yz;cjy4m4sI~1Q>XOx0SWEV>)l39RhKcx}uJwS6f)k^i|W{V+Kmi zl+MucUf^G=#&}I>h0PU~&OaYgvzyIjER9W_H`XEE?W)Mu_sk|7#Y$IpHw+BFM#?h5 zBf(Tg72iNuiHeDpz$;xFZ?$uDvK4oZpM_a%*9#a2Gi+?_f16V3w^;0($hPXTb^YMe zwX)O;LB(if542s$dA*;?YgARS$Gzg!uz(3TO?Ez7N0VDX17UyFXKViDRaTkT!!EXK z)nwC>k+atullF0%KJQwV7aSh`z$w9OXS^GAs_jO!+Y`9sGn3@_4^^!EMb%!>a`WnW z!P@x){igk@Sic(SH?pRJ4K=n(6TGUJmO!?)xCenaxbByTK(6Up?G)?&x@U0$db-AD zO15*a=yEkl%x5tq6ZcipTda4Zk94)V3(!}(JU)G~UmXf=4rs?~wk zne2DkjbsttGkj9VD^NsS9Vj&+lN@0APXY5Ii;@O<8FJUF<@aDDwozCl*9dfK7qt2tKxZ)MxH8=g|fNN~U`G6kD`b!(tspuNBQ z3i0{+_0GqA1#>yv?KKP$DI!$Fw~GwRT94$HOeW&Pz{cOk?7qb~-9*)|WujM@ZgCXpFnlO$nqwzZN=k*2B1 z24Vl(!`T^vxkI;}3%C4VWvoY0;jq%p+q+%tWIZ zi|;p51_%5Mk5MV>>xZ|fM$DT*EewJU)6;%#$}H`;YjX%BCWKhWw4$fdjhI1rUDilC3mq_y-rU;v4wx9R*o?DreUP`i`89)h zWyvPM%53=Xrq9yL;Hk$3h7;KBRC$_RV=o6UC+OubI#QbO>)Y`)k?hGuEf}x%TQa6f z_NQomAOphtP1v-9)oRmfQm)LDgZYIGS4;rGT-=#COC*SO`17iyCl%MHF#h94ox)bN zTcKY!>sIc{>F6w%8m^B;^_4rgHhH)+X67kHG4XdLi&8t8DY-MGA<4?+Qg+i#;*hnP zMMp37cNS}nQLkR(O^mxHU!UxvZq8PndsY6ZS+UoRWi@mGl@&%_gW-3&Cj}mYQ@t{^ zFnxDdA!+M#%Pvg+)y>Wm%!GgS1lzO^EMlLanNb@I=6_v9(<)e9G6=Lj1Pmlzkn z*5-}zubzb*OcD0dC}Sq^Q+Hb4*ywQj(xWy?WDZ(EknZvs-)G}MZl-pe&vZU|4n=QT zzO_75%#Km_7P`ZbwlSDm+c#BU_p#$hzoEyf#w9J#cRFYFg-E#f+-X1a^L7kBDB9nU z;$$m{wDf|?{s)Ui=Vm7!-C+dM{7#s%W&g$!(dyGdR+QiF8W*cp$|HiQhSdFgpT#vqn#dAzkE+ zpszK!cR$}MDw02Ce0#lD$KiE{fHm6&kr=E_q-H-vD@KThMOCImmW6tq{VWDt=wZ8V zvswpA;jJ85)v-Q5p4WMrs5itkdQ$MKwdlwfkZANrE3-rYdHWQcR>y?@W2agivZ1$7 zFzq3}!?0Jp&sHO}7#Tk9HR}1elh50=l`PZXvB!%w_PRQd*}laWb=b;4LA--v3k9QC z5~~=xEw{^P)920Z?B7i(-N`wtcDlgtba z!0w`0=|sGZ1ga!DEbd&h#TN&qfU9%V5=W-t8fSHn{bFedZSQOWr1VUzEX+>rK$3fa5z7oBOVuLfbE= z3Jg;$=q_fz3$c0(tFUx^E|?ZM&gGF#$zO9XgrM88Px;z|qK8n~IV28Az^m%Fb{Rb(kvniLOmqglUX4OXW@zKL z`>$XxvOQ9L_uy1Bl93A=pHGX!+$i=A*Lnl;qz7gy3!{{NAkrti8Km6p4^%0en8gkr z*!0yrdNm*E?^d)OI7y9#oA07^R$S>Pi}31l)FAfg#r?zf_`qQdu`gbK9Mu<hMMN-HrbRQzoVWOIl)GlvO+|P3}ILFDKbS! zm^krh?lN6_(BS*ctbfb!>Fx@OX2ZU;emB=L$1z2#w!dmiz&oX+m~&`_PaZaD2+ zXOtebCD^1+8c_9K>~kab_9UJG``Q+WKLX)2R-)Fq^XBF6?lRy7e4wqtm0(wD>-U4D ztO&ZIP;X!5ApYfk#UFV+Pm}w$0s6RA84ku+kGwVVPLJfc!b~#OgVBVo zWU7szEG4@GCX#S-L)3FZe$ykYi9mm=gR9w5HWYS9N7d@^!e!jzMD_!^jT{cdFO@ zGItq3s3CD*w{4Fh%eAjX%QcVMmHL~vxZqN#uBs?sX|WG6p8LnPBIFL^hh`JSg&h{L zDCi=A{uA91<60 zT_@$2JZ=|#BE)CG3vhz%y#;x;5+Xyb{CDrXGh9{l^l2&d^`#$iX$+`#{4A7WYQHRq zN9w5is^u7VJmu_HVZv!}=anF*pWC-@rPwO3o`C)U(Zltb%do3i0gb*zJ)#2z{hzS3 zQ!cw**H5$Koo#K!j$&l}CxR3_&3d|Ja77+K!@FcM)z$J6nG!E3I}0v#?@rM}&O?9M z-MXXm*m|!~oV>S(7w;C>YWbsC0xv~;;r)>W$Zi{`%HMVZTjemp;@W6n6VMG`cdY@7 zO{2x<+H<{_UZU%#?nVxk$VmSPC7gA~lY>OCk9&83OH#rpnp3g@Y7bDWtH*=SjH(^U z(mW73o=LTZHGF|rA^G5gLgTD$EKEvHg%5mZMyDb?y(34grmh|>X|tpCsXz-$VJtJB zS^nbkIGAQ=FQc@@hrY6Wxx#ix!q^nD2OzGI{VoC6WogKXE6H|`5MaD`_Qt?ec!-yq zh_`75+D#+WbDD9ZFSAA-4jEVoo~kjJQi{FTl}@|Z_L@wy0T9vVjUrzp$E08cp3)^S8`=$H1l6t5v9MnxK^Y`n9H6TQ@m{ z+-h)ZuxDad5q$3S&P!1<2PPl2LSaUGcUfjgm5~{i6i+psoacnulJSDs`6fY)tKcpQ z&&Ji`rVQt%g0>g*z4uqmdN6wjAqj1UG-aI{Z+J7R!ezN`nUi@Fr0#e?Bv<=s8MC*H zAyKlvQF9kL^0yr8u6PQP<9Nl3O?8ascV1^XjQJLM?JR=iw=)zHJQ&mMI^Iadp+0Z; zi+W?b+~EU66hwob!i~I^u0g9(o?ls=ca%nfd;M~CPj68xKWIvuSn%uIiT5SzHe0{t zEvJ9Dbq#X8`xlreQ^!G7Qc>Q!`-v<4Ie_gZs~0~Mfpz)54&4(P#<;XQ==iq0_C_#L&Uvp6 zH?#ks0~F}jw@?>¬4AWsv{&r0m|WOE1LP^QZ^sfPzomMIk@JOptIe7-Q6i$*v*m{8w*Y3uv)1+-mZro4g zpP$|?5Ew$O2tqu`4gb*#pxrvJVy~dQ0zXw`23k^@rs8V}kJQW86}W@s9P&Y@FApSB z+?@9erCtReeNJq4aG5i1i8F$bBd9IZ|3HtN^~RHvd$a6wi~`uFr3IJOh^i1|0N1yb zEd%P21<882f=fxz@*%%oSwdILVqWuDvES7w{w4|dO4o*TD3K_(jf7b%E$&WJsk|?C z^@~PIvzOb!QRuTv^sWAuiFWMiA1%8*g0E4N9jASB8?+rEE_3dk)(=Zcl*i{Ud@Thw zLsja%LN5D#;u}&L>dr!4Fj3UQM7j6Uk^9e*!Wyu>LU7iz9ylc;Im@BN0SO71gEWK! zj9;@Y(U2r2S8sA4G1rxz-@$NJoEFauoP|{{It3~oG2uOz);8cYQEX=hNY2{j(|`qF zyXloW1UA8B;2PE_na6y$NkMn_5;h`16&Nvi)9qc2Qo$J|BQ|!7TO)Q|sGF)9a;whC z7WvWp`8EP;J~)Tox{-y4(2^7Dl(htKdPPvW)WV*uj3`*@80##AW62vmv&gJ8E#V{{ zH&bax74;iPH)-kyJys-0Ej@7?55vK^2dky7Uz7x!3?M>vzY7KyoQ316D;`(Cq~@aU z_!1XZ+6_GuP#*--eisPp6aF?uLr6SDfk;+D|A}ys>0VSEzl(o#@@=ED;=K6tvv<{> zGa2nc8u{Y*$Na)-`HE--z?uFx9+%Tz(+9p|ydkHnx?`KyUcpV%y!vfF(d@H@M}j@7 zyscgW(*BZ*Hg?#f`@9Xl?fX>CGLniC48ZHPU@-qbuy( zYWcESf}d;f)5aES)OoFPaKaR8G*D0wee5H7^_GM<$=VZMMbfyGG2gH<#u#vK#M*w- zm){oU_GBaslUO!0@%ntCF-NnR|FuEgB-Wf94=lf@Pl<8uC&I4Q&-7_rrIdl9=4qF@ z$>T3|Q^r3BnoWOlVP{X+f$#JCU~t^pSqH~JPkOEw;O?t#sv1U$jec=_I+B_7n(i2+ z%B%C-G3H=tA=z&hC5M%cP-CzHfkf@?UTI!5xMo|Lj&z1>rGh?+Xy0aL1gqi1 z?DW_O?uw#{5dtCJ7_A@8=DRj)cqyaGb1V_`tR2Aw#5u+}hE*#yy``$LUnq2n@*rdf z`*+{^OC;g;<7#9X)y+?xcumDyQFCNhlp*zsM+zWJ~Msk5hi>_%_bmOqyA@n1iMW!l&tt5XzlL>eD`dmJylc{gwo znUj^m3|{duk=yq^AYUAK8zH^1aLXLX=WJ5YN_H#K76c(HoF$l(t}h1Et|G9X2rbo4 z{AGr`p@b70PPb~LvZ&3a6xU@vfsw)lT;*Jb;YRmeKCHZu93#Z$#!hZnlSyw3Y;CTP z38>)eJZ95KopMJ~y?3WfV4fXw4kd9{8u_6+OAWfKfe&U=tw3#`>C++eXgyzaK-tA5 zbd5#1YDZjtKYOGa=YBn_Pm=C`(Up<7z%hnucDw@FS?ghDUhlPc@_;bwnk7?uex% zWw7M2*9lizTWQ8xyE894#Ww~8l}1PyY+zMucauvt<&UMVWC2hDTR0kf4EIcX10x=8 z>&|TycyLY6J?vwYgOqz>?bf_n$bm-=aDC3yzT@oSR`M`L0&JX# z!3yyZIH$0H5e2tNS=(FDnN3j_@wNQbN#4XZJ3nWDUsNyM?lbaT95lf>R?@RQ)$W)q zs+Q_^9q+Wsb^C%?cHP;b2ImU3BN`hp4sTo1{gII=l-15j-5^BO)-@a^yoXATzkJ59 zx=MxEtFi$Rwo|%CQT=?`{9m#Wp0Jj%l2;jK^kzkJ=sce zf`{Sp&dQiT{n{rl+4qG$LVjNkSl(_W@MHN7xTn*R>Kz9H@Jr+B-QZ;njE_O2eEkWU zrmwbk+7rW-tPv&dxGkua!VIPYw4u^|p;lfTu_KJG8l@(fZc!y9(_L!jd->Uld~A;^dP?bUPJh=-Ed@u=3pEHC_?x)TC;GCG2J%3oU_5+J)bj^wR6*tY`PZ=bhh!kfVF#Z{N-x+gVl`?pe@ z=#0l`*(f2$cSJXE`F@GE1NI0RZYyd3wT3%2zy#}_McmaoUfHrUwd3B{WB)k$*_dC- zZ1^rNH_>OXGBluHW!+o&^mNxG6GpZk)XjpGdIi>Ug(w`nGdL9HakXWb?3coXSv4g^Q#b!j&#OjHwy{9+dGFYc?D{O!$5q@+6zg~wk<@HHLesUX+ z?w#?l9cl6Z&7-^;-Aj%DWD$zSKUg?$+wlupeq3;OZabRPwJIgGxBI@Ho-ghJVo-|+ z6!FXjeZ7oj;(`c5H}Ulm=@2F6u`im%*AWe8gn5~`y|%e+lm|Y{_qs6y7<2iXq(Ask zQfeA4c4&E!goOI*m-DdS3DugZy)6ksd110keUQFge zx?lf*XYLrxne4HTb!sF5IS6vS77c+~d4SiCU~}Qm(ZeguLb-6B2Zi^~dp!t1^OQ!X z%hbKtFY8C5%V*xi8ViaWQ6K#Q%tY(sZ&(nHRQQAq$Zh%Gs1`{aaa!dOoA@gA`AZ#! zi2MAZRI}wlk4nFh+igK{pz*<=UhW)l1zX$UvknI}oQ0of(px_-$0H1KBdF;$wX$8K z$8%*N0$Ua_TJVqmxVJ(eX!7Vv=ii!v)+sF zJ^*fi>}#L22iGJ5{@tSy%t$H5LsSh`viId(Sti;!86Gehs zW@ZsGEJe0}H4VicV<;B~p(#A)!0O9DWZ|mqr*0Q+K(zN@2KPNYH)mtG+TOixUcuav zY>pa2p{vez?S4D#k5XOM6)MPwwUv*$7v9*|Pa9RA1rlBQmo7kk8R6hT6$Q*%jqL@zz36{&!l zfskt=HGLGQauBpy5NWhm`vLImcmN;{20Swf<7lABBlI$&F4Tyv7>Be##lCGeQ-hNB z=c?0=(RbHQx*AmZKO1!!iOBrm?ARl34Uey*oeEMK5a270Rozhbcc)7)}}nctqh!+2Er5?_k@eL8fc##r@gJV z8DhJ_tPQ)~h9ep%R#WnseS1U=L5)9TF3!rfF;+WBUQ{|)*Cl*{>Jm~6qUR=iMPNJ!`6I9-*%CZ zj_@G>4rh;uiPhrU>F&(PxjW+_4p8G?IeOe~H5~3|M@Ufhkg<(LzwRFJJ}4aB`bO&P z8NfVTnu9aSr!b{&#L<*3Z4!w`v2G;`%R&`ymjQ8*gwn=+-4g2|)#uV1&YuL+dK&2A z@9$%~Ain+&U(J0d<=D(AN?c6#zLR1GPW6Dv0HuUt7GJjjhRbeRuQ&>9-AYc z2hUU1&UPj1Y-XS12EQ?A1ZTl*8-N<MMh^`#BSI8kw`F+}SxXuA<~J1A=!ZO7VM3 zl8GNs#jH+TUNsxxe=yKNP4}n)0pHpN7|MOev)xg>p-%v(O7dGj$y)H73078&^{ zY>94m#N_mSv3Tf@urLp20WK`QlAP}{R=4B2+yzALzyX(^jSDwN(SC7eGlhcPjp6@?G0R>aA2P-pAFyJ^GcCTZ( zRN6^EoN;5`C=>6~#h#4dh~8U6ECI7ZvHKIuG}97|TgZ6RhnKOrdSM$`Kztl4ogU!b zf=HW^a@}o2i{_v3ytzxyEEB6%tkIwAlrL9qHM~&GrXC1t_QE7PlXlAXHI|}kB6UjOC*G}%mm)y7G5&1W*2`T5DW1f-wALcOLQ8o@Tp0y`Y>o&F7*rz6O}odb@Mb`_B8pp{I6u&2m0SQ z~bL-3Ts8OD_=1%fY?ogOys?P^GOiezyTn=Ne|QrzZ=FVyEbWa$wcrqBM3 zEyjfAK^V^uLqSnSw0(6YE5jZ9rMj?tt%Q9emcvf72bC339a4uIEd@Qb^i|~@-ID0O zo^iO}>NFt7+tB90(&Ze{ZNQ^8Dmh9*hw2_S2xfjAh?1-9P}O@!eSpPyP0kw;Z^;Q$QwOhs**8zr{+u34d!V)2j&v7}7)c^Ik8m*oa?!Kz^c z&Nm-4m(icYyBnN8uVj&ilLIMNL;Qh26lAsD8-b~h*YNc!FIJNCGN)KcDZBs1tAg;b zp7lL9qb=>V3>#vmb?mn<`ab6zvdk^)W$a{ z@Po#C$@OrJ$WbnA2fH6dMz3TaRC`KjlztLt`ql+Nz8UD-ixeB++%F?$bLnQ-ixnCN zn4z4Gv6UA}jjIUdXraG z_bs23N}qEYY|3RYlyDP@TyVYZL-3gkRQGAUGc2stb!n6ZXu6;R^%g>F3!clXmOG+Q z-^w&o4?VU(u)*O>>YHk2>AjB$=8`3!EFZd}NR5d=#oc4cw_C>eRNdxVFO><3!wPrE z=&k_)jpf@^w^#!8+Y|VhWm5j7^MUv){0%F@j!tsK@WkEuxMa`eLR@J9;DWxTJx9^y zKrNE8@Rl4%R*D6XamO6Ja5Ru&rBPYXh;&Gxl1;Mfutrr7>4C-!cUx9tb)&$#|y9z2uoJzfnt$sJpVZyDbT}*&>_Ed3(!U*op|0HnCL) z-rU$`khH~I)NSxLkUARMBXc0KNb~k=eP}asASh*a{HE!V)S3)}>&lgvR&j&fcrv?o z^RoacG2(hcLPHfr?C%bhu;S2zQ{C+tp5tiXMwYOB1)VeWvYL`6y^5Rktpbflb;h|# zt3V>RyrgNbVY+|#NV6jLMp{GAI9T85R#C`aK#gUG?!L#OpwuxCPn}Dvo~Xd=0m?9} z+6mETNcDi^d(HctWCPVM?}A-tD2D2$r95M}R5F{7w`9KgUO zjj&h8#A@-|#V-bUwbDX-Rd^J%T#>n7b9 zy&LMkznMJJFD6=@Ab0O1%|^HxN;zH>v0-r&=@BX|bpN@u)cwprn!YV1Ai3_Ao!d|- zy32%B&^6U{f};PsV|f;{G3-bf_#=+4ERPp~u9@BBu^1{Di;?m3%r`5ITo{8lCP|rr z4G+V{F3VK!TP#UU9qA8&Cb>e~fYX3sv9wR|plx#Wdya@x!%41I9Dji8E4{+IAD#$K z57Mt3dSkh??^{`UcgqONwxmI=4P>bLj)X-Z!bcSOdVAZn528OvoF+Y~!d3V!&G<>f zOKVRg_4?6Wn&n-KgJyLpw?0t-&F+dm_J!T>Iydw-z|LoBSgFEgddAjfSKchQ%n$uMr%1Zaa z1c@dR#rFx`_MdS@%3Jek1wl&_gKj4Ap0#TntF=oUke%VhG9LlRUj-(b*y4{wjQMqR z3N(a&JCd7@xKS0=WRd(*Ux`51cV~rDY|J?g=t?GI&Oq~$Yw{qBNY1~}y>{sg3=i6P zU?}OKN%=hWy6012Tm5bXtH``VZQk*%a{;_V@{@j8syN_sLe%3LJ%vDFrp-^ynJo^O zNwEAj!fl2=Jus5of8Df~Rd;L>SJu-N26l%~jDw*If{q7RAEDJ}*o%{>z$Uid5riU@ zOr6=4y}d@A?T;sLhi?6-h86AzFu2)9iIJ{*%wl1tw=+XV=w}vT9xV^)Cs*KiKfqZ1 zx8~fj&0u$_UCe}lh?djk;fdu)yo>&07^Gh_9_6@)JJ8u=TE}&o6e#KS0;Bn8WLOF@ zL#H`yS+}XSKzB;b^H5@s%dc-;B>}Xd)LYeO-~3A`!*D1R)XiylysoGC)lCF5KZi&& zVqx44s&E?Vlai8V0sPMT>uIlnw53>LRPTW{w&_>l*|t!JOeT{B+~MFzJ_Pj+ z{K}(V$T$ehotFY;=9U4O&4%MS)(^>5}(z^$4EtUd+J@BZFSi825 zn>%@CIL-V+qfn$=!~4T+oSx^kaUku+BiCc?ZNQmI+rj7*r+Q6<40=VsFpSK3{_^r0 z>CW4yf(e20T2xCI9R~+%R}Kk2!}$}!koU;vrO5=YmDU6WR6%+IB3qLJP1L5XZ}hw z3Uy1%mPeBE5Az)~=Qq*aiS61}|LU=x{pRplm2rg*XTnxl)oHdKVjZo1rT^{96VrOj zF)xBH)3*k^eT<`fi39C)JchRc?j&y5d}0en#9?POKMVM5`Br|Meo!aRGc26J`qvNveG&Sr5Su3^BH~e(>_}>4 z({--J2{AB*g^IN^hnlcP@#|MR!cX!}Q(j(0mKkHt>;hSg>caP3MGn#zt;;X_3DMVb zIq$%1_VmjwNV(1UB?u}~&@4JeEqq>pv>)+R^B$A|&g`>^mgK9GXftp;5S6AIqZcmJ z-U>s+7Cfk7lN*k(ImWkvQx)nKa{xCKy?oaRZ~QLk?p@wRZCC5DWme`Um9`OG6)Q3& zO49jB&MiSS>OJLR*C(~0Yu$~!uK}JrHiVc`HLc%Hr60;TvQl0`xrds{eJt~32@71! z7|Uzc@Nq4qP9RtGy;Ck{^C3HT^f=ZS$_CMeSv>~_x?tAwj7JlZW8Z_OOANg)<~{@4DHd(gVM5S#q^}Q4O#mN8h*Z<=I^7AA>)j;!&KM|upeevJFIH!B!saQB` zeyyga_#brOzpW>m8xW(Klz!M06wUBS+6sZR)aFyT*`BSDdGRk5;J+^}J~to`WA=pN z#G5iVv;Mm#{7`4!4_sK_pw5fYR6i zaiTv_l7HX-|J=}zza^uSVnu(gI4AN%Cz$^4D$e;`aQ<`^|6fo?2AEc=mjB-<{6A?F zfTLpSo$@c$0Xtg|swRpbwMyTU>^`?Zj`$3<)XLlE{rP?y|5W?nTXg!Yl2-}E^RnJC zq>?&o62JWXKdRt=T49SPCpvYO)Oi=kdHn^mCa7>i#J>P6lYA!0U)GrY=d({NL=xMH zMBfGs3y^{M&$s*8Pr>Ryp)HE|GVl)3SN|p1@aNV-p7Z38A$SeEgYqwxrqZ0`x&w*0p*q7&qS`hz<+ z%iT8)D~A0)B~nlNe`@o&`M0+=`0;(VyLbqVWK&wh>PPH&h?a$v*MFn2(eFEWG713^ zIZ3fm)j zWRd8PwhPsh*LDAQ17+Z+^5_*Zo+EAR`@BVPM>EJR!<6!3$RB6@Q|}r5>^laXJ_m^7 z%gaCYC;z(t7Ek1Xxr|g{XbbKYzaqbIosCA6Z>s>PL*L@f{~zG$@0;~I+5mFkD^B2# zJ$}k9D`1Rby>QE4nbe~b%urur@qdwhu_$QjR!eLXBRjABtlTok1?~%!fR?BiL)!oJ z(EhJqkQ97}*9_n74u=}a%1E_Y-3|=psJ3QYJG+wyiD8m(&8~HrIIhrV33PeK=jI!+ zH& z*TL7=u8_fY*uzC~RdscLcxUCx1mA(m5(J$5e<*v)pg7wtYxoHv0YdNq0RjYfcMrkc z-GaL|?oO}}JkUsR*T$XT9^9eP#@+S3^PHLYn{&>*_09ATD5R>Z`rg;xYp=ETwWA0! zKKyu`Ll)V}Q_4}%=^3?m7AN$0iVe%Nh~1c^cPm70UUr5IXYes)r`CtIeni=y3jfjf zzqn5DPx9AN`&-4O|2uofQNU-vOao6t+s>!)3F^%Qd7YF9-jmr zMmB-(Q)zVO^34L^=Ua&<-B$6dqeb%q4&Xw}5;WM)7u$Ed?1i8vpK`#WEnO5xCH3uK z9t}3B%=N!3PJgbDneUSk{yVXoAO-CF1^eSRJ-A+pjaQ(>0Nt}wYX$`^f5!&$i0r*& z#D`~tr0v{X@?DocJyy`@tSg%ko{&0i5)+J*XOdXK(jmVLq5&(}JR-2~zql(x|L;!R z|MPbNKau|)paw45M8FiQ4(`Zk^~zhAhl$z@x#;l@>G|A?4o`+FQj?D?)FcCgrP%s7 zbR8b^$L&M=#(hJ9#JGcd?ay`dU-xIq83cG;zYi9uB3A5+v-jVPAjO;94KtbSMqoQh zH##}lZLsAjEwklLUJkry?~=vYij?Iv<+4CO3q~vG-?Yr^kJ}ouT^vlLVMy1mO+aN} za_c!TZq8`kY!(xedYn{v|=3+yYv*oOM1kNoJ^L<~B&`=Iji^XhXwF}+t&3Ovz_5HEwd6J*Wgmp!K z#gHH5Z~)8|T%3P?D3{ z;!Jku@ygnH#diOW)N#K(oyOo@G&ol||^C(pQVrdZs*aVY+1Cm3bz^GG7f^vK!e#(kZcQ5>4iCWrZVP)QCQR-u!UBV&Sz* z5l1G0sxc*Kw@}^F5<@u%Gw@-cd%6(;ZV9wsoT`9vC?D4>V3?--|2Hx~P7LUxa`@f- z{(#dBVYLnvW3$|*QwR3*x;{w$?lH*k0g~$ce7o83>s4-tqX*zp&<(2~nTqm93$WRa zP8sPIGoAHA5Q*FPbk-1~Mgst^6#73nPVqnUkf^!COf;=>L*+2|&M@My38uCu@$$CY znXiX{cPK-^E#so;QGUMJ-awQmy+59>zU89Z*vYl)^_nT4+hH7|&;0;eByJe2F?Qn( zc7pAq+nf~2O>3c+3SYVKynubP$hJ4%Dt5H{!9rX5+lI$>*WNb{GyaRo5;0Z#KKG~Z zXw-8hMG03r>{0Ay{xCO%*LTgwvB2z~M8Ygmneo91Zl3b(nQIEy*M-rW>zIrm*8b-n zhkS#_eowo4?}#$*$4z*uG+8_xyCprWZid!wS1?fWXJB-C)w$_}`P zf*f**s{^OsY2Ph7n%oNYYa?t{t3;l3&kg=+*72R>5ZK(-7g)}1v4x8{#=EeH;H~ku zTP=iTA|qgr-Ix*6yYaR;?Lt_*$T=v7s(|{a?+)=B^V) z?~E_4u->j}%#<4t>rJ)AWP06@So+$TeDsS5{M!)o-&VQ*8VCU*m~+N6PhOLpT6Gsp zSijRYi|AOzJA>mxleb3WRcO^2Xs9w&MV+>zWLyAk0eG(UzRIgh3|D>@Nzy9Z~c z6o^3Qa1eF<&wpk5CQ4zgyqPNb{13DPdjHDxc%Ez;oMD~0c7Z>=b)OmcGCGs+=BKXf z+AEB)rTful(>q)UdFDWFiEwRd(L8WxpxuUeV`Wz&rzB@HQug6OTiL)bU5&O3p;a>r zLHCoN1I0R@@?-pl%ur^4$FeSkCvZx5&yRnzG-1N~WWLDh;WnsxyE|*pHFk{R;Cqt- zlMUT*bk%wp@1Y5~wfcF{a4lFlpXk!r(R}p1WTB?c)aVtrS5VC45Y$^DbHxd;uv7!_ z)(DM|ziNDgOG|F6qf2{fHH(P?_3=6jbL!CbDAf8P1m<3Oz2ccE0zm-=w6%-kM&i83 z#63VhR53OQf*;)`Q?pHddF*fstetm{Xyxe9Ge%Z92E8${G34-xiNy(FGib;{>V=0jwRNt1Th)hG!ISt z)x5P(}y!u+7 z11x@fGsIiowqgt{ZlO5L+$ogk>+#lG8$$%q2ya6V<{Bqo>y7z76Y>p5p3rHwK$wFt7N2``z>%v`JO6A9R=Tew&_cHVNne@ZqTl1;6zUqTNP26MkH`;D8<-0 zk>v|}|MPU8@??yS)g%-!=YLe7uuyQ$Iex7L4ZiF`o}LZ?`u$`a1;&ffpiD1eeY|+@ zs$U&^_Q@p+bx0M@GtI|R&l1G-lYi>sQ~d=b0D{8UI{f;zHFNE2wPq3;8Y#xc;u3m; z#)(m<7o$4$_Q{^O23A4AOkBbhWBlwLWu#3E8W(#Ek$!qsFAHJbH*I=2_WlsF_eVan z%WEIktGe{xq2&5*&cn89>w8m6UV#RuSaNcWZk^U(PbcvTTlr7di@TlIe{S2>XI%Nq z+_s>O%kww=AT|MK9i(D^LhER6kxDn$^qPp%LgQt^s;_CoRG~IUi)^M$rb^gaj1Vm@ zrvM*mN277ckm+v$x3Ho8{tc0^ke6wh9C*=8Yylm-7ImAcBct&zw`-^UO;OlqU{i+c zVJe%W0Wou@md1UOTPn13oG2c{V zTz=cm*t!Lo_1n!q;uDCLb)G8--T3Qt|ICb_m_ao2;~~%F2c}f@@}A zjX5>xSEKh_Kj#H(o!I@stRCrd-k*Re+|7&7L~_5*mS=`9Zv%4Zf6k|!9|$=l=yXn! zeZPDK^fJ1Q+Ir_aIHVms`|Zbbzx>|-lYtx!>TD#|NxFoBnHO~lbb1TlgwD@O&X%kz z79CNXTMs(}FbSbcH7|11w8(Cqt zvHt;uPjTuy8}OM+f{-12?f2LlT|PN|yeP1biB}m+jMjwN#xLI-NxWY4D1QVDYu9_; zi6hcwbY;vhstkPY%TU^8EJv)$gljzEmP>8DZb6s$1n)~<{2K2{q5Ml1!r%N!8wmSq z9*8Z@+K!7dvcr@0A+O~GE6+=p_KaCcp!o>bR11pR*LMydNh-xF&1`7*6ZJm_j|q{< z?ajt$_7QkL3>~C!5~ATKf7=rJUuT+@H* zn)1z>x$1R;b&@^dwlBMuJiYp`S;NuFF%I8x9u*N5ij!;5>v#=o&u=Jcv-I<+42$>p z?N1jfoL^TGXZrx8^eE(I{(?xXI%J#}b7`5RbNTO=mWiX=pO?-OA{J$bOtwYqYA z#(*p<1{|85PqSr7`JUh;TU6SUj3eqA6j9!7yAMo!tvQ|(l;_GIvhQ=_xyEVS_cg0- z|B@9O?(-)^#1|wV;e-9r(fU#%*Xn&5F(@(!AWvKKJKloh{=_xLUkGPba% zTkfbHYVK8D*Ew1EnLtoVwivTz|HnzY-o7s@H&xW%=?WHlS=nWJ(nf(r%P5K8Z59Rq z8(042S77WR1=?b=XTq#uEEQY-*V59`?@Dowi^mcy%8?ZFCdQ}b_F3{wf@_x1S<7V{ z2^zVD`d_nzLa56B!>9Lu|1H2D<4fc1S!=TU*-e@G7dCUn0u1~=ImD%j zL~cGH^J6+8w$lX|NIM*@whP_*JO|{rzq7|k^D7DyLET2cym{C{p1`5c_S~7t=g^+1 zh@|v$=@bf!`RLu?z+LM@H$p64awMRZ&hmJlty~nh0&kA1p^-bNp*Ve8A_l6lT~Zt& zc$PJ&MC6UVooEyI!EfItZRzUZ&~e#?FdVG25`&ekmAjq&%G{W?dI8!+oc^>Ah7M1- zB1%Y1D+fEQ#nM-?z1bQFDaB*`p$+TaoVrC$N%Bb@_)f_&Gm=jDv2&i9&RS`)T0vgS ztdm=>IAM<2&+a?Y=$Jv<=ai<))4dUbArV#`!4^lt_3nr*^G>bEId46Ovs(K+rwudi z>*Mw|HI>CfjQm3@Nm``;#mNGm%}o%}lE)>6n_b(*ur((xc>dL-6R5mp0E0u@y(D*~ z^)adop*;58$fZEA8Pj-5(zJf(_m7UM-7@b-tO%-=5g*>drEPv#*jgOh(Y|{J*Dmk^ zBkj=rtC&OG%Q)3r21HG)G%Y)`ffT?()>`S<$M}KQ@fSIN*jv0`z(?f5hQsjfnpiMf zyRrKH_@cF_BH#v6GgX;Sy(l}j{`H~nPJ5y9*hi0sQ-u+f?Gz}9gp6*V+ zzgFV5CmJSVqwR->`R4R~8Qgx>Z%mD^pa5 zs~^R$?&2BDjYMNGgqgI{+kDQEB+!=G7q?@f!zC@bF!q!U5?%MjZ(p@uR)2&K^!MMW zKSmTAdqo2zUUR8+@oH>|&8rsoOX=_uCWCf$`_JH+qL7Tk z#iBNDtrV;zV9nkudFkzm!VLrI>S!VkT2wpDNJSEn;>(A=>1h-XI#sy(L?Lt3;Y?;! zsmUX8yT$L^}}T2M=qqUEJs&P?1cer>UdYx0qpKDxI&!KHcb63Lm6*B$&*% zClWAGE8XI{mU7xwvaK!D!1b%QE!$pRTKeATPD7!JJCKYE5sy*bzs#iLvMYXPs(&=o zgssh4h-(rEk(V7!V`NYt%P2yVlDm(3A3|Y0TjcTlX0&(T$i0H7P`@orix|sntn+lz z9D_$$RF>k@=WOZrJ)beh9`7s}*w4)bj#D;d)z#ID zr8N-dSk3X#VJ}h98Bs{#A~wzk50!XnR4CFk>Ix%|hMaA_rlJAOH<>dnTpg5cmn1Q~ zOi~GocOJyFkICw)a@wG*wfzMRpw=iY)HbMe;^23GcyIVQtaqUi@|{!)v+_$T==p{7 z!JV7$`$q!gj?%%xer_T1Tdmh8S$of`*0XC+s6Unt7m4&{ ze?Stn#beg4=6G2nUF4oJTB$-CJMrN_zOVVq7VTRiE@4}1)g0cthiR5ifoxnRveHD; z^<2dXG-V6VH@oqeU0P1t?1!x->ctoPj}kF718H;&DXd71)n-bUmzl9D1%`KlVpgX! zRb3}d8yNli#iem$Jh2`V--SJKKm3ge+$9gsdt<}`ba}>q`bd1>_l3hC3+{`RBMM(? z>1yw(@7^7wu` zj_p|9dZ#@~3yajO^sCvS9GFv%hs1WFMrj|HQTOZ3=?0lf?+0#=gKy?~Oj#)u=8OyW zU-7u$Fu5hpOR2C(PR^0wsm6t(hUzo@K03jFhH{a941^G(ZD*2m6BqJx|BeGQj621F zW^7;buMTV4U9PgWyCB{7jQH^l>8~Ja$~bIGr>zXrjM5W;47HN-wDfAAHUJj0mE&{k zSrG|SH+%{6rdEQa@+(GrsN&;xE}e z1TxLW^8K~+#I_6l2YYR9;(@gEm-O82&nFl_8t9pXjWt>t#fK_aiL6~cdbvNHC2E@| zOaw_g=-1ow@m>=b-o+5IQKs#5h3<24T{buc*8c2urW@1?W1(>2o!M0UsVCUVtnhjpqjgL=Pd zoIg+*KDd8<&&bcH*KDS3&VT|u9JBw6yZ)zmfEdYPX?_3{BlQJUoDp4XSe3hbIOjLM zfDGLmY#Ahp$WPra`!j0;@l+PeO|=Vv3s8)r{an9Pr+#qVe(nLgS+M)@dAQqPsYdy~v=iP`-XV|iBO23T;*sM}Du zXF8M=;?8^5G!|3na5~Yh*5$Rrb%Swwg_+VrL^|FeEn#Q9aT(-78xhb&BNU{{Adnvy?;D7I%Lf8q2q~Krfm!n2lT`-Ext!**zaZfdM4UT zi!^E@7W$l8>5>0ycs#s^Q!r@h7~yjp#&<(J@(1>?k?J4JL9ZbJdhRtb%_)0unpxmO^K1IxY&z6!gnq zea%?33^E&in9RlUdGJj+vnTlAjTIRv!;QYJ!Nj$5zRz~|VW{2l#9ZAvM8u6zzco}w zqk@!WdMJU>_NQN+&(#@)d^%r+FRVY6q?MP*y3a_rPnapWTvsLQ6HPot)jp)i?{3`3 z_x28(j?*UFSjOK5j}4yp<#J3JP0Y5(*%Hr*+uJ0v6#VP8)f2|auN5Cs-wD}$Fv&$H zL6tpIxQ&h4;(PVXPg)%V^r%AMjvzM!$LAwH-nR^FUhhYUj39MjOX+89^3M6rfLfD^ zUnE`+?KoCigIgX+KduiNq&-GdS3$TN{1v&hs(~c>*^*x|r~)-|0+_k1F5KyL8{kza zV?>@FJYHt?{Itz>h*fq!U8kT`rHpU- zv)i#0kzdrDqmfs9pV!r1m>jq++j%$cJjIx)6%5_8!s zTN-nV?Xms|!a49sXRF43{))YPrDsHsdOX)oymBAE%phB6ekRH)ChiF&a3_$4?&-2#+*nUYR3F2{H zhnLU1ZXhVrT3+ydf(ymUcRS$o_mTKUZAr?uEY)kkEDj*6spB;)D#yKtSG7xRmd2BM zA^C4I*s#n-lt2%rYDAuIOB46&?Tm*q1RCVrP;i*O6N{Z~UAhim6A57|U$`=n>a#wtn)`GrrMvh;tgXz=$fl#!T;9I#13-iv3DkWQ{Yf zan-}E8YWxhfG+-QaRB04S6BT;D!*Hs^vX6<@|#UozoT$yPE&ut29o6R zKw+rZ(!`z>Oj>LwlYtAr(^tjw;r)UCg(rC$Lpp(l=y!>03Lrn} zDkZFp)mbi~TdL$%YWG#pRV+8<+5Ora(UeR#TCNE9N|l=mA!kHS78{n`_l*yUP)EC(n;?CSOlT%hWxdfr64p}SnSz8 zOs=nxHP$ub8xPQsO?m=^{Jx~y!0aGSJ(FIkug1ip??pnFe-jFRl#bhR(P}^A;i7Tv zP^3nAwqwEQUcInhKwcwbnf6b$pgYgRipif}Y!3Ne8KBA1{*8%zNdt~AJU!1Q*zSBX z(SCj8rpruu+WE|gz+^hM6^w`>vR^AGFXZRcGcHmuuw;f7uQh!gm(~go`_lQ8^Uj>f zt+OxmaOtKQaWtzxZ?4&#j&#CpvhGbY=hmQFXX`zhY9cWMH>6{j6x8n9s#q&62I*>& zc`mQesmH*kOA3NO;^*t0>qD3g+DW*p5=r5MKXo7hh#Nr!XwUpNwL-SGZP*sqXNC_O zc#NCt34O5AJidk<3EV*s=FhEGU!&}0%fL`q7Ff#BJLe@uwnP4nOKZn@!N7e^&3H#y z$HYGI`uY6;m#|@=t!EU73I3>H-Yn{c+NGIyQEaqQAcDblv+XL-nUoUPd6A7jUC(sXdvVLgk%G@YL8omGaLF~NTkIt|`1VSk(!nL>3=tn$> zL*Lz=fWA}#D);G%P>)##cRK`oLK9QpFCr{ac&ga&J;0cqD~NM z_;jq~%$=+g7xMNDy!q2gmS7bUHZ8d}myQ41cfp{$b+OWw8nL{Hdi#m@gkUa?jHHp< ze}oShft=v)64(A$z}N-R)_vwHk9!?e8x{+VblIX2!D?&;WB`tQ z!8_KDfYwti2d*-{jr&O4+O-f9OUjL~vzCzl&q3B4t_M;M~r2x*+tp^W;7l=$scS=kR<=M&UsW zG++Q^sz013(78TPBm0heFcU)5e0OuR3sLKAR4?I#Dk*6UrL5a^(`*z=!!o9I$_4Jk<*LMWGLFaKIiGv3zn>J~)=~UAAco;ds37VD2IeIF@-=KTW?$c)QQ`MC{P$cfhSe1A=ho@}Q{tCjuU zsFyvt)#W|}8TV7ECX3_Uh1R8?2|f=$U9@WYAR1RVHlwmM&6;<-R*i*Hm17qzI48>O zu7m)PU!?<< z<=+YvG7VRsf*Ry=daid?)C8V}k}GM={h^c* zx&QD}U#Njd|6V9PfkP3S>mj9L`X|~saWdgyZ}C8deslXXMd2VA6LcbP$l@c?sJJ&P z6a2p-&zp8KD8MmN<@#O7r}N`+GHHLg{6IX3n@ukQ;#J2i5T#cC*dCYdV)_(tbu$ln z&pEy!uJa^3M`H*_9z}pl@dkTpdJ3(o`HHxs7F>UF)a7-VPrp;;N^>&43Uh##+acryY}N} z68mp9lIi?Bfe2`JuhsoVrNFEdPG+OujJ?>vGMm8zYPLg$8BCRZaD`py#K^d6^``FSCqRevf!F$_`IghXkxR3;_uP zaYmP=cq<1E;Mz+qiLv*a1Aoh{o$L1F66+#>3Ya`$F$ULN6q4 zo%C0cKwMnAp{de|RzJlri+7e6_bKz}05~Hvo6+F6YD|vmYv0Q?b>ci#m_@B$`gNDu zT`X#Q`N9XmOmcoYVe+n{G$YZ{0m+bkP0#oN83cI6TtQ( zOW(k3bzY0(RGp2gyAki%E17_88c|6MiyO>7u_S>IuHfp=|AI(fkev95eR}4$q3hJj ze2doQb#Z}bKQQ!Q%z^fz<#~riPfd>J)Z^3aD7B2pl~YQU1&g?BM)7vEA9s({Q3H$B zbe{F>?{ahEC9fMrhJB_4!X4zTaFN$V@YJnL&*-|9(i{}bF$a?}6`XRA1nOvuSTbW+ zL|(p2FEN>iX%&9$R7~uj=7OruHi?*$A*;Xozo2y>DcuAiRwfKgND;;1+=lj3s8Et< zs2>+O0FCyiwl|4Dt0OKNgDED}oD_KehtGqES)WX48udkmZ%yX^h>ptAiz2+m(!&!v*UTvj|jO8l5j`_i8L-&=P11k%uPf|6TsWqsweSP;vWLj9>fHzNu$b4-(HhSRGntj5-xBZmRZ| zi-L-lm23CchGv=O>RoFs(;f4zCbzrSP#F)FO|=g*eJ@7QGWmrEv|VT#d3cf(G9%h9 z8sPJwHn;Bcm}wFU;DDlYTxv$7jh^YS;|=G9z>xPGF1$Cd{iBh$Q}-C&_izgpMvs(ESv_BPj{Fb5gZ*AJH&T*z2quZ%Yp-Y%wyDT`><%T zN4#4y3!OALZ0BHmTV5GL$C%-@H~zXdFGAQ%9TIA49O8iSxr4?-RmJT^I?~(D6p4xo z^;T4(aXGp&ZBCIYqYw-#`40yL*T`fN-_$Hey{>G;-rapmiNx_R?*1xo*{L;Nszv5< zKAkV8!(u)P*Wz=HI+VotR{_$!dJUA{^#C!ULQyV!%DDIKb=_i(C4|{!LH49>x%tWy zkjQ?QtJJelaD)lN+Ic@Jtmqa;m+82E0WRyMRrB165X*N3f*L)IY~p??m7_S@{WSaQ zXRe=)5$r=fjLs^AnPdI|BID`7hsA1iG={t;{ojjGMmkcV2e^#DpswhuZ_sxkKYd_v z3&HM0e&^PZ28-pHGcYEL^Ah2LB~KjG2VAPaoF0^M`Ek+7!7Z&voCk zF28dp743d{WX4SEPmv}5qYEI=W2E_nzT=QlHUaOtxJ5EFrsx0&bldO4ui(@7te@e{ zGgF4t|Fy3S?s!Ii@wicGAIL_Vph`eQV=q&Q4CP=9aj!Ct zNAyYcT_TFEKVB{AhULo3I}~J{F75Pf{{g4@>CKfc1D^qb?WeChDh4S$1+R~?&}$+0 z2lzuUr+Crl5OgMan*#yo_d&gKKiOeu+;tv7yU3=GT|!js_ZkkaE-qvPP^1Xel5osd z$X}K^p7v_o=Zya7WB54ZW+i@(7?oNPjaa`jcqI_xU z0iAP{4$b87%;nXvOQ5J`fAxlpDvxoGz>9%z|93&=U6i-?K4m5OzE0IljALdm`HGYF z)l3ia^p3QT_%~GL&UtrwetQ{*qq%yax}*8X5`^ob^FNA=-Ewo7$*6pPG8ux;vEsm9 zbYI%eNcdmfD+*(rwAD!vuhGrP&Uh9fp}Fgl*9m$g4GCDHO*)nwsgB_?u7B#q zAn~90=ujYc*j+*JzRkt5YPB)Tm?Wyc)gBAF)GgZcP&dKT9@!(5gKbrPu28@KQFj{E zm-Ia=h|Dr|$4m#V8vT7BTVxfV>w!8~scM<>_<@6XOo~6S8OV5H3aKB=RlktGwQ{jL zVKko2V@BpE`-3|~`3Gz^|E8rLcVmO*Y4xDoY9k2C z^I~UT&cXNUpb;~U&2-S8UW;{CZ@NZ&)& z3{nV$I2PbAEUxv(8aq=(AG184p#{dK4gV9J`wyOj0q(BPa1|Z~Ff~6_+Fw}A=3?-2 z*yqa!PF8Tgirg}La-7#J2YL^uOM-00j3p?syNh9`$8`S4T(f5gX4>W5n6nDQ^Y6o> zCp{G5lG*cwu$e6q*INWbdGGCPzZ@+JDAHYVps6OzNDEn3V40Um+c<#@I!FI>sMyE`Ei0ojmKx@ zVSsT*r*yWc)d>lWP89X=-ZzW?RJ?dz0$sBGjb6Ia^2+$`IuC3pvu8GyAaT8>6 zwPo;72q-<=kTX(s-1;$Xpiq3F1iALIN^^4wA~=J4EfrbKI#YUlh7v$+y-vv`{jwE2V06g(d|O>(AOx|U!HEhyA5Cu_Zk4~ z0>aP`7-~~&W125BOC$~5R<%qDZv@mfm2{OnX-iWkYQqH~;4S5ib7mSo@Vmb+TXgO& zj1^I;HeoS#;SjR9v7py$f&+7v31^EUwc5h4hmy*=e;ei~<~fO79Ip;p5qhadjG3WP z6p_JuQvo!{aTM^yR#HkQK{Re#-vGaIX*dp2Mj8lVV6E2w=OhlCf)g*yN(~1=*X_4f z^;RXm2{2+<k- zp(8Q&#w=I{b zX+^^gVSYI1LH@AZqJ(l>c0E1mjZI4ixR3IuG#+`v#58d9qeCQm2DGzliv1kAv^OY) zfaQSRVgUN5SA#RY_mbiUC0BHO!523Fy{e~F;6wp&eWc}1E>sH(eftb&-8VSkqqU&mx2p#< z2Hm?`u3084nD2_6InR$~$Knhm!=$`yavW+H5jqcug_hp?$RAIme&CKGcsFgVrueN-}%<}_^Cr&|LZhGUBy?)`Oadj?|)%@#v8T+*zlL^CGbXi;)_63-!WQikN@ zkh$8dT1ba(LkEBrLWL$U%_^f`RGqhXmjv9d59D^sP0 zXViAsuG!&c_sp(cU;UHwWx<@tDg`5=cNe?pLi)P)VGhZ22Km>;SXAGHvjT5hZH4@v zNbxzXzU4|`8J}i*AYu>Fy;#sOC02*ei4;1}I37bIc8!$-LTv6M5aB4yz0-b80H+Fi zCv8zDYgYGvTO@*U4=`@$JOQQ_$-grnWk_>$+~;U64?%j#>I}b|3}5yHhnVAJdT<+C zZ@tS)bNqaEE5Ij*Bo3g5G5xk~p|Ed-Q|yKcI}{{ZJ=w;+LL}hVG4x& zBJulIb_G!s4!}QoSVWg0dGRmzXFEvT9j37gXoz_{q>);xjC(IAxtkjsNo=#S6W*sG zAya;<_jDd2Kh`DQw`wj&-qs%%JcX78L=ClC~#>f#Wz zS<+i;mF?Ddwc-k!Is~+*_5N*?uY_cw^@pBff718-c&BhH>dw<;0@eUeN?OJBFOTz` z>3?`%A5`J4^_!v7$hghlNIat|9Y|lKOo9(-R@Lml zQ0Oqy44F94baZ$J$=#f+(Xep8NJ~#gC;1FtyOIj!H<9!MHWKz8w7ewc28qNM40lA~ zLFAN~WM)i<)P%D>zhX9Q>GPrisJ{E_=0G&eRKo2lHeMo!bS~(Q09pYnpAo zWLM?-S#f?o&9Q9E@a8ym@vu|LF3?A<_z*c`M@JX6$mde6aM9D&T#Rk3|#h)aSVA5AfV?T=Fom5}}%;#$x5>qzw_!{;r z+3yb8^2Y#i(41Bq?sTKKGe6{W2gQMwxxhmQ$Lo&Acbnd--mXOXd)VcH-Au3_ zygB4GT*DU`2Rne!zarv|asipsWvx_$U(sn8cKmQOTNT!v86Ohv{Bdv8$pgQ1B*FGpo(# zsAC!4p)=Wybs5)lG09INX`*ZBBu>=Js{RFJ;!&OJ6hkbw3!6gEmS+_DWL;KZMAodbh7aN6}&sPdjOrVqh1KiJHl zn5ej!v|HYiNzrEtSknUgV)M~vXK9|zewTU>q;We+=b3=CT|Y(S+`9pWH#AAC>I#6b z$@mks8_EFQ-v5Tpzykr`1RHMaC~5*ApTCsa`=_q(AF!GX4v^5(_|=Q$1rqbUjTX&B z!{13^Sd3rVKZqDy=|7m)*RN^Wl2lW0S1i^Mrr)E9n&GFG`z5!3r}S`|#2rNVMkvvR znm&w`Ml2xSIjE*oq$(AaMI0X;5@shaA!P~p7CT3=W#qu6kM9SNCewPiC((}cUz>>k z3%Usq3iQu|4rdaaX_N-G@Kgt>#DvMEMQO<-8Lj1>_JakQjf?;)XK=Vjn%2?da7jzO zvFhXfaqYg<6d0QDmhfHRdHG9(&djZb1bS6PxoE$?Z8CM%6Nu@lTx_=MW7ZGAZ1uu) zK7rXBN1U9cesX3aCz5VDz1g7P~Ya1R?) zNr$gmDwAK&JV&4qj2rJbygeKzAEwBlc%{FPXu6pR^n+x${5Fp%Zi6wrZv-jZ@# zWBy6VPz}4nF-q1$7-S)$)d@;FtJENd`IU1hISf3zvc9Z?D$7_b`V4=_9h&dvvu3KX$wG%ENxcUPS04}c?!*>mA-X$D8vL7$jj zfL{%&?Wt)l04HZFF`&j;LiVT=1@d`dMHL(b&5z6gerP!yMmkbd`Jp<(nOY<&{O z^i%sh3Gg|Z)I)X4HI-!F^Zzb2(E8UY66hHD7aDH1S4n}_it7hgrjbfH9(1C)SwlP? zS_OE3)}|FGAeN~BSdy)qlvJ^Np4Z6E)JjY3j&yl#tF?h60RrX$s)ESPlFHfgrE<{v z*dK3pv8og4gMP4^+_cq}$ildMDN(?J=8T~iHv~X`$~+BA@jT>(o`lDn1%mzu{4cYI8q-IX~Pv9s6WHkt(2)G$1UW%#xbfZS1>hEbFoz z+TQ5X*8{1L-EUP+D=Oc+<%`RU%-()D3+q|_93n%iR-8b-JIX3maeXB8X(aU$ozq71 zV7_&FYBLOt_&cA1VK6EozV$4<2pm4)_o2j++26%#UKDTySZO@VKUBWmUmShgn=bjW z-~o7{{($8BDFwwhqr4ovhQHv8j~UbXehLRQqXjrJXOHaMIKOW70^5{(+wy{lL0Th>EI8fu6+Ue7y*Kmpeq2+}8 zBTmO+v3f~P+rB}QiQupC5eHpB8x8xG-V``l2OCLWliOBLe7K<+p0!7`lcc`hI8xRF z67Wkvq(ou18F|_bWCK-E@T{gb5^|Z_)Q5~NdIFy*Ni6n%b1YK*89GzKyqg_i-=rnh zd2d169i`)B7^I?zSTRGh_CJ+3@*T1bBLM|Ev_FA1YiOj|O$mS;@0c_@Y$T0z0?#jA zMKU|>vA?}*rrT33&joj$Cs!!uh|zJt*C;t`zcJnC_>1^up%#R7A1zyQ{(jsiD#XCSiXx@$`>uHwB&M~ z7@#T0hWBAiHkQ_4dFjahmW>!ep#rn#Sl);Bb`jP%SCxr zrezATTFhvzav9U7&lJR-XERm3bUtH~83#H6k>HcH>APB;`T!vSKk@bgMocBU@iM+V z>b|}xrLJih8&3A9eoJt#pGz-{$pC{>cIOjz+8=9^pirp8+W<94e#wOmm}<9it*bA_ zAx|liG`8|+XJ<#pC6RV`_gA#_wCM4-L!UCgkO`_=i1k!|Y!&DhQ#6YP_laJ+*+oaU z!TZ%e^pfp?gf#twWM3OX$;{ZCDw7fI1I@Z_u!Ws5G@ecnqq6ffK0l5kRGSZX{w?s7d8ztjVPP;0vHW2J+uF)DT zZh{*A74lXeuIF>5HH@g`r|Prsqh~g-64}hw66uFTZ+|*pL6g${=D;NFWwJ_Ibl~}j zPQ8>(xlFe6`^396N;#{g+-ppL9ToqS7wD-*vgsJ)ydae13=9dSe>9o zGn)37e&H8i3dVM?Ya;7eTSeGvgX2r*y*YJ`^M@W})j~xSVjNhZZX47SYddVtp97;e2e?8c6y z&APt;^C%Lm05%2xvlRDtop=z1({_*tkarw6UlEyyOx(%5OmauKk#4n#WK0NFAKBYH z1I;8r^ezd80T$Pvl5PBw0Be0Tkz5(Mi=g_{={h<$PW^)~*^&!5lPGt;8f+dz5hV~w8?H)0FNZQUpBHs0J`m6RlaDPOhv@`HI4xo4Otg`{=*90 zWcPpA`|7YLw{~wu5D^Jc5v8TOq|>0KrAtIWx;sV$K~%aM>F(|jMoGz`8wMC)fB}Z) zd))7NH~YM?kF(!@&iO9AxGn}^*4*n}cl_>JkIZex_eA}v21suonTRT$xDe1W2W-fG zPW?{8;nqQ>H&6q58w)&=94~eOA6q(4-Sl( zwLaYp&wBYIV*rWX9yR(cKe)wpt(t(lV<5^*-`D9B__GrmO0m{ zKhxcr)}FmO@t*Tp0Pc}wi=N;1O6IaUp|+T;zEcFuJjoSzPlV`Xgj~^;6 zwL`1*S{!usQsNiZhD^W2TSj%qav8Lo#PU>WTU}dP!)XL0P(QN( z9&lH}Lx>}}{;X|o8*kX?1L30;MLkopj#{B6uL<1z@p7yzHa*|GJg{6LvjE=u2*wzv zTl$j~PaqLnpXg7p`8`WCNeFTDIM!c;RqX9UJWNXx|#vew_G5vPL;;6Fh{urQZo50hBN_HObf?~jg zQv;18W;%IS5Xh({Hh6EXCw_(70=-uyj2lD-4gG8vl0KAF$kfQt%(f)SMc6XN7VJ0l zx1h-8e1o^%eIrs$mX$Lh-1o_suapI3nH9v+yEwE$6{;fj&TSjSoK9RBNxA^03h$WlE=N0w{^216%j=je~!yzS%sQ5UeZ)pwDRju19}ZSbB|=L77z{g zP_o1-K_Hh^-%zy$e^mComrCd9!%vedGnEz_V6a!bRm?{c27*+{VBGfQczv&G(GlW!W>9u%N|YcGaqXLC9ML#D}&cxouNN767VBloR_7o&%WWkaX8l|0CF5b{;I$b4aJzIIPmS!DedrN zW6eU|N2bgv z*TX}1Mb2V5yH2y%=3C05-`{6IW4!%!HQD&pZx!T!YJC3vq?*rZyv6e%*}IEC56J_| z&WNUN{^DW*$Fs)|i?lI;xjSCLLcr0bz21uh#y3g)kZT5&EcACB*6jQCH@=t47o@Cf z9sRxDgP)E*U4z-yhRK@SXER3{LNBecq_Fa8xtWj13N|;z=5pRk#UkL8`u!T|2qk-5 z6pyc13ck(BKI!Ml`UF2Q)93eZ0+l0zgpHc3XVfi+VoNPH=ksQGgrX4JYqtnAt8h}> zv(bvSPW23$Ty7|kxD?39dW+9~X@Zov!1Q)4)rkA4B)_~Pc4l`!#a)z%Ca91h+%1Wx zW79Vrt&e>Dypf1ALu%eo*PX}fZ0#*s_;Nq~z|G<9(XwD+N8F?aU)4H11F86{@A1(~ zGlaprc@WJyB>R3SyH?RNgIBT}9L&fpHvR3hL6a{*giBTU?Wb-c-kzRhLdR^n(U_r~ z{Fk*0w>qxdVcfNm|3mp)`?q-(R7C{pZiAC z7^N@vt@{DMTBJt77IJDRm@S`Y&O#x|G}GoNNipH zgS9@{L3hI?4g%qlwuJT&d*#64(fPXmB2*GT=6@ZmAy-#bvRZ+**ka_C0kd`$8IoMM zW1S;SkPW5?Fcav;gw>BC zu90%4o4tU4LufR?=_MRrqsfz=f!NOYlbh>BfV(Dmy|-|#GVg0M6|IyI@tHFmw9TIT z%z7;KUKwK1l}fx0%Jken!kOZ;y&Ig=^mO5Rdnjv8c*Wo@vbKs`s7`7I%zS*TL-~@vkk}r9IOPshOs>7wD0Q`BaQsB{jngb zM(1O$moz8aRHx(4zRI%KF12V{IXOzQH;QEd1)sRRkmZWvdw;%_t`bT(Vb3cs{`c@=|i19 z(Z|JoEN7DiqQ@RUwea=_mpuimd7CcgrXUcuYo0TT;fN--o+GK zTxqPXPE{CeK7D(7Hh7n);WbX2D&xJCo$@j1|2DAmkCn6Ef{=rc%|Kqq=BQPC5}16T zBq0q22Ak8RW6?sXwD8Wh)agpYe}_^p?p}^c(0zg_d2Q#d2cwo-$hW`V;H6wv3eu7wSxCVBFtvd%-L$7gIx@F`J^j!xn2h+7)n%rlD!T&A-slWgH}LK;;}+9hvP6| zL~r<=)o}j!HDS-atCGF?Es=G2vn%){_rmFDW53^PgBCqa+i4GB5GPP|hRub$oXm4G z-tEL8}xcW6;%dtNL&YtJU|qICwdI{Nd#D>@`G9Qls4!UUDI&4Z`j)GPtN`Ywnakd%E0} zr`KQz-M7{v9$!kZ)rVcZsEvOZm+b`2A^sJ^MFxnV+uZfSY0vERp4$QQl-g^HQn%?c zbK;COOJ2x3Hg=D<56el%7_FR4*Pp+)l$4(%OjOCn5Bi$Sz+z{z6LG$eKNd9FR8lV&yUL&}KlOu|Xe z3Ar%nT};H8F1_~EbUC0^DodEFc4lO_lRgK@S#g*3+3Q95zcnaz9d!Nup@;>APv z6A6O5mb$#ZY#YXNTlSS4IEci#*PQSd2Ncc`BgqH3^=>mH5;X4|Ed~Xi3V${i#npw^ zthh!|)<6*8^-IEc0kIG0=yp=fbK9NdJjY+%JR|j_PmD{QR;nKp^x*iqUTH>f|>VDL-?EOYu^E zEFSeN!N6r)N*|zNeO+|-S;cVU8wI}V!jh!v=&Z#M<&e~j_Bj>vQn_m<8fYZl_dcY% z|45htKLAOG7Qkc(Qbrfq z|DKrZeqk7M=Z*nQSTMGcT#*ZK&Cp1?c+_eKvsT`8dTIGLg5^VEXM85JS2~TJGtX@F zYCXvdLwCDUxw~3)kV8wv2lCd&Wm(dNi_7`cXiZ+o%XhPJqLKw6c`w~Q3cgq-pr+-h z(u-|fJNUF=3+jI${v|e#cYA$Ln;?OUp+!}+m>JTmQKpKOX}X#)<9U*PWb@$`b~Jnv zscLg}GK^suolj*jhH+176PJ{ex>v6$5Rdw5Wxn!#>04h^bv=)E?kr?ao5-ZeDcG55 z5ZJIOHdGDiAIIGpaTGbU?+R%XgxIA<~Kd5b~)S^4uD20(pvKnuyNVm(a^igd-_|I`P+Mz@1N~z52swC;^X}d4rG`8sGi@^QKh=g z8FUCEpGqlN+{&x|WE@ZQ{6J_Ru3Q z&u^8C-RIIo+FwJ@8d}D8Fc47APWTZ?dkOtEsRLv@Nkh^*tsl)|OII{xu`0GQcBR$Y z46IM9Ksl;aJ!$%dUG^W-Y(-Jui+m@lE5WAa1K-7St-7i@%*e8}XQ`%5Gp`j+NXI&p z9p+WXoA392t~K{mBgyb{91JK7XbuTFZ8P4FV)E-|5LQmK4p|p4ZKtLAgljr=iISlp zrt3OE1zvb@!SS+V6k3i|k#PVTV|z5r#=teceW9fQmS?m)V14wG2D>Uxsn-Ea@iB8D z+A_udY%_0G$bN$-GJWEmuem(ta;x3!cULB<>`$rGDtf$zI1u-@n>cy#16fl!rzIaD zp^GOsPvaq~Xobe#N^P?QB8LQ~L1?6ST|F7n^vhs>?5+q_pUJI~erM*W4$Fv?MLeB_;q zt!DA^E7P-416N%e;VE!7uH2Ae4${chalza0^^=@=Ei|a*$$_1qXWYo+#cO0}rPJV% z=g2OGh2f7&>ca;dE5eqa-I)0*qL-{#nQQxI$|sl~jk1AX!aR1l!QaOCg~J0>{dd8D zA2=HcFQ2G>j8*n!nV;AxWcF0!Tm6w6#BwEvFu}jd+?RogHxK61u3|W6-X5t8ni$|d zMSgM3Bqpj367g~;vu_ge-!RH9vUW!|X>2jL-f*>eR4~Oz*YShbhX0c7vSF-Fuj%wj zt#Lj78^JBWv}+w-ZXi?8uaWNaHkZ*esfopIa@tRY*vMNzisp{u5F3s`$SS0x*+}gM zqg-P0@sZ;M5F<96!%V~kVL?IyYa21yAspHK=jp0E-KOyI+Ubzd&+hF z6G62mkU9;%)-M#d&G#>32AVGk|x>!4rwEk;2%8t7eP!e2YiJ%26F zVjK|tF^cRs)Pt-?rFz6slgRlR*{5_^soC&iWW3ZfTtA*+;V4JpD(lS^5z{1s9RBUN zuIdcfcBHS3e8})XMqyD_6a=FP^0)v9VeDAx_z{jTnEMzQoVM1DjM3_C>ndEQtmz0e z17Dsw$SiwFnO6I(Y>yy4cWNB(@nX~1n~3KKX{y?L@e^u8+a?Qi+a|r2_P$Nllo#m9 z%k7cNN1chSu7QnBAB~HQ#AQ>N2p@NORPw8!FJ8nnDbSy-P6q5=7f-%(PM$EzOE_=!x`sirUrRc z=9u`f!oh5jU2@^;&5CZqH>?--6d{p0zr%Du^8OiZ{4uBn;d;XtldA=syB!JDJLVK%}9se6|#MD6Zt<0bDZ0eAbJ&`2mV zKFqqGC7+-OU$Z=VLp`B{ouj9k3$@S-4)@b!%-7+=yKXFPLXZ%MM$h~8TLp}MFM2Sa z$`?UUsQKOao@&;CS-;bBaOSiKS^kteI*{N17gyp8`@i3$Zd#_aE%)$Fm=1CsYUO4$ z)#`#X$FWY9Vs1XXjhSrz;pujWu-x^s=f+*Ix5O2+=w~4NCCB^}hMc|hk9@UoG5E5%Hz2sU^LMJATx!yfb zMu{K&e3v~lS(N)i&BPHoQ)(L`zAC#!c&X2w2(<&>mTgWXfhXDe45kW&gR_7tuQe?@ z;aPK6z}z8MR@$N~UJZZBDbYj~EgxnFBI5Yk*S+lE?Y(0>KF}%d0r&D98jZ589$vZP z=Dr7_ceNo2OR$87B9-85Hu)MTu>e7PKn|X`oT%5~b@@ujxmS#31P0)L{|9ON7x|P~ z%-`x}vi}8P{hFOQC6B{y#EQg*6Zk>a`EL-;@%3nA(U#5fIu0A#gjoOy8vJDHIhEPx(Eh}@IHHw> z$(|i98rtKC+5EmTQkq;^St12MB;3MXo>7yHEQ&=b3I@jqD%X@{`ovakZQcValo zA5<{MOD#`>mLKrR9STkbj2N|AM0|~XH}I)vFw0d`K5`lMIIDd}fRGuP-1xrBdDEvT zlKt~zMt%YH+KBqG+)-1xYsS*GTp9NEFD*8AbA}qDK+1gWlC&}um09Rd>GxnI8iM2^ z650@UYp(0PljL;a_PI8zLSrsv$sd3+I{JErJKn-*yE*kkh|203fM}iul`|-2NRiQm z5M!7OS0x75G*KY)L!%4Ffb z%rXxdbiTknGJ7R__GEbY4oIcbfu?WQXyL5GRo-+{LMZo6kF06=vQSdJN1y>hEpDK& z04rrON*~?CxO?91i;Y2`zjv&8)!Z$b5~o}h=>l>fZIUYfA)5@l>_ho?!yS*s2Zz-j zXCzVaLpycB$;I!9gk6r5_wKazw*V(Adu{nCjELHbhgp5d?AS~5p5o`B-_XrEP!z1M zZlb?QC6;A)yrjc|0X2zG6b7vM5Pgz9R&ST|rwwrk*Gx=v-;=c2?$goZnBzTu3$`cb z5MJzFh;-r7E87_vu_KO3OF0|dPf!Nr&F`>K8b|pUYw-Yh$E(AVh-&gre$G32vlehz zM4@_OA&vau-=;DCvH9qaCjVb3Apo4we*otcpzrcG;4D77{ZXAr1a#-RN!KMp?z*A8 zx4tEYF}=2&8O~nRmn+@d%rYGlN(IgOwIb&&fX*tokjp+NA0J<~h=$ivt0m9{I}5Pt zlmI9V*Cv=Hh7-Iuk7Ujc3O%ns+`38i{uUh}Ah+@@XU{xZ{c#L)yX-jc$|vL8Ywa)h znp+lrVNmS4q--{|b;zV@W(PS^mDbvwZ6Vf37jayo^$Q6h0rCwRx5R3a^f90WHKi%f z^oo365awuF^(2Z^$RL+^)ZzT;11}F_UWO{>e2n@YugVzUAe|U-yGI&gYAO;-8LCYl zPH0vTkeEQ!V>)+5YltR?w|)~X&xyPh2v++myzV@u8B@R8D4}pGxSqO0GGJC!-@1yY zVuTH1=z7YJ)!Yy_^l+k}ZRH+)6xFTH<*ve!<0TD?#9%CbfdoKYP~e2Ica8m& zOy=%9GFc=(U@Z#ZQ+{&QV&enm46>@UAtP3x(l=FQ9n25pj-ka3HwtuTyly638Nf>p z|M9D7!tYRqs+%FO&|6b=axsScLQ6b{6(6i+q@})fdfiEI1-a{IRg=+`+1u4oHF&}7 zme*L^0RQ$@b6^nWX49bcPlNqGm5pfb`z?(b`(Nrx>U|SijY*TAho;`nuB2o=Gh#U~ zVnj#C#5i4H*%vT)v-nui&Qelm^YbsA-*Q#CF0>dv)2+Gpji5utG$kmg#dfzOTE5c(E(cn(4;dFY;F*X^tPWe34$P%piHL-2q@0+Es} zv|r^1%mBJ7II+Hre34(~GMl^qN+$H_w}FKFz4Pgt@xeLQu#i|%c)6&kQ;=Tdbb^;c9`4g^j|?lQH^9ZMC{HWD4exDt&b5 zo87cau4g4Qd@yJkC^>7=L#zmFnpZW;Bisa}K?|(JTz7VGgdy^CK%(?Lsqlv;U>!Zc z@^eH7bn#Ny){m}n2i%&r9xy)~X`>5G(Cga6A(n{x7q?pcjI}v={Ns+i*FYQmZ}H5{ z@2M7tX#7eK-6hHphWxhN85?Ajt5MLJR|@fuX;wG++{P6EZRx^l5DZRKO=V}xa`+H@ zeFzvIkhnP*MKW&Ls`E-N@iq`*LZ?(~_O=2DK1@7k1D--WpXYeRn3a-}4R*tD6W3i7bws z^GT2|iOK#<&SWVtQF6OM$JTbB@erajv-)v(|1t|K*F5Gf01wU$p=XcwYT;aO&V!i0 zMV-^0N!F=n>Lj#?Ke_6|mF{xw9RX$b^>?hKW`^|T4qniZftFF(4rQD>onh?{pFHE} zc_36h^1@O2GfhQCwQYfhV`2KBJu5v{YnZ+)@0^H&2;7ckbutf*a4lS1>aROHsZ1>D zvpP0$@pOt85rrKef4q?ZUqWh%B9~!KSD;3WCFC zAm{sV(ff5>qQ=$n&Ng~Yy_Fz^hN=^)pgnT7lNB&L7}h1qqjCPI_2}GN^lV%md_^+2 zP`~rzrORmO*YJUd2L8#5W?gI|KI=y&v-KEZl3q+0PDQ%p(Pz-Cg|1)`&Y&p}+@~_Q z@WkTYD(SR))r`-L+QzzFfh}FkcUA5MJXObgjuiEcZR^qGo(E3JjYtH{K6!I9NRr%b zE2wX(hizJ*To;^=C%n}RZ6w~=Gor4;txtAY4;4JC1nz{~;dJtPc{r^h**no_#SDM% zX>`xZOKo{?s+5goA6BPNE^K#jSAWO7tB$|OaDJlUY!(T!J~<@!JauaFP!|Q&Xc7QSVK4|i}wK>+#A`j4$Y{Pm%KqQd_-*?R)-Iu|5M#tpPR?I2We_x*1Y^C!Fg+poOq2&&=V4{RDa zyrADC;cJ7))+bP(9f9vx zpm$c+vg#J-L{G#x_)HUtA1|nd{0Q97;+=Qy7WK_74pSq?J;oKu0Ycyaf1!QqJ&B?2jAnw$A^S%2Dj|4pk_2-$5mH`N7D zBTxw-P|1-$3I)v+dm~o$#@pVxX#h1Td{4k`6O`Lkf*K~M#ki4BcN_^kpdN*diA4Ye zKG*$e%<{Ljzq7oUot!FiK#jmu0D1gUt-l@4UsbJVpu<4XSn`u8I&f7J!fa>`#gh|D zQFw8*JlNNR2!qwZo2H`WAD|#DKSV1PJyX3m#sm*~Q(`w6lx z_TVi7z|lr|nga0VO)|a! z2_>)^)CM&_>HmP&$_&qv z??VW6)RdVV4M?K~0N#j2G&^`ezi?V7f!nd6X}Hd)B|0yod7=#>*y{P3K&z#Ir>;WP zQ{Rf=iSMbe@7UqkbiqUohERM5@}-l*Sr7kU;Qt)2|7QGB0f3qYvlsfqA^*<{c(Vfd zuTUg|@+GR~3^sLLPh=T9u43*bLpPnjnu5lEp00MuF$Flv&q0_00K*HrGKMqocc zVE-O}(!V{<3%*_KK`gp?y7@5Cd4?F;o3tuJi`uDOMzci&!bi|a7Ci9d-Ci&$`~_+> zqt~MUrv7l9>eqlSg{Ud$QikWmq=$w(BWCpybH%Yt%$25fJ;t2AQAkm-y`iDU<=dEaj0zQ7>-; zdbzI3h#G;R0D++u{Bfx1Wfl-PniMIcSZDzV49f56{{vHh!lMWP9!1bS@&jrFrUL}N zvJ9gAS61cUuJj#{+{v-L=cp0*03eXc-2NeIk}D5LZm(VjiaD8Is&ygVzf|i&d*1(2 zt>4wp`~XY3kISUh3K#gJa-{a|2}|ue>Wxx zG}nH(nngwqq(=hqs4h7FI~e-Y7h`CH+ljXjM2KpaIS(fS2x{>r1wI>lqL8CoQglF> zvrQ0O`4Vc3-bNcVF4Mf)i_j41@V|+g@xSD40mM)1XF4cmR0Ajlf^3bs@RGRO`Yp z|5B|#=<+Ysx147vS}m*ZPC!{fmI`kA=KfbhfC0jmu~N%Km}zAx-OfQAAc^6na1a6TRgaIHTmr=y0lgF>6RPiG>vcY0 z1mJx#T@>;D~)+3s4H35 zcYcvimo99??-2LTUi=C>{s2Ax_lF(7b^82G&XfO&qyTNn>hvgy=Ou44@70WG{n%)T z%V~wWyRtND!|89t(tpVJ{?*IJ--g?f=j;+>q1Vv$$n|p?{4{*X8q0z^j~K1!ms`ESr=aHm(Th; zk^vf!|9=G-K+`M`G5IeP@XoydD^(@n{m#>#)M_O)oSw~L?NT*9J8yDo-aZj%Kwa4%lR zrMPtSCPBzeet|PMAojT4tb~inEoF_ zb#|Qsm|f>bd3HSlxYU3Z<&peFpa={_`SylVfEh+8-!vcu$WuM4Hw`EPbMiGPU!V{I zT%eGE@&yW=z!XO`%J)L-07)SP%J)M2A5g9X^h=k{&%d|5yYlzPajhp})ZY<524f<) zOE?rcnk7m~itB)XFmCqZD}u#$v`gytDmd65T^7In6yuKe)(Mmsc}}w2oj4XVWxv^Q z5Z?!#o+=YOOms6|jvaFncHhr`owlpmA94xnKl&1rC3r1~0gd|7S1kPoUFtvhvOs%o zQ)#-D^dBuEl_HL%*is|D&1FMCarr;_iobLx!Sa4R-G8(Q!D}U8S5`_VHxmA{Z8g)1 zVVJ~Uh5v^<)`|djrF~oU%73`6Vt~nGnJw=B34;kJya{QS>a}N=+w6d2yB_n!^tVpIJ8bI-K{3-dXS(kfzP4JwJvw$>}U7 z7&>vUrib$&p#J@7!{nfo9Ah|pJhA~ct4`Vk2N4y}#yfCwBbd2z&Q~59b4Wq(W`flQ zNrm`PWI&CK3oiU-#FT7HRssE zcYRZSdI(K63mOxA-d1zcDY7=`ot$$8IU~zb`}MIN#B~QAt?#9JzQW0Q>M(e}Nl~xy z9*44htdbn}FevWu99Gm6yXzcdskWDV{tyHg2&w)wJlFl*%Z)pY)n{Y8!oDDO26VFR ze*}Zun#JX08=QtoYWAJH^c3V${aHR0!| zN&}x{dkxq=)~kf9e@j$NSBObN1J=&;ow>+zhN<1wLu^;ue|5> z61#3xmfCAU-CNosU#1?aD5M$pZm*cHs+T)F#zI&ZzPh{jcii&)+y5;Fvv{=$JuqQQe&nz#!pzi^%l*<4-J<@iy zj4sxJVCrfw$-28$bxx5RbP74VS+|)IrJJeEjYXH>PR!yp3O?&Kl70mR`&VwIn zeuH*+(Hcl4N|_U9G)RTxm8J7ytBii5dNQ#`E2|rIgihPqyi8@=t=uu`p@7hC{YIcr z67QH#cNp-Kp77n-8U?`y5D`1>1UeSHIBf`_6`*`+7d1MvCeBQl9me zs6t}dg3rz;4!mV)3U|L{SqomfWbP!9SiC*wA!pH*n(!@>KwqD{bhZMvfO}1{e0$jw zMk7`A8yZ?5s#T|=z;1P4mWAYd#ui1!36v8g*~BKM$?0BF#Qg+_S5A|PR+Pc*-M5wv z7V-?8geKh?V4uCN@Au;phRRK`doGv+__H+CZf!9}4jSN^ zBerItJU8Z!Vn)oZy1KzeG1ktm2c3^!vx*OdKUbRrjl1x`ViaDp4sr0KG}Qyz_=^n4)Y&4iJ0id&c1qTY2I<)=T^|B z%}*C~0B7HNHMx#>M&PAa^MIv5v&&cEEVR;qvq5jQkY;S%S5kSbWaD^~hPjTklacB0 zq>W2ogn}d=3tZTCG{5RcK3E0*1W%vorJ88}(j%9A(mM|QaLNo}CMjL?UKQjjG&<}I z_*yN!FpITG`^hN7a<2em+%swDo@t$@Mfy2du$%mq=9MTBI!k6=%FSEj+2s#SEM+}R zLrq%2R!kvK`YvhMx8nNhaMB@!J$IEYWKIi!8K(S3t4xM%O}$09|2kL`>& z(3!jh)Rwb>aGs0EKEb|)#%%Gu^>Cq}by1YQm1%N?+F`y@ZQ9ZuTSchQu)?DrV&te&VH=pt;3c!Aj1Pj>-48S|qP0lHZ6>AJH|D8-Bt>?$@o- zCsdT=enykVOz*^wKW&`X^XX-lkZfJ&tIN0jKF4@2^mq=_X>mB$TSa8Xyu3c_7|`yp zP*b2(`8bBHX(vt%_q5+daOt*6t{ktfKG@#KnjK$k6OJrCS{&@#b=NJghh>oE~2MuhTBvuMk8O>qW|6S zNE{9!+=6PgDzaw-DUD=nJjr#N&oLPuO$OgLSFBvG{W2d1;mPgWtDV|1r`^5I&28 z=QB0|uU5x7S?5Z-fsOo{oDrq-*s@5_SZUyV9YMJk)>YU6Gwx7(aC?rVFKWFurgme$ z?#|Bb%q&8Yr*wFE3L`qOY8>-3i_M^0oHJAqDj*wN0-IPEc?bq8k&}C#Ftp)#CE1P#s6v%zrrnN( zxveH=hN0MIG%GB;Gf#WCoDmMAMu+dB&$9Hq4!ODAx``GDefm|^HVYALMJbmJu=b8C zATx!2{!Xt0xz&c397!GLznR3iq$a)6Ot0$di_@YIKStKDx)T+*h#h)y6%^O*D(m)G zA19|Tq$1o``GnIa5}cpK`dz|&B>(K$;T!E_!J%N9I@ZsX({_W#d$RKjjQ-0k=3&Uf z9ftYa;WP94&)mJ)pDgDUHS$=_`ICEEXqsjb>ez@>M6$Zu3mWzH!iLpW3#~TXqwnvu zvkxkykS5j{N*37Ec`APM`flRWGrJORE%Ce!mJM4?s)J`wdm?>aV?ZyHd*zc25aY66 z!sm*d-s(4uj2x1vB^&LOP}lSHIXVbeaBP^CDrhjqE77xWP+_{#m>P-Je`Zm(x6BQH zd7N0zd6q=>O89JXu$DZg`KslMDw?|x&Tq@ZPPz}Mj9OU?KXWo~dO+$e>o)wmt}ML> z8Z%MXwk#>p%9b^mFxwMRXy|^Re=xwj`F)0@dWI~w52q@(<%wVJ2Q{8iQ-90n$UD(_ z9dvB&wku4|F}pgRd#Xcmc}0vP4uXR6ltGG}g8M5^9OI`;j(QtgGKIgm!-bucWLYYF zPZ=Q%H<~}seiCp#zYAHugKM@DVCuItf8au_g_$_kH@=$}?m2~#GE3QiA zI7(X?=tvFS4L3P7KVV7Hr<2*Pob^l_zO)~5p2=y$oo0i$G{j1d_-J52k<;g5!pPPx zi2%ZBgPn%>C}-zUt2|?3>M4)Xx9Ua^CI;+lPj+xm0e3)}ax`rld^DS;Fa&qwz`7bw zTlt-{Kg{jHh{?<4+tknMnu<@86!mC79fvL2SN9C68=ia=Sm$comYWv91T~TDN7f<= zkB7Ic8#zibk_@Iz_Q>+G<#_eTMOEZRT*f)|fydVOOn%=ga;$dCf7j!o5ix3rOl;^} za~&6c%GY-=CHF#jty0(e!$?((UF^_I^=kA?;8%6urSK;lc*Wl&U@$qkF32Ykwbkc5 z8!H^dD+2HXgs-jkPbQ7F41*~jN>onCy-piEc9_t81~RBaPC-et2u<7Y8y8uA;2POwpVHnc?C6-9 z+tLcJNoJX-8n-P*IP|kw$1I;|CpAUXznKg%IV0cP>1DXi6}nm)BwmDCl{_q+wa>ZpCHCkNhEi|oJ(C(_#S=}VjF9s%+QlcDu;NA+nF9___wIw< zXryVW?z2g1-A!)mWLPcow6~3t0q^BV>AGE*2?_J&x&T#YQRBmxQ)0V#={TX46P0k- zdC6ww8#TA_N8}Q3u9)bAJ2TVS_Hpg^jYn16mVsnv!kwpNCTq-z=bf{ky%+TL4_sDE zZ|g87>W+O}Wf=OdE~e+X=bk`SzV4hbE~2&eP2LQJ>kSMj|1PBg@05pstn*vqm_vac z^O5%+uyQ$`oxZ!Wu2_5g^VS2ZovKleOYl8P7vVlWFpPBMTCXtdIb0BF5L(%q=#JGB zXtF`)`?;u>a#~rJ=L!+UUH*av?N3 zfn7Ua|AutRd7ZR1pEt&YOs|)!=-r!$R^aFxPbDKBV1W5aU+;agcRBX%+7(V)zF1*v zP@7!U>XoXiuLM$VAR-&s8piX%kMXY9XS_@?bbtt-KBt>1By&9Q&aqcl4(5PVOxK!! z;EgRDwYcJWcc`k&Ju&qv9$9W-?ef*ZOmfFBj~oF% z(eq@cXA3pH)25{>4|I2X1Ch5+-eJ|)pSD;BVBd}vbV^WZ_EK!P!inbvAr=lhb$a`)L&a1<4ieMj)qEgd+MsvXm?f{S?%?ItN6*e2wF~{F+XBF(pIK`c zW$fb;N3hk{)<}7zkn-?N`%`;;gPqc335dq=*7Hq*Mp?-O+P0jQI6IQrFi?6v31eUX(3`MHz7A?n_;}$%v zOXt6H4||DE`p$J<-WeOCz_8U~gSKJ9iOt-*dGSseY0qZMpIyUvPj@@FsAlKEt=nGu zjy-#8GzBLgs+%pEHVBt%m)G*{wHusNSHPzvMyD;pafb4ap3du+hw*fFJUqHRRkru^ zt~}gt%)P1Y6Q{tYE!y^nT8kS8vxbQ#a#@2<9+6yEeHkRO`Kn>CQaYBfk9ys8IZ2z; zZL9HRPrA`Uk28CCFjx2Y>2%_(7jJ}}uv2w}4_6wv-=4X6<{pepz%!=@WDW*bZ`G*x z*A*-e%NW8Nk=||#1!ocr2RUFkgzg%I(MI)(qgAg+QPr5IUjPPuy7OS?{F6QDRt58= z2UAknl4R9w5qNxUVmIpgHtXy3hwn-gcfP0JlP2@&>ZzLktgvTgU7fnAZ`VK8!9x4m zY3%e;LCl48PBma6rENE%&M$-pD|WFgU&yVbLeMEH0X*2Vsr}d|g zb<(7m&cMr62hmjxAvT`sLlZrf<74#tsfB4h;$u$_&or(1T3Sc0Ag6L8X$4dhW%c0M zyqzzy78J&(Ke(w2I(=+j{+dJgE zval^18E;YVt+fBRF)gsyOtHgh$`dr`XDAjZE$;g+kda;EL^uuV>x31qHyhw}zU>dU zBD`_*HX~Yl%7Vyc{d3(j(wnm4M~Ug2UWsar8?GHSv_#TZR3!)Jvd&} zWw>Q^4-sh)09uKwvb+oO#q=^%!R2{GDXNJ4o|Nd+iM8ze|G0YVs3`cZ`&&fm?(UQZ z$sq&;K^ml+A*Gw4yKBgyL-L|Sx*56@knWW3M)(ch*Zn-t`~J^buoketpR@NqXRK=C ztXDkKq&iLuwU?+Bne2Zn5$Ky5(|$Y3<$rD5O~Mc=vLwzoPrR+zxXyGDuSxB-QHIMY z5c0^a|18z~bo-ob$A@`T=FXpUzPf|r!&G%(TmQWj^`zrYLZQ=Mc=XKvE_sH}=c(os z`9I^Wr>cohzrL4q!aUtyy7%uAg73sYNE%ol^R!azjFu&%mk{Q7PtOhfeNg=zYFILMN}jlTu;{G?gL zDV?olqt6`*pz`bv9sn|>pH;eP3i#ZdhVRo>}Um6N{RLu17JR zlb$lJ<)I$s)!tKkgu9nQ>hp}|&}qTxWq1tHg+WU#fe@m8R#cjor)O8-;XS6K>;5!bGmiFKT zJzO$qDil8sjEhH6c37|c$D@*4fw3Iwk?G2v89UCBF?$e+X`p6juQSOH`!l* zbtktaHTvi`nwiguo5p@H2Tetc$%bc&>1CxERh_l>!Y46RFRoNbPs01;W{iU< z#-6ynMPR?X+locIK2?sWT?y{x)?9p>n)Kjw@lmj&ZaHLVgSq6E21*NV9hiuHmPORu!eZw#uO)55V&6m50$ z>4cBrX`1opIQz~pF1KEBmHwTV>2oZ#a^=Yc3uuPV@^Zgs4>aKSAY~W+%XqrGbUmsX zOWYbR`tMI3h48AMy@u1wI}m_E-sm?+PbZKXz$J-Y>s`9}MMlkLZJQIMD%)(G<0xiX zBM{)yaLQigjZF*a8=O<+i!@k#fjriOP|PJSU45WH9lh)q9>>Ceno$-bu|-aujlY{Tl(an`{7;avKQ_HmQHF1;C5h-MX4qI&_1Bzy{_Qx z>EN{?DAIQjmLB_b2t< z*82rg>y^EsgU_i_sXQktzBb?Rr^cMOYHXE7-IiPpFOJU`vuh01>uXy;9Y?WVGs3Bq zEs6XmV7Cef6}1^?tfG~&<#|3w9_#8jzZt}QM82Xw~<|GfVf6B6aoO4=Ses|9!p$;f1(?MQsn~EgE|HB*$ z`vu5o8#sW?-j&ETfVm@TEgDIyX^%FXmS_L5;<;MuFiC|^dOKS|9_nM?I)%cL%n_zw zskUAPwE$PLXk4!KyXhz(4!7#g7;`mvTZTk`ZJq>NzGDhA=^{RQKUu<`IS6#ducZCt ztLsCdHbZyvQ)8Ziud`1KDyZaT9A>+Epwy~)X*9(taNnf%j>+D{A5&r((Ohd1?Z7LC$_YBOvp2B5FT)Gqr>j#wB z@Jr2h9^`-cgySk$Qi>%+h`H*E$vh_6B}i%*OJHtiQwbrRJHp|r|Jy!GJMjRgwp$;s z^|=;n?ou!MLRr4_u*f5zefEy`3jONKLg%+wkx~Yx($xVnzprD?chvGewFWwVOuo03 z9gjin@8_lN?}g9fIlP4v>w~mqFd}W?P4-b1%%E>;5f@|@^+03jJLIN}v$tBG7bNcX z?*mpv{Y=d0bDrIZ@*21ro-qsgQ>=6Z(Lx$G*&iU@lf~P7lCQLhk3{=Fc)o7D7j1~X zJ1;xrWBl?A94Ovk=lMGBaae(yzG&H059~%jkEt zZx+IT2&GS zu{vvKj3v)T7hm!vlkLq@fx;T3l8Jqsj*053t(`PQAU<6=aI{Q6sTYbYDHwo!`ir@?jm=zOZ zda9g;<*rrteRsRYuJezg7io4l>df#FFNaTw!P^5OkKIa13Qy>nx;uoldntsdVoviX z<5~KRbycU-7+@|U>))sO@dvM4R?VP?SVIMtJ7@+Re_Zb?LDb@^2$9vU8PWRbs-jt$ zQtho7K{>pny=$;!BOeCxYm)>RWzf{h8wj#~nZU}SdzxHnvfVb*G)?5VL}>e(05mHm zB5ozI>?SQB1&>ki_~)PO6L{@7-D<}s zS>9K;OP>6*B7GTkc=fh@z6;7?g-(Ua81Zn$zheFv)Ddh;1tg6vzO1EGa=D3_B+vTY z1;Q!tygt;PcRJXK%%EuXP73<)4aG`+U`@Ba=M5(TB$omp7RxVDPWw3#Qc~=^@p&V zTW3N$lM5w}JtgM%s5p_{NT{D@8}duF@XYPfz*1oYIo{IcGkQ|8ztp?1*FD3sZYpay z#4d|b<-IR0(zn6QCjKEF_XOqDNE_@$y($Z@)PuF!;YW05=w6?+J@~%$6RmsdUxawf zqv%(s@)=F$1oItBM4~A72PTjqlwZH`N7Am`s6$p`(chn%vqnJc5E$W^5B7%(nfS0r zEYV;+$-qk&SA*Pb!C4}0V3I#y>>YMr z@F)mfc!m9K?QljtVi7#>o>{30r7i)($FokWqPaf>(@Lyy$z0Putv`R2e7-tJAx}Z@a;&e}94@*%+OU6Ob zwjT#Sff_(jllVnrTyLQ5yl+b7`SSpL2js4?yFH`>vo^Tv$Tx2UFAhvGicj^ABO#v6!`G0>>_7-0QKD}Iat8SltRBs&Gy=Dts%hawSp1QJ469YWItc4`mf{{v~49%0PnwUrH<+b z_lt13Vnk)cvz>*-SXKrEzUEWf<$4i%j(TnF9fc(~(S~I?(oV+0io+Itb05^yk*2$F zIt&lZD(~boq5Z^(wbNJA?xPd`j&1m9jL~MauA{b%l8+KTerdJLTeMrWcs$;l7Nc?- zRBz`T2PU{Rh5*tjd_wj)1_hl+Tq(O;o6|J55SXHe_BYo!>(*QxN$sf8iP$|;nL){` zP;P#B@Y(*)j5acN_Y4t~GBzJ^8O*7#k!8aCI8$*$rZw-RDz6=5JJO}u6xLrtBbOV4 zG{(~x3j|K47yeapGDIn3MNHr;rf&-fq-bF2a%w82D2OuPjo6fFp>k6K1GJ{xbxv9M z;`)edvsLSk-hN3!cpD_2!(z8=jfQAQ@c|)Hf24LqHruhf$@u_n5;BANnbf3lxt3zu zbx*x5!>DE_{EBdtb}_7?FM6BqxP%lm8C_p^#ZisRJ$bXG--zQ*O}wPZfQ9JTc=Hs- z8-=r@3if@xo6-5`_}HIeU**VxGR|awEbdty-r$gh-thiO^n5On*ABRYzZQ^`s+^;S z8-5`BWTzinLr07n?MMjG-K3zr$*g>^rG!OOPOGB~ih1K>?jL#>e#0q&K!LUWYgfK5<~8u{eC-rDKj##L6(VMQ+PhH4H^sg|GPTPhCM zV+wq#ap+blu9Ak70aw0Bh+CACq2GPEv?icDWYzJ%!jDOZ`4`BEMgb#v4A=<<=uk{h^sN3{i*jUYvT}HZ(IQwxkIN$|o1x+$ zdNr-K&0j^2ZTW?IY?12&^RQj!;nr|O!-Oy(kD7vaA23z&p#+K8s(f8E3Or7RUTA-= z?PhY%kg(S2LexMLnRysrucjm9pi7l1bx|;7-IrCi_6Q4uiTOMzx$~>7b7WZd+(Xjh zDLvR3(eDTl&M0~yE-u$lC-5U~01obHCFpd0Kw-fr5~-Rc3mf6LNv?7<#<0kq+R7%1 zIQm`})eIkvUKqguwM(!4T-}&>7zHp}Qibzi!|Fwx=X@{)dNAHyhzz%7e%WP@EJTOI z0?ytiQe>2oR#%@q!;Bow#%13|$64FEU)GA7l{$ltw1F5NlEf35VLmH`3SVU?7O$qR zi&ul<(%&}?lQxj2VL4q$57#O&g?gp6Nv}kL1MFqMq}NpOLGVw9;-R`3(hQ5tZ;wu=8T}t#vAwETv*i1N!h| zPgIrBy{T?Zb+qpQ#wrHS(Gx7i=dyBPJSvc%Geh&e8+Onk%Dc&H(7zXDn%L&zy5f%V zUYDzDl3dOBuHZg}W5TgL&Q4pYyJ`>4S<=6QD1Q*_->Kz}M(c8%v>MD+5{Fz9zAx|% zW7rQaRhp4Iz?3}eayEQg2mKY+Zo?dKeXCE!Tnr4a4h#eogP=-ivo{#`**Dq-A)9kIO|wY7$6xyw)L4>*P%F*229m z5{d{k-8vy|A5Zx*!Dml!{2FBtQk6I)Lgg9G67^AWn-(38Mp79?q)09^Yi)`LS z+I0k>(<#oe2DY5U=9rZyxJR?vx~#mwGqu%_7kZ4^x)PXA{=9+We`WSq;KQ)Y%9+gq zgyTzUL5$VNMJjh?RN8Y~ni`&YgM8R|79St-wm9u>*Kksz$?#VjU4)Jvk@wR0+B8<# z%o>&9veqfw)Q)c3PEE;I1mzL)(tH^-Y*Gp(@uvTA7u)p;H}M=y?7&AHr4 zQ;0{Ah|<2Zq#aP;_B?k`CnYJdZPUncw~WW?Pi{T6+alY6xN8 zYC+S#kz|OegJIeSVJLahuwKxW)8K%>k|N%|4S-u|XJ0Hl@WL07!b-QQd4PoDv47pe zByyBR<|TK;2|anY{mxv!IlTNZDoPmjJ3fu%w*DvB&-f>pv;H6xim1uVlE-bnnCt~k zlQ_t))MGn_F|eVfGxySA%}sZa-g=aSN#r0mXuDcjL{cx|R?chFnq`~rs{CB4{1A^8 zh`O1tSwMOuuJLwvCU+q8{YVS-&TNe+wV!oGV{@8?M0#%g;eNgF&7&McJE?7hkGTsZ z1813EWo%#zeTI@cWs4wP|C zeXnn`0IMzAJY7+x05yLJX$F}yK^!i|z0G4VpUW{dd;o@<-dhVnW}FJz>?!rUk5XDC1J+N^d^rbirP!)?<)A|7Ydjl9^m{xIujg!i&$jr(K9{ zTqCu+!7|Sx zds8@MUt?8l#Hy2B8ZKm;u4N|gZ#K+qoom$7h*RPLs2iKumplrwWps&C zTdiS=xTb7HJ1b?|ON=*c0Jll&UlB9$<{ldPn`ZFNE-wjF3Y(VvY1*Oi?Wt(}pbH$j z56GG!vRR+1k(EXmbbEk%P)?Wd5E#=%;}cj*^V;~(eou}}s5y;#NYTE&)3H-$0XXUC zoR^BcUBgTyZ$M5)HY9FX4mrN4KbpJ=NfMVdhE9*Lo`-OZX)nnNUsR_$?fSY(K7MX7 z(c~UxU0NVrIBjO!;u2j~4+X>B!iGKkOGQyLtFSbU&O}JjRakfUZ$g+|@#p5)l=K%4QjB)ZVXdaGm_FN`wYaYmO<9c6HAo}w)JtJNUnd|aBrFvHWY zE9}>R)DE76@n_HwOAf6n=>Dth$GFC9Nj32%owrIif%Sn2mSdWAHHJt4jKr-6K|TIKO`u zBO=GDHet@WO4cfH8S(rperMnkh)zA?XhjchP zB(&E1{+JCN^o%cK4Lz-c2m_m8a;nwwep6pZm*&`Kwx9#?TJk`{>e$=t^%H$FF6_Cp zML9Og4UV}+2U(G-ICNCs;9qa1dwDbC+1U!#47LSUwN1Z)oc@ZvY0u>LmA0+VHFaAgrA*V>oU}@^sYPTT6BjbPPW+fuIvkRv zfu@X54(a~TR_^HEuxHe5aDq;8Lwcw4pkg5O)N~okR{QtF%`fOn1K@BI>bPpPl|b+y zv=+p`uYBAPu{B~4xheAXCqxYpFqgN2FqMi1YHU_i6h$hofuY}L> z;(eX;UTlM_qV|T;+g;Wqn#1@1#Lu(>oDT;RHsa!OK71Hdk>SH zeRm%llx^W?yZt<LJv=`O4Oq?pVK+cs%Gw)aAFvi#R>@;^;CdZW zy}^6?5lK`HJw?_w?nrA`NrnioRgP`8_trYc6p|7>0#zFFZ|o*XF5>GEM|7NG@GT)wew?_+FfVfabqeycjp+ z)tqn8o2`3nHqb$yjmEH?__nYjOw7u!atZ658jhG=o)#NC`x)gGR!;GvRpTFd$ki8z z@x=yI%#5t8>>ob=8P$m_90_|3hO^k*LnaTrV#fpqfmSy^Z>XN;zMPJQgGCvebB10C z@9`+Xs=S>Mh@v&D(hE*raC$52>CpETV)cIEYyB3Jlk8>&`xcYG z?BJpLD=mRjS`J^sF9&ca@b;?D6z->_Sp)_KKZ~w-5DdH5a2$mci0pkG3CnAXHiOlA zCD*CR&!wnZb8yUrZ!RgAS!DN;+}(&J1!P%AzP`ZZIkT+!WsD}az0Gg&bO#jnX!)LR zK7nAj&jdD#C$1eenC@nz^#YNb|Z z!z9C!nh+umF&npZ_0AXgVE7$>#yTVV>7uA8DS|s;gI7Sn>2_F z3}v%GLDao=*o<3+we^D28kIOE4R1Hh#YO;%A&W1Qxvj`fgQ@6X<+nU9ua5RXy~u(A zidEVFkI-d*54n>AlOEoOHnsNiKVt$wyZ_a@4`d{~h@sWP@rfoQFnSF22y5CTmzPPg z^jy3&?7{qH|Ij>$Ut=W|Ze>QpJPzB#vVMCYWNEQwDgtJ4ypz}bEmf6Z{V`I!Rtm86S3)PTO9noyxu7gw!^SS=MJOFDs=GEssW+& zoE3vPYbu^Q@AwEZM~`lid*a{1j@R+W5LC!M<#ex@Um@}SX$$iBsm`fP;` z_8nE}%5bZ&Gw9c=*ZiuLrash0W382XA2p-}K#M5GNeq&Alkqc3xBHn#N+spfK>cY9 z&0~6BS!>b8>-Lc(mdpiKY14?dO>W1WPZ+4bcy5!SVx{eOSF@|5gb&{5Z!bIbX!nv^ z5?t`-WgiDwfKjdavoF>B7lb?~CoSAk+Bu^lyuU2oqI26C>k`kg=m|?4F{TvaZva=N z(E0exG*CKYiZC;OnrjM^Ts{=jZ5-B|2v);}Azg&X-pDI1D{EUjl0v}3!lZuOEo>p3 z;>-6O7&~4a6H)jaGEY?-cN0ojEf^AH%4P{J4WO0(oXdynXRnR#RwSA0#*pN4!ZA|a zb(rwMBym43Xv}Pbo6AqY{O>IQYTlTr<)Okt!wrVnpWg-`r)RQNbV1#(CoNF-|1u-s z7{_b`57Q6@%&4M>C`^4YKb>s&{p%Jg%Dw6?9;3 z?4se_CE?^D?x9Vq6UR1Uab_4A03J{%rOAu`I7H`&0-Kqk*TbJ>=C zcYTVMhBY}2e7AWoNOO+gjSRjb^9@52?F+jPm6hh2Oj$%-JIBzh$9CAv2>*P5Y3sN2 zpq!?VsBhy}J1p5i`XKr*B+h?rWsve1bQ9=?i!x|HqXP!R^l@>tcKGwl7bTsTFv3!- zZ8O%XdM6?j%)Jk*G(h5chw8qWR*2_THU*Y$Kb9iSNbN$Va>hF|t`iUZvlQG6@GR#8 znM0E?t*v0CrE@MPN{~|ECjU~ht&VEz#<{ff@|fNQZ+ZM zd9wAfRgS&2Us2<-y^0)XV$zhpX4hb960Z&;15XKAtjJ!8bX9g`V2}fcLl3xN-Wqr0 zY%7*7i2A2lFs&rebm_q9==O;9mj!i4Yp?t&TR__cA}<4oBT)xUtoaYkky+g_(}K{oPa* zIWj<|i|E_@W{77+ROIuu`o``!+@f3VVnFk__5IQpAH9z$rJnirwXjxhgXW{mM$XKN z(GlvN(>7|(xODU(onw-<2C?oTVS9s=2w#Er47yKbD>owc!Xb-cS`V2e0qr@&rweCk zh%i-hy~$Tno7nK3lFot4@}MTK3c}PA=k7Mi&%Ho){dBPJ6Req1QuEF*uS?UVmS}XH zI!o1}w#f?J?M3oPCUs5(V3Ma8OsXz(u*~KCXAEg7cgW-`anq!bv(?~FfxBoXe(@%} zpv3kcrxG$4#F^Z00oSq`=b%ztB~jX~QKtBAn5)Q|6Wax~rbtZj%`tS+p@&Vv-B#c8 z^}ezQ5Dtl9GnuU~Vep%;n?u8kLT(Cb-*lrS;wg94DolPryOw|#Tf}^Lf^E{2ze8Aa zvG^Z>OG2rBn6NN)q^}CBLn4vgULrJ@9Q3hPacG!Q)@VoB z&LU5-2iFZ|N-)Wn&VugSU;+D{oK7^e>aLC3wFqC)a1p}8MScVMs*kF$v6wB{nwkUq zh9T(R}Epn zh3_p&ZUTlLYE4@qO?}tRqI^~p5@g;WWm7hmicG5P|D91~o^#Z}sArMnBlLabPd&Gj z(YuobbhhV#A~~`LUJ?fyPAz7?bNR^j5wvHMfeAe<|E#sm`i9*nSY}@}KM~;O)J+`d ze^)a#=+-p&hoX21Puqf(hGn20zlUjJsfkXkA}2o4NRkxJEC#CKP&wOD0-f*pO!3VJ zHBJ9opeu>x;YSNBWia$vN}S?sF(cXx*cU~h)#G?HmfyXx*xF1KnL_H zmlpXN`CONP>u$H`OtE(?{6*PR>QQFDZb~z`L-Y1=->! zeXAw{Cb7=N_zirRU@j|cL5inib}XCPPD{=|gMZbtIPg36YEA*7>knRmN33HNzNu&% znJxsZ#kC(|f=INo91=_68#m0wS!{6&(C@Gy#rCyyw~T6cGgVjog*5)8r#qSq8aWB&JkPKlntp&x)OnYh9vK@=>JvqQjbjjW>kr)V%5NZ<7@V-qTrd#70 zd~>^%#7feHGM<2elyPqL&XG~mg+}YKW;F_sGa1$7Tt&o6m2HvbRr4_Xm;wiX2r~se zh%@|Xl<%9MtV>kv35atlhdk%V1l~pCe6sLda~J53Ke&Jcv`KV&eMn6K=Ug6nqjS=3 zsdm*v4cD9dJ_%SXDY-7MI=M;@NRpHU#aq$>FPa?t$(J^!EGog(k|^rcm~VZNM6MBD zAw6Odc^3ooY-og;QY2MG$BbSDlP4%UT@nl8FKE085)p*MWz>0)*P*~9z7L6Dd?QGy zxpk)4dSr39hMWIi=Ie4!d*52^UQr$5j&*KINO8NZ?hu+0A0MHncuH|rtMI~mh@uv; zr+llPyXj7}#*$tAJpcqPH~sTI#-j4O^9~u`hsVVFwS8-j!<%D~edb%E>O&rgR?h>B z`SL88@750{s%66OQDB*R1&8nXVQz1%t&{V0$a%?5BpllPJ$n@U?$IYHW879&Gj2G5 zgb#ult+d@~GV`Jz6FcoP1aeMXALC&NlhC;a@?u<*pS06AdHT>b44)K8{uNGbsC5Br z5O}%K0xV}Tzgb|brUAc=VL1C?;Rsm?urf(uLddb;mo&lZlYI1PvR(d5?nw^=%WUpM zqLSadPYGu)15plOeLL}L@?Mf3AQ1B zY43PHkGM|hcLD>$pZO&JGT*0jx+-s`G?g8vbm&bID<2HjFWXtFn!A@C zOgZOb*nn27BbLquJ`N*hhaUKQA>gr={I<;GcM!h6GXhW~~ z+?$o6EEWX1ir4fCmU}fn#v?r5^`GFJ6j*D!wvOHt{DV3fyL)7xFd63#nx!zIlhsk&foOp}39sRhmqs zdA|KwG!GLwq{D9^q_6|LQF_U$yR57twZpFZc#7=fh*xFX$6j{Y*`T-+z=x&y94?ry zzIq1E)=my1KVHWKB`)v3ntBg~St63pjJg z#LT_a+Cueq^dO9z$^a8cIe+WXCshyFJk0ee4kSElv|^BSczUI08|HkQZffm*t2u<~|NL~u{3qXA`~+>Uzb)ou?fq>))#r%p zAqL&UxlxMYQ15jdAkfPP=h!w#QWBd2BmS~NbHQ!f0K>6IM?RrjhZ}=G*x>b~c?Qgj zUk#4>t~)&y->vDpm8-a_83ikvf1itg!hY;$KFLQDSqoYSB9Jt0OxcCik!;!aDlCLc z*7O)#VDog$Z=!-;Xz`JwYN2heB@?5up6A{7>#{YUzm{Tuafs_2!1N`lu6c zuo!-7i#mJlwc3SwM;#QtXxF*eU^g6=IXL%FTV}mtJj)@Z?ULU??$k5I2#x%-!7rL- zLW;CQ<4@fQi-qQr@3&AaJ;IbKM|LUJ8Y^rfZ*IMDAN#0~TqV$k`Lcm${d^9XW=3s` zcE#<7n%!RBZs^yJlQFtG)LLtJQfPMiA%=oSM7Zz}EhhyXir$V)@0jJE*SotlOz=`} zQ0qSR8CwP;`ed_PQXMM)bB1BKH;o25f+sIlYiCUmnkz*Hbx zAO0P6ls!|cC5uhsi)9($W*_6q&-3UY!RHa#4PfGxevPm|y7*)#tTd?tZsA*1h zmyvA8%LFdEWio_%1duaS+Fo4NaX^MptG_QE&3Q5p?;6o%h%1NwXvjsQJ;5SuwV$9# zMOtk;%;OAn3ghH5Y8cxM3ey;AX=Rp%xqpvx{cQKVw~o-C&&ir&m6jPbNNKy5WN9V- zt9B1zV~$S$@hxJ)vY&|LWWCPqMh-Mkn6=IG ziJU!lk?c0s*=0!BZUY$EIPJ%P6;zbgcl$-Z4M(adYB1Kkk_MvXED_e6u?O&-1$VERy<<2!SS z;;QX(BEA;m5_Rds2;|ctSmWv+{yMpl#)omrEI?>=gLP7H=9S`tU$KZIt@Ds#At_#> z*MH5AGw|a(#a(1|7r{9P0gC=x`BB+$rvq63pFC2hZuft+UO|W4IY4&Nctpc&q=@L? zuwz!ds9l^iMlFF0qsN5s9;03yR&Cteo%?t4cbH%31P+v*(l($cwn+gDua^jRh)w=& zogUVrh}LY**S~B7(|_j*SJF-EjLv$A&e6E!NHc=ZUu7Eaiy&$=Wa6o502jGWksSvB zka*UgmkoPf%yupa82;!}$%H4QcoJ6PAWylCh% z#mVTm`uBb1uxlp7e{Y$XY^6eAN%{r|EIPYu|58xuI+=ShJzsbJH9fgd#*scL()2&_ zEEsde=T-BC+*?p;;R90SB3py^T`9pa@}un_(6gW)d?U=%~)BK zs**sS^PVxr-(LnJT9hHq{BUyfY+me29>JA02hmb5f8ddA&YKB4u=XfMLKnhoT@cnm z#w|%n616VkH-m$G9E*o{PH%^QSy^Ojej5l9Vmz1?X#T}lQ!5`$#;`qGVG*u`^UQh_ z!{i?64Ogg=6Oo){VtQD+1Af`#m>X#^mgy!wFJU^|_#r+Ue$F}k{+`c$42K&%DgBJ1 z#gvIr6rLPs)ViBOG1wP&=g7L;IEyeWVO?bANzyQ-%JK49?4#@hEcLi(uLL2qlLfEu z^BU`aNvh`J@UJ49%fS1Axi&K5WT|VpLVV7Bj~*PrhrDT}n}g}>0qgBst5a@X8tl$O ztJoebla4KQja1&pesN&`%(>Q~5>k&%4xhl3#X>RwE2;LgdLOkv8>{CbqEKNO)IsXU zh0Tp(9DWK8BEu;Q1UYe&Sjd~p^~S+fSdVV33uTZ1s!>CBeJ});3efkGrxIZTdmDtH zN#e@HxK5E18d6O~p0TL)k}Bmq5_A(-b?@pqE_KR(q!m`_cf-$I7iD<ZH;wxnJV5m zHhYuAE4;*%foh9cek94YaF-8OSffP2=uyo!{67`Kon7)}2%JY#J+M89%Qx$b> z<)pFWwdL7hTF>=iIo=Ir zn>3NrGF6$66VBx)y=MCPZ}uO2aXv@5mzdn~U8S^^sMRba+gRJAw(1Gk^snX=)NJ0h zu8OxQD;6$G%zqS-prkp9+meEbd{q6>Ad8jc4})*ZOJ7@;M|lQGI;K9E?9(sMtF(l1 z^DGd11rBT4*A3Fw9D?oVAoNE}NjG^|(ZUjbNDKnD?+aMt#94>is(bA&;=!T_XtgA1 zC0)`?!2$Q7PO)kpbmK(1-$a~ZngRfvh}P2_P<7HeDHywxJv4GWZ}}t~;;>*w ztX`|%bFG?Y+j114`#-d~N)ZrsQw-PKzQ^A(N%FGs#Y-y$!>rKSgt3VZC0MdWIQ1NM z&Yi-%k?;kUYRQr~;F}>$=e(FZzP}Ddi=qo_MyGb(NBwvflliK4MI%rtz{MMjO-7kO<2~ z=r}W7NSY=F)1)Sl+h9vP8tlUu23s)%k8w`CZrwjhSx#_F3Ri<;C{&=H>MhZB8k z5v*VE666;bR#+|*Jdc#y%WhBF4b+Z*k|C3v_1tFZsT&ZBN9H^h{TNsA5karuAaX@8 zSXnPofZsD4&5@N`iKzD_%U}9W4Ydbql4_G6L234EJQ@Wrcsk1O+aS0_OuOhxt&~Ht zXxL-sP5G92wyNQfc#Tqs%KD$b0^>UxxcwrUBA5JI zQeB_K%oVvJv?u52k`TjTo;PfzfbBmk1*{Q(TlyiK9zspPqp*{dybiX3YT6)<1(;r? zlFRjQpB&M7e%^Za*u!IfZ#!=XikUc}vI`N1&kA&9ES2krKQ zXX<@lN7ksuaoz+?8JUf}y?@(L8|b7N=c)p*-!eESicDeJEIDAhfP6~0qb;AE`vY$J zZWbZ@vel=hx|jhKlk?*77*tNQD>iNDg`n+IW}^=-*U0 zB5pHhZqT1yF$?sH$ty}~57_BpvnP3Y=K%4ay+3p0-3~c>+hGDU>3n2JMMpyP*J=Ds z;!k)$7qCK>E=c8(t=0ivteS@yB|iZ}kBizdM^t;%g3h*bqjC-%{qd9nRWyHzW|Ro@)3NRH=0!sKpP$) zwW(F$UEyDL`oud|FH8_M$iw=1p*V%s8%Zv&o)O0R8TB9hZ-}D5kx`E++etkQzS38e z94lLk9zv*>wX8C&#oFsC{edZbNfc;nJ~>H^VKR)kWO9xl^l?Z{Pel_0aQ8SLTgUlMu}N*QlIP5YpYQ&|VguXrN+apnaMG1mWNDz>ELdKGf^Yq8u( z7R?K{; zE;G03(cdO0|0+pJVB!%?9FBL-z3EAy<#C?#3wHTm-k{gXa$=3r}?;DQ>eOCBh z9Y>QT)m5%-fm9BY9)mL9fZj&!dCbVXwJ4y~&rQ2?`<+p#%NlN9C$4)J9g5YHr z!2c25QrjXPwn-n-?XRsKo!p`8W2tx=h}FbqP<#hRYU6N8EIiM z^beA&kZ^rRWI0t4xN`gjDOvZoxc>4x!x=lb5+oHy@QDILynhP}$<>@3F;Z$w(?;2E z|0IuA(82zLd=7~O8cdDi@Tn=n2jH~ZITVVzmvLJuX z^U9Zm?)Gm&7p3kXtpz)jYPp}0GReX-WJiWo#ZDvN3D8zTto3cjV&pTJHd)ITgJ-NlWLH2T^`f616!KpJ!x#dQQ%}n{5M)7sy?Q>)htamI|T7*)1`oLW-MUH;Y!X)N$ zdvQzzTggG_zm3;Kfsk7ZEBiG>JPok9&7dL=*UNF8)^gAmtf_$TRBHz*Sc?|a+ACYC zn*B_XZ*%XWlbadKd#CcP1Gg+fGrAef*bqxJjGA=s=s!(ph$iQ9(vcLyHTNZ{FBcuZ zJyS`Pe0mglL2mT9jdTy@p-R-#g+&=?>t-LA#Tn(dA69)?6KX{C@0B6PPz4hyDVs@|_EW5C-GpdHdh4cMTINCQ7puJ0rZlJF!%W)=>>bz|ta`EH zMM?8cX0LCuN1WST+4^b;@q$uSQI4-rl|I`KYk*;y+tEVlI4XO5DQj-D-=8b~um6-j z{XgE`1E}e?O&_(q_QI>af`WjGf=IJKkQx;Q73sZ3K{^2`0Vzprh>CznlP=OjiL@XA z5)~C8bVvdT0Tn|@2p}N|=_kD3e!FM)+c|swduGl|CX*qB-|s2ceck1`?h!J zV!8YvWPxlmHLq_B4-oDY;M>mOpQcUG0OcKvDzlDuH_u*ua){bui-h^rgqW?)-wZ6z z@wI0TOjTLCWdR*}R#nb|+g71Grb}RPg0%gJ<#~(p@7^>j($#X!PXNsEd2@5$P?p!Y zd8-Nb={a3`P2-)ND{m*APS^fXwikE@4p~jx7Dt)0vR3QpHiNh!s$<_zF+2@i6k!r6 z-vsu76(!O~^_j#W@6lCP!sCgYuKPYiqQN~{*Q1_gCQh7L5u|`OuMiGqXw_bMV;a=D zpAkW5$E9Z-YO8Osz3Kc@86}qCglnt^!=H5`fnH;<$ybs!L`%n{lyj^#-f@QwPc*% zq_2xqn_TY`7-|&e7&WL%2XMXvz=Du1IyMto_^Giov^3;T;8$Z1-00Jbsx$q4=U_4| z$>`^T^ytFfi;^^98Y;_cn+s~977Df{5A*Zb(8G;p?C z8^@ldsdlP!@^(FdOqYwB?WtN7E+mQuhiL8uF zujxB&NfPCYuIL!5BziF(5V7RVcPA=x<#CrLDm=V-54C}Gt|#G<^?;0LWn*q|L-+ni zk2*{Tc1Z`A?>~^fvY)ZxE8STpWY1P8%CXKaqF*m=MrZkTOv1V0(^Nzny+@*w`pr6q zm^RGY@y@KsfUbVt&-8hN=Ky3V1jn*IWhMl{E$Oqp<<>>4CtDMmycY@`rj$*yRNeNP zOH^`WwC)b-jjyGgRzIeX(>(JnX+*n#wlY(`Zqh4aR%IyQyi~b5*ea6ck zM!1}ihdW=BKPtGs>BPlNU`faCelvwQTQK@JZ?bz+$FvQ&nGu5M3RJ7c=z`K!A(*Rg zZte|uwk``VO88@SHhJ`PP;lj{r7x>zSL)$1Z^b8)WdAjJ*yuCbQxhp`Gx@NW13Qb- zB%M?)o_yN=Rn|RfCe3@~?lhUjwRaTLXJT+ZfF`ti+XaUVPcP4VT z-wjbyM8=Y51U?-*M&SGXkDvMeC>3v-2lkVk`(#j69woAlKo+`bb#c6^wovS)+EHlC zPo)P>{;V_u4ZeJO{!d9K4Y<)p$)Q=0)Z-13i&oa(G%By5WnT5&T>&RsMnRYL+`WEx+fx)P) zx#qs6E}8Xv|E~11rSCWQqu;-c#l=m@D)+TogX{b1c)PZYRN^%CoNCeGoWK)z-{Li( zOSY0D^*Mn^Yzp9wSPd*1EeibUy7nZ(O(TD!<#IkY^8T{Qj*M#?Dnvg^2Zp@AR!a18 zTs!!a8z6+kD%Bp*lbVNmp5%p8KiMl^ZhViEIcn$ksBn49j$bFNQ!z*V{_-zapV7D5 ztv~ARl#`cCSJSb-$Ei`0P575_ay~_lN;EoX?4~VT9jW>v?cv;aj(L+N5gWB#8{O0E z8dO)V*|z4+G&bMVx;-OdKSMISg1jAqbs72JA!>PBPMp#kFdTV7`vHqS0qgN6%X74g zjAN(aR&iDnVqNAPlBDO!s~w-4^tz2#H2&Fx=VHnBTkUth?dz)ErVu~+VzadSEo0_k zW~k|l(Hs7Xr%shy81!k=e(s8`+Cu=f*$m~*b^e%er1yCBgHUyXTfxejn?l>UD3T;J zVxjoRxPol!;ed?`bx60+uhK1^{+ev(A?@Y~ZvTO%4JQr!CjBtkU%Bl;?CKWQQx)Nf z^;S38k_Isz#|0LfPMoRV_h!*QM(@U!*%fEP+ULuzj~)Rf19ZM8BlRy z>`Y8T+-ueH*OwulcVW$3U%4tFCm{EGZ|-}EPK0#aR)`uVVEKT{{$KdiVIQf(W-- zEfYQXIB(_g+1v};-*mK9axw#bp>y4*)sU4bh6@`S0e{|*2Ml_p1d6x@W&J)0g6|Rg zpPsai0{tl`g;tV`$^*kc2(F>KbYfTiS)^%x4e#jz})icu_#aYXUB>+8VA zLt~Tw=`E4b-#_fxB4NCKSLMo(RPK$U%J04B{_hj^fB0H_T+#|kQ>*b`Q2GDcy-P|_ z46}5M>t7PSf0z2-_hWipxx+fTseM6WE@RY#;=PX5e*CET@3{Sc{Njww-wSix{J(<7 z|1|RJ3;kaox85pWqztVzWbf>!*4O-gme90y8~lGaZtLIraK+;9O{*mdC3(B7FY7&f zMgRMDVGo|=Zh)BKaGs|xWOp@uRgoT+JtDaD-|g}LcT47g&PG~FR(6MExup6eTd9}o zTv&Vb-wn&Z4%C157v;deH(lKSe+%!yADcR#>~T^uOSUbT$#Glve)@t%BEBLm-D(>` z;{MobjQU*C8NZXR)7ZJ~zb-Cot4+Wc3MG=VErf4QjIxe6zJPZ4FQM|fdHR>f1AzJ8 zQz`$Srt(4WAI73PZ(`>&iKEANMcWuTdZ;!UxZ2YwoUN6*ld%%7Bs{G7(+W=7^d~MU zZ2D^zJ*3^DPkVZIAh}jo>xgNN!tNuhIkBQI;;Mw^VIs}Qv75>|?C0@jU4tD$Qthfv z1~?Pn3x9VNjTUfL?HCnBo{V>tEC;dV17VO@Boj|?3qe?3U@y}mCT|{HX0(*pV-pqZ zcc>1U}#z zIOxrE#LTqH;3au#L^E-rSV9ai?fo6>B8f1W&K&btpoIq!J*RJW{`a5nsi?NmP^`Ev ziV|R;BRTJHVYnmci)xJIm|q}vbE-oJFBFfp)*9%?f!=9)1YXX^roujlc)4HxBnT69(ulW(YX zR%vZ4avvEMMy1t98GS8h^*#8yHt9;5(9W6RE1FgdrSxq(B+@GmX9kQ|Z)n3Y8;vRe zVFQ_5n%BWip(4m1J&48t$RcPPs(*(s27mZW$A9T)=Wi-m7JV@$R4?t0 zUqi9}ynp=b#oEXV5>l75V?MTPfgWmFxlak>5YeZTDDfxu_C0KqNW{_}r&~HnTAeiO zFD(@IO)@XN{@~-ZYBTG3_nH1(amT1c{dOPf_3vaYEf}8>jrXaizPLOcVP*VssN$=0 z8*>pxo13--JSWN&zZ2cwB4Hi9oH2QPg$CdKSMAY2cd)M0RO>4d10RJ+M{uAfBr zrHmcrLeRB4crOZ7DvhGZs`|3q6vCTg=h9S0sT~6jd7&tqCLpS_GsW-{g}8JZ@_S3i zD8LQui$##3K^jXC^SirZCcv!<60K0U^}2%xnLFdQ>&wrn0}FF!y+gSVDzt)*lWpYM4Y^e=`}V`J%V8#cq+en4LWUi z%({F+6P6Xm7>!X)uxNP%WQ1b;Uw)Bn?}v9;O5NUW=?`%j{3FJGOV7G20`CB;`iz# zp9Cv7N#Q{QE!ednDDIAH^oyL(S_AkYga3p}uMhogOWpsIEvhPC>o7!jA~2(v=n}-r-M^{K^Au+?n$s7UK+UR(aY|Kmp6EIbwB#NE+`DM> z@af9=((i`8y{R-wfCiz*+M2kTO;PFz4a-a{>VHR=Brp1CfY#R9%&Exn}hHn-ShFkKtUiE}uOo1IpcV zVeIM1D`krmCaWSjCDc6Yw$0rq=Xe1X`7d9u>a6ZqWA+bKaXRwOt4?3f36DGcdmqqCl5zdI=Vx>dkua#3m7~DJhyM{DZeh+LU-xD- ziCPuBM`Ed#O5_y6=**KlxoJvfRX%OV0DuP@KD|W;*LxoS#_F_;j0PJUFGcw=~XL?)U4pHap|C8+*JiT4Qfj$S%Nk|`c)XoYohY#pEv5fzA z)+g2sd+klgAD;1C{NECi!EMupkX@bZwR~n=-m)fU^)x?>vuMILb~-k>q>xU^Wr4?QqB!PK((oHtCtu-I+No^qiWlX+NDVsc2^i3!v_qwn8K zZo*dMzNZbph0pElk^`M{ z78NrcpX4;St)(e#?a8nASt*L2Lc7lisM(+#-g!wH0=4xo{Do3#p4CjXen^>aa#pLW z>Q+HjSivNl_%DpHzOcQtKXCfJBc*R|;`@j4Xfqc#^X<^>I_Lgx1o;EnH%% zz5W>fQw@bg;eO8-DO*tC=K0_?_0Fu2=&z9eLZMG z6-&rUGh;IHUY0b7Q`GdYv`$^yJ#R6vBA(rtylr4=E4O z)79-Y3Ob@xxZ(WF+V=%+ebrK4%J*QZe@~5Iz?fX~ar(-90z1Kd*5z(@d%HU`X_t=M z-3CFoBx38gy?d6i{+901Uz! z7)Yz+EhM8)10F_H?zQRQ1v*+b=EcN#vc|Y&lSe*|r^Fd33s*b=@0wJ)Q4T(D18s#R z-riNI?Pqgm3tB|_MIC`I3?TlP%f*X}HVR=CfzbM;U~y$HIc%TYM?FU@FMxqnv>nZh(p>11ff;^atX& zRAhH}Zbf&!wCqW9di>lEvj)My_yBe|NI${!E{vDs*SI3~_uQMSz##x^I(`xTsPUtj z(Q0n8eeV6#>$4wb#jHrKOYdC}#NN|{SY0mN4A1E_#Sb-{}5v`3B)nSlE<~|xlu?Xw^q17- zP~pBYP2k4%N+EkO+KM`mi*QawJw~6#w zu$Ms_fVBYD3v9mFxCzwLRR+XPR*aITOEdz>wsmYiWut?Jn}A+UijFlyf0@bsq#Gf15f6Y|(Aik|w3xZc6qcl!nv`b_Zcq6-Cu(1PL6cC!r||M>IbKb{^8shn9D8yaf)V$wSVd+~k3W;dgJBi{feXljWdE zj4n~2*l$)l^i4j0*JC6J{<(rD8X*8^@2z`7DB7RJ&7&rPZ=@O@$^N09! z87GgCkyEv0MtZsP!4aCnWRY4MyP?0}djs#xuDdkw zyWd$x$&*Bord;c(xY8J{wtf5D`w2Oa($y$36bs84s?5s`X`d&+T(inMvdlJEYDB}n z?v>D#Io_SXbS%6uex+(MRxV^HF9P$0>C-G!gNCyvHCiTbH>Tbb5_f5>MsdGRc1ZO% zI>Vv*rkEyfREPS}u+IVq#WlXWPr!Jp2>GJ+-3p>32o8l}-Vb}kby)18yZZ_IP7`JH zw<1!saI<@`9eThhFle4NQe8*Eg=DSLOz-yNhLqh{9!)j*Y%%*-hx%ZT;}D$1Nq{=* zJdBXuK0;o)|0r&k78{E;r4P<&as(o&tSi$*oF*2*>2@Hpys<`MpM`ooyP#Hoo0aeQG1n`;@V zz<=AWz&FqiY>=0cD13jdzr%#xrf9Ji(R@$y27FMIT}ey#3|X31b~W3*J{pfFvb~fp zblq(JSvuPYEmE*5z(twKD{)@0=Jei--<$7-&o{N)e06pQV+40ccOF8^8O_U(*R#y>aU0KuBQ!oLG z6m$#21NfIAqK}E9H`$cnUB2_t(UkOTy=EhDPOFUb+vve#W`E2Q6Tu`%*0s5IS0XCd zb!kkW=4nExXV%D@`2NHeM)u4b$?X^uPfX(44aO-Hglj4le>n~{$E6j$*(g+A4}wTvW(t0VrJ{{|2!>KV1O?6rl2H z$$mx9x{3(2ykFpffI#$=Aj0w}X>s0?~Jpeyuw zbxNiY+xTc`<2azd@x!YaXgSJMVpQ@xb?q7H%EUQTyO>m(GEcj(n#uqOEx;UVqOo8I z);fZ*8Q<;!k_yjn0h{n=s2p&MRaa4_UdPuua{(&@ch3V9bR&+$M#7W5$;dhz#GzU!1_zwYPt8u|@tFC|$x9GJ1?h@HO zXq3w2$%XXbN*QV$Fj^z%8+kJd*aK}vc|1Y|EEv*}4mpx9 z*e`lRlBj;pd<(wi$Qz%gKFFM0&@cGE+YyOHIaIr;^SQiyE!m(qKH2Vm1fX>R^7OVO z3RjxAF9A99k|6l0%&n}IJr?!lU>45QBVQ8!>>0oIr&ODR6E|a|vY4tG_qYkSDQYEE z<~7{2<@+XPE4{%-n@7KLprez49a!Z0qebC7Sh__^m~54al2+~7Q0jZ2XdafmWqGC% z%r4tQ*~n)sY7Hx1x|57|48wNJ6BdX9Ewm(cc|vi=+MCd&^wj0ITN*4~vMI~o%n11H z4{@3ur8s?=36JQhbg{mW8B{|{(f|4VGRXmcXXClWIdAx@RM~f*6J;PtQ2*N;Yryb; zgOjkQHCQ{R2z2j>oBu~H|N3Zh@6(|Ay-$3$_WVovh6i3mlPC{<2>wB;fUO4-BR z9w4jA7~VCo5L#V3Y8tcF-@*=$TVyig?+{Xb_)E{Qz*7+{c+~eRsQskJR_i9KwG_21Mrv?w+!Wnlk&cRw~L~}n)!kt7bPK!O& zb=_uT4XI#g&vZMpw*Vf@pJi6V0dmQFJwlGQGGW zQ}pkOjA?Tm{g6N)1L!fc9^L*}t@eGq@$(sQ3neGJ3UA!<++aRqWJTE4r5nX|%28Z$ zsqlB4nLLyXO`vA?p?;i#AksZYoehoI46~CDod)zDLGWXHp>39Rj z*@8%cO$as9@!gjdCy7Ojv)>l*9KYpQwz|KbVy$P);0deDZ5C6)ehhc6T*f2375A`_ z@@bdGcTMcC1%ZJWm^t+&Ti|T$HR~f&l$p7-MTj(OAK1igOfv#S%zE6tE-H|>YJ`2? zyr^anj4uKyVS?M|QH#2|6R(L}bItmD#^~N=EasS2m+Y7MpG+$pa{^R3lWdCi@UBfh zHw~Qa%Ec!TJveTh8vy;)egkk`QM5hKBmCpMaMY_DOgmg4Q8Dr;7M8vk!%VX1259dT zqhoRmxG(3&7hp#64OV;G3~B1UdgZ72wF}hG-li{1BP4Bk(dvH!rW6XWSuTOk7A7E} zi|>BzPogR@d@oflj;5zZDB@DA@@kAR!t$VpO1W-@f|Uv%6I|Jmpe&K4K`_HoG=4aQ z6i^=46B;#gs#>ct{AKdj(lp_K5%U)*trMIV{KqV#g>oDjZ7=PxxBT@_k42)(`0#e4 z5kDmT%2diq96rvGL!W59OemD>vH?HomKm^A zfs_@0K6RIpWkOHEFvW{r;4W>Sjr3X!KJmq;A!32T5gaqv3M+!6xaD%_P=;HuoUd-~ z1j=CHRqEv$z_)Nk7g9h|VKuT<6G_K(hzpfSZ@^@TYs?LDgGEPr=&e3DrrERyVKCtP zVACSoU~xApiQqmR5rXNZzj-7jMPS<9v05I37lpDTG8%-gV<1bSGzmB^^=hTUW-hMz zOG6RrN*50CesuxM$9JYT8yI6& zb}nWdNKuU^&PZieFKpYvjls@u?~f8xGnVr}9>=x0aPkVy+mWWgI1FHr;1o?c^l9Cd4L^1s*kSF$$|gr6Dels*e;7A?d&t$fZ`#nv zd|I~$!^SfNzaG*Xp&7O<^obgkN0k=UbXjaID;#in#j_1gk-z)Bn9D*<5l-JhtS08n z>Hk<(Jd$44+N7DyWTN^#(7uzbAOmzQimZq2AGW(|UpLAgE)4>-uBG>H47x*92@IvF zTJ}`*l=Mtu8&p&(`@(+20O_eP3M9@s=I}4si@1u>^RAlORmMh3XMVE(H(U;y-5$Cx5+XYTeU6Ce{Uha&d?WF-n#@9Xw!8^bQ*y<>%3{kf-&9H)TS@a^=K`~820mvVdApNfUb2) zHH#r}pk9=>Gk6;H53*F7bdFRmT1i!$RJy-m)ApnPc=|B+*CEq|xw(nq-!Iin?BiZOiA1n8#HRDQ(dRc|IbOhgypnw!`$;MLhy2QZ6c)@=Dl& zn3_k$z?Ev#vcPlWPm^Vj-6(0~+7UjcrT+P5pN3#xg+@{RzLQ?qUnmeFh_8L_nJM7i z^%B6sN-{KDCi#Vn!oDHcGo+7{;Tpe#=*gejKE`9AS1OmZW2=DySpU&O&ZOJh9sYZ9pwiutwX#~Xg-0D6gEOFzOguq#jU%(Sy&hVZ zrW~n0UK;Nod5dZyjudc}HW{xjgZN6JSq!^Wr254+pslI&%R&B95Yi=T#fBeza0#KJ;;R3ZOYq?+ znnALt#KuQ(ov}C5Prd^1suT)xnrz&#YDfX2#JL);u!t9dO+dMEd!;^py6HwcSeCP| zk9b-zzm3<<&^t+Wzti=EP?2`0;r>UM=cw-*FC>l{lK~mIzUasaI&p+~-fTr=wR0VI>?uSrgrsoTbW2 zp*9L~V+$ShA&j@^@FKMEG>24lIsztV7Q^dQT*9SZf;z!S<{q$Gi5WZJ3<7(A`{7lP ztChRScADBdUHpf+H^_2w7V1qm=NfmMeb~5CYqQF_lQ3rAGPRIj)kJE-+Tk!*%oxNw z*lmG{0A$bX$lUDa_-_$byw!N>rBxfaZJWJm;I1{IPwvMo<^Zb-(ngb1&&N)?nh=de zG4@~$@e%isAl7-gd)_gl-}>O_C6VZetjB<91c5yt`>{y-=NoVfsSz+A`^&q?Swn0| z{3K$%oD7-@)DyQGykIXMC2puL!`p z!~lr}Uh9sty3H@B(rLvTxTR@)JWa~hz*8K__->U;G+ku#oe%_2w8Uo zTUr-$&B>vKqI!QTt&y1O_~3xUi~|fgk5t`)U^01Bt7_dr?kVAI6Yx!F2O1m7oRWh> z(tC<|FIh3Ish@ftVC;lHfEG5UQJzC{M6}E=D_?hrmQr)-$nYXf5EMm?vb|AMK<(3V z%fcLy#8&sv^*;}b)dzb*Ti&1ELfvUv{yq|LZiSFw-e}?zRlYIvI zJXE;g*EJ$n#uKBbuay{r+x_bo}>in%!Dg<@H zzP~nNz<7k-DFu85ue%5WejlE(Q9)4ryG{RUOjQGci$d%BER|+Pj2(?vk?mNW#x3pi zeV;^kJ{Eyfk&@#bIv^ri4yHF2PpaF74noZ9lB5t#{>gjp|R*& zFU@N-E5qy0N$8xdWWhSwfnbUT; z6-Vy-Gc!OA1Z^zs(3Z(;l`TKiTYjGA8dBfny$-lxWakw9g`F->+<9cD#gy^#pyASZ z`o^6gdg2LE$d>K|WrKcgD{g)eD{O9}xJlg#{c+K%m};A%V-94SNN)kCP1Q5s7OBLA zesn;qvc{(2jj4Pj0t;;Iag9L(u?sT70hXz*c-5pL{N>}mjO6cmS{2@uB3e%zV81m? z0iS$P>|Py}jf^jmV5buZz!j!H!d^6HDjDdET^~Peq+uj*#0O%c4!0J@7v^K zRCiW7eQ8Np0_00p8NaVT2TY!-@|hwyK`0lzlx`@FHn(2(h3&6z2*iEq=b$yyTWMF| z4OBry2R4;7${-cFcKtL*+OLJ>)>;u;)N0xTjjdX1eVC6uHBway<7rQk2Orl!PTwbp zFXi|7gWFbG-GISE68Q#^Q=EH1T`T)*)=d!!5_j)q(EZWa@^_ae6-CCW^4V1^6z#8= zCo~(MkEDp)iX>De*t|6)a2-Aodn%$H@$xUH!+zB;MzETj^BMVv&{+<~UeH^`h9o}- z9j~ODU6zHl$Qe|Ef}P7 zu-Lywli8wjr)G)HJ0j5TS!djzSE`IX7@POhdgP ziN(Rr)YH9Q@(%SMLi*arr44@Jd{cP~T0u%<_Q}eA@_y8du5K0BRNjubP?~-vIm>vA`S;hiZ(x%b=b{CdJF92oOixs6+%R-AhiHE__~y)be&VnX$rmHZEC zH^HIbiuT**Ba&)E60%5PE|{v@mpZ8X0)#%)T1M-7b_`f&;K)1g8`@xYZU(1?8tvMt z^oh%iWQn$!iB5r>^p+OB3R9UW#HbeufNhBp3X++#h`x_v$8|~ zoGEelDKVhWybal}T|f)@B`=$C@YrU<%*~tr5t)IhA2rq8n;DusHWJyeVYaqK zEhTF~3(0&n>>yDi-Fcbdhd_Fz%d4QD(XAj<4<@2l6)N8PGFiEHXb(emgQ)T&;5ZgL z^-pBW8xd`PEn?HRzCc+p0Xa2$iub)C-7F4wYHshx$pyFj6t=ANT`o9~;U~O%l)z?= zSOyTwP#ShLc_326A7$1IR|d)8Gx?*87WTEg2+MSVE7Y!TK)a!&=kf@1>L}!D&2Bfh z*|+AJoM@1!^B~oNA2!&%DL>N_kKK%6c8CxQYg$ux7y~amc7W40Dx$0?I@TU~jS;?N zc68u)3rhzawpNwMgfVFgu*&;evvALG4@l;K^^eqdv z^iGKxpFwto74|G?r1I)rV}`?u8P78}8o&^BDm?+DFl5?_pu3-GLEj7?ALNe)-M>W3 zu{K`KVK5QX=b9%8=~_*xmw~dwXx8REajpqdqkcQ`U{>m25JiGf4iN!lqHIqI)Dd5Z64Dj1C~!T@!kh?iPP-%gP_-WZflbKg z3BTKkBZKF=od7OV=B=q!fkMcvp8wx~{Y~$rKgh@~ zd?wT?HU&;)a?-(0GGN)->V8Q!q$UL%N4ZUQf`P-^nrh-0;}p{J(wn zuSfsRbIgL)^BlIk1@%zac*Ck04)PdJ0WT$uN9`^F2U9T?{~&erxBnBRd%9uvn4VP0 zOT^}{3I>QbHz-Euqm?%_D)ZBu1Y%r9{x5;(o7CNl(%5|QUOl2LRn%Xdc~8IM2K~ZMs!xlLUX|V zt2IYdJ(BkxlK~D+nX>TL{OTKgl<--#UNOy3&%MJ#0_pEHX&WT0Q60ja@h z%$V4nqURQ*klMz0yG~QW=$@?C$svY;is7_U$D^&AR7?tfi5luT zu9jWB&noh>QfttXobhUwaIS(!A57gO=zb`-TG>8UnkUUmOE5jPYIbiAgYN8M?T|Q1 z=^X)#1N>L*^Hw+SX;r=UuFpRDfa!y?H#iwZ1&1vNAq#=G_%YS5KZqSq;;(RHu;T{oYD0MBn#yrlg}Z0zYcHDG{|!TR|ml$9-i2xc3ff1 z(R|b8v{uo4qNga5ba-|ZI}#rVL{Rq2%8zCRjnv7Nf^%$_oE31$s#%vZJEA7yG^dIj z%1B%BccU_P3O%x?5WE>BHJGxNR5242V=uf!d+vPLQO=CIigSaOy$etjugZaI5K*zW zA3O)De{ccb8%pMn0kP1@Uu#leyL760HH5Db)!#&A$p*qJvV_WXCQ&{XM4Z`~!=3-! z6G#aM4A*b^C{S&3Plx`6ATwOulw`ju@4BJZ((J{gGp!ZCuw8fAWFWi1Q)jq5MX?!s zIlZu;|vUM=(SZz>af=6|Nf^8|P+k{Ch&f~0FbjZ61uN#Wa{vK?X8}jL~k^rU$L}3?8 z7tb)K+fDfL*Q8{dJ9d(Y#;Tkj&3a1}sNz0@Ro511l@H#XwA#$$29ehRsYRZK>H`&Z zwA?b^(n(rnQ6>*vbs+AFvZ$mR<}H}5ic-y39(W$Tf1uvGlMui?1^N#}qK^c9b1rhi z%kwaSfpN9_a24_(S*jsmTq+g^eNfM6h$J-yG~FyPt8{mk&kLs%L!!dkPs!(bk=-GE zKDe)jtj+S2r9lAi0Va+Scgs`Pkz)3)i6N=>IXj)fbgr3jfvJCJomgyMs;yN4=G7co z6JBH&bkMp6TphtznsGuMAK0GIf(9!RO7|Ry2HkE?_fI#56bSB z4c-ahbL(P3uqL-t7SQseSE+tERI$ujWhDN3x#bp!^_rcBvA4WDA#FO<_h&^l2VGwc zQ5z%`)s*xoo*fo@E-}&?WHUc2$MxRA_Zb+?Skm1g$*kad0iG~F^@nQj@{A%xel)k% zqIFkor#RSoqqOXNBf^k7^~F?~$*kg_2HJMsMR^~kuGwxVL=_{{kA(dSlXzm@lRasn zf5t7P_Si(%WOQifzShSw%yy#oCIiZwNH=$`HJjhF|xowo~G1qg--+^Dac@^wfM zeABN6#er<%G!4A-Z`h&QMxVE$k8EEeDZ_#DGnK}>GIem?bCPnj7xFUHqlw|J?CG3v zs7s+0oP4d8DB|%3iCjO4JaQbyh}xiV9)B53&q1TrP(_#dq^XzI_ZyS%y4U#$sjHI@ z;XRWOR+zd~J(Z*oRKKbrZgAqNOz#AinL z@hkX?11(omH3_SdlGs@|IXM=Wz9rBd!fNJn+(P*Hco=^mr2ruqj~1}|(ov$eUtd>U%^la?E;iMY?Sr|JP>j1@Yvkcgh zjvHZcJeLn(=o=t;0ZtuSEgEEkfzmnt%r;+-<`8~Aa3JtEI?BeO)rtahtE;oz1rGIq zRnz3;DC&_SwN!26YaaX5I%8K7dumAb=rlw(4I552YGPOzH|m=b11&0)yWswLx$1r~ zmE_R+h;j`i(fWQw;CN_*F`QD&fq?7Q>r85D2pTeN^+#Omx{Y_@83Y1e>V{XkDciQg zHYDiF3~^TCP6<>W$h$f|@YSDnm9KCrz~V(yAzdCT1GtU-u5~!}yDb-5I<_7vEgS+o zE=NN{SjZ&@)?~liPIr8Kb*)SMP76IL_M=Hn(%X7)l>yAo z2ZKf!)%sf*24eK*o_hu6r>s;N0c0#}%9DahAP9MMW=dCo z>&H4of%yxUW2$dVp`znG&73I|HpZC}<;-nnf+^L_$>3HL#M}Xf9t-U}T|%0z=iz!b zBmYytOMlCgg9iHgEm7}VkAH}{hUwHSrTG*5m12Lj7>V2pd}TF;l;!qCCV`_GQUgIn zaf?MUv#<1X5Xy4$pG+@~wil<|1g@7WX}nPLfOVR7SV=_q;Hr zWGBKhiqZ24GN4|wS*K+)yhZh=C+%D_ipA>=PMr7WL&E#V0Y$O$(ui@(9i5mucNL_PM`_nJ5F<0Es_-!}#h@9|=gSNgKh zJm-(iu66FjQ4Qj%b2WKp%;FaU+`M7dWUugcBp~j^d;K+V!w@%e^Sin_0!L);Gzs?H zVjl#b|6L=y(YU2!&X%t)-uFu3BSMZ9w*b8;>gcHfmKECBW(S1_KMwKa#%N+&m|qpe zk#KHx57vab_-;wDz+iqWf8!L_6Bgk%Ya0Y|@i-vC`OJB;0(Dsna(L!khKhjF$v^PM&8y)4KVQa2qYmT zQ37n)jN7~JhQ#eTbIr&*gi2h;p-{5%J~zu$PF?Q8^|KyLS0%R+EeRgEitz;Um;Z~s zH;;$9{r<{9Jc#PLvuXCOAJm)&+d7Y!hV`P5OPMbZr zC}?P@xYH}#s#Mm7_$e>)(!_^OR~+G1xhxrF4l;V5Fe{-jQ5KL$g*ywQPFX{6ijJn> zuD-0m^s)>vdlejn0*6(aAYYW>S2dUy1Qx9@eLyzNcJAf1{LciJ_=SyRk@swW0i-@p z$AF<_3yIIw%lDO`hA;UW;@`|A>gsHujM$#(%)^CEDM1Dlo}JMmpzw8|0TM;01btb~ z0r-xI`OdqE*?E<~pddYU$R$jKy4*K_-y6rS9)4`zKN@aDw&+iX9~@T`ko1cgxB&Rv zP3K)Zqe^@Tn&2Nv2JKabjjO|}BKoUJ*`2yh+DnsZO zXJ1zov*}vqRXu~$x?sV&OOoE0eFH{XO9kDP;5jclJN|zxz~Yhex~GM4}L_zaz_QpLTJRoQPT8+JNW4(W+=!uG4gp zz6tuQJC9QLw%IpLoGLaPSe`(~fp#0Z1L|f*m1F%)#Ck-N;ShZd->PPw(VEfGzwcXy zT2Hu}P0MONMDdRN59ytGDl&PsU(*9q7PghKpli!Jb{HfK>J?tKyFq;w>}Hc*=#w?` zYdB+#I9X>(a9Q*K8_aeT^h)IWH(xj|bnq4u1Al!x^=T|AA$_T00_2<11NNRhQS?q2 zTDr+W{jr9^2JMX8d$bT-R2KSRbWj&Ef$zjQu!9$tvevRK)5?Mf01BKtQ$BcHDEm--LU+VVEZz_0E;jXe8-zC=N|;TD8lFz6O@f=;xnOdtTqoTGokuFO zGpy!s$uKPku%Q~j2Fst~l-4IUIt%8;g#Iae7F~9q(%GM2VQzNlPUQ7o$V<08h90EA z-@ppkf^*LC*gb`Hm#-O6SIDa;mL7vISt^Gmp9T9@Kjus_Xn{v9(Mi13tf)#iz4Fs$9(9uR|al zd<{RJIJ5dWHF9FY3LMd&M(!+pgRtaMah{{MmBD}FxOHF@H@ck{0`hE2+n@%L|+SHzA>_) zr_Zr3V)e@j25X(!$Ub=!lfbTSH7LSnq;w%9k0W6Xu-u^*1&k^6{ixXF2_ohNR!F}s#PK>LMAfy1kS>*^__f{8gL1R zhk2jJ2d*EWiP6Kk%y+LIUs^{Cdxa}O#Z@+4a}<>LGW{dU`qUSFyPSjE(_=Z>s_#TB z9E|%56xjvP@*$9Dvw74ahPa-nfrje<=OUf5Pcu)-z_#XjI{40Gpb4@|G2yYHW^{&0 zvL#j)@q3!Z-UCP=iR{vXZrjCy#V1Zm_*z~$mz4o;$jA6T=G?O1<84H-S9^+PW0#H$ z^MQ)I(>{&x2X9KigYXReLP?Ex`yJZNuS+rJbDx9nPnzCtN)fHJH^@2KRnwOZdF=RV zqCJiG0`Nn9&55W74f3OfZciVd5LTxRlH8^cV3C7R*F3kzSK?ec&1Zl^0H&BUY3|>- z4cYtV{U&8ZQN$_rkCw^l6VrT%Dp`a@87*4!oCrxi7K8jeJQ z22_}dhgr+CVa@(q!_5ZHm691_Y}mIqpwH!3?PTT8Noc*<#DmL6HH`}csh9ABTw)D1 z%fy}l$^41emte6%BT3M}WRK3wN*u9o_`S!Fgls~$a*lA=DKvmW_Ch=}eG5CD_ibx~ z(@Q3NnUggM@4WPH*Kjf}9rbKBr|Yv^e=N_cpFMCZgu@O8yMIw%R=y0r6c{kAz!Qnw9I+!*}WtNz=zam zizT0BI%cbxnt}s#5BgMk!N!P^KGbG*%^C~HPQ|-t%p9+EE2ozbG=NH;Iu~>m+;XGR z0H?(`b1QP{UO`6a3I|q2@6?o(Q7lPIO1q0MNYAOyl$$O>8q_&}r2)BBKV~p_d;xiUKlRTxh zKnQO~h{-lXPN>t&`X-Bt-b11@%7|77dVjbbE(E-M^SxX7O=m-0h_0lbir(Or6#(e` zt=MxnM9GX-b-a>^J5+ddpo1aJU?97ZO)Kb;G&9DzY>H1Sxq1~gbbY^Z#?AE%sJGA*GOVc18{ttpk+@KJ zhxEiU(YFJX0qW$9_%#ExEQIXa)0^iMV4c(CYcdGNHA+CP zNr5cM<#(cQl9x@A$+6r2KqTXzQt|a!1bE3wW3)DxO<)vY=wf>3zpVdt^XIp!r&&^$yo| znv-()``ZKwr5lP4WVBaC5v0sf8Pd=A@KF1cX1^g(Us)E&VB0MN*OnfQD23PEJsJ$R zjbw(OLiE53Z%TIGBy-=DJh0YcUhNrWaP(HpW{PX_1n=s{9YvrLIF}Kr@v^dqYUxmP z-&c0-CR^ptR~c1mKfzwK0qJm0YutjyjooV&RD=@2h-+Fg-W3_*R+Y9@DMQp+h^&8s zUSx3kl?nme=_g>3Ki8soZt zdPd?V!zfx4H_=q8cKVI*6CudkD(kG6<=Wy7X!jAZ1lPNU__}T=SpQ&Wr2XAYsHGjM zDNW^;VnJyV4L$QiqP)-xuGb`nSG&$<`O=ikEc7d6KYRz-rKQ#i2vSk;jz_UGo^KNL zIGCvE%n^rV_kHpiWEt-0{*x!N&}(AjQ)=(_sFJG;@?3sNhZiP(uDM39211GeNKQ=4 z{qTaNc|AmpN6jsjGIs$>uZ9#&zeRDbcC&ca*P9?(Z8HJqSGbTDf2+)p`=tA-U;mzl zj8jud;QM!0S^ag}KB|ZsU;z@sqloZ@#t=8FXyQD(SEG>jXyydF>doA~{v5I_cfsT4 zRv{PkMUe zAgx^Q0{&E@^YYZvthel9Ev44jqh{t%sd|DgV5u})@tzFYB~Fpz9cCS+Q8}-SYb^`;tDK9M&SYO|0ZW7OCzgSOq8e~eq}|;Y-Ldfgqx0DN)?DxtE+xMn(K3aa-X5Az zU9*jLV*Tt|-wB2qP|k}yVkr0?nZ9PLT`CI!AKA<#y_FWi37pN&l66ff%M*~wp13s( z?)3jy%iFDB28w7kSTgg&KZgQ`M(j_)pnrkceMdcw4;b=O)aK8WFS9E2d-1+m=HHcS7IB3_kTE zc)RQM>7sgz6fRWCp1$R~R#mnh8g_3YzWO;V%<-RD6`E1Yr}seU@FJII$$kS}){wDf ziKS~Z{!&I1sp8N4FFedO}~{z7_sHUDuW?e){oS4dWnN|B_4K6cweVYF6ghrDOx>mN-^# zYhinSD&*1^&0xd27*Ubjj@u<}(&d<8?q|=lXGUlav@Rp@9F?`j75wp7=DieOdU{%| zE4^AJG54ski0zJ*9@@as&}eI}Z_(oT0}A~3dxo2yml-<&c_HCQ*JoTxi1iZjc~gA- zO>ue*4qjTz=VNee(L{>$zDT%dX2c*Qi5C2zI$AJXR8Z*7%?lF>&Jhb)BNqUgs;;Z_ z@<30Y$eu6RQ~*O=OxXt5>&?Z0;RAaui((2c5i;YMr6GW4HwN-uqJJGX`*%URgwN zQ1R3yGu@>ZVmpVAiTRJ*rL8zh=cBF=f9QEM*mkDuh3!WDu$yXbh~NKHK_9T8$GX(6M7wfEK8-cLUzCKP`ilQY@ld7#DMy{MjYugx^h7vuH=ZuyNd zk4JA!7u;C6z49vYBr_Kr6&*orzndDGRW+sd+AAB{6w@P@6GUHFPocqulTGO-L0`{P zO`g}($a1@w2(b#gCX-$kSpZ3zs;P6n)MjMHnib0^!iQpL+CJ}oRHmhJZER~I!5w5l zZ?Rs&JzD+nY{pcCnqUkPE%MG68_B6Mf#oi|QD_@xh{W}U_0l@v?U5&3wZVOKVoy4_ zr5lP_FoWwTx19352Mt+~0!;d>mh|eL*Ab)>2F^iL;FcVYZ>!;^RO~?`u`HOt zU^;M&o~U%IZNSEtRax1R=Ux>N?!s88OLL?_3(J%M<`TwDDua@mg~IpRVZy4N4~C~_ z)&3qgv&6c44K;3AEv1@=cV>|cN|ninb^{_tp&j|mgS?y11uZ<+Z*!~!4!ITAbIYZt zu1;p9`h?%*Fhxz4jy&5)^gBV6) z2i%@|k(0~ zrk4V&ha9XSO6br>&Bk8%JUg@Y8fa&)!$vqcPjwar!bxS9POibYOFYwu03QFili? zhyO*;=wcfYwY4XzwAmoJaB{hAKltYi9A*}vK3wzBdx*=~+*tnWo4M)SmPXE}JM$7`WTmJoZ84xlU{UlH2d@ zh?t<^-UO4E%(ncBg_#shNQuu%NcVd<_yiZk<<(_~vng~?W5SXYsv}ftBr<8H zsd#sP`y1>t6Ri-)_{q;Bz)fp|19HYrQ6W) zUQ?rcs=N#a?V&lN&q3WC7y1sa4zsUEzIj#6r4uW5Hd&vtw%9Va|9WM!Gih=2g{2jn z#DrI=CP9xnZ;ZU-^pWcC{j#qj)w-lSehp@2<@S7&!@C~ZO3{)_i3!iE60m{CYfjnL zJb*3{(-Z!7m%>2ST;2pEXR5KV*~h5w&(w-Hk!7+FAK49S@;zju-n0F&hVq2r`e`LM4;%8vj% zT06fw&gE@Zm!L`PJKA{e2LqtO39TX*3S0xV`OvI;bnYf*tT}UF7djm;tP?TcNKLKm z>XC%2ta7|WY`$T41yjjsbMZr~S=m+pzCfr{PG=YWBj4*DzXaED=mgqXd;PL~-b{bN znjyZb$p0dc3p4<|U5GR+74cggq*wX zS*Lxi{kM7|va{V^rE5MagV2vc8+!Z7@}5NB0l`X>UvZ0Y7fwIFgorCYx-V_OM|+ot z(dv$Eol6e~S5O-5%4YWy6E5Pq(}q6_!l|`c>m`QO7UR-d+|Rs=0DyNfvoPf1{{(NM zi`Rlc(AdHqr#kptAmj|F(S2R#fcHCNab_`X?Jg%H6|A*yhs2a-oF0IrEu<}=fqN*4 zkxF^B*_=tg0w5THgEq!U)|M#)Tz~R9m0E1Tylgc?UAvLe8P`T^@NTx6k}o*0pNQ@(E^NKfQfnjyrqMD--huX`jeS zS77L~d_@b2GHROomrB7=R=A{PT-h7u(knvwK+UV4V@KL@KG-%9Tx>UM2+QkjTfl(u z?qbW?-rU-mwL_{3Qkx-t=@QITZ8TW5WuYsLQsCc`zh2vf%P9-Dgx4$0fy}75y-5J?*1yne2Jy7 zPL}}DY)QZ1ElSzXW>OwCXA=~&#b1~VNDIr$^`4P{B^d&p))3jVgEW9rVH90cxsG$S z6gThn19d+wtt=eZEp14jwy7N`^Rd*7cFSRDPnQ=2F8q#r`bqGB(d^VWZ;jR7({8m) znnB*aso6Lb|GinDvNebFoT2Pp)NfX)v=CVP%3>{x9ntv4GOcnSb1$~;LQ5TP zwToQx270~4CVD-U7(H7{4{z2~xcE3=bdK#|P)k&IqYY?`oJjWR9GpnNzV;DAtZi3Z z>YnoyOM)pVcKkT!h?;srFVb~b2Krs0rQ#Y}RojLr$#Y48^=fbUQv7Mk%aqJ?pQ!6+ zu1=9y4g`5eOsk@8vwN_)sKHkoLxhCFWrp+8)-2YM%m1EF` zhj*QG0~$toc(rn>EWqB7dufXCMK5;Yk4CB`5FxYydx(J@o#OS<(s%5vtZ|f%g6Waz zZZG32X^R*h*{UfPeT}-_!D@DRK4KAOt*yX5iyoke`Y+yG{>UR+Svk3O+ji`R*{4;h zrIk*vQzZ7i1qsDv3-6NQGO4&>yHqF&=rwBB(l&hpSyhqGjkk(k!W2U8A{$@Qr8_nmXG`mKs$N`Zq39u2z1`%^jS#zD#+C`c!0-GL?htopb>yw$ zuupc&173XRHF$rrA5G$e|6pKTY#4?W)6+BOI|pCB_2F{y%~O^FffFp|9Q!UYV!6^} z_yLcWMmGz7Ott)r(%2$}SzIN1#<)0gBk7nukbsVN0T{JDenbAT=x;*?_`PCQ@^Mox z`Lz_Eow^`=Kro@4UeIEj+X}U%o&xcK)}90q81~JirT#z`ap9g+c4U!Q`1nM|jKz#> zUs@G|%D1Yql=Koc^qRg)+8z7^=x?Ut?)_-eX(4_P30$aqo(Q9v3eP!+q(<5=brqW? zx?q(Z{MN@#kh1_t1LI;cmvno!?qsN$mr}}b`6W5(Z(-SGti>)pQ6P103#1~2V`REUJdb%ExQnfj%Ob3kK)a-cQ2x4Un=8yABsU$iFN)JN< z`VZ^paiH=Xq+}t0ZI-S2iZ%~g$T}=F1X8hGwJ9y|k?A}%OVe+r2~!yf+|>8z8DreL z0_lOY1oN?abQA`T4J@rCZC=_>-wo;8<*VBbKd31dhm6e|_Uj?siWMUne1Tt|wNLTX zTxQKei%9T8EuS7bUdmYKrgp98{1g()J6-T}|L5RB7vZIevS~3LutQIe2r%Zju~gIF za^Uj+`)dIFEA#ys$>xS2-ZXLT~(M?1yW9Oa#u(PM){jAb?*T% z&^Dv=73s+ZO4tB2u_nDFJa?!()7*wW?Nu9|e@5lVsng*(0U=V@OE|8h=d( za6y(wGsCxXtR`xHvO3JSZKvMhfFI0QN0%-OSRJl->H5$jHv9%zMuS~aRr`YyzZkFO z2+%swvn7Agbos8rfrh0%s~GNKKG?fd53V z;hPFm&z+a9IZ;>1Wt3oVn;zzFfnN0v6dW;@5r0t=E)5aJaYL9dI)rs|G#0 zkUf;4QTS4Ie&r~GFu!#@#=ZTK$8k|jy?SJTnBtA26ojq!21IY>>&F442o%4D_xPs! zFM~8=*XlF+d=LSn+sl_l(I{DcX zPqhUGqy+E*MdJQ=7-K0p{29)AySp=2w_Dd#Aw$gLiRxGmb;vmzJ}8p7bg#!P!k#^i;dbx?DO0v7wyAmV0$s}CSQqK*!n)S-p~;#8O1l5^_t+0%(x@ls zks;&+akMe?^ij^Sn8yQiTCLL;ikZUB`&A+ZV_1TZpp}mkyawrl<(q`< z0rqHafR9G;T-`%>WpMp#4Pwe~=oFH^#b(U(JV?#lv2bEREO}aJtQzn$GL<9GKJtxX zykkzR%xv>W89~?W8dI@(&amOPdU)vPi`ze&Fu!h4e+OlDy9WhZ^nCB5k~+B)tait2 zV=G9w>J*Mq!h2=yo&skdFnfuVSG5+X$n=>L;TLt?!nv)4zdbyor*H7In7S9spXh1s zUYcp{9NK_8{&9%^)_KKzU8U(;elrJ&6&O*5IsELr$lSe4=fKa>yl!C#xcvK$z`pGoHI6>8s~5C+rGvc2BCuam~YTz znkpXf-f=|NtjzQ;n-hJ9+^&@`XiGkOJEYLEgZ$NAH`5!8{?v!pQ6x-8tbRD+Rv9Uk zoUpT{x?#(dJ2Z@Zx+v`YgP;@v$haH^&7iQu0s!I@b)S8b=hku$+P)Cj@^CBAc5`5_ zzTJxpT$DB%21sO`6IKrRVfoEv0WUOpVZS2v`dQwDS#zkrrJu+O{u3UUAQGbfC+c79j8oT~>05gMK&ll>6NclLkJ}7*p5Prk2PpzutkgvG;M|xt|fI-Q+xIU*}Y-P^te+ zCZX$|vi*;n?w99p>f7T^fVIW^u8{mt@UU+Q8g@R&+b`zPtCMM5ysV$w=lyP(|MhQ; zCF>SQwgM2fSSszuBhtV)n%`1+f3IS{gT-Gx$LVygCn99G<-Na}aWLqYMt*;HCHLi| z&(Avgp<@9WjTDc>%_}^I*6#oM4ZrD9@ZbLS%!$KB3yrJZqU7gbdL{It88CL?x2>T5 zt0aD2@$Kt8yHyAAl;d@ksx23__jPh|BmRyMez!zhvvpJSl?BTCrULqJp9?6hZ2r5F z_~mm4x;X12e9Hs=W6vMNzAyjZef94?QWqz%t|Itr{|hQItX4-C)6*bFD6f8|=pR3> zb>!Xde=Hbxv5g(-8=JJc9h%~;?#He!hs2(e*fVnd;NRJu-;l33whU`>0y|w^{J$X~ zRW~%*Z&n9{FIhiZkz@?@ZMHZx|2Jj(?z?|e*n7CH5}m9r!{bV^;|KOXeiBe-)vd>s z|FI(O+fYR8k$)BOXOs9BtA3~cKy3IkR<&+Unqx6xlqzrCIY4v0_UZ57_@^NLjkagQ znw-ZA{cUY3#HoCi~A7thF|8(m#NS7xBc@t zD{>>{;qw1r$|E!T;%YO3MY5tVBkn-vsP9Nh(`{`3Sc!DTjUdm_RS@A>nKdu>vHe?-F@)y z9+m$X)7Fh}3f-fD@@poopc96QZb{yA@e9eOb<`i5w(|+8CRB~eRg6?r(}pu*btpLQ zK$o3v!w+$37j|T7g5i@$*-EEjW}aGw|3@&Fi{>8``G5Siu_ow25|YseXqSQB3$+C5 z;lfl~>k7O8kki6u&ktojx^+|$M4#nxLpcEzPKQ*5v;R0K-?r0wgfQTU)dMZRO}sOh z`BKwMH(W%76;r*`dp?NzoM!5L1a?aG^zb=)+Or|GH0H&k-z9VY;;{8w;hw*K{-|>8 z+MF?dao~I@7^qNpl#%>Zzv!($-b?!kYy!lF$W|>Pr50+L7v$1f51alHRQ``5`q|D# zz{kIh7t5kD)8oEWdly|b1N&*uI%e;$3b#K26pd6>-#yo0thKlmT@|-qoIFD2e|h=u zKKhr>&1{i>N<~<~T6GvHem$*bD;mdrOP+5+EOG&dA_E`BiFeOg%xOgBbkUJk@sueM zy#3?ue~JPeTZFb(0mf080kFkeLUh=ssw66T6M1uYmgP4O#smK1P_5bsx-q?L5bo#n!dVxM`if8x8 zP{tHZXJchJizuOgzB29W4$Wt-gaGfzi{82BwEiUfk!LUe=XLyX<>xOPKAwKoF8-r7 zJyjOSK_lPd6eS+X+r_(i>p1^Y#ofdyA(t%Ca#%#*aGmn(nQs{zPr83r`yapHCQh#V zui~(;>~=^_MAiT&YS>dZLfXF0S9Yk@bD96#Xw&L_(xrlZOoBEM3tH-s>FZD0iaojg zw`TYEgtu>tzKx1Oj6}{W&?=C)T$Vt=qSn=7TA?rP@!xyIZ@%;=1?&7*!TzqQ{i|TV z(VjmV^LNzmUj_S)Q2v|TzNcVi2VvtK7_EW zOH{e`d*1ACo#sfc>^ESpCK{RS>sxa~*nQ=)A~1-TOK}FE#d(zvK|>Pl2Q%^ZiLUKP zqAZ`^9*xgeR zADEbE7ZZ<1k#dwr17IGrk$PVFRclWfv!{?g)K*o$F&52#GUaTXsMrtR5%NyspPbEC zw+3@X2+(@5WKMsFQ}Y>calUS^tXqM9P5)L>E}2)^vDR@fnp}faWq$Fd&b4VU>#L}} z)*59akrB#6z!2$YZsD@ecB=78d;d%Ts$zUZC884DE2ZVb)~ONr$Ow%gU@A8E!W=VD zjn+QL({i$nK&O1A(1~or6^jq8Bs-B7!dfHNZ8Y@6sX)wM7vr+; z%OKVSf5t)|vzCv+>Yy+{YQ56c-Fqoj#kv#(rPkCZbpH@gNG7dU#wpZ+O~>G%auAYj z#Uy9vEl9AB{_bAhFZ%pU{3j3D*fm)-VRQ+I4Sh{FA__3=oPGmT82mz(mdAFOli8g6 zo~E&h&B4_`*&=NIV(*VP3(a@#zR*zh{^ZstHivKD5kH||!}Ios6J^`_llv&=Z(O)Z zP#?oHQeY1@dt+lrEybV$2m4%hLMierG6vKqpQLJaX!+hT!UPCJ^nKwtSY&yO&_tXR z&VRZKS0-cTq2rtLezuGBS+wxmMIKE;l&*w4dzIuqzs4i`j!4QYHu|EzZ1Jd!KSEUn z|4zB~M$&9q)ojtV3^HL{ARV-_tR)8?EE$U9X?P#t7H$_~mE|aEw;5|}0kPbvw3Jig z+i_VSqD(jkNf}(5vej$U5`(?8nj3QJQzAWMJz}UHJ|DD+gLBKa-M{iv)t-%t-|a{#x;$#Z)1+=CWYyD0gw5v^}UUXBET8ay{%jkh&50B!I)W8AU1w; z*H{sIwumu2Z$SayZ{lsMj?~C;HRzZ(C0gtgvUQyvr^)gZ-XX-?l;20W%RII8AgTirv4=e9 zE@+T`Jsg|t+eFsI*M8Ifa92E1^09>=pzpdWl+S3lr%MTW1{EQj|Fr%MF|QD}SFsw+ z3xTQV>HG-iFYiB^Z=%+Y4^g=>m4Z7X8pC5kGsMJ_YY#0)Ko7GZpKm^*`V-^DtrEW7 zlXq-??IEpv(ymUK1P$oelKlHUTYvnk>xZwUpS`Bj1g`@H-sT|gWpWg&rI|Y0`>sxG z#_9n4CR(IlP=Hx@}wM@6KtAxECly2!p4MmMp0X^36Jt+KR@MyYlUXJ`o(rmy-D)4f`)M zm8vioJil~=2r{4EU7oxvNYto?Hr@#lHF3l(2Z_#)e6qD6z)-U zzr!OIMxW&VsGGNGdq2ON{-S?2AQQWizuzOpGdq4JenSlmqM9Kx=%YW@&?u@Qn#D`K zt;oiQMC+%?zGQtRJzlq(`K-)yHw_>f?#dgHW0i%CS zX3toPf5mBPtQGuR8O|wkQ3o-W0#}N$@R~0n?!VwUsdN2=_*c6Dl8%15gyxb>CYnESqVY+qMsL@nwne?(MC@j1mL#3#cCf~b#( z$UVtu(F04QzQ|awurg|^qWkp(dF5VkhYXy#)F?WC3|_ zhmzlnfGtwkx~9Zq`1OV0dv_PV`E*pyesnJirXmrngW*ONs$|uLyhNMobke?=Lw~l; z^}p1y-S6k&QJ6SP$}`yV^xn;S8Zwq=t@0-t=UdAO4gL;DA@m%|fa7+eh)SbmevN3; z2~jOEe@AFF`2?;$L=1iiBj)FGD6qZeQ4PYluF~Vv8E(Ve^4RIo=YQ0g^RFLrnx08E zb=F?tXLVfS0RXct_7}hu_JH3RvVCD)q9YI(^@y;bYJ7~R4$;SpMUy+EQ$)=D>->98 zWdF#HTCIN!HqQW62G%K8#t}Zy8(n0>bQFd??41_H|H5~?ult6c3lmk33_Z3_?apl6 zz14xj>8)Z-jR$2x>>s~$72H{E8HUXT_ur+8bdKqw{hk1R>y$ctr2{htOeqgPp<|@0 zIQaSOnk5G@-rv>+?pCw?Ptw$ue15zu{#|-W?I=qRa}3YQgJ#n{Cr@dLDB!@HL5`921wqI87-WeuKbu$+2J0#v1k+@sy zP_H^vq~|EpNer>zSxpvtRq8@ULz;~*Y;pS-0^Skdf6C(-$fq3;{AFEGH0vz^iKKELDjhuOSP4eRHuGA5ICKx zI`|EUxUckms-@G=SB8Zu+As3#W48Nw3p*YIb5m;l7FEB-m4G!jW6SY!KSazz0;LM- z*5>uyykD%@i_nDw=NpVW-s&(GeY!uJ+mcIrcY9#RR!;m;o7W}T1hG+4;R(MJZhHGvu%SFa?i+CU_bL> zqq&lzt{y^kN_gUTDPjO(zHpnGoQi;j$NkUjh%{ytpOsF{50zSI8t&a4@;XZE3g5Fi zR(qCyjAQyp-_Y*Oy6TeZwGJx&_c*%3*fnCxDrKG6g{?EFl1TRlBy3?Kr%e zzlka;qFVpDYgKSKNQs_-91TJ&UGM9QYSnrfAnva9#r?|5FL&&R<4k<95r0-QZSkLM z)2DU>sa~{hoA6(3bM1i>5R|2{$bQ3NG;_+W12aE+80|JH2XReC&@Q+kx6l3Ndx?Mo zM0wteEzeru&U-yrUqeDWvdEM3)VA7sI386W1jiT(dFWF7S|OK!t`Te~{`RNG-WcGl z6|Fngr3VU&)L_Z0Q-+*kfk2nzuLZ^qsapqcu)^Rejiy{~ainO-p#vBi#dO6}Jz|Rg zlh%+hS;3WG5$|F570Nn1P+du1O-vKEw@|$6;o4hcDHNyspbI!@`beQ?JL!3k=A;z~ zCEq4L3}a|5%75lLo)B^nYb7pZn(3-;VO+b+?wL@$1$)|urw;Wqg z!&V3CIDQ3-Uw$}$m8B-Bcq^Z0VZTmyKX)z=^#wmUfT4 zq-cv(CO#8rO%!MbvBtYr#-rNa-b}UYOg_1=YfQZA0KkbfU&mWH_8RqajRwJLV0qnz zT$8E1chjJ#K=msnslZ`_-@CQvb-NMmaIKM5Eu85%?27-0f(&TbMaG)g z6qNgAnQN)-YsInCKFN%b!%BAUC$(%$7OLtjRJpZ3O)cuz~MVk@>d@CV* z77o$p%+@NeqL;>=UXx{ouwSurrybUzaLt|RTzuq@$oVH{{U5YiqyDr*Yt(nzRGuwK z*T#f4py7Bz&7D63$$$RukB=Qq-bCWEV&*v$PUh4c7o&;R`@tuj8j1*`!YmAy^Lc=Ng9 z)}Jv^midoXnJ&8l5B96YaWHdIlE>&(#H7Z$g74gLPTdP7yR`ly=Kh{I z;U-{V90px28w)#^oB{olUi@Nc-!cFDHut6Cu~O$6{2(!cfJeR7z2FxUB7aRsS`X*( z=NSDj|8jr8Y9mN#b>*|<3JUVxfBIZ?(t6lx(B+@?{gCr7R&`R~r~M2xcJSxmAXfb- zjo47ye?{~w()=r;?`ZPx)#+am{eKjZLE>4-UwQ!)cmHd2{~F!@s7U{Vh@ynV{*gyA zl`WXv%Y*GcR&-a(VP?r6{lT*2RP)`0>`Q~ zu7AJ%FXM>cqd1^h|4RN}t^3#F{x8sm9lII3UNusq)f=2$Y+0Tu()x1FI9Qu896*>_ znWf6w&?kf%1QN^hOCrPm!Y)VV_O2}TS=+8W2^d@IBRmTQu@1CM^yZfjJ5x=-Jtxkp zKk_34fJ8;8^**E+Q)p(tZ=!StPMn}**W9gCNRYURz$)8jODl%?k_RUfPcd1*1*Fl2 zSgTNCqs|jPjg`~ACa@mI-W;bKG}i29hHkR9%Yq;RNM{bgiQleJ2JD$Jg}r7j38!qxln{+P&>7~fA2#|6Lbah~H zbutmWH`W?@;ij1QU%vCcd*o8Zw~RMhjlf_HD<2#RLeh-Iilh>*VvK}$%9tySCkT(0 zbhXI@WPbEa?9Cl*B<1ILR2T6CV^u(T6$=R^PO-eIji$j{jeue}?Nfbd;T)VB(~@6h z^rp=h=eFMg8PB8Lmv8+>y7Kv!w~Rx5vKp&XVxBUPPGCF`vM$a0qVBhZj1TX(SrKhWDi>}Pir38EQ%DvEFxow!jA8?6_+ zROgR+DD?fK>+5<>WZg@T>~@}MLP1x-FAnNwgPGs_6H@R_%h19BKj--6sc zXXUi8->$FnsU6L$y<>9h{%DMJqM=!N_ugyulZozTPUZt)K=#9Ew)>6u8jCSEJu%Id zR7;=4O`+QE&tp~R-qqTn$z@M9TV(k8)a6f5M9!(yFllHL1XBP=s$Xgz{sGRK0MdXt zJfn}_C}#8#ll@W_{dklF<&^)Fx%g7Bt^o$oX{kZpS51zxwsR1HGJPls7{2hB3o7Cr zICO3;#j(HO&Y?cy`#2h-WC}F0lY@9rpc}Z!_Pr=T`^aZ)#`)BV2X2M07>;)@Jz*@_f2cO8a9Ei-U5j2NOm5ACPp zXKMx2AV=4226%tHRQItBUE)G_dV&&rUULnRGJP zn>6g&A~H4>2@2H=j|sn!v7{G_O{E}CZ%Lv;|6Eo%D}H!H&D*LJZ(dhfcA0&4n2ibY zF50Bprbexw&PX}bB2|0~r3Ce2cLAZt;?BPcpRxp37P9kgQ^U)jfmkbNMFZ~MLcuL- zoRKgzCQ478-;y?aV6(19O`!KXB5tlcen3H7t$9<~%vKtb4)(KU zw%o|tm*Td|fGO^y@#N#Z%g`@QT55EBY)dse|1R?0FaICcb9SJ?HJMq_WUJrsOv zKy+KLV2-`u)-oE*SZhdc#Wki~jk5jd9og`JN5-Va|1lQCUJL8@O60v+?v{LSgtcdW zWlT>;aU>8&{?tX=%r_-~5%nJ5i9b3m53oG$u^z?lP*oA{+mgi3^`}Vqyf2XBnDs5a zDKTOp{51O6Jn#}0EC+xmA?T_5{e%#rye^cf8=CkSFQ6qdaFkblrB2(O4WLU$l^wbS zJlhpn|29lU->G)3K@5?!k5t-!8o!+$9h(xqj?u$VQgWPo4gIZMn7!QckoN>ycbV~; z5`y^@sV2Z=EQRM;s$ABnv*_W)VvtIwfnMXZimxb1n{_Z4Zvr&{=p|$gThfttFLRT3 zE}8Y2GcLOcY|!KHY!DByK~<(;qisrk>Gtu3TasFCD;@ibwBzx;pF^IsDdi$&Yj9Y@ zVY}odg(pb(ssyEmG-(3lO0#EU<*Vi=%?p_kB3;Reea%a-g-nMXOrSo6e4NxKB!4=_ z#Ag2E(gOE#B>W_=`s|d6do^UNcL>`FA@$}ej0DzQP#^7((;N2<5ETW+@cOJza((eI zmx^o*7jAf=&KD};*RO!cYt)67(RVXRofkqp1C^8M*VD4UP z&@WQi98vi=L{yZYty1AIX!rGzpthfZ6?FRDXsg9H|DhG)Sgdt=y|!fDZbgER&N1=? zDC&exgmn3^1dyB#IiS22H4GvJ-mo19C}*Hs!RvsL>6ce6$qFoNI0 z+Y|(OPq^jn{q^wo+aKx(;ssqY$zwgn*7*t*LzOM?`JsyKo)Z&-ml--Z9lr-x&npi! z_>Te}vEwn=Hm_1>-zs7mzx?9a5+z_xD?6rj9SeED^k9BQ@?OPVXldSQ`1wPy?mP6B z9?Hm2(b9<*&yax|z`^W}vZYiKPjy=d;M)%X?1r29R@DcTM`KJ%n@&9{Tj1uzPFzm< zQ`lgy3%uH7l{30L$QvBSa;QRFtn&vr0QEspk@@wI_=Qd__~XMaYIuqCM?Sb%Pv(hC=Ab*0+-aw)TOLh#iP5d;Y$$LPB?A_h@e6c9k>zmcVO~&9z@pmh$SE!Gu=Ci;*AOwAMVMbhvt+)BUF|_|zik`L z7Es7A--js*U^CZRK--AP%j@yaGlg99{kx1U>->SqiJ^KLOev;F(_H4PH+w2%wEa^= zL^uw7z}Omc)y}fhg%fMo>G)f{*?5^R&)2=VTy#zp5YIXO*TeD+&5@5`GGRP2D0$!M z0^6i>j<@4u*rISil{&Io({BQWm3y8aqo$V;=EWjQB!t7O zl#*?GW#MAywT9sUNZ#ph;Bx%_xoY+WD6?Nf>m7l6w%iZ!%;G>z(ysQN+t5QL>NAc) zl8mb1dp((|C7Nm>V854BHCr0vLgsQy^t=El)E3(HM*-hIxOcZl%8~S!ki8xcc>c)NM2tLUf4R+ew7*MNy< z(Bm2_BcDtH$meV(qE8XXm1CledVoD8HJC|X;on*h)ZlKyE2dLwsgpc_v%;6O#XDH< z2KtLZW(WM25A?GiYLnb5$=IoYKsGSEFL(4KwJqm==`Gf1z!<|Odt8H{b>KlQnwmTm zxCxi@B=VCzJeiP*SzhW##h}2UbpK@8XQhEys_Sm4 z89?aDPGXB(ucOxrMEvPvKke{&39bcIdBvmvO2yYz$<%3-OGf#%kF8pVAg)88)i)+y zO~*$&LMWlDOC-VcGozG%v7(`fCJ{f$kPaeJl%}Ad zQbP;9Nf9X#5JCtcK!5-Vge0^ONZx(n+|Ro2LGSg>`o8bSv%dK^ONNkZUwfbBIFIAl zY5q|)ua8(MDJA+Mh~%KO*cbgvl#rUpOmR{_hg=$Z)_`z)gDG%u+oGp%lDni12~Ig8QG!qPKoO){6;#lwh)SBl zI&;zHqf%nbV;V4r+~e)8a&X9tk>YAa=yf?qpid;p9n3FyDeA)wHO6gG@HWM7UR46AnOLcU`-R*2~UKmOSx}QTisn;(!YO#A9k)AD@-xckNS;Z64gK z*cK(bn@q`nxl>Lg5P79eVJd;7>^pKPq3P>=L~~=P+OdRgZDOA|e%ai+iPttKzV`Mw zNJl?PIMTvkTVBvON)#S8Iy}G{2O-3-@Mp8vIx(Ir?k-Hw+mE&wodPy+m!%vXLvZwPTiAfgXNczY0c*e)`PYW#zkTvLb}OX1DZ}S^kCvct zAwtKP2SmkC@}<}Z;dgvXj=dB)1Dab%qZc8dq&u$5`Y-g$lAGEXYAx*#Se$h+#R0=a0A@N)@?)kvS<1F_jE!2&vYTiSxxV& zOxI82%oa<@FsVK>=`&=Ibuspss6LwP|K<&b8QgpUVB_rK#rLX|M?zg=%`^_(fRD9?;KDrGU?i3kO0#O2iq>ZdUvdC9+3Ov^zR<3n5nk9AM!ybYr?(6L;Bd5 zyX*4Yr)YC{6L%!Wu|`(!t~3378mlA_h4T>2t5|v3|A5$G9^N6_s$2~^%|p*G;oc<} zoZPMY5Z8`BzM(l^U$%f+(_y@}YUB)T;K;7Sj{T)>B)E6*9VtZVnw`Va7)_$;%yZ+! z*b*Gx<;~ddYY}O4W~AO+Zw;V)QG^W~1)??CiT85q5eGa?m6zud6Hi;(@b}W5$xnw| z2z$CZMM%qVL2c>5#wH+V4pw9Glrk$tYAF*K=OvYlZsV6cCJ*0uOv7(V;e2S^y~jj? z`TImsWF|B;&1gBqcDX3iwx!s*wnYaS5m}$GI~3=#X;S!LwQaD#$#08lDrCUSVE(0-+v0?nYiJR9R&*mo-Leq67C$=Nyu z$(wDsbt*?~^lG^AnKu{T%RSszC8JrOGymv83RpQ6H?w0@- zS3s(|01s?X=CrI0=2%|s1E+QcLUPatmDSI9IFcFswo)kC^PxqYk_gXErc1%XVr5p+ zlVraypGMSot4HFk%jf}&Sx74O@%>uo$Wr5ox*(rzZn@W> zKhpk^{W@L0x7dF5wB@x3UQb(yx_@a#?zd{0D5d*nb8v;8iuvi6e=wrw`d*3NE`zgw z0HAiL;2-PZQ>-UnVmLYCrRlGl?yo$6RxW5;zlD031tQD>?wypu#qjv+lu7MqYL8{TZ0l&#!}(o9}9m$0aQ)HUjaPG9wT&N!xY>-OT+ z%@_zA7FN0~=TK}{amuL1w3|>#u0_t&+?3?L`>>wc5kwn<`OH&9rba#R&25?CsG#Q4 zY;?**vUya4qn3&2avz4=oL=5UDr6t9{KEf}`H99(ANsidx^@QOpv}_S5mA;21{Dpz zde6GuQ)Z4y!(k{b=Ycv-(c*;KIiFe0V5Qmaqq4R*i9Ej3Wj-!mLzZd&O~i7<;oT&h zTq({y`+S->0xRZY2P-J;dPr6zr4A#wXD*8 zU9c&nZ|rx@u6q2zR4!FXnZMhYHCVFfe4ETUid|B@f=TrwF|VAjr6Z3fn8ejkXMsK! zX&Tn6XMA*f>YyCYQwM`GJ?3j4V-OX4p1IU=Mum>`KCx7z(@Etm4$c!(A_?|k`L%S} zD1}l}Ww2E#ZIU|OrGVhkTx#gJ7ZS1@@<=`-TzfyZ19k;A>oMBp9o~FZYsimKJ*nn& zqv97PKv^?qT*eyOlf;T(_%;8s0)S*7o_;a z8|K|i_Atut_sXs{>Ir9VATfLogD<(uL=oAP`{l-wh9wWVVMFe|?-r1e{;KY1U-BKl z@@C3dBID$Y&Z3#0`Xn)5kbg$bUcc13qNSC0?k=$;a{I377CuQe!{S(1bi?Wj;}Rhy z=IP6}QcW^IhHF~}S|vJaSN8!$r$xIO%O_}-e#5nz5qh*0PQ)8g^HYv>0zIT_LMTY>jRUs1du?-ugqyWJ;Cr45 zy56mw?DkyR*@HwK-DR6oW3#fe+Co3lf-OpTa**{FpN<(yVRxg`UvS9#&v%J*y}=Jz zgZqobf9FFmY|qg{2k(G$O6~_X7Ia34LOHWK7Pz%KO! z&DiEw$fp?IKm0R(LFQH{fC<|+web6T@DJf%U$gaJzHi2>ti{__Q&Ip)rhqGfkkBgQ z)AH~P&tWv_uz`>ty%*Ju5OR3topfK6E)bOp6Xzl7Mr|vExw$*H_K~~+T zuC+P`=kqud-9=H{sR_79hAm=t>cK%doQLjB7b@z&daB3uUC!?an!n)su<^jo0K0b7=lU2is zqSQf!m8IDl)e2>(DESIU-czZk7P}H1hdg^qSAD*)B_mb*oszxlO_gk47{3S15gQWh z>jrx=N{}9ZO8auYrjp(cp5%GLoX*bA=$RE5O_pZU zo77=8M3;Z1ekw1SG>;Z~)_KGr30dn2x5EJrhOCwfe(>SoL$s?OQN~y)Q)^8|uwz>W zgCF*2U?FVqu6yM!%nNZBjSbMpjnMBBtZNQ4x4;Oz{9^REb-hpg6c|v@U-WcFwLM~g z(;C%|PxE~1l;rbT9-VR+J#H4|wXhS+3^^`i^x_su^b z#|%2~s42*LSzhfjHARe9`_-yW&I^oz0t+14Hn!y8aqe7iA&g|7u~jY@WqsXQy9Bv0 zbRY&4AU}jMSYwBEtwsB)M=NCx;|NIuj|B%IAz=S(zZmn$x9wA|za$egY*) z3m3G_dbhaX)1rw_e~aoEKx7v_pyevPO)?Nx5`t>cxV7gCm|m2%091FyciNA|kX|KM zHaDugeESR%|z{mgo5L;I~1}TRvZfz)C;ew zQ?ITVc~KN}JKq6jVOH%7&`X9|NIxs-Gg6UICP%nB=yFAZNu?{fOZ53Gi=`=WG4$bf zl!?@G)x$v#^v7$)8`V_Dk9^0|+QkLp_wI3$2)vmT{Jn&nNam221i9UrLJ!<|VReO= zPnouLY%D@*hN!nF?*+V?gFLb|VgqB`_Ovvg*B+Jjr4emxX-Tz4%UawTDBF@&SJ*Kv zZhhFyH*u)Oerf{qyqok#R8nrOwOFkY-s4fbZxH39(r09rPhGh>(%ftryYw#n8_&2O z-Yi_=H1-WOV2lbh<+oK_`|X|4M+uFK7zf>{16HXn_l-O#T+Y{6_zV!tdsOL^=h$4F z&p0~KrBgxln*$L1Pl@hs0SZ?=6YHRzGe0SG&n|3+OeReH#uwg8lqKg}7a{hbc9#pw zI@UVJZK+-B^zQO&OW8RilGZP>yWcUtfeu=dy=+46u7h&Z^;i1PY2MRs-s_X|5safL z&*e7HmZ~G>cZEXLI;m1S;o&RHRXW7O11y?Dd6gp*^D|G=e3_RjGLv(}^%thTS+YDP z(9RT4rM)$NjyQbGd{Mc|yB$#PxW25Ls1OIkN10t5Su!q275&vkqT^H^vh|feL94Rq zrV2-t1t;eQPS`I2Q|#YeLzrR@aC(sXVP@SBrcgso$SKvtzVOtmt%R%B0B+ z`6#(hepSfSc=_sSIAYMJ=M1T;>uGM!9ydDW`wPo0#fY1DuXUU>J(?IHZ8r8^etvEU z&d5`O%r0K-HtzMXBv(Q$LjaXATXPh^fn&&meO1`7g+K~51NO*!NY zChk&|-UA37HKh;5%k8=V|1jJ(ulB@>p!aQl zYW~_6*>^2MzVlZjc-L+~-3e&0gc`#{#ds>mWxq@o^te0MHzla=(^F+7_n0A!-E(dj z_67-`s`e)M_H;{q+5k_llX<;i%EUru@bu5pT`<2KF2BwdM~Hmv2GggsTpVe?n$1M zFKg+#=`O;ti9o&9`9tFM4c-73$pid^RJF26aqn^R3zEZN@ukZ5rnP=m6sq(TjIK(j znfkBE=o#CtU2(Ven+WaddWg29YF@&jjO5N>xfFKn6bTr4NuO7qE(4HUj8ye6Zn}75 zw}oBK)Vgz1hU|{OM(I4leFmWod!f}=gK;SE8}UL1hO?ReZk6nIuFK0E?Y?0}fg99! zOCB0dN%15^q0r0Vj{Y+emA}|TK&rA1-?cCB6wYt)_Lw)8R<8EJK5CN*%W95zHB#~W z4*3KDF@wZe=Zk4)LaIF(k@$f_32!@ew>^jQIRf7#;M_D!GsD=I9#^_OB9FDir?;8i zW7WXPL*81js`EEz>vnfq%J6UC?m^}2!6fKnG9Cyai#M@p^;nJZ20X%(HId{2VsO)r zH7!j!LK!~ZHoLlOx4>cqkBN`Mc2ie(QrTKx+@c!kvpGI*Fq1bc?a?h#?>qAW0)(j< zI#h}>S82$8&z70=`=821ziC0vD)}JSonoz0NUrri8(=*Q2U0tO~t`~#O8DB1->ARGk%kdu@vH3y3|M(pP zx~&q3uDk?_GJn>eaE2uC4SR|75|@jD9M@)R35yWA%9Xi1Ky+-rE?LW)WLI^+VmO)n zjFcKfN_RGva2@pPai_Nhk#)3`lmxv;r{-k>TeXm_S%r^jFHY#-0R^_vo9w?QSkUKe z5Na;cZ!V^v_4-rBpPn>5&L4Nm<6LPo&h5@3S-#||g{SF36K_v>ri@IK8>d26BV~UY zn)C=r+OHzjHvgA*bhfaEx_zcJ3UY9Rd}pHMbG*CGUS+r|&FsyurRYJvEB#jaDi8iI9~LM~=@RbAVi*JKiO@9=5${QrB&gH9$T&CiFeGZTHUfzhM-4_2|T224nJ|qs%iWSD6ovz;pU_kVh zViqpW3mbQjhFbF=Yd^@N20~sSM(=%!jKMQa?gu<~?lFz9ySgmm;&!^VrjwX#0^Fd) zu~JrZ*MOREYeS|qoyS=9;=rRe5`<&_xVB&jQP?m_L%+ zB!>04oS2lBt8kiwF&^?`xllnCs&)zQYUO{fUD(zfqb?-0q&DXyTzQV8d^Ek`FiaQ! zE{H`}{~Z3YIB{aduvSo{zyQGs_x2aUa*Ej~wR8g6?k91QP!l}+G& zgNlNzJ5s^uL%{ZI?K)i1vpAOa1t44t03w&zp6sOujIPAGOFP9+vltoC<|G$#dTU0tvz;DSq1X+`HpaL5#BlO5j2`BrsX}z8b=4pmm1iA_jBbD@)1a+rvtLW#d)b*j*~mah{#8 zH^FII7_MH$jkwwSgxy z!|t|m5m{^Mm<>Z1AZbfXHf!s`z7$fraz>g3Y8(_X%Amu12e_!S5%KTVo@<^{U0w=~ z`Qu|OqqAMad3hKobcwoH|1RO~qX*X(HZo&Be>dF;FfIp!h;a#-npf+-C-_j_4dBa9 z0EqQc_VtzZ@>R1-Gi5*6#_an`yq~RqP-T(q2G;I>`{Z?ehqpTpsBiRVZ{f4Q`@*EY z-?S0+`!D|z_B}lA$9}td4!}fbZ&tFce^7~%ZN^4!=*GnPKmW4c$sYQDvN)?k-Y@~j zOD|cQR@r5@--97Sf`Y=ZL1t|I0sWytAz0gd>`c0?8RC7v*}a_f{!LE;*r`g=3?(b_ zNNQU>VWDPdFY?S_RAmv5EnRl7i>bY`%)#3g7L;UrL172uYT?70qs6#g9^_5+J zqWQCKe`j?M-Z=K``IN?o6O|tVpF^X*bZgixXo)igN{0T+2HyCiTc=Y_$h;MW&V{@6 zJbL|!y0-LgG$i{kkMz%``RTj+XVd(3`~T~}|1a1y9X116m`rQ7^uKq$)r3!(jUlB7 zVl&Jen%CUtIk2y{8H)`YWvID-x&ErR{mfdi3~!9X4`Gj=ZKPwDhm*+aJLR>1yB(Rg z*C}sr&SRIzdu~6|OKm*ozwUVt2Z16oAlp@P*l64F5`#JgT`hrsy^a5Jy8rc~*Peh; znuG#UgUdi_AoHYG{O|m7GYEGw+w;1sfeqU0Zx1#RZY#L`Y-{}MFaBFj37Fv8R&lFdRBHNNBhE5Lp}E3dVBkV)m)5r{U@Fpm zoOM)2)KrgbC~vtb!gvpvJHK%J_{edk_{JZ;_&ZV`Nxw`#VwWF#}mF&e*;ydPuVBd(=7wcLG* zLw+)(jLh8RJd4h1evssF`{&QRx;c=omXkq(Ir#vgwIxSvhbOsgO}2}$oRO}Zt!HWP zKR>*L^9T0&#J%}7i zHcLxoyTsin=uf=!tGua8@4K$j8CSfV!p1qv0c{{NSANEtdjsNG9SgXQ*l_{-@u587 zQ(p6q6edH6ThG-)nw8={U`-9}1Wjh4Ye(9B^*TCoPXF}f9~RqZQtGTesYZ)e%kuA+!@DuvO$9Clj~0IQ&4i zWUk1w+_TPu6p`k;%D1@bz;X5nMqOld~{VUNnz|T zN5BlXdYO3CKh&+yBqImAu~#??b^YZn{wr5Nu`udC@z~|^>AX`HDhGEOn>B_A;Xkwp zMJP-p=B9s}8~J2Dnf1x+m35{>&Soi?>Nz7y14WCQIDoL_bLo&Fc;v=$6mf|>^;y{*$X_2iznDVFNr;2S6L!rp)Zjtthar23iseG!VjrH)(88#v;iEsIX18v5 zXLYrpf@0Zs@q2?>=oYJpEm=C=*c~I1>Ni1`CxaT0AGzUd?wn<~G*jmk5w?7vrCg6}tv3i*)vs?fuF3N|+k8QQ7ZzFs-*g2XeEi}w>0Q$ zSp3rF<1Nin9fX7-#j`|Lsby~D2q4REzf@Y-2-p|<0A_f#T^|42y9&7>etttoj7oFF z@0@_L8wyz6HFGCwr*_^ah#WTDl;O*8cV^_Gnt;W<&PQM@b+iA6cx4@{s?OP0>kAK9 zykQeKuEtx>XK3We<;F*CuvUzT$1J?ib24n3n-+7vM;ku7-58H?tQF0P@NrR1@a^B* zqH$_?_4ASFmLV06Fd5Vxai&)*pvD>o;)Pht!V_u!Lz4xKqr3ce!Q+us2B0gJDv@~e z_fM=99rWP-6|u35Jp<~^!GaF|bTBi!{KbFqF%3G~VJWBxQ-1*g$=YQLi)=9~L+*Q> zdgUIB85q89Z7I@(F~gN>h>$38Bg6HpRtESWKZho`x=?Bwk_b%C)8sp*-B0FFeT?px zOk5R7IpfG02_AtHXH{rBhSt1hxq|tNP5t~x{cw2KF~cS{A;0|?hic1tm61s(BvDKK znY1)GD?t5XsZBHR1WC>xOP1Tu1gm;&c_VrVP~bXNO&!QL&-^#d73-`t_Zhm`&jlz+ z^e7Y-P;m_0EpprAUEEx}E2$oLjbNWT;7cZzO?!g{5>4-l5BqeK$X%|DMTP*Ig)@F@ zOnGj{@P*qX?%*|QM zZ$p134rOXj+%HfM!g{%-^V3t`y(&v;!oWRfX0mK-Ul*%0>Dn~}c1X=j$mP6U4Z7L6 z(;(|bCl9Pp_mm#7&~H`I%e9Ky-6J(q+%u+$7-AyiR}yRBK3SG{;1Y6NBU$T~SgSt{ zL|6IcuR~Ha4e6#5eN*1tyCq+cR@rCNC}TZv%4H%d+jAZ+awLf)z$yv^gT~??^DXDp z#Gi?Ok(vbR`>-~Nc6|+p$@@jyWQO>ten$Q?mR+9XTznNdB%jjP>Wd|xKsZ(rtRg!~ zjla+to9W%$2`q8+jNBF$=p{6yoTJ5?(b68xq-a`4!=N{};*dwi!suX;wj?{R+<1|N zIS%kI&7sInXgDm4PCvpXSl~#iCOB9wiFx1q0p@oU9QJ*eib$X>3t0o1uv5^tS65GB zheh^btL&qY>kYj;h{xnqUhPC-q4t7+w@+;uJzi@;zr6!ES;OuPQ5h-@nS^b! zxJL}1`Y4E15_p(y^(aLM?)aLmTL4t618@x$&z2&Wk2PPuRp)HOJzUpcCVq`_Ke+REJzt5mJnm ze!t|gOGyoz^8CYjlG_!rn*PJiNGu{wBEP99f_Fj3!t`c)O&BzWnJ~SfJoAF{8g1!a z5@ut>Jga}{H9Rb%X%D(W4p?yUTQ`@<&O-r&Z{+>xkAkPnQM@G2jTZm@Wzi!{x?{@wwFGn^tfnXW`r;lA06{;&X z5h6w@BB@a-i_=`Onz9C})=DTNm-rp_WmgeCT;qKCm4_!1%}WbQDIA$e4iyhE?iH)^ zmy_(`{F}*4Z*mB@Naf+`bx}co)*1(tv1mI8QJ{kXtbR?gUibwI^H(+>PeAA096hej z>6G`LApFdmTlXf+pL2;-EF!WM3@ut6AMT`cq z{7)x|ASc@N?^@IEm@o8B8+#2*5yl9GlH3nWPbb^=860Mu9u-1|iC$Z(rDcAes4L3l zM&5#Hn71$mMRSTT|0X15mabozXXu#r*JkhQb5M6DE#->?P9&w=s6#HK;hBtoRS9#-RBR!Yi{V2 ztX+MTKb+j*Z6zwNtuVgd)5Sb>TnKmqgr8Y8l#{vBg4{qEOO98Jf zjX^{cza~02PvNIUC0oh9y}hWG)~r-6s4hr;^eAog-8&qgLzT?Ryb7pPgw1~3rHgA1 zr?FP8a^Q>2P~F#5&o;|by@<*F?T{@l9TP?AtR!E;sj9bVDLEMSnM@7O*5Y{7qWBFy zoKQ;$dRJuMy8HS?ghNu3m^6}~m|LmmC&3_m9`YTHQX+?OuFOZJzy|YNYsz95{6F7U zZ&|LNY*udbzAzL`vJFG?>|s-7+);&@RGCtifHWdPSKm|(XuCL}BC0noC*17^MZ!&} zNMKEa3ZKf71{MDE)au=#n>bW>Mf1L%;1K3{c{yTPsuZ24?lU){4krj!HpT=Cy$Fs8 zpFtV>yUfqP?zW%$G*7?wdyl4^{m>bkZ`$zQQl%CK+5Z3*HlP}%GBKBE@%-vF`|92D zXQ0lZA4bOF3odYcMnc41ICy0jlRjjIi3s^0%uK8CrN~mcU+bGKsdzVOtE~a2oLle< zf87j_E;Jzi#0{f23Q5{E%!)Sisdw@1hYIWF&IR*xaqRXGevsl857Eim&Q}|JUG8qn z3$rgBI1Je?g*GdZLd?@y#!Jor(>h^efiq_14BS*uSQTH)xn8|vdO zCw$F`xfSufS*q9jTAwDlh9A118ltP1zuzj=8K{g@bUu)ynG|PI5&0ZtM4ffqr{1}` zfrnpI((KmBf(>8);mG=4-u!(p3p#ZkQZ~;KEX1UwPukmz17u;{-xR9PJ>@cZnmUFW0c;^mSOohpd;1+jn`%Ku)zj zQS_(h=R7||34C$tyJMTNJYRHQs%gJYBQal@4~p;x-WS+F(S}}KG$14kB`_C0R)t&d zG&gWSTaIEIPIjXQy-Ee##tuO}7>U78=YJFkt#8tZV;9N!AWVg zLdy-Wm*;zs)El(7v0A+05*1GOC2FqEPnW~cKK}4d$qkwDKhY6Z$0kCi&)CSPU_!QG zu|qEe_%VGRxtp!9Aiv><_svRt?{Ay*B&1Z@v;zLOUg9r;`#?(&7Q}6IF!Pnomya?M z(Bp3SGvGu6T0|k?fS4?4Ohikea+P! zskS19kCtXv@|U%{;lKPdLi=ZK`v2d#>BP6kE*NXf+o4;3XI%gIEa$Hg2k0*C3fC*u zzUge>A>7i*7$N?hJ7d0@ZCfV`iF22I3H(0G)`Eysw_5(zr7ve+7k;OhQL!mpPI+BQ zEczkPeST=7F3@CSJf3K7M3)@f9l;oI`4!{4X+>lCwZln;N4ruJTJok#V-2WAEG z(fsbWRc}wo$ne(ut&}FNcMu|&^_T@@E0?YF8zvwg)iF1p>Awu6e;D2V=l?m$zRoZ^ ziNhKI*r$5S2}$c~nvqZQ4Sxv6{Ft+LHk@mndRKgB_#}tKI@3b|VwhbONK^b9AM;DV zm+#!WFEE%MsPD*U0d?h{Ed#CKKU?OgRVA~WlIWeMU|w#S5T=z9ANm1POuyjETp>R5?^sGnuUEDeWTvk&&@Zc;HlArV~uf(oBMz&7;+H$%W;FSK$7ZiF|n7 zDy`4fcCztARthH%3$)apM2Do7a;p?0O1CqZ*XRzI! zJZR`<7N}V&;ntf^?tpZ`(kUsRWswhP>6kCT9IKDYy@1!k4Gh!8sa z2tGjaw=frEtuV*ZY$YnsS%Z8*WLl~)E=dk+#%P%C2s9rF;to-Pk@sE~tGQyaL^r%2 zG`WdGjeBOGBocJm>jB&vuKaWThUE`d68pJLQE?gcoHi6-iXxcm&41ZL)Cs3+2eaihSevXwtJLyQW~E~!03bw- zU*#4I66=ltb8^9uz(t(MZl$;SA9A}p$+cGZrvye91Oj zZkXRrf8P$%IP^;m2w5SYgEDVV|ATF3Wgt6fOB_LoF3_RRn9yE>Cco_ z&BKa-CgN4i>H2A}HnOGNj)}crL&l%{E(}YvPUN)Hil9=f^F5>vsxt zGnr?z2T|S=1B79{Zx32+aWT zHl(~F0lgUzwdEBV_$(}-f-Y$~bOiRK`BN-^^O<-Pc6pck+YNV3d_DhXnfwAVZ4S!h zqt_}At`}f@*xKwcQGJWF!ZuZF7`UgU(&*Vn&$%`Fy;(ZR>+j)Sftje;VNdcVG~GqkxGu0 zvU)~-pALmEWqx{i)lVTz;J&Hh&*}p5R8MfSKBr%nk4M zXR3?lC@&mHA5h%W&-tvE=fOx>GOLcPXyPwI#^<}nXjbSA6mm()PF48U4A6>uG&qg* zi2R?_BSHcst6t<^u39z=Z;V74bK~$bh|NdG8y7W>hEIqZTdY~0Bh<7kMHyE( znpRj!y-((K8iE`uf!I09-5%2nu;O;LBg+XcS{)1)r6ndtliMGRc2|emB#Lf}ReU~A zxp@DHf>MGC<0wxm7oTI9iM29Xt906J$#}7~MP=%3Ebw#^OUO4_Tjtazh?-d(B4CM?xSA}DC$R3S2ouRA38@ia?i|06AV8`?ZIT)uX&LQr5T>aHwLaL06RtOH8p&r% zf4y9K;4=5c&dsfHr6ZrzF9uucwI-Q}xA#P;!BqS@pfnDQG7l>PM0n79NqnJZ2-bbV zs_9*VDQC;P>X}8XEsD~1y`^%tOH|NrK2)CGXr9s1i2*#hg|vHh%Jln}544Cs*#y89=U!VKlOa^Xw zII;^2ZoD?FYiXK)Q)6G2C~DvhI!o>MSZESC%s%k|bC(E$N^!ZivCG^QYv`rs+Y$0%I$V3lzk<3ZuN=umTKU7#@owbI#Q^fUJdlKfQ z8l%emDA0rydXTMQm92@w7S?|1Zp$}WR*DEsNcBGRAk{OoqGNFb-DSO6ytjFC(;kWe z;Ka?8qSNa|d1eaCB37odJcd&-X05x!5gOx6aOl z4ovep^#g0x;|F1lkI9E{5BA%H=a%vzmebjruC8g>)&Hi+np`if%rE=(h{G;&_jMez$}gvyW}mcL zGxng2I}g0{56!K#7bK**Jp!*a7GWSS&rlu$R5)G7@i05CNfS1K-;3JM=nT59S$2KY z&8~{p63yp=ZvHazUpV@1M{h-nkf$qd{l<&AK<}ofd^xdlGO{2HRoFJgC8U@x$-Y~Y zGH-4)57JFga=&vJjMXN(>2`!*p{IlqvK^}$VG?p8Gx0DcGFt6--t0X1&Gu7ZN;Oxb zG_7gG0?r9RkC%p|R$GLQB*~}xmKxA1YPnQnJkE}AhNkG79CE0f`nYODNi%dL3k3^0 z8$r&I<(iZ#(;t^h?^Cg`MvMzpSa2PA0F~0ulbCG>FOJNs3%U# z0)kbp^WY8Pc_)dX6$fPqgM++ttzE*0eNAU=eAP3@*ubJ<1*H-i5;qyOARti_w>O=%}n; z$SaJsE>JI;G3+awUS=(bo0u%fv8zhz+<8Y&F!hf-hfK)tHZsC`uGU`Y)iVC!v}V(0 zc-I5W{+g4^?!Cj05caT89uxi-lFKpTt2gbvs|wCG8-Sz@`9t_wFOJs(`sp|Wov^$( zl%Y;4_g?y}_W&@jK}SYMF&A-;NO@|(mg^m0Bxu0GaBpr$(=4(zguLeb;M8kv>as21 zt{RT!6Ip3|GnkTa$cTblURXrKR z2|n#PKOC3dSBQOt$HIaeMJ}jCYOC>aJMQw6yBsZ;f>QAy&ve}nZKAGmuA~U+8@c(u z-I^7i#!d2Gy$RVgjj#DKuuJg4Fx!hs4JU>2xf;fr;?hOr??wJKkbLjbZtUxK$Wj@7 zX~ouzwwe#OM+~LA`mLsv^kI~rRXJ;B=%s8+{|Qp^ZN|fbmbB= zAUQjq~wDcV3MccP&Hdfa<#_5=Yr}1J+HLjY)fcRkYcG= z3ep|iBPhR7`2F#FJp8!(ZIH<`w&WwtP0#8B?;byAFR1h#6kJpAFPHCBAdgo0fZjh3 zrRhBC3Z~4XHI~cxN;wO)AI7}9L z5o~{V>eRDVC57Pe^_!v_6XE99mmEO#Dpsk{;u91BE>@JC-kuyD`w3<0dliaF?_PzX zThD(3m1}}Qh529)E3<0s711HeoRQ40WQWjAcAnnB*%N8|pFQrs_TTHMI#q@M@6wRk z!QBc4R{W1{AQi>OE3Zpb+6dn8BfY7F`q|@-{N#smGO`gsHgZNwiN4%#9c{3O)idFfhuPygwI31!*n~n=mQ;?*4&F;CTYXMD=4x$V2q=oZDD3{01%-De zXUpyyyT=0ivZLE;P8|l^zK9`em8#DV9X z4TA>nl`wwvaa6-9ts+K+u!wOF-^QJoZ(=Q*XNP&mScI;w1>9^gOOuM!=Nn51|L^@b z(^uWw4bb(8HfKispbk9t^jO07>R0}Pjvqf{MnHwrv6rsT!?2`6?|hGurgIRiVvehJ z7dLk4k7>Ci8{ow(&uWS{mOGJ>_O+8>>hsc{DZ%&ySEQ|c>dp1Gfhgq)$3YKZDMSrz zxg4E>5VdXFy<7P$Y!J;#RpSR77SDvc4kZwI+U&-mI+aiJGsFd~^?1o>n`EtoVZB17 zzB`iC!E($MsvhMcc<1(Tr zea{cNwh8R&yKQq|WrVK_fNE7!F)gX>Tlzp9@}~IY{IX0_kaViYSn8cQR$x2S*^42% z@8V4r>+ydADyXgbU043Ccw3LT4_m*Vg?KyVHOsn4{I46GO9FAi~E_=Mish`!di z%N~ZKoh8BQM&O?t*qbu8tRzUZ9kVM6#n?5CkfdR;T8h>m#4)X^n1tLu<(>%X#$708 zNip?esp4gBv6{s%;gB}$MXv=LdU`L~h6f$SiN1zJyMwzyFOki)BRXW!2PJBTbD&BJ z2!4+08M|co6vF=1MfdOT)pPO##Xh|Lm|+VL2XJbu`I&ymE0YH8yLs+{ERutmH@wPZ z6KVo6jT_WELL5UdZwt-|Hw4_=KRU&WB!Kf1X_!rBIk>ZllU6>8Xw*DYU2 z+TG68$h~(NbcFO(XW3|FX1p(4GAXTd;uFiA;RZjSWl6i9Dztp7N^UHWA}J|=dxXIq z20t&2T@j3BJqY(Mw&^{IQ)4x>jYbq9qn|$M4>a);uW%^Q$?5iNDqCn4rM)7h+10F; zD$Sl0&>v7JMYhtXeE$!7-x=0qx~{v>K}S%TQ2~|a=m^q5dbgv}RC-sCCf(3Ol|hgy z(n~-=dXwG>N{Q4^q=pcrg&rY52$1Bw(Y5woYruWRv;UkQ^JlKPBFXoDZ+q%}- zjo*V#C8cH%nC3g>K=#^J+HVvHOaDQEKukZ#CgbBOXEs>KeGWMOw2nv$ZurgxE$VbFH`O&uQ7t)wI zL#sgml4f*&Gj~qqb=5UUlDPWJdyI*L1ib{#b>de?%0UK%O=8zjE*GP^Gw}6vVy+|-*)w1|PXK;yHr0Oofj$$b{M@^AynJcV!G^WTDB-r> zDSCd@PWKPWj12u+TJ*bJspgO*wliwMFeac(+rSL%*+2y5J$xnL3tm)R4Q3qY1uI4X zvF2LO*~9-xaCdRA&DDc$ zH?K_@(b)uxLbuHqhjU|@o9afISB@aqbZf|^k5Unt0k2miFy74lsxm71)!VtfHNQOV zmZy!U{e#)qCbDX^rejSrS?Oi~OkGt$LZD&YIPb2Lj!E&gMvpex+p=E*t`fPnRG0c? zT;?0Ra(>sRo$|fzp(x&#Y&(+M(|_tf3uS97>f=J?u;WAXr=F}FcQSiSul)%v?W(Kr+*_J2Gu|ZFjnL&s zJkP2Nt?4}N&+^KpdQKMidS%&ixuxC>1IcQQ8zyDlE0a3Xp0Uws)UPNt7q z1WQ1MM=eVBni5ikbKjl+2pIOxgz(kfJBhjDy4U!9?~^O0-|7ZoY*@+@z>*P0F8R0T zND_9HCP6|i93%{9o$Bikot(+e}4q9 z->Vfbn$5}(vjtVz^RMlkX3?w)^k=l*wh&xq_q zU9$=OEsBg?X73*dzOKQ^2$x&gatu&uBw{#KO5&*dLA33m^R@pxW&8U;|N5JI zC%!f{McY|sQH!FX>GQ|G(Ot0=r~J>~{L^8${LS@T{twsle>H1=vv2)tk-kp#-;{0p zTf1MrL)n&}xm;$O=yLAI*!8xqAGmOf%nIr5TVWV@`&g0Eb4;(N1(dF~O+nsOw@bNS zO+WqD`5$TTyzUv^*6iL-;`>9vbO)j z(fR*{QjG!_m)t+TWB`oYsRaa(pBC2zJrf^#-!z#}*;a-C>q|38sxe0>tZH~}0>r%B zc4a&$$^e9vr@!VbQ*?I)z7O72>0J;~lF=&Wq2+J}; zIIA9I1?e`at>t+4%rmC=F7d+t2R^RTAMZPM*t)j<9$+zSI|N%a78*}bWOtV(ueSBf z)%DR8H=hUz8`{P-8G94O*xkOo`A9xaZ9(dk#VZ$I`7EWaUjVTq;CPq!w32o|E}304M48u(3(lZmep!1k}8mj`mBq zkz6re93WXJHtvQm*qTkk(U~skoSwQ|+;fdyR408;Tg{b+9{_;!Ur)Y!CsPz0aN>OM>tQq+y*)o* zUHS7}&(cGLh8>4)n&ZjWqwimy}#9}?F?vB`1`rhc)NOJisRZHgd&wNgfFQl@Usv33@k#DlO4L-_7r`V zzxsxO4;oHwzTmFRFL@5sQYAf%A=ew+2xz@&!UXR^(t)mn_GefvV30?T;|re$9(+ko_t7Yu&*z^Os*ezBhv~Oeo9s! zm;hR~wU=Jz3i@n)MyzhL*w5mq!awWP zIf?*%H7gcF2^!YKCR+bStLI(vc z1EP+H*gbR(xGM%D>i5Vx-Dy%&kdt<|isilAM~f+@a%VS-E|8 zK1~tg=A&}8mZ~6mw`nF$y<&Ij2%F(Ko+S+U@+^*?@z7>BV1KGl(XlPpY!8i+XYHLC zuG&f2!3*Jhy@wEq*3m_$sr_axkJxH&)kH!^xzXtx4bWH?iw9>_74UVp8Ke8VmYpDy zw$>jfXkrTWU^G}72~47bhmb9wX1QJew$IhY567t-cEuDLwx>fdmoagH`mMST=W+Xo zszNk=b_f?FPSGByIys`n;uJkhj2?LaCY`B?>(XX~Tx$g2topTsuT1n{LP0F+_ZCt^Fmczg#9P8zGPL8sx z(;|+IR|97vT5H#WLrfM6DlV1Qtv9jh-V7~7t9IIRDs1-{W`uBB$!oS)*0c?}Z-keT z#|s!ji{NBV_9}Zz_P#>PmkV@~SxLww36!vp>;zzF;NW8Qak`n*foj-7#uaZ=@UI24 zMkqYBKSQ{yUGg8{;(x7G>AOa1qCd&5!8Y`ah&WGeQR0fG!p3^3J589zZBvteplb3f z^{PBB`w%|oeXa5-x?jhtQ)Ae47mAs3ZCK5uaRihh%`PZAMt!?tT~?pTHljMHSMQWv zHcWM!oS$==7IysLXlRP7*SiI7NE7ri@gh**0d=}o^78SpTY%uC^4NmDdH#n$(kS~Y z=H05v68g0)I;PS3ygL|oz zarY#k#!&GDO(%||a_#2XB-_hZ-=BIp?i8|&cfc9C9uEfkEjh!KAOXbVfh3Ng>!TdM zAlUjdhK+iaJy<6)^=+5|f1&M|^p{p03s^btf|iy3n`??uHuYCWwk=Yj_YEMnsgg@Q z{%NRQMU5&WXWHZ)hSR6GU{7y}8bP}}K@no8&;c6IdQEdNz1AhXNru(Ha#MBx6=3}N z6y4)|)LX}~r;t=Sj~hVvE3bG+e%WL`BQ98xomJ~mYG13y^aaM9IX9i1IS-AjzMDkP zHy`kSR3u1~32%n zn0{++ULouC!wer+UJP*yD2Z640>M8BI?QpmSn=kLKPpC7Yy0!d_t+GNAH_7=D<}4_JF5SD(5}l|E{EUj3=@s((5(qTi&2Q9B5VJ1cv~!y?7m+rafY zX86&L?c|yg5;;U%0*b47wnIsZ{ zFD}zMZT|UU97}o4fo>Mt66d#yzkB@jP|D=kf1UJy|DSjhziUBTOCoAva(c^p^KRDS zl?7#m_=3a$6IV>=ON;9_QU z%;O5i8puvDbLT&pB{cxCvT#^Lcx%{D@Ks&2`E@6F&aVz+FlH)jquu}0ywUEPK-SX; z7nzdI2I5}#O2`rHy%#J4kAXVb**;k0!EWGqVf0d%pPfAW@i|F3@bP7X=K4S2-a;qUFzR%tb4W=6@iv`zmeti=(lK? z-8Jgv-}1SzZo5X3@=VDTxL?ig*ma%?X7k(&6S4rdq2pf8saxftEE}sCY*#jYBH;5r zcMuhZcSY>yrE;~}uN|l0Kp|rFZS>}YWLCy_fJ9)^9Mlv?bwy$}c@2Af8edD659t!+ z#yvI`ZqhMiS;aBQw9IF11RXkdHrhq7o3e?*6)$hrHb*FsSeu@YzhXXihu-?~gr}F5 zXaO}y9DnhYeZ=X?8Bv_<)$!}Ui zZ|B#v-xXd8zv@1ONOj{mE?Z@qUp~9wxMNG=D%(CK$U8RHFr5Z5!NMv8!{eTf0Xy^? zoCd~zM-l_K8qu)z%7+S!gSo=23#Bf|k%Uy}1Wp+>fB)>6z5=gK;H>)v#l2ib_8gnxs`#mLL;A)m&7UOlB@j|!k z;_l_ew+lVAPh_sPsLTQ9St| zfzw9;B3^eK00*VuN;Pn*tdkP=_|cTCIz8Z6x;PTcJ6iF7T4h=Iu2W zY_ju{vM1Hn83*kur|edH6;qq}B?vE^lSB*4jP=Ev>2{LvoK3@5CeZc_&<9i~LpE!YPi~EU}Pu_(9v+}w7XBsRm zv%xvLy3e_3S74*gX{Ok8Q2dq;bFtM|M6Mv1;uy`Jqn+)Y<~=;gmbJ{O(EJ|R_XOY= zh03hd$>^;#C?!KTt=utA{^j$n`ZIU}M&x%bok!J2!iB5p1?!h`tENjFuPz1KjW+;= z9b%7lVX@xPGqSHc&2xDt>VRJ~*^eX((uby=w$;OFnzP97JxkP28l#Um$*;v%VaE?o z26eMde(Vm^l-7ybkU$=6k9w-+0s7&>kqVUastN@bOQo)PPfas-=42{|MQ(+iWn@76 zC^DHF@!+E5adjOjtDgfo6yZsQFEUxep&yJN`GU()z{VO7cRHFxsGIf_TwlwQiX{_# zuW{_-NPpGOu@e!C1+~4GYoWh5j6a^rG&|7!ViMr%Lj4)$PXYE~eOWh5kL4%Jw#(P3 z7$cjydGScQPr& zp=B~eP|HB8w=-O(G8j-ZzcYw=$YUgmT8U;oA zTPhNlN;?n#set}Z7wHo3y-QBTJ9hk&br6pj!!2YMhtDoI9HU=kRPVMtlXbpcTbz;3 z>8*TG-PYr3u7ga2*8lPXzZtc#JXWtRXNlGktU)|RZ5#w;bcP|gO7llkr@~eHR}s!# z7zTLb!(5NTm&|2E%pAv?iZ%1gNIw`1shmpma7bbxN`N(A9NiA>w4GZr)x)M zV&lYRw|Xu(>L3bG|YHhvo#2mg*!%9{3Q?CctcdpA|5zv1*HS?vd zdMvG96eyvf`hJ1l0%u^ZJ0FgKI4J^r3t`Hhw-o(Geap17-L8#5EY>xb+Ik=V9{G z1zo`0u2QnoJEYdL5+;EON9(O51xBl{hO}o!sY+egW;m^p`P7NPp49@pAzuBA51R}a zsMr(0u>83M5Q7^7H**W=)6o75$=h<5mz?)uqrZNUbI~>ylkNky0B>22!_9pf$bXxV zwG}+@bvb|wL8VeJ8->8@Om5Ik7Xs^L7MJ`W6omFDj+*zo5Fss~W#AZ4MJOzhNU3E*U3h+!jVZ6F;mP`P zzYJ)R*`VWSnhTPMG(^i{(2c zUvrM2W)>i73egy+!guNWv$Fs4k(Nx}!QEEFf5jJOUy!+#6&@HJd8fnGdPJHEGm!)~ zEnQ`?zMYSQa=nHN_N)a?-qx4hZG1mpzC(eTE~sq8wWuw3!c!fSH1)~I0*g}Jxz`(=66&@t%ysLgnp1u@ zTztB4O z7c>oIDp~=Dj7Y<_Whv4L7ronk3e?#B?!&!_bX`p=H4-&WY|QeijIU~4 z=A*S+F4=qmsn{?o7W+ZAw5F}<=ZRZkv3z`m>t53zBa1%+8zS*q{Wf&Dd@#Grp6+20 z#;JpvM5spWdTowib#E>nJTLdkp?pB=1wB$mB-Yq_WU_14$&(rvKiG%>vd0z?3bJx8 zR>kVowj>P2i=Ao(iVneev{F6$j+kxECjmY?pxQ0acWhoGXR5tJFtng@ykt!B?4DH6 z*Bj<_-IX`jApTObK@Ady>UEUF4_))@>s|zfQ-J5}wk*gE+y^(|x3R0olk-KCE$DSjd)W}AX+#HoY zvr^k!ZILSOmomkqlKl))qfp-qI4tLSP1n2>Do~MT;mcI}4h(M)RaQ||CIjlt;giPl zX>Tg|^(COT@;07R1b9x(@*wb0Ag-5E2|r6&($n$5Z8=Y$&PUtDr?_m7^cGS0-SJDM zLlHn>fQMdqsDnpqm#e5fnTZ-aS8cZ|nfTGp3L4gT?k-v+}5C~vStM#s^6?*pMLUgxAa+?9=lrO@wEhCeM2ujppX)~+Qr z=I}VrueUH$UV!|<3FBRCN1fEN5}iyIGDli5(LBkvqH60ysp-ygMYn?X`r4pAuK72% z@JsCd=R^3C1FNaY)2+1*L3~nA&KmxXTsS1X`ur8!KU;#YZf*9BbByD+-%L~WEd|d< zuBafThLTr-_T+2}|5;JXUq~er_$@WQv9ytJV*R+r>!}Ux6ajfnK}~yALcIEFm`1e8 z{L;K9Z~i+^ql?@A3<|*-j{#*Ka|h0+2sq<0OPa*lg`<*Lz5OTY)FfA<>#ACTTqz@O zQJG0+44Z523BxM*`0Be&M~%4fmJe(*guC>6-0dqjLDHWwN4Z`cT#_Xzy})K=K#@km zUt&=uNfjL}a!rS{b^Ioa5viez_H)cg*&WZfiffFSGqHx8qbqRxrQRn)bi5MG+l=E1 z;6A9@+)i2xKJ7!=(0(JuWZ>X8Y#o74C%s=jk4}bt$|1dhO+|vvK;Tq{2`IA4_YS1E zttCiKylz@smcvi`j}#Abm#WtM>e-rror%fm4_oZZX_L0bN(3g^Re4)M0`T0Oo~BwE z#96<}R-lWlu@seSECFKD{AbPx*=R$9htKzf;9s2x&N~l5(Bx?7l3Gm=^9sHx*%zE- zGbS1?vzeWJDaaG3sW!)G8?u0$m6g#_y3_TX9t42Do}qI1*HOrUk*#X|4|IpC!B|>W z5iVUg?zlpcV&cMa$*g6s^`5Gq#GHgF(9SF-BS;0QmyjP@?dHLs?vRLXpx-= z8de7Ca%(GnAjW_$K4$Dh*->XxVzpmQfto^W ztphyxHExMW1`c^yvS8lmF4|S088nNrDG3z2#n?Y&K7Jl;sfwuS0ZsYmM<~ks?#y#S zGY4`K@!AxUG?kfMHZ+;%teWLEiJ-8Si#WLuMF)Ay#UW5Jm)4K=TI?gKwn%$-h!riY0D@n18YNT$psB2bdf)B;_`UHy)qBbjI5#0X(AHsd zud({5;)WMs;*eLmuzx{q|7Fh11TpWYPXcg$plQxI(06{qfIsxm-ub6GYkQ8u*H4t4 zLC}|@YGh93beGJ<`TIgPbkFY1jeYO?znl{i1AvvROc$apP#KrViDja}K6xbi^Zt+9 zH`I8TOC@-8Uc1b4%0Zy%z8A+&$$7}W+dnjw++Zrj$~)}y2K|<$!wny=5+^%FJYtM< z%}T)VxRSbqf8^(^c)-u85Yq?cp=_5cT$Y2p4;H3WKrctx8aBFf%Dg$gw-@xg6BKU- zn4q(ka|+9Gc34Gz!F=}e|Kf}PL;A8b*(H6dt*y{A%Q*zy0taS z3{a0jf?ka?d0>BAy8p2Hb-w!TzIp`&DLw{C0*k{;89FM#X7NC`$jA|A$-h2VjeHlt zRN1O+D9~WRERp4G36D8ZHNJRdmuqbM)86kpP@xXMxTA(rH_Yp52*ffb?`#6;vH$fX zA&T!?+W>asn>RbkRj6m>@dw@SK*#nURTy<|FCWP+H(m@kMF666=hQBgc78Wni0KCg=X-^b)_Z#)Hcz#{J7K#*pI!VRR6r*He7WhPAv#t6zg_Y=flw;MkbKtdiIWB@`Z(?QNCHr=s zI0e4!e2MIRxH)1HzFn$P2e}i`))pR{Z&z{SwN@<&0ZJBQfR%dka!Y9xG}4>UkJD{* zchOAF+B$lA|0A>ik+>)3r%_KNe7<1vhB%u$8mxk8L$zh2r%r`t&USoaK)nk0}w zd)#$_ZF1V^tq(~~x`DJ>sbdLXdgI-rCw^$5JmbtHnY%A1Z!6=W~82?QIn!f2D4S}^e*twfVko)m~W3~-C~{%NB1zl ze}AiVdK};$mh}YN0)#d3u7y&Zc~2o%ZSR(*2RPO#ev*D?sP*1mxp@P;orj#p(9OhL zDT>r2Zfg;DU-*fWOWJoV{&((Gb-sI*s>^&m1`X>|fJ|Ok>hj!IZnd&qny$dqJ<;S}L zWjHu|UYW$(6cTwzC;M~q{YpF76ZR2@A9zKY1N~8r+4s|@YEvmeElkN49-{?kT#SKf z90I11x%=H=VbB5|0Y4hCy|DlS=TAxOu1NVG@veT>h-&2UAv0c}oMpr#QLhtkC6@>0 zIDhdoMBD-yfW$@G8;4cA;<>KBCdsCn6l@z0J>LDQVGZWG^o8Lvt?@*%$;ThjVz7FVIUM*}#<~ za*}PNOfFh)nPHo2!V~5TQArN|^@3sx1ALeL7}oCfD~QTJSfak;hVEl8&Iuj+>A?(; zqGDmE@08K*G=i^)4*Ms#MwLsK;ly3m=ZbrGYVa20>B+5%u{=_gwVhO97n|wdu-m%V z;C6=GBr8uno0K1T-o1-{_tS47KJqbxghTn0o1HR4tUMChcO3ephosZ+j;54d(`Zm) za#y)uq|xQR6$CG#hiZVn0Zs;4wO(JXy^@3b9<6U9isw1N-D6PevJCE5w51SKPBY2+ zUcFYO9alJMCM&XcM~^Ytx@V;4|FS3%cH2?^;sW^dEu;MGN&&l^M<|J$^sA#84$t|f z_Q%#3JUd&;Pj+G#RpbL|j>(1)ec{XB-*Nj-HRsiH43ZQQAzpH4N#y@|2(liZ;wABX zpt5Hw{-Z;PI)5(@N?0*fj9_Q8fH@DoT_5@Z_x;-%igDgsY(7~VQJ`7tl03dxb`}-n z_d^Z|u(Cqj2g*IxWvy()k4UNo$IAa-miEuf`6-*Kq`Y=EdH;FdW5)ma^M1L&`P9Ct z^c^;y?{3j28o+?uazFN+zK!n<{)#4ucJk6P_6iXE&%+i25tl)F-J$(i1HQZIUh{x- zj%IJ)9-5dx&)$DtFn(u&uyf&kBmccUHvjpDe|qpeM&NwsTF(E70ydc8D`8Z{n@dIS ze{joWgQiA^X4(3Wa8!fwzrF>c7VGGtiSI4aca}oqCRlPFz3m^-nFibLUOHF|e+5e(9NX}Nlk)em+h=0`K6d*} z%HPLspBv!sW4F(w{B7;_F|&VLyZt8RZ)>;LTlrBZ(_9Edc8D9FeM}ME4`u9KsUF8` zJ=!M&!P)v*E~`ER(r#kMIVTLl;-CM;e0<~jc_EL^;nT#S?AtibuP0on0d#Yfr+L@c zM2=@w-f49Akh!q=R|5_8OD}oEDN;{AoeDQ`rv6+!<4?2M#u7zyvT|&BzW8+~@uj`R z%)K*1{Ur0=IZYdO`AHl%Ys&)F+Y#bDA_|gS^SQ4|eewBD8&X>PTuJ-5MdluUr|eXJ zvpD(H(~8&dlP+o9C%lUNcwb#Icz`~A_b&K>CPZ?Ss!9depIhBp4G->*Jl!EnW}gf+j=um4<{+-T!$4S2g3RGAU{6S9WM zhh(n)walF#H}e(=2{&09=H@{B{B^b#k9^8y6dq#T$P76M$iM_t{bApYlKwO5YK*4e z72hr5=4k%*OC5BbXi#X0l_MVit5KWSZ@?)&ncE7eLK^d_rIp2GYsI^N@m+F&M=h%v zZ~F(>1|kKZ+@TD8!gl(vu2LBILYZTwCfFREe88m_=}2&5-Dg|;=R1GfgYl}=B3IPO zM-tTi<`2RhC|6*Z*v0>Pn;o)jX=PQD>Hx>?{Ct5w+|9#->Hay0bbl9Eo%NA^~H}- zXMv_h#e*5oX>|G99jGXh9R?Poz??d0Vq6oast6Hk7LS zr)A_woO8IhsN zAS4Fzw8l?0$DS_s7;F#Q!AvU+T@zadcolX5@JT_M&D|5=TC?1<=#e-^eph=VB3-O< z+yMphI%MgSC(aY2Wo$cBHM-VU(A_FuKIHC5I!1z2im2g!Eg;y6sh`R8 zCr(EpC;wQfUz>VT^Vy?!=Hqcq=~gNysAm06a{#m98334Ybwty5FM@_qRgcE%Wg0es zo%mH4SBJj=n(cVPMflCTr3ig)GXq0-6=OH@JQb*`5`vEBMD2nmTT|FZ;ta}e%&qIR z^j7AixAKdQXd{ajSoZ80P?kRg>-?NKK=7$KB1xEYf57e_Uq56Obw|>)XmgYWcO%I^ zJXC>v+DP{0AxI}XFWI^p$d2MlDPH?NvD6li@zg*W};n-q^Rb6YO5Ruw|PjGd$o!35EADc$68i962RU1Ifj^*W# zi#}(dh){JY-DX-8c^1VrM;RouA+l?8QN#*%?}xYZ-+d@iH>>S$y2W|QA>~9IcXjuZ zK0pY@>CEM|WVJy8clSC(!&EMiB6YsPx|frTpYg%l@}VtX%(?GI7SGJ8qyut!cNFLf z;k28Rxm36;B1AC&o$GE6!@fE$3Ae;5jQcwp!F|w?@Q&jztv#m-xQY6bIR_U3>LZ+~ zwxYXsxvELrafT7AViPmDrZF!;YtkRhBPiG~oVL0G<(N(tZ1@sNMPRQH@3`x_mmix# z@?pg(p2wyJ;y99`HJ{|G)gF{-Nso|=iJ(l7{7cWLbqozxWLEld{r5}X-uu$X+O6Cz zS6l;eD@3s3_R%wLDtvW42N>;&P%p9&6K@E!9h+##kmOXfY5O{h*b|L0$07e!pi_I3 z?j{mw9~8H@Q-awlQI}&`i%wBeM=l&JY_J=wvmwU&_Kx`6X6+qNJJ!^#MqP2St3Wpw z@$!u;LI&5sL2u3&1UTA&?DkGUnjK7<@oeVNL4O80CII?1YUuvjI96EYSvGDF5pUn% z&TGIW*04RA3s?2*?6;16qyW1sBrvCF5}R@C%iA`OqxSc1_AEeehM^eVNtt7@Ys&g$}X$~`f-y^W!osx(@c=Ju;eNb zaJ?+91HYq1Bqmo)r1>x-Y9{K8dUMV2?OOa5#_j{hgwL}~t??FyOOw|eS*M(qPTdD^m~1vHnvG2nGLqO2u&XnjFZ@?zv#WSnzakTsxCv|!3Vn4C)Ysl zl?G9oMcCx=*k#Mxb;LoLRcdF~30(olrelL)8jtGk^_08)5+xZGZhgfC22b)*v>wQf z(YLQ3m|Y#7Qd1#l)xTU0$SY5c9N$^s^N zwTQ%40{5A(xr6Bk9n`{ude$43u4gg^oGwu0`pW=iUlKrhb`Txeq^#! zJv=0~cIBQ8c|Ob71Qv!M9V{K?;k`vC>$5grrLY^II?vF31TEeAetAGQjcNrxjVDgntYQ``|r zX6}8UR{NaK_}_j4r?DE_W^t45_e|~p0bI?F-&6Cv!kPyL(2bUvXQ9-~-to&D!;;rC zhi3*8O0W7?%{}WktzAxx(cJ_C$r?8g%x~Ya7+?ox8qMhikNdb z;TR9vu_wFUK>##`Oq@%RgLi?mIPyb0T%*dSgJ1lG{E)GcL2Nkoiz=cyki+|d4R7b$ zUzUV*ecg6*do@?(*Waa20cfAVQQcWN+9C{er~0z(ln>a-sp*bU{A;r_CBN86+jCFo z$9ot%-?xclJfbpEM_(%BM&}xydueLWbLTcWgsbh(6XYpaBuNQkzJG1xAtSjID0!x?{>v+G z*6oD<%np6P4uC}q}o^9e#<6m^kTi0GV=H;|()v%)Seq^0X_5F}DiN>YWkY(09vdKmB| z)Ta&QaGgq`r@@+BZ&L;sYV{yU2BY?+aL?TrS&Y1r#%|t4mI<0Ruwl_08iYo<2i9;i zOe4YVT7Qgoa-4Rys#orU+{LToW%Ye8Ew*LJ8b3?A%51*PFTwu-@#Jonv9Y9>v@Qdb zLd5LIY6M4+0Qwf)4FyuO98sEE2mP2-u+|boo1A1@s&8~m);W>=K)37hT0wdT4r7no zGqZ~WdCacI=Ox;H!?J824N_N?sT`AJ6bt28P$)*sM}Fu%nHX!7dtcT8V@riVhoI%R z=lU%h7XK(3SeN7I!6oKjc8>C_2fN0x(S}nqVI@w}Ku!H74c7DOMZuhHx7=`}k!Uh| z@)$mT5aOOk2R1`7i8NV8tYd7Kk~=c+l5yh9Ztu?PmqqGZ|- z;eBUt0Mox&K>iHCFiL(};AnMvNl5T(M?5~{hsdwmdQGc661 z_}cf+-p(H5&>f&X3+KEIm-g5tUyW>^Qy_fED^{yjC7#AVezT;?HhQacDUJ@(J#uaZ z`lYyPqR)D5Lp^Z%rEDJnPm8$yrH)*ork|Aob~Qz=rX?I@55qxFimokS&0;o$9V#=@_gPM&8euj0PY*DAkE}$6PZTTK|oI@zd%Az)}c& z^`nv@U1vJ@(3fxoK%Jr(VyKVB%dEUpdF%USS~Y_Pt1(ic+416VGCfKUIIZ~HB0O?P z$jFOWd=0Yd6;XdY6zGV=i7CavP%CiP{3y-S-atlZsx`?j{BVh5EM8+7d+ zaoR0Qc$96Jo4t$YNY?!Oox_M=ik^IG>add4!1O2A6!+Qi72TFm+luk?yh7XVMutRx zfEBk3dKP$b;G{|gIynLga+syN6m@|Ir66}v!0ip^#0d0jGaSzkn$=Zg_gDGaZR1h-M*%!yXBr|F6&rY%kg5T zO*E7opglQ;7vxueBn8X64jjQiwuZb^d-a*0D8VJyR#HR6dyFQ-q_t`l!)(AUfMApt(3+Be?=W~xpf;v z6OQ98Q)EAx!o^3upC`nlgzq*h>Q;Zb>W>Q#EL9luKWxBu3=7NZluWX(RP-dbXuG2h z7b(P2=pOI-;_I*5g`RpUpln5auD^03M-Q@XivP@_c4OP@1AyF^pjBLg0Xk1iP{~`L zl9%SVnZXP5NcJ#VhnrcBI}PR_TE?MeK~1$z$ZNw9GqxoADpB!m?9j1*jqz^w#h$tW zKcd2%RUOKH1txgHLEvZiztl+}WcoAp0$;8~@f0?!J-ckl-5K``VI{p5bUocVu4#?f zLGiLpb|udccjxO>LrHqL1Y}kml{4i*kTY>6uC3`MOFrgzb!R2UeKx`a0+jbVvG*n zIKbmy%hpQkrG9U{;;KaSNLaBcqV;{C(~CfMZ&`amjjDoN*GD~V$JSyG0eYjctduGl z1_S0+6&%}fT43MIC;%7yniSud-Y= zCBBO5=s2{jKC@t3X|I_t2{SlrA%8OQE}VsX9rgi}q5c%*fr{1jUM;GtE_;Um3=4lz zi*pyU=b+Rko`r8scI2OhhSa5Pj_C>yY8!iqB9&u}^0*y#?=ig;Bz1>_E~*Oj8%<$W z((nb>){+iB#wZpC0aA)FYG+bbITh!$d3kQ|jbo={)-(rYqiAOG{m!AdBLP{9O9e}Z z4_J9zL{Vq~>3j)gHk8Am0he;Mg=gJrx6IQ&6E!=?&AYvNSP-k*TgInypj%F@og5w( z)(?e=t3GCOuU|`Sa9p`Lb)Y-TNr7jy!cNmxQQoBaBN#HJ6-1EChz(g@sM7}t;xKSQ zEU00$yEB$k_nwqW2Y;=GtS6z1=sDj}2rZ8o!gzAo^W+oycNR)4g{|oGGUlmT|A`Pm5M5R{dIjulVH+~|b@r)^pt{Lry+xap zNqb<_}8}rekk21?hfYoCS zWOs3DuvT=E|2*p#?iI4hc9gLg@VkXCH?YNI7z_VVi+|9FjX)$&;j}SW_ z?FBG%a#JAr_^~L&72_@5Q8f^9vW5^$Bf-#JD{cf0yNwvO=}qL&1q)gHI) z1T3DUMzeY9F^5mhYNnqf(DK7}0i!1wJ}WMd3|~CK zcHA~DHC1zHShDDYyoLXv)luGv8SLwKs-Npc^T&Y?Yg($3bZfqRG<~NxGWS-5LPl;o z#*<__Uf0gUJF8%n9qx-q*~_v)%mzm2`+;;Ov+XACrO%i%&`5k0(n|Z0zC(#XNqKb8 z6yPSq+rGNJR6Nn@-1Vv8A*PC$ao3RomEyjw5Tx8GdcnGfj>W!szOxPu&L7c^HUvjz zb8OS`ph6fTrc%)8jAr~9skNf2H@Mes3nOuBKBqBV&pymE(@v=&WiS7}Gt>3~z~QZn zs-J*FPuG@VdY7MKYd^9+x>8}a(YDvJt;J;S-Og0@R4Z+7gnXKp(?U-n&>WKRBz0Bx zI(491?NcVOAG~uXWUq44|D|>uW!D4!N{!X`_^VFMwZs{PX+havN|cY)yBh$v)zf?H zivqaI2tj}!K&^O(iq--wX%gb&mi~a@rJ&;v^7EthGBr-iMoTiq2^q0^ms32K3hPkK z2ciOs&HwO)X`!;69WMr-d|U>Sl}6S|^F2L?4{46jHGso&!MdDv`))1N3KIuO{nBO=!RXS-W>rSG`V3BV zxdGMf*h`y*#RtUo`BLMG=0xkqXOL~XN$O`ERB3kKR1)}YGPjICd+@?)ZTkK6i@_bB z2+`|{pVFm+qko(7PU!PQ6ioHu)%cB0X(j-we55I5QxT}Pc*y|>qW9M3QpCdfRj*Bi zUA;;Ymc&ahSt&me%%cG&ty2d}mlXEX; zEp!5h9$spkU-hAcM^c+s9#OT0vZ>=JfNuHz#k>+klg(xKCMW){pgQ zlZR~Tba&S}ihAFDu3O#ajmSdtN_fSX0m85nukjS`UxZ#`24=RS#oxBif=rQ8rv#>M zb+m&q!X4j;x?oso#q(3LlK&@^eWj*>*9>h&6BG6@MxL+6g+&`yuRj#tTdlrR7wq;~ zuO1jIkaFc^+EKI2OoQAXg_ppk2G|@@#^6#>PwSWaw_m!=jxet^-BoD5t?==Z@`3JB z%SNioUkE1)?lho8YU2?3vLHgx+FE@6A>h@x`X#A=sQ&d)2%)8Jr05U5@|9lz($ai9 z)@absqRX=8Y7|hmAmWLqZhHea4L_qMCn_1M+@AL?gC{F8moVKSjw4=TM{Omkkfhl_r5neoj*4x3rqgF_Nwod zGk#~pCJZElG;}v-{roQkya8Ne=*uzgwQOi^X;;}QWwC){r94O$yLWbvHs~-GMaBxp z7aguN*qHdWefWpSwMJsqN!1X)%CQvO)4bR&)3%EgZDP3tYX**&4{jrhskX_OzOyP2 zdS9Ie+t=QxBv9>E_GBdOUE5s@;Gt0DtC~1q7OMbg2dJ6e*&)^3as6$Kg*|+^)v}~p zPnet0S*fX*qi)CZ6S6SXiqSzp!=ccy*2Qz3ti))}s@@0ky*BZX7rJ+WqS2Vs>JLElG^UTz+Sccqx|NM^Motw?tgD#q;yTVI;g z)(Qh4_iKzIQY>y^$o^}YIhyaQd3+#X#~`%^u+|&#Hvu*|clkHNsTWx($N>*l*Bh*# zT*l(+&bP)HxnJ0{6e;9$f3vVwi6Q;MVk5s2VYMxGow6txrm3a(rgNco87*%FH60VY zo>}$`GB0ma(RRqu6i44-!|Mg)nNp|xeeULEJ?Z`GiHts8+~wFNV(m{0%DHPRFHrmk z@MZXPOrv9j(A+;OU>&w<|HWR^_y_G>O6u$pLK(H>x=(4u{ITC1lKra4n((OWfuGEe z`kj%Q*ll4pzLkE9LO%}(YJ*NY;3rIUE(aK=dQYaft!8Ma52Q>T=m5MSD!V&tj*r^! zNL?3EZzPs<6ukiCf#(&LuDx`6aFouAbD*{d4m^(ZTVcU5+SZ8{V@&Zr4Q|91MNX-y z*iHTdN)v-Z3qY@H3=ez_dGjFx;i$l|n9Ake+?#+h%j#=l3 za z>D3IxoNdhCr~y?QvWEW0_l{88n|`eDJF?0#r5y50In@u1#B`(O#vAP z>my}bqG~ifw^l8RZSjNdC@pk|bMMvLzzRSh6I`4B0AcA=`+`Qr4KPlNqIuwaCbfHOt6gY(r)mepl~Ck8gc? z|K^W*%-p%>oY(R^&)4gox%YMcC(?jt6d;f*D5if5S)6&p1>{E&M|N*LnP0cy+dXK1 zFfM(AI>sqen&$FiYbCkl-dybc8yATsHP^m>UzYor`&xK$AwsuOCn@%!s;(D?W+&|J zP5A6I&gIx|8vty|2F;Yq`4qOC<<-CkRJBmHcLCO`Pm?yiNEaIOo^>fEz;`i^qh*+Y zV$8Q+pJF|AF@@ntE$9H#RTkEFn$wr=-nnz{4qN;g2=K5&mgdIb+mk8>GFv3cb#x26}TfLo!deu4>$Cga`Em5ZZ z&h?&;RJw4FYi}l?dgZEIgYSf(U=1FqIi<>(DsZ7~2w|wNXM8%jvkqHo-Rih3`ml{c z?w$WQ)P%e17%*1I)>zqPmfQ-Z_TZM{7I;(e?6i1MgNjqJxg>e3@e zHG_py&CQzT8%kcdFYlwaI>ZD#(8h9i)jwU>`?UFe<)oYezxy&!B`Z!|Je3x|@|wRIN-Rj}jzLkUUX zq15?;vH66UmE*(dHsl82sUqZHj2Y?!ADc~p%a9$k-hRrYz)*F&1Xc|3gt zaCBn6UOBANp^42`vo#Ef&ryvkp_#FBC-d9cI6NAwlS;fx`K(t)!h{;}mRoy$JbtJ0 zbpKIe7}6~}^!1@zOV9HU*8)Yo-pga{&(`xHH=J9J5Mvdg%^K@jWFe8D>9SU%j*KIC zkA~Fi$7@4d9%1<>^p@AZEO)1P+VZ3hWhyAJ6o!h~F?U@iH0zr=Ee)+@ zJrMLnFXBvk#&VfUM(%wE!K-0ri`KI3Q} zkwbbSAFXiJu7_nx;mu4EW)NV!}9Q8Q+jhT^!7s9~EqO%9|S$NOMAy60t z6UYC2Gmd1?CLDNCyEi&Gu3=rw3#)O+gaw?3uXzN3zvLB!SX1g-2mCR&}1=CAV&o=b);U|%AUSv49`I;qwU$G!i!ag1|s%1iJ>H{}( zoGXDkUS+q+iOr>9DzM!d}Kni_%Youa4e5$Gp%(>&%G31haAwum_%Cn-P6Y6YY{u2&8 zWl(pkx+Si<#R9yU_=hGQ^-~4I!7&5+-x|4xOi{8tVgdDtz&G+eMb~_-P^$*2q&Mvo zeHIoOnf1&SZ2sgAL&5(1_L>2MKI{TMUAzP3{svZ*!JQ{2yMV2_x~86x_5PhCdT9Rw zahNUhYC=!_GEfF38w%S2DR;FT;ufxP$2W7eIe;# z=V5o_py{OWgC(2m1p8L9^7(Uh(_GqYh$?i}kGWz4toM(=MJ#yP83~|^wx7Pw5aF#E z6339Y(O|eZ@ksCWwCo8q3MMC8I>JH%eB-dZ|sQ zb%`ih9;>?^IP*}Z`TTM1xlebGdacNf*7a^W(ZWXhZKOsKtar_R!gUnI@Y;Dq|L~o^ z0XQs-#Ke*vDxD8v3f}=BxP;MgNIyGg8hYY}EYO zP+}gjQo5TxR2xPFjF)opYgAZn-Olr}2W6R2pY~8QHQ?B? zuDG}nblk74!~yR4`NiXotYwJX+B8reR40+c;1fyLzcrElU~x7y%!f%+NKUdf1dKVCE-mTgL_k)JAAsVaQ?inKt*y z*D=mh9@PXTXDVWq=4uS*9kfTa8LIMXrqE?S%gpuEn&qt$>y!0Fcuuvo z?=>3bgi|YsZ(?H>Sv3_99G|BdQH83VkA{kQ+Rm;qftG0b?4XNhT@J#%eAW60^ZexuZcGYCHngMx(cRf z_ia(>q}2ccD}#n_%|c%7xA<+Gu^!I4+4e)%s1xXxeA^&klUk?Gsu)9jsg>7s(A&q@ z?&xz?F9Yw^jjSuQtZUlSx;T;JRdw$vx;IH!+gIe)@2Yu(E>WMm8Wk^;UV`ECx|ML0 zQV>q~-A0GLwX%N+jWHs+eOd0PGC9Y3DScY*+GjoS5jfQjY`;T=`uwX|X=Y}8#fxDaj#jn_`+#sCyw0zhPbz>rbY@Z`LzZ&H9=IT|-I4hU z>n@goaw3;{v1UZ{?emnPAQ|NK#Q2r*X!J&5I(y}BW&ttVLS7O zV5Rv)hwQwfB8M5RP^8B@_oaC_jQP2Bziz&3vTbkKGULxY;G8{VGiC?-UvhvS$pJ4= zero1;%GamDEN?he~>pFd+3<>)zkh$gBw_I z_&c!g!?hNXnl$r=iw?GG`ms~gym=qA53#{xU8eHQWoAY5+}e-F)MMG6De_zNSHNgD zZxDvoGTP&B4qZ#m4RTf=BCn3QB7t3*wf1z@&h^*l&w4>6i^8!*ML9QF0;;KjorNbH zn{BL?-xpSPx!IeP>(u*wJ;ZD@Cxge{FQhI!Rb2Wu9(z7Rv$P7go)jTlBXTMNBY3{A z$xcF$>5)?CL)Dt)F7@~F4KV1i`sN|*yMgK?f9J$8pvvS!AQyO8U?^k|!;N_`+_G{O zzpKwmd4#%^BB11nxT0szq}eUU=Ik_4?=hwJqA%b+x$TcdI@b;(u_K!ssTwN_=M{`b zU?{uQfnd4uCyF;j1W)CtlXl1yuGfdGuJb=}f70|CbHe>HJVPA6*r~jkqUX|`2B#56 zdh&cERHqx6oxF-ME~TSuM46G}h6u3ljnHoY_>N3lq2aDo-)VVw!xGr0u9+`nGr2N) zZ6|Fjn85w8A?t4`T{H3%%FIeutGfa^P z$CCoPqym^&5RS@HyEuS`XMqI|!YM=kh@X*!k>#9jn76NMhlEl1U$YGcrUQ`(D^?CC ze$XglC<^%_O^MpU>wT=_F6x&EA$vfB|5xW)ogl=UNYKZA z%*kuaovQBT(xGQ}!1J$49?}Vec<#0oA8ZSW3_H*Nn1k1wJ1XM>V36 z#OZwwvvTwVDZK5@5-a_y*&fJ3jQfGxt_huq*pEATy%S*Ez{3_Cpd=?iT>VFy&D+Yd z9D=A=utYHM@6Z2hZ@jG%kVfDnjo8TcihV#{Mf~dE4Il+F%gUkB0=u^#L}y}NG(?p> zxP5(~$=-G4?>R2SfoW2BMV8dVYAhXzw>_nEdh4{0=`d|K67{58wZUO7)Ne| zOnun*kI1$W{|3su@Jx!EYp5louniI2;7r-UUUnNKRmE-#Q2={ zh-uKi^5t{Ee?i?AFfQe$+kaV~tYee!%?2XI18ZvYd=-~Ke7yW%_O}$n09SxT< zugng!qDPEn!RB5oc6T^`oO4b3LvE#xlDnlqEaBj&aIEVU-k)uai zVvzX3njE&du+y3YSCMRw*Q_mKSP;YqV;Xe3jGBUWh~p*W_P>D@B1Iaur~5-(I9tt5c%yawnSa2 z1={c2ay%*wX%?|YhiTZ(aFxhgPepY<(NO8tAF{6eqoyiy;7`;2I3J)zyUuRmDw9vN zVq@XvBzQqBePM|AP9iYNzjd%m{3Pvg7S2-wmu36W%d*LbY?tEWiS!}qYW{*+q!cd! z0!?w0=|?wRzPS=XKjWJeq;Wt_f9B_zkbxbi_U3V+MqW^5oW$SxoR5mj3jW_QYb$Wv ziGZYtkJLW`&&o#)<{Dk~4+Wj{# zIHiZGQYZfdq#WPmc%9EGPor;_*{8uS(y)r6UEATzXZQa_Kd*cEq1$!ZYXW|u!8%Pd zE%S1-{ibmA3(c=Qbh`R2<{i2eZpu>`O%Bms$)_>CS;5Sb$jIso%K@KlvJaeS>x4Bbt?97%W7ekLD|j%fMlTna~#r>(D8 z0Ng)5V)^f5n1VW9co>IDm3mI;aJYTXpe+deMSLJ2-}r|^=8DiouzqnCG@g~bYk=}a zm~Q$cmMyQQOEuF&-+j)=-DO-wJYXWlN(LXvW(&5?LFoQRUNOj|&F)1jA-TvFo-_7u znr|8qBhhDIci%XX z$n%|%yA!zZ$ERJEE|0iV>op>Z+}HU1LQ!IN+XG1g3CM*vN9E7_9EjsQsK~fzghnyx zP81wCc;s3A0Hug%Biw8Z+@+HE+X%nDhFOF^i>T0ECFw!IZYqRUQcxGJZdk9+t4!X9 zi*8|#b3m*%uKYw_|FGP7N+6J2OsegxkRB}e86YFdwXp%KP5%|CwLz_oB_ma|BK{gR zPi%&L_h^9^ZW|X4Ksw89kO})aSAj&;B-)8KjZYL*6R;x&LUW+nEG7!p;QyavqONZl zdKl&gp800dBx+PqZnA1SYgs2oSH&AhlLq1iq4a?&m`qr&)25YY1Ys-XJ6p=t?D!pQ zn#ahh*Z*oWV>>1jHtUFKyn&${bnX&gn&4{Sy<23**>@`~4Rn`6(t!k|=~IE+FC5&1 z6pMwPxzv;h^M6AtGMoOg1BKCPc^67=^2;goPmzat&A;`LV1u80dk^{^`(i5LYqPFJ zm7fm@^%kLPbVmoY*Bk9T+h=NGvisfsX+HW795$fY55;1%Tcu;w4fudn$snj#|N9WG zB2Q;<$aa4J2Rr5>SZ;H+29*+jG(rSJ%l*GUb4V;q>fa*O^Ml*m8@Tr@D?)E$^}qm( zD)I{Zbepkw)x%#zRM=5tiyH8-q3U*C z)sWwKx^&m#VKg_d{=+MM#K4_%F;u#w;V)QB83}0VY{@zN=7AE}SN3g<4M8QwMPjsC zK;!HSIu+2^2C(mwhU)GTi&nxR%26kqU(#64O-L53(&dl{8cJj7yT7k8c*FkelhSXeq?ep^F*=pMYM z9oMr)ok3g3jkjB5{YS)qdm7a0T`X2L^yyH6As~ydxkuJjQVq4jzM;B?_=zm&N2{ky z{z@WiAcp?&d8`P^=vJfG7DfDt-0cVDSg-VS5i~@Ry-# zBwpQ~Rx(-A{-4yQ3@-d-{#O$?X({_hBQkyF3YBE-XiC*eLi8&e%tQUFYYuWB^|Jr!Sl;)46i{{{dOk B5dHuF diff --git a/site/notebooks/images/selecting-high-pearson-correlation-test.png b/site/notebooks/images/selecting-high-pearson-correlation-test.png deleted file mode 100644 index 78a4cfd98af2cbf7cecebdcd57a0bf8b3c559ec4..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 484028 zcmcG#1yodR+bB$z5>iSD(v5V7f}{-HrG)g*%}_SoD5=zdlysL$$4EB=(lB(4)NnTX zys`fCzwhIB);Y7*%-(bFyRMG=+HW;MN`wT|1ZZezgeuAk+GuF_VrXbMH}PMOu!y}QCQ(adli6fO$#p4M% zVw}WN3DJCb6N^H!qF}*Tvpj^*j6w?$GI>if!gBWpCqZw6;b$%vZMg}^O28q+W3BZn z{S@}z!*-_`6K$K>fO}9;7_Ix(ikfBECDl@VJXJ0@1f53?({rJ7W#z4rmewL#a{I~d z%A9&g_p~FCKuPe))kJpB?d5$mxoif|%G$m=gA5*;c?X%=0Xjpk?_uGoCDkhmJGCbv zBz%dBp4g@Fi@MnJT(xf_p32W+goIP4aH-(RT@ox)VK;dee4NKEPk6&ejxJ{4rq|B<;a#tXk6PCG|ef^=7n%hqbR5X1IQ z$@lQz#tl%dZhPy4-`y;FFw3PH_2z})jUU7hTW0TCCdS_*d}}b7^QuWoq@U{h%gMJe zLcHgbBtICmycO!=HPp#)+!$VQVP&xPuYG$^jJyBvf%%i-;b#fVqKX?gN;DEa8>ynC z3>cAIhsT%@h*jeCclbG=UsdBovJux zE9-p|Ia($X>~pT8@-QJ4&B);vAvPD+mdX5^_t+l3MXuR?d&z$Ayn3(h;OC<0XD*Ce zTtS*jv7)?zuDpQ{(-)1NG7KVPv?4n=ejwQMq`|D+M9KTlNE(+ZzHxT_ zz&~|;z@BX}#q0Qx0CcSm5Ab0l*E~b5@vMy`z6rTqu9~jQIEhk##A5a{8I@e2UcM4PD#dQ|RX#lUl{-_dk;vGL47R zB`kg%J633Ez-jL2Ug`;x_!9h+Ca|L7@v8Wk#IYDe>uBXhe$V_+NCsCX@2z~>WkJKe z2!u~c?!ere97^N&(&DJC>D}*$zciJb zj+0bo8-CuTDD&~or&kU+iUkxBq&C^yLN}x2*|Im*$*V|s6jMGpes}P^Z5)&Nknf}D zccLTg=@|6K6eAsv`EeuPmX|AYQ>lGro^&cEFKYW#E>Vmxv*u`sJ01b~_EwC}c7ezh z`}M-U7;#n@tn>EL)*hX&^n94bSIK>u3z|^X`^=MMQuwrA#2!jK<^1?U^TDK|n1UE4 zQv3EIU7YeLi!s^UqWNRSXQZ|yx5*~RkR%$UGjC(wUNV!aWLaw*us({f%MH}Vx973H zy@9pCSr~Ej;E`U3LWdEHGKYH+*)mR3>Qis0!cb)Ejz0 zSFs1cJn_j%`X4AQiVPhi#_dbGJR`v%+Y!H`)1#6j!6RPh?3!v3B}tve#{sq8Y2I1U z$3Z;-M~Lm zJw92V`+lAa`%XLx8@4Egf4FKx6a6gsM&zBt=k|rj2fMD@i{@YbzN)?6r4O1Q5v~%9 zw>~Ug#1aVU=e4*OQ&bflrH+#HC>2M@s4!+%x5jbO;b&Hl| ztL59gT5;&}NZu_eE+H+UG!!=UGTAcnD85nRSh72#=^W^sKdC!oTFWvm;9Ti6F=JjX zziqlLfZ*Kboyj=?pFBK2KG|l%W0F#Qqo^BGxAI}N&7Iw~z-!%e)a#}Xx4V;Tp!l$* z^+ERv6D;atb$K&!yJ~D>p1nltV@|lDaFqYH(^|hrsb`B!s)U5yQTk9+jz#@sBc7>9 z-K=+BW3-pq!ZhM+Y+z3PhyL355_mE5hxtlSMBzZBCGWFLB!lnIfc|~4&HbT~0YSJ& z>zai*k-3q%WOGB8Va%<_DT&qc+V`Rp0hKzlD4|_!jNP)HOiu!_mW_@^19qV^xsyeDbWUjO)dxeKRssw#3JE;zRyFs@~pa zrFz_(>FFfkEZ}nxO8!vQvUf0%Z&n#BTq~QDV4O7FWp${3Xufy(lj~F(!{tH!f>ElE zv7fO|gU?y`QRYo}WO$_cJ(D`nPfkH*53AMl;_l*YBh0C{QvvR}?%M7|5mgFC*~z9$ zv-b7dmH}>W>#V}p67_V>n4iWw`zgcRAtd2u5ZgcN(6mb8%FZ^^f)IN1Y zVZ<0Zqi!K|Y4lWc(M52Ns)=eqP~CPgZAv^s*zwTG+IthRD!%FwzUDbVnAvD`*T#Ij z2_BIi-WxGLsF4oWb<&lp{khNA-a49GkfN*#=Gim;>~^n3v+lF=(E20C+Yqjm6mxD5 z&LZ~nig&d?f0X3}nH_Gm#aFR7tIZ{Y4HZpUjlDhUcMs@RQdeHC#27@HX|=|eS9q?> z=p`7K6}T2o)}>Y$H#AtO|4i+yrgTgxxvb!|nu}7s>Xp`er5EvOzHF_mwV~Zbeoiqb z%7#YPnH1d6Ebi+8sn+xAp!|^>#-mZo>LiV%Ijl`}#+m7@fz0tv&zCbjQ&NixP#*w& z%}{e3GPZYymEt#+B8VpE4A(Sk{bF`9E1I8_dF~!UIL^U4#53@~5Wf#QA6D8w$sft* z9V=fRowdA*d==f9+j;W!rxtf{meGD~Yb;yHHv_85!5?J%uZk{hoQ9 z4d`7_4RDK(|Ln#V&k&EZ?U%uHSsKriBk}QivbAOyee;6f;@+dxj#p~nluZ*e6XMIA zv-6K4wKZpbJ1I0}zNYpL^dmtRpLWM|CNN4fng{EHHWa@_ouuuHy1-Liite9mtownQ;G;5F=_>A;0Uj4DK&^Uz%0AnIuu-a zS<*MNiYh?8-gTdc1X-LtxcV9$(;u@Y!xH3w;!#!S;zF2zm%RCz{(`G&*Vj0;0^ZQ)sBIN_Szl| z>#sf_;ClUu1-{o~eqS-;U!maw|44za?+1*3^u`zafccMmoSVQivKtrQsx&ET7Xfy8t^H19A8hRS4KbEj^aprw$?P6)e>+9@# zJrA0cuLN-GY~%Tq!PnUd>>=SR&G@T_1aN=d%*V*^tBa?jG^3%qCWE|-yA6W~?|t6; zj4}ia3=C54*3Trg6%_v(4*Zj5wDa_ImEhy^@$upH5#)7ox8>s(7Z>NdFTf`tzytK) z@$dtCKK11RdocZ;_sY2$1Eua&?a ze}M%6$anpOkDvEG-#_ODhDu$xN@&{q+Bg|0*gFF_1J;n?7Zwte`ZeHxd-Sg*|81z@ zzlI73iroMAq5t;i|2tIA!^U0S#Ti)CQ|4a;`)lyOKm2Q;6yG)M|Avd-i2l_IU|NPi zitnFPlOZsClMxIMlGa{9Ll?LLYW3_z9^_7 zK^@Ypz7w|{tha4MW_s+d`8A`i1P}WKVf{^O>297anQZDYcclmMS*EZ-7pF0PL=)DJ zvk+@v@EJnTFz^^c7yJAdWgz zd^~>#M1YQ!j;q%=M^E6Y}c zQpJi^RhFo5LD+U1H{n~-^WSTdyp^NI0LAS+assvU!i*@fJiiqLc@>{oygfh3=qFNlNjM&P=x?ck-L&v?UCY%bu6gA zJ=9@bAYT)`kE4+lqH}Ye9E6XK6$mVGnyN`w4)lG4ztN@^5|VHmuSn|_03SuC!q$rr zo$R4u{$KPWzuwL*!T3_(lHJwvYh^r!C$||EtS6p`8v~onW>Jw%1`wncHerVk;47K= z5}PqIDlpUHL~3Xd@+_xXNMi}9Sl&V_7on-i>jSNI9R6M8`wF6Ie?*# z?hIhBOrl{U*Yp7zvGIU8&_(%i_(HGuo+u{Ef&p?1!nE2u!oLP?F_A>B%>dYZ8NTqX zYf`<7C29gD@(cY;O@WR9(t7&13gk}esw!k6p?E#zn@j-WbtA=5G63+OlT_EM0PRiy z6i#=z71aTRjXGeg0DYf}vJ7gx4be%5x80F~0KI*6 z#8|LM_9z$@mK`6ee*^k5vgTs)0TY=s6e~Vr0LUfC7Y#5*$emqg18GIq@O_hS8~3io z`3}(8a@iT+X+l+!VNPJe>@V-^G61>@8UIDIY=CC$mWsm^=uqXO$=iR? z3;}dSP;)B*ysJ~iywODpFr>;JH4m`wq4->SED~j1;P!2Anz1Ax-p0FuKg>kYq2DUi z2c)|!+U&!&y^Ruvjr2HNuF>c6{|~UFj+ItY*YMt%bkHx7kQxD-X2isukw%9; z$+LU?3vp-|_y87EmEUYH0&Gv96a~Qm{uAVWt#Jid`ejb13?4)5G5~dax|A?&Y>19S zU7t!o6!3psvYM9XnoI~LvAm1M#3IyJjw{Cchu6#sw?l7ocd@%lWP~2hctVF#nXWVKkr}pmr+5F)VrI2E^6qUX4AVfzLq?`5ZC6q(g5s>YqI>dE z%sGMoDO;tDyoedPZPXW;xI0t))wt=ydFotLWsmuSuOO-@X@~@((Oo4~jX7Z4k$I$= zYGrgo@L+3*%>4qH@DzC)CYj$Ek|1M1L=7v<>GasFxB)A4WC{d5s%-TQlHI6>)`X16 zUZJ>N6WvW*>WWe8EGmts6?a#D#Z`)TE}}RMrw_ikXB&z9Sv_G&b|U`S3PGT7hT3bX zutToR1wSr>W`~S)oCkjK1=%FhPFV0!30UKupS+lXza?w!Tm1TZrB(2%$7{lz>%E{& zvYO5${GN_`+r|YF`n(f<7~N}=V#YcZ*}7)tINRl_R&^%ne^hXBJYlgKpn63MvS+Er zPVZmqfK;@8NG09&{bodMJ z8UHTmvBenYK@#khYh`#Fw5lm}I^h&8!nqJ_|E1A8uG5vaG5zlam7;0If7CbI59h{7 z?$pUUOqB9$j6jj2xRoT^s4($o#kDE}=+NT#rOArIs%ooSrSaH(t?|buKCiqQ4`$!# zIn4UhPQ{G1*L$p4&k)P@T;!##Wa%v8LW@y9HiQ94PH=odV+2d$6x%>U^U|)Vj!v8X zfKjzAHF1r@6@50eaL*d;o;OXoL3Qe>V62D8O)Ht_R%!% zh>~O8C9aK#^bQ7hz1e8cC3O|pJGM?*$hngeX+P2sI;bKncm=e{n~Af*!Fn8e`H#7q>9XB<3pp05f|~c69fla;YkJ}kb1Hbz!vIqi=F5yL z^XmSS=1*3-JGRB1q(11-4KmA@yeflCKF?n#EQ`@%{g!CuYpKk>-2uS|taU=^FE#T+ zf1$ga>|ks{GH~$f^_*-Em6a#-;?%u2Z(~@z1|Fyb>PeG)ox@vQmJ9Ig1FNHOtsKR- z=04??9n1M?V1X<7Ljwy4OF+~WL+RE)nrMm|Z#{ECtzy;rjm-~k-`T5i(!qSOCJuT{ zCc}AZjwm!ra2toos^jv0Ck0E%5Ny@uq`M`HnC1#YI%C$$@otYZR@Mpi7MX2nVrAj7 zk_My?njex7U9ohTem3#*fm6h`_^}-xtk4d58#m^|2M~WXR=SAuJh!0jpiuBAk4{6C zxaG%ak#CEPnkVKND3jo$BJ%7o-(t^}zvywr*ZDA&d| zK&-qY-<7Fh2al&A{l*c1xN65sn}~SmCXP9dX9k0)7PD0;c5j^Nznso`l_qmX$PgtQ zrY<_~n?v5Pt#6f>v9(;*dX-EkS<|;89Wc-;soUHdo~VAWJu&f)t3=>>Y@`S()2H)} z-_}lJgi26bgbzzJy7B$f>>z_)4^SNi31yrB;rB3tdYV7p4D9B}AMx^?r`U0DO#srUH+ zktfTo9x0iAGOe#j6b<4o4@dQN-C|PgMqQrg1znzLSJ`Pf>acoU*f4cy z-T5vLcgIdYpHlF-y!X!KBUMTsk`p}8z-CHHlXzv%urRzgg{Gx)?o?>_N2sd1e9e~i z$!!|zEo6Z6jP{A81A!RCb5UHcXG@&L99GEqKn9vUXQAo`&3hFx!eJHO-vT*Lm7T72 zn$mV(>s_gbuxVr_qIO$dPYVBF8st3e*(xgxcN-GPC4!aSg-2Lq{Znyw@HJ-jtI(aAU}!*Ukp2sJ}s($Z7krD zSE?*CcDBB`)qt>WugU*`fK3B$d&HaE)mn4l}F9GQ&*uEj*<6T&s$ zhM2oLPzU9-@hfHds|@~94(A0&RdsY|j#5l7E`8t=lw;p*i@>44>T?%(iy7>*;TskIgh5j$IpVir~%|92gYB#Pwj}Fu>?9X_Loi1ROpk*v; zJooPIO4UtjNez@R2EmED+C0s2=(Nr9LRo81K~DrCP5a2*0bh*n&>v*BBW$xuG~@9} zkK|gqQ{v~X3-2vdxs9;C7j{e*msnhikzVyUUkygpi^uEOTUkRl$3C4pAxrUUTgh6l zz^K!-@t>gK&kr!`kH<~&0%GnyQwF^LTK9|6*XO7`pN^p&7~IXWMdl@LdjGjpK=f>) zXqgns&?9SxKe!!ZNxWK7HHDXDtVsWOYRBx>HmVpUODmyEL?^*rlIgDsuDgd)!UEZ& zL$BS0h0wLD%!&ND2=~*K6qDM)z{{syMDK<6FW`rnfelGkX)~%-XCss=y?3Vgt}ZUitoz?RaG#G=2lu3?9V|D@uYF1hs@EdA@;AL(U|~c=C&RiW63m55 zbT@7Uic07kv0xRiu#FQy&e~sW{V1Y4M;-2Ryq9n+w-!=N{A5uSW|dZ5ag}0W)6*Lj zFKR%|V=mR;YZ$efd+TbX)UHo#C+;zIfIqh+#l#yx@CyTJd=x03I zvR#@SqxAdvJATJw24v|i4AYn5Am%{>ZId<*t5-+bb(-cz)ozXa-GkyBItcwj!?Wnm z7@*n}7zd$aV!W40Q0z;tnz|OU%z&y_rVuZjjl;zwT|sp-fb}*!Y3Ye+3ej(8 z>db{*J87~LqSop9`GD$~^*gklYw$YwE>GN3f_>v`mgDlU)Zz*iKY>SC<$K|lwLMP} zcT_trMx22oVlrO*tk%V6mp8@SpF?NVkEv_l|Jr5fi~i*@L^P1!bxoDr4XyxR`ui1Z zJV#~e>=MT|`5oMWalqw5k86zba(JrxOV+ z%mT&513ug*{i2;-nf2%BtMJ1H&+X&|z;X=1H*}yjKR#|4C};BGunjY&V_uZTaAAh$at0flzUWd)gVl?n|!P8tl3B6XQ) zKV|%7bz<*eSg{TsKh{cF3_l7Yb~`%?7VtW;6!%8>^teGNds0u0nS_pBQyEG)Mh5EU zs8ZE)zzCx+*LY1?0+4jEoXDiujy}VI^Jv0xiAz)y>FE z$kpjuV&s{1vZ)PLL8@rUMqw=<;(X)w`@_=|cQur$cD1DnET2g!rf+$}qI>S;jb0Xe z-Z(-Fh;5{lXT3Flm>M+F(N1?w1{VB5+8^;EP$b~3pA7Z);UzkuFvHfn;?J-?^QN$R zaZ#-W=0RP|l-1GME+-PRXma#Mi;Z+t6WJGCKkLh&ep*d62h>(MPJl|APmGJEkA?rz zk8mmxK@dzz0$PF&Jy)Frs>C3R(!#H$$S;KtUO#65KOO3Jj^c#Ldfqt!9gCfxiEO2_j``+HtFKu84K|(jJK$EY2?fnphO}@v@K4kB(o)0QTleD`aUue7xA0 z<;HE^SckEw*0m(htAKg7C7EPtfE~^4()O-7Fe`ezYU+;4{^GM3<(7+6o2Y#{m&Guc zj#C?lbL!9YD3(#=**APLm~5GGtw^Q;FTCg`2BB-#mH8NN`%cc31#)K%6O^GnP42b9 ztNxuonfgKu5@z@ye{?UksQqjCH(33zn2|X&WRRWIg$0ZW+P`B-aI^}9ugb5Te`mbR z7~X{>1c`pXKNKTNpV&*W6KCSmv-DJQ|2#{;uai-*cytmE%RaW7Q@>4RLXD#J>rFZ% zi%pa;>g8(qm_^SC{YlS2C!XdQr`b)8M@$QzCH&4~^uZ_=8?|{BAIgR)2^QEE{OW>B zC-QJyD{XUvozEz8AxY1B=hR5XqN?+znXt2A2W`XbbaE;N=x_~|BJ4QcjY%xU@RER{ z0?=h6`RWNJCbhj*ow_m-!wMfrw^0~$T=O-zTnS_sbdTp5wJ%C91eOk~E1cJ0rL`^d zmn_RC81>d)`Y9@qYhO#9wT*26#dQhu1(gVA>Zk)8vaRI{deqpOc+#S}F(GK(O=6{* zCqG;+kK5y#C%E8b7SI}WxeSCQiQqZXyF-krm$zIicI0~fqE^r4^#DI>Ru#GzG7@`3 zr`5~ps^V%*_X-J_LrZC^X^wvEWL=Dz?Oz^%JfxNmn-)$Z4zeC7pfeLgx*RRO( zRlOkwk%8aAciJ|#AFX>!-f-!Cp68!zdnXunezsL|V0YNnzwtwg#*~tATLVg-DP^C1 ztA^9&Md@G+`wG68UI}YnBuBrtu?itN^q#!uC)G>yvH7dRPZn2)u%c44s5QL>hNnX3 zObg^RS?q*jdHL~aKwhI^&Xfz#M6n-2jwOx42ovfam|~}E8yEBhmEm9K6aKQ293+q1 z?6df%htpvOs;EHmJpVlHL88d&5;*F~i=-Om{p?zm9?RQTn1vLsy~)A}LWKrBwymvE z_B)C(Sffa!&3^(lD?)}b><;SziDak!Dn~geRGtt4GcFNrthe(_Z#Dko zM)Cm^kiXeEXPp&{0O$_8H+}gsz^tdmf`C};c;|5NbWOLm6fj9lig#dz7X2HGq<|@$ zgok0@SgC~rcKm5imiy1Wr&iXpf~%u`d8$dVruFV@=QH<=*?N>xp+y>M+OfsjVBbmXb3T-{dHQ)4X zcIP30?b%$t@)I`AGB|7gq)q}a%ATM{hfex(0m+=G^ERX3!rv;#{W9YRV5Q=ptB=pb zg(R@FU$X*Wc3LpnvC_v~P@r+=3;j<996YCo^Xnr$A1z{fM&C!U_zWVnp=ZsR+xl&V zPK95zm!nOjj!#}L(!*i}uQHNsjfL(9sY%IN+X`WTrYA;X#ttuVwa@vz{5qRfDz4g| z+Mn8)sRWlXtiL^w(WfFQSctyd7pB%NIwgE0%$0E@ikts#34xc?T{k`ek7{Css6pX7t7eW$1H&Ds^CGn@} zn~30sUbiKgIJ3@COAVPJ#yt;Cf=fbrer=+=QnJT&^I@JF!z!7x^(*5hH&pj6JD3Mp zLY1t`!6wD6*Ivow5a0?UJ&SCa&yOb>56Xm4nYGsYjD?1U6^UvKyMA#;qCM~Kdc!j3 zh7Wq6v?IKvuI|k|ZaRJ^k^9ZP@5YfA$WMbouz=aFP9!J9{5`zL{ml9HHo&W=d7xoJ zSrCnjWF`tJXEpy&L&a~ao<`tN#@b26U!m=jOPSRM{rQHss$>Cp@t+vYwALIxWJ{)kC%Ur4clD3$b zM4BLaaonD{(Yi|5N(Fl(CKo;83{2f#F1OrQU7S&D>j?x=F^Qc9gQAikvM6?#?2N?X znr`CR2*(EChJry#)z&@QrvrwA&M)NXgNyybMwU-aU2p2=0>t-aDsBuw^7<_gt-9;T zz=tA_DY*p!1_(SAC^RanA-`sT#gkgMuQ=~TeWGV~1qkVYeE$W*BxqySXenTfC}mgA z;10a#zz|;4VDZy@jHm5Twbg(M#&sB=M>%sV@HRY1(jPEa2gmBssfXy$!>jN67yH5s z*4Sx?vhk@7q&I*x$TG!sf{Wo!=rVz12fzV5-xE1Jo^fb?_^>R1k~dH0He!~4#U(Av ziSt~Q=J#>VujJR?{NSm}c}N9cdT9eGN=8|#E6qpp%U#iTk2jbLF*JJa0xaXiX_j9+ z>uqw6{toBgO!oV@>i2<lWmu0(T)$Vt0 z{FlD_??2z5Adn&IAeP^<^(Pzr7Tp#GTWgG7w$6Ae z^c zoe!=F=1&Kvy>c}_&Rm*v)BF#JP=i(De6GJb@SbijH~W^m>QRW!!^Qk90h{<_4H>$< zhf}{|>c835U#BwVZ1)|OT~D_s%R|^G?YlX&FhTe)(~f#olX#g{c6jLb82@LpeRl6~ zH>&j43uB{H>E;;10JXmWW=oZt)IC`4Nn3a{{GTcK*U{I%5MinI1m4nou(7Q2AVatA z9#jFyJJEoLxg0NN58YRisGc_{Q~ypL{Rymc2)uNuQ|L`9DfV?T0=8xH;}9M8Mb_qu z6iLr`SCEt)1@-@crrCIQ`s>X;=PyQm3XDD$;G{n!6tzDm%4m4u87lPKh5HjoLnm+f z)@qc+;))TlER$D<=!h+{+ORH=OS*^*{?2{>Z&>*ke*SS6s);SMkjs+zU7LM7ht!oo zbT*ewwM~VB{ZCl^myiOoNZ~qMsSz2g^9?wM{G#r!g_W)w8uk~H{Ovq)fdrG7Ab;#3 zB5@XbK#Sdc)W#tLf0r)&$0EnMK<7H#p!pDlUme8`s1?IuvLl1pX)@OX=p*Sro~4|R z#J)|X?7n6H@#xxTHdX#gR=x?X_3+~WKay)X^ zVZoZ$Nny%-(A&9MUcPOvf^X+KwKgTkO6$Z8I@DDr|zo=PjI3;4k zo<3PqkW0Km7UVh3Nu=p(cAqw|Ux|NAK)Ad`hUPmhE zKPJ$(w{o8WiA?{j=(ahO(^W4=Ilq`I=Un1EfBw^iZNd8O{;%q-KhQNKd>HgMX*n@* zh~jGas**+cq{xy+K#?*z?v@WcAGUGeF{l0;VgCDsvAP(iH^@(DIjQdfI?LLt!Lf_d zk)PKbC~oR-{=xjGs#H57#D{?+K2}HQT$T5#?uMA3h7#0u-fLV!XOv}PCVc5nt+oOq zNz@Zc%VhH z@6Uj5lGEDHnpAK%*V|cfqec#zsGG@Qm znoul5(uddjZY>1i1*rJVs|HVq@Uqos%s)1#<*X#^ZRB<)#b!zcYa>NnGx4nI)Q7?vg#GXKZl@}~h*Nq;NSVNYWdTNL@{HN@DQI&|L9mqu8ee z4NT65r0>KXOZWSOJIaaC*>IKGlK5H-8A~#q*X#t{e=-#M0=W@fg1_0m8_NP6cvvnW z86ESrZz)sW%k&>I3)b|~-;8Toy*EDD zbZ!5(jaqoL+I~C*ZS8XXxJHi}pUiH78V8q4)0A3gRMv7`Ivw0u00U%C=e6HL`WqPP znqnhp9AdV9#x3Ifc$?#RbKHHjoU{cfjN9Ax)BvjHPRm9AtHzK+;McuOn{L{R^ve#d z>6lM?;jI`7fH<3=p-GfBkb0X5!`@#8joqltA9{ z>V;?zPwgx>R3j_-IWBQeatXY9)-S6s`05fmK)W(00&LOO59 z!=AAklcG+6;N2oZPGIQ zNx;Hmxz0Q}kY9!$!}APduuAY`sUBK&B#Vbn(fV30YFKj9K5yccAK1h!d|b;ig_pkQ-{)T0_a zGNPU-rXPNri+Z#^ND7?Ybn5M*G=qM5uDoqLM&||Q0#09(j%h@CYMkMHgBgu03m?C_ ze+ z)ms{w!0SWjMW=baU=pD&t2)1We4NaaK1XB*Q6Xu!0JNAGfh(Io`UYzFY4>+8# zmiC>sOjr3uEVin7?@U|$w2g1_^H96HB<)@6h7z5P{p!{Ff>R^?rx{8#I5ZG>NpuJa zP)QZi{?Ini=^1^RtMK**P&lSP!E3qL(vbO1tFV2*@5;o^7s3J*Iwmc3X}n`%Tj@zq zQ%&K#ca}L4upkmP(j=~v_axMW)~w91ej3sDsfiKd?{`*6xw_e@ZqXCZ(l>w{2OQG= zhTf#zgk+w(!YL_bhAe5V)0W!~Eiw~l95A7m&tFIGy=Oc;wKc)MG|D)ex;YiRa9in$ zSP1;2skY2#bjkpz_)r)xHRF@Lv&%tew0Y3SD?G2}$Kq2;llP?EWO+~$Tr28hHs5Sp z)}MJO+af_@2(p_0 zvKGA5=3@aLvNE!Q&Ww)g3$ik{%3!>DoCY@~+NgUfjbIP52H=p|ulM2h>0h%tbqi1P zLio)bRF(wu*GN@{&g(nZEwXO;%Z}vMf1~9=0Hu(E4_lCZNd`EmO533yHU~$7&a}|` z{>x7aHCxhi23xX`KXYQ(d^5eqMd7w+H7%OH=SU5+R_9s7yZsSus{mxoMerE>w+2&E;P)nY&QgJqTTr*wU1ObvHi zRhG`8&*ZrLA!ke{P#s&WCK*EQ=Dk3!LaP_In_lysJ54{{m^`hPd%ZukN-3~!Pptw)IOKY))M@WhYpv8Q&dj z2)&oGN!8HoZF&C6%R#q{ajwBzAk3uwrP9&m*-ilcy9bF^sB_qShOM~sbk()XWq;Y9 z;69Nt_+8C?-T;Q*z#Ulc5OuUO0<#qo3D6;=YdHt`3~9mV1N*2PF5u~~{*aVrNZI^u zPxlagz^ChyhxdXW6;jAAeHAuaqM81Vi+?#3sYTKO$4NlZoKaf;giCE{!qEl{S8lp$ z%WfVs;5k2&Y~sO6-*k~_%JHFj(=z1Qn=XeBPm_-wlk0FE3l#_leIYk$vMv;N&a93> z7U86u2bXxX?ZGUWMvry5*6_?6po{ILbMpcZeH)t_g(r@ija$u~{SJ?LgP??7bNq5E z>Gg+s+YuSGJVFbeh+4RUl2&DyZ8XkpeW4Fih4PYF0XuUd>9&=q_u0E?RDx^WS-}mm zX~%v>sQHQm^cyVK%}rYpJcjc_-mdsNbvpT)!8%z!?k-k>_N^E$g(G8B{) z32LM!b3ZmYzu+UK9^gpd;&RjAr?QE>h+7Qjl7f8*lQl{6$iS9OP0jX?VkYn0e*q6OqR=Kdo-Ser<;+7xbM~c4X z4eNNyCqF1l4~aN7)`bkr-#>_^J0EB%vaR2E9+lTM15|y?1++LuZ>~()aR)ZrM6C(_ z5b*~Ae&BMmPr$T~n^~p^B-wM_6D6kqT#<0AN3t1Sp+8_inA>#K{#wB9vmn~0-EwvI zV$EHb)>ToG4a#u8Pb*de{)oDeMUJie)ZwGIPL8B$?A>VERd(BRdvu;}8pCg3;vOZ1 zBj>hiAqB8in?8fy8_tqAH#7X{5kt?^SH{2@q=6z^mxzhr*g!GQT%X%P6MT&bfsHG3 zzC!boaE&xgXu;M+@0QFWD@5@oy#PbOlS~-C=I?i0{-ggQ_YPy|ChMchvip2&Kjwb0 zF67GkH&7JK`;UoSr~fuKFG6QaVtfl`tRGX1kK3vD!%83WSa&I2uGhO8+b0K zyr%clMxAYy#WoJxKWA0HvxF};DT%kDdts7&%XOHem|FH{f%uAU!P-Zid!RH4 z^Mkd<9xnsBZAIu|MJ9S_lqsk1=;0J}X}QMaMyg4|SxZBDe9mS@3#Rqt-cWRdtNtto zXr4)oWwZ=}1$_Z{n4Mha6-^hYbNWoM*SA_i$CQTXFU$7U+*Ni{)9FD2=CGW|7x^HB)&ae&XNv} zcg}JJ!s3e1psXs3$Jfb^I<|yJ{f`%Nn`I?>F3p?mB{mRs4s>WM3T3jxsXguCOj!Y% z8Y^+^3eg!ba(?b74LPeCExh`0vG?RV3ewliMiDYG=OEK)=j7tT8WFIOVWNDI&2js`C;I_h7^ILhkMIaL$x?Vh?Q3~?r(r= zo<^A{i$Q7n$)^t~&_IbfUyatq5p&sC7B$Kgph&TM%)o-hp#Z4_U97xKm{M$34WOpd zv$vc{DLU0`XRZ;}rF4lPejTt=Z({Cro_+uBc+A7R~ldfCC@l{q+Jn zJE?-Ib9>PL>n zyqbeh^!nu&sIf2PLYl{#W5g+Po8h6D^GT_~pQejw``ADiyvW2)3qPFz?^eCVw581r ze)%%kQ=(kJMpZR5?J(^@r%Mfl4IWrI;C9n&$UEZpU>uE0|Ken_W}6Bu&$BRq*`&s) z!UY~v0i*zi7BYPRF8rzgF;IMWGj@C~zo63R!^DT(=NQ=FWZ6iDsS|qd-qnE?3x%Wl z4-=oV1b0ahl}55LdXA$$bb1d8|Z1W2eH+gzQ;Z# zXOB6|KD{_aTp>p}rm7udip?tr%gajF-xx(u_KN_o1#~{a1c7&L2t_x@)ATsuwq9L- z=M1T^KEaO9z6GuTN${Uloq-TQ&gd-JEH9wm58p? z_o29sIXKU7h^Sc`Vw$gC+$uJ&@7`x>08elN8nI1n+MW@SxY-<2K7w^mD(Gf(OqP2NdK%C)rtUg!}d%zL^Ux zAGvpxR#)oJWIHrB6nqtQ5pCn$vL^#G$?9I*b~S2 zLfy>ncuRz3%>yRG!&lvSCwW9pYvq$yV3-3knTtodv6Yvax9WDa-@(QjX{lh>e(!Q!Vz_DBc;8M69`ZHa^K30OvStRMx-)q#>EIXHEXdW^{q4*@xoAI)S_w zf%&Gv5btx2G*~qdaG%X;5LRzi!q=KnbKZU}>H&gJp6wQR6(tX&RzxyC0i`(LU8^tZ zcOO#z2X*_8tnM3p;6*`xH<~cnx1#e1mZeS&g1WE56Q<5sq2X%#KOW`zmS=0{?q#6Q zn9v9-WTVq{d1`mY3`wk{;}l#Z-EBVWUFTn-O%s0eArvbBe^u$I#3I-cOCmjIY`vQ~ zp)=-IGjDD7gIuM%0n@fe%+fA|b%q_=XF{Kkdz6^^hhmDo=gYvdLByAzTW2lu;&bdR zanpr941F9ax{hTtmo~d4bXXz(i@o;@YiixvMqPG6WGPLNvJ@2Q(jhbp5PFr~1f)0V zQW8`Ilq$UwdJ8>-P7o208ba^Ag%+tH1j3o@_P*!ry}$MSJ=e8IPn7Obf@pLq7~&RbBGhxiJPc?OiQTjR!H#)(F=K$iX$Ai0t_>R+rhTeR|x<1&^&^U)_?<@Cj4S8cS) zoVW$l-%y%Q+&|ghgBH%tJ?On8SpH8ffR3M5^tJjmR)scS2RTTilC$X7c}9_Wk{ZA7Mbnmsm&B z0{dBoMJ$Ea_=rP?D5|{DCF+)w@?l?r?2>TGXM{Jc_RE#nS!%njS+(6eY8nMjmMjQ` zb_2tqxtehkS@XZasT&$T-)pQGIS@)@#t<8csGH1u?T8dozl41|D@NnRMH~x_|x)vK0l{%X~^w6pvCbbmRn5&xlPu_bkQ8wv!CBi zOI8Y?JkL&F)X=!62IE8O$8h0>qCb6MuSZ@|h+;pBw<%6-TsIsp($|)HcQ$>Qauj7D zk|bqOou6j-tex1D9qz%(hl}!)NV`i8`mQ{}rQgGja;kgO42iK8E8K>!zOTMg^K9{J z%c(5*8QH6;nT-kPOlqMN%T{gp+`d7L>aL!fb$|cXyOyh|LH7nlQG;IZv&k9bz%Ydw z*_WL78QtmIHiH^|d8(u_Och{Hf2;jKX;x08Mr_uH+-I__!u0hO36bqLJ>x}nSq}+S4>%CTlJ|kA1 zzZSZ&4&HaADf)7ip*V}>T^N(tm;uF+LGyfTL}8X%#_FU>m23U#4k9vX?Jsbte;NZcD=nUIKDDr>E`z_bgS>UxZaBMu@vs~ zjsfL4fs6}-F(sSz0KW*nQ}eB_;C8?giVW{_c=~r#K?q&XV`G!6OF4_}VhF#ZcViV* z&6jtL9BE|cH*>u2Gl)o0RnskUumq;O3i4IK&I3_lAK4#uzjED#6Kk)ePQ6h4Acv}C zo^&*=xt)4G69rUD9-rdfIqE91gC`mU5~k&Ly@|a}yT4wlNH`wc%S4m;Vuv!*Bu-zE z`^IhT!u-=?8*rG6>9%hR0ww8RZ}HX~RC$q$HjxahZAm|0rVBffkN6=E9XS&q+X8{9 z5Z@QwP5|}Ha|TiuNYnz_MK49wZE*Y_e_+83o)HN@~+A1YEx{1A60 zYvP+7+%$3X0_ECS#83gWIXj)^PfY>}m_(;P-D3gSzFnPkZYD1(qm7h) zIBkNR8yxYaXHT2iYOC(Mo#y9RuU_$HoDB2&V^dE$pFSB){jgSCf8-iSQ}fFJjpUZc z^GjET;o2@k1Vh5xHF=g9lLJ|t(OZdN2LMfp4izpLR{A8ofgW zq2+ZDuRZi1rwIA(uq3zTeBai_?tTw3{HAcG4D5Bf2$TI&`P$Cs z-^u=u%@=@(0}NlX=Q)Ajuw_5LNpZw=V*Z88xcakAC*OCRxbqU6v z_f>2F#0Zk-d-GAw5~N)HFipxLq6#-QjA^@H<`o4_5jq&My`T~B`7kT9>cgef!OxKW zoy!2K#O#as9N&~bbW^ZIKjV@L!M)xcdmNPAX$^xulqH+R7LZ#4$Y|ZJrmyu8xK*!3 zHFJegW{7-5IB4!QNl2^c(Cd-gHLmlZZHJ zcD;hcY^=Z?t=XBO$g90F?JiTSoK(8Encj~W-)a)wZzZ)C*z{?n zNif!Ptrb`u`p(vBK+v&aKtc8lD9o0}HsIl=ArW@80D>0YU82Ug?rg)qIiA*zxps9oQ5SUsFZSQnNxev;lvLKa`X8S*DaO6AXX~U zaF{Nq`BKy#W#phLU?c7M4@`U#{?MS$dy%3N=p#r#erN*TFr>avt6c%}(963|r2pg9 z7Qx@Gct_kYMM`Vqo3wZ6C0)y3Ju&TDf7mh#H2FJm&1dtr`SZFIEp{#qFi=xLP{c!S zr`7E;bc%Byn?Sxw^gfmsKl;h%Pvg|%^Eh=GHWWM!bah^pjXቶPVKd{qg_z+^~ z%)5QP?@f2ItW#IQ=%Cb4pmn+v`!cTSZZcr*=S34I1-;~^XjnvXw3yOku<;X4=6(MT zh86`p=60$(D;s>ML`& z;OGOmM0*bjfLXF&)D6Jc4Wr#Ivq`RM!$-TaBgOK6+xSObLO7hn$-r(OE;z>`|50j8 zrYK^v0+%bhc|3-$G(-z2(cbGnTZr|lpnmI`lGW$yNK2?aFVOTbqm?n=hCP905#^^@hd9`wjr3XFU+r*wW}pQJO@Qr z1>vs>pLp15Cy3*ou!2A_UbZ7Aor%{)J)p}T1o!ouXsof7ZS)V+VtEeFz38nI4r%LjM}OI|~P z4GXj(_r(0ZuF0Ez=!#X#S4t>#7ESg$#Ik9!e*6OlMCsW}$0pbFBnD2{^_#5H@(Y9h zsBtn`)*lUgo8IeaYQybl@>Vr-ZEdJJ>$gcFd>AqOS+7O7#JvHQsrq4cZK^tpr4DH~ zJ)*<1QUHbd34XsWT^glpirXH`<1F*sdJ-wUa?_n4n04`=_{4vs#YVpZs0-1q!CvKc zd$jT?o{S9Q(!;xa*eUEIu<;AdWlo)hQ=doVWV#> zPcV_3`TF;;%{2<3$qti;P5DxFa&LXgW}lrWlW6a?kErH-Nc@mxYkTkR%48EGkIya8 z!&e=LeNSb8W%C~`Wl^2huZ^{E%Y0GOwRtaw(SxMZD2m_YW(wT;$Zchh4YNhM9@L;tL!ulUcnl+Uux zHR7m8);#M0@|0#>ksC)wmh5&k;FPaGGGNddUeJq}rXjeY((%C-w$LXZcj2D!)IEtr z372AOUY()A_Sx@bX=Dz|9PFm3;#e+?DUamJPf(dr)O{gu8Pvosa`h=_;%EJ^l1#pU zQbE|MP2uC^W_vozO#<(fie&Dry#x4*P6f5p5F9Sb!1j((W)fO+1 z|C3B)Q?b~f5-sd#08T?zIuI1S7dlsqwfD3ghn{3+J;*j`zb+le5uX6m;oaK0miyz& z0NbY$$g||+okJZaPU0Hvl>KR*H0GHl`{Lj{prWtZRsLzDZj(Di)I{WRlM%wxVlNrM zY1fk@P_zXA>A|zm82=mW@mb1V67`d}h-{QE+RL&>&su$XgWI09l}1-|qORl6Z9iM$0-L-HIK&>tNVXrtlw8-nEKyb-bAI z=)PI^OdXWRe#Dkb`lP(ugj?f$e&|20Ui+yRkl^W`N5#|e*E?=!aulN@Q+jmT-8 z@w&mgve|iV)D&sEwkuU%ZjgSS%)N{ZjR?t>4Y-}Fl<*aW+y4qLLv^i7wb*{G%DieN zC~Cp0LO+@?BRN~8OW?Afd)~?cWK)7QmOyU0K3vFNcK5*z_<7?^?m}T>s+ikI5kJ~_ z=et{bcE7avx%phE83l0y0h8`;Pa+ch@Y`x{HSGC5^8L z+=-d2@&t;6CW4ML;V9-y`W_|_=+Y;*TmUZz(s=M1msXy~djuhbbD%pl2^JE;K~=Et z>tBA8^tN_f2CU1jdBnr=Z3siPuaB^5l5kXhX*uohSir)E{j5w1xjGU;%e#7A)b^Qk z&reFm^H}b|&|}}m-Kep-{gB_~#HRT<|5g@Zv7`rk)L0+m?Uol3@!EoS{!?u6Y3p>& z^Xy>DjZwN*A07V@_{^d0y(fp@TpX3*p8I# z1B&}Y$fBO80rww*z6z!92RFseVmx^X5pQDqpFrt*;Q9?w45Wk4r~Ey?CkC}_K1zH) z2sDjzH{i~bUIMMfi(!Q``%nVsaag}~V3W;t&+ly6o7w{iUDJC)?e``i^8}rN(q2S_ z=V9DnzS{lKF+F!O3HaJiW%>}L;!`u^rEAbA&;g+R#-^&*kW>cDSEnco$Q1$tG->}g z_`^RSsgc_rAnvr44ikXQdgpI^-BV?yX0vOj^Z0-dFJ=8%Yv>qnE3`fVxNqgU-K~*T zu>h4sbX{>uOF8J~HruGl#{G*Rx0SGc@BC&o<=}bS@Em?wV*~KagKJidwp~&N(Vjqf zbl*OT-k`+RqxX2L5&%Igv4~M1(_witJgt%c{5NNEMhffiY%CBi_YOvJ&#nMUX&(BB z%7om|m5{4po3dbzx~&Lb8L(evNQ8P;_7@``Y_5*>SebD>>H+9sXw#a_OVIjPLrlL~ zC>zWp%r$!}3mBvN7IigpQ#QbcUVcw^=4|feJ<*?HT@evqvqm$`F4kl{h_B}LbiX_Z z4BE3O?K2tO{Y|cw0hARvZ4%%Kx|lR+W8%Y5Q&?yqLHg|On;Wnw2sks)treu}(oYR` z(~6)qd80$k(d)Q_Z9SscuCTcrwnbsfq*d1ND~!fq@c4~a`=`^zgqVoO z$6_0U8!8wKb}A&|VGisD@zuKtRlM!g+5%BHQ9aD?%!%bOR?rEQW?0+ic$p1e_K#kp z0X$t7^0@k47EX~hgT|9 zZ%l0Gs%F6+_!8z$z5d2d<8b@L#)$1~aLnjWo)oBfemY@QJF+e@=3D15d5t5yU_(WNKiJeW&7$mvIl6&N@qzc=QdD<1 zGs1tYuKOm(TCCPgu2!(;?!Fc`!*I*V=TdRy6e%T;yU#GtxI8D3ooE<=Gdh#@uDUg- zg7XILy(1el@8_#9@&<$@N;vIoczsa#SyK(v>7lrVd$zLF;LGvJs|F@VZRXx9<8+KFf=fDi z?h0;0$z5?AiFlQK+Z16)wU~JVE2#zeJFEj$0;l9d1d>8@9dz+gF4{ z)MjbrZ%rX50p6&`<;3Vm+kAO2&PRms+a9G|i5d}1;^$Q0DA~{iv5EB%!wYg?_7riP z8`w}11}Ie}vZUG09*?WTD8n!Guas4TCf2)#I+>M#as#qXwTC)p=z*KBm1B0deJ?fI z$qLFGsSmcdTazBw^<8I9=Xa3ipD9y{Fa2d}ul)m{cHjL?XkL5q_RD1s&xtYC< z-6xFix3ov2kM`IA&Zlg?w()C;vXN4cAj^YVr|ca4)oh#_q}i8_Lc{hgyYa*x3;U_h z(q}(@Xl3zid1pf*`iW)cRQH2cIfYrg)GsXJ%(qF|0!Hzt(@f8z&-G9F*MCy(1MH`P z<2t(2po$jl?Cn>>$!8uwI6U1$H@8>4AC_tLB83D^32s{P8dgJD&f8sy^nEDFOiPL! zW!UkTrM|Ro%#iOoUq|~NI_|7oj0#;PtSb-?@$@EDaBh)y&6hRm=%{!-K#R&l;#B$m z=&$4mjvBKqP1tJiByvOEISxMJqU^gi(r21>^>XmdZC;?UziM+w@ku+%0$8}s4pkd{ zu<}4u%#`92^h&&%B(G8eff7@9;5=;Xy&PBhmUOc-T{1;``Q30-SJ6Fm?Rno$gL$)S zF;XZ-x53k0kn+Gpim<;Y_24;Unq7O<4XHtK(5)YT26tX(Edp>IndNZkD(FRTH)DNB z#3w3Mc9Y_QoA*J_N}E>bn;Y!ES}hes<~TQhh+uC%*?v8lp~V+|Pe`T|bd17s0)zys@=|;Z}d98bzSW8yKZiRW!H~L zquCmtj3217QaTH`#JgW#?>S8!AJkM&d<#X|YtyC8^q2q>CFPUqBjwWjWlF!uIeK}ZV{|np0tCE=a&&p-N8Se_G-3EX@t(5cy;>OGT zR~vZJYyFB{B@2Um1SvV`fIiFnDD&}o<3DTgV_h7u$~A|h%savKW~`D4pF+fXr^wNX zT8|zL>D2fGY}3XAPgsVWOntNBBkaVoC59PG#Q~y74}y-A%&tvTg}X!mb^KO6ccLPN z8kQP5>pXkAye>lc;Jiq1{abOjpkV)s30~6*#hL9HuqSTZUYh_JhI$-xGFox~O=;*P zHV9bx-zc3NhSV3oJ#SqCc%aJwbbl&5ims0^i3i{zfhXq9Yr-G3B26Hc`-v`xR=d6N zNY^$R>^WmTY7@?)g%$Le8O{12O3@yu={`}H*}iYS$Oh#IzQw5eR+nPD@zBn?F8^q{ z)X}4=QKw`QR2@#w_cN6qqk%>4Vu$Vx5aLD|kxp8cAbQsFtOv+yx%b^B@!SNM-)<-C zJ+b4(_YZmjLehg0LrGgzOPkqQUon8w8l)z489v|e>=D56-F-Xwc%X1G^m$W$6)y$# z;pVjNV--@5wc>M}g=Lp}TI1NE*kax6&jYk@+JYvS9Fc%wxuz7jyS%9#VVUNMga!mN zw5U6(mpa(8d?RLVxG*sOU)asLqO`|A&W3t`YG^!HTwJ@2=q!CET@7%{O2gQTXo3x z+4sGY;^FsYlU^sU!)U9%F$`PaOCrABX;z3|k5(G!QW05>0x-Bg%G{lppN?g)eD68)%Pcl+ zR@gs1XzepzxvngBvn*wAHRjHv5QR8J{Fl;ehWjXJg-s`cXg9V&I4{tsh6d-rz4XI5 zF4;aSd>^-TCm*dqC{`p?(?FIHJymuGpz)nnRlk|Bn~G)}E`6kBTVxjxyJH-8BpD4!a#&+OTb~cWveM3o-?t zXa=tJi=c^4v@|Y3I^6cu^?tU1ZSDI%^{jvY|2fnEu*<%z>dN?MH}bC`$ed^Am;B0e zRrsGv&a80~-*TT+HXF~HPHF2-4+~xUsIMV&n^pel!>d0oJme_&oWV--<2_lTfTH^I zHZD1t+p@~?!zrVk5`Hu_#`PUDw58i*&s#{`3|Ms8X@Sp$sfu9w)?4A`zq!^&2II#x)O3{v)pNf&Y*99|y_4eu4QN7kZzaQYxup zog59~89JOFe6)0Kc0`@*mThYoA|1?h46tX4KAYWT3>Hl9ewhtA)r_oUk%JR`I7?n> z`YpTfc%2<5=-2OC;^isUoy~=(-D9GeaBT@DeCKjWgj1ZqbuzBfO~GaOM;mY&m?`4* z{6jDJ!3YsCZr9Cizfe7O>0;sSqsdXzlu=Xv?zWm?+LWY&5mR1dt?rM{15VOKF)blF zraEk)IWA}MGx$}y=GV}Zomr=rI&6eEt8CSa;B1vr(xxvl4d0W|lAi?JevAjgmU7y@ zdeU>F*y7~<`|Cq*>imwPI(pXN9Vy2tP7XYxXGRF@Kpa z(tf^lvMjk(-40%4*3Z6A)c%wAPMfYWZZSf*gVDE&vfY~OGU{YO3DGms8NF3KIZ@m+ zV!-~RenED}a1&Sju1+SzTMaS$0WYoRBt=*f@OqnIC7z_MLx6a`e+b1K()|X1)}PCZ za_AF3yR!$#nJ2jg!JG`LY;@+xRA2rt3@Bs$vijR1Py>C#ZcvJoP=b+D~Yh3A3 z$ByCW*FSArdxwIVTq2LVLdkpW;3Wpj9njWoImryS(Vm0Rh=wlrDQk2OZuU4qWY%YL z(|^wFR%^xww&a#e#JEp6-4DQ`p%?Bum48SQU)_mbDEbn5c2I`vy`{oQP1(TU{=RN^ zm;9fr^Ui7nv2sLpwf2ao&L$S?qn9wWU~-#95pi0uy=sbC?_OVtVkONE5K)E=5i1C^ zw1~Jvb6OI}()Er6r{CR)rS;zLZA$UUi^Q8xXZ?tdBjGtR{2N#>B4X###q?(?aosQ| zX}rpz$y#naZxtii^Vv-m-#I5>d9h~7C%OXI} z?F*>=_cV$L^nx}{#5g6Si&W=O#C*^d{QQO!>&aCj(hz=2D?s;^iQW9;|G=8^ z0T;+rZPklcMBhlAESmP<T`6_Oe<3ed;ffg3G$(`#^WIG&5Dj*z-1+klz5eJu5HMG)Ze~nDO7Fz%$n78*-mr* zW&FV_X{FNp(i58qJd?JnmxpnpCsD%~`>OGYsCGYh9E7a zv{+Fg&Y+AMd4+k!l-vxsI$vJ_m+y&V;+^$_F=}Xj-lqK`vD)qrR(rkl?LmvLhp%VM zy=vL28JjJfxUYrM^%!&W-@|~J8^S+NeAT0SRD!SKk+$cvI2EKxwIL4rE?VXuv|M8b zm@H_uJUH@aZQj=2s0X_hPe!RUIX2*5kG5Ip;uF!5IkD0YwO0%8a39)Xykw*6MHBU+ zXbW_68Y7fA$yA?+wl&c)7`j{^%J~lFi%CbyzLm9;2Hea$r4VfJ&=S$a5Le#pY(UN~ zg`$~90MA3`GLvndMkv^EVtX-5;u}YVk()>0WT7DLxzV!1dslrZ;kdtbu($9f+Lh9i zRDe*4WPy6Dnn2YIyAL^AZT*1@8| z7q9r9PqFZF9n-%s`XMG@ci2TjIJ|hYTNFwkGNtT%BD^L9_u)?RQXCwr$6Vb8>n+Lm zR;A0*M8hYNPP^b`K!o!hr3>?-jw%Tore{{wQ|1B) zB}7`!EpZ2D9sOmz3x-uH9`G5s0VUoha}&sx#~!fbW`=5TEhyr)rSObRcB2bj!|Sst zx6s}_C6lK>JGPKOSUpP)7n%7fIPKy)Sn7Cx?~F8i*6mxi))`0+6A3ldTTd*j?BdvN znlK}_Te)?c;Pg|e5jFz8eMZFerj0Dyv2xV7FM4`0ORz1zZA|_S!^u7Z-Ge~)c`dI2 zkyrp;M!H7qbAEHUQKdHH3ddtt3~}yWV;Iy+ZUaGqvuaDL^_PRa3y+j~O;N{Pg3i?l zG3<{;3#}-fG#WZ@qY3KmEw32?r3BO)6X>!Ki-`Dv$n@A|1y~$09B^z-1ij^@M9}DiHn zz;&F_fu_;J-9T=+`NanR({71_A+9~rHRgKPEZ2&b`Ik6AaE+wicOu(YVw3mgYG31ngp3@V|XUYd5q z+pO?`QGf>BGUYNVxqb9F#o-OeW2!;(uKye#kqP1eldUw(Z*m(6_dzrHEb-AfH8|D) z+~^wj*#q8E)T_F5;VnMhQhySGZB%%>F=jb7pQ&5su3w?u{aE))h|q>} zIm&jVxNsauL1UHqq;P?zc$~BiXeEg`5C+Cyn>n0!0N&f1j@sUHv^Tcy%s~1oz zv~g-!bnF8KJ!-2Y0x=z$pPAQfd$^}U-<>0g} ze;);Sab7rxR!@pEkHA?~DO-e8QNwul%Z4>>3;$WqjyEZzXMK~iqv8VqF_V9kC=9mv z5&+WGD@@zll-SF0ls58Tr1XEP8;KI6k~+~G_LOt^F^%V(^oP~ViS4Cs5xX7r?@vZL zHBJjVv*MiosRa;s{MqMhwI>aCyk=k46mM?<8n5>bAJ(OeNx=J7SjtEB0<>G?*)g>L zw>17`-R75s?gW26t$fM#TP4oj@2u|&Cw$uK){wj<0xzvsbm-SS?sqA6TXx}F=C?H! zJN&TO@-}R=X)?3al!i_R(t@8cbuvLL#_F2-ww`P`n>I-O(lK#gB|rHobRYKh}21{BY%2Xx+Q)~Sm@?Z?GEr*^*(61MPV&K)A+TUFWP z7I;smK31MI5G|1Bp6uNh81Bn%+}rjQF2(gmRKo^(`kp&Dt_r$_mn<%*H8fLsMXptp zQ;_C^nN=3sC-3%>7Zlq%u@vSQqoFpd*gQ%8R=<-@w!SOY0+To`*aX_JEthi}!c%?H z(FV7$J%of>lD?DkCOuLzbytX~3`=+`HGtMWa>Jkn2Tjv(wgzZ}VehafHEXz{u!BYF zRMhFV_}*tmJPNsKyodHbX?3vZHWUz#n1%w*yzfKMftn-o)Mc3~ye_?HKX-7Qx5|06 ze$G!{&nYyiCIa>Hn6WxdVZ{Q@Pxrf*X~gsYsSo}q{){}6Z+OevR$DNbv0TYn%IBqm z8nqdC#F6|}dU=VoQoz#CaHD(;AUws~r!#iVy%86CA_n0||8VIzd8XswCS?i}7%!Y$ z-pMAS%3$q@`|AHtVD+T#{Z`($zH2ec#QLH^CBFN4z@L2 zwGkhK18md$e{e9SiTSb5;3-$-9BS`>Fk9{nV?P61rfiPa97g{I= zhBR7N%(u`)HSYMxyWVDU`rn!HuyA@|!B#YV^4jU($BmzRJY8H*38E+YGXWcAO&}dC z8#XJP_T#>Z-GZLCC?O(c^Pzv+!>VRA?e45*DQv!{Xn6CXoZHzYt?>1z0(rh-;|Ads z`i7&sHRW_tZm^jqu%M90Oe*>Q?^FktX&GaA>+mExsYz2 zr`Np1o>K87mr)vA;YUU#D~E?r_d7Ltzv#mc2J<2*LgHJ@^3xoy!u;yJ-P@J@LwoYq zllQIv_fxf-m*ljkBRT z-g;Y`l{K?BC>wB^Z)$gUL#Hl6Gy!EI$JZMp4HKTewd2Ja0gCvIspYpFs>Lg4@9)JH zGzPn!^I(QWYO^>zB0b&2_Bl^4Ud>}SF2VF}N@~4cKJin0*B^cz!<19sIH;Lsni8TG zz9s;2K-2r{{s#5%XV>J{d*BkP%`FC7>eIE0Ax3a8zLT92(ACDdj7PZpC`0#^~l z-P~${5D_2K@ECyDo1>+lb-0dda2B$kaF^F8Q=5IRih08M_4GVarvU3^OCxUpR~i z`IS4#y@p=V)o-&S3eV(B6h6CU38}8?jKwmI_naOh|BsWOP8LKtIpt6#@exa zVCLR5kr3)Q%L&p6Z!aOOuJMiDW8ZA+1gJLVJQ8O{g9T>GJ_hsI2QReSgXa@@c_>I@ z5U<&l*iQ%Iwln?*!=}Ah>+q)t*RUO-8Fl}K%I-2@=mW_KXJwx)clXL>yC?* zR3@hlmP6YQJQUO5pGOe%ROg65g`14#;}OvZsxDl|Qo6Y(U}oW7l(Tk{slcI+0Km$^ zX=1-QH{JFh9oKs25-FUNdlrqAo%x(Ug&(s)aPfm)n$iz3%VyUxJGsz_Lf=BS!>bp5^JBu-WmqVY3D%|VZM!0qIJW4l2U@SiJ4M^ zh;tnlR8vkJKB=~33*MU&iv+7R%u-Go8@H-Tv&C3zK8cdei7UQ1Ol0Qr5;vkzw-sI8 zG|OAn0RL#u4wzVmu75?OUU_(`V-!v=^ftuWsiv6Q{tow$5oGvL9CdfJa-RM7qoD`Z zqASS`b;MEyCrpp*Nq4hmm-$__OyG`*(mkYWo*p_`)H-3pv&B*4uUij#r8mWzTXj~e z^j!DCM|CCuJv2nuq%W=srU8mI?_=H)M}|(@ig@-chqNmgc*IBYj#?)Q1Bfh=>QF?b zzV}MPK{>^i8}wj!w)kQfIEg<)qV9C$%OPZ1RM=okYJ1T6t2cCqe7li$MO%%dly~bz zOmQ)xvTjS*P|V6@B=vix_cSuQ_tTA-u*WjF--a@tLqs}xL!~O`mUAJ*>7Ii{aL2Zu zvZjuo1mSxSHYDYTW}PGo_pMl#t)h7LH5Gh{fbbsO?iwkDen~eBZP$n=GQUd{FFSr1 zTi*0;4kga1A4|^0>5(?Jkqp2aN^)ZI9oo)nFdsG5k++WZ73_+f_C(?>17ohH z+P&dce#jX{;dok6q`gF%3y3rVQph?|J-qj46kP-96YMVD1f$d1CSCpw8d1-HXDUddmvTmvIXyc!X4$( zMOKFTQ}IKVgi2&}*q3fqg>=?Y+r~zy=fobCY>_W-;PC6-IoOedxgJZD(?G*$r$Jz> z=0}-j$n?QT1@1i`c{=1Fl~lN=TvQ-la(gVhm;Gg*J6;`f1gNQ_YW2}SmKWQqg;2d0TL(jX5m?l{l>I0OvaIvbHANf zfRKumHhY#=mP_R|E7mFKJj2Bq25sEmg|B>Vjd1UXtkAJT-;?6vR5pODef7wcxHTDQ zw9KzyDz@?5_2fE{!i7xr9B^YK-r_eMtN0>wkbCU3-Ykz9=KQ_TsmXHS;8C~*Z#Au~ z(@ERM-0j>e2KrzUF!|r(YJX#>MnLm0@%6^$@qvhTx#Gh|t8ZDNBUc-o&z6gMloZsm z_@+N?l)C8nWJi4h0#8RUy#>Wx(I2vDWzt@O+Lbg~3ekI0$rW4nXPtI4X2DEL!gB0J>}0A` zPPJlMYMeamSP{EdvpXXV>w9vLtc}eEdh6ewa`{d?@pLbfa`MZl$eao0b(ob*T~sG# zgzMNBGdVxv-6BP#!@tT*zLs=;2^XwgVIF-OuK_vDsNQ_l5T%~7v5*5=ACFO8w>YM} zyU?D>v7u@8DrsnJCX_4O^sPzY_hRWhv|yM2UDIU9tY+e@5J3ul`CX)vfgW4nV)x5i z+eWt8R>|11LVQ&hI(a}sF7eD|7S2BcoZsNfHfyIUb#RkK>BoA0QQy(3bTiTe-I3%? zt6$L6aR^zM=z3Z@VMlCrjbAZR8$ea!O~qXU(%Ec8mPW};_L={!d{B4M zCH)Iw`|bJ=&3RyoOD!-50f9+yGt8~`k`|MzSL{X14K@aQgRP10b$|5dBvKpnHUAn2 z^GGoYRh_s|%-C|LjyCGbm$$*F#~;)xsT}BS>|8dI@|g*6YR+I-k)E;gSqFy!q`3j|(`M2C>7fFGkRA>o;yv}i)^ELNvD?05NuLQfy zf{n~;hUCiPwO?lZC#LrhB^zOGePXjVnsQW5T1>Uwb>r7@G`EV8YiAJIDHK1TJ?q({ncGY7#WyXrpl z^Qj0ML$L&s`-%Xw{gvprj7ueL*SG4aE^V#E!D(>SYE&pKhw=V76Y_6et(^v!LtH9Y zMyZp-cF1p5KFetPqdC+7U)#1FJ9NvSiPDw1`t`P!eyJ3h%9Q;#XeH7l%iR`p!~<7; zyqs8&qd}YP(tf&aA4jPA$`*BT@XBghiuQ_S+-5OCZd~@v+9>fiwzN5}mOAF<6LOe7 z)$f&LnT%^4`=G8QN?qgI@yy2Y03wygvhABFyh9qR=W!Uz_~1MGg3iuBS!~qcLJYh2 zml?_?HdIfrixRs=*OxnOE~ElCesU4eW8F9wbc3kH7lqKgnrZ&$`IW3(Danai^i|KE zx-KQ@Z+E`YmjEDbd`3O%o4-JdlEB z-&LGC>c0QX?LjU=J=8KRqN^q&L-gFvOv_(KiFNU;%U`9e3@N@;Nu%%KcY+j2gqJeq zm#Mz8dei%3RFYIn%WWws+x5%`j6-c$`FBB%{j^W>+w|C5OfstWq#pY9-2^7S|V zfVwDRHR`jXb?#HgKSw?{xCxr7zd#hk2J+bZlD>OH-5&G z$9xBcQ=Bz*^eym|3H@Y0c}4Qe#4Ua3&MC`|as4}|Vskx}VMoOT>KphOd`6FXJhk$* zLtKhW3#+V>bCjmlssGXZHT?FH=3BDKX9oeQNZP0P8T&}(lt1unl#=^4E5>o=ix!bm zUoEdF+u99Vsg1T5LL1N0$t;C)MK~(!Gf4@o2BWV!hzUwjS*0UbTfXbFE=bJTG08Vy zeb#msGS%mWbi)XPl6760L)ooSPMS@uZO0|WAEUVLY!A7E@hCYQOhGkWM=#Z*|?7E<#(Rh(X?6`Si^!_~)>-O7=}G6v&J>*6nsQ z4gUyyufFdjXpiY*4?S9}Ow5Jazk0z*={4mOfRUr)wLQ#~++g))s3vzghn>cy!6@|G zFgoLmH@|!@w59=Ri!J7o!?BR7Ti;?{zGd&C%QN(nZd7xO!FTIN7=LrR zENx(Ryvui=2jM$r10jq10_z6@W$*Zeb8TgCw(|lSHvM|iwLCF zABO2NirW21s#Y6%q{YnP)a!M#V)22CmZ>(Ry1-mh?{Gij)IsFAjaz@IPNx2;(U+C@ zpqr8c0<#Zf4STg@0e7t`NUEJJvPRyJ$0f;FObVbGqLJ*_LTjrOfj)+)@v++moNjdE z{=*vRV`0aXlA6kqWw|k7kIK{{2}?;r5bjoMzUCcYHkSRY%-f#do$;qjNLYF30qB+O?Vm<15KeV6|-p_aapa zbM!Fv(=)0MWXUe5_4z#!(8$_o`Z@4vN=?Xd;zug~T2Ar*L)Ux9HPvn1!jA@)&`Ss%{AwkV~lm54_XlX>%G;Pj2ywpizfb)By;==z`3FK_1cTHKN`ucH}*2F8mcv= z5)&)Na}HJaV@`o|1)OhNHDhQeoy+#i>Ma%HwBrtsYTYJ4$I z-=#8LMQ-_IQRwi^+fl$)bJ~nKmVSambwZ>GxK&k!gm(S%Q+|(_;X*ljd%%dWz=307 z4F|Pn``$-OI1}~)D{(ENAm5ux1euw`!KDKo;0;3IKUTEne6pz|Y;G-=BcK{*Dg8y? z?Td6l(?WIP!;Rx913Bjb4#%MLbcfXEd@c^|LWc%1flg0e4&9Wr%6++T8_Dn$I8%O` z)R&h=?5s2ePOSTyzuah;Db(28al!n#Oeo#qI(xZR3LiIi-r44ppxWh5)5MMbr{5Ih z#V%G6#hWdb~p-&vU-w9<{q~)F-$nw~N>Ad&82%EAO$*S(XEz;N|5i`5N z_)$_SG^K-5JM%}D8;+~!Uj9J706hmqLXG=V2z}n3@zkh2Qq|jidoC^(?4pGKB?uxs z&3s_3x_BRVDWzCK#=;uXA>UMPT$dH2qn7x+r*~__vNj)GF`m?7P$BB;a{?d#y56t1 z`KZs!)neTLGhu?uuzGzIc6Q?}tTXe39g)AWG7KZWi%Sy_FdchkrzW<0Lfhas>y}$^ zR=T@eI%2PXK?>Vjnsey_f7AZhXuPaLH236cr}EIirwsSLglRi*K-)*V$%?T-dw7V8 z11^_Y-mQ$}jlyfp&l$TT+4Ez&ClvJTiJn-h9a-=yi@MXu!xe>~6HRApJq0m3?*)jF zuH&|^I~2-w@rWKV=zi5$-3Y(g0aFLy?SuwQuGl}2U%BgtnJJ3j2 zi|6V0Cc%H?i8PfZDSM&Stubwt1zM$?Z?0!?rn3dLZJk8LUL5S_1rHN1db0&*kY696 z(c65;Aw7XBYg&0z!@`w|Q1zVr9*8ISL5rLIg)njp^(k&vKF*|DSBJk~wv&97T&s&# zY50?63r^1XJ;9}~@!W!*?d3cbVYkkWxEbZDDb=Xu&KK2G!#DFt3lUs2EY@+EDrd@C zU$aUh*9ZLiDSdkHqfeUVCXI2nCAjALY*{CtyUYPH1?y)U7Q%RH)J%P%E}eBMFM@W( z5&EPf+9v~17;efHNix9%`HI;Sd-{(lqp-*rijvpsla55Gud$uZ{k@-jg9#|y6ih{r zjm?&H%vIuBmZ$l{+cY>nvgMv17BJ@N@{c!z9f#{-M}^IWhiW=8Z5;y7y${O6T54j| z%^*y%<%TNyxeN znlid$D=_5CI`m||Mn3PVtf}K9yyjsHy2)nNQ6+)47%zkH^+sXgqRR|b%x>V9lr%iu zMt)Df0CiAm%7f0+1Y9TjD7d?{Bh z<`(X+^dDc4qi#C&)}rRDZCPqtPYQDPNaD64`w;P006n2z2d|5D_mnGkyU_CZMgF62 zHu4hMFh)whFm+(I+ z16#V$)Z*=ce5;i%L!65%B)0uAM`QVq;ReHJ+CVSBSa5_phj${o7XP@iu&DWQ8ig8o zn7oXslxN+PM48K!7%Kf=tl{lPKo@#j$xwrtzKum7?V2N!Z18E^t|ci_Io#m=$^9Xd zO42m?)&a_otEDutw&a*)h6EntgTb$Hj9orDx{-tZQ361S4KNZ&coj)nmmL~rTxZw3 zVrQw?Mjfi;yX`LUPd{vOd1dc!WX0#+ZjcXSxJyqs>lVcQ9OHZY3f;8f%RLJ=iq`++ zHV;90%B&ZYx=kZ6o5cXl!*4;PrcEjTjKJ(ng7hsNp?stGQWe?}0ovvtz_ z!HRj?>)NAYJ-lo6iB_ipZ#YA6!}m4@W`0}ve)}a{F2g#uO)hF(;3ZkaRV60X)nPqa zl<|R!@YlB0l8FSfiu*vK-7KG;K~(>2X+^2B|G;tm{J}^br&>cav`jxs@+W$_>QkH= z;zvHTP32rqo09v5-!N5(`rG@hETyfxprZ`GOZ(!draF9WibUr#w)K!_lH*v1U+VEo zS$2;P;^KMTx4N_`nlOW^Zm4~aVJ+6r!h44_jp{s()mM?E-^#Jn;qMyLvMPB= z4m;4-ipcp?H}Qh1Hvm=Eg&*yTiz4xepoZxX;hdCdSml5t3!E5cd{mepqfOWyH`6G5 z*4oySxh8~rzB=U!F3)m%b7G&@)q6Rux4LE_-i>=?lnJ>XZ_4+5!C_J$qJL)A;?CY8Qe8a79US|1bZ_@_vpwTLNN01pO$ z;~kXd{h~!C75zrk7qQTXNbd2QzeDBM&C|4~3hLBeNg5v)Wxe8O0jGYKiP8woHXPo$ z1;rR?M2TaIiXfiJ*Rj-x{#btG``Vb(=ELjIj-c$_LOj2~*ziDe)-!A;1)WnN97{i_rn({K%{0Ea-q4bCq00o+Qw? zN}E6XtYm?xN?)e$6GbdP&?W0$(O-j4Ho^Vky)%}canyA(6LbKc@h%BHH|KRNHAeM@ zPDe%cxGg<8OonV)7N&_a`<_IL+s^ci&(#AMh#8W9_I+kWZhK<%O3BO`cS-%J{ww+B zv_}HH$74JeMrJdHeI)MpQBa=tAX%I<@?ai@>P+uQffCqG!7ogh5Az$l=~dU}HlH;Q zHnzB5=dUD%(j&zL;`CVpZ6|xra~eQyvE(*#K!D7#F_*QOn!)_GyEo(!T+9Ix%rQ~m zwOR~z%~ov=A!`0Z*ptrJ97gV_HGZ}igc0i!UuWe2b+Xu4W8GSC>*u9>wUBzB**oua zZ~DErE4$TWH-tg}wo&FisQ~1%=P5rtraRDm44kt4)D%dqz*TcIxkg&JoO>j4r=khq zOP~2TA+jV==c`lzgdUdjvu%L2iP|Ll82VwvGh}D(v=NOzN6xaIp83KBp`!%?q5z6p1sCYZ3MGC=!IecXia#9r{TZPyfg}YcIoNScf?Eey z=OPrV`PI?Q+|D3$DCw2!Yyo6l!{tDovE_!tsE@>Y%VKE8?}6M!I#z9b+NjV<=Rj?y zz1ynK#J&{CZa@#y`)gxW69M|8(8%67sItLHI7(T(=mof8@Ks=f!-q6`H+V~rXOwH7 z45GF`vfqSW!y(!{MUo($`B@u?Pps-5EndpTx=0*}C^gCl!dmK$0{y^O_r>Zo(g?ym#v?$couVneeZ4UDBm zN^Q)|-8A+mC&>Lr?D{wm zJ>M8gGb-oeaT`?=#()sDte?A92V<+{nKS2i6Z-fhI}cT@TbaSzhzTm92kDBkGdt7Z=QB=tqCc+gTTEzq z3DDwQV9ti5ddd3&5w+;Xiqjn z7}ZWUI=n-+1#L<|8Cma}W}xv8b9A#)b3{+l%YW z2oG6&rL*^rXTB9Jk{zdKXp-<6mC5hgnVvj-b&HDdDqT#B-kM4a78!exVX?r^zJ4hf!z7TKbUE#=9O|Jp=4jn-dg?`X zt}xBVlwRKdy-M|%&?&fB0mV$ew@b9b^Kyg z$guYry*4Ff(EsIBKF+7QhDje`N73Lg{}}!9T}wAhIi7#scnk*PjX2KkU(DUKPHjGr zaxEcBM1%jsH7f%yC1HR83_?6PVc`W zph$Lkv{pTXd7TYdv**1t13ba*D{G%$V*R`iQIEQBT=Xt3irl+9J}p>46^mS*!-{g^ zcJ!8_rQ|f#N?EkT@n8nFbr|^iCV_0z4R_J;%=#ajza?iya_=#j^Y4C~_1K`AdsRwE z)_(kZg-{RA|5Fc>sD9?a=53^YTI!&^O1ol~ezH|m^LHSu5X{j|cqpC#*5t5tb(FZn z*+FFLa!N18D;U)O9!QtekaW93tP?(e_Bn0Ox{C<^I7J_+~5(DQ!w zl$O=x966z_sa~+aa7%*r<0EX}u}3&$pBek0zq_3^^tJG>qU?*#xWGO4d>s;qAd9>& zd~<{SX0L~9x?r^%^rru;mxsK)Q+Cdugx|=NLC1`;YsytrZjSfU>R;+Gg?@PScZ+Lm zaF=Hy;?NszgA(c0HoB-%qe{ahZz6%{VL;Y|flkXn7d-1Nly0z2^A~K@?dHkWB!z+s z9AwG(E6NaoPfyqZDv%PnL!pPa1=N3kB~p>M`D(6vX)Ug4Af0<4?3t!9WUGJ1u6UJ! z*=0@0xm`jxWIN;&HOfgE{cK|4g9x{qmMe8Z-F&l^h+YEfi063##(h|*Q7y28k|{6s ztd7}Qn9^Ao_EY^6fa}!*9E19{#aed=8?cwe$%TJeUTq)%LGbg1$PDzA>K}NpD%MS; z>PDLDi+p@Q(?j{R`RmnSsdN&Vk!Kt7N7J9-!#TkV3Zb2ykm^BE<^1vzXXECYQ6S+t zERI=qoCM|TYgLsM<4B{+sY$V^KDdW`%Az7YlC+_L(k^17JLV*xYU~>f=o9#%=r~kE zv7Fozw;io?sm*c(&OMrQ{MdLZG6L1&44rPGTWQU?Kw8ZQnFcz19bB4wRE7RD5Z?M7 z(u)m+a0;~#xRn*l>_mziOBh18hj>so!}Y3kZGd($UuC*4Px6jjb5JVZEE5NvHLHNn zaY9+IbcXl#Ij&64MwQ#*jdi@=X=c54>Qfb+6N@+JKc<{Q`}8Mk{W4k4Hr<&4@NM=Y zy8|)w_PBA?|FFOMSbW0j`}^jiqIAXK0m3;AshO%hA5iWe*c4VV6giMc2Qb(3n)&8O zOwxIGMa!DqG6$|}I_A9od+eO;x-|s_%sWI%RJimZ@Q*maOO;pp3!(=m#FP0d=} z7N+X7hSx#!AMx+UAb=2G)hbd|9~i2F+i46!+Wpx~bqw+yb-vT>_^Q7yLZ3NJX!QYY zjd7+`mR}H6btNceI`LKs5WEe#@@ISy}XHCCRSrkjF|W-^JHtYdI7hDJQ!uY~rV0`i^5d&U457 z=NXbdXnT_3cKt)37#1jxcqEhkgfZ!7gdx!dl~OwKYiUf`*sConxA_%csX5S^>23+4 z^XX^Ts&yyk4NUfcE0xex@RTyEA~CL3OV3vGFuLk-3|%Rw;gv}0%Ly*s&FB;o4*8Nb zf8MLwysM-`?r?Y__d(}3H~y7LG%BkbvsyD%>j zq1144kj18*2?Q0Mtm_JEt%w_utsfK5(NI3xaMACji}5(}&1DkkBJZu9;g6I`I<1N2 zt)6U%A~2%bl^zaoXVrkaR4z_2&f0k!pinu##v40cqUI%zB`Zm7=K#d(XWUM#@CXJJZ?wvj5g7M5GyQL> z)agbh-rMwYfe6@-7pbbi)^>XRKXEi5aHjud^NnEzz?wlm}!@14GIQ6(7 z+4E}KDJ`AhYYxOy>kh+d6r#B-Kcpa~MK58xpRUnflQ=a?>tGlpu~d3m&c1hnf6S4< z<~e?Rt6e8|-lpAfgRP(|EBSr*A9xM}R269!Z?mrA-OL{jtZ~I|uyL#FyXK;4?eQvT z&fqg;Qm9wP;6lD3zGP@WfED(#Z znP~5Ps6_*K*(xjG=2(fdvs3&-3do{?ECOT_Rh+w;Swnxx!Hme zl648oz!(mFBY?68Y6F1J`Pc9ijn%XC`Wk*B#;{<9`eDrAN9_W9C2|hsG@-dS!uDUP z_AVl3tW>rFn<<%c7KvdW9=FFS|GHk-?hE=060C7>Qc%&7X4G`CWeke|$u@4C^$MVH zj+meinS=e+fvzini|1Qb*>&+UvdEggXkDluv>dK}w@>`Btko;7a8YBFOAnEpvfds)UTzuRZzYA=0(Zw@wm!eRRvjJlcV-2QKbhba0<@UFq0)K3cK)Jo!^r!I z$Xf9NkCn=TuG-n3LyjugTK1u|3N@ucr#()-f#bQWm#QI!h6iyZ)a`!j z1QdC!rTyPiN<|qTxyCnodta)9c*c$FMMiRDO)F2PZG?N?3u%Y=j~HE@^P}KqZhoOz zVvGx1T?nh8UIVgqV}R2*d>bkpvj&$ht#6V1dKX>as_Urc-SNUNI?0rMH9)T45UEJY zK&q?l>L`~8EZ^jQ?|&`cFp!1RNX`Olp}+d6U7Lt@rfJ)_t?>&sX!MIzsc}eDIU!NiqRyB*K2rU7|8hg@ife|awEX6*V}?jy)k^6B z3{Gmea_*V*`9ywFI~eSLuoV5Th5P2@l*r=e&rDFaxb*?{cl?7;+XK-@UA zK-y>Oz>)0mc{OaQo=8)rhh(D>tH^cTrBl)dR5IOB)88O7yD@25W=DqP681I*ke(Lo zvmzbl0wp|Hxrjn`Gm%ev?ZT3X1c9fc@gdbAM%;F^(s-)V@>hY1Zjbh5tFMOWBV<3+ zX91Cj^P*REC@eji81({QcP9r>7=ul2pt{5Da*+AWo(}CzZVd;a56uxu6`6y)YQrP9W9WovVk3|LrwIXG zCjM>cvxR&g`lxc>HdRD=-HE+eCpeM-VfcfpyLziUUn5*jxNx++#C;g z0GjX1KLj~CO2>HvU_lY6lWseK!&kdDR&UEA$QuDecshH0f=BmN&=Bq&(4zS$^RR$2Gi%s6P! zxpGJLXHLC~qbE3$uJQ{wGUL-thUya9%C`(2`f{Ks%{t zv|)Y&;NpA|Cwsqi5$bXGfLS1S>JHSLvo7&@-khJ2DoJXKTM@%{HD{zUz!NLyhR+-n7*SQpTgNcUvG`r=-T@J)f|5pGgX@W0@$W@p&xXt@4usCq=h3-ze zq5nXoF6=p|if4{e7$OeEhg6kfI7PROBglscTt>AVY7B-_-NWbah6X(&z#N z`@E6MAmj&CyjYWMs5ky2(TB3<&iEATJ_zNOl#yqTEXx%9HbcJ-Y-Z^CG5I^cJB7gC=r*c(+A%`pwu zI+~7r7zyWg7K1;6iKft<`)=tZ#;dUjKzEJFFQ{@$zPlH>;Hg&=A8IOD z#X+v}y3L*OM3y*pSm7^__{L`XKR}|oLRu7{985ex%=orZ{oTHas;S22jP5Pq)H*&j zHL*cNwRS?EXEVLI`5#v%6}M|;rrSZ59P@pW4=q&nj!{se2-OKQ$0d0Y2DcXp298x8 z7vVMf8cu|a{EnHxDi{Ac5Rq0_tH;1KQQC%J5Uz3}Dy(AywrbPATlE%QQ?XQ+XY6m~ z*WJgd#7uONqE5|Io<-eq7MV#W0uvnQ<$$Y}FAxn~8gj`~(Y5JJibjTWpvgQ92(o%j zN2tHuDLC?4xY$`e;t)5U9oGAb?&w@~oB7BiJq~-ym zHj7yeSJ{MaLVPjQW5}{5%$rm*5OhT&w?7fhm ztc|~sS`RV0n_dtEH5C0keF5`XqED}9-jOudOn3WVz^3BHEi89-S%UzP#&KJ^7xkv^ zg7BHvMzRzLO$Yz^QvQKsbXikaAR$?EHfVeJ)7O^Q&uXlmDJBe{(%j~UX5Fy1LnFby zfd2Y^k8viS8vnprRw;l14w@T3b!+5k{8Ol|G|K&(nG{N3Y}^Rj(~MIV!}TIRXo&SH z@AN!OC+4MBEpojSTI9*HeKE6S-Hjo_{>UkUH_*)Lzv`$!G8 zmie2cXA0rE>$EFO-x#b&cDt%^);k>u9M&}Z>^|yIOl;3+>nwF5u_DP`4Z*DC60qu` z70DAKrMuTSbP^_@bQ(O=Z#wIKRiVWjiPbp*cJ)dDvW(;oCnnqHrmh>hTpB9hs8s?^ z6t-*p#mIl6PFh&KP?3AK1?Z(N>b40$>&X2)C24?{INo%%0Lkg`+pYUjY@jIn7tzq)0gFQ zGNz@lL|-k%W3pk&vVdcA=EI8a9+P$^Si)>ZYM!GeiRP@S4wcffWvg;>b z4WK_6%+!S)MB6VLVlgxJ<=oh^;B97CGOR^7`Kf&9C7T8*f#l?2Vv|MsT$*8-p$P0T zZNe?&>d&ur=EM{LbX&M3#M(~v{j+JXe`Gu~b)9)xa%j6Yu~Yy71Tfk2P_DHtMnyi@KFr<% z1N;8RI=h%GFF-NMtU~6XDOS)#BCZ-`q87jWC6$uY%&`pwt0(KF zA+I?xUBxFOMtbTPBZJbB5wq`QO4#0_16>8b=EOSYK39elq(5K?!~~h zCcSsXr+wB;I^5-qG#KXH0c#BS5ZXZv6=_Vllr;xP?f{Xt-HY`8a}ZR886o3qPm;xB zxc3y?5YAG2#<{d1nV@fNm#mS=#D*t`{JO8E(D>cVJJIEdu+coL?R-^1y4SiXA-Qf? zKYb>J6*QP)r3N*)MHc#Hh8+2BfZjNRskBV(K^YIq-kngwD&gsSq07S=+apEwbL&E&K5*3zjrQPsJdg+k? z%=B7pAy1>bJjUjfPfe<&4LkBdeE9|l_rVr8yfh(N{?yLO;8 zGehZK-JVy0Es*i&a80Nyh!y3X6Ie%R4<|=zz98k6x__98`LBw-N*S;jm>&2Vt)>Ei z`*y_VjW5JbTp~F3Yf9GD1W*HDGaK z`{_;cy-CLnEvatdM-&J=}%jFt7~4mf5SfP_K%ZyBVvbpPeNs zL0~3@*;UJmTZ<-l5=2YX_-a>9DCTZ{9(L^CGhRw8R+*-c?j#Ur=;#T;k&13G?kB%Mb0;LR{n6Y;JN zaD9PZ^;9c!ZYR7adjuT`iOD|Y*>``tVW(vLwmF-W1(7&7j4mHvjbvL?s0`6UGK@8! z{*kZ5&EFL$Vi}Sh5GTTBKHp&HTj+h`a7CqimW#;MBqh-n-6t6E?8h4rjpEKvuTjML zq?*LvjLTZgL5&nv5^K}kKa?=e{Z8AKs{a&R6`~HJgk0v@IT}t#$E{{L2F6S~)`MH& z1vV=T8{Vm4#HO`@VSmO|sxUF^^UK^Dxynv)nt5R?y|qLq8T#wi`;uzK{_35K3*w>O zQEZisxOp5}6iJ*a&Z?X(m1-U#bl^ZHdaev&<5gt@pd>=eGkWs&tMSGvRGo3_xfH}? zR4%{q;-M6%hHPGEq)R_gdvwtcLnUVYeYg%k}$wn)h705&)#yabPzq@eQeI- zvt(j8>L1gRH+EeC*bQ+em+}y&0tOiYYvn2w1&DQ^+vm)yTeIa0ccBR-bJAs&CtF&s z>i0T>K_sTPKDWLOO$TBG9P~>50gITT+w8AdV|Ge8F7AX`D-v{5bx6~;JbJbm@RJ|7 z3rO?8_$M7i%FLbJVn? zL-TR#m6^jlUIIDtahwdZu5Et8HjBjkt57zs5{o%T4xfGVk;8miz(!|auf{cFo2X|x z^qip3rYfH)$>}Q1qT`wN1kfOaWaJc+RIAi~&nYW*kCtiTN@wftOXr?wFeZ^IYXI`# z3rFhaMC$R9mV;Ejw^p?8xnAMfGJljedv{Wu!e~MKGHXA;IGX|_9%>k}aiYgicP>UA z#vd-3)D9aid9<|CiGH9D&oT7IXOuA)%Gj|^GodGu!WUd`6e zj8KY|o}0c{!BNNOVz~dzhvJI@8I%&%~}qpd%5vl1JCW+101Pqgxco{e>q&+U!t#h z3iEO=ohfC6#wXU!rcqW$0NypGF+*KVA?H`%EYfyV`4QgyecYYCdmWC41+^yI%l zg*u6AN0S&bO*z>Ld?k!u`RtD02W{A=OoMuUMaqp>{Ve_o(EN>Cfo85Zgii_P{_O=| z=rBJ8cyt(W|7y80$){((#kTcZktvYK7Fm1;VOLWVr4~?}`sVm|t1hVHH9y-@k4;&# zb%_ks60zQP2_hP~f!3Sz*XQYW__dG`n(~Lug0sXu0-M)Jzj^ci7m%9{BM9-)_9VIk02g#h8Nt={Yi_Y||6GYk9^ zmKwigm+b0{Q?~}2Y;)ybq;ZWrpw!4j{79=)?oHfgxYPO>sv{ruuHJ`^N(ceL__;S; ziT+z@BDO}3K1Y?Fz+Pt<>-X*Wm$(W`_k5mx`Jlq??%K(_SNt!X?k?vJ{%$0V#2rzX zS@ab;u+wO(t0_ES#|C(4ebhbL3c;mvIz~eTJ;H^LfBLxnj;B%Np&YA>Oj$g3j4fP@ z)k<3%iNSF95>hLL`hzLMx^H~m-03f6X{S%w^ZOCzk|3=3eK&)i zk^p^9y9$qNg~y6NjXsfy{;~5-W77I18@7tC{QA@82inl5s!>mO28~GsFxbwSBbm z+Lu@v1m{L>*Q4;=daa5N=*TVS@f}M=jOdgWe?K*(dWN1lkf52YkfXI63RK1)|5f7T z$u0odhT0oh9F4Ic+o%-FJb-Tk@r#*V4e`AN`riDE3w7ce&O71I-0GK!$3qpV?l8O$ zkV(5;h|22o)F|>Dz|S`4SDtBWcYAKGxDE*<0E`lZH-hJ!@!F<%zX%Uz^`=!xc5h0S7 zg;{SUQRkq0PMrr(c6yp{)PrN!jt`)M?G7r_k}+p%WIQm%W8TVCvmhPio$PrZN*aR`5c z{4`+J*UGX)73VksE5D9Zb2>#zwT>;5ljkcD!Wgb=p-tskZXLjeLSw!u({r`%!C+{<+1mh7}}0i^5!na{$)ILXRG@<&J<#R;@%9S{qpHV%5_uN==MCN_Bjf!Or|TMBj=P&k^)s{;r~;25>_ddh(ni^m(%gP-bF8T zu{0XPjG`|D2p4G3#A+VtwD2Sz5;CFN!GiO)a|0Q^!x zNKBhUyh;{{Nb#+!A}0Y+l7JDa4e&h2?*;+GbTMX|1-HWW4c%oc+0F%4L?zU>lWTWf zb;w5A<8cf#{MC@QdW-#NvSPjL0y}IrX;*`q6wpwUlP8C0&2mddLeDAym!?$z%mdvR zU4utLgP$6P?yPed`IyJauGcjjhtfHkcXd!H8Om`fV@_UFE7;dAov(9L>-SfYj+NG6 zf?apzoRHl20mlk-y0>S1+S{ub!3ul(k#Vr(S!IbEP==Yp65TM3>Zz~4ET=eJ@LH6K z2yNQL|2bWWT>tUa&tMSTR>vn;D{pD@%ZeZ+h+4=kz<$PitfgExN1&!SpP-aK?orPFi!j=9qeHGW_pLvv$(dQ7O4;?c(6DEu)E#AnUPbV45(zH&V2VXun)Sgs^0MskBAx{@4J@Vr1^eue^S6NjuX7Si9CRNzP}X%T1}9 z{E#&rz@7nm2!c2@mXCP{iym3TKj-K)$m_3z}``}c<7K|z)Zk2b;(IvC;eo40Rw6EC(<@ycf zH&GLM)U>u_8_m0W82ji8d;7VfSm6` zM?8gjo4y`U)G`Zf8IN42och-IDBuuGuNPBuy3W%>VCHZzR+06h>iDm0;Mk*0pn#4F zwP911jb)MiRgm`|Xe-`*%CbFavzM_Huft zza*yEQ)Zb#EcCeoYlO+&1Z^YdlBk}C96*6@ygFU{-Rg^1r?sWv$T_s7`@9QE7LmVd zNJFVZ25wM%{l$QZ3c$aL!$*s1qM!TrpV`syO;p{&VIiei%l(YTB?e6$*yo8;1(alw z5!U5CD-TV3we;+gSifnm;wC9)>YF{Tz7jeG4&7(g3aeko&%TdUiDsK-Y7nr7l^o~@ zM>Ed3mss17X*T3&lKOZfMR~g*OKl03u~N$|);E^in)RkSLq5|z#A9rWkxI*T2UJsH zkG2~!w%mr9pXWx^+Eh8CmP4g1W(j^nqq1h!v?G{x7EZS2daQyP4+L0-;AdNw87W+* zWubR^{%BD;t>j4t=)YHikDLIp5o#<6{2hVG(+VYVBM>~x_$DZS4q+&KXZ%Mnaed<7^JZEvrqf*R-Ytfv5 zJV=_`sZ;PR^h4CSODE)~+Z!;fG%89I1yoVP}0R1q-eUl%BIB z_{Ws0Q%u4zKm2zKVq->KrRTb(FeM9W?y>S39I^&)##V9K+5kE288I#b=go4Y3*>)w z_TXKGVIeMdJjUEcAOlR0sH=yvw3Rh2v&WQ9Ovt@LeGH;9T#{2{T#M${7T)AgD!PtA z=oG|t3_2P3j;RC~D4?aBdy8>=&TYAUuU9$@wuoDuq0BDc{gsEkpT9_Ock>}zu%e|f z!B(()2ugT68L9Ql?B@>}I~q=HN}uIic%)Mw@IV-3CUfomt~=i#|0;jNpehSObek)Os9` z{A>6#i0b_Q5n!n+B(GgKK&P4JHl!ndprz;5IUy{Y@#K zj#g@pS-dh&$B#Z6cmpM4mFmEFj(<$LU`dlbxHtDDv!pDuLSR+}QXeWwUvf0j?gied z%p2MWKj^^Mz*pq%nI3KN7%z;1Febh?yCEm1?^FXb z5XkKG@3u(ceGMqVL#j5j%7?x{n%ZXQh$fdVR+^+qQhAmNc)}f@T1rWuR;!&#sO&YGq&(rqET@?Eda83SG6B(@1vt0B4lGB{5^94_ zIm2S*kL1TFA}Jy=dR1yUg7b&;O>M3sfQHcP=k(1sy_zs zVyCmip@EHby-%en?lF~<#O>J*@c7}5c6u9IBwgvJ`KzBA&S#pYiW*&r+$sjnd_OVE zU+mi6m$eGJkrfuz8D74Q?c_<@DXou0I#WFl{;{NIF+zsOCq&Np6)I79PCL}&F2>l_ z1lZfD##S7-T{d58qk&F%7cg@Vlf7Sf%u7lUsX93_Mum!qxM`Suk-{$`6ya$cewEo> z$JUl+%~+b<9W>;ABpiN13!3-tFte7APvUu!&jSd2wE87`Q*1?6+3`EOW2OhaqdAPS zC!jGNgLbPCj2Bb=`X?5mirF3t<_T-r61mX_2c@JssX9DJibFK2pK#P!j z_cf2j?6K5%5}hNuefB<1*}P9rH#U@8*Ry*~bl50+0C3Kd)kXQXAqb#AeaVp#VER*E z%QN>`9-fizNBoX~Z7lz%!Jj)wOD{{?s_$b+@2$NJ#Fuy2=&=NqB`L=qlb@>6FVOlq zr+Y}zgqfznV~MGwl(u#@aqku&$xzP=IFj$;=0_VDpMW9}RmZ zKOSaAR&Bsv(7~<~-yos=jAy&J)@4=iJ#IZ?Au(b1kR<~(QD#q8!B?Voks})gw4^5M z!v72x0~EHr%e<6z2<`gFbN;Qh0c$LRL08LYmoU#b`@BClP)xfQ7W-xN;L`5|eN&-y z8+~g}tWK5kDH9sZW1Z<e{|)6*`0p$|ElzLg>en5{?iO-6>+Fmr=-jouel`4@eH z^IFY!;#P3q11a^ko%JdbZTFKv(ObRS=89&8*rJx54g@g4Q5`K4KJ3bH#k42uTljkmX8ykJt34rvtEo zi431hN$(^IRp;?ptaB(wg;nT9G%J)!wCkebppDE(Dnnn!B)E;t_Q21V86m$=3B#le z`_ygZF{D|hXRI|5+9qYx5M<~v&*0F0GIKH=_Rag40`k1bF2bur<6h3z2U8PW@uzOV zCVaVn^uPN`ot|c2wJXy~n|&$c*)^1BFPO*p?HM=V&|XuU=GMtS^ZwlRk1tfdY|6d# zW%OHUAKzMdHPX!CeZIB9<#d#=#g zrioAcunm-Zl>5b)MaKxT>q$lzQFOXzZ6vcm*{z?4>Ia> zgSjX>=*jg{*Ix+1apMN;WNZN}fBpF3y?X+@|EzFcKNG9*1MC~6sz%ID3ycl-ZNQ1Y zNv2srv1edCnYCXK6}vfM-^PsIYYFA(xV%gC!dfl~q3>{+RIwk3QdaH%pIp~fZ1u9qd3F_&WD#wodOYhX_UWhS zddx~m&3rJxNhm?n)T3W>Mu?p6F(_JVyG1?xdlv@%Z`z)1_(zIf((yejmkcW_DIpSC z&u=xsF6`*n*nj=k%>sV7JTN3x0fk;quV+XK(PM{?&IWlcz<08W0WLxI-1ez{?)}I4 za$mSi+kdYinYT-8my@4yIhlR6Rmhf2b)CM?_2B|S!M1Yy?}%7E{8IhL`B%`@nZWqW zhsrQ^yl}wO1}m`h{TBG^5cl7Cbp7GeNuNKf3!@zRu4m4d<(n(*;i1m`jZDJG(P8;^ z1`1o1h7YWi7mKZt%hfaJ-t-1zJ8*agN2LbqXbbK2ReliP6on;-4~YPy$0@s5Z{Hj$ z>@>3C4Y-{k#WH?t8~IQHCkH6AB}=)0U7~c%<1VLoTpRvpRpj1BtYr{1UHQt)I;^l) z?7-za@OwDO8ojWEJch?K|DMbFYQWrK+12TL_3!`udsbbqhW|T+|N2Aa;mw;$V}qg7 z|Np1{m;d~b!@HeYh%n_1{r_I)e|hoex4ghz(v;x%cZ~>-}OHGAJ-Kb-1qvf z?<$}5S!>Z(Wztg>@n;`CnymrcT1Ajn(Lp|ZLctXVd)qyKJ!fEJTuYy~(KC zdcw9v-J6B&+e!jwaqb(L+A}OZFj zx6FEeziDYz(WFH&Lg6x+Yi40c>@`Ka5_%NTmH=v`<-MurY(pzswZ24xlPo5ak8A}V zVc4deMTibQC!lZ38k%JRBqqJ8HGdExpG4@l!Lrja3~>HYUVr$`i&-G7wArBPwQ}T4O-28H@q3HbHO47uP9yGkWUA%N@TQUV}EqLfBcSb*2xp%%F3Dd+QqoWgxxki=$Kb2Pd4XQH|Yv(Pu(pfJLzFj z3v0ajfkJIk0z-zjzB`k?Bv`nxE?75%*|v#xdVd!ubG(t7>kLWHzcMjy^c^xAI;`S) z)#?BNtgpq{Gq$nJDB7|Ni9-kt9HWfl)wJRk@yS!GQ}9xKW(kbH2z2Y^lWBI$x-;9RAcSvD~0LR#Fa^JgFMU21AQfY}HMM zhb1@-^A9oGW;a>{93T@({s?Ge>(QL|6W!-*72%Z0BxFcYd)12|vv|i^!NSQJ8-)1C zd!KHAeRgqZu|SVGUQx_t_dBK$7!F`Y9DEU(HwloOk|wI_Pgid4uB(B;;Dz@l2a)n3 zHSY*@+ABU5-w@8hLCD2C1uzwJIG>?)r~l->J1qD#yJQc0GmPwS%q z%Ods~m8&yDf_4*i*4|^Mhw8p_sl;aUv!AhT(%F+cfCal13cPan3G85uvQ-UtF9~Qb zbzUfIh7n?IvFu{?fEjsOYFM^9S-@ekXebL+;+Nt=J1=W#XWSYqURN1X z6{InMd>L*4)>>Um0Ks{rp2rR<;`I@Y!`m{grmw*4!fBW^1 zIf2V@@uA)TldUDMbb8`Jj=K9A@H5hGc`kWAVBQIL;h+2osw=oN03a((@{qyfs3@Oq z`8aF#{%DJvbIaFzwk2+PU!!K-@V~v8DwIWTWF~dufGny zlC*bG)U zTJ0|CvN+diGO-3UShqPZcS>&!n_!PMGlT5DhL+Y5M;W-OPq5x9rfOX>FaJM!g3)TM z{Q->lp~Fl901B}RNqcz^4LLfP58w)3x=vaqou?x8ZI!Z8fUTBZ%!!a56sB3<}e z@LT`=)Boug`-)z_bye(+iVd_Re|6HzWT9;JDX#m1N|l>aFtq$^Lbk?v$?2=ntC=*I(Jw&cu1TB}NBKIJ zh{k*-8O^B<6g$=`+a!P0i1?@H{8wH6Kc&&3czEY6I2H+=`RFEx2(!qs5dc!NWXfxb zu%!I#>$6JTUGQZs>8}B(LvEVC{ruUAC4ALV8_l|V;e^5KZbW?F3|`jNpC+D@C98le z?AHBucuv`39#`bh^Z}%peUx)*jq|HQNYkLO5>v>;Z(|zc`Lo~DOib!B z1?6GcD_-i-KX=G?uAiJzu)_~jak@FehJ;?96xZyE(@f*P9}pg5H)-i0psVaSKb-aO zWWfI^p#S*FJuDZ*szv2zj`hxWrGr~~Dip|(HrA>&S+$@AQpz>s+E*tt5IF~bG(;Am zztE&b65Otsty@I+b*#tw^P0g5?(MjsAo1J9A(wZ^O3TE9;1joQiaF>6T>5HVv;pnD z07MDE#Dtl{Du6C$uY?^)QvN5?e89S9EVh=#=8~r_ONrCp(Qy1IUibwJeh>S-^vT}> zLu{%wo{F6?u$cz}7d<>5{J$%v|50ihVZdiuF_!pXpYEPeR=%raJvYK5e|}ScuXCOh z7g@1d^bf4(~Z ze8)`5Lx;EyAwF0399A(k+B$OpP3#5Gg_Gh@)9K8%Jak0?e;V!n>mSZmxl{89cp?Fm z5BTW^=~q90_a|^5Km?x1HH#*`sdnM&&t>eJ4de&sYNOI4d?JA- zGOz>$aLfI~0WLHCA4j78f8|J`7;Seaf(0gl?am*UoPZ6=0)!FA97KM(`^&aBum-pYK5BI zR)ohUn!dhh{w6&}*WL{}uKTky@8JB$TYu70ubI!z$qxpGP(a*< z1F!4*Zq18NeIZV)7nJx=55D^U-194+p{kE%t<`O^91r){MyuFp129fYtym4{G*RB$ ze|xRSfj6)C-la9tk)UX$<1vQN0kng>ZR`Uz{1r9!6-VfMv_UtU?QXjQM`MA}*Ps&o zYGYDRbAA~QsI-~?h>71OmQNSy40Wp%rhk^sM;vzWzU11IoU6+Mu~h-~0qRC+$^Cb! z`_CPyuPBhcQ!lu7$-BcUp(5Me?K0t;eWSnezI%C*JGt|{T#dy$^(?;!%7rk+kLp+} zw<9W{y*pc=r7)A@2_Kjwb~C@4%qBR!$aSetIB%O>#;{T!pr1 z9&W8!fX$P%W%0xofO_q4hR~Kafuard&iv6JL8pcOGU5oP9_SG~2kbb9UF|n_;4LyF>w5kykXEh%oGaLR zZ~deB=O|BW8BN?Jbr zGT4P@NX7Rfuod%Gn8fZIAWCjzobC4gPC)fW$1(8{2d!Xrg^dY})WzqFvlVaxvw5PO zVs1s-pjzM|g!(3h4OY!`{Ia8t$Ii4i_(Iwd-kTWxc=*iN2VNEzUj1mDY?qa+0-)7p zPY8hj3|zj6{>;97$fo**ZVkhC>WnX7hY-S+U9Pi+@tx_N{xk7KFpu|aPc61rlJMfh zeiJM}r18UVt+V+bLg1Wm+-#4yX&k%M!QCQe@pcV2pWBa=n@2yV;*W^{xK`OG>K^19zBlh{{C+q(qV=ScvAdh>%}^bIwhr~h!i zV!66K%$cBdUfaOSLiC-+I6bZ31Fl3{>%Ga&#>wwoNo_$ud1!Cogu2D66ybYMbh{ot zVKK0E%B|{S9oV1prst56+8S6Mb6KsADFSE7WGOq$$k!;Q-c47_UbE4Y(r#}Ss3@Cz zq2j+OIXyL*s#7x$lxV)clulI$Z%d3C6d$czmnUAv6{)3qZ5zEv1FU0bx!oE!!n{vO z0QG#>t$8z2+a0|H!%QxTJ9PvONQcSbSNdHM>=|wt;G7`Nqil^yx5|*-Xvg_?RaU6; zmH_rV5}M@28qsSy;yM$?1}l>^OMv&G(KglfP6JL}8yS0n^~{jpFc-x}9U;GI{}+$$ zKSU74VKwzJ`pn#7xA>aZ+c9Uzq26QB#5*4Q$ig-wz-nYQ z;cff6h{lF0BAJi3bl6@td@>v)ss5H5lQxf6x#%G;MQ%4Qx^JP8m4n}XD3k>Q3 zNcr{N88&F8=_xkRn-!ci&09G0d~4k=j#vCM_qS?BO`bp<5wWq*@~!YT`N-^7KH#if z+0b85Bgf*NdxE+j3(4E-UZ4HQVV6}>BHAWJs6@}7FCB#va%J&vDWwr35!b?5-!DQO zzesFmWX4tIdj7T3Iddba23fHf)jIS3n4AUl8#hfVKQKXMZWNYYwiVmltUXVWe)M}p z32~9$~6iSz=n*UqxHZ0P3k$FXQW z|1=ee`Lew%H$qPSjuNnME6jXq;h(brrdOK{x7zW&i1thHSvtQ5^Uql8Erx7LazTb= zb~Sec8m@&)MYE%K8mvqPOk=pE*eM<}1KW#3k+3+?U!^If&a3ne)5MAv?;}1 z^b3=fuo}FGNP1WGb;pSrNgN<1TMOr-98#g|Vsn|?BT9!5_xOhXI+o5n~1jw_|My!vm zetufpAk-J6LaUd4HY7}H;_au>N1Ljt(oF6uMnGp8%}h83=d94_p5~Wdz3XNzYl~gj zOE#gLIpS~^Sv#HV#}FQHd0)Y4E;JtXST&qVkG2kpjaK+eC%BQ4Y7?Ks|$iT(>BuF%*e5 z%-FkJNQ{BKS1xJ>@860Xx-9MXxU;il4Jf6ro$LFFKri_UR}CXE133o%vFGe#rH0=F zi#U5@Vmz}YsZQEW0H(B9jPL=+zFSJcj1Ju}Wd&BkIFAOGJD74L=2{!oSkyDcgcI`P z7$5RDeFY|gUw0aGB`)-QIljo^zKl{RG=?{JX+YhVL@YGYirOVb;^YOq*8(_R&%Bhe zGOqwgltCC-TUA`K{myMLa1#Pc2JMTiV~buRFUokhY*40I2j=-o`-P7s+24b7K7WgU zS1TMWP^~lUN>YZ;vJO66R9sX76Q%d(`~({>6=9k9en^ok$wP8$WVR-E;j;SzYXH0b z6CAu=^s@Wj_kL%LUlBNu-Im9cg6Pnqg{7OOM`^vWLZYSOCT@W`LNG_;V|~+a{Iwr$ zy3sNG7M%Vn?p(x=EKNpZ5L{ycMA4z^ZPKY~SfC~%1`whRJAoH;nfLeZbyeUyZOgW$}YGIIIcGsKZ}ig)W%^CS*!h$p0kG|S5$stE+6`x=aWR%<@pB3J_s^LSUN zR!RSc6I^U(v;mRid6|%hc~pY^p>E!+Uo2CVOM>z-`aH1Vuyb}#>f`};>Ap2y$O^A>#FKki)Y?~-!?B$vhQ8?R0i-v**glLXPM(ocU%bth>R!gZ;H<}H6|L&(4+Dlk)5O@P*nu81IAy}1Qw7&i71jF`d zr!e1e5ZNm#l)snyqD0C%#p^0tJ$LAgW?5Qi!ERo$ac=2IzDa1Wc5rMg(jP&vp%d?& zEcn4o|Bv^}H|YEa-EUc4waO=-{BQ>x4*@R9MV9laI_^HDT`bx_A) zk^XZ;m@a5cLx=bpKiIMQdY2Lhs}AKwR!XXKcu>ehu>MGe^RhUeA)tXd4klS&?g$jp zZv^}O$JhLEPcm=#zP_iD^)5!=?P(R-zFHLc>>Z4af+44(Ip2FJ-jNhN)-! zbOUqbx%tVHX>Yr@x!75h0FpNOxX5QL5XXHmUrG{D|8?Rk#e-3xaz!23HqATm-VI2r zD84bh+J++hq)GoxBn(J~^2aDnzzyE%{dM1NQ6#)Ev4`u(6wXgfJ76g?6YFe%%`C@qTiW*!4}t8Ru!JZ##Zn@cB1} zfdql7sXZO5)?Pv)Rc=61dWjkZIgJ`p-15q{YaQ^3ogMP_4=-EnG^4Z|6nXQWXcrf{ zrJpV6dic;nV(ra+<5Gxob_KG(!K%EY-Dzm~U4rGJ4_JaEySZbC#0ZLApXiOqm{;oV zp>8q;Ki0l6rm>J`7xLE^IEuSN=UbL~t@0z4-_`o=Kh)ZdLS``TZY^zvR!=j>HH?8( zGAYT-{iJnGY{S5;{#%@DzO-|DMs~AoM&T8;>c`{gl(X^bKAg%&h>NA8rr-wYJ)-+H z-YaA%rn3^e0JaroON7RfHm~*ma2?fTPFJlj6PR{F*h~ zIQXP;NmItVIE##nlNoHtz@6Eo{Qa-ATl+gLMXMe6DQ|kU$eOv8_LRASB6Hk;wt*{O zq6a={73jLC|7aeszN`;P8Eyb->X(NY;)O*9YS>Fc?>6qIEI9&K8GVn2jh-f6J>m-@ z@gjEA)4L`MjN2%wqi3NSMC!$QAZxwA>YFv$2P0Htpe^Nmvyn!07&->KRi zHGS=2O_q!U_YoE6*^E%q*Um;$h7VRqiOB{+ewd8sOeUpfc!WXrRA^o)POy4=qS;zF ztzxN6g4w_&1#NGh;OV6g--Hi%#p3yuUA3(xX8OwSM{61;ky!=*YiJ;rYoW3WhVG|2sYUU%Ze}|o=d{;Iv?d=0iHHCg}!g+MVBVH=_ zH}~K-V+gPyuqueRA`LbwW7iHTeXu-IMF)$p;c#3TU+wrx$pjhv9;}KEEiu9>ceMIS z7|$r(E&Lu6aZRD2hTZ!-K#uWp!hgZwm;i1AtjajaV{Nqm+1J)06DoUhHqjGMPp6;Y z;K{dX)W31_kx48ZjA%o;MYAoGZv}K~^kW@D^X|um(}IX)~O<*zxGskkC zQ{@eV%=Rs0)z)j5g;lO1C4jYeo?+@RKfv;wf~j@+fYBC4E15vAZ6OKf6exn#;oIF> zj6)GVnp2`eZwLcs=qWpA(rp35LICUT%9I#?FXV?zT6@l#bTS?u7wA(sI9!YO{el+K)!-^Jc27FF=ep(|1YFt; zxH+XU)rGzfWXSVnV|zvRcT;76z_p?0<%3PyStnMuFWiP(O|piO%7M`tYF5?&XYxLS z#H~0p=$gH97M3{b&EdY}d8d-R1&TB@D3zU!_ndr<(f|B@a&LfUyGnp8fKj<)DWU@OV_FgB(~GWn_tPyt?^#oQt=ww zbc}E|I5x~WPei+YsXJXMW-kj32U9~T2R7xC;^}Mc6q0}0L{QFTQu`jSmHD`I*BeFm zg=#CU{Mt$XA>};mo3qy@+O;8UVk3@LK#p!voa2}zy!(W;b*{KuCFrU7$Lnj|<(i5B z9j;J=>+9*?Q3XUH1q%<{_ewSdaRm0MQe(x0bU_tx}M#-ZxGx_hzQ!FMH;f zx+ttsDx82bPx8KO=n251w+StU+s46W-6#=_>PwB?Hg0u<%n2JI&7NAGJp4N#1kvl6JkuCMhFV{>AS&!b7FgbfWHTU@yrDgH|r2~F&Q^1 zi=(s3##{wx>=+NVOgoJ#AFq9Vip#cIhA?&xVB|)quvvGD<}Eic{uhAFyT%%JS&092 z_=SXCt#4VOj{2b(K5ehZP0uSt;K zvwi1Q){cp0bi@yrgB|hxYW)`H0x<_5kZB6My{C)re*8?@t!L`k_eX*Vye`lKZ|j|r-2GwrjfzQQ zc=+I8prBw(Q*-!2&+E8!{Lg$TRpy~5(oTQ_hOG?&Jmn2C0i>X;9Fu)Ybo%505Ym61 znnpZ8uc?Oz(*D=5fWNFPzhwbh29$ITLjMN(`R%BTW9)UF2S3x0AD8)#!JQgz0VCyu zeYhSdaBJ$vWS%*Ibpd1rl++QGuS5>^VJ3jmqhSa6bnhUiaP-8~K_KbB1^ySp{9E9E zG@*C?*6=@imj7KQ#l{u_GBO6ZWPpF;-~d{mOeS8P6fC4Kxql!EbS!_zRqf*YjbT75 zT)^C5xkFEnW1tFqcPCyS@IKEl8PB@EMVJ_G^xN4KY-K+_fZcs!=?jHa0MLkt=0!(< z__sOKRZC<`U@M$QoJ}RjTjB@P9UPa2xrz2pAz_;Fbyh}Nr3YwEDdzS)S$@dg(xJ#R zl*5qCq%pJ|pV(lF?=8}EY!02%P^OZ=@vRZ54GnM9Og^(Bo`IG5UAcOCUQ0gFJb*9f zyYjYH#l&97ciK-yx&r$kdPToX%AcUQ*F4Wb;1md6kO zEH4Li(-+Tl0p#Qn!YU+c^=z4k^*g|~q^~SOmA}U^-1lzENIws?Ji?w4^cRNjAOGd6 zb?#e0e3v6!sJ(|*68>1v#vS01A0dnFAvy&m#X@vqqmk~V_=_^R^c^|B{@jTrx1J1P zl@^UdBK5HRofX=pqS}=2yf?>$%OCnLB0A$K`HgZLv(182$v~j1g&klC_CDmBVcG zk8I?YG~|EqXv&brFm{u^k8WD(y|ECMLkP3TDB{cI*Qde*%N5%C8zLR2HP?j?q8&!% zAHPaYUHFkzH2||{R+dPjEVhb(F8e#bKaivzUjfj`+Z8_GROxP`m9Z}tzkK@CTo*{6 z82icNzUg}7I>FJ%3zQcwY1Y9=v{X6{+-h~$LGrC+dw^ZqwXF#@JQ|+??Rfz}#PY83 z2Vm=;d2#0l9ObK>S{y(VuSl4O9zeUde*K+6=?wXsBK=37_V3^M|2;+ee~y(Pfo^DO zZ)JLnLEGRkP!rXGmM4i-kbw*;gj!mvKoUp*}-eH&C&OWUka z3KiK5N^~EQv%}(u4cQpo>nbu9PaN?*StC0$3m(8qz11(rIm}{i7S>G&uD1w@%&xtQ zsiMFb8vI!dSRLAhoa0^gfR>11S>7F?jOo|uwmcr@LA@zTG5nnXd#GI2^Amtr7yi1q zrjtUBci0N0Vg7&_IBn>)qe&g4H2|N-#flLT8o)M@s4KT9Z(iotJhN1g#B z-wRnS(27OIilxscRqL%dSD<1zF$BXj^ONcYc4uX_K%zwE>jH*&(df#T-F9DV!v_Y; zllYjQJ|8#3EgH6yL7Y8(^2(*q!U@`DWjh;g*%(#3_3x*Jw-Qyfihrx{=WwWp_GdXD zKG^g`O(8Yf2sD18%8&mcovjv|_7-w*%7xN(MMR zmzD-mo?iKqsC?MGd2_WV>AB%(#;keqG0>Nje`Wyh{a!dd7In@x7N7v{xg~@GUALGp zMjl|DUx0DLMRV)uFUn%0tXv5mx=yVCn}P4N^i+)*WK?GyV7JRs3Anl6?t5dzX1Gu6 zRIV?%7Xm~-EW~S^#=H?Ip#b{|&yUW_}qetMwhu+YPEgi*O!wmlYi~uHe+XM&l|vYCYKxEWYRAoa#*rD!r4!xXo73 z$vlt{V_^BcjLMEDzo)VfF1xIZFQ5T(rnr_KJj!e|HHEHZzzN|p*;@M9&`f$O{5YPe zVo5-(!2*$#z)l!0caYkjJ0fVs-y@jU9HBrT=iq)Ul1F)aM)`@qJ6W#ggHDnxyg>T)4Q-y0n7HuzK- z-8vD>#|jib9$8~>qO}Ri=d4Kn07%0Iut8w@&N%wjB#OT(7fh3OcS+oq)Nj>3Dok=O z{=TefLsFGm)!2iM`h7M_uwPp(1W3LpxAQrxR(I1Wl^y32_JgCaAspU&cG;yWlpSNB zWD>quG9>BMi5b;s)&~1#czG&a-Ke)W94#5KU+;IBq<8RbHR!EN{xis5SM!I5xGC@~ z{b;}kLwX5C*PzUiG3PbEiKqrbmyEw7Yof&yGN>-XV~ z>&KTcLNiI;)ts{Cb#KlV;_-y%96}bw9~$>ppFL8LvAik1HWjJbI6`_%E1Pka;j-rx z>}lrb?@>v)&m@khP-~Zao6$o>tP7vJYjiBLu5;$Jrs057p1ZF*|7^U`I_roEaGWwC zpBhHGvmhY8*|=T0v08V)v>8X-tME@5sa-G0xj#lN;!uv9&SIl+u*_3T@i`v`;uIaW zC1)|&h|M-%v_+G*DkgSV3|g#J#h2!e`JsAuQOWPT4PFGYRA(JR>=ktrvnm#(ZiR{O z5rGOiA)M!9zP1t)F-_1b9exUU1PM$Aw4J%vp^1m~m~?v6S&(<@h( zaMQ5SiKi2&O9@=YmKo}^%U8Y&cn@H>ORWnYbH4=9k9e@Eq^Y601iEKFx-VV0^{i^1 z*6;DxNw&%~MiqafKuvFy{X)gE!DgbEThE-eMxn5ByB~v$223M3N=x@DzZ7NR^W&@A z09|JGp5u|v=*7zJ({>wb0oWXKs|Cie4jocYz-D$K##Xty z>?H7TVQKAW!)3M}h|1y2%FmWF>p$D)94;LHb^4Sz6PikrEo_stV#?FW8q(Rgn&()s zVp7s`2+@sob*?iBT!IZ}SqLL<{$4OG%*ZucyA2>Jnu&NaypIOT@s(e=ujg!9 z-Mvn|x-dPNF=4{ZqB=R21flwCU5!hi_ZuHTAFlK@GAJ>Vam3&z*5y zH&m1%TZ8Tt7yP{W=ZS96#M46a8QJiS#UU;+NewbLJ}4?F(Ig6(=##jmv5RUYe@Lj5 z4pmkYa~+5t5twz49aP_@8_QBo*#vq*>QyQh)L6_O`DTuCRC+5#mU(}qQUAqk7UIoD z?bq6x0(jz5{8dwx$1p@zSuYQLuX;_p&U){rCL+F92#Ow$)d8Bkn)KJ3-1V;?i|0|z zaujLu-p?&gVo}7_QZQhdv7YzK7E}(@fFo|+1f<-!wKBRKEL?Ao5@_Aojjo1O@ew)1 zx*Y2*Kzsr0Zci2pNl3o7--gc*6u0eAraE+VF!i9XNB mQdW_A4T&QoFt5pBUxR z^+7yN^)Knuo!|Z{_vEbmUAg*~Cw`SpSYP5$73Oo)Cl1 z%B`dm;`0gANX*GBeBHV?n>ftI=GJ?K0OZo0mESQ~-E&7&22peZc$07e-Gwdt-C~k8 zMVCabYMeigr_Hj9>??udB*8@OA~Tgo7~4a$+d=KtJV)rt0_&Pjh+n@EWD3yVzkFN= zAIi-Gf*nL(VYl5~?RVb&qCqNHS;83$z6l8UYzd9Z=!q2$&Uh(&uOJwM9SYK91&Bf9@r#c zhdw9we$dPrDOHsr_h=o*t!8$WY~4MMM!GSEDQd6Lino3QjRNa5^@adBFK$nw)AT~) zkUqdN6=^82GpdjbFiuKfXV&^DKIDbuBMgq1k~ZG5BKzs!Enx5$8lGTp@63YNt#U(I z>;{=96}5`uTJTs<*J+8g5>kMti}^n3OF+r9FOwPk7NP@b@r%H=yf5xBMn~CF6PLoY z)s?NVMYj-5WTl*OwG84rQD!WWqrjUFjD5)?P`wc(z1ee{1ay=gnY05?Lta$&Lb3Zq2H=R@#s-E=U#S!zH^GCk>0xO>6qeq_5#mXS`(Y0 z#>ZKu=IGXLFE2^{9`?8}5qObyiS@-Hry-kMwXestDz<#Rm}oA`6uM+VI&wr6$|!`#?!i88WuA8JjA9MGXFudbBA&1Q=1{3umEORjqk!uH)D%4 z=RC2fesGrh^&z`P_QyC;ykvZq$2ona%ien@OWy$Vq{)pee1*W&-Bw5CQIl{S(57TH zwC@Y?r=1qc<8ODRkv1m>ty==58|>P0b&&$4I+_Bs9hD=e)kM8yg+nA&dhNTet#i7MLb?bn55n3ie#hr4^r>Q*@AwbC!Oqq*iUd(NdC zSJ2ub4DSI0hV@zeVKg`?^2%g*$>iyw$d7411mcKks)QoLQhZFkrQaSl>64~vu zDOaqs+i%AGny;k*x5W)==IV4t%U0bmxy1uz@sxdnEl0CZSUtk=BV-q=E^$P2edaoU z_gdC6v3yp05Lkc2T3OmYOyp(%x|^6`!Ug^g@WieV#XK24Q05Mo((S6?gKK&9pic*+ z^9-uc+_p1H+wtFKyof+nIPlUxc-Z+hK-F%@*37pAVKdLhA8B2brDQE>vq^;53pBak zRokmNv$cjjQ#C-OkbJXf+Dk{h8c$2D{F1}s*jxr~LOsOB`hTfOJ?Y=hOz11sN)&Z|f%S3p~1MN2?X%=b_QX~roXjk!%W%|K3SI9B)4hpdBX?Y+Bq5l5FN@kZEkWf9m&4-_Nf%;Gan zzVRl;AZc{jZrsMQGMQMv;$~dcPBXeX0zox-n&+ygE86A{e?2WDzKv9Oz^-scHqcqU zExw)jZ|3F4Ki@w*bQmZ<2Tj-9v^d07HNP%UR_A|PSO&J#BE>rlMejIaO@pE_I~fsu zlG7ZE_>8-EBRWkQ21v^k?cix1%hgAaA(WN72GTY5t#%D}_9scWoMoXr;p6U@qh67j z4c?lC^8$SGg!bLrwjG+#{STd`I+GaQ#XY)O(pnc?GhiVM3k+Bc@b~DwF48lpe-K_p zKFE;-db$EaFSFMcF4UXVmxv=?}g>joQqNIx%+;!XAQpyc@4AQz3zq6kKJ%`5?J=ue^_oP>+gckg+EWFcY)cAO&_)Mg-l3`Zljn3-}a~VkkKcgWiJ(WGqev)5JQ9Pb>dQWC$hSGIIc_l6md?bR2+up)G*P$ISkZoC;T^UN zo!w0oV1g*Xr9*k_Dy%DTB~BBM%*)Osj2e_#P4rl)8#I&PR2ICbWB&;jQznE$A^-7sCTgB=C116?8yh}%MKcnd1=zP2`eZrPup za9Ru!vyplTzL{yIW2yE*e>MQ=N5i*6FFp?oxzCWleD5>La?wgC1D9mg?JvQM`W&4X z?zF6Tm13&O7qqv8xTEYR&v$P0zPn*pvs*me4e8nWYK1(HwnCPKu@5Qm$3Ts@H|N_h znOI8>kn{YE?jfxXhn;TKv?>haj62*5)41cwx8T{C&xIrnhg>XL=oJ};U(He1h zduUhuCJy7#ZaWEi9mAb@*M&Xg+P}sYzh8dqe_Z|(e1V6!$cJ}71q(PVV`j_E2mGtr zl&n&kG{}(<%;^#PWp`Hi}jIdfB~)s?t?bf>sqarS!V#{UhW_O6oIN4GSc{ zaoGHDrzW%2mrTXgmFmo22>=&V5Ow)ewCg^&3SGtWs%(Q`$%J|$LjFRsUaH(dt|>0t zVIj_it$r(!UJ*#>MUPH_%1x!^o^vAx27nO@cogZtqrO4d$}1V{@^A9+|G2;7i9=}* zbuLn{wRc?H?IiuNanQ@R6u@ObyTW=5S zU9Q3=2Cm+{>u*(NdNScWt!5?-pW#OEPJ-3Kb8niYc5lCS2p>o0jrSWhp6BgR55^2* zTA$VOuDBQ^82dOewwy!ZX%kbQ`1%6Nd(5fy*w(~@wR5gXM;tFl+&epz5nE?tGD z8EH#ttiHRadh;_{x9oFUrK4KMu?^wp@3sJM-h}$a?nMWWzU4`6OWtRa7F}fI5K969 z7)BdVvE|8_+Z^CjjCAJ;<+Uw9-kQzbswz0drFkhw-E$|%{2Zei)*j`NPKv?A$U&OF zHqL)b4ieu#-y%XN<~G(2z0uctCW^WW(f|>mg8hk;?dP2rAOB>E^JJ}p}%#}Y&-US>zZ4H?#KBd*faHKYNJL?#~S_6mK&zP zzE>1TVf50)7pGxjJkQknNr0YIn;!A&$Wz=lD1g|8Su=T1G}WxYV*1{pY}otT@eiFb zBp_?do!w#K(zWcCNM6^TO+?CCMzkh1l8(;4ZmY`di?3bAh%eE&dH=99>NSt@a#fO6 z-;zkrfMy3l{|9@8h($uI({LeV*!eCy*Zy+w^2#fl zc~u4!WqD}Dm&Em(&8LFKPg+ikaIxQXZ9Q%A&p|6b z0;`9#5N$>@t`;eV_iM5qYrQso&A^dZQSO5=%=Uh0rq{(sK#)@UW*bEoWS>1LiMtWe z998ENd5@o-t<>K4wEt~pDN(Mx1TVXM!?m1DtFo)wryuHWp|B&id4ZYf7s-bqMsH0d zooy$*zHlY#7{|~s>1FpnA@Hj5K|T<$731FcjMpB;iMDc^@3`#IIx7V16N)%iwnK9x z5f~=Bz+-`XHFlSY)AH;?h!J<@6t7+?&-tm&|_m_f{(qGFV$rT?82PA1Lj zoGu+4(W0U^TMOR3=>@Cs46&M4IIr=s^j1NiZ0^_Vl~p~&%g9q*vc9u^Yvo>uLa7E)5aH*a7w6Es}h22M;%jigGa+PjBtvd|D zhRzaGt}xY@Z;H#Forx83P@IrSwILYbc0NEekn}{xFb>Gqge3#R(r#|mxOLO$>E6Lt>1JaYr8%5{8u8;)6+GA{$(pnEX-teU{(vwq4OdE z$RkJI>(-9$2&-bBvFB&;L3qSHo>w&>H*e8wF_pEzxWAEGVfV3jAq<)uPjk{g+XjAZ z&KDS3w$qgfBN^)x%LB~YQTy6cdvEW1YiGlYj?v@Mr1X+im1i#4?UTs*=Zo&m%&iN3 zdeDv+$afrm`T}N$f1oXQDnQKtoCWYN@XLz8nUyNi-0{g>Ll>_g{N&~}qu!#-7{mlW z&_(cEPNjyMH)Dg2YLm9MfVoO1(e`oUP9FFpkqgOwr{8#-*FMhl9h@LqKs6|bTdv?lxL`dBdJj=P34}K%ZyX&t?9bC88Q;rzB0_7qQ|PFYma z%Kr3bc|c67m3~s(z0*;=uyHg`pk!Nd*cLk)ZKBTPuPR%9BkOvsI1%T~ul6I@$*y&7 z8{X+#Uh`ez(@pN>c;Tcf#59`>AaQk$Vcpw?6YVj_^t7@gX%Qx ziB~bIVg~&VCAWZ>W}_5@4z>b9r)@tl=XDydJ(*_1#$?^^>(}wROW|O?)O8IUF^lGH z&zkpLrF#&+&G}0fPcX==oU)hi_q2Y=F|C>58!9g-Y&(R?J_{B3?yc5WojGBUv72KB#vN^>Rz5c)2ddp^Zc&ZT_d)9!V^SZ?Is?L#{WO1Ab=O#R>ef_*`S2(R3q=&21 z4F`+>i*i&ABm95ty?H#8Z{I&$A|gxHN@zm}rR-ZsLiVk!sU+JN%h;DB2_YoeC5(L^ z%UB~6*~dD@lHHiG8@uOF-|N1w>-u*6ZvCF;ujh65^6Hgo&htFZ(_$uvE;1#j4^-S&+h5g;N^d1~VT-YV;Pq6u_THyim7LRj5qr;0#qkN%qG6MKUJg@ z$?$szK$_MnpY*`XtsT;u%tDaplEUsP5tj~rWxX7?5yaUa@KmZE5AVKlRmX@=$j)!{ z%bh_U@-XeBi{}!Wv229QuTr^auC!~GRdJHF!?g_?~~$>{sQ6~ZV%GVaNe>N>$x7Q{eIvlR_r%W{ zGYRhI+7bFefE@ocHLHVcUsg&>-sBfTq4xwf)6s@vQ7qdSjUqZtAdM86t&4PPJtkA8 znX|S*%K+`FEDQ0q=J8v03ne$okf)05RudoCGIk1J4s}!Kkzf&LS&&VhOH-b9DVq$3 zrmo{8|B2htUaV7@^}0=|uxOO@bP8ukF{K z92J*%$@Yq`2;#KSz})h)!0E>$_}g>u|3Gf9tZ4asQ#@Yw(uL*g1?h&U-~}K#%!eWt1-bGj>qnGTMN=!*7Wz2Hkv)&n z^{I3|5H`&dmGp%wF{tMeCf7o?V|y9&Y88#1WyT6erjG{1E)jh7FN-ZrbQ?K;(%_YT z^%Gs#sczUL$j%AgmKh>v;NDPSe?{)i>UCR=U`VKEO7Ir5dApo1Mrq{S;kRinTXK}s zQ%~C}pqfI*ua}YULxf}dy6we~#|7Ll}t?)q}U6c1gA=hG&H9fg(qC<$Y znQs>ZFrlSBGU`BRVlbeX{i>GU;3n^8d5wI|_NW}T9AEV*#Dnc|%Yjv<|Sqn=aHi=*0Cr_BM=&9-Klk?byOV{S{7Kk?Znbvr$o0gpl- z_#qKpuuvLfsqPJ{yc*Lc($5zo@s?{BW}5j753q!|yz~%qk&JCV*mx?<_aL6A$+?&d zl`O%|q9qkoYrSZQ{ZAG|h5PHTzUC&}`!HXp7~|Z86E63@slW?w=GC#!m?&M#gqo=v zV@_nsBGxa!j9irW&-F=vt7nJ67m?k;{`IsPrcsW0E8XrUgYB%I^LB()aJKgxzfDe? z-}sL0?-pBThBiN~a2}ET200B2GFY98y@SMi6}lRs1m@l=^HyCU?#MKAsF<9>I`F31 zSj2J@RL|P_vpC(&H7ksN+gEKa##Y~CJ!hUvUhYO2Q~r3h_SlQS)r@GKF7Iueu-0TM z|16QJmsgspP?Jnaq-^x?r*#>l#j3?RWr(x3VqWvsJ@z4cMN1k^!;E;Yva~g;*uj#( z7sTZB6QwWP2X?B3Dm_5gf%u;I0g&aQ-U{}K6LE{txRj01cT+MH`F?v`d!=VN5zjWO zR7T{ol)PE8yN*sZxQSb>bYoxBtKcMD77^dios)}u% z6gDyHxndog@zP+XY;4Oa+X($hfH%P@0l$ePWL=p_t&QRz8<&cOgrI2OwK+4yAERK6 zk0_R^8NOHF<;Ism(JZDQEbGm&P|_`B{BY_ zAFoV^LCs2yrp#^)hikuBFak+ih}q%v_f8%f8x9_e!m=2o!iM4|>lEy2TVhQuUsZOE zK_2q6ma|LsRsG`2&lKl0oofu&QOQZMc)K+Bw{^5-NC<jNnN+rnV&gKw>p3R=oi|09mW}Hvk98~%@?&xm9|Uo*A0{Ft7v{zFy)SfG&v)9 zNKL|uGok5M5Db@iXLvF{5vlI=p$BW4F6p0C98_OSG1-^FC)UKU>Tt0RN0EHq*3G|? z`4TiZ2xYcen`1eH_^_!&)_Tqka}5i1M5j#Fp_M%s#+s8*5SDP=WkSU9VEsYo5zlEP zt+v^tvM;~%9rXT1!TE5!;MgoY(Jn-Cdq_#E&o{4X+0aFIaA}9tU0=f#n8r5pOLCw$B$V3+gP=n4_(DO@4Ho{n-bi;Qn7VPn07z88{|@gR-L_7)bXTZCOR@P zDqLUxoA(}mBzncIfhmy3<>?o6PJaqde6A()Q}qc{Dd6CFSA&tp7G3~4GlZ9%Wl9hh zA$yf+knbpuIh`pk7l~R+aaWmJT@=k~F1i%@8n4g42HV?QJyT>DL`!l)WRu(m@n*e$ z$A6c*o2njTf5)h0*Y0djgV~%?J@1f8qEjNy#Ht1*zwE)9d!!A(DNN~x<%Jkp;`3TO z-xeyoGbwAHrWh!hhKRwaTFROey|+X=l^!7>iJE%(Ph0i3SBu(*5ZH($t86A?m+Pml z^c1!vxDAmc?P8b3@VoY7W1CEdkAk8kW;wX7!T@rV{h+8(*Q#dM-Y>I0Z)N;I%?Uo> z70E!{jhCBaz22s&%{M^oU)PsX`M&1bOfz;#UX(YvHC7xn;m(`0c~3GX1`Ey>u&e7& z$(39wGODS5F*LF|C_+M2uBPkuXJX$uU=w~mx2swuGIcE}$K zq;qG**S4)U1mnfn7U~pScgfE0#GvfFrBYSZfU;W;`&`ZDwefM5@ta`P`KCuA^a+D1 zvypPI28w5?qiyAJvt~T=pzkCUZBz_Rz?F?tL3X;6m>1sXI3N*{D{r)nf{MzW_<~N1 zZ5tR*d(O={N=l3-iQy}J_H?@n;`O**NA*VI2xX4w_61K3EZ4jz{;Qjck&9|m&%`Jx zxof2Oolr_-McniY^}Nc}d%3;TAY9hB9v%8G!T6tmx%Ek+fmd-#nq`W3;<&=B?X)1I zY(gr#=xBUMJkTd%SR!y$ZqD^M%00~e1x4OOe*Vn#wPLV|&%?ukis zuZMH{UPNw_FN8LBlf(tqalz0U?G&1ynT*>zSwaznVfPV4TC%QAvXW`2WKPlfqnwDd zLc$_bJ@IWIT-S%_qCt4aFe=nq^wvAf2OgCh;&?^P{DVjqueeA9Dd)^%6kLs|!&gv; z+$M1mU@02Y9YYyKB6FIvHW+z}4bz@$o%@GI;%>5J3#W6k?oUd<2(GVJ4kuGIAfELM z^G2QBA=k5<*2y^ST+4-4+FWE_jqC#OWVw6m80Ao=e#x_^bAi|MXz|Ad=JXS0MO2KO z?MUP3SD*D37p#Gi>)1Aam^X?&6`6MRu43P#xB8n}>}C2St{^L9{!^&56UHY@+y# z!HfYx!|JJ}O`m^XPbR#YurXD%CRqd7<-k|WyyYS7UuuHcRS25RfjpTR1Z1Z=$>5=p zGleEO9hAJ({@oVgsf)>v#pZSQxDl|(AbmAcyn|I=F2a{8WY6c#NYgzz5(VdxoTptE zHwEvyU03@zr2u~Q@eaFXd;uQL2d$C8Nd z+(yklxoMJ|@D0ANb>gXkaGTPBMoU~xqE426!!FWzWt>bTqe6(JpI%Dail6YN=0?Hn z1wG1AQh2LlJAfwiYXi6s5X;fKY2^1{!s!rF9ZeqFPwRT!9?JG-L&-%NT+cl*;CeSJ z`Klek*W`H9FMOHXdv6nNg_CbdedQilp;>Vd@EFlkf@fK)I?uo4xm84|tKtcLWdvIj z9JGF7K^>zZ*e!V5nUSE!)UE5itV%`vh;Wv^-^*0a{+*V%aj&kNGFqB1bjMx@U zOZ=!yiecm%-228E|1iqXLq?NAI(64jZ11I(q43KsJ8OTwkReAwJp{n&bud)!#=+L+mo)dM;%VQcd2tO*5vy_eDE8Px<4 zK0W0}o7ca#m)vLkn^8m4f10GrFK*SiC)}y*UJO((p~~iE6n~)IPZQ@ySu>(7oE1v$ z?+H08ScjsF3EcO0*<6l&VCj5+OgN+2Oh+Kgx=`_fm4mhwA+%S28)qeGIseQ6Ql1(M zH<9Lv`v?TE1qQh$b1f+(gU2_W<_-E`vz$ons_h9820txwtBLNqm+W;HDo}x{Hdhb8 zux0>;k-r(Qh!1}je7niHDakBnkd);a2tUEGv?;QFAD%vL@5+TncbSLVi z#QH#@0L@GG%yPcBj*yI=VqYq;+BB(a-)wHIU1;-cr$6^i+b4r>9o3Vdq|L53CxtB^ zdEW>uE<0Az0bO!4x6m;M6<+rAhL@GinK}y({8dGb+Eb0$f__o)Fg))aNH|fH5ghSr zmD--S0m*OQf?8i4H+mp>%`C=%?*QY9cZ}>Eqx`}GZMG$Wx|rPbLgOCa!^mcwtZ>W6 zdEep{PH2@&Rf-JP_tP}nIDt_s0@25Ts>-LWM!xQz6wa8;GsQdkpcPwrF68D752*$K z6u5b6^vjQ_BRt^O{E^|SsT!Y-3$3>q>}vHP-__JaH9E|`H(1EO{&>M@VfuY+Ua2l` zEa$L+&?RyK&k(f5FNblvdpK#ODF^0HEuXDx236NeuBda4tHZScIBSz$X;_jp0s z?Llafs`K{8U5(Pjv5s)+3mFuT{kYdk!}Q_d?(h%=^Z~xr9Jl6B1F1S+LnwukNejNE zE7dC*8UKOgzjJhIL8`QkWN7VyO)0KK=;`-;ypIM}H9jr(43BR1hnu1Xz9~#g#D6F8 z4JL})esP+jO0s2&+qVxS^jyJ%%k9c;AP_il(|V{5HL18VrGi^GuyvF2t$^OH*Uv)ZHKhLdjS3g*Wy|H5N<+z@70z{KZvB#O^-=t~XTkD-GgIFR5vTj}!D zk=Dp}C_Cu04x8xSjPSG2z-^JQC7M^d-BDgb)ez&ILTLejw4K7ujZ*=E>f~o78%@?8TL*@Ls_LM2TvLK>oDJX`yL18L$u^K@ z7STwFaY5B1F1;|>R%_ctcpCykZKyL_o>}W52!~q@6q7C>l=-o?mN+F$ngqg8W)PBh zK$oAx?lF8A+&>;1gASG6dt#|$-dU$$*kY^U6?2g4avv+iLRX}9Ox2FVwQ7FZ30w;p~7rD_6-iZbIDNQD|9L;DkzC zI(JgkYo#ZXS=!qLxrQ!klwiWs>c16ZM1Tc+V4NVncBKqxtb+*}gX=Op7NPtfV$Giw zwG?1*IUuzeb0+ux`fbgYtI@;WnGZCG{X2t`19Q1CaQAY{9If|IdR>9rASRX zJI}&=QJgw8!=^Io>-u)H5S z+QegZPr9*#P;dU8UkHKM>(J}~ApaiuPF9@jR7NkHAg@?K^ADNJQ5AhsO#m~^hujb_ z*w~=M_aJ+*ZWTcP{M~6f+2nnP7}Pdv1xBShex@e0o@)BMY=>3`_A5yG;gP z%+sE_&md&#Sd$wv7z)i&j$zZ`s{To%fy*lKx^MGe)`=ZEPH7h@Opf9DQq(@|&YmN7 zAVsN4ucblG_82|ZKC%BpTnj4C5Ub!kW10CxJAh|21;#~|6i6HC|5V~3#`oZMCa#HO))NrRBp&C4$<+!4*a2DS|6Mzk*SY|+0H&r(oxUoVdgP*E^yxxoStVA_+^ z1X6X77vYqq+y3M_F!f8YxfMHArRAiS7@M&=&TLcPE!|QG%R$yT9Ys9o)!$|rYqK}b z-9&do^A4CMVr4a08B9TBr3)aQyvpSmE8ZnAU;u^@^&@aq&VpQM3=$^YqcX=D=ivLo zq5VMk&Pt}o3cy-2ponU%+3jy2OJZ!qctV}mYi1Cx9;2jvWi-*gb8ligCWViz#|K=|CNQ-ap?@Cfw$Xf90bb}>qZ_-w?>({qm;C<8Gse$#Dcg-wL5Ber`{}Sw1_qAbkSBwa1VOy=8m7U7<%%f$xy;EPX^E2WKEsL_0rU zH^&o0m?4bp=LuTgA88CJ=PkG=a|htfU7lNw8l@lFEEr&^>x5wh+vtp8wyM!Q6< z^Tr=_A|q7ipB~Gd!8U$kb>wBS?-49%;S(I%m74PcJb~JZe)V@eV+T!A_scyr|I0o6 zlzL&b7?e@;X+Do>(#zt`oNzRH8Q+E3BD7?rgw}OoL(h?Y z1uer~#%=cW`-jN;lc`nw#YJrryh}r&UtH%mVVS8UEZbX?#dXSC4epk%I8d!N`%-IY zVs*P@m~TP)%2E)F_ndhx*AT7~cq_IC33KSgN`2A*hmjj33d@%#}UH$Xk| zktNTV43+M^4`P2eb#7(OxVbAqF_ck1a#3yVqRhi;(s-U!3R-BF82w2t7df*LPeQ># z*~c6)W_d&X<+^QG%C-$aTJ9Mg*#n7$kb!h7*R*j|)s3d16t6`seY0o1|1sgqqw$4=)7m#j<`Seu%yB7mK>boPSkj(P~ z=84YT+OD4aRdj|-l>vS)!n~X1x8EjOg}a;S`nNHE}#!+T17}2pVr&yEo)TaeXo7v zMRapZ3qSm0P1L&-kXperq{@@0jgXF`w6eB`qcQ+*|N&nbNL+Ov%#BEcNms4;v zu~)kSu9QWTB3#dlJQ|9GymoFgEu!3D_tdD6h7uE_gXKcM4E4q)N0T?>QgC(bbs1~` z`OPfDer)CW9dBJe9u=~Eb+`?>*J*rB^yAmXx`DGVP3lGd9^iUjyH8ASubH7W?Om0# zV~dv-zf;TQOvOG&$92PFV=+!!>5C!T{eTaVR+>n&T}NpY#ctWFUDA`yh}^E5A8!)J zO}JuSGuw1cY)3rn5qrb^;YK91)SPmz55dSA^PtL2VK7Zbm>iVC6y}iaeaq?9M+pS% zH+c*Jy*l!PU2SlaX~HoKPCop&!l|S!UaU;15CEs2-}8I-HilH75hpteSNoeFH;*5vq&#iX0}W(>QUyR`%yRcJ452#&2g+(@cMT>wUG3e#k$3h-e8h0n>i&RTr z$uu~x&GpO!+rC&-hR5ErRTfWrF5f$u0S;oW!v;f5GFiO}XwF2hss6!ItH=JNV`pI$ zpCEM}-=-wF)ibWealeFR&mch)S5CCq^jVoJj)*zCGDg4ORVi@ zK9=zy>Zh}Fb&yXd;TuY`FOJ7m@9&W^*>9XrPa?=)@Za_2JELPUZP2D3#7*9sjv$Zx z#o2jiYUM{j@b}($_RSm=2UTFSzh=O)81c%moZWR8K^xZX(lk?$i&o%uoa5pgoZPj? z3*+TK`|6+2>T}>m>Yj7Br;uer_*&~IZLl`Hy{-wx#9Uu7oo23vFaWo!JvnixP{gdWa^^;(f674h?)yOfM^CT`B|jJLmRYISa|Cojsnf{iP$% zZ$s6(8AdI73d?)Q^aV5Se|&+}oC_8nDTF5)Cu~qFg3naX;!BHmgny8d*DQpxLwdvCB$%Bz>;*YXOXS2F;HH*#l zs_R9#IKr_7d%6dv(v;J@Xet+8&GyCfDl(XudQ7+BCayBU+#rvlZLNq-&Tlp(yYKQ$ z0wgCk%_?tLOIt)}Df#N1x~77**2NKjxArZnou0mtesSBHSdl)2KH?f|oU*SR>xFt6 zSjQ9l(aBNyt(rBx`anWRBYS!^Eu_;kDb z72I@8uncPY^UZ&c*BwB!;k-Yq`wGr|Ol2RX-&4qYxSBOFP-L%v?kn-yq@D>0Lq1al zIm)qRHZealThsXnH*iL~+}5(@KFTPc*fMuTCgWvRvs_^N$+S9L30zQmV_S=pA{P7O z_T&aB%+&tuUAz%uMi|R<_ERS>%O*j+I7q?YW@WU zKkNd0(SB@6k}yssOuGvaD5%?D5>Dyhq;jk{~XTtJ{QXwbs^yX2!8zUv9iT zX|VzGGQ1##Foe6mR05`@ep6jjA{lbHT;h~rXK{LRN05{j$;o6TkR)EWAlI~XkZ+n} z`(!KRhsh_rQ;*IFz&+L?F8Vj%lHD_`|xICDYz!>7PyD#=L7YP(=^kU>=_c&$}Bt7%h)N2$F`el6au zv3PWz-hhZ{9yj0Y2Bkp)ROWiwJ zdU+)>dQ^ypphQOu^reY*DgKnx+FKvpf=48M$hT7rI<<;eFRw`oA-7c#FzhnYE#`{l znW^v~_i~d5{yYIdEeWqbk5K;sRT_PI`h-7wGNYFKj+0CfnI(e^uLNw)v6%S|FS8(T z$YcJ>;XGfHM4tNv%YIBode>u^)_7cZA60wVq!^c?LUFfsm|l9(8>!1KH4;GAICu(1<-8r8NZ0^=e}7q$y*s3XJJMF; zbIm}!Z?A!>RCz@OM`mPtI&HNvC9_*9?1cV^qTY{m6jfr;zIN@m8fg}$P^pp*sl5#? z!kGo&0;Vsl-f zo5E$8tNjF|O>&x7sqmqg7gl{ea(TgmqbWA&*y|(`C`~Pqp-&E;QASQBX);z)gWjHL z(MlHLQOUzW!@k@-J%JOT&hdTm-G)F0%o1iLuBfYoEPC}tO(H!!NGLkdO4G%A&1-EK zt`ZxKstbCrk)*@BcCswlR+A7o)@NRX5y9Jz*>KOp9}_^TfO7=wP4(hVcL1ODiUWTK zA=hr4qYPi6>8ncah!rnw^K)d6B`laQgJ@YEkL7*f6D^PHgdbBC5XEmq;MPXPlLB*1 zEM0jFe@HQ|?aJ`p;95oY=o$cd9jq}_;t`D6L)9~0D>Ld0HG zZ|}MHv3G;cK~uR}o310$dm9m1ANsCm^iDZZB8N5AGt+}gz)Xn5fhZFXGo37*0?+l~ z7F6A1ua-tAAJ=Fyo^i5(W2O!Q3aadkD7jX3IyChS9pm>VBI2w*67iE4_oi!H>V13- zza1;vKz=JL_G_J2UJp`=3h%6Mg2>dtcqZjz{25k1jKSpjX$S@PSKW8oAr~NdAg`qX zX7UCr#pxgV;#M}Eum!1{X03qZLiqC3t{V7O_SSp+PA_^JgJ!Z$uqy{8A=MW!O_5s^ z6OJtwZa)p`ll)m02ljOFAg%Pm2spJRw&UWj4;)+dA0+f{EgT!L$+xCLHd}J|-S6>| zk~S^;?0b|&zJISs`4?yS3C#f|aMh2U8o5fOy7Zus<6vbz!3LvXz8L~kau$9*#$S5L zH+QBh5?I`wV$QQq!7N;a)#sT324&ygZtfI=xY_^VA@B76lC7icAQ$et_0MoX(mRTD zbea<`w4i2htHm#JYYk_E30a828hLKbn#*8A^ad&TD<`Hbq7RaCtVB6}SH$~IpXFaC zqxqR@*K{3Vu%=Hh{ah5YPc1+s2VRGLqFs0bk9R8eTpwszA9_r^y-OIu`BF<4nziWm zBVpJf0R@!k5CjC&GPMW9md@V!iS7R9zxqEAlO1$0ct0Kug3H&6PW{@q+oNHdp#wmj z7J=lVi#Ty^t+Y4snDLWXn2Z|wNkpMV)Z^p7;mChJlhj`=B_4@_iRla!H<)D(vYz^1 z6ZhsCSvgEeJhH7>$@SbAYgz~MGw(-c>R9JV5{j$V2kWVFJ>T|ZA{OqfhPd+l4@YvF z;$XArJ)eA$hy;xmlCr7(9-{dhkO;oA5%t$D`-}b4F8>b?165FWIRCQwP@v@6FFN?2 z4(>PQng8{_%<)H#auTUF*sIe)xf!1dz4+&i`p-|bb|fN>BNJh0kx>C~=Z68Z|6$Aj zx>J7ViC-i{D)_%I_tzQxzdt$e?Opzo3;kWXkvt47>-~p3i|4hC7m}VwT$Jj`(&g*1 zof#aH%efaFmmkhHN4aSyea|yuHf>rTvavYUJbn8;C39C>Gs5HV(8E{-n|BfD4`MLc z%A-F}IFQ<`)V=-KOg3NGAwhr2DWkXo~jTPwR^QNC<02L|| zj&GOldypD9Y13^VOaFwS^7!v>^fy5=YQnKNZGyF4Td=YQ6_y1#%C7CE#BoX+pDVTq zh;;*KzD)j}nkgYpf^m3@OS{Whw)Lh#X^Vyo zdJo=Sur1*5Hv12UA<#sG7$a4j!2Bwqsm90U*A@Q)?yrf(5 zju3WQoR;VI-*Z&|cJ|;aduBvf-{H@_*}L7ndz(@2)3JJ2Z6moMC7oA5S0FRCRDJs` z!DrtTF+CYqX0&sMI7$=;3SY+C-*q|U1BBd;`o;9-nAm#nBKS^OYnD11Sl{Lq-lPVj zDWk5)*TX&^MY^q>-M(~pSmSpK{OxDyG^@nA8Je3(Vdom+gmPtTGw8h&$i;F_=sW6f z_nCk$M90xIDa#%$s;L_`1=#;Ap#Mw2|NP1sVtP|CKF zW*Q|S6LCH9>hKq+{{ZGiWLfA-7TLoQ)WLc2->UtleC?0~0xs~O1zI`di|6Hen!|DL z!AZCPTh92!g~MO)P6S-Rj=yB;|67{+9W&pm1?I^z?a9N6X_{L}Ux4c7wqrXX{__8L z5dT6;z}}9Cdic|b{+G}H<1w+T0t-E0`)>b*-`MZ}fr1u+DN;!+clZmg*numQh7{_k zE*x%EV7bqk{+Ct$msJi&WB8yLTd%CtJ`VAZiT8aKjN&nSKac^kDNlu5t4UQ}nvgJecCG z9P=Lk?Hebo=z{MyS_2nS0NUy~2dR`Vux4pZcR2Ffb=Pd;-LtexGnn~7f=-Ud(||}W zgWkcCg}F4HtP+IeP6z6X*m8v$r<3$>h$VWUs`^fFy!9+y=;dAp)-$jboKw?`9Yho7 zOO+qpy+F~f`iAujnFBPiV`toQPmwR;aB6A2lq2i69PnV1l8$TCv2Pr6qo@WI*b4i48R=^xWcbZJ{3b4S1;A8Gj)gH5a;1|e1yP~O}Uuq43$Gas{ z(6VUcMcee}iwmyN{i7Ga=aNB7Wp}A(VTx!~JplGV@~ZaIVsoGovZ~QUsw4yO>L}`x zXzErcbeY!|b@4O@{}jxo&6=h7%%~eAarX^qUt`%CE3~eLSw!1bBQDJeJCFBd>cAFw zEsp-S4cot3g1~ z$3Je-lfW_tk&|!sdG9Q7_7=;#6SR5Nb3B|fDCF32&aW8b3Ndik6{dc9CISpSJD3^~ zT(-_#LjfsU{RYQnoYiaQ0ZdhgeBD6k_<4HjqWOPcV4yhAUv)?Po47#Rbhha%Fnd(LV-sd32n zT3pY5*0SWNjl|{74k9efN{7R{N#YGGX2yEaP5r#3!?S5|{ktL+ZeNcGvpUz#HnDjZ z%3*KXV|87WFXTeYK~@9~@6O@RIZ(YlZ?-Zu@nPc(fENoay3Nbs!okQkv0~eC^XrGT zC9eI#kXjR4>;F zWSuJ(lmCgL{X3ie#0VMO*D1IB>tRd1j4fO zd-x04GoTtE>$YY5-O1H%)j}UV!>L9J5Yh^AC-uo1@Fva8WxI|79lk#2bW2~c2Hk0t zf&u=R8Z^EA;M+Ef91~1`X-#<8@^RxZuH+-=DWemw)c>r{$f;m!Ei`|d?Zea*#tFOD zyPmEK+4(PC+QwCsw#LU7y@{q2DJwxKARftiq?OSH)?d2l`i^!IYw*foeAi0>phE-?pokGtS{hmkrueKxAX0Ch0Z}ULAo67RPlFuz7cPNZxvzBn=S%UHR)#6 z#{oA$n0W0qShDq=zs{c*HYP{V$4bvaxtZqICD zh0Mae1DF}*YZZBuBDr-yN}O70Bmm;y zYhBB7>!?aZ!nK592Yr02wz}+v9Vr$M>!heZ0e(&38=-0jeh&{|uSi$mJNT++B(ShO z(STrcW~e5GGWq#CCitfQ7(Ikf#Y$&@Sx+w*vk2bI>weeZO~Zt7Kd3WD$&iML{Vu& z&Y`i+!1@rHn?hqY(wTD$qMbjUfzmgVN*@j5E5Yz1giFsWN?=LTgUUb{kh(Bu8cStR~0UX0kc5h$+FyGO{!~ zuDE*9%fBL`Th?y?4w|%nM2Ka6LhHh*aM>E4wt6et8StB_QLBY3ZV8i(Cf?(`SjX=t zAUplJ@$(&WR3$BWWnGdrl_m?dOWMM1hZZ88% zbrWd_@$7>r@qC9^@KEyHQx#0k<}!fC9(+o(u?DT0AYC(Pg$G#O)?I7itM>p+&g3r9 z*?jpLJEVo+)8TpcZFVSusbYFCp(Iv65VSGI-F$``tln5w)`Dg#g6f*bQ}vLQB6AhQ zt`!d32!;|o^glP;;YEeqK->AV%rDkc7l1ZM4lAtD;`QqT?rQ%xZ)L0vGR=lSul}c> z7S1iKJHT0-iyMNb1~_8hsmwe7th?M9oD+taky9QNx;XWJMA!`PkU+q+t)WM}F@lQL z?V!7YaF?=^gP6mH_R+jI6f?^%m<`?a@6oj`sH?KZ0d;AHX1Vh8KZz&QM3^Lt(^ zY+qjK&n}RDC>0}1uIA_6mqPi9Dfp3BEh1Qa@pE@DP?fG?kV#$8vPG#BlLo;-?mgi9 zzcSJ^K$oi~khuAzFXZbihbk{Uujiv`olNb{s%R$#X9OUG$wu4LqTikRz1)Ra*YaVJ zIo1nR$rX^;wM9d0xAViU7FDT?Sjsa)(iSn^;h>1J-;+EKZL7AB;{gA(t0lYxjNbXL z$9cokFfO}LEULh>>*k@2NqP`Mf+rUR+qBBFKtT~vO}0jnP+IPUHecnaVZ}^c0z~2* zhdrHu#7&E?4tzCs{VThdLDe9YbaFt>gCdUW-i^h;1{oyfNbpyvyO1)sp8;v-^}cvr zJj8PjFxEkPaQ5B|$TU4(iPvTjR^Y5x74;zwQvQYL&F=)q?>_r)d)JgmRZ-?6$o4B| z2)g8KKP6X`voPt4sSIS`DbtP1ByI7pNDuNbefDU)k#au4&iQ59<I(92KYiW8Gj}MRll$%G?(trwmgT@nYrpIG4TN{i?@7 zjJ$U3)Drpc1ucfIUFWxavRp-7_KksCyLdbV7kB@1h^L!#54Zb#L7eW#^(;gv;#u$I zg1yDBphgh=DO3{Afh$@t?#+}hTC{FFEPxnzA6OQrjup8b4njGw8?TfBx_W_LQu>2S z9v#@0C%?}fc3A+|90f{~9MZ(|2ba7CU^Bz@Ygr1vF>ilw0sFiFcq?}qOWtAaeSjTb zWCX^^m)dOBL%OtJG2{o7GwM`Hd-zKN?p3;-iyVhC|JgEnu*6>e2cgIfuSd>XIym65u#h_P>AXkM7wYpo& zMHHd5kLrjuFm{4f;YFX5aey!&}G$X$*zU{~w?sYl+neI^a-YNb|FE6q;C`@03R^}T(rbwAAR&{VAy1^A$A&9$KP zKX1;z3B8>x5Zu_!l7b9hJ^YfD50-2bD)qGflt0T8U=Ovhk5`1Z2wr+5`->Ko0WCOc zF!_ADeE78o+zKO&a~jO)eYHzKx18;jyvh_neQ`48)b$IPLN6rST&kC$9Fgj&5BClo zhDmxyMz36(VI5uAJ6=4VZ9P!<|H*mXCsMVa=}nE-8F0(GOv`p#DdxkJ;;$EP0WS{b z5k40Ak@v8Y9uT`wIW}^^D#%+!o zK?!li4lQ;xu0UD=1ERLOT7WlJi_o6@idMg3x+_A>fP|bXcwK>6qP`K7R zm%u|tpB8QD)AI+?4uYw{0yEkZN8w}j{%{-ZucTpxqUnEi0iJ9U9O@@h^(N*O+TB9$ z7dbPIYBC$0mU8it?Yx*;R$5k3A?G7o@B^RQ5L5yFy!)Z*)Ajzxux8+{deVk8XKB&p zf$a9-3a086iMLM#qQ6&$wPHNBbgaq&*@xl_q$>W(`d zIy6U){=v^8=}|^;$x}#LF2}&LqO|g-kl_S&aC>%zaJAZ|y7BvQDL{H9e zkUc;02S3DDea!dGj*nB_aSx4)12^s0!W%Ui*SIQEv+jLLP4z9CWnud6^C!0i&p&C`BG6Rp)}VO*Lo7mp4bI8y(~ZNVDO(~wDcZw9hTdfhp7@#Uw}oyo+qQI+^(#Hz-h zC=w_K_}<2*}BtjY*gC;r3RjlQkjP8J78KX!rY z4Z}i90ORSKofLn-PJs6|pk$T&NIG!FJ8_;Ss>Mf|-r)a?#%tXbMm;;@0bP&6r?%~r zNkpFh=^+k_6DfhEKVeIC9d6kZfdZR>wU%8gd46>`x@i z5g$UxuifOTRcjp;={#Wk_QaYLraa?LY@h8cE>?I|sJ1-m4;Mx=NcQp+3l*KvRX+-* zR~KJVtA$;ZPlB43qohxrL(=W94)t>rtKMNfS~}5A7wo5^mM8@Wg5hMD-6y#}+=2@< z8&@6?2hy^Ko4i!rL&XzM7x>h2iCDN7@O#HU+b3^I^G?{Yaikmk zWw~Dekpu=`R#zJh{+hkEyvFRd^n;m2Mt1S^vs6 z+$g=~1ZF6&nGUxWTe>b|$DNDW$l}h0op@`n!#UV(&Oh_5Xc}ru{&qYN@@Eft&hSrKLWiNWMSE6tX!v6y zx9sQyXMSdk4lG&HW9OZ5wTDXQC;7Cyr|PoiSO<^(5uXG4y*s+uqPIG;bR&@F!nuam zj1BPjI6Uw~>J=}c5f7whb=eq;0fZ9-e30)>t5A1VFkw!}|jkol(dmSWT< z`dX>tOR3b{C4RMvB=XWFALnjYxbI)DV%UAd%oD@r9+W2YM~d_au?KstTiCVaFcGOa z39vsNETMU5xzPz88EM@mja>1BG^+F_})Xr^Fl?{!CrVvLDqEqt?N_ym^}} zPIil|tiL9cPDMwY4;j?+M;aqakLnS)CU&pO5a zjP;*C=Z^OWZ?yFq=&>$^+J-%ygFD%E8f72x+KZvLa*HW;N7sA z+TU#;jiYRR@sp}gpLlu7d5COv9jEZwyy_vOdJf9?l8Rclt-dFV2^!J=Qu_4S(wzv? zk~Q1a`AL9D|mO*L2IK>7$$dM~GFWh*k+xnV`o5Zo|Fu8QY;4Tgs7Wtlls= z-xTjTbNZUsf$jPKu=bWgac*1JXf_FfpuvL$3vR)saS7hITX1*x;K4(158Aj}N{tDx9ZfbTLt~Ap}NAdbBlaYlLUTvOzffJ{?&u>=Zj~iw++@Mc{Tamgyv1iII{mZ$?`pcIyy#XT--&ei+`-VPpsZ{@W zkAR5V_f*dtXYbE)$p1Z?sXu4KFg}@kWrp;xvoZUOz#h~l@t24XI}jTH_s!oMTKiv_ zz^X9idy+$H1Yt%#N2Cw@e{MgJ!ProfHhad*L74LYezuTys@V@Pl(6rg<^uk8^aQl? z|CLMgMNJdvD(`*cS4YJCHvI4FAjI_-iLzHr=l@^-T#VoG&A&c)6V4l*rvAMN&cFTL z#>7`lOC;YInD5V)LVqll&?ge>zl*|c!QfXg5Y#86wp-v3mvLa;a~AOa1YWiZn%I;( zJ}Dnj5fS6P`ETytU+>G&zP0rS!La-><@bf}1XgpdPXz&B3vxcK^e#rJCfoDdNA}zG zZp3@L8$O_G40RxZkt0_kZJ-ld?r&cqK+KBKwj*H5=e`U2fA`nlKyQd`?Ss+x|71%7 zs~F$fB#xN;+9c>@w@M1oDGJQRP)^(ZgC{Fx#Gv^y4m$%9NAs*4vCyj1(HP3f9Cdu* zsbY0z#RRYwvXRTla<-(h`PfnGb7fqe*;F_%1j0-cB{Iom^f@<3Cj6mTtp#c7zQNu0 zaM_wDjr%5t-+HB%mIEi?9Yzpe$j3DOuJ3|&>s?tq>#keRow1vKGfM}2>6pYKo6nlDzf@#19Z;;wZBW^=JnDh9x#8AkU`uvu{Y zK993NT2L-ki!`6kH>N_gLWD#8nE5US0-ZV%M{{yXBd>j-IDQ| ztQI1$h?|Pz*m(c998wluV0H5y{Uj`KS$x~b0S7j3@Y**3q# zXsL-!$)UjTVF1HD$?4S_{NfojMR!JY9gCVz*E)d>fizT^d{CI$#^PrstdzRg}Ac41l>_QcJ(TA zCZnO#z)ZSoddu1BxOtb|ZRNVVstVT%(n+WyNHU!+MAgdE6#wdw_nVr?o=jI1P82XF zOk9~p^_QVU`vQIt(K@?KN{y40(hhRU+6w9~b7fk$D(c$*+X-jmJk@IQRJq z1u#?jI-1w}&>W;nz$DDT6aPv$MD5F-!| zI4tu}(%f$+ov_ufkCxcF7V*6HJwMwl7L5v!7;Eoc=-^Mj1i3q(5Uj&k+8itMp+56k z{<64faopIJS?C+j1Is+!JJ-=aFa!6Fw>eu(_IKi03=%Ey7wE-4EV#~*u>VARqTanJ zY%?3|z^Yb$luBaS6^phCy18(*1VTIz|M@=s?GfkyPmkF5S5QEk*+l%&WUV&N9IRC` zoteUXgArRGPJx>r?Xs9JMwL_&DVLR2>muaxOS=vKY?w~TcH^q-= zdQD@E7Q=L{OEU!EAY14coaGMK(~xPmN7KICoo}

izYPbcfjX-J7g7GohJh$Fm(> ztW!ytvWW$i#h? zPzKq5BO$R}_d3vRt4tXj+QXnyF2tN`x>b~zB&FtewWYp+k4J;CS?RrM11aHeLUOBzgYGLdMDzo3UavEV{3b)-^?#-vGU7tYXr>!E&lu zEI-j3uscQKf99gT0u=8}VJKM@;k}Dw|b32IytdH=86Y28r*@faT(c`UXx;WH3^;XS?dVS zen!kws7}mqotgD0$NKa1N&n;NliRL8U4-TF+q{UHP2`|NSCjiv*wV-sqHDLD-4Gf_mGopaA-iyi5W_;#FkdkIgOFTeQ=& zR>^CFmy<2pQZ!om-g})aoKD6@&y9?p+-pRW^R43-GJku)WU8-rWh3p4{&geq zP`5p5;+;z~WsdLM!B}=jUQ4Mr1uoT^GduHBPiMs_cPe=Ue%*2)x-wb1RzCVC^T+_F z=kYzgMH@ZHb}U$DHS3`ALfdwF<9_||>y2lH(*>Sn95b{u-tHO*@YS|s3cv`EwA}jQ z3USy)vH*p@gjWrDRF)a;6jwLF89^xcGAt&zgxP9ejjpm^xy&P8as&A8J(|$Ve{v6P zY`y~MoD~)rh^eWf;-RUFSe*UO7f~QYW$Gd*&e>xy_JIjQ4v|XfOi&uNE`#|jWFk07 zV9s@BL`~p@=aH!UJATRlNB&mz(qL`K(_Pc_5F=~lFCo*o?D~MLoP%l zs{lltc;}-<@x#@i({uq-98Nd>^BbGN4$;B!h?F|Z*Bcg_zn_VCB=^`%`{0vVTrx|L z({b6W`Ms1&DuFQ0;7Vru)1cYy06VTs&i~`fMHJomv2JhFVJ48M%?*6Mi2TL@LkYd| zheIt8EkbcTyp^r&W7q!d!o4+|GFVHX#R7y15@g>PyvXBP7~h!29Jzl`G%yJQ#P3p5 zp&WbfA{QF8T0dA`D6C@A_fSsemoXWET-Kj;eI`nzRQGziX&s*@aWz{9rpqL9&6TU@ zeW+L%pS{3hF2`dElUoeRv03XL;4;*wZ#E0Rfd#2q0sTMa>Zr+5YaI@Qo2zZRL`qOH zAh;LXjvD;$RU1G-b)*)9jKNVB)cf|BjN)f}FXbzHK#G2C3Di{oNB;cz(;G0OPSdY~ z3FAQq(}`krmc}^qHQh74X3s72hEJ(svmw5UnV-HBMg63d{Q|64ugirl)V=)Jo+Wna zC@@WCyHjB>R|IRuP)q3u1vX*bIs{$4l9XO!{&iqv#6lk8UA^ zOiD%VUTB<8ZV&8RjA{d){;QoyDn6Wn zs}k{d5V9uHgMJ+f_O~;{1E&LaixJgD8}LfPXF(V|HvRI<7H{7P?6bkFs@4ojzL(Ab zWMMM}NY*CxPx)hlIgvuLlX1>7H?Q4Od2Wwu0_}R+5H*1W8YfA5C;SY?vyJWvm(I1v z7B9HH&4LW812WN2@!@3l$@n`@-4@Rp=hWlxXF-Vgk?5boWp%xyBM0K>Wuj%w{4FDH zKPwhWW36F2&;1r!;?|vT@$&dEphaJ8u@c8_y=lTVOptst+N&S^5dUEsO^4^Qm2UDb?#IaovN&Mq2{K(-OXBc+vuTZ zYA)6ev*VATEqxMEYA|Fv`nYQP(i(E%xidoi*EUP^ipTvBG^miP(j>RZEgU9jzNXNk zH-h&5oU3hN1*4!Gg^=qD&WLW#1rCFP$eKp)jOKiG0k!OLDkhsGX6ZIF%$|D+j}e8P zJB61ksnV7~GOdk#M1-aPguq8o*}4~ELT<-cAjV3_c2~%}=WnSW8JEkI z+@j-0{-QE2#rma%ZgyNVL?Unk>-qi z=%&PydE`kEp{QiLjAkwVepR{CO{ok6|BD-$?%acW8MEVKE3EEGhKkidrNz4YJRKkH z9nZ!AwaQRpvVqEg5;c|0u@)w|q@ns$HZ>%X*YT258~x_qVTVDk>pNN(i;lv*mM2N{ z7tVr8vh3_t)*6Rx`pOa=VuA)WItu?2G`2$3=&+Gx*Lo3JnQ-#Ic*@a#nWM>Hc!-}_jOZHR0j1ORGBD_HV*P>ol86D zSzSI}58)pu+V zmuAmmkQr@;oe(xhg;x;kLe;BqJ`fnF8QQMz>IgewhZzo~_R94L{mhHNm(z_N=y?z-YVVvO~z`BQn+(%tK4Yy)2DEV4+sOc(-gRV==Ey#!qWb9%a+%c;?iVC)^Ke4jL+ z`(ckshncTghYj=t`f*;Rf!ZUHA4%_a$1uP}7g87LOCSpvbTKTKqh%?XP7S<|RKC67GXPfugF-j*pgQaJ67$K1k}bM-{xegTk- z>8{xA2iUNxbbcQ}gC6j>PDe@|ul*`qD8|e!99FBiHbYO9H2f52T4MIS- z;CIr}Z1c^6wveP!%z|tRnIw_`84kkQLE3?M1_^EMx*AOG5{>4_2IrFlM~{nvVSH6{ zeIbUMYlm+q^cUOoc}F0h_kKMb+uEjsbfzV8El6ySwUO^XaAW)1ZV5Cj-{SSF=I30) zK?xEYg_KKewrW-6Z9lNwd(5R;nL80yo6p5+)E@7iHmzE;s78&{I@;3Z?Tt619xZva zO_#a_O&kKcYFUp08=aluN=6iL6F)ZoWbss6`GL1rgFLU@NqOiWplFX}`}g8q*lzTD zH%7Ij}ws5e6@m5N3ritmlcj@1o|X7*>c^>Gahj* zK842__v~9dh$+DlAEbICxxB6QE4UDmHQ6ns&(<3$dyLy(*2xF=g0&k7msy~wbhi7H zU4H7!Cb8O;GV0Pg`Qqvel^~XpQd*Z<7nNSmmFYJoM z?-~o$nPb9jlR}22d(%;Vt4~=fm z%=5LbannvKqjh%4?0%RJt*>H|0{QMtRV!j|CVs6I;c)pY zz0)?nb=3o9LglmkCet$n$D@0v1$L(Hvg;_s9b}hmt40RraPY?ku`&J->zS_HLDd4i z?D*u@KbIuoU~yZd#j1>@ooR^m+S|Vgu1o+aBSE#k_M$=)=J8}ZmHtmQpIbXrHY;FS zt+}Fe?VT{^mYQf~j)d8cb9P(E#Ld+{%Ws7M{i(Nz(Ag<}-4Re2ug8tzg0nVTMNpOd zz0)D~@Uf=KI_?4CV@C6r--XnZnDpPUoOad|yS7e?{MhRoq z91Ep9$y)rj%&FbxMggjrHL<6c^X~7Z(iU&c3m!yYaR30j6!hoX3nW0iKR@s%* zR=+o5yJQ}O6_#|<-T@uU4pIow7ey*+LPNHkFCX9L+t}NAb>i8PW5(*dXei8|N*{nq zlpcg&Co@HZMW}+9a}WViZ?q=kw%@GJY=yhFavA)qR5Qw3sHt|k{3umDrr@iuXArZ_ zXT;Rmij(7|#x8wK)v;VfO&a;E4^yQRat;cP${D(gWm+u)4!-uvGgI!mu;9lzog7!U zPHTdYAhpknLhvm^9R!^{d4Z!CUSEK9HMI_$>+I*4RC@WtFhVV=pwIzji+4=hHbGN# zpO;#uT*I%h{=i;`*tw|K(e4M*eHkf-*SS$vn$zYetoO{S7a;H3=;S(6oBW4KxHldL4I5oE`=j=L-5=w6BEAQ(A5B zRT~}`P8@X~XIZuCPciu=Nbl)H6(mWPecaG}beaWQLoMICh`voM*XvMNXQiexn^(#L zR^)AknRsU(>@b+gU>9n|cLbuESUg`9%~YEl8CxaYWbg$Wkin`taplhMy%wvf9XA@4 ze)qG*!E?J~qw#ty`vH~Oz4$|Hv*}+Y%X1r$>R1C*k-RYRXfnm8k4M$^7e2CGHw2Jt z!n{_SCyA&)Jt50826FFlkyx6Egf8kgOA*2A!Ic~gT6rm?C9B$Fy2nxk(Mg{y+M6)% zHu8_zH!43y?c88MY79Cqh(cWOD?~D?F#BXGFuJN1p9gGqc7fj-rvqb#@~MK!`ckX( zD60V^!Q-7q73<#vT+Ita9ljZ4>i4Wcm`>XpCCcMZJ4amk{L@M2Ta!IqJH4ZDQRB+d z726rXQPme*P*?sQugLdIS=MDI#=M07MGmLhgMtoaf$d$u8_hg9a}7U5y9GrFH&`ZB zTMTRIwicU*-?J+NZYn39>fiAPUk2OtONvMIsU(jug0QLr$M&(A3UtoBnMRgTMzZM= zb$thNI}>~{Cr>O+_1sq=r}bB&Vb>LU36C9@DcljPVtAbh9N3&#^FKu$=oE8K2U)$_ zM;I}h-(i!gg>)Aaq^8zn@Fb13n|0=uObX`otv$P;KPsdvKIURjmlgY^18rrguuohq$P7DzGr zn@&xmaTqr^TboE-n3jGNpTWP0fR*$lni31XF8N-DVbWge2xMoU7iZK7TyEPKq?mFc zGrry{bjMFT@}&O6s;$fB>Dqc~o0}H5$yt%3FXvcA_zh!C>SX$3@qG0k`8_xttj1?% zMyvYV7q+LI8C;A8RkIx$Se+i{JkM{v2S@*E(W%;``g(U)>2ty^rH6bkO62k;Q41)4^>vrCbpzqUsOrw+ZERqcF~SNO?ar(YlPwpsmE<=yxx z1Y;uTYVdGec7R(Y%LO_zrR06w?|$;?)$;8B&H_kj>(?D^fAkMWh_f&^3nd4+Kn^kc z7mWl~KS_N3M>>20=lcVj(`{q-&0hU^vbTAA}*!{|zfwC{_-)w#J~cv!bdKlxX7NZ3I*$7(2Ex_AGcH zd;a>@!49MR`XqJ!gzFhuBN|3XHy3jjyzkiWI0t~iG4dU_p9|Y<>mj^6??&w8*7>?7 zH+QeA=!=j`VEkjvoi1FR!+`Hdy)F>v$ZuioqDXw|-1&fq+V%E^cfQ4tjd1xndluq# zjs5%_-K~4&Q9W~CYLlYKZomJrf6in68&-?IH{-BH1(3Ty6PjTHI# z3l*qK(7WKpK&t^uO|KPI&~a_Yi!sk1GnQL!>EsgvhQ%gHo)KMS2VU%l2|HH^y&d>ApCIN$CU?T?|%g+nD8 z$Bo|ap9}M_c>_Ltj^?mokW8RpTf!KjY<+T$=QK*-vfm#1B882oZ9U|)sIHQh6`7_w zlFDs4TT_+5C=v51lTMv^9I~FfVYGJY2p|ShCRrM~su01~ewb(XJtYbor|#aaeQ~rP zyN3*T-=V7mj=kT}*m#hemQopAfD!ifRXnac=rOnC$n?vAAtle`wQ?@s;5)CK@exp0vDXHPZOa9A@EtzgKNVINE2S zpv&pG`)XJ(YLEww(`w<_Kb~7=1$(p#o+;y&K%9aPPUky+b*WA>#bQO|EYz-o*kma= zHy%Z(^bNS(sky(RqMlK#wBv0Zm$T}<-3~}hNL`h`M!$%~oSatC2q`O}?!*&%Ns908{pyxNj_X}K|vv_t6J*se&MMn=e z2L-vPaoVe+5#w8h>fBuR8s&!oSxve&dC5;AbpLeY5Hwj{8BuGcA=k)0yp_eig_->a znuz7HPpjC6EZXj_o&FT~XJ z&=XE~C7XPwv^2;0sE?)`NX{f0pFLrk-S5V08ES9SqgpLY#7J=0x?D(Q9$d1kJIs1$ zj^E8dBzWJVHxAjJ-i#a~byu{AF zx~XsJlJnxud|nWOUy`1JF&dHj;TgSOrF~Y{*Po$jUZr;0W-QqH&F6eV=L&N6S$9^O zBqQ5~*zK~MX+P_7SdT67TQ4kIRJeW@62n5>hFFwTGz}ye$18<@G!|`HRS23)9jfGwd%Jdm+yKgtR=f&fV#VJzXc+B5oCzz z`y3y3uav?aK=~^UNPRM= zUQY4)bu29HoIN({B%?z8w|Hkk-QDA`BLK*ggCejU>Zqs|szq{M!L}+*!9mIomQ23o zCU((?rYR3l&*{VsVC4g)xRB?2VQe#^S{CEsY`xCj*@4Y)GBH@_a#{gUh7YkyT|d^s z!7OJe)8KnU2N>0_+DU)#H8+Ifj4C6UANX_!KhN65eD5tr#r}4<&>Xk^Jz<4;cInsG zF=pq(xhK=fJ`!gXRn!3YI-`_G$IIP!a{=sx^!w*$Q^Cri+&YX8+O^WupadEyZwAJp z{;w#4xvKGiicaK$PS>X=I8>4O7Hnt6hgqk9C9%sffk^|sXuew`49K~@Zx~s1LWNbD_C7UjJ3?JAV_za_>*XZi%_yr`C5M!VI)knuWresCL%TDL$tG*> zA>!k~>hQmGA?DA3fJAPptfj&tQt<=3wNljW<3z8oe{g{?RB$n6zGk^La=2pi-TnB1rF7wvt(D z1GU(3N3FA4kNUiX&dKr4s~K9IhM)0Ye$>UR<)6%mAXHj+s7L%c@sbkl=E;%azEC0? zoW+@xma;?6UKm&}OSQU1g+R5T$O_Ncc>dEM;h`mJy%D^f+}A6IFKcN%bzmlibWopDJsMHUXAfV?oF#h z&AFa)3antyprlSoff^0YRxU{!v*l%T_1rcy0GW0km30?{J6X{^+?s8Is@`2Z&y~Hr zZ=7^QZY?Pa86Set#-5!iS?MB1*)du0YOKROzVUp*qE&A{W~mT(<^cq|&g!0pQM!TT zAcztRMBniecVGd~piB|h_tSKN=sqKqo{Fw+mhF6{WuU_}kavZR-kyHcYA=WZxv)K} z{Q}_Ey#Gg zrXh9(57qJ-8uVVE$L6O1hLth$vusloY(Zt4rNs4Jg$%ofCQ&lVu@#{v3|bv6eL>9N zxv^gDc6ghE!cv|0>qG4!6LC!bZzM1jyT7c+jXM@l+fP!>gG8T?6?@j@WA^$Ym*z%p1 z)%94Xs<)A!z~sFhl`RewrdgjfUt5n1-CEbjTSpaH91B>be7NouDk3+?8&4AMt;Uv` zDx_G1m}}JXtuC-9yGC>a>l(w$i^&k@V8fQmIq|NJEXCz15BdrU9prq9v%_3w4k@mc zi-KZHY`rxLTYFIhI%j7KH10)~eT3|q8f9v{S5A&UXr=i#We98f{K_?eoY8rqgYulM zPMs(==J;hz7uJ<8H}-!K*^7nUy}L?ePtMw37Pi%%GKhUp>&TLuEcW9lFnlg(DnnA+XfU zYINZ_8d%uofspjop+SkQS>RsBtjtySjddS z>T2x33L{}JF&M}nI4aZc4%ulK5zw3D+HF8Sgw6pQU*(Ja0ts^mTO4}6nuIJjH!7TE z1zJtrksAPTbHZK--$S@cnU*OfW`)%!W6mQ?NX9rEqJcZSd<0+d8=N8Aji9ZrhlA2Wc8I46G^*yufc1eBJX*BGQ=kRU%`H4s(Z`Q*si-haR;*e$mF_i|}HFV`7 zo!%emVH3NApr(aY+qNr!VV|os%Ioe399G}w1wOK|48A!7UL~W-M30Q-L{#Gqt_t$U z+2EE^vy(*aWONyya4bJ$&Y)wlO1C8ak@+xqt}z6Dy+$xXxg zY}0Gkj{n#nj(dnzMi8w6q4~V)7}Y!sBJ*F;P}?*>DeWCKQE4A#@w8zlHo_> zkixa=PGAD`b>%$im61gSlD;uhJD8vmOj?D$X=I;@1kmZW&erI|4#ySV#9C*s;Z3#4 z_w3;C44>LK{h*AHySYUFlO+H(mE920V0mIe4jXU9x$*ESw@u9`8QKXicG%dam-$`n zGQGLenH}tGp>MA)3l=VF>(3Z%PYq&PN|-|$t$ySJNiHf92J%G^L;Qvl80*9FEg+tvr7kP}Y;jgz563HyYEsg~byEPcQG zAmebK(U*v>nnQZP=1{zJL#UOYxMLSule4EFKi9y<>ie?=S@q&@zAG!)f&WSR4Bdxf zvGJ|?_03!igjMc)g4$tMq>z`fX5-ffsnwi)g2^zS+s5{JLe2TKlc8|M*WN2lG4TzN z2gsVdqu}!1E2Vd-&uB?biHwatXV08$I)&1@rwoVs$GTK9=GByW6?LS#U$Rl`UT_X$#SPQwhlGx+3J?Hc~&;JH!areN-RaPgu%; zr)|oJ#D#sU;s*axN$h7Gs2{(NB`Q_!jWy&Yluh0PcoFk~PV%Wf>tK zX}ses4qzvj!pxq0C?4dZc zs&9*wz$(D419RB-s@x!A`+uBmTpAe@weD2JJ~tdxZ#x~2Sqkon!p|%-<{CXlIXox! zUn2JAZW&#@HFRo8gatS8>h`8STTjGWIV#nAm*`aVt=^N|kU&y{Dg-je?d4Orxlx)c$2hMR3R;#gtuDgO7wdb7yjT-2XQ>dmjX*AgdHR{|)4^yOo-%NtOxkC)ef2M+CD_$zW$*3ZaSn@+W_ zu6Jbs+kHGI<(=vVI4tKSk3fzrlSaMx&G1v@3Us#fqQ*J=QHF$K9Zb%j7oip$R_G65 z#HAqBnW8xS!Af!mYJeKGSDmjGD?9l<0UcQ7zXg9>v2iV3b8OtARY_2$gIK80=Y``g zKSi55A=~4DTHx-{X<)$*s04ZUE9@+eCi%9Qh4RgXnM1mVZj>siTF~J$=jR5WN==fL zBE$0UZlh5tTUkr7bEjNgqtmeGJ8bpg0|VRa{8EW>QZ|LaaxJGTMeMIt?=^`cwfuwt zm$J{7x6hTBamH*oQql{7npS?El`ki@8%{=6%9rCCpjSY>;2ozSwyn(aWB$+d310h+ zkmPE$Xil0#tUDOpQFRqlivbL%W;w|TKvR$2>(%U-V2l~D`d1aJU*3XtkgN$dca`cT zxeCXJ&Tev>hU>vz)v@M>mX}Y3&`Mk@Zs-oNj5vnMV|j8QKOMF`<^#KBc3h0{oC;hh zz!O#F4yMmdGhPfeWp3<3c;Q%scum4HKQz!WY&X!+=u;3VCr?8HzBGe;3&Q2Ff)pi2gp{KYYeUS}kY9CVG(b zHD^%nNeA7W-|7oS@&T%{I2!eE3R%a;%r>h^kkUM3q+!RD(Uqwf~VGgHp zwZ@V;>VwgRy1hRlH?Id&efO!-$f+=sDm6b{=d&g$9t64DcZbJiS(p zUd{63&$NlAiqBF+hufW{eD~$(rMokOZbdjRcioJe!6nI)!e~NLGHu}OHA|i9|L|8OX0rE23*sD|qzu>&d!2+N zKU!KK7b>KG6;%hS9Gwq#7jq$eSX@BB?2~c;;Gg?~I3z z)S6kHIjkm2mplq2CfI{0OokK10}-*x z-FgIn>>HMB)$Hapk&j$^hY=F*wMccl+1+Kq{%q3TZ1hqW7?-*DHn8i8;(-jJGU6r` zFqp_CNU=TiJD2Oss?B&Ap-aFt{bp^rc_5_)>hV%7DLR|4$7@boN^2tIA?CGi()kE0g{#s?+vf8)${(8s+p~_V z_~pVlB3}wpYd^-0TmnPs!w+|{gXLr0ziP6z*7;T=FF(dCzOitCZpXVg7fN@HLsttrKrvhn%D)IKgZm*t9i(0HrS`M084G39XN_Wu;b zHk+`LQW@KJ8?3&&&W2N{yfi~P7)Cr??E-@dUgOAS@4k4o4xk7$o3oFn+WpLtkRa^E z^%1eK-npwXnXDNAF)0(nr*?#$bOi&GopX~}ZW0pa>_U5QALm+QDJ;I2Q@e5_f;Fd0 z%)O3+eccAJA<0%Lc!j7|(;?QnjhS&Z1oOSevma>SIys-+*b%ga8IBk4WQxS~9W#xC09Ap49Z6D4x3r(d9YtUoxGI z>q+~%gXT;e-F_?!NFM<;(BwdLu60`3USkU{(p53G{Wi7GO+a%Nk_%y=p;u9X`l_LF zZyGbUrR7rPJOi6n0kiV%dhTE?g}vdZB;kBbB|Ru*uGu7GDQy4BB4G??Z-RS4TC;Zq z?(o@B6uzPqUVb{RNbuztA@J!A5eIpJQX)0c`N(ret&_HdAa#foRhX;+_SuOCa>OK7 zZ>vY0VmnQx-b58gul;lVos_)GdTDvnP+VPpW+B;FaOSuVBBkuJwC)sZVS>laq0kC@ zzCn>SdHt$G!l<5!9yq!Sh8?@X6<-Af9JW20vj|<;8osKv5?i?13T40tY{EIe>?ofb9D=zW znU|wDWs;VYdiOogL{=|_9)_3Kxg#JucNb}ePqDzR+3l1}?T6VzO1k(E zkhncr?1ayn>kk(-o4h`n$YG-cXd?u5drm_MxGK>=`~9mAUH8a9#HLsWIqd%s95T_| zYY&%%g9rylDx}??h^Q|j1FChgrHeS;x&y!u$oteU&OAWK;K0w13`)fOSkwyNBoU$s z!dR9<0LdE$gm(fCu_K04e6Xmj_o8IMY5GI|`MM|~2IAHbng&3xd!y7!iVYsl`j!>j zT5mdrKe~j2`00ML07fGF`r=Rj+5h?Tp8H)uKwH1EIk$vDX#9L!DM=Us*uS$|g3;D~ z7Hui?gPnomQmGZNI_|Pz{MX#vYq7Dt*McimSE+nK9b&f61}{~V;IYtpZ>oRZl&MdxUTs zx~xQkryrb!i90U8fc6FFV5x{>EP3yu9xo7)2yDUP;LD@2^qo}7pWnY*%)@P|+R5;o3 zH@1|gb|4_nCRd%y0}%i7WGAP{e!_!Zz2g+`v{0r2ufkNo2mGV}SZuTV%zric^^s`g z=1avf2VRRd6>l70uK2vvNWpEFw3$5y6tG(RopnyD-xD|x-I3$9O2_DJt!|DWIzyY? z%&|GS+}{~4kIdOupkwU;ywONuy;cIN1bQ-u3ql2cFxTwTmyGk#rsavPI|qoF*CnvG!c4wHpqbFj4SA{%}`B$5f1W;19vnVl-QpAXfJ5;23;KE{n$DNmjoE4Musl6KVs2qy5v}O z#7TegyFD{|S+=qt+nO9;+CF6LMWWsuLIMA&bS+dF?dEFhZV(-k84^JD^6j22YVRCV z(-_MF=nvDq;7l*_tQ^lLao4^$PY_4i)jq6Za}Mj4C*GqH81IfB6S1Q;#y~NEC4Q4# zyFkz-!hY2cAA2BkEymB*8oo2i)c&u*lB3%cr_0c_E4Kp{Y zG;Kk?Di8TeG{_A*p`uTks0ATaiIcN}7^=H++@=3jwSDI4nvnnZbO~rLp{f;wp9P8^ z8Tq=gvI3NX{?gb2X<7i++wNUE&p{BaqE=4JRFgx2SzxfK-9ZUuu!kunHkp` zU=PD)5NFB5@Ak4!qg0Ih)7p_-HEQouextmabN}pibJD1!!K29IdfK+8w`P#WqUtR3 zLtBE=3mE$TmzbUtzpc_tn!VWG_dn(S%4j>CbTK8t)orpPCY=f`)grl690N(D8Cspl z_xGG!St0iSNXdEM_-U?gmU>`d7>=O_7b!Vwke_9=ih>pB+8?UWGkLsE`RA}g(}+I; zAe%-Gr^n8$jV85m4^W(|j;7DxKzVhWoeW5bsSBn|c8Hu`t>-URf5a-#&`EK*UWwK5 zx7xc9@^1@Ke6q`Eeg{V7G&PgsW`**$rljy&@}w;~$YGl)VuQTWX<%~)<3lR3N5nEH zBu&*$hCA@r#J#U?siSpHh?m~;gP$IzGm@)fvuphn|J3fA6}#R1HQd#|7IQ;6@kn74 z+CsVw(`dc3uly&7?D0fY?Ow`S3w)l4FOaLg%3XcXzUof70;F!j!)NT`@;iw#0>$!w z_euun$7PjD7ofA+b%?8XAMFW8@33luvyF1Lxd%ti8E5Yj4w->((XR>z@q@h> zyB+JKN*1>=K>>69q|jDzsE{4p=s|&;8$}?jv9zY_;Sq>e%JQ9WHLJcaMYdMGBoFTkHocicd=7TqpMsb-5J!PSm)t! zT!hV;cDKb*s8thRlEE_pQB^gUr8L4h^*3_dixsWABp0XVtU~LwW#Ufx+N3J~Et| z>P$LPec$Y=(@So6Oy8I|@t!%(s{;jyq{MIvTaF~w;9eO0&K--x=_*>a;hd!Z!cQY3 zxWLFaU@6g*=6fVoOL9WAVKeGEv;9k1I94>$FqT;+@gN?Y;#7pWy)eR7Ph`L#u#eBL z@}it9;>Qzob0%=i)){;%ZnkyEK|5s+ln_Jf@h0So`e9{T_hAL0Uo#3hl+rQWaoSg7 z29)DD{XL8^hcs?-KbS@B=WC53masGBftqu)oyBB0JCn4e3e@hCa=L>CNpak^gabwc zvC98LZ3&YiV2hZ#=wx#qq%jb#E0?l%&G52ZZB!G1$;oB!52akbL9$-d$qDlyCrBcd zD^oY12EhH2K~oxZ8<>aVwNxLYhT zR2W&|Q!fo1xj4fR#@`J!?E}T?Zd+bja@fzQsE+ZSzN}o0n7voS3ym7aI|XLV!J(6Z zH;08mSo{RxDW6-!WfC={tole9B3Cj0j#n#%-TZ*e<|CrEz)4w(?G0-?7GjwetZpOW z;LLQP(Lh2N-u0g0g2>eeD`eVKK%_w;J%;%oS}Ph`{b-mltV_H5O~HFkb0x3nPB*Yc z*ItBSz*}F%5QVtiO7V?tuzI6}=f~s93&oem9eSAc_(QT~ctKIT0nlO^}JC)KeRVLspA=rz? zDGF76NQ79_$A$eW@$hbaW*AZ1_PzddEcp9CzeD^+j*76D){azmTE!klv#SA7>l5qa zCQ(+~Q&4=<+I7hRmFRg+Ejtfj3CK%mn|%%UW?O~27>F-QqHK00s4w2%O1J$(3@<5k19re>lX>f9aI< zLDm7*Erw!pJZpSE>G!m8#sWfzFP*Kvpm~V?b*swV*38PG!cwKjKvW0pkoth^_LEgM zc93I2sjEhb%O_i46rQ{NCXMB6fcE2TTFJ||6SL-{RNB^s?odaM0lkv#uB1I{O ztmMyTcoBpgbG68$xBIc=_=?oF%!s3IM8P*FIrR;UzaB?)v=q7QapF5C^5`{L^T345 zP%1o?=epI=2g;N6{hEWxVmFX5o?el6A2YaWp-@!Od?Am^jz$%iu2|1EqVOxnZ(jZu zDKDx8WT2OM)fFY)^ZIZ}_9$>x_obL}MqTF41`|xKCYeK4jR2oooc{~yKh?-WIKIHO zb6a2Dq!;yR;Dn-c+GVK9rb&5Bzx6VS21JbXqG?lVJuZ8c0c?UR!!Bv8sT~1cH@uN0~;1Fcce2+i#DCxW0N4-?VF1ZNj1W#kry zF?{9}iWMIxEVyt)@VLInpal82^~G+xqyo5}9 z6B^TdjVofFjsV(hX~yDYGoMd(K3S$r8Jjg4h;6Su+>_3PxkEB*FBTEl#&q&ru8cwB zx8x25<`76q#CVoM#W(xYVnVU$^9gx~l;f91@!3qmOg9L{l!QCYk!68_%;O_1aB)9}9+}$BK!QI{6U4wror0>tV z@4emM`|FMI#$b>dMdh5@`|Q2e+H1`sN)CC z49*$%rNkWH#5f>zn$6Cv0es3v5R@lkeWnxf-PMWVcS7-zdA zMVn?bQboH|K&4?S5$HM>cy=k!xQRD|eU|C9=ifB#4D^d%-pPgB0f4o$@^4Cqn1d9s z?G3wYxubNW{_*OVg+9tEWxC86hS4|H`qzdKwwSbyAqtC?-TCoZzn#_gk zgsL6xkKuccq8~pwxj$DOJ=GtqAau7)I^rgii0dvcKbUgNvG+BUdP>UhZco@|cSNC(41CkM{lFDq8l*J5an)NPg`hws7#%;B zX|dd@%2H}G$IMZ&4r{7z0mSQx-gB1YD}E&HjlNnd85&cTG6KrpYuH zVWX%>v1<|jp7_mW{be+To_wSI{zwwg_Bm1INCZ#kheRlN+agaW2z#0Say)-0Xf8?L zAE7*GqKej-QZF!tm#6k}Pn24*sIszCJ4BBBqVCQr<`Bo_FzmY*jot0Ae|8IUO(CZj z9ssO+8!BFDF$`4uqOBzyj#D&mc6a1cK5FztP^M?5BLy7`$Szgb&5P`eWyTVa)2gr~ z0GP~A6xJ0by4wgl^Y&p{Z#lh6nZ5=C-9Wm@XL8j1p*GL7FJIePMxfE(Xa8))_f8K| z(@%18!uL`190j;eaz7P5^IJZxd1wteZ#!V#SYlR$L5R)bbf0G^(C#D$55_S$v-A-0 z3=)gpcaQT*BEyvL& zD9TQsj3Z4N-q_LRN>Qhm43>n$BD&li*=C@WQzyA4raMwE%W@u1cOwfidU-dn0;$hM z-t`~-a&~weL~UPSQSehuIqilQ?eAHXUV&Ar-3M8K2KTC}ft{op+fgHlXe!_NszwZ8 zD3QF$-s;C|&Is8lb4?F#?!M{M-`y-?L$3E5#gb##)HB(MEEv(hGyW&hjt>v?cQT8U zj)rbk&&Vb<$Q;lBhf`u2fLh&$c+JUSyYYN%=Z43Qyrc14)o^`n`a?}!?)u(KG??2~ zEpTgl3zeO~4y!rnI5Zls9l*7XHv5^H=J!LBYYaDOP}kBo9$o=(W^K=y^r=!CBvVM8 zkm|m)Cwge9`TnOi%1-T|!?>G)#H}33o}TIDK>RGWh%8Ad{Ein*z4n#q-Q3ScY9jT0 zaws7&c65PVE~^xN)jLM$axmGC>CshL+Ity*58&(zWF;*9vcMAKvi#{xefq-$I9Zyv zXd$O?!zLDTA?vrp&|{|n20(1$=#)ESBWHrfc|PKKGvR{IBpu`x>YpRDJ45t% z`N+v=kg4S}g1?6ogXh?6*7(AVl34K{Ap{(wLp?t`Z&EC1tuSfYme1T>)O$;;JWk}N zUd^6dxG@;@J&osTXL~KJ_Id3dUuqtWO|K6kYONA!Dii>Ui=mAZ-5$*h7q*jgySf$@ zNk@Ng!O6Cd4mL)5*@(_(L4H1&uAVIg^O!Z6T};%hpXuWXD`J(qtZd&LOb$A&_4t03 zRU(!=UFvGQa+A#y&uT2B&e3qws?s#Tb+~Q&(UyMchL{iq$QE$l#B=mZ9IXT3p_@@&+gW%X0>x2_2(SazP>)_^Z&J-=9gdZmNR*5OwqB#EL1)9O-xwPEiQ_WWTj2CUdsv{ryGeLujV10g`zT8a(0hqmu9lAUI&Rvr;^Ni zFL*{5Dn&|+7j5L=#4jk->vLR3nl?l?8(uWa*jzWG^;_m_1~Xug^K+@XvY+m(OTL`U zEv6k$1YB4BASSy0$o&`oMNyW$`2fcCL9#k+R#%$`sKPjwcYXn&Zuq~<3@=vSXN2+wfW2dJ7Pr6R@*c0c#q*lfE2=fvlJsacM%4sX z6`bJ7)81R!KX)m#%~eyv4Fw7un+(Z=c)5(OuiE@v@$cah@ash2Maw2v)Q!)SUFNc0 z?&J}qPf)f$0Hg(26nOuR;a6B{yUSNQHxB=>QLvYHA$iWtYl3B{#Zwur(m$4?plQ7^ zT&yG05s>bDE=&k%HvG|L$!zKr_CeiCxk{H#Z!kufjssFQ+*2CuDRye$IGwPx^R1Btb(<}WW8rV}H z(-eh6t=M4Mi&^LT>Fn$AyvbC#?7^uR$4{-b>)>``HUAq~#rmN}0L4T<({dZZI`kC# zR{MHnox1A&l2-wV5)YRad*OcNB6G zz0oB42Aqek#oYLdL*w7(4Ttf&o(tQZiiReoe6S5(!Nj6D>9w++!!~ndx58w>IQ`)O z32DJk3T-EhgiHh`s_R-9^>Y({A**v{N5t~Ppj1rKsCUrH?+D=rjt-1mZ(wK6KkXKn zWl2`^R5)EFiz!w%rC!Bpz>U}PZTHHa0eYmA;csj0cf|lMtVtH&+q?&1aQ2ig9MSn4tYz;o`A~ zk?PH70S~F7QR7d_ia(7Zsw*nF3EZ)m{2p-W{Bacd1CN&9gs%?(xSXUUdA@jeg3$1W z(lvx;?3<=i|FP(-*yZAQ`KF=TrrVi_3wA%U2)B?#9UL0Tl+hfWf5XUq=<2z^lG>1(@(8)7a%pdn4FG_U8lY~dE!wed)p~MfhA=5O@b*008v>N zj;F!^1SvwZP2=v@QQl;YMFQ)%V`cdp0D==9qS53eMe@PElA&m73ztfs&3lnh92R3{Cx;L!3nHTMjDAnF!kzy+G>H-?D#b`I8{qxpzlW zlb3P|S8NSel+_}|k9_wQt#G^_>7Qt4&>mzp{HQz8u+&Qs z!|t)q_wpF@H*}|cF0A4-uXnsPC+Hxok;)EB6dUwv0(Y=4g!vWe9iK>9SZ)o;+|wer)h~}n%uH#pWVN)OyDo+vIVgD!Y-O4uuTx#3*yjFTtpVL zx%$k-ch>6N(bRRReR5jIAqDx#Wv3!Zx9!i1a3KqU>WUGs#S4m!SO+N!*rG9E zPkSKjBlH1w@905RYi${aeEBSi2D?>>*Tl4MUu9D;cRo3me1S#0 z4OALaH4o_qf&BF7^%vE%1f!KV`YV)0Cgbkb5;R5F20*W1H=c|6x^qE!*O8vn3ij%O z-!J3A5XoX%XFx~c#6#ovQO1X-Kz4@RT6@J*B|{b9Qqj&Pb_MKQX#!#0H9G>#IUxzmzp?{_vV+64l#>Z zEsiYuyTt)encY)8{r9d%`V8(_+d3^j(B!LZ!~tHYKa!k6e{H^@uwcxNk@a0f_e4iD z^<&=836OGlsN}3lQ6u}3hJuCQx>$hDUGFTts~y2ixg9h&O%aW1&@TQ$bM_*(wbSr2 zzCZq30QK9eB5K#T*&iC7y_4^fMeCHzS5yLt+r-+96|-yvm@(CL!yGL(=`4Hb?f3%m z-XgfhQ@m2Ye@o#U5Z%0eQHjf3>q_r*pA5wp%vMUNy6l3%q@ngotX8Zk`A6IoS%5H+ zaTd(oQ12o%IRSQ^a$#brhdyP5Luh#BM~(hy@~c?G?e$aav(D(<3e#hWgs;AGd$uom z$B1ZF567m|cNg3b9f24nr-mOuX~ht^5^KvYrdc*>Rlke;8&vsYT02aE}YFzVB~&tKBga+opSha-;&Wxzvwoh`uQk31Yo zP3aVqPHtGi-){(()s-`ue`@5ctJZb9Op@Vi6K%NR(p%T#8`#ER3&{4nMpFGiOAyFr zd~Gk6Be5PRqEt{di1gBg)!*sTDa`N4g=r+SH&!9167g2b)w;G$Qgj`YoSAtHQh2f1 zuPHRha~T1;`6?Z9vqZaNp@eI#7=xkq9ET+kAV1I>muAn=D5OH#T8xdi?4p?^ZAogX zHqBXB{*1@FP1=XH%80 zg38O5UZSFYRA^V%@|T%b;17{msK*?nTUl<0w2k&;4DwW*66r*R!|doMxeilhBnRuU zJJUxDIg41mwurE*TBawfUqWHZrh1j8)mb0?#uKE`$1*?LK<#W%hRqi6rcidAL-8{H z%<{OUrsp)N5TD&Hdb8t;d$_}Txzx(Eh>(00WrKdqT&yJagAgcw6jJ8hEY7p}@~xQ; zW_%ylvuQ$}GqUUj0ic1j0i~?2S-dTj_JeInZL&}2S^%L6chj1yKudF6qyziyPA@zL z0l}vM#T#O?mp^VleyDJWn`hUDCWoh;+#M$0-w_vqa~Cuky;hKTZ$d}pHUF~m<+N+e z3%x9tdrI*s4)yKmS0siVeS`&lv_EWFqN!D9s0Lk1+*)5>-}~3xf>}aA=;-~~V~t#w z>sRIpjNT38gH)Q_y!ejUH0pVzcU{off?Q^wYyiEc+M6U;gTV#Y4U_A=QIs5opal1C zk(c=dDCebAB&Xs6T=oaLcV8&d!yIBT4g*#!dm$MiKK@lwRcUtYO(xy+x5!F!RXNfj z+?Z_Q`Yz`Ldt2o~PsIpBoO=(rg|18eo6a3Cxi2Htl&oj2CqD#s#f?HT7tGg!o^6a} zTEm_TKd~mb_@g~6qx0+ovB7)x!O3k>A4-k5+dl=%-WMyl55=Z0Myi9thr;MPN;k=N z-@ZgkeTR3S)ovY3wYQ{%DoSTKn~q!A*p%;Geye8&^nV@P4T^E0WX87bj;6P2yif8F zdPCA3-YkA%Z+f5b`Z<1ji*0wQsnKvDx;RC1aEmuPy3e=O@7)IO$HCLg<~=slqQ7;p zpH?YP%bp$e4pK10vTI&Q1B6zqnhyOSfuW0fOFW!tf2W2M2nZI(@xVl>q_cpX=C{bE zo6J}t3k*C^0hgP2+}Ts`mK=4(taub}VK(lSr&KJFsjVa4_FLWMp57s7bWfmHtN~loAtuS8pt{|Mq!>DF#I%{9 z8jwO41@IWDKYSd>gc7)K^g-W4X=Y|$Cqk2#(n5!8x~UtC2h*HY8jnWIRGA7R;ZCF@ zV$xQf-{sGj2o&JovOZOuH9$>X>UU*85de)(*`$G@bv83t(tl zqt4gxxW}dw-!=!$KaR51%d_jVnhnbVUVq#jBCWdFLvltI4e1R824`a}u(>-t77oq0 zVzqtJZvR7WNo4j9+}Enr_fm_cVNycEA+k=;Az;cw4cy5>8;U>+WlyI0U-pQN4YV6Yr4oTf`?au9LA#{v987XWjIcLR z$y907y$HR6s^NTzpHk^wptHPriM*LGW(gS1O)rP%%>IVaP7#a<)E9h%qWErEpY^}> z<0@uTQ@tzSTQC=WwUalm=-X05YYj@>t~uSLJeYIjTrR6o^L}POG)8HtvE6so;r6M- z{!(Q3}bIFo;M!j7PbG`I*ZOZ8MOzk_%sy))#L$&u;Hg8?)?r`sG~X4Q#Vol`q; z-KBTzGZg*sX}tV0E26fWH zP%cZTy`TRl;^jBb0)8SU7%-zY8oT&RxozFd&NW#}R@pd}94FTN52_k!iwL(D#01la zLpN?l=e)aOuOsMJqA{)AGZ!~$OC78uT*Qebs1$0T=TCLbsql7h5!F^O4JiFZBg4!@ zKw6Ti0D!79B|Asxit(&}mXI^LHuOAUA$5(y=yoSGnyGx^f3=fQ z|5~rAS^T=>^3dG?$Ju(O(JX!bL(+-aSPlN$rflz%M$&bBvSXu4n+dfK)jwE4;lb7gHxV%m;a{#3U2J*5FeZZx-s?ZL8Q?f)D{L2^GyR!*eL+m#-{NVp42fff})+ zZVicg3#q!@b3PUICDJtr^n{6>vo#57b1$t1u1HxDW5-Byk-NXi9Kx zgH7kJ`wIlX-NdavgB>{8+%q#y@Sg(%(~x%EPP-IEqF;lP1k4-H-9&^LxyUt~vs`a{ z6_hz^00XmQM84ED6y_aS3j?ltrG&>ErxoJ^3zC)(8A%Vl%wBZR8j2((2*B!3*1=5b zFNws20gqI-`iBn5-N8~VM+_@TR{?jg)nI@bB3kDl5WMTI}XP!irKDiuVe#%g48c%aW$CXp}{f&w*f~IPd|c^ z1{0&2HNlE~Dn}R?7(&8Vh`lQ{;`R}Z1lRV(2mx5=s$K{fL6$eQ@i(N}iKO=tMm$M` zRT|&>)^#-nkdij&sWw*VNdw7CgCmOAzsleyL9F$k59~=Q*Ny^2=j`_@8Wfy5T}t0l z5a20yg{HZI4Cy5Tfb{%V$Bqn4%WjLXSgts;us?1tw@7D?!L#1{U69J9ryV!%(`;=>h+gU_UXoyS zv-nJuZ$b4U&SsTz7M?q`+eAh2$ax^K2CIhS`P=L5*sBn~7&8sa3m(#-p=fFa3+30; zh;(grBgQ11GrlxdY^bntD>d6egS=VV!sdR;*6-ixiT}vjWIK~no4Yy-Wb1QZITke- zvVIb!gewa3)6ufbp{>2JHL1>5EY952bcq%+SO)hD(zk(#^ND}DIAGalgw(Y_9Y3>9 zY*Ns|iupup5m9!zaEWTV7>)%5pUGCY=TY!IOkeiudQZEC)5nS*h%TO;hsxWtzcn2B z%;TW{e0}3VnN33F%B_6gJe%&py3797-#pSsz>2))an+MAuXdrwI3%f@&q7lV7p0fv z;~ZDs(L|O3H~%zv6_*Szxc81;7s_PXMtk6JEv$Z@)2=+9gka=g+vt$LqbLOzhvO+& zc-pvsmqgQ6WpBm!;(#arX>4&)@(IE5Tyv~=%Xm!!hpmC=VFTyud0snwU1_^Pt{Z6o zK-${m)f zyoq2zDyb^?T$}9^z1oIPg85b;4lw~Zy(oZxoL{afF0~DFB!*r5MQ{vlLaupP5M1*F zwU?l0r2k!*-J!wIoJ7V@o-ot(o}6m!pyczBJ$R9Nhm|1|F~ zp?g6!2Ildgxum(DN+h~u%$yiqk|62J#l^&Zp4esRu`PzfA#6=>C1a^+Z|(D4 zv1NYxdI{@|d+l8ES6x-+y&Fn{kT@bl^TUtmh zkRp@5C>zxKDfNpfJ3F0H)pI2LlQdzz)tun9=|G`2)s{WNM#k=&IH&1Ozo4zo%ByQy%6ui}y$vbO8jTZ09?#1X@uR2z!Haxq;r} z`8wOwQuAO`Ipoo}QT-fVM48Fr#jT@TrYA`m*|28{n_SD3UVWCUyqWabE~{?}Vryg^ z-7egBES-L&!uCzu-ApgLF>%nYV)4hYcQ;gO%_D$hyg0k(Z<;<6}ar=GtH0%(3$H}?00po!a1`TfGzn2xM9y;;BSVK>xd|A`#xCZoMde}#@C!~%P=uN(30@&xwp(e`&lRdv zc2(t!m-^&KlqiTq=(pEa_SS0~PcfR~`<^o;4;nO~`Fk`xG#=n(mDQ*&M)f?uIN-_; zX1G4NVviHvOa}jem@Jxpsp?lf{_Tp*AMW6$Eqxoo@yiLw?trO^I#qNX5osU-uGk;R zHEw6DXuBZC$sCVo1u1?&OB>&nXR=T=D9iQKSB^6Y1K57L7283VAFtF1nJ&2F&+klP z8-<9x4r4XlLIlZGA2j>6epdT+V6E7+(PY^r7NzbWtqT)&HVBS?tOB^b26kOCy$4~u ze{(m=&@d^SE;JMX&lCWNyXmF(Cv4y&ZKfHXQ3M%FBb6Yp%O8f+p7nhZHb{q1&-N-6 z4e3ZfFQNKS-m!nyl9F9`vWB}0u-8N>B)Q0`VYR4~gYOO@tzTVy+% zR>TtDyuZyqd0u9|=yu+J3Sqa-L)FHlipqAXS01T%M2`@Ds}w;qcHef9HF!G=-g-~^ z8eo1l7M4C*fqeECf|Q!xAP1?aOkoydZ&VnLVmaM3SZ>gefk5W62c&VVD`d-chUHsN z>&7rry8&o2?F!v7Id8q`Xw%0k9IKyk-q4i$u9u3=9} zokTPu89UWFQ$6}A??$c#Ieyesfm$XzWPdv~>5}?;AFzQwk=3Khq<)r3B!fqy9K=uw z=C-?K>vq-55YwuppihGpUpR_lc=vJ8BwwjmD;1!5iS_n^cH|K{<&H_7sW9S4x%d_; zw6;m3mwh9}b9v3p-DM<(ggYiDK$rgx8yU!1t!ZEna%_;MO@{2x#+Mgdkk#**L~hA~ z!MA6EAeoqK@r{We(yx^j+%xQ@LE)wOhA)G-O}Gz*Mb^@!dE5rU6n3AJxizH{MR?0z zOztJVAtm+M8P6BIJTA{RO^A^g&x(L^K8w8cjDZbFFSgaxm zfR$3YZ&k|m2i>aNHC!9pVramXKh*CAIB_#@324)++RDS`2&t8d6c0mX9Mf}@ONngY%qRz+9qtC%e%3*ChT;5-X0Qkr_AK%GX4iL~&>BlRMmObO+GR>A~fnXWj@<`mP->qUd-6Z3qZvjc}p^zk22B-aej!Do}I{Ec^;Gx!!BKd+jc>a-nM3L5yfC|}GaDd+HYCa?p zdS4Tx5E{>F`Y!TyNOeyu)W1V8{uPe#FRyTE31HCV8&&}H<&H`Z@tkDXUF z5trc>6ItbqH=hgd-!LHma?^i%-=p}aj1zU4%H%;7rt#L%foq;OzSF!k|M%}I)#Y0_ zz>N0(6gb0)mV^Pw%;6&+5Vk?Dx64*}K}FSEE9b1+ySa~Mx7%d)a{7gV8O(YYtc_i&6G)|Hi2dDo5 zPxfzv|F215BLdvn6N;{%+L)duVt|xY12z)yi>p#~li+9O|HBIh2_j|-ZY8Pn3XYv$8r{0^&`?ofvYq)!TO*EJ zWvk;KndJfh`9w%Fe(3G6&GhlQvh6HbLIU99$CcK6cjc$M7>)7rrG#s;L)5b~XEona z>rCXH|KXPZ{F4VA^eazyH2R=jMp({Y)9gkRCXpKT)v8X3cK5%#y{q0$-!>Do&VcaGbLkn3%$a0G!!P`aD%zIA|uRgW)gDfG&dPl1C!O3kwFh83B@C-=gtV zSVVk6ge$YEO z#KXzkBY*t|l>|O3BL)dD?TP}q z?KT%M<3?{@QURp8P-*m1Tk!_9dq1oSO8%N&s#YhTCHaW>*noBSuM%zv|#P+?&A*5BCbdqO|< z77*cr#KikMm$!Qy=BnD#07 zp7>C^?(wP8+2OYT`AqzpkAL~ij0OkyIh%+>z!OUUt%?uMuP=rFR95y>h7mJ;Z0rsG zMuEwrdC(bW=fwkN2zKj$Irrxbt%4V2u>Rr`AcR8WfEiIiND0C~wMeWEFu_YoJVWy% z`+2{v&wRh}S_&%nVGPX9MG0KF5SuVMrATh3_J$18yPEX{Y|1#s7~FymiEaUygeQ_iAd8eP{iuiTvInudc0Pait=Oh2VEYEu2kr0t zywEHQz1NX}eob>`UJdj)Ag?2mJpVv~`V+s5fPyBNU{cqdILG0HcQT$U2 zUp@uTyua=UdvvzuBJlZtclMwC+TX!$SH_3{M8B(Us_5}rLGvOEe{YDtL7~9^3Wb*6 zR44$*1YC*57bB0MP(Ov%-xRCoYbN@CM(pn_u-~BmpAq{P*Zj|I{NKO%e}fULrM8d| z7f%Ja`STAp0)kT@?5&iq7y<^`)ianj2jq;X!q>1VA)#L0-jpBPx+d9sEhVByieAg9 zdd%CUC5~+3;jO3&&-)}PG z5_-ygQ9%=Cb0U=8l!}7gnBkE-69s^FI3^|v@}s5XdI6pH*X`P`Pk#H|gtOKjM5Zmk zK?1R9-*rkSrM|xyi=ubeKtM*wV$xsT8d=ZJE51rU1+4u3uek8ojw|P{Ma=)SkzowK z%pc=H;o+bYpw_s6N%u9`NTHaIpPxUkoRTuPV!7CTt1Y*D=f$fXjAw)HkG|Z22Q>B< zbexALp;=!!rA@kz)5tbFSDjwEWG5gx$P^lvHLicHwIYGM}9C^wm#(k5)cmUexmM zT6xyrt^B9UO6*IKCHX3O_&l}Lj8>k0%?ntv*QCG`2@AoEdyEpzZ19EuZfwEzyXA)! z9Qx3X>nRrb6XsNe>Vzs;QQTu`lVC`vO;I) zMfbel2Gpx7G*DG_9=~-A7#id+;9HdRz;XEj38a3ZZ_*q9IrlfA?aKXNu!X zOhCeKQK?z{=5cu7+Mv;YZu39Hod5Bs=_tOP%F`$pJNirwj(0X!q{HW@LIFw?Z`nOT z5wcspaW}bc<%$?+;Ry#OV_{auFaNj>fHeM~X066-Q8uK(>Vh(!h7^C_Y?Vwx>{c zPoY(oF!0*E$nR?u+QE;p`&LunfeOG+sjF;n>-IGG#T5r{KVJ6On2q$b(wpeYy!I~-dD`RxCSqgzEMUf;8u@r^<7o;5zXYrTJdu0y#ACe4ic;~{J;4&br_X#4#m5Im zG82$tBpUR)f`T}w&~SIV2Fi>hs$ReyrPm}_B+|=6(#FY$l|&} zkO@(f0Wg*Vl;ZGsJt4y29z5r9_|P;DJedVsj7t7yBH zJNvp+-L1LHT<&vA#XRnn3s04UmI5F=OlQ2tKGWhINkV0|_ULQA4clV;_0vX|z*E@) z!mJGUjrYx&Bs69qeh*m^l{V<^ZgHElHuevdZYX#6gYX=-!d$E{)}=&&1L*uPX(#W$ zyEx_;ymJerR133O56IdgF2zV}dZS)v;kQDsxfEuX8|6YGAeYoaMrj|ggMo?{<4ci3 zUCjsuua5l^^v=`M&?|P2OYt#PaNA!d6Itcg7*LnyGLUI2uD(;Ti{CuiptCEYs16*j z5&RE-MF%ddvdTau;2$3He|~X<>Dku0OKCg%ju`8=r%#{BQo&>ez$l{rBuF1!x$lmP8=$=#^o*Z6buw6v=)sO!b*wlH2d~r zvei7M|AXzj(?wknvv~3dGLX{#p)!#*jsO2|d`D!S+&9S7r_LAIrt2vtxhpZRdt(}F z%)8r=;XImY`H3tqtt6)Jm=NQXGuu|OJczKKbTyJ|$V{6#YS&Zf)Pf#FEIkbtsmB%k19$c7 zaItEAf1I#nM`LN_zP{6y-MUDyOCq6fh};fRwN1$2%vyZU2b?t*c9vPj3q4MoDGOY| z>D~7~Eg-q`u%P*$8+jsU)Ub9D8j4S$#=aWAn;AYk;*yWshk13Uu|>A|vnLuOMUJlZ z$=`&LeEErXL1J?a!LoLx-zdX;308(hxL8WN0^BJl$w7GsJ?ntbdAs7Z*|8#r8-u6v zq7dcYG9@4dgh2pG6Os}}5Rtp*s>y*vjB61TRG!0lHyHt<>E0Cv zQc?VvcF3q>`)?A^50ucW^*n z{ub3kkkCTGMdKqdttxldvz}wsohD!L5sK<7FR$-9TsA5)6xQHE{7R8piu?tfE$u7M z7UeUHo`s|pK+~yrb5M5JDmr%$pSGr-h{M$Y1%LjMK00UG3n3;ig8R3$*y^XAdoPQ? zzYqM`Umy5j{YlYF^E#~aT+^Uu1YhP+*~Hf^&7YriJ)`>P(ZmDwO_S2sfUdegQqkqD z;1XLa4S~u11$QC%Z|hSKq7IhPW*#_Q6lls#WZ4mkd)^=iQ;-%qVyJx1>!7XxZuZj4 zM;m^Y{E%WEdoLJ^E~l~-B|wQsOG!mc?hipsKWPtopZ}p+W*B-nik^d4cr75UL@gXI zD@0f}N=SboDJ}8launub1_KN09m7>RWmp57x^}3PaIaFP?AwI06ePg1ZcY2+>5ZT~*@rza5XP~Bf{ zusDmq!6I(2^O_R+S>5~j!{n)#KQUT1j+?7dm^YScOL9uN_s|LdwO#52c^EJariY*^ z0rJm~`_)LK)Osen^xS%s-@KKYXd%Xy|}6BUcw zvM!A3I-ckc>Gk34%FjU((K_8OR3bQDB!6l9)s znru?#Zt4=E;N&b01qnG%uA1%ZXvw|r+bmQn=?uLIvfEQh`!*kML~5PZ8P%Hzq}N~C zq~u9)=;LN>8C;N~ww&^k`W$+jMEr_jSy}$#9;&8+nr(0Bdy&JMddQgWZt&z+Rp}H? zt==I{q~U?AZbmFK)teww>>|fABeuC{9+r}aA-pGwMg)z$_59^!>RW4N4$VeRT}4hF zp=eX7Q&Z`>3q1Kd`Y?JFf4JTR(@w7tePqw5Uj6NBX?htVM0RKap(Tty0>}s#YbHE| zR977}^EJ85+?6D=Qw#Km^FjF9h%gE9y zgG+ z@?2ileI21TUc*&o&yBU=0y5K?2{)sij@+#qH32SVPB(!RPXPV6c7-Q(_U69SPb&9&^0_PF>p8^)W78a zVuU%~BE>?0tX^lAc!AF$0>E=t+iBAnOWSjFP0-E@U6eXATmAA)GnMELT7+kumniVL zY#=~@QU;gGhDj|uRJdz>bOc{z`4ybpz=u{(p;W?$p^z^-a7Yt^$0^>cXqtkcr$d=# z0YxQK@!sjhZPLSGlAz7;h~9YUhiauN=PHxw#sm;LXnse3yhm@o_SrU$G>}AE&u1O- zl~l?R3Kv-0yPt0~CRDF;;^wxFx&8bZBA@Fb-dnl{;dJ-B*yrsDDi^Z->Rmj_TD~!Q z-Orh0Th5u|N^N{7#pn)(a7jK?>Q|a=oMIJc3HMIQ+;NO=HP)$avm7!xe>#*}U3jrp zZr4_VmV~VVaS5Cr_{KGxgnQE1V?J31RjG%fE=ClF=9zEYlr`4SBohV)VGD+wtdK&J zUuX!@h=EEY3&w()4IfgO&dxZkx5qMtYU9I%x?PBUgk{*Rd}K-mu5yjyM&#>1x?!=c zr$N&g^&~f;3S-fTbyIfRtj{Rab`WR1HzTo6660{OAe4e!p^xYG5f~z_ZP?!YMy1d| z7o9Wg69EEk4PDq)=MZV2{EzneRlcQ)7@a3AMfQ@5`i zH{;F6RcrV}+nyLyzMJ*g?W8QM&?40f107xP*IZsDaC`Xf5zjfuPKgyY3-1D zDZ};gSW7vy*L5gfRlx#kk~21!KTg*3nU8m;ZH~I&>+-bfQ!eJE{}6KTr(>8Y{T z=(Z=Howg{%r0MuyC?QtFJWoh`rNdKhQO&Mr`$Co2@7aZGGnM(1AhGO$9}4jGJTLZD zWx8RKnu;6VhU+3Xa+I)Q_7R$nW8;(u-!+{TR6(a~^Z7uE8@K~jR?)Yw+_;tb;%t5W zHeTsxZFSv-tQa#yI|e_T-}Pa!sn=M2`E*wyuVt=W{W6+*b%ArO(5_*kU{;{+3K4~U zx_q`FQT5Z+GMQ$R7&vR=H^e6KEDe#h zwsi?FS-zbMMNhW}{W>q>faBi*osDv@`7W~HAJ?aOhga=?q(uHJ08xj3f#~1({6@RC zgC_4p7iw5>Nj{N#jB6x4(v%5@py^@S-o}G_UF@2kHvUitg@5%mxVq%nCo!{(&^{s} zVmlwFX=q~KIqJ9c_nt2DwYoAA@7~Eyo}W!un43-(TjSmpOH$M69Ii055ModOQ} zm9Ud~W!ymmSi8Li>9R0taW>{9Up3M0Sej2IM?06i)gR*f1EI6~-4D|a-9LyBTQQOY z)H0zw$$DBwbjEe=>pO2j%ORxrKf6zCP4~VPkKwV-^oT8yUbdXJj{2lr0Y_wI7H3!tF+It zz4=_}#sJPi2!m#2X)i4J5@S1`{P6if@bK&KEA6nZ0(T3ctOCA*%O;S6;Zl}NfrI9@g7-$9$)yaB(@&CRJVMiqE2A0Y%Hi) zGu?Z3a~DPCr~)|7&cSTn2Cx0zCEDT-Rj=VVORLt4+=QhzXM zwZzvj$>huJKqAM`P^veRixOkcr`SRVAYuipaQ_}DYW&qmksTV?WWe+@R_}^`o*yDK`!+EnoN(JeTL5kFaU{IjA0ORy)d=?YaT5B4O?2eEG0dnQ9 zq#7~Tm)l?lJgRmv;`dih8(cx3ccVA=xM#(ti?asPrBvig*qvgm^v z5qOThvgBn`^ySUnv4>QCKYCv@7Prfb{C||aWl)>{);3z9|CZ81@#1Z8Z-L^{;z5GD z6?b=Uk>VB{0u*<5FAhP1ySt|Z2@vGXKKps+JRkO%+2=bm%rN}cz1F1y^LXC&!^`hn z(R>7~<(STa|NNerT*_4>BnnIYS>Xqwzv%WEsxiw=x(ENCuQHTx+V}W7U%sW_iBUi1 zmDIokGYWFa6A<>UX3~^VNR=56DL+|J%B9|ys^GDIOKq=3`^J=PJ z9T?QC$AiK(%BA=a$<_bBd#GPQCNV&)+FbcKtNu&TE12TnWUM2553b?cr=Hv$_xJGu1ZX> z0MIjsiE~5;^@?}wUS2QqxXi=tc`-J;uq8ChWwgBNq|8e>_-!;vF4or#h$wvT`dHLz zA9GhIehmUmF9%)``S)gYO6v)5<2$o~ zTL@3jqj&`g65XYF1>p#B_9{1_|EV(B$LZq5e!w+iya3#bjT9{=CedlloZe(XaeNp#__u`%<}g~}>ik3r|=)(hbLgX^#@7JdEQESZIT`Zr$8YkYWgj-HHZRmuM*T-wO7Oa3!+^RJJrv4lAqd*e%=)Ku=)am)KN= z&U_5GJT^FN7pl`| z%VhEUgs&8d!x;Kb{%;n*YeN_nIGm7C8W)!bPV3xB%eDQJ+A@m-z!t%npx|lYyhB3o zCZ6JPz0b(s_3YbCB$MigG&G@ZVkdgyJvq>Pt}$MtMmF7kb35$odF*CZtNN?_50IZ8 zj(egx&7xxuewFS97s&~lYjAUU1rcwmwcQgo)o$Ahlj3QPX@D=wdgG8Q8M4r_Xfpa` zHI*xqgq$mi=d>az&DhP8o9<{Y??CAaFQy@}Fc)X=yK*#*uQpe2)mOu2VBjfo>kj{D z3SHQ1ZWeOCf+5TwD0h8-^i}ZPEGQMIe(rb-S9&FalZCz;ITGW(>jjMhu!(-IdSX9K zdzU5s0!v)!jrUqSirW5EI%Eac@Zj{G=f$1I^}XrsabJl|u2-WB+^E?cxUj)(+q3Q@%hx4ZY*(+vC6UHNoAO~K0V#{w zdIUY+6znvIq7JI@3whg4XfMkc)brj&6hst~fzCgKSadktgy#z|&)MzrxrOej^RCR* zN71g$hwXkU>^IMg+p9e3I~hKhtT~feC!j;0w~F|^B2Uo^8z#gqsCdg749z|S{nb4Z zb|LDcFH|&o?O*&;1YhM^H~Ozbi9tS|8XLMe>)>0$pjF=eK=f1xbx5&kdD?WftoduR z);TtVTr_I|?aCelKh7;8UnF9}er&vi9uOZ^==P(h0gTEWO=++TNXgUuLy0bJ6 znk2urG`-r6Iv_|MO^y)*B(oQ`uv5l8+*&hJ@VlgxsF%y3Em`iC4>CJJK^Qv{2WYnk zRDFRXf5A?xgNnmXL^*isW+pep`Wb!h?QmcS+PH;vttS||dyk+dWVK>%;D$oomKrGP z{2!nC(HZ242ES%98#_!nLK%&iq5A#aKW5Ebadz`2t2O+aeMrYb%XK@RffncfA5rV1 z9^^lhNI!L=e`ujgk-I#Z9a!FyH-s8^U>7DbiCP=;g{z*KRikfpsSwbn6@L_5AN>PJMH_QA`&An#s}DFsVYN8 z1<M4OW+e$dyQ`lYP`kzh#M86$fr0tjlD z6E$7DyS+lakO^hjIM?4_Q(0+5o0fN3_2nayDsxi;ZuU*F=9$57!2Jb zX}VLG2%(Dbm;43ZRHUX;OQe}%CO+&!_fKzRh44v z1hAZJVyQ@tlREpE(6lwJ^nQiRED`niSykw&Y;sPnk{XFqiNs_a7 zJf4TTuGdiKcFn(F7;oZ++meX4<>WM9$;fx2eE!$wY1vQd=_M3}0@or&;9r#`PVfJP z?r7XBFfgKXGd$LGVca%reX6t%wD1pli#*u^M>C&lSe62%K;zz7KGuA0rsmtj3bCQE z5uo^?(>->zkY~Su4B+s@*We}Xs*{U8Lj0NzE`G#g-{?a*pG0W56`D-t&TyMhGuiPf z-(Rn-Ke`w~+?DL}5T3HeO#%;WxH+Ks*ZI9N0Jqipn4y15qPN!YLu~2=XZrWc4ic74 zI!qP!o>`BL@EoO(I^x!oh%pPWZw0k=!~> zcR^o7@?>F_d;vO&fANc6aGyctTPjgP=A=de9;03hbOaUObc?I3*U_jFm&e~v+BGhG zv^7sAqyZpO{%zC4@$8Ja9R~s5e&of1dHnwIg0uP_H8u6YL19*ocdNUr7K17Gdc_|` zHnwQ(#)IBW+sdh0*hQ`s)rl#$`GP^GA9WN73j1WWeE>`^0FbotF&FbXssAW*ZR$F? z`$xmL!BUcfNx_`u!zlFsGC%?U9RuVrj9W#>n@Wu7`&m;eA{-s>mwC7nl~lR#NB;-5 zqQpbPeSG`rP}Py71tl(RqG?ws3%2=o*U&3xI!RL{G|i6*408z*vJTb!Ms)4Hy0mI% zs0&ShA+*t^NNYrYHmeL4;?Zky3XZP&BSF;D=M1#h|6QieXY80Xf5A1Bb5q4bUld#y z;@MS7hXpcmG@eh#PyEi1t7t!2;K{vqFn)lv2KJM>_8cpUanxPzR5$t9&GXCKz zrFI#<@I8C$VL)qWOgI6wvV-r@=kcW;9&fRv?zL$spx1j=RzYN2^eqa2Mj>@v*$3YJ z^QFQd@bqAQSy@Z;Ewu%v@xTLu@u`VR?k3j@P=8Gl}d7J+B}h?*D({8r7Y18 zWKD836QS(X7S2k!4cXS9%*R9Ww~Oe;!M(LH(~nz}A*<aa4rebc%+NYSkE|P4wD{wC*Mm?Q?pm(?2;&iB_SbE^1z4CHmS2E z;uY$jL-&wS1mLJWvU$(d9c#+Uj+ zN*UBv71~IQspGMQJ7oAzIzx6ubLMdb&%bc6Og8 z{e}Z7>3s>}eudhJn@j|@C(o2?^))bS9nQbwCCNVEK(v^gHaSMU?E>FGPC6DEKKdZg z(OebzZCIg$+JX2R z@qSn+6`tygm{(zW$02weimcL*^v^jKbWlilk(f{VWq15ONBV4&^LECM&D*c0ersP- z5aa&uIftvvY9Lc0^H3%gwI!8zUBVHyG4VuY9y)={J|;^>N;84!QueR+uUGC)65HFT zo!2MZsOfhBGo`*{D{C{_k<^*Fex@-2l3;IO<`h_*M*YvvKdof`tBb#`5@0LIQhs@d zKnh&*p;ki+>3812tyWfvec8h-}r=*k#^Z*TcI^!88gyu%oWo5B< zOJ|-SqRrBwNlYMs6ZG}LU}RT&pE6_Vn%{Q^pW+hy6NRmqop$*6BrwPW7`jofrH2sVWzvcM_@$3T#Z;RZmt1 zC?T(V`zPz0Y7_o+uj;;tQ1lRId;LZ_XRhB@Zo)I5SO0o(?^~3(=@4GIkPR5#yCLpmm9tdf?I^l> zi;3xy4#!XCV=yk4d1+S$ICbK@b3*87gU!tnAjxWR8$>GdAe0+Etb@+=FuG1PH~4Kj z?3XJeNN+ZF*I!YdA@G>5-%e#5lcsUA!Yb!k5*{6%(<2of3yD{4cA`x~>|pr0@m`(o z4}V^AP2t>Lp@YC<${_#4>M{SflqmUAG=`{>5c!5y4YC;F!9T~(4~vqsdqBIBf|SnHZf!3m)9#P#W#m5JCeqJz z$JWt$I7C# zf4yx6)m+oBKIyHEVJSKQ0rtJN0@)XqMD#-w zn-6Xzdfyp~q~ypgWICJpUw9jUJ=9w=gu=rHDBQsUED!U4dE`-!acS0pS-vSbya}m! zj{$!U2CL&QutO9GKFY$p;W>+q8cv~Ff+>T})VVaU`s%F?^@uAo$Mn)?^q#XroC>{C zcy!Jyl4k^9BTMDeo7vIR54g{HWzaU)u_4Yf!YevFBpR-IG zI%r1Uf6tkaNwuxQ<;X?lk9osZ)E^)QroAt3qJ$`M!uuRz?73g`#!ceNA4RF|1+o>uD1b<0Az(p#(ZyrbgrXmWCJ z46K93izk4dA*rzJPqfVdBaTB79Zxv!#;=b-V4rhTnBC9#P($(k**Zhp!|{T+9euDH z?R098n(eS5GLq(FPs7Qgl>4gEiWS@Yy*`~~R2E33yU|JwI+-iPPgb}eg`5&KY;pD# zF^h=?lyG6Gu>R9NABW7xwU<}Ualn24c7JLGx;1TYeSIIaRv0VMf?}&I_F#l9ZX+kJJqknj>4zIJGy%Vbh>B$Zqw;y;JaqO%kezJUud^P;7ez;dT z5K5@^MotL56wy(JywcQvIS)81i9C^~_ z!FN`=0KNE2j8f!c$G8TNWa1n;ylQ^Q*Bci{BMy1RK?A(z#0=bJNF7j)6k-*eLp}|O zxcfEtUJL&{$2k9S#=@@+38<0|KceIMihi+sW-K*-7_-OF zz}JVRKI`gM_#Y+G0ewrF_Pddu-z z#0YDxw26-n9Bpcvk&DNn@1%(Sv;$YstWtKqhgw{$BH)yRJ-T^`UhF%gv&W~9RE{=} zCME|`ZntN8>kiSoG{~Q!Z%2x)L(Hc0zR9BlfO)VWG(iO(n6>hNY0@6iaL(f!>Rj_m zTcPta=3HU3jUHF4r5wt7>xpq6Xbl|QXxp<(QFj2>n{KwnXBskGwTTY}^+%&*Ok+rw zH8Al0p|kvx{_Rq~{ZPsZis@@_}!-ePPK$m?Vsu2IQD9@xOpf zEgF{JwkZM{w-%?c*1P58cNV`w*w35G8qf;UJcEOF&iBvgdsMdVu0ELOITe@(yh@`< zm5`U@JKyWrK0q^%qe=3I$^+cdYf9T~t(K>EgyaP_fnKwrjZ|B;cDKh8EPo!-+xZFe zh&7M5Si;_K?aYZ78GZ{mM0Xi!aA3?DDRggz?@w}vF-W&P0A&==6L{ci{&;4b#9RR# zJZQ_$&)qq9X(WML1(@R@z)8Fkdzy$LF(zW)i_y zsmGPB%=FOkBsMYPO1<3b_(J84XSmbuUUVIOTFk%X)HDw(!9MX zgqLe`VSvDVR?`lL%R~BkjYfS;ha%s9moz$bDX8WRj>1;a**D6ivlKVyr+^h-cY0Dj zqoPFLt*?v8e%prGcbgKOc}Ijp4EC#D00-@6r!Nr+nJQ-L6`H-PY`H9S-XaJL+xaTs z<<0-0B|ysw=|@*JK@5;{cq#s&WmdnNhbvsd5s zM`Z0M#)$*38uZR=5O-B|Rj=`KDxUcZ_)Z{15|AG-H8?i&2Tpo<@MUZ862c^V zcs4WF_=c`CXg8kbx>p4X3Ibftg#L0OvJn^hs?p@pnOk z6I%t4t*lRV7c`M4rGx(W-jz7Yd^Cp3&ro>VJLRhYnmkdt#fQz~XuPncTvpv_-)i}nc=o9kc?+Qx^ zqwPVUr}^x1ex28ffz_EV(Zq_pll;!2n=MSV^w*UrbM;=-pP|Hu!Fz6t`Il|H8Hb8L zHo`qD`XRJFX#YN@;an%I{eaur=XB8d@2KH(0)Ix9x#Rq}t(jVtwxLiB5_d4OFV|A~ z#g~XPO%_7*ZTOQ#a}+wng3fro-Sq9dgK%o18{mQSvm5Le)mYwUb$df@LJ>%sgo->3 z%~y<@GQfU8eETqaY`Fay4W0hV_Pr$7LpeM$QW3T5o{FaC41y(9>ACK1gM`F}%v0O4 zhR(HaP0}MDWI76;(HAotx#PYl+#>-(r4jjmhHP5B_lTzVqnD;g+s{z1LVkIt_YH#& z=ZgE0E%lAVWPOt>p$_sGHsDOpJK>Pbyht=ua8~>}E;a6lD%6br=TrlaHojE4%Co{x z=z%6x;Ly<+fcCEZp$ul=4SJ?|O#bIQlB$%;_%w^lal#EPI_%9VE+!zs9Wxcx+9!~t z%{f=BB~{xWADXJOh2hFXDHH_oohz1em=@{U0HH&fKB9o`{Wp4MwLev(+l45;!&C9i z9;83tQ11_Y>tXblbF`yZ6dF!!0=9iwxyEha2?iA>^y1u-Ph33{VJs9drEze|XFcP- zu*jnK)O~t^O53@MPQ5CNZ9y-^r#tnX?0Z#CSMaPMmLC;*g>i`Si_=!)V%Ji1XyhKd zrZ{`YHPd4!L!>P`^%=$vL)gnAZd-w3m8n8sd9{+J2mPI}KbB}L+~x5(PU)3dZD_I0 zLY=kiW-k!k>XPfF0aO?_z3?CZooqR=nUD4_@ZI{&x=W%{R8m-@7LK>k$w$AWXN7~% ziK7NolmCjop^w-SMgG5n^8f9H{VMR(K3Fo-ZE(qr0u%iv{H}HI6F>Y>vM9=Do|{s@ zQdb@$-(bRs^EqGARE%%?EQZp0Yu>99uO^)vDJ~)zjFxLrTRwuYwI<7rovG7_e-o9y zbHhv8%`cpK_-@B%x%hd}YON&hHHj&IzN!1^UE=YEVY~Jq`3;zbZIvKH`Z*SFFYJ!C zllkLlQ5R3oTq+q)Kfra(?}S9)X=?Macb2adx}dHR%h`B7wu-HWVK2f&|Jb$Rcd|Z; z-zLv|X~xI6T8Sto)1!$Q7)r8aLe%nx!sLW?xS@=kjN3SZSRL?1%GBn3*|6Gp&M)on zOpKM^TO%CHzHovXjzTfCvPWBzktd>!qh*ex#2Rw0y8bUo9MfG&<7&Fe&kJ8{c6wxN z=(%{FIcWDt*iF;_!OSn^rfwL8$8jQdMfWeR-Bl#Ur$;OwN#?_GYft)cm3dh;>~}frR^o?9U*F4 z(c63>KCyN$Iw%cashDQNH0UP5*ccmlEqWu^T&sWSB32>e>Fksrvy13I*A#E7tVwPy zgV>;Rel?^vb_6K`sU73zCs`dzdN+;E_P6Ztp4NH1;P<8bSH@l*1~UQ@>+&#ZN*Jzu zMSEcvmSYLiJ1e2vprTEc_|(awE00#bsN4uGOlh8wW%kq5rTa*R5)0z47LT`-Bp^E_ z+fl)zM|OpYfa483pbVF^2}fA(;usFO5rs( zycz+}NFm#Dg%@|j<{EmI;pkczW|%c#KQ;l9CaTj2S3+g*m6@Ll92NO)CAalpo8x$I zh5_6B!k^Te)xU;feus#6k}sI(ebQ>rG35148hvF;JXub!Yc|xY3X)qRvP|4#$x9S2pGQ|N2wxbK`W;gXEDAOnS!S zCJvziN84hchDT_{Tv?%QEj`B37aozt82Tqn$ z9sI%nIn^JPUaSGHL#Etxj%6G5O?E4e-5(H>a+n0wbeNn8GN~5*E~Y(N(W>>!i>r)L zN@dnyYw#F=Mt;^oZtvn^fOxb_LWEvHYOdA({|%2kc9jChrWe>^Cd|L%T>mJ+{VKW{ zhD0aPk{2~SnDjcA1p|A_xly8R3ai@0?%cp}#=5~3=sTI@h~`(f-H9z%{W}3??&!sO zo3FO%bB)!fE7>f&p?_t&DS_{qFr`e>N3gPS*5)#a7nv4`szd8_wchZw84&IT&+d74 z%}7@uK0#Ys2{#7vifa=G%dbunrGxBF(c-OT+o^`TN?G-(Zc|IF@y&&Af`YscJ{^qRe%D)9i5<^;M%S!^{UHxY+J2(PVmd`M}PDUr4KO-8Bigc0@Ssyh9 zP^e!g7dMT}2@bqx6Riw1G({r?W|*MK;sl0xWz)%DZ%L$mz1)VOLfYOS`Cx-CVCqE3`Yla z%4vg^uHP)GXn7B|VwIieAor_vYHpxP^Lr-c_=M{CHR^sk-+fOrED+)b4AYWWT#ORn z&32+579;`j)PnRUvsiOYsD~T?GDAW-%DXhSpfu2R?Zf@HOCd--_UN7w{jpuMjaE+F z=7Dmu+)LNe^8z`hCaE+^t@W_| zJqcXoDdg*Po0hRvPZlzQzcGI>)On{m66NVVq`cVNsQQ~@%$@6`Ow~{i57Gc@N;)Lm zX2OKQ1|ptm+JG_W4IZ|JYE*=XAToKnTv(fZH+>g~oSC{i*j7T%&Vl%3@?|?O-(UnS zgTKE0AYpqmmoZzpZfl@ZbEL)Eu$OYDUvjk%sbnAK&~{I^f`bc&joV zOhCSR6mZb*ANl)grDEN;`cE&w$Wf_FYX}IPKI?-S$6C-QZqih413R+!#6L>ym6Z(Z z;zFF7mRG|GOD&eJC|vf}4T(y(82I?|U5Eb%P3_$4>f(1jB3#+2OyHAzfW3irp!c_J#1zun3r%)G=@-8Y}P|Eg5CXDm`yEJ8_%mMj|E}>V6io@c- zB+Q4+wIIFnZrF`XK3YpV1S?0)DE5-+R6 z00Y#ve&Lqn;7iRYunO4&+qk?5o#j?0#mOtC-`3CjMf zclz4jzGdT`nl5Ip(+tzp9}=?>wjkGWU17fY1L2JQQe=F# zrYd6F$!5uYGF?_UfAEut)si6i7Momr<#ug^GCoHOWcWrJI-s-!LnhV`!ka%Fe#Ol_bt_D zFuHEudbxb+;G?FbSoC{f`&f;#1V|#_scCL=tEuaGU7r$P*Wd(JA0|?|xb|(-X_@d%VdVve zl`mikEA8mhHT^H{))&-u5H}oO zbYwE#wpVZg%UoSlXeH2pHNA5eK-{ta?s?2vpsZij!SAp)bU2Xw1=+8)tAfnTQkYn? z=wY`nu-I|*Bzf{;=c4I>{PoAjs}39dF5qNh-by#^XIC!?dOvBM)B?D{Y=DwHEpJ$g zN9-^IpGao=Am!cg4PmhwO*r%wsQZ4%Tvo;E z*2%JYxTy}pEU#R7ocKPW#zd;wW^yBU0ElOd?SM#FCd*p@(=jpe9$2%snBwDoF5u?L zU8lhVbFnhzhZwJDr_s#bHW(P|)T|={;wLFrH_+QR~Zw^~*hl+)9sswFi z6<7ZbfWXt@mgrGXQm?arM5V5zY4J&(MC2E@{hAn+PeY`5nZPI`XjsEm)%1k)OA&01 z)_F$AXQU{-cPJ3 zEkIz;6^j!|`hy-li+=j8<^rB;KOyrcy=06+J$n(k1u1uit|4Su6~UlU{IV=<>QJC* zw3j}3a~r62u;2n47E#odVKZ10f7+qXbi(+~;+-Ft~lVeo~D zc1}grhpgAhz zjapG{qNwI8dLT#z#Do5vZTV-j=72!@lp z_5Sae<^zb^L2bX7t&1_4DmHu?kdrj1H;27%jO3l6*=_RLUJ4(8poxw z3!{FogdnW==xd~*qRG*=(kVv;s81>mGV;UgT){Cj&*d4jkb@~pyeojO@X34)&RYjd zDQtCcLUl0QKOcI8**HOM_x&uVZP{q0FI|^X32NtRGndvM*|y+Sz0>}AX!!PHlPPUx z(VK|TrR(e)EzhC(Wag=g*WW5GOlSEC2_ZBTps}pd19A!z}4atE>6M#uw_#*fkYd+=c$YX-E^}~y(Bw@lg z?CmIyN-qi>4y#l`+qQ}P`RxtlR7|9DO~|DcA{JG=w_8+`mq-Y?@YUV!U9u-JX^&v) zO!hv+Bl%< zmp(ehAjx9jY@*qnEdXwNBD<@R7goe&2i2_)W3Guf)uADdv&>t888zfZE?8;b#hK0` z-Hb7zmI^VnH!OG6#yCPE8*6tcI<-r7bdWuZXV^YN_edlV=dak^WQ|Z_fG?gYZ|F1^N_=aWCnGT#v2B+GA=>p;z+~*ExB%*Ou)^0*gxfocW>(uIi>$(qJ%ECVnimc z3$v7lP1>TNhnAjlc@6cQm5P>%HHPx!wWPmi3=rO3Qo=a(O{qJcDKjyxO5YrukR%Q@ z{^}VgJxplBAVzWyO!`t4vwbn zk$=E>k&IjL`aIXcGd?^>$L)rOcDoyM@Y(Yi%$O)Pm1li}1tVh8Y!P%gGcL|*`x7|* zl_VaFZ#!xlSKFYKWR1rFPYYVx&8hT>p_-M0UK6j4OXM_){`8T)ejVez4(o;s z$LKoPV{+Tild+KOKwTwZYOFnE|AaOlhxbINtTKE(58pGdZH}VHX1JvPd|%GN%M{vU z+6x#x!eV1>{v|8NZKbGxJOE6jVZdgMqE2p1G?R7kI;91f8D%fTxv;HC*tb7M z9rmQI(>%MruH9USwZ5H9OcnqgCV=9>Dwc<(@S+C^BZAZkb=2;_49yz#Boj67kXuV$ zS)TjQ4F@;yCLV*^vH_|~I_K)D9iye(t)HpOD_=?cIKaVw(RVJh7>nnsoC(5% z(A784igpe+3*=Hf0U>HQVE$in~sdnG-}ZY zbwz0_tD7-pL<)!l%|S|Zk4PyquQYHmze^`Xlsa88&i&Pk;LKKgjii;y{Iw}!1HVzA z(N8Y_2b=UTQR9AL>7zOOya$bod_HIW+T)VY;eFSm)sUFh&;q$kaj_8WY8Flqi_ejs zS~qAHy#}y8n$ikATzk~F8fZMA?K@X0o!9G)AzJHv^gVa*=&_GRH<-8srM zdkW=e1X32rZ1zqVV=@_3m(j|w)Q3Oe5&2ngY|#g=ShX2efoc6mR1(1cBYPGkGTSi}C_RdN#~S612smV&x6qj|MW* ziQBAklaUDUsZ?RU=lS#DgZNW+YVBqY$(5K-S%U|DlagzG&w(;>wsz>j5i>RG-lDHQ zp8$CDU2?wSSlqaB+Rt2KpYBdx!b63;Je7>JetFXjF*~3 z)&-}=$yfW`Ia^;BMy4CgA91y-u9;^F6J`=PqWkGYFR(b*RG!6+G}5BIQVJE0jAAH) zVc%JKVfV^0ogBO|lT^+w;B4wA+sxb3+6c$2|2nuKm{lr9Nxr^{!c^;E^^J=OEDX-uR@zncZGWzzG_$ajc{e<$`F(H|Zf zqHZ?lUEDi#)j24I`>W{*p7*YxNk_X%fkqfio=@V)e0ih>kDiwV{poQ;Y{AtO7zg?S zRNO#MqYVBspYSFXn9wOG;2^e5Mh2_CKHQTQ$qZ14WQe9NG|N1*d`~Gip%irxV5T%( zZOamTE)rR^wG5rkC{Bt;+OBF261%Qw5Ztv*n|1D)uHEsb*a{;E_r-$-4F7pR) z@MTEKUMRnBf&cxnS1Y>FJN5JGWt$w-?DX4O0cevOFs;f%#c! zyk^5=Z}1R$*SjuZC zJ2u(9y*w_(+(C!S-^vYhwmWGeynUASUd%fMG*LMpAFZ#sP7Ux_#fxHJmVhO%E^>3B z(kwM`S}YeH$L*Q+0r5%#AW;ag~~!rJxM8?z2yy*;7A_p$;+7067&iqQ`b2ZKR#v;>6%mMZA`28WAM#VqXlI z3pE;yo)KpC5=sp9lE6o9NiQ|8%ZK9d8LB%pKQI!%}txQ2FZAdW?G&1>t471@+1 zt>YlFC87Clb16~W#}-JUjm9PDtv!cR_RSK~Oud1;@p-j}7TQe&Lz1V-U{YFnVHJS( z$tp70v=RM-zBsXoG`7X?d_C5%C`*V}Lkf!5XS)f{cbg}NIbfxQCkKOUY~RR#-V?E-@5{T^JOUm9l1oqG0z ze>f>nR#!Ub!v; zM8g2QtDO@cC%949K2n#zaX0n{uaOA@u!_|cRtCm898tslTJ9>VL-#B}-Zd`Uqo!Lv zH!)N<&NX1-#|^BXZsfc`m~T@U$b9#FUWm83=0X7j3)IEJ?pZ@Xl$G+0WWn?9GgEc} z#LVrZzeaz}IzbOhPj}D3C+C~1#*+)=nFn+fRQS)whv^12eEzr;(G9_sc4Mq?%LVY_ z#e2DLaia{JtYMqj@a()^&V*6q;Y9*LB51nUT;#Rgc%+Y5dTMt?Y~|T73S0G=il!!m z5b{_YUGxdlegEa1G&0H|4eIY zNFGa0jk!-{3xuu*La4An_3i(2ta<+T$J$pliCpNj;zPd-^#45D{X3ZgTcI~eH)9y| zTILJk^(DO-a>Iemq3E2;V_vz7t%e-{qSvq(%OKUN*FLq_Ha8AACcipFONSa`&v4jtyqz==al`ov$ zzX8wwrmBJp#tG5vK()K$P0gxfJIKG>rwSV2L0m-B1a&DN|L&7|Umv08=bjOIHc9tB z!3E@@t-h4Kz{N1GICF7^8d~$2ymgwK^-Tpo#qRO*MdCCVv~v?;zbQu&7u_UeN} zu|z1rg0%2U5+4{C{Bx3KMpFD6kfRp8VelI~HifI->ocW8y)AK8yvh0`lYPJ2vL=MK z?ZY@!m1WXy$6`;6Jkn8;IQ6e;ITX?cUuB`o*ZB;Toq*6%ItS$lJ8&729pqP&Shach zKG~76DyxY`^Et>ewbzohzTL75IHqEQ_C>=oU8ibP(rfp`W~b!E=~l37Fmhke2`O~S zJWkeKY@?T`{xh;mKpYiRD>WEOu#(e<^k87_e$^5 zc92&!f&_}&Ln|1^d+vJY-)2o%=7&ACcrci04{9szSe0>%Ias96XLhSkfo_e|3xwbC zDu-?yjx}oe$oSy|Bgx?s?60pxFfxleC5oLFZ1%3Czvo@knX}7W5?6VlYj|H^|HW=H zFRQrRo2wjD2QNLuqGOxtj?eZWCSTs)s(CZCSyK_Ng)evW5-yl_e;vVWuHGD+8Jm_2 z?0qQ9R-s9eqND&?QzL0U%TcYoIlNYq%=RdQ5Sb$N_*!K_=CP?0$KBkh{>P!?6~om> zaTKFv+@6@^gfM1yl}L#b1ljAwHytawZ?b&i@do;|cVf%yY+VjruixPXTF`jU$-15$ z6AI~7*v`~YDk?iEe7(JfzO4dH0&I|!*NG9Zu6fMVknDdBjASoXm{kk`4P|jfrHD2y zut`R+C}&vB)_qc|pt*2?x(oR9v4+kWDTi)12agCh9*_NFk;&)H6v>^+ZCm7IDg)B! zs!)|QsZ0y`r^A+Nl40|zENp6u7KQ_8=OoEf&2wKj6{zr`T1va!bfl6Z@EK-Cdn#Ec zHAn|rl}t^*3Q4=jdMdK&G%$F@tW{YP(Pb z=jyd=n}t$5Df~)bD|+o98G;9@nwpgIZ)VRJO3+=Y3={61mePX^Q%7Bh?q0lT*pI3_ zt4@gLHfkpo261D9KRb8I(E-%*Dw(&MOMYy=(c#KJEr#B*PhffcV6qYgO*gnWs$Wb*VHYdz@m_S>HsevxFuq}8MhB>w6%PAzqm@BjTt9y__LCg&vz3SP*{Ka*Ds~q4_UaQ>1>`VY`dTh*L_9GJ&aC=Z0kz{p_ z$z&ztnw!9*K=o`|>fyVS8-J4WtMl zP|iI@Z1fR6mT17Pd0 zMqar?*rF;sZ$d9&T+L7YH&Nd^%_#td;9g^yM=-G}WFVdwEoAj*hT0LTl+2VEjSl%Y z&Uc-?8}tp1n#Psb!`0o2hhoYH6V575#CZ%;OlN}XLX7V(U8{BEkGrnW_w-t|s{L$* z-V7^9T_DyDDOWhjBD!Rb{OZte_{FD=?dw?R6;=jjz?EZ6vBsKL+zRyhk zA_KKwYc>y7`4{K)-_G4!r%c3kdb9xlnin!bk>yR!1B&dli zqZR}OCudBR zUKpum5VD#HT=)FInS9S&ZLRfg(bqI_2*}GT=ZqDolp+p6h;QF{j4fKVtz+s^LOIq9 zzai`&co&}h@|`#cg0~%UkdM+mgkroy3v)IU$bOgf8r^2H}@AdymZN5JGuq8N%FXI^qcgdyeFmg zb9&x?|MM49{d^pCeqUvHfbjZ=ukrSzgJy&x{TUMyq3-M~h9kCy$~5C94bMk>-72+b zUpAA@s^kShcGG(ddQ;&{r1$!nveelU-q_r_t;JK`X|Pa`gB^|EGnFum=4DcT(ly)R zw!Zl>%Slfut&xBUDHrpaX@Ct5lX&_(Us)NE&im5PY9gOc5{i?~Gq(z9N;G4p7h2p^ zj4?LjHE$ZY@YJ1qo*)5FZ`P8^$*BOB(Kl{iF=Kcjc}GzVux=EObcs>qci;^6SI&6Q z9Xsw?QNUFx@+?}>Fln!WW=KQDEeAa2kFSGep|hRWD$D??XSV_CU{?v9P|$K|s& zhT6!&%=j>yEG$TMCLoe@zod$ZtF}O(YB*qu44iaF0C=hyf`7G9hyt5Hm2k=EbP-arQ z-9hFd@!<2@($5d)a5Z>gZ;;H5i*=2HRXAaF#^`V`#DI0p^=ha; zy&+c-RZNHSMg6|Q#~zQ21pFnYLdk|)*f=Yr&@0S~y+_su;_;Qzt>y!huInoPP4{Dc zSW?L%81DB;BR;Sk=$H{_=08x~{c&hHks~Gdh9C+8lxi(QD}BGMR_3rF7IM1rD50Sn zNj3{nI*k}M{R*a-epI77KrimDAE#zJRY8(a;%a~(BXpi?z$!$Agol63&TbnKQ}Iq= z-LY@I~#j`CM8f`uNp?$K>(?#>arw4&~eWvl~>p;g` z;@)3m!F2D909bARYZx}!?PKCdB<#D%zZvQQhY(CN)7(`&9!9*>98*b`2xAcOb!L zec0bKvr6-5Qzv{-7VsOF^>gupG4Xji+Nkq{*2h@grCF6%(+}Zc6Ma1tEAc*WOwB}> zClpZ!H?2CheP6P-%f?Z|?N*BrF-M}*bxkCgQ^lS&ozGI--jr9S8L`}9kL^f4{jq!B zkC)VDG@r0MB!Y!u?t+4fs0wQ&ow?2mK?3)}FzRN?>aHHc1J$AsLdOwB()u(s!McJM z41TRtf)uX{9hF~f>YTCs+VLrkQeZdCvUO=GK zdeRo1y`KY&oKt=UU)0o2;%f!ILE08>f_2vsDzEv8)XJIbI}KJ=#)Con>jid+Eg=y| zm(y{v=v!I?3(Y6xL7$t_W;0-URP4pTnB)l5;!vgDycK#!uxmvlFCc>~2Dd{zvVePg z^+bK)>PLzWYrU8lO^yCg6&7%v~MRARI{&Jb6yaU?kvOv`@powf zJ%OK(zrKjMjc9!)QdD7TP~lvUe4`{6b0-%pQ(mdOKrO-9F<&Wq0zgCWxI~=>VUd+0(J0>56qQW<2~FafV`+u`ju^k41w9H_h!8)KqkPjp|^^Tg;{a`_^H^l=M9c%ZlI`XoLbRIyMW zl8?yo#4GE)-x?VWY{{N-LZ_5CNA09UGRZ)u@;8z3y|mgUiguSS6_Sg(8?(+xZa&lO z1Bi=gBl&=Yull_ZZxGV~54L+`302iI3e~X2MDqkFnfR0t1=JeTHnw}`F!QlDD#s3c*}PVv3lnpuWJ%)mVSKxbc^6z{`%Q*FSVfgv@93iZqm#b#2!7%WXmo_2SNf1%#7l=r9GbP;9)c5My0PJr|N?o#@LSNQw zhMaRLQaS;|R8=f@(}!EgU3Z_}V@4rf+&-eaNAWzxB!jJXJTgYWnJrM!Z+T^<`0Jwl zWB5Epi!;dT<@FiI4e1;M6c6%p(;VZA%LG34_3MZ_yI)VLkYnz>?PhSm(42haP{hpX z$hR}B&sy%e@4cHEO6q#b{bNNX5vlQGgyd<(G05Z-p$aXDC*##?9lA$X{eH1d+uIm**K^^|&OEcgi$Ln{~yQJ&VKh|+1 zKSaB9AAfG|hSeRb8rc0&4fM{Q9PYj)`P_Ci?wN>!(oD*iAW^fANal^WeIG1G_d!|R z6Gska_Xhn>qbm&PzpfPe&t-%c2=pHs$Ci<%aLcpmwLMFS&kHo$qLI(Pa>*_Fv)-P? zkRpgBhLl)&Q^YB_CT38O1!~mOKf>GuBtS9xG1%OqdTRIPYs13YJippwD35{Lmq#|E z50adnOV|={#~!gS_`6kiP-Hp*3W#PO9@F*>p>HKF1sTn3?CmfWq}k3%H3t9%eM%5s zsEJ$FAV3@#B5 zUw5hom^9^lP#nUj5wMC}Il(yQ)LhuzJwS3_Q;qhMxsG@11Mh3MwSh`zk;jmB?60vvrD~Ro?^s`@S+yxb%aC;lgO204|S_s z_-na==)E?d(<1Ul@hfsLFjBJMP4;}VPkbUjMwJS=Je&z*Royvz5TB_2(!Oh1-A%tC zVu_5Q@dAyQ9RUfQRBk-%^N&8(OS7q|aKUBI*?Td+NflbVa$=Q)!}?sNx3&%zF5uoD zpKgWZGpKCwzW<0=)HngpSh$_}+xg7`eBR4aIfKeaem+SXtJa<=k9O|rC66R}S{PVa z9Dpo8ze@j$d-HDwp1BjkMcu=wdXj+*^`}(CDh(7=i;5TZGkD3c10jVJphIG;dtfvY7 zze1gH2KM;>ctHNA`M1MD_;FUj`wIzagYQP9)y|^pxx_HhjcL5C%oYQEUkE8CKCy?> zNWOco2&JT=c!7riB$7$(EJq+Vr5jC5%q~_#d9Xebcyh=fgHMp%NG`VNT_b$XF2)<5 zB+)Nh3{1QEE8ZPga30IkBVgSKBq6$yE^aRy028A2bof==P|^0@^Y5G8E$wsS8&f@q;n{w;#~RJwMK?vG(;7)Vy*#2Ti$vsvY~{8lD>FMfI%@oW zhz95aWz6}Ddzii}(bIDY?!38X^nOK87sh@33yVx`!-N1ulUGoM3HpYEnR=EsgH`vD zmSzLn_1CQX8)P+_aYG~lFc!7*kzdbvSL7|1zgA5DD+2%L@ANPk|MGhl^m^tex$aHY zU9maL>gccI?wiw!8^Ykyo`7#=kJap(f1iQmmcW1h_22od#QggVS@inMVtMXOYj3Fy zMX5(z-Cr{^Gw%bct|4T<^5ye*9G8D{IX@*NaVtl=QPpaj+)AMy$LDtHK zwvN1N_r{yx{_)BDzMfxeNfWz#^YY)H@b+^x{$dKBeetRcln-J4`#b;lH!W^G>fU>l#~=`q!{J+{?0ACa zwOuXBY#$z1%jo;xv-<;4e9rzgCbIFqXWTnHxS+oN+K-!&6pDXm5B}%x^f3OhGVA^W!pB;WTQ?Auf3gPRy)4*GK8pB7&< z;eC05l9uWz*Ldq}@%@*#DCOjuW7Lu@qi){HFwEf;x`(r<(Y)b&xu^{i%2gLNVS8kMj=b6~eST~j zXj^m3AQHMT79oD#(z3D~A8K^Iq(?m>&0THvku4|~7@jfgaV)RlmqEN_pBo;2=S>R4DSJa^9vFCIpj6V-xj!WD@j| zNS6&zyy`^sKOOr&o&44Z88*qt7tWVhi%gt^)Esp^K3-8yj!#a!mZbu@ZRqjwZO?2y zzJw;$``6KdPKz{u+JLz=0iZ}j@3Z3r5kG-BKOE^^Cbn`g zq&tyh)L_WxbY`x13Jv{cB8fZOOC(7yN=drpM=UmH>MD6MQos7L(N0ZUeq&7C7+be; z=Wp5&e;H>qgl`7-t}~-X!}1Eh3EB*2m*@@4g*^9UjNPNf-!EoPu~A+k6*pHu?kmEo zI!&a1e{~;{|2#C8nq#ipO|v%)x#@3XV7CqIW&iG7#$O=?(=CPmhq3U0PWCfVB9*td zwX|ij1E>^K+(4nOe}BJR@PFY&|8a3xO#a0`7xQ0l;y;%8cX0EMW&W|uKRELb0{^>! z;UCNVW0`;A8GJPHkE8s@QU2p7|8WQZxPyP(!9VWc|A)Z($5H;{DF1Pk|2WG3=N;vK z8<__#khFKBsL`5A$Lv+-Uoolo9&fe{9@nZvc8`-;%0hMu^HZWYe<~csIE(|g52;P4u2ZsRA2jw+KOhP z?nKSx0(JCx(f7qTZ6gu0Ae#c+t zmlN*}^+@cUrTtPliod)|UP<%o+!Qx<5HL=wS3F;C3&fot`B}7q`m3_3xmDvFdAc~v z>To7|CnMT(j%vEL;UhhR-CPEOUJ`RH1W)3FSR%YM15W4nczco^>KDw4DrYAxI+^3y zDEjVkZbW^QPZU|H(^Tz~gv^+X>Or==#zg&Pmfo{$i0-20vn1^gr~E#Jj^0Iu3ra#Z=pTf1SUtIPzgw34S*^|E8Lz3QsP{P+cp6>d z6E(6^&w`k*jAO6f6rO_~8A#quB1jQ-Q?Iv$%cQGcaxqOlNh;s28tRi&1lC(kS7*7P z@Rg-Ep)8IqYVFCvkeLmhy8Sbd#mBc)XBZawTd!r53C~s>p>2il_KGmA&_!N{do96ZFs82(4$^QG z+y7+RBYC9kK9G0WW3{FU95~f*>LF5 z+7iF3V;xnCPMzuzv!yQ0x_e`lSLp?aLsar;dT;q%vb`I@$g}3 z*vh+G9BZh}K~vJbdBvhCGhj7PL8X!9sIL~Cyd{R-j6t9)r0TssM2tEQ;mhM}N9T+s zOIII+o|(Y(fKulsS|Ck22&5l6k6Z#nLl zc)sM*ioOt`_%Fyy5B@_w3x>@xic-L77%}Ds*EPRB>CB0}Nu(l8KBVQ8)RB-Zx#v_m zJ|Bwp84@34Y)c_W^Ue9<>P!1zWd^XuliY>TtzP6pyB8T$D%K4Zl%*oTTlb?x@1XX? zrWe`(pr&8C`>iuMUjn|@)wN5_KdTccq2u%&x<&Ygi&Q6Pz{b2lH-2LbLh3R|*eGFp z?IQ*J{HPI^$gKwk7l`|qQdJ&o^q}Zd@+v-bO+7PP0vzSZ8{30HMqU^p7aSJUqXpJY z94*k(ApNS(Q_W5ZH)B`$-f7_T-7{;*eX3_^pCeN028ZRLtxrY_w97Kh&)&XaNu@T* zRIYU4X3EOC94*juO^@)N`{r$T@0|wVEHrwIqqE(}>=HCH`bpE8NYW0Icj5gg%jxu( z^Wo`3-hO#dACIUHB5gd|p5zaqknYap5IMA<~&E(Hr-f5U=b z-?y>pv@V)c1&||A3U#BKXPD;MsN#jSpg+X&ZwWHL1@Ur=fB;Z|3**DH7gxAr;bE)7 zRpIz$J;yKWJn0%Alx6e>TmJ>sFDM&7X0y7|tDZJ#x+YO_jAou4M7H)H9Ep(M1Vs=+ z+>K8vZd-qMB)xR&Im_z+5PyP=0k+`KJg?QoK2E%{a^`+@e=qjJE}t3DN=%1MREW;@ zNq~&$^~Jj)$3jP_lC}o?N|kbKrUhJF7?4iv%k2NsYJYS;{4MF~l0Rw683jb$hW-+C zpSjXb7iu)E1u;M$!B7%rvmW@kU~`XGsnmiqIb3A&vt7+vn?>|nYT>w>&Dhg9oG^$U zZZqrS5`j6+RPBlJ(;v()U#3wp9L*iKO!e%m`!U$fqkOtQHPj=Hu3MgJ^tENt-2gh& zV;jH%@1q2%=(9b<1eVmT%F$o<3t5xo&#q-a& z&lbC?TEovif7!0{tsNC^u&r5WJX^ALIZc7oE(E(1ix4(E+ax?wF^RHFMD0B=0@ofPBRNqw$skiJ<(Xw9&ID+#7~cG5iX=HQlMvV zeQq_e%#m7!b+D4u;b-1RWXV&`WadAB=;&u*H26h8=CN#&KBA@_pe13dy=P44IPOqLYTOjF8ijUM+ z_Y5|B69$=N?5C^nIh%+ludYAZ1VeGBS2Ea`&Y`<&sO1JAX_dEUmwd)8^!yaM&N54C zHyg8>8QCk1!vtE-`p}g&o%Hs4Z4`p3UNTDZq~6)vAL>s^?C<4g@H-2}p#5-`S}$LD z>XGI?;A=U9+Ox2C<&zb?B?Xjx4|J<-hF=3!js0n zrAzoq3qS0Jk;S~h5q^?HGPvPvowKynBAKSX2v}H*pYpUDMe(k1%>e%rs^=+>ui5$T zO5+FSc>U&4;=E+KT34P^sTpzW5MKr<7hF-IqbzX$X9OWO+W?X;9aP|q?_0PAg+g)I zxn7#QBAd}NHv_%e-jTDteug8%!?S+jzN3k);#qI><-M)Sx$50dAD8Njef^dbS8&s# zDKw8@wP%A)78^I$4m2dv3CCuvpvT{ha;#@e_LL3u$o#Zc!h74R^3k|+$lfPzOEmuN zC%PBOpYjVrHPQAZo$s8`+Y)B9(G)UTk_j3P^#IF=>b*X11Mwe$z6>h#d_DP9%|s+E z)Ye9{*A=UxC`7gX%nLDxL8x zC(lyZ1eF~08zW1mkwlmEA->#yBDsHlMF{Z{N_Sz8re$;2@keD-tvvte^-wuys;GDw1(ffwDGcS(ysTT9& zmHU%2^`z)=F$|@iUk?3{Na*@-xgBA2(I4#%@ z*c$USi!U*yq8;Z;;@&(~6J*KQsu@|CHO3zFw20etv9=jF*oh~{u8`fyxEWGe+09d# zKk%~U9j!30_$;4ko51ptwvmhVWQFOqQkzoEX{T!GljVk!q>UaN^lSjiE3(NO?_D_U zR?ZDWS?RCI?!BcOJ6%xs6XcxQlB;!{96jqiJQ&nh*Xf)}<6#v6W;M*XVuM-JoaZX^ z%I&R%=X_vS=c;o#hp+mWwUe(DZzZ;!534GlMyPCfw=aBH76DJE^I>z!KF#mgC{>4c zfAx$mi4s0!p0FlT(K*moPrRZ!Dsb0YiadX(BcFQHuaPxxnJsZT5_m%rO>>+5`OLcU)5p$ zbE1DTY2yF2o(CVSa;iDjY#jzK=a=?FVAf8LX(X!b&aNyU5%)NVx#oke(o{J}FQ|^T zrwz1ERj(MCHJr_NA9IWB}7c%{$s$he|Bx42x z#-H>VdVLRRZXFq-C8WM;Z9kl~)KQh$TpUei5k;~ZsLkskKK!F3WWWL^Uu?o~q;KJ%tr9wv&eA@9yXd7{GEr z+iN!y_A^TiyXzedYI$FXX;i1#u)UTQv)&|mI6KB3f;nGkUy2Pb*>%C=k?_NJ3>y@E z^ZeL7+}Tm%a!+ZWz}TE)sHV(GQ{z&^R0SB!b4Os3y_`zTand!kV!6l<&{B~s3t3&uf%o@&O#AU~Q z+fe*-Hn+16WKOa&(QT8d934_o)R6DozEkCE$BMc}|EPuJb6Z;UrR@Rvn?kh*tMrYU zX#K!4YD3p*H9uJl14KRDZmcjIFvsyYw?3pcA0k4(QYV3gd(Hmwzg4I%cCe&I^}(5IXRCD`5O{@Eyu@O zZ(?bA>VBvTtQ4ee=oz)dXhrX}kVo)|>mj!?R4$*i--E=c>B6QTU)#2f5ViXR*j**h z57?;h3qA3LiqlJK3vTBXm8&u3ZaS!s%XQmRyv&B`MvJ8v-^dU|V5xvu&zM@zDSX7Eu#xOd$y~cM*&Q+P zFfc1TMfA|X^n@KsC%*OlbX}R}^uT3Ugu!c-&th; zdZqV0OjV$$;qY0YcTIY5YxI{$k-UPW!tt-e*xoXqFjl5s2Z{c@!v)y`mj$PTem_Dd zV!n!)%0kd&>Dh_r#CS1fT5*J?dnB!Ja~)u%;F!DgTR>yw+)EUIxm$urMQ&=sBh}U? z-HHzJ6j@$bQv|M7tpsb;_3wTiwqAaHIjAcXWOm1C&}qABI*m2WXL57K&ks_IPjh>8 zEaCcnw(EGLwH!w`Cfhe~7aQfBD9tsprg((?B(uoOTe}rz&s{SI=0b19*#0ycy1l2C>VWo376EKF;#T4hHbYdOZ^+0E9Uvw2tun}4v5=kqVB^^f< z&rsABFK+bgeMyvN5dz7&G2xLI#qWh&sE2yF-OKf5ALQYrhfa?0>Wp%Dpl551>WepF zn-HzW2JeoaK+8}ySRB$gUL2(bu5d%gQ3a?ppl`A^PBU;D+6>s(YHM3LZ9S}TNGMM) zhI1AEN!ncSpS$!2-X;P58*5V(OO9hal8mH-1OxmwTak+DPpm4sy0*Wp&K%7`Ld0s^ z2d=E(1GG^c5RkU)NLq_og8`V7j>U)e8_F`g=G??HB0CnDb5(ohRsj$+^zuxI89QZu zZJ+M43#a1cSJ6X^?^({zfNBL*fKip)PC2Rs*L+}GD|Jz+8PihsUZ13FLdu@>Gi;=y z(;bP$>nrUENaFLwy$N?l)g723?JXzz%`}It^_0t`Qo~jU4vS_fa-2hgJ8X&)(Y7Nt ziUQwrcY;TA%lJ252L{zN9d_V$ySZ2L%j*sfYe!Fdt}!oBvzx`z-UvxmJ_$$3Qn#^Ls@m#yPUW+{gA=TTbVL^a>5(LK|0xEc<<@ z7n7Pt7_sYl1?%}=JNaSrxp5WF*XTbL0oLEkPCM<(6Up-GbX%lNjyN{#MjB~{-I2;~ zT0i%IDx^{j$5*1`*XeYu-%sC{y~!^dbni~e=;_GQZ~lQd1tN9x2&47f6U}=VX1;h z!=*){Ze9m1Lvy()D0Zw4T!8J2c?)09)^n(7(`j>|-b{-EKV=%i_2#aexo_uk^6;Nd z1xG#Kk>8pp#T}!n*V^(n%K6R(i1PTh@)+8k+o7T`)*0))>NL*o+@dlf19B!^=Jf>h z$Dn&=H}AsrSsF3RMHcOal3?j{`?uNaWc46?z!rwdtEbkn(D2bM8=uy9!^n~3KQXVb zg~IoKXkhM2?Fx!`(mTEE(FIKR=+YfGb67T7$8l$#tyZ~tyL7gil1b{vL-`X~ctkUzfc{ULn3%s?t8X_C#WQts`v<%$L$0G1% zJCO$$PQpULNtq+9z><T|V+icL9I(~!-YFqtSd^NNkC-}?q@o-kIT_g>x)W>xe-m}5^Kso zol_|J=DI+l9d{D2)DZWi+@H7pENSbM!9)tTNleWfcf~g_yWV?gWK7XlMw!6KcZvDn ze!rDv1zTbqLGXpYdg3}>i4qeKqnJTGvZ{K=4K_Fp`*@VO2GEijuIpDQ)sn zvXv9L2N5>l@QzM)XuwQA_~;yZ!)9Pksk@zRcEBOYs8P9ZDw1@hKYU($yK>ZtSg0Zf zRTz5540EtoQf$hp_VOg|N=M)|2K)0@r@79#mV(f-^exr~%Yp*y0snQTgOq1sqLL#4 zp!e>h{VpT*avO}uHP<|2PmF3jgB#D1gHG%Db}c&mL@|0E*#>KyaxL7{Ig4#}P|UMp zys3Nrk?F~o9O99(&)?(q&>D}H-`N&9`o4mwH!@Om7If9S-oq(ED`&D~U3E{CH@p)hu&zK$9RHKJxA|Vo+>$!|C*}{HrcmHr%p?yd=~-Et=gD({!+;%Rv7)c-y13VSgXu8{ z+7+Meg=yaN`1*0VDTu*I?$uy@Jj+S0H5Lo87lT_4X3goosF zhshxGLeV2?1E1;{UE{SuMz89}-a1c2yEqg&tQk)_9=?0o%~w#M-o-Fuwr@Mig!bT6 z9lgBR#U60XP?xldElSC!H>Hf$9S7$NzT`srV;@FH^s$eu>NC9mvSm}og!1<(sEbbq zb9ZoAP*%jiaeV8Xq8O;qMb>UX=MtWWoB}`|ung0#-+nP`;uN~7JH>`G5PLM&US?B_ z1Z96sPn7zkde_YY_*ii*RO-#cyC-y>U5R}m!4c)_stVNSA+L(X#yKQ9HfxV1cYoBI zE9WXci`g5>^cQ2l=~_NYhY(#zzCIT6SVthnuX=-3-(I(3IwQ4%t1ro8+njYc4hqmY zxMQs?b#ghVbn8%3y(=0?%r5klNY!bBSotoR&VBuR#pyup{(e}_0mZh@1)BV?ucM}U zW@>AI^f+j$X&fp0ji>QaWdZwD3lWZJgZ;BF?%g$}k+g>DyFa+Yx#A`o%b zOJp-TsDr2Rs%NVkWD4%1D-(45uxrcDu5mOG2ddKuo{?F&cKFlTxJPXejc~J_v&+4 zL5L!nRN|sXx%H}iqVM78u8%Z4+fh6{Sjgf>5bEiCSDzv)IF65*!_)A>tl^>sX3Y81 zwdtChy1hKCSF%8=QaOgSNJ`X~ci~(l9XBh;M@#fq@!7&yqzip2Xw$tc2@V_5ddQmM zm_K80hE>POgQvfJl#l1igi9}TY0G1rx?Jv!gz8YihsOqlKYUS@7 zV)w?|h%%KTjb$d;4#3ALr5nwKd|FS-;qWksVTH?KP1KoGLU81;;9a$t?!7el`HP7P zE-#ifZS%_h)XNIw;3Rx2jh3B*shtVl8mY|KD#6aIxH0D)cm1&^ZITYsD&OC+)y9x# zq*0aJ%zooPpSBc|#hcf#|D}I7uY@|6G{>o|H8V4B zMn}e+W=vh@kSZ3Lp z_cQFB0~gOzWmk*-wELz?1R}es_xPnvvG-KI%Xze$PBN zMb|Z2u6KN|-#K(=KcB++-Bgv_&X{Nhm7JA-(^|!O#tUexv#HZ ztu+I(+j~Cn(jp?iHmO;C@*xI>q-7Vw^x?k69b|88DO7au$Dq-^;%k%D-8?D)S@KxO zBq5F|;VAG?f4lkY;xm_)B2U#iwoR zRvVytl~aA}X+nDDlzU$HP^y4j9G_;UtP911d{DM|pBu$ebH&pzF8_=B#o=@4K#36QUX>Y+hGhkkM5L*UD1# zWbd@*e9UgyVF2khe>Pgp;Z{kM=uF*&TL84f&8fEK@$K?o-;w2?cS?cn@W*|>0ulH7U zZ8YZGDxLLQDf_6ERKz-6eRNnwe8J8VD>HS~=JYcY>r!oM{b`L}z1OzQwT2*(A6lI) z-(QNap)TrGmlY7UJSH!o>3})IV&Q_oJJmlGdKEEOh+9VNU5DBw-BReDTjmtfNV&s( zW42v4)uYI@prgfjw(MP(ymG3QZQJQ~^5(o>X?fqGUIzzr7ybU!$^`*zFlrjfO7EAR zN@l#sYesv!v*%}Mu5$T^JU>IfTJVjXQRXocd*TFZOetvY^<06c9(TOiuorx&D)nVX z(I!yvvD5U0j8UwHg!XCCgXo~L) z3vUjtT&SVq-}XFp`q))CZ&a2>RdchWTgIt>;WvsdL@kM!F{p!kg)0n0xRO)&cL_4V z6)<>(PMynq|AE$ErFypTi#h3#1UN;85GHj|kkF^8YDWbFX$fwRi?w{%=s9JNIocQ( zyuymngsX_Vd6=Ya-^ueR+NxbFtg?fzeQhWZk2tp?Ybf1nU-ee&I;@=YGO#9-*v&L5 zC#>Bc|DtpLzL0vYeR1oW)u{`Z_9044&$ogJb>)fraEzkv8&KZ4g|1ZFQ1tSc1MEOw z>o?Qgj@Z*}<=+1dk&=Hvq>rua6l8W(-=hft)lg`kY%^4dk5*K7BpT^!+pBv{#}xX8 zW+*03{h*qMR=VP`BvLTPVHBJ&YkxoK#F13mBT}4q*nc+id&E zqzDVr)B!J2h$XY zIW?+@b*M_#AME5lw6EmX*Tg=!+ zG35J`0gEMy8!V+U&C%>uB%)Tk`aITO&)bsixxsEsE{juc(_+Bb+N`DkY6zaw$7lqf8 zhUy=NfXq_XdI2t*2+x^#$;m<`g$E5KO8P2gU5aBcuJWgW`Zc_;LwrHVz^jbCJvq$T zbWb~3+;~QiY&|cuYUwHmWmbi|zWEZRBGp2M$8jUA{dAX3YD`Ib1^N5hI`W{de&H=~ z@eOA6;ktum!r+6KOt@*j1vv70${cuawsL-vlf?`ypTlkEJ_A-2`NXUv$8vQqOGrbC zqro|MNDz<#=Ia6||LB@-oDLr|(CH_7y07$;N>0@5$81esJSBZ3&j1kgDr<7+I4t-7 z@rD>WEhy?$q^SZ_=MSf+tHMRay!|+3-!NKmFA`5lPX=tt>y4I|G{8d_vsHW+Mpo8G z!gbJv_=qi)NIL1B8g?%h+#mmPo-}xXW6)l~ScBJsU9_{gYY_-xYI{tLmbJ$Wtlo5kKpV zJUy8gJkS!R!)cd(@owRM>7%C^m06n(9G9#ycOs2{W$)|lL7OXW3m;q!^Qtgt^Y_JZ z`89p2-tk7#yrE$QR{pJ2++tPW=Ha45*Zu-?Zc|QZV9-75GGwNH4894|D;x#b5-X9= z^T6%vfylH;0W88HdnGsTYUe5S7j2;~>t%~**85MSg(`x(`7(H-P z7n)4S*eXeL4=)C)iT7g8Lpem*?}3}3GY2BIjFc7gw;&}oes$$Hcu;S>bUy_J{zprN z=d8brFZRc^a6CjRBbNGXZ+=yZD3HUD-w={pcm)aAC=FtOE@^Kgm~9t&b!+{zXKT?G&r+OCBzdRMR@eITg&qPD zX%aejzV@lAPTHiiO{chTuT}I#Exoe_`Bj`FPp#qRNaJoTI(=QC`y5wLf+!Dg6_C?% zS?7G7JX4tx%@%RG3Hju?SGtp<2|8MJ}`oA1+a3TTB!%YEs9#KpV^oXoxX*6)T=Ln~T~KS2I6c zv*m8@Hl5_o+0c0;q@mOy?sUBAQaV$i7E^T5C@xfR#Ua5Zb*9F}vR2b<(X3Y{a7!0> zHotlInI16$P#^)t#Erod9*_rm4$|;53=L8=ikz3^_2^`_S9Pmb{2%h(JF3aGX&1L) z0R@#NA|Ohy3P>+X6_5_0grd@-i&V@8u%7pP6fBuDNFJ=pOUY0NeZbhOM@e2)lU0SZ34*ybT4W z+(gRm0FUju18e_jHQtdCp(1pCJ0Gj58g(Nf`c)|H1N#r|o81d5-SKMJD~?4+?H;g0 zVaqk`Gb33t!H*pMN$Fpqz;w&uGRHxS4P3ynadHJmnXKAq8bvMjJMz}-mTLJ#dqH~H zmpo#G^`}N^SJm$V~pH6 zM$|}*F1@u-7x32KGQUzT@pOZ>n#_ICDx*V3cTnY-k#|Ew76hBxmdlVmKXi-?_T>z) zJzpUaS$BV0z@bpl0aNEJ;lMq7Ynh50i?MT@JZL$ zBWrx|<*`3(JyWbB!KMHNip97LJkkw=5z9DqxnD$VJ^P)YSe?H&5@{ZoyvQ~8s(AW` z3~8`JRecpp1kQa&qO5@`F+Q;nQ#yXgSyVj`Dq7pA6bb5H>G5Vg(YOqJ{brG?#Ip6) zENORPxPFDfw|=NQS50<+HuHJf6=EFRVF2WDo9iRCR+kLj(4csf5!6J798N3ue(LqI zJa>{#?iuHUikn(IR-94ffvMKo7BloojwpRdKOm`K_9bwiftxXzXEauxD30mf?DISY zB=uvM)QH^^-o{dfWj+}ye^OW}E@jes?UsDkTQ3uiD?-F3n1tfviEk0@K&$8Fc*#I= zma1nvqIIygUl-gD8O}Dv%_Iva+2gNSOtV{7D?G~S=QBNMh92$Xy9RVFGQ4`{u@$(k z(0WcWIKXE#ed>O==(zqmAg~Ck7q_|=AE7~*!WcB{t+A#is7W66L6$@g_Tt*TqgWC# zhj{SZ7U7{~e6s8O9pCl}i1pO81$d^WVR;}&kq5C7e9*PNgmThfHEG^{MBkO^BxN^q z(co}Sa&J?s7kRk7qS=JNCMx4y>Gf6pzcKi12JL;X(^|zGr)f=e$0mFXQR6smV3fIW zT&;Li`aFy8{-A3<$PU~Ms782f+`qdQ`~ohNnUA}npVvn`>$n0))%lZ=MOfn8lGc(L zGvBQl-PvyuQVy$~`2(fc=@zp19jVj`kkc&6rK;}dtb($AbD)moafX-c()?6AIoVee)POnaYFcf@)BAe z+~3lGiIGP6UAiOOCobTUKw@(x|LRCKjI0cybISm;=Wz{r-8JRr(X6dWd7gqAD6K!qu6=M*+fLCMNtkIfGka6!R;{;)sOrX2T)yL$;VJPW9gY6+H|ygQ&!)-x zNbW$>3Y`?Q?MBl+C8fsD5Qm0j+sy*4U8lnVR{CVkE{=8TRNtQkf%(Qg(jwfSzCHQZ zb{{kH%ka4yN~*4*`Ag@CvKQC6xg>rUJr~TqC#+wpagJ5s+eyk&+bLGT&>*srHp9K@ zB}=lASQ1(P?|@vxdI1Tf6_1MfgaDPcr1!f8sfjO|9-Jb(Qw(v=ES-mzW-kf9?uS3R z&9spmoh&%?TtdtN+vr@uP!y(-6*LWqn}Z08s1jmT%{$$Ka*hCF+PFD}CJaz?Mjetv zY%(`aN|pw(;1DFgwH2#f-Qf!QV(Tk|it$nxg^da$_tYj)9VY&xrJ~Dm=2!KbUxkJg zp&l1k#CC;l;|o z)|T)fjNAD?(*@9K)++`lkh6|k()+%AS8=o(tEC$8HRxDSzP<%D4@Z=h2Y_Nj3a&Jp z)X?puUc0fQ#F9bb+Fi;?Be!qWpL3qW{Ifu2)xCwNQKj(!0ZiJ#hmkxDu;9^&yl5`P zo$Hz(dJejJ>lj2lcF{V2%6o@lx&Xd^qZc-2kqCWRF87g2jH}AbTRFWBCk}y4h;%C^`Fp8^9I^`5IR z2c)RqLd;s7s%v3Nk<>{e(fV|+83vFLxKx0Im}S?E+01SYpv)T$;33i9QmWc~Z!=Kl z-uBiE-mlpI$|M-srf;R2;1$>8n<5#^yYV%Ihkq`dW{#(UvY@DL)&pH(kz8t?Aos|G zzUgn#`C|nCVz|M zno>a-DBmgruKV+fxSb>B)c0N<^6^6~y;SIYCE$@eYVJR-(mQ67M^yw%5i4^t+@C4@ z%Nu=wG(~ZZmEQXcSJ`sU3X)}KXL#p1EXi!FWvqE^F7rl4`<75&e5$YQ$e~5ROkCHR zwhS*^^f{a|j3s})5+E-PE4@s+?lKT4x#A@$E*Zl7jfS=EbOW znM6TEyAf<`jv_UfLwIkGhKzTqCZJ;5`CEkx)ds;sxTrt+M6!4XMtDm7>?HU`WioDP z+&CIAFu(2gL)~(ZwNr6*oKKk-A~;%Hl{{+HGwQqn>1-R#nV6ol5l?pSoWk3rSP}&F zzzhgY3wQ-DXK7~kj%n*5Y#$ZM6?b(A_Muy8yE2NhNe84qtiQgyT-Fs9eP%jAx?b_; z)9!;-PF7)MwuILq=zb8yLO#_cZ`dv4+9b*`IlDi^egBTeT}Fx|GjFx?Bjkm(CFKYk zCKHV2q~4$%^xjg0DR1S_m^E-q?{Q%hYbW6sVd`7N9gCG}bRk_ZY6=+kI4S#yg=B|g zlA0f7rGmmlKi|BlF}gtzWbuv!x1c}F3OO_Ijd<`{g|slyA9>zB2Cd^lGA+vrnYRuf zv2!fxlS6#XI>At(>uLKHsbeYA&EEoVK^X49+(%aCX)s1Pv2UUR<;vS4%fhl5The;GVkt~y(7)VRKjx)5`&>9^kUSk5YW+%q|QsF1#V17=!~YX ze=NVu#NKIT%DJ8j8f=&`-&1L8tK>RP4hDCxZSX}9*|*q@*@@i`Xa`_|ab#F&-Ozy} zdtw@el5oXabc#-yx${mFwg+$qhDsE-rL-hc%*NQEpM9{Anafd3miy#tRSmDIsZG=Y z$5f0=cIGZ>$K}1z%K4LJ%hfro$6krau5r=S99BlpMr6nyNtd_jkER!=dL+Kl?v&MK?2^ywy+k>Jl*Puqid1A-ZG-sFv?+pLx!d&4%i#a3BL)jc*;@1#g= zzvOovWhU)j0G)OCM$F~NZM)PFV|&QXN^*Ni0N39CEz!0#tqMPDGqzPO4zJ4eO&Iwf#v9Fqma{~7- zreFdrwb+H(4Mp@x-M5GJXRGwP&>k9)kI~lCnN0RmTe>35T(Gn^QdV0pQzR8AqKkyP zhqxEL>hgAAqc1#rzjDM0LB~f6N_&Xmf{I`Tt97-)hs3H!XrCW`2O=}UY;-d|_f@MI zpiM5^W}|=(QaZ+{L23mUUen0{It_Q}S8L{CGS&Dv-Kl#P2Yz@r*B-1wQ@qx#;+E%9 zlkG|iY<7u3Yf$Fn+8lY@F*n+Kr9HkY!4iV=t*)R-7k_UUNDAqoj2l2qM-(mysbT5% zcCZ0%i81}CU^7|FOf-r1Tw&R`MW^+M2WWUb1-%%(moXRVqmblPUA0}EIEoO*>qK5P zYTH2E@)m2uOvW~?=Mq}E&AET^05h3Os>?WoO(9A*1IgHXJH7P_17Fj)%vvZ<<92F# z%$}B_`&AlLj@R2;V2yh+{UMP$ignrSVs8vwXU{YVJX8ya!{x!Yt(1EyzWWQ79rT{7OVd{+-F%&f()>pq zMp91(Ipk9eQQSHNY))ipB-J}#-lFyqK$W6z{Y=xWQ>&0;K~2tk%uO$?)o312VqS_- zj#!sm8m=*yE?pl`lxx(WP|a}_;lvZS!Q2UR3mLn7E(IP>%wL|&L8g?NMwG6sLTRgy zeGWSwDhxulZu!PK{XHIa#_!$l&S>w0cPgGlMA=z0Mh>rj4>+^jRjya0dRyMAz!MSV z5Pv^T-?3`N^*{v^srGs)YitEJ2J5VR#r8K)s3Y?Se zdvaH3fRguuqv>c`Af2fhf&Ka=d@JOICb7WOV@|&cV&`r-s)t=&9{0YX z9-hBQ$`>IiI=08j_wbWtq=1=ZXL&GN?M{x?)fnlH0Mjl(YtP6e&`WxYn1Sg@|J{$r zzujKL9a%*%ly8jjtE4zX7O3#?Y zs@g1b0Nlo!UuZNl#)jIN9@-?+1jsOWw1;&PN`dwpw=>;`xF?{Rjqxrelbq^DK&;k$ zrYj1q{>fo8X4~7xbU-|R;?H#9gHgHCv<_NIUtpu|D3Q4uhgn|Vy%EJqXM&zk)k;Vk zK)oFux~9D^4Vjjkaqm?T;zNxKFcN<~L_#pNd(*vY$vub4yy7~VuMu-CVnNYcF1f#1 zdekjiobgzQX5qIJ-z_QYl*EN%HkGKGF0tUuDZ>)0rZS~`7)qX|00cct1HTVk$f$kWSZlhbfCU@)U1?6#RL?n!U5KnoGnfb%m`BJ z?yL;0>;01$YD+1~@70QZu{p;F?NM6E((cD+FNx;E7o#NhhWx`w=aOMKA*{=|ev2c? zYWW%7wmqdOAtPTnIO)2CmZE-8O3Qu+Cx7%CEweOvCHQ^{+g#@3y5U(!2Mp&6om5ir zU{R~e16a0gmFwxJ890vpgk}?j#;Z$a_A4IMrcbe@jm#93bE?#}{jP^|O-+LM78#7Z zT4v<2`zBZlynE|}tIzL?X24fM;0ULMNq4yeGx1C2~+dg8h zVr4QgCYBy&;ykKH#43R?nzI9DmvBx>x5l0Ark#{-JTl}=vt=D9MJK-apR;PAuxeDPLIgNOTMXb)|>AYfO)wa@$heeR2| z0~y@=mVLi4s&=ZQ`x_Zc5Q0G3Jt{Mv`nx2iDvX}aJo78Um(WZ4waA8LT*Em2XCf51 z+xgXSy?z0E-o)l?l1NrtI9f#19TmHwJ|HIk>$6OUKbh2IA;7#yWW|r|=O;yJ)`Re( zamZ^S7wcO0)X*uXUT*BxZ9Y5t&}D$Fbey%q$$E7fNaYo+G-k(Vwy~&GtpNwtE57NX z>L}np2iTgBr;kiT;x59IZ1)L#u7(PkfS4URS>fZ5Gx=g$EJ5HJ`jXqloT zG@NXv!VSb?qA3)KT(l7ruS(TpVuI#$T$?qn0zM{42jc^F6C`h#09^STxEtjW2|G}s zE*VNS*b<@)I}t42@m^Nsk+<@qY*G=#1VFm=gnJDIYPtM%^huU(REJn8K>dpyJ+ zhVidaUYsOWHQ<8RC5(0&!5+*dJ9!Y+Jm!u-Lw^U}8_)9n2}Fts2y^% zT=)sNNQ&RoAcH&iVC|_`C-_L3U6{_LJxH}h*TLh(cGSrruxjXdfj;l26RnD81<0OQ zV?{bHpkOun9!?l2nE6-{s>C{hfwpmFl?^1O^vCY|Yz|eLqfuMJVUQNbd68{TpTJ~~ zAz-qz_0Cdo+vJ^p?alu3>Cd`owg0aiy_^?@CN&y$G8l8lj3I4XnB!jERiL&2MkN=< zvN2e`$8&2>JdHMXyjFj2ad06&S4|Du-yJyPdNRulbgw^E3>uWB#GNGX>Nt+?Kr=T1 z7!KyH)>#!gktQjo75E7ve4^h3jRsIBnIczfAhngdthbAAyX+5Vxh%zId1D1F;>~?Q zyL%JB78ZXpiOF^{vJxd$&Gx5`w;y9)^@3K;>vFmXJl73sQ&KYo2~-s!ERiM1Gz|xdqNvzipYe(P`xB09I^$yNeJ~dyp>(W^PGDZ|`o>sK#r% zSS}54FJ2q5CXq$ljP0$xq8r8w$(`fz@FhGkGvWb~0$8Mgw&Xr%T`xsd-+(RVTcI82 z6L?9|8?A9cX}*&l#>Y2_>)wD~nR!4Z)%&Q*^SHt#(NKdZ`vX2dM7@y#b8TSTR%pzv z;$(W7JXKSor-3xK^=qP`VZDgKHBPyHnh2>+Y%n$1KPhzV*AXe^?B)BbG5|M3 z0fnH`W4<{Cl1dC+rEz{Orn+B-JMtd`QFYh4iP4L$_#8?&?mj3ROhGF=89cbA<0`Y+ zD?NDDou==+<}8!!K9i{S*zIbv+!n?yzj6sg!QB`w^&Nx|y96*?n>O-gxXeT#LQ-(7 z#v^$~KqJ4y2`KgP(l^Yh1Cp~0C!iYzlHFtE zP{*ikjid|{@4-!4`^PPdS*PNmc=0}@Z>e2&T|*-+@)aX^-e2KR1w7Y3C=Fj z4(X74Y}9r%OAGBRAy0Bq+5$volhNFflTr=HkKJFdj`1eUrR`TKI)s@hFD*44E{bc9 zN=G2r?9vt+%6D{s?((R_KRzDEmhpjwUS~c8mmG1 z6vwQ1pfS^_+;flVSl`!AEAmmjQfn};5MH?G*)dk2&~UEJccMOb_VW`LuuU?*M>IBT z6vIpP@?N&wZG(c3)8#6u6uL9F^Cz?weL3)nUwD<}T8)}ZKWrsO#-hSiZ?lU9V=% z@%K{s8%mc*L)e6^D!RkPww%E=?6xr5OBvLT$we=$2~!Ra9wIjQPi-Q{yB9*RMf-BJgxQ17a+LnUjCs?)gb z)}zN_IUU?ZcW_;lu_oho!;UHva*gve`keE1ZS?X{Y?|WT18?fA9N9%&Fvq5*6Mo6~zD~rbU4p&x8aF7g@#>q#!JBY%Rd+Mxrh+EW!>(-%a;})OPhv%ORGLwU!!K?kTQlHcG&f( zfUZmfh7PP0C<5ny^iDd5Sjngong!;8Wwse9c`>0p$f0yq#T+~cp&TOJcchZA`C8<#-6?DZm|e+) z!qp+4O%zTB=Q1Grkx50b^i)e8LZuqW%5qAT6CC||7#nW04>dnBfzmm>9Se7F;}+y) zCtsl50ahNk0BU;L&p_D{0aSZa{1BWEr_(~i94m)te$kL;M0LXxYm+R!ka7~P;@7gm zy7@<0i@n!`>}Kku;N{tJf*GMiG?d2xsKx_)xZM~t-Vw+!m7g|gBs8I?P=Az%h8 zs7gF%AiblyzMKc^C#k<15fehvf7aAAhX2w zPR~b`_gJaKR)&VlZA$`%y*5;qzw}}X`jdq^DjlYj&)b%EIyXUt`VyZF27iGd$8s}Y zXNgBY$`KqG>Ag8{ZxC2rv%Id7za?Ow9@HpihpE4oQ}x>EDA#$S7+@LFGyQVCFOhOU z#L{d=MqVK*1|INY2ziP21CL^q;mUjWs8er27TwTK37bs}p+SW>r$KIuT9?6aRgy!} zQQ!CRqWFibGQ?Hw4!O5up_nD)P7j(fTl~{Ygs)ZcgO!smi4SK7wFJYYcMI_# z&R9BFhHq{xOT{LiP+Wt+8JB~*#Gx#D@RH-hU8w_}j-^)1gf~%AFS7h*3&mU=3@ai9 zBrmVbWv}Nd$Kglw=C2Ka2fgeN7uypaVM8mn{G$1RnX5Jlpam~7@g=P@VjM^=Tpyj# z^B`j@xeZGjh~@q!YloccYa>V3h4BirubAfVpMhl_OXxt6g}3We1nKFFMG43aHf5#T zpLO>~fCT^sc2xm;i(MwbUKsE3X9Bgg6)QDKG2=2+32vbkQBp;zu3?`6hBV3DtHt1HEKqYov~T* z%9G&tbOq91QezETIjnpvuy{|yRCSl1z-A+p(n$(v)0WqKB@9@dhiC9+>b6Tvrb6O- zUt{(uK9T^L;?^k!-;(9(@sWY^YLDerzFU-0ie-@wyyF;M_BGc@g#BFs&WjF9eQ4Pt z{h6hbSNPcWeu$al$$)3MUxGT`HYOfkkY2et3ZX9R2O*wu+CZk8jt}Q4AX_VPt=Vln zWuZEybxpur3`u3>p6~j$7Qr>`H&Rb#dB9*cB6?UI1=ZFk*1qEvR&mVCBK>2Y%iX=r z#DRt%eL}llY_~U&Q;t%;i^iJM@RcfMRNK&^$D-?S35c4=5ZpI1H79S&nTXL{`G z=hIV`Hc91S%lrqncX36EN-TLBFhKgd5qU@-*PFRF0c^JjgShLv)f0t9f77YcIkqj? z%R;Jc!OPAhWlm7!B_jm7$`Y(46I$Wv@k84(3> z+hh3=G(J5gvGVRav4>eJwoPMRu&>1#?UJsB9B(%$yUKOO>khUb0V{OIxlp5gv&vvg zRa7u(+>qQQ4UC`_RY~3k^{KTL8AXRopn8~KvdC3y6@+2HmeR`voMh*#3o28 zFGzn{*~f;?<`{6nt+Bzc7Gb>kg>fYMYH}}rk^DWsGb{f`FZM`=BFQ@WJk8HQj#p{Z zX0oVyJg`o|<0lCbfQS!eGCRr`2(?Pa3E+s*=fA82Pv*GI*C~lG8s&Qt;D-t zFZns3vs!adw@^0ty63eoVR+uKoflXN@-zspsk`vyg$5C8gW7q&Kzi45cl?$Rc}9NF z;7?ZiC6LzfO<)gQ0!A(|98UQI;G<*9Dx*Gqk~bs(HyN)6S>#o zy}8z8f$@`3@_()9wZQp5IG9$stQ49hQr4FgFyl86N9K z1z9FLlvbG$V(eb>1CmCt-zIKn!oy!FGDhG^HbHzc=LK^9 z-;rD|@A;xla8G7*2APDZ( zv%|10Q@u*T&lJ_HXw%wws_r^eR1$1Hd-pA5D7SVg@@4Ar0Xcuw6Wg&OY1hLDaiD$BTYK;DQw@*|D71b&kJ$POjA0B$CQ)QnNMv*6G#YUqBpTc* z#hsN=!WK17WkK>+EAgUEV=h8*i!vbOD-`ZY4Txa*C(h>7;eNA+AQZIW(pAsa_`B85}L| zpt7%0I-4-9Ht4!g5(s{rW6CYvjok9UNsH)zzwO;Q^j)-Ig`1;J@M?1w<+ehO)yF6FnxT&e+Dt^n(^<;ia?@p)E z!+SXs(RiT`OkTP)9Whdh>EH=hZ6a3j_kIR?7wbWiw>AaNg(P zUS9L&wEaO(`usM{c?#g#UCsw=Z}GI9d9ZiRwH0N^i>6;oBMQZ88lleN4Bob8FPHeQ zieu_#W4M^wGzGPO5@yILcxIH#0xHrtuW{=#$xXcpciY`4 z5A?;W>%-@4uIg|k=M?0PrlZD7SLGxk#O&v#>s!7<_SSVupZ|QhCyQViQLN=>^L!D* z2BMO&)5Nk(r@tYev`CyI8{xUSqDfiM>f!U1r1krL?-kB-jl9n)`*RHRJ)AMj3lr?f zbg!4w{38?<`eD|DdLhr229+2W6sOGL720V&pLo+zf?3M9hsmqV)|H>0-P+px^qoVO z+-I22`~tS7PaSl#m~!+IgAdaSVpTU%&RIMBI5?KqBo}qzOGP*yd%4})lPJnf{oDm@O-*W2Ez;!ZE z1oHMbfX&OO@YR)u8a2OGIRS@awjwZ`Q|R zoi_^C+H}M^%xSL2_$kT^b?fZ+aD!_do@HNK*dmD~Gf5IP(oa25=j9MyE_QTmFHulN z-0r^+oHDN_r(ypF!?XC9P4oFp@>tE~qw^FcZ86-OmPs1A)j9k0hsh6udQ?j=-Oe#a z>1@=Wc3*-qxxE&7)vKmD7ROwaob3~>qfMoft^abRDh5?KMh>gLzh1uG3M3uAC z2p73Cj0|!r}j@;7v+7q=roa27(S$eEu;bNTg z(u|pv(CYA=smcg`&Q zp1%I`9NAnjTyU>kXT=h7|91>pnik~h^!n9(anCURrWE1VxAKa+&nU7vK8a&3xSv@1{@hXjQVBQou*h9VI^kWNz$Iww|?yDvA=$y z{5940P%d0-e?GeI(baLgp-TOOtvM6lJjAe4zG5lgo0bGaa%!A%sX%=-C311p2vTLA zS^7GlzDt{E;F+X))YIod_{qo^#3!z!=_tf{%n0$>*82)~d|oN7p5@dN@+w#KA?&dt z?z~4>J1QF1C;*zF+O2Sqalp-6NW>a8nAN&^s_({S+F;NndOgE!U2w6v(sjER{Y6jw zJ3l>c`Yg8Ya0M>xxtliga!|t!w>0jir50L$do;}&t;e=%uJsF&P9RK2aT^59vOOvZ zfs!hn!{$FuyELFMt^@v=Zuo7xC}5agdi=>)sZ}qFL{2z2@s4I|S?>F#N*M?1Z|_v{ z#wt?Z#zYjZ^VAtp;|1c(A8TLK7>{e)G*7p68GiS+baZ%bTIF7am3P=^L9|d*P>ASt zAFqPeA_Q1U2b8%Tm%#!0fz4te!Fo#Bi8UN}3Hrz|-T_^^s}_DM3zgqtCazBZ(^K9P zKWE}IfREq{ymJ$6+-BiP4hFsO#Pz_Hv!kJ!Gx)sZMC;ubT~)#07oiBVbi=ErQ?-}! zUDdzJ)AMN5RDuY8Hx)T1^HNlpmy0bw|ESccc!9JX&Ko&kz5t5XQoK(A7lkck`QPT- zm9$bnZZCe`TOC+;l|n}9VA6>_Ymq2gx#|IF94NiV?N;kg-~2h{7;&~-M}^z<^FFfC zg67c=3}?{CFOBUr-k}TPGHO%2+d0= zxMA^nKat3{vsjxxkmvZ6TY|xjYa}8RH}9Uk+3a%e39)JwJ+v|6#DT7!je|U8`>3Gd z_FeFi-WKV?SXsW|_~cD;I(pNa@Ar?YapSs;;hx3^l1?MvGpV9p%KLXJ0nfbP*W3v*-e1>(Rn$a0#(?O+B8Yqxo%$I$&c{f+$1fjbY- zlH0={1IW`Tyy)Dz{MYAy)$CrS&D*Y);)k}SQq)Y|XL@{&>BGZw+V-uttS`!*V^Y6- z!4SzPb+GBQ;0Pwm)!bvMN|QA7LX2UhFaRSUB#qQ83JN-f?yn^Ac>oYe2zTXk$Q53dQe zQCp**G|P~YMmu#DntU9REiQbKagQU0TU7a$y@i)mtBl zqyeHhe8^O6+B8B$DO?lMlMEZy7liKX85@_&vR58|Cco4`U?js8Abdxcw-vbs})(<6%EK?F0oL z^IYFum`jB__KUW*Zl=C$B)VS2lPuP}1u5(ZVB;eNz7&?xk0jBE$7Zmwm)+k*<0*cm zV!GvqJGO6ugdlV7#Abg#*AriJN|5OueOG-wbqP>L-^#IBJ@vmzRffa+eugZ+07|mO z)Hn`)+HANTg+GtM2epYm{1;>Ra7{p|gDRetw#@9x>k^ zrJR1zLUQ1o3Nef)_;NS&^^)N9($aXe4~#TgCx74_5odO?0eFRO$YgsAmZ}b`iBI|x zoy$Kx?s=(&6^y(f%A=Y2UX!9y-fnqFm6z?Zb2IJ9*Z%P0T$xD%>mUjrec2^D*)`G* zHO8bDEys8kK%B34e{~4qlqwRk*&AAG8e%VJED+7L(AA=;izN+i{p?jcF~i@_P|B6c zo(=mMHKe{)Jv$b9S8=zHFM-oHQZ^2y_#DmjYlhmSRkp%>Oo0KSek{}qtRrDDZ@qJm z@_Dx+V!BW#KjC?W7tiJPt|Gk(16Q_E|2z4h6rdv|d`j zu$BMm(vks32=jrwPb}#P0^ZfVTxloMnS5MYU{ECw(MM^Z6k$VY?~y#joEgOrw6nKc zMq@jFwlxWM4LSY-ln}kx3wGKArKCf?QvH|Minw0UF~)G{q{F!aYZfT4e0VK(E{Aot zM!>{pZ;zVYW5`e*-9qQ|Kxc>6`>47OJ#flmcoBc-R05nrh^)%MJ-X`=M~ z7aX0rk}3T=zGy1?w^;%v5oy7=JHwGtwVr)F@+t4Xp!;tO|BoNI&J}Ko8}6CPadTz^ zSWLVs=7ViGF3*g^Rq4@HtjYd#b6eI!hk#i{H(&EX>VD)HU zi2N&x{>L})cXIsnF6Hl+5v6q7P*R47HBNw}p7*py5BMPK@lSvA^Bbpei-FTxUj+U} z3skH?)_3|cHb3<;|G$s_e|tZVIsTw{j6kJVW6Yag6waH~#Oo+#xpb*hbNTgovj_>- z!EUY|jPEhV{~lon=Z7K|zd1c%D5N0%s?%Y4e-g`t5e3dDh z5seW;Sj=*XXNnvPIT@QU!mt7~%d5xao9~Of16zLBQ+!0T4=|^;eC$c5xq~9(6|cAd z)I2W!HsYWgThb1T*%O=H3bf{v)6_a{a+@G0L{Y7h>!?GM*H4*xtNHu)r<+oJBI#mo z_Bk}sb5lm--4(;vL@gu18pE(u?{ZHL;hAq#$APClu**I{{;!bppFQcX|LgGE?$wei zso#O6h_W%qM_@^9B~`H0;j?g&eKwKj$A`oh4jbZ3TB0&i`UfL~2@k41w}=ppPMo7m z@+y(V%>R$L$yD~lh2s}#Es~221F5KSAKk2uPl%#Eu?>M-=dTU}W4>y+6Pvst(VeH3 zml=Io638|G`Vs%F57*;=qnd2E8>5JG+C*Qb!~I&lS*$&;`yi8 ztI0T_)b%Y!DmSx!hM!sx!r{4Ght@7|Gy7`Y3&;zY^9O4R{n=f=961iXnAzOp#T;=! zYu1pGz8iShbO>)<-yB>yMw?hwVcv-#QvH%>*fc_78n~2dg39Fz2YX88363j1r11*< zr-pX#kD=YdOFMyTWEDknWUo{e?8#jrVr65I*!*>$qy3(e{sNm-UvUcQaT_lt@?%Dd z%I!F+IAboj5?$t4_Zuxu!=I$5Xa2S+7WPk1ric0W$t+~Cj?Hn>dlyhhO`H({vp=L{ z*Oc|7&Yaln0ECCr)Y&#dCSGegb)Q41d1H6IJBPR6ZYAe8p&Y`@#hI>?rxsSB{fqsd zfz4gNoIh}9T|?3~_WH=1JXocLB57IpWUVnDq|w$pS8+-nNV>k^u|ZD2A5;II`eSsI zNyH%q_kE2{YAj(Ts@jvJ>7P88@BTiyv)%1O<{ZN2S6!^nR-GCGH>FqbKQ#o7KZbC7 z^{b--QTEq+OU)*y0VzCE4*IXa_1}wc=l(#(V*!5Gcw$u=nPh{<|1)r`QJ-7fEV@2; z65b**%(`j+*_qw-JM%m6l3(u#HOhLU)u^6L_2Cp6`!wsb|LN~czxlgh%vs~%f@#}x zaFvD=PBWJ&HvP?a|MV5S{*68tWcf|wfHde~UoO$~|A<}x|HahyN@qLFizIfU9cNmf ze!M5=T^z;uGBKy76xo4q~)-1JlCo)|wgoR6(k;4Kl>a;1(JMVE8Hk#fU!ZZ#& zg^IP8YJ>l+K>Ke|+2K#NH@b^k|Ohxz?aaB-Mg9g(0+G9-6dPCYphhNhMb*gyTI^uLi~D#*Ce7s&sonx}|-^6ZIkaP3P4*0dfvzXBkx0v zh;>kw?qKc&gqj@I=jr~Lfqr-$_#4sO(&N@rfs#{r-RlXT;nh*blnF3F2OYB^W?6(=9s5IgWeKCzj3OwKki{crLJbnj4IiKHA+-!+`g1-?m~_qKLZL5Ynb9hNqYh z7eYeMVGg$wxlBq7B#(AYhY4#`?A1$!nkzp_9d};@E)6 zv1;OkBeM@*Mg|93fP|9?y;2>bjd0YwC__qW-P%X9{>5J=vz-wRH^HSiHrK=b>7WWQ zLu7_y7$Xb!K!Ljvh8v3U67vXnro?!C_h%r(%q-ZWd32Ar&1B${dW4-&Tfc4v+uQ34 z=7l$#D;&9XMNV#^EOqR^O;-!F2wKx-T4$}bX^QT!POK-kZAoerHt{xWFojy9c--^r{Se4)8EUyVK;Ark_2o%oD&;TH{hzyl7R{%uLVvV%zy% zAXh%{!9m zW@3fr&0O+jtZLVq>Y5#+N~f5@3OXC3b$lo;mTL{!16Ji9+RdPmO)F^S!+p!v8E&4B zz@kbPoW#+P&lRt&{z1OQg@=3|?YmyrU}6n=X+KJgY^ruANrdBV{59raW8Rpn3~SMK z)z4STy_Dee*)#|7EqvUjK`$Qzu!z0tXIt(9^h9kqF$mB~a_?@sthx+CoylUVi~OXJ z8j|kZJoRT^b%4+^Tl=c7A0bS2uP?_%KAq0BVWtB4tH>SD@H@*L8B6PkCR?XUScP_ zznx7{GhWExaM}#8{RX6oNZjMdU#DS+Dag(b>p+pj0=t?9k5y|38k ztM&9vENr5T$6`NJKjNrdQ;_}$n_WYr{W3%oqJi}}npH`UB-N?_jY7L;xR7E2Tg2y> zuF}xhiSx8Mo6t9CEpw(!S?>BQmW#fq#xaAJ*YP(};aSZIN1|+u8orP_KMxrL3tBQS zaJnRz0hqR433?qoy?W0$+R*);cr(7_ISvmFOF#$5koeBq^B!2_o4LEpcyI#SPe8B$ z4Lq;R(m=3>rab!DKn6-HNU#d}9TUWe9G){nG(m&Y$lq3Kwxwn*3b~xkb5-sMI&s>j zN#(AhGZa6BH9N}fQ1%%CK}ZHa9>~tjIg8xJmQH61Y$T}}=6J@x^QRolm;ILL9&S7% zRoKx8gf&+fH0tDNpGFyZS>y6c=}dQypA%zm8?>ZK6Y`$DXC%qmoP$xw2a=*L=TN-8 z-pyYCEgul=_}|rivG#cR{*r<3fs%KAFE)uc4c5-Qqn*|4R1SCr<3prOj5{@$8eKp% zb+^ZD*Ng!@qp{oHb^Ib}V@k7=Nx9NlfTNWjC{6q-!kV2|cep{-f0xWH@s#hU*J4J1 zTEJ#K`dp(sNQ5kZkR_WNJY-}1A1fXAzd8X}x94%Hx*9X(wW5=Lf1j ztZF5FA1zhZ)p*haK1%xZ+x3$qDH)otvau%9O~B4vRlCaiE=zt`1M47Vc=y9nVZ2*X z`A{#w!WW-M$a9S(N+Js@4M=uLzb^mYpu{Lf!(TF`p*S-O)or&s|L|9Jw{x+$`1maQ zYjbMdn1`4%75da?=fz(?`rmS#0ViuRhld->6e$`xM&KcLsshfd?20CS+V2u+in|C3 zk8;rnlB2gkU(pCwyur=vV}hIR7M%KGwX7uJP3cG7{r+F9>XdL}WXw$xNAkG0zLY^B z?yH*0d>}|L_S^BUW!uB%{H%(~_T%^_>%lmE_mN@#qSP`j76@ag5}vmu20LcV3tf|G zBmYZm@W`Ki8_;m`o{1W=HN5C+-R_a#QmfMpeq!UOw^kL;m={KIiM9U~)wb~>x}#!6 zk|{ac5KTRd`LXhMF>UA8mbQ1ijz7`dbvkB^x2`9z(;RjTsyaU^u?h{8LOfD*TTfU1 z?{CMozV>tG<`B zisz(1frk5HPf%DFU$Vjo#Rpu}$fr`ymMAV#@{uZyEWdx)+7}XQ`d^uMssdoiCDnVv z&VNAp)ij4Iq__Obifl)&5k?1lgBOBk~x2Dob6W zusaD34l&%V;TQ@soD_@;xk1n0g#smdM%6fn0OD@AOcMSOCYeEbw4THC$ap7730z<_ zAY3M?K(VVRbOX(rDtSZH+<;4`vEpv^63it z<}8OEQK6hEkqRUvNGZ_hNynry=ghnr1C3Zr19y%YK|z-hIYN>(Z=;h_*npq2AT_}K z-0S0u)ie`xnHbev=scHdmC!7=lYF>SWKqvgPv+ESo!MGCwp+4P{mDe@@>OTigdv3u zM84lrl2F5%fTN))<1B5lT20-)0S8H9xX*@m-Ta$_LH^FD| zxHh0!E#Sr{vOQn`|BJ3^M)4%_{!m!U}uv67n#Y|)J;dY6N zXfct%!noi|;ZTcj96O0ZMu0(Fm3Ip>ZVV-4(s8L|zHnG*=L|t!z)4+X zV~<&NjhCfY8d?o5R|2Y0Cs)X7CJ*}hA|1{ny#p^-EeV8=uPh5ShrII?6q^tSOp^%| z)gAo;D!6--ag|2aLqh?_8aip*ou8mkeqzG@D3GWvY70x^!V%o79!c8^% zFqTD6CE!fhJ}EOtswA+ER(q>Ycaa0u3IMYaV!81KMYa8qoV!v~ynArjFuTy10Pg`Co;Im?9 zGXUo<)jMS_HP|Vf+G`gC$jJ}m2FoSM3|;pJ4$CeN$Yj$f#e&ESne%+)J!O_x8s>nEOe*AT-s>rIP{P*=BON?!d)j+*&xNx%+49 zsQm61z&2FzdfWii@3yEltVFByo5ua;zQTKbn-S-Q)WY?XQwUcVok`MMStt-oskWCv zlpv{-BAk+3xBpCjRH}>n$l=mz6oBq5Y_5|S2F`V$K4BdqkO|x<^zpp?JnLw>P0oqP zFoJ5QcTAs|OyJN-k3h{!hobR9L1qBs;%)ksn;#%I^ZON1+M0C@LqK;D*4F@9zNk^q zZ4F{xaGcI>lha<@>BRiwxO$`lXD4wAng<#-xLp<^@-PHD&qHGKkCy7J%m0potoKFR zN0B!BfeyD06S~#G$~vWR)UcYkRc{9y4_2nw=A_o?r*4&PTFb%ls{d}sWiQv9S& zJT6lZW)nW4IouVow%|5@e4^;Vi@|iBgIKjhpyF(J)9}E)O5R_|EOI$4c2+?cZK<@> zFRtCnR^LHil{K?=sl6N1_B$)U%ly+6+c_^~?>18E<)B!kFckG}sBn!FUa)&gmpFUM zN6qMl*`{d+QpuK3Y}}Wef0&Nqu~=MVOiT~(&IkrHxLzUMtLLJfHwOy(wf2q90r`@; z28~QAV!!m9u7$@oanH7Y7Qb`gf?f6j!Wt_8@mZJKgMe#Up{?r@`~5^#z}$O*+u7X+ z)oy#jh=gicbB~xekW3Gj>Rh}kHE484LF|T^isC>bRkgQp;hl)kz=t&-cdY^XBKIPq zoh3`Q$OSI*C*d<4_ecWmhb?|7((x^llrYF7@ zn`lD9y7CU#^;Bga+N8qy=(&FIn*TV%>*tuA&hb5j`FMTV%brzJC^_p@NwUo@8jqE_%%`~qhE3_?|ZP)z5yx@ zJHg-(qmy32kM7i4@tN8oJ2vd3T0vli6qPAhe|Ci%o+y5woLEta%J)C{bWj=YAD%iL ze1C-ReTQV97&jL7}wxSUJp z5`*vGwnVz=$7x!Q?aoe0qyhWV5)ru2n(2@b#~BJ^{ENz}7LB8ug~RK%5gX4huqqi*G%3P3K>p)Haqm8`ruwb|&%|G_-_UZ6{6s zUgNveBt7JEyn&;x!>&p@lDyjVDtkhx0A?!E zzV+w&h`Vn>b3eSrcJF#%_#Qw14=ovoN!hJjEFPcHI|6$^4y#i-5EGF631_OFJY>AN5vC>G(>PicU-K9V0*BG>oljR@ z735HqufeNtMwGdPR6$Wl{i0oC-RdA^(NfJeo!<2KYFTlnm!bmJ*|Ea(f9A;mU!cK7 zxM2CUSQ2eniG<*8-HTu+C8BCa+W(nC#Qv8)|B`>PF%1Kkl%ZF)Iyxp1#)W-A+z&_(?BAtKfRliaeT~ELyhKJ;Y2y zj{6Tn~lX35oU7Nu+f%~o%C_GD>? ztc?+3ke<1B?O{r-r$0`Y`xMus!C1uWMB6AdKy($uqxhQjb8ZfhHtN>X{d~z6gZ*$$ zSf55?gn*gPskSM;p&GKcE@gC^`Efu*uj-tL?0WGDIiOyc=C#`-m(pKrdU#_T?A!MV zb*=;Lwg|ZJgM&rXS?-TRJt+!rRV|*ra-uyw1~Ljc78}}oJK6X9#X)j8wBGA7%e9Wz zIF+9hd{@F;92-~#PZH<=fKW4%*yqG_^L&IN zgxTjXpRv)4&tj`kQ$8gXTw)d^&A{f`(5N{G* zKe=A#KhT9lF$ym<#Oz7bI%mvLW-Q%?_Szpd+xvVf5tx>QZj9j&Ye`Uw{-{||@MHdsr+P0h1xLuhmlm?mQTF908tx_c* zMM)%$xp{uwbjT5Pd9DNGg|vQxb7$$;Ctb;^Xm>^gc?A8q( z$~-m=M#^&9t!r?RpWHM)+}{qTPt&6fX|xBMPlrEqX?5u+N?Jv9_wgQmZ5OC@vq|t3 z^dV^OYpQlxeI@Eol2zrhrqSwsSV_j?3yOH@Y&oxTFk5ce*;%ki9bA0=w_%y8;jkGy z@8=avD>g!^o5eyxM*0bM6hWPS3hinFdW{z5ow5>x)Qnf<^J9i%Q5fpDyE`nh8Mi{M z?1!_Jmdg6-OR&{Lh}XAe`&!v_GqI!6uZg6tn_*-(wQ5G>yxTnb5BFFYz1F=91(!WJdR^!q z?67F!&yF#v0=fybuvj`wb1H2Ie}5C~-p^f4goIRq@>V?&A0;09NI{KOZ-{7CoCxAV z)_hb$SSn7TMlq(XUM`!~pR$(VESi=tIBzwBy+GNfMvc=lPZ^!f5rik7rcs!W{4ute z2_!ue4ScV^cYj9fha}d5+;}FiA8<08sP~xn_W$(E*zH9OjNjw;-XxaXPGR*pi<|UF zhu^W` zbAbW5BPZ*;t?KlbSb_|`b{mN(_@^ng%6v?cRfoj<_6h!vP?q<^UG>h}Xz;%H)?a9o zKJ!iRgu7|#vVlX$55(UUPMkG{w31W$DOpV%bH`Y_SC;bNl%mc2`bYlH+99pE@b7K# zPD6=3y~dU8Fhx`njx%8$$*09ENrR*{oDo+>(93tBA?n-d1&QDGw03rlLy>2_1MZf4 zknMnhr7G@%D~f@5ak~$yCrkkg&Tsy}1KP$i+UW8sJOJ8)RW>2}0!Lsrci_$8mluf9 zjAGEi;RS(0L09~yHJgVN1$rtSH}?lNfw>BO<#(khEmk_sfoK|J^_3f+v|V+Xi->Ib zWu`s~Z*RyH>eV6}=ZjRq+m+HC71slGx6W96ciu;M4%@PNQ#~UH0utPFaqwT)M4;wS;N(_Rb~O(-kL^i;#rn|t zhyhiyok&9)Cl6gB$`6yEA|)sL63& zXUZ>}?zvu2FF2E3CChJ2N3V#lyB;=5d+TFj`{~t=>)VHYWN$ z)%BH{se_6@+E@C>5icj__w8hQ0(Z(Rc?s0#h0lxF8{TZTA|%40pV{tiwVaadE0{M$ zdNT#=vRtAD@E5CPt}6`yhtO(lG6>YemYw!xh(JTEQxg%F!ns;*(h;*8>f?gG@JLQl zROR7q4xhzLJalPfL8q*?E1C|2N%wm3_@sNqqU};}AA!5mDKna)UEU>tLr&h4nx^(V zj;+-74SCf8i=Mn~Win{`t6*4QytWfO)S9D&xIIY-~mfD zz&uCwWbtZ(kFmnM@lVZNNTeya5ofDO_{Xg=dB{S?DcGNjnrMf+(>$52L^ZEHo+JE) zZLxSM@o`5gMnNDOV}+G7%q1_cZUi&7qJ=o8u>{m;B;0CBs4cvlKK=;&m*gdjQ#3!y4k$_M zta*_D4sQ-_wXo>D2uS>F*$UHi6zo#bM_&O_U9X9~j#}HAG^}bzU$u#yIH=N@XeMhH z0CC>p(=^f#lsfEY(F7Nj`;v5%YwKI2jc+zI#8Op{;6z^Ypfcsib{r|gU}z9R2iZ(O zgcQ2M*s-!K;sMzlU@;ojb^xlHIPYpSlptIJZqVOwEuiLeY%|6~66f3%& zQcIR@qujX$G6+g;Olg@8$mVRe-bv;m>ybSzu(Y&1!L^W5xJ5rvK?B(p&6Elf0HWG@ znM6=##tBgH0S^@3Lj{Jq%*Dn}bf`gFrSx6>Z?k;rQGK>?(^_qOJcl55BXGUW^OPMG7fh#CN2ARhDSl!Vfv(b?LpDzP`zhbW#(J z+l&9y(s;Pe42*5FKvP`%*A#8{EL~b6_nu_-P82XE>@Oq^f=Y&@2qa$;;N9rTW8h+r z07uYMZq?y=jzf)k;cznc*AcnS=Z4SIoL4nVA_? zWE2fD{_rBEbgm<(@cO=?Z@6FY^Wool&1R2`%*oT#H7Az*!6gWZ=l`s{B?9B7hB8`U z<6Wa7RwqYXK%+xIcEh1-k~eAL#wYKOv2JHdCj$HN?n+GrsrI5=`m0Cq?&|oDd*&HL zBcKnCPp#<>rq|}Ox-qRQaV?=&Dm;kkHpC;xMpb*%qT3cUPr@>RT35 z2nBRO;&vOhs+o89?5rE9F4*E7Q5TIfGP(B`JRPppWNg-p-zd!UoLStgh}109t6QOKWZl6job zZ~isJL=WVFb)v+BA&aaO&Vad zIXG_)*<=+m4;TXcGgaZGK7)3xUmcA4M%4|8N08 zh^QYKL2^ZYcEHKM)GS5JzbVLJZ>C({MYU$%=GWUS(=PDuH7Y}zY_$%3{4;)#8IfB7 zch|%!4N~Ah;{W0T>jU@Bp850ht#;c(r=s9Pb`rIH7l&$Kofv(tal>3tD5_?cs6S&T ziP4>b7h{{(VF9tqo`YbCs#^7x{9MNh%%xYgXW06lKm5S9{0^S@P#PVFGh(2^Rr%hr zvxUWe^jprz%BKup47n2}?R@u-4}beJ1NX(tF#?%iGs%9~SWsJtcebZr*?D* zeWq&XlfIEtUGX&oQOs2)$$UPkZY_}P`hv{Vv(sqqPoQN4oV%1KG!-U31UTxU!OC=0dr>LTkyZPe$V%T<)~?)sltYiF{Q(=s zKHCn8d6NQf^}U}Lj-X(J&&dUM<%66bqBDp7?q|1d_G6+SFT%xJ?LjZcf5R;p*wItp zPA5DjY*RuwsG3FWwAdFhC!eTy<`YDL>uX=f3`a$mnR?50wRwf$TVIhgXZV8Sz+I}_ zn#^GWZOx^1Z+wmO*QLp8&`&o4bP$VTs?%%OqUnx_^{{*2$mN=~e0Mum8< zEmt?KkF1D=b0Aw0zYCf!`}mow=uLt*52!vu)kpsT0+Grg>oCUHX^D~w#+05VY;aVw zy#AYu7$@^hz#WQ9Dn;R@ztC0&Or&39VMe;`HD$ErBkE+iUGv4dx@%1XEzjaG)(dZZ zeTROh<%vBas?J@DsAa{z$@uMp8Et;3iyr`b?BTH3p+1Oe)M#Uz9J20tgT!ZB+*q#} zeoNz}FL_fm9{@|yHAxw*9-(rG_=W>6mh6(>>TkyLMtLKSaX9xQE`l}~M}>2Xrr6o% zBB+JD&pidvBZQCEtg6ec67P>0BoY^E5je-D_xY2Xp`qCY{sO$+f?fTEX{A9YH&o@> zaHsGUN#lJ3geHUXVf@3L(R{}`Vg~^UR?Q&Cbz1|SmERuSeRHN-OU!^a1{_lrj;gPQ zGaR>R8Ped8znj90n!$a7y5vFq_el&<@by3nd6uWOX~#YhHQsR__a$8VY`|_egvVUw z99&H-rddb_y=@moYDq*$nZK_Q#*lK9?TUuXd*n#ra8#}2ctgeqnja^-3`t7L7Z?-} z8-|m{EJ4#NH>~&)R#GFhhYgCeN=Y8^AWrX^+c9Lorn}GI0nLj}ZE)!Mx35-ibx!6D-O2)yv;PaD#5n?Ii_V4w9hP}I-+YIDs$K;t`MyaieJ_puQ$fRALW3; zh%m93SYsb?>pJ0zka8rG5PIguDFvHXbfYGHO%PnIz*Rn{iXoPnD=b#$B`e$rrgDsYYwELd89N_n^%76jw&a z^hXgaYS@g!VV9JHW<(;2PbDLs>)p4WR|Pbt#MgudGLIO04wI~aV!2x1U^4GKHav>t z5$ZkSV`s#OuKZZ97u5D!&mSY)+9qpW?-F7|TZz5AS@+>OtNS2zc?K1YojezjdG+Sn zD|mL-pBKk^xPpSKM9lNOMAy+Es}5~Ku>V_$2)Zrbg72=S(33zI8F9UdS`3;SM(T%y z@P+XRoyyMWL)9+W~BdMpTX8ANxJoEKawKI)tf7XB0XhB6Hm>!jTiW5r6l(5e)<_|yN2QO5MU_%b432q!=GyYUOAvYWajjfGFr znY#IdYqwcY7g4yB-~RP!0Zy__dPQ&32QZb?osCj}1kVfS+kl<<$_s&x=G~^oV>V~-AJH0hSd3M%Q(1vd}c-emZ;-Z;~#fhhq2fdt~-&j^q4N${0= z<<#`;Wz}O0`DVq-%B5fi4~Gz3TfRKlF@mF8HO=}(Qq@&@N=8%#4e*?okQH)cd<5fNgXDZP)eHP6i~Up&EsVDdWQHLL zHjB(11BrWRblx{2Zr@(}*>st7UVQIh82kK-U=-j3_izafzvO14wLGkgqE(7DwpN9= zpI(ZNh?p7kORb5U>Up>FG{;=hthAW)Z0}~2IVFgolbV_ucDy!i-B`aI9i7T_{hkFR z9mgz8%1=+`kHIbg1~&m5|HIo{(m zcQXsQeBY#MHNAokk1jT=pEcsMg$tRZu6H75a)eBt}(H!<|^I%!$g|WBuCPi z4P~=ccS7`mu;U9x(2{la*r;H!Z3rS*LHANr#b*LLwmt7NPLFgj`)>qMwX?KZ@JWDx z=Hh=@F`E1MI8it%QdaKUt^di0^pwnPax%p|>c0PGw+Y@AwWtA7{S6q$CboX)hSDk7 zu>i{kxpcz{vr?@KsFcEMY*Q0%dMI5I40!!yFPFhfpn<+d|dW!u__8 zwUJ0ChdcdxBtMO?Tzs5w_XJH;AIn;UiRA?PZ_l9q{n;# zcD&ZI*Qzd4-lN)KP_oc1ceA&^MR{_kssnkpOymxhGqGmw3YCC#9D0!9s6X)7k`xjaD-CUr55Y_xDgrlNn^+{ia`NaI=Q=RwtoW>&0*c&bm z=O;5O!lkVM`8Ct=b+#D>2{FxSe6{gKHG>F*BiVT*Rk-;hZypTOa5r!DV{do80?JR) zTV!?#dvvZQDdPuJ0=P%UW+KU6aM(PPUg;4@O{w1qg@GMm_<0R>*o55|u^&qQ7K+TD zL0&X%wK%e662PcH#{;Dn0Ua)I&inEHtb}@PTUxe?|7>W2dZ<@7YWmsiR-`qY|4PIo zg4-`gP;Q{EHL^98xa-p#qFesC3_xkJZ`c12xzw|3ltL?-)X~&O7b})XjM;f>yZ+ zoz&So^zb=;hs9Y#L5oHS;yjI%lILTH&HNmZgLGCA0<+CW3Ug9~TF&GM^)2VsC~8KO85m{GFuXPF27mXm2OV9!JOv^=w^>~nUy`{K z@;|JEzdqs22CJAih)RD;*V-R%|6AZi6T2;z#`cNO!t2hvmu%!nYwEV@?xbI-2}to2 zW&!6DU#Vd}KbahncA@xaqqO3d!=BiuOdP#ctjNe{)P=C!%yzydzVFX?)G+Gl60q5E zahRfEw4P7g;FVWGI;Seze0deVcvFhFHL7)#3LQ{5L9+-60eCx|>Z|d#B*dbQukX>c z)QWU&5Hu@lKHj@E$;r}zD<@spDL+{^^~|vrdB*$a2c$a0;WonifT6~|;NU6qpWH55 zT_7)TxHGeK)@B$_a6Lc#wBCE?jOq=Y`lzQPvl%G*g<~q6IHKuo!Spy zsRoI;!B1bQ>Ggyq0P6Dv6(F>1tez9>wMqW}X8~*iC||Cbv$NWBsK#hFsE~ydO)RM@ z4%k|Nzu<@5fwiJg2M_H7%*M#xa0|r3qCY5Q=7wZSUT!*d$#(rk`b3TQf5751`$QR6 za>5_zgE@hXpIyVhqMd%#e587PA=Q4(kS)*}?Xfhyj!LZa!Q4sU{S=wq>h?^3V3BUm z@P!clrutHQ_>8@&&oUO@{X3*Y{B2DElG^uHf?Gtr@uoCPgz(zwCUy&rgX%@xC3Wj> z*q-Zg@ziD=Y~Otq{d7DvZ1zt^{Y4f1XS7{*Zl7NyZL}DclzOksVv%c1nyv~q6WCDA zc}lN~=k&YRKUDlU&0B^O5{2PbEQz-}SX-Qia{tm5hDouAI7zaeL_Zh*&(IS)Ok>tx zuSU!R30)30XSjLIUlJj&MPIZ1G365Ml?wp)^DN=WmWW;~rFKgB1-n12LKqp5g7;FP z2d{P{MLeE5c14VkS`Y+oG(pR7j08~`waez`Q+TV)am#lU$zJI=tw20BBpBf}PO0i< zy^Y-Z4Zlg8oVuGs&yx3vGaN3TrTSC8Zl!a33{f$^Emexf#bA<@r4_vgM{vQxAGvYk z-VqY3m>rXS2Xo|10R*c#W^U++|cG@47O zK`Q%>zFxduq|rju-&hVonPa7So*IIFu&N6+R~4)66&y#IDatxlrpLUE4$umLi9Kw8 zHrHtL=77pMjp7D}_w2r4W|EeFw#gBzQ)N)l#MLA2C>$=;&s_`AUOJ~C$|!xm` zo+b@_6YagOs2S8n1*6+gU2574dX-^#&hZAttC2T0{*2+hQL0qc*#|Wv*rED)wfOra zXG5l0SV9SW*+p$7SYWJ-JiYhj5re%tF8au(ttz>jaWF+l5;ZNgga+pNd9F8Rkw5by zgper_G8vIAcQMOP^cU|wiLDOe(!V8k-cowBV%_TV@g`5I58dg_d)^4V=325c-;~HB zczs+QtB7ezwiN14j%7mm4>E>pCqr?|h3Der341&OA2}@B7t+J;1QubQeK>5G-r@bu zLyYzJ`_{I_RQFxr-ITy`8)R?$M(^l0a6xejr;aV9@x>g!TfdHL*u{?>#XY8=c7=;p zgJ;+Zr|H=In>SP+yMVgbT&epY=H!J>`rV8Km@WIF8Z0a-g4zxX{l{k;36q-chc=m+ z?WAY|H=Ko=)Wbgsa*q0^c~!8mJ)JhP-)r4Y@qH|>a>B8Z`O!>?NFmImcIdjUR@S(# z_E8#4pKJ$|27=N502?WTS>IUFBrD7RS9Vq1DKC^_Ttp)nQ48Ztp}_YtB7=qplr} zTaLmhBnyH;gF#Ev_xSBjpD?(%zBw4|jBA6Py&Bio#m2ZCWqgef6IP-@(y^o9;&L~l zq#Q>Up}cXmt!=(P0?vQw)8xp(OlF3JOR)qfxb)Hoq|`Gks*;Jl4I&a{hi6S^wVco6 zlL*?UExo`L9rzTT9ddUjvs9nay!TIYo$=m7k6g_f+uAmIWA6nq|6{)uJ1=S7g5{LgIZ^T#Ye8T- zzTm!>Pb+4~PiMRN5ja~wr0QcGi*|^!-e@fGt&a;BE4jy@lk6nKQ z)39WdSO)#J%z_~}Dm9zoJ@X}D;!*M18Gt@{$sIfV-P7uwn%omrapX~dDkfkts__g< z$u+mov40`iSFP30EHvN;1|-5iW$ZVW%U{{LU@ALKhMYuoDq$*0C zM6)!(kzwW#sce5i4Yt2h)AtF}raJGq#w&y1SyuB#nd|x-Dm}j(*!#vbD`u~Ua7Pw6G?P$kC*sG|Igy0hmQeXp3mUtnOXEU<=&d}tSK@i`tAF%gg$US zsV-Q1)MNW6Vb{@>8{5bUM+Ep~Tho@thL*+#3?f|z7|z=p|+ zWsdkvw~phT&9q~P*g8@K?lUOwC|Q4nEmseRP*-@T0G?(y>NOAIWK4M$G9jnvm`qNsgOe&2iEGYlQ>$dd=D#qtaiuJ zthRnAuN3LFbc^Nu#gIWB|JBW3q~1^17Eg=>&PoJ30X?mWWP2CSaLlmI&-^qSh0<>f z9V%x?nuohrPCY1OTg*a5|Egx^djrYMVAK5s^OmvO_mQ}mqzQn|xd+t)C7p6M>{F6s z|D^RA)r;tBQcBWGz4vN?Mu_6g-Lx73f3+;w;ZU}ojG;#;#zc$98Mv|z9xw^>uKfoj z3j`hSyvHG2PvHD%LEKwDbMV<_0@276U2ueVSHugG5Oa)YZsAuf4b6@Lt(1a1c0W_Y zhKU_oYF~aIOL%EQt7l=crH|@uL$*G&o@qi`E z;nGOT`opW3i$r{C7t-`YJ_|-CSPT|ocfyaya}>RK-($!)3^^LF6maU-qvQTAZrJ-A z67UHl%Oe>h^aELTb*20=0uH!tb9qwBy6sVQo>(aWye9+Ios>ZsaTY0u39D|SW0Hn@ zL6ymce~yCX@|8Io%K#t-a>oNFg#Bqn>a=X0?G8e^3zqG1@Oz1~?kG7;RQUkiQNsAV zt}$uJ-3PchEZ(6GDrJp{new5;9X06?-d=A)kcxi2161JP#K1F(wHFJ9Wcs1hfWlCt z{DYV9?NR+7u$|ug4he*n1l$+Sa%W`$_c)@?t*0tqA11HvGLP4;Htd3WM={v$R05-N z)iSAJm{PqzGK1T#Zq?mz-Me5NU-voZiC|ek`hKOaqr0rAi_y3X^Cj>YT{x-gy|NO9 z&9>1>*&>?jYB=72!(~qAR~%@)pp>m+vB3KiY$Cs@!Y3P8kNPJM_X1wLma1Yi_qKE8 zCI;e$v9{203497+J~_bbN07KgyDR1h6t(jA*kyu1J!8hDVeXOn6!8yT;@}bz$gg zjM*}hx&G${neaU%32(+1iSKYo{I0E!suL-ijmSFE7tI3D_u@|4V={=h*_q$P0e>AJ z+;1F^^_UD8;|Knvlpk2V4_57YJ%Mhfm*LDV-^;st=#WZEPJYoTc_%>RYr+`2<*UKLfy7TXa(_oGqJ&ebG;n1T#`rL@vPXEK6ULFvx{(U8mB( zx<_F2i=tseGGJ7)a|o@W#O@)DxJa=OD4dOl6$U{L|Fj^P0F2h8emL3eR8=GKm3qi( zFCzhT*ea-h<0?545GG1y`Ql{gz4_(eN@L~ZNo3PSPb%{~#6md;7dIiCzi_|_|8}I)qHjYPr zyZj^HZ`N;*oSb(=-(3%GG9UJN1iWZjB>xG4w-0^pzDzi)(zcxZDopma?Wl zw!z5)`@ZaJf(ZE0^QlWE^|&!2g1hd_R;>RR6 z9R^}al4R!l$&+0t#5$Ze=*+-C1XK#r%oWg#hEyn z`hIl>(0-Gn9X%}BXj+`3VH~+BsVr7Yk9rKomy*MTEK$@pPK`GT$-zgKgeHG%3j9S? zuV+ppGc}`IgLlIbCb39I;Bg#hw@hS1bKfq7XSV#?l1Ew>wpdk>*N9(k->Z${KNkkG;F*P)jmzcszsFRMMkn}rL` z_#Hi-zj!56wlH^hH^AR!biyi!E?)0D%-chU-(UWQJPJ1_@!-C^x>Ep$D*(NFE29j6yppH+<cLv_wv_ zoP@Ir_en`_%=9a5)rt^F7b+=H=c|dT_A}o7nU?TeJX^)A&=f1@f?(XLQM#lY&+{Lr zOh@GW4~$fWr;*PD(&4RZzwjWp|5H*b{*@FmaJI7-67O+2x3^gC&mgmCLifAVA!gZb zBvCnmhjPy_f07lnJob`BXei^+6SmoLpQqLVMCZX=t=l-2Pd~Xu0_nrwNBxTgFsomv zlN!Fw5WBvspR@*|*{yHl^bJB8o-Ii-Eqy07 zdtsw0E~+iX(*p-Ye7Z${(gv`_$rBBgb?VE)<}L&usTQQ}1Jd$Dd;3FS=DKWcxF?U{ zql{peY!Rl>pF>f-rQmgRK&(^G3k}#uS9^)WTbGR`d)~COUeuWt z6*XJ7dK(7jhOAx#xNZ?=`p|nc>$zFWKf+bZVZ4R;E2PvqidZZI{6E3K3AcA7iNiBaXU)3ug1=khGX+NX0o8yD7K>0zOh8ua+72S3A0WSREXY0A zs=3Ibq`q{O8Z{lxL)xG(bxS2^3Pp%*Q|5&O@~C$SxDy|-b&aM?%#Qp5Z`PH9+E3pD zj7AW^Xjp@OQcc!C|Eo0YPP-<_2=pe-LFo-u#v4tybH!u$Q-O10%Mm=ZO`|?~i?8a<&xddA5M(h3L7XQd+)7X}J+Xi6x z8mvtuBS`yR{$8#*^fshEHj2+o=Z0PHDX|H?z%MSn9Accu7j7^T_OrXRBB~7ZSY98P z8C0)>OB_*CNsdk5zw3xKgjG5xaa0N3XYdYkNsMLwa!T1R_2I5#@pT6w5erNli13yEQJ#y?n-)Y!4m?LnPbzJKZI?glA zUmG+6t_Pn%&UEJX>WVh3s&zVU%1HfPCG@kXUbnuf^dHN#8fU;LW+v77Ry4jYfT9rT zcIH!yb~)JuBOQD1(^=|ZK}P~v%rmOA(U zkF~dqinC4DMibHm3mzbZ;O_2DaB182U4(C~A z2BaU}N<_gGpFz0rJGc|x|G_4_zH*GPf=qH#JtF&#^^YfxhLl-cH8V_O;%aqzcJe>_ zH-XIc>7fi37&E3AD%`T#SsL}i>>-0XCzmpJIB0Dvyiw#LetGUT{S=YuTlR79Jv~(4 zkNVS!?*z0PTyu(5mdB{2bO-Zp;EFJQ=7Pmtakd^sLK+YG4tQveXNT+GP5dO+f}wye zU03GBRlbs2Mj+b^se7UeZsqu-d=!X59g`DKU3n8@ zo!{@BA;5nx_xe8R9qu+^D*EP=YN4%HrH`hd#kDq7#UNP?N+*c7Lz8@6Lv8LWKSsbF z5DD-%R;w$W%Y!BV%VH`<`Q2IRtjnwnQ*KIFO{YMixl}>)R;8Bx^Kf!>9q#o#%c~@L z%!^-{3;U)oA!ji~wvTm&yG*y6ak}XG%QojBV{_wHZ=@X7ARb^q1Bu{fS7T8KFPhGW z{^UgsqQyVeCPk+q@?t{oNDa1rMEv!fZ^X4A6zuy(O-a4_Q^t{$-sp?7gpw*l>W>K} zAU`A*ts@t`@++8mczqYbf0#<;m2IK?L(D+8tU#lV%Bu;_U;9R)7hHO7Y=D?Bl0!~1^A$A(6RWWOV}qx2kKV=ZH{ttv;#y&C+sSfkT9ug>GtOZ$ndgu; zBBio?4aVDMcSAbHbg*xVbo*wFQ~eY?zr}Rg?udToRv5W^Q(x+mmCX$pO#A?hF-@ig zCh4wKjW$a&Zn;g2#y}q*eaYzvK7BO&L@n+J=*Up!&w&n=7i%s3eDQGfd)FpF#HqdI zRpWxkS%6sdKy{>NWME?oN;w?er&=rp6PvZA%mY)`kG$Ff$M$J<+Pz+O* zCm|UvUeMj?wDPFWhRbd<+-$q--3-uqetvvp$UH!S@o7&vOwJN^UP8BPSPfdXuKS4_ zg*Vn;MTy=L*NB0@RB1Jv?aaS|u!dUH9FVFM?Cd40RyU)~B6=86xL4qjY`^THm5u5Z z(7sIh+FP>R^_T4ST8EFVU+Du`^S=!@sx3L+YjaFXBJVG&N6ER5V7Wb|GtFjSg?9!* zMu$VurGMc-gwK=Jo!}L(sE`)$FE_GmY)YD9O1j*V*MY9}#EUK$5hbRqW_h90{iK6* z!RL+&wZHKT2TM?Mm_$Zc8fI3(dhX$r+y)>`_GhMml8{uVJ>5>8R+5Kfk7q?p8`GWU zrLP_?iq@4dmgCA(=3kAd)qy@c&^R3hW0jTPw2CP}E|W&;YOsSYLVr z%5E^q4KS#*hCy~x!gHmtV!a1y4zVa+16|C9uGRz2EqRZ^zYfccv?3@%dhMPDH_{tC zF^mKV#xLL2eEHyb>>Xea-$L!Gs}t}Sqm;GfgyrpQFUI+$aWvi0xvcStuIVXh`Q5vu zo3^*C8F%Hbk9JE_53i2tAPrx~C z9gR27D9s78=@QLJ`mk^QE&D7%$n<_W#5L{^$*Ru1H43rDw|U!ecy4vpNa=U|Y!cU| zh2I&}}dJRN*uq{)A7?Tm(cz@ZBafqHZ~lvMiH=BY|N6)#x}XOSFgfvP1`$!J_O$zhHYJDt8j9v)f-jI_eSF$5g==*5&{PqtbjN$OVd{G())Q>vbjV}H zadrIKxX733S&i8BJ_G~Khz($k+JZH5lWqnRc~$3Xj*C6`?k`{|3yTkhJkgjOhJv8X zeu>J(rb|QO;k#54wVh*@wvK@AT=^EL@HW=!Q%OAgnJM@xD)Ci`St+l7mJcrkeG5L* zIVe}?O(Nra*T_);gyZE2Fp=BfvBA%U^}sNtn6m1!5^4tUE+S4;SxnKy zl&0Yg?M$HJb%Xh^^ZM>1ZqCCw-}*JTI(9-Y&t9ebg;bH-pvo0Xl<$?>as49r8u0h9 zuk-C1_mBINKHoc3{_*&zUU7$(4&7fgYpN%^sEp*QmDCV9OQvLn*{=waZUCiZXV3=4 z5w9vM6F#C%Hf+QM_WA%j^cW(1AH-rA>p#rM{2v(+v zF**1Cm9wAdnDz_kpZ|(9j_yx~+<^YBYOkopbOEh$d`^{7r_LqK_tyZ(>Ax^M^-A!J zs@`9ImV2M<=1!zBQ?nVc>XewN<~F(E<$-Wz_<5~@s@yb05nV<3;l0(2l-nt_RF3l` z)$MB;8BlrTN%8CLelZ-jc&W3ghRp^=!{h1DBti73Ey3cUw(Qy6yvfrPlDZaca!;Xm zhGB~iail@pJ-{Rf6Fk}bcp$+kyLF4w_{(N}K3Uedj4j|Fq8vM44lDNI;9PJg^R7cT zr)bJKAJEpWeEyAUv_}g?HCEiMz4d|=LS@gV-T2N77)1>gmryl&uxM&Y?bl&e9O&?1 z|8A?+M>6thPN_2StJ3#X{Zn1rskA`RDFMaa-Q9Jcvuq3y^DP9MTl}CMGXFZMR{i*f z`+4o0ej{2D=L;#%1TM3&ba2%m-a{2mUC5+;A@V|+u^Zg=iMuJ~=aW{Ta+%3vsVW=S z&|LW?w34}4o$IbwPizEAs_L3`6|L!P(R#DH=q)bh5r%4YENNfh0MNuRyM*>rkRL%s zjN^Zyn?4qRZa9|{zh_gNC$UYB`oxrOck^0gCUb}exU_wZj4E$3=HL4ycMRk$q6b*`0P))VLSLXjgML>A!viv&~?tY>~Q5Au*e$4;kG; z2i2D+BVa$x=y5(yaM`n*rCc|j<6+-!FyCpE0#<{e7wC1fSlpqa54tfFL$P!%O#bn~ zv|^(6!rLVn_-LE-FaA)SKciieA%|1NIaF z5)dpz*DXFMRVXP<{b1>)EoC7OUDV!B(cgHEA*)Fyn?RP*XgePD8DU!{PH}H)S_NvZ z+JZFx%T-{1<&qIDj*^Pwdf4(+hpav|NBtUfQA_y|g2WV3BC|tFRsECb;gqjeR0Zs7 zWN)CVbol9L93Z7;Q$-ds#no=HHg5rgD)%gZ|Bg|8ZWK?j5rmer^wG)N2IdnP)+O>~ zx7|Y4)M05WeN<|?!OS07TqeQ3+$fP4v(T_BAZz$6#;G zo)_(I)21LZ0ZZHklyBMGVQ=>!^FQS;y8Oeyss)Qj^3&bxk>&r>In*P8S|13f_=Ytc zl18E3cs+Xb_*O%!niusu5Fx}>za$5k8-G9b0nZw$E$pKmzr(b&$@8rlUg9r&1hWqK zSD`X@gN#5H00IVB<&kk)FDLe1@B*+LAA{j=!!ugZ}6LPYP-lSA!>M2%eh9 zs>~2_{N!duuSW>$i}5~AwDE9e-tY2# zY`tMABoWBO{W8^N;Lp;m@O};~HrgGhTpzg%f4ez7>N*%WsEsmi;Uw3^+i!6{*-#ut zg{70cu&vnN-!&E+Jz8`j@%+_qmud+{{pwcdRATGxyG}K^Ck{;(S~1gNbL+)0zFP;Q zUiXF4(+~tf(zuEA?L;HhZn(*+3=_%*7nX{EW(N9rw!F7FawX#^ zhs_IJYgunQ_dQB8nY~X|OI>f(ZrMnfSGv!!yV-6e<|=e!Zn+CQh`iOa#!{X@#S8tx zvFz>Pm><4HmW`3sr51s6xwe?sCPGh57~4p70+V`~dKrZGVzZ@eW5}wmz}t?VeRSTO zyq`Sxb&=&RxAyd4UM%La(OmXv?ETBUle*VL0{t%Y8iyNiR_98btEmVfla^dsRrLA3 zptt0A!lS)HA|!PLN8)YmmoO~|9cI^fM_I)NYiZmL;M10v-7PO@>shoi+exr|9nE|t zr7e2~h9?KdB_i4TYFe~19EgNEQ{&Z~bfxa-`bgL#SrA#jl>+7Da{MJ)`Rzvp%O;`IadlYKFR%%OxzlT6fz_ zZyUGIix&Y>9#%ZBD)GfVEHkH5z63OLLEv_OC6a5^*9U-aUckW`5@<_TBTx;cTQtCj zE!P}!kGmXOV6Lw|d@^W@@!`X~BiJ#v6-@AD{Thv4*Xi>Fs?=NBS89U{mh{#vDd7h_ zG*n~M8OLRla3r)z)kdy}5sj7%HlQY7INg^MJ{DZM(H=0UZ#O8fGG?WcnLFE})iHNx zz9#`L>KX%MD(mDEA;0_Dv?!S}HtQn)T{F8@gHn;a>T7R07vB*q1Ed3zFMb;E)L+Z5 zpexXmrMG#%EL#@3CyePQP~5{mr@zZTA@dMa^TyFJx3RRj)~+(#h4-qaA#c6_+AD3& z!A^bHTT%~-rO}kYX3`?%)#uo?jgW2S@?#LZmvuYvY_Y!R0hv+xw@Wl_34>P`OFM}@ z$nJG&ugA(@Zo#ltLRgt$aOmIoFif1j;$-b}IN@3?HhgT*`}l6h5tH6k(Nd+F{}r!l zQcai|-aG4Muyq@&qH|Yv8;;Wzdek?D7!?K7C2y4P$7+J4`AIL7r{AKa8O0}MlFnZFk;@U~XamS}!b=`M;!!MW0VXG0i1mEoc z(01VgmltZ~UE#8PZF@&+xRarq>Q!UsKN+;*b6;{~W$eUW85U!$3+LZ~9Kgyr;pVUC zB&WZAX%@JxPnGJpnm%Ua6(vNJ@RUgPC3h|0E!jATJcTi+N$cC-B#ZsPH_-DqiZsg) zI3bkRXAK?F)*6keT=shfVw4U9V2mR^TL(+IaM@&*!L$7U7$&$+zZ&;rf2c85w!*wt zBp?^8$RK+^g8{|gGV?V96dS&zT`P^$Vtp9T@XAqh@by$10}a_AN+U-fAZmCKt>twS zdB-c^)-Lu|6>U49_gX$@pg>GDtx@O{YxYrzj8_U!)QII%>7Yh~AMhdrEp9L!M9*2J zpJA16#66;3L3T%0iIrOQnN4=dv>R+$-16PQm5t$Y&E#uXdMW2M-{w z!MvHehTXK0;wRl6M7^Z4S5F^-s$4-`Tn-9gHfk2L-R^`S-u&|pJnDp7o#I3%by`&3 z0E%UQSKrF{cFUGO{X$QTdphso{BsOO2@a%S8`=RMw$6EzNFLutb9e{F-6FB~EJ;by z?(xxUO|LL*?6hff3u??L=RGhV-Pw2*ZEw*yy=g;q;Z33guTHH2fY|(5ar%u3pLmmx z0^jwiQs~=_E+6OOSY_8+wXBJ&bse>IIn+kBY-LFQ;3h?BiH90KvQrNzx7J6?Q_A8UmUX)ni8q8s8q zz=3laAl%NAs*3B1zBx8wQ2Vj_Q#3VNp#}GQs#fD{{`HojA7&?MZ;OsAOXaf)lU;m4 zb=F?DKu5-)4fPUk%Pa|c^13M{ed3Zdogi{}u?}9#192$*MLGg4R;$w&o(Hcg=-=En z*W#r~0H8R@F0;FvzEt&PR+KK*Ao*_Sfm12_gw{D+pxfRH9WRm34r(%AhxOoglUh}A z+7fQVWrusJSnG>Lhk~cf)XNU+DjC(?byd~mqJ7I_P*H3-LH@=aPHCKv2mAE*+&O1O zH=47OzB=>@(dM{28A&hnF&59bqbJ0}zN+4JTy6H2t2<4h*2~}=sc*y_xO_HBXl)U# zuP^t+9r8Hr(H0vJi`&~{a$8nWNivO7aXyY457gi}+#y9=H2e2VII6@Tc3@lF`x!rW zl!YkWTDNuKITRaB*gk(5PJ7g<4MxTqw`R4#QbRYP5N}rdrMy=4#hB|@@XlWGvOgNB zUujmA0|p}C3oWstS)q<*(~g+%kqxv`r2zyw6sNQW~R6=KL{Uuys@XSDFk5%3F zG~4YYX9sH4bFl5eCOf&%oFEC_E%WU4Yo~64Ww_J$5nqx?SE{kIHC9n%FT%E+E;xoM zw_;V*3hcVz>IiIuS!`zAm~RMfCG7z#hLXeaN{%p{Jrf_KbK;^wpmEtKI;A` zV$LsCYh{ScC?jq})sPP6jrc-hz|P+cx8|~ccpJIlv%cE?8m+4UjFufLB4sV);mrkS z7U*=o5u2(P?$;aDr7$>ewlX6BaQ3#U_R0p<-Q^ml@{rC9Nl5H{H&MGend~(2d|`wxNgpn2Td!eTb-6bcQM&PqOw~x; zw?fma&2Dil5NMfIl>@}5k2{6o`{Wi`aoc9do8^v{$KTxEYKahf=w}ot6`*~JGyitZ>GGn4Ga+gKRXbiwYVz`loJd(S>xkqDD2DJiTW5`lb{@& zbP5O_D#(O@$lo}Y_vV@T-#`f7e$R<4Z?siLchgxp!TW>y$p3lSj>u#RrYZxT{wT5_0MSm4Xpk&dis>R zD@_e*n(AKztkDeNgx;y)D()KrS7NGl5Y~^|p~yisnIuMUg=r~ffTSC`!N*VR*stF( z=gCz0=WDQ_qB=diE^=;i?=J6|I9jE{?1fp?&nUL(g#RReNXF^UYqJsmVc`=>F~~c?d9+57uTh+Y*G<+*bz`u?t}v5-H*>^~?(BM} zewbohxbij_DXq=^n25lNSGu^7D!Fet(JdC<3sOO4$Y7P(+C}{$xMRZt^#ebg3bTU`qFyt2WlfFMdlag&`gW)J>4 zor9n$z#YEha2`5BgL0_I-wL+KHyqP*Tt?UABj`S-=#H1^m_w!l@4=Uv4iI$7aS) z6}OV!JIsvRab277H_!3>)(c(~d(zdq*5^UD#QyKgbAx0UG5TV1#+TD_59b}w{V_Ca z%n^6}B8eezHwnK$r;cu7fPTn+H7c`Z@KL(`&aShzPRVl56XPYRy9L+QlimKJ!Qjsd zxV-juH2m%F9IcPy=>lD?3Pfik+{NJ|T z9HM+X2|&*iFUo^mukxNeZaz&$Qv{Kl`*2Y#gk@R06s-B0^}}{NM18k$JkP|t91J2V zohx3O2rGEFdE5h$*V+HTn6+(F#_j5=Hd`dba*|Db_z{cHpIJ8GzF3I{E9G6zu|tx( z_hJocE_eU-`=mULDK%s(2KSTvY;Pt#4M7-sY-R4QWMuQs=zt2@om81w{-|p$ zYmQ|nKL1w7nmcccdk25dpHn`j4}4D^9jvKoDR8TVS|-|XLj`Z> z>F1~ha5Bh8q5G~y9hRl;Te%464P5OV)*G|!7=lmBT|O9aq8wcZg%G5>PDENPMCqiHo0 zmt5U`7?|73BDwG019g&Ue`^*SS3$p^)4sHVmg&`)Hah5Auw#!89q8LpcD$^JhE7x0 zj{f(0t?xRRrPT~{Dqa=1wnQU`?WV|}hI5sYMATv`p=4Naou$xk6+_%A*zsG(NmS_EwNjU&7?gWgdiV^z+0W1O0l&wITZE;} z?MltPBx{&gs)M$o4o2I+#N7S~-|SRf7(?stJ;B)g7~J87jyAdt#o@QvOWWoWK7R+- zcC*=gj6;k>DmWEf{+ZdebTNEyra#7youpyC{Yim42Oc9aGn}wf<;V_gCk!w26{&u2=hF z!rKV13pM0^F4v|Rz#L}!ervN2A;8<{29H$omksH^{G3oO{lexWpg#yXR#S8_ry4|C z83?ltY}+!wLqj4|Rpbjsg^$o2`P@9_{ZR^uZ{fhQHtLF?a*s6hDc_&0%TJex5vf*$` zMQN1e@~Ec8_7vYX`#iITI?tCo-=CVb7pzFcX(Zcs)8LeaNFuzloW!g7WhE_f51{&n;+ z?`W~QCRiU;*}Y43yBTR%+o1YLP3SWGxNRe9Q@RKvCZJ<}0HCTgtK@Wchp6HH7|n9U z;ze;T6U=?E%k_sNS)M&s=WQkcI1=_zcZ>NNZ#4ZkDN zw+=8Q%WH58;oj1F#$S|| zcg$CHK2DL=fo$JPX};|61UL|3{QXIlo!hR8mi&mfAo6SzD0*`ZsNI(!TXu+OSZ|vt z^TL~tO|ak~-^P5tHZMlDTMCGp>GqY3{x|XL#bPHM)kmBj|FRY)5pzisVyx}t;NY0bYGsacjKvhlEpX2_t{f#nMbdko0A%_Hn<6kDYeBF2J(n# z>D>)%ePfpCFEZj^33ab$x6^7jkjQ*sPh7%CdNH9j$`>YY5PUov1!^A`NGI}6>NBbsbs}ue)A1<-Q6&=Oe~HY z(%zwQByB?R9Zh^Ztb_)f@Oh$PPwg2e-gNL+Vv@IscqaxIg)8+iL zOul3_m$Y>QtV|`}CojdyCm!Kke~I5r2OZ6i^i`F$O%+jJxU>D7nFeq*&gv#&R#4KjoPw<;S%9({rLEPq;uTXMThv?fR6DMXA_6 z5LkIWA_eKAJl^wV z)}Ss4mc1>OL0y~*O7`BOdJvvdc8W1~#5-^0Z2W+9q~Z9>_a(Z|sW~JH!|i8sn}M78 zF4se-sBii2zHm457&TPQ7^aaJ7{eB4{F-P=681t^^Ji(2#A;Mbzfi0Cf-U$pAHlzA zMj!dQpj>)|L~yM0o!-oe@g|WE9!#Icad(5n{!4Ga&=XRQRdpInfrJ*@SxKo;&8Zp# z#vQ{^;aeHWoQ6csP4rU{rbgSwE-3c$f~vF6(|t71@x!O+3X0`!#_o$+=}A z)H_<2pEQK`D80{B5ug?B?wn};z+6gtAQz88=A##D@KP8u>}-r*UB%8qgEGS2Vq+4` zx>~v)iLn>@85T8};GS9Wy5fc+oK_D%qaKCDV#8wP*my~Q)2S|S{5-!7Iur6yhn4m=bQ)g$&>eEJ`8&1y83GGm=QKx@ z{NANYNb7Ba z*`1Fl1yE7J&*&4Kp94am;P}=x;8MB3CiyrV-w5p8oVwE3EWbiV=Mfi(rE%75(EA!F zTZJjSRx!Wq%$J#EEvj%BvIf2p&lGjeRGfE77&ULf8t70%eu zM%_byXRw1EhQ~`Mj@7D|A;c~v#aBKQ$oRP6gy04T7Vq{+MLi{9TM{Qm3$R1V9g{Q# z%q3rp!UCbV+c_L7UvuZS29X@eO_Palvc2_L?-geP1gVaJESm~r5`$Xt3o?Qm=~Sli zMB|KwquEQ(O(MR(P;$&-v#`3D^Dfh2DcbA@JUgJ@d7v%!Q5@ues+P6 zWS(1AZTez9ELsIMjh$gTNH`37@^_^^6fzFd^JBxnctfgG@zpkxgOk;)DT5NtTG*c- z3>Sk>UXb0Wlxk6{*DXJvcrPUAR0_+RbYV-rWHHy<6DmbPhm*dD{;td`Qm3z=AP`?$ zrkNMZg?JC$h@MLpLy?xN=h({X;JI*Wx3$eY1`X@VZNE3MWapK!HsVpi01Zr*4FE;? z$KM!2yfV~ak8i$kPitR0f0&8+xsuXZB|y`Za<|28^7+Z*xiQiP!6S+^5GyKZ*IMwc zFdKs6LJzWim&f&@L-bZ83kXYP(-(oQj>(JlMXVf1PjEk9j?eXckdl+ZYj>x}x3Tm& zafh#%|Jg8!HC(cUL%_u8Vuq3H>zdi!W7u-e?p!jbjc{?*_e^{QL{)g^0(OKo0{bsy zHF_sbTEOg!WDZIU+1j@>1Pecn!b}-o1+v!_a-H|n+pdSq!XDizXxzf!0pqR_aj}j) zq5XNBY$S4QXqI2Pp^)|DwcN)t>AKKnn9nRWb)vNv7*`~rsyu*`c_ennbjXVFHrP{T zt0G+ZZ=IbTtE-6)W8^D=%R1l{?I0$OVotNh_>9isPg)- zelL^|OWJ?g)wM4;9#Rchj%?l=obs_j(SP>>(Ak`xs3Y#-Cd*^UohuvWyBDH*6cF`k z^i;bKbNploh2Hv*^daq2vfaur!CCOuTQ_#t&!h0sk$#;`*y}Ujo>S53QmxRkhmuii zA)~p+7awfVchDKG`bp0e2t0ITqaMItNv2roHyLRJ4^Y>c^%51$%X?C&#iWC}N;<8J?nsDK7kq!#?`A`% zBZu;Kki6BIgNEXEh3ChJNa%{{#Qu$8H%*&yvnhsI3i=&(GkAvmn_%q=WKsS!PGa^g?G!?9wd150SVP=C=)5Wq(#R^yCjwEICblAv2WfB9(Y=erz+st0=a zQuO=3ym%T0eaC`O&VEj0?t*223VuQS3pU#tBp^+#TvOEL#!uBd*fly2$ zlh8Y1PpqsBlD{}mb~XZzK@7al<$y`>o>vS@Hj%}WqG^4j#*zw%l*g3n+|%K{1@)ZU zGzMS@<1s)e--h`QOJlGoo(exe6>PjD8Vk);ipa9GH$EyD8MPLFJ_*&DQ8^OP~g1{7X0liFc)maAuo zDiv(ajb2O$wN||ndXW8Fd4n90T@yv!NIbZOGj@|R+xMK#$xVo#m|**QC8SSzq&v~A z%mu79ejaJxZB00UD#rluF)^b2wiUElzI8a6vuS0A4$3=j;g&haq=}79vwn<}qRWxS zE{m{ukmY}*X%1(ANDyLfmKAt?rb?yICGOA0b}C!!g{8jP;%xj-Dlauw<^A=cOv#Z2W^9B(8&S`??3+x0 zzw2>6)$_T&Tk`^x}-*3}|q;dFu9+0+PpITjo*$v?1jl%ulc8{D?-fIJ zgS$)BM6O4STj}3SN9cT^`t<>73b#Y(=a#eAk3eS9nMy^bNWOEmr0ph#s+wRJ9`_L4 z9t}$-4}@70Mo977Ii8Wg*H<1q;ZofDgna)^z<%#4L*NVa2zSt~ z=hXFBIK=H0Z@VWTnZVA+^BJ%KMeJXwD?ZFxWeWLPVw&MW?_Bdgp&SJwjpoYWkEmC` zA{EGG(1hS~N0tso_d5STe&_j21(nqmoEJwf6J_S1#_eM`Wui1Tdq!{-1aRV+%OYEg23>c{TpE3z6UjbEu+E zP(y*D@u*en^U-^QA4R1=oC#aQJy>zeg9TIyEr9!39m99|V|D(J; z!9JW1^RsABd?OSI#LWG1c^vGK%K`HTjQ=Tu!2-#sH}}V_q^qmv!`Hu|eHTa^v0bcA z=+F>*slQ0dA48=a2p9PENi^B@?BffP3}UfR`L5t;xdr@nb+d+i(fn!Bm>8-w36g3OPGX8hU16-^bQs@8foB!Jrq0mD8rGl<_rFpRs zAtvj&v=^f_hfD}1nygr*7(W|6`@w-rM?j7GfA~bp; z{+KTnd5{71c#oHx8uVl30uozt@e~_oh^fKSoCHkIT1`x0;dSP-cb^|>z5s@Pj{{Bn z1y!}}NEfXi?UWdFE&C#PllHyqzCJxT-R+IY-yNJk+HMWdUtG_6>@-ME2X2?;xH{SN zIiiUA#=u~uDo)=azRRY|>(TuK0F-FLLoEN7kh!=lIHp?|U#qi@ zCwD+9l&|K|j%6z$`B)%sJBE&)j3y^R3$5a2@MQEK?Jhy}Kdz_a`UgbFt{=X8_U=!e zD0O^5dduD3RB%h1)H0%->@$i8WV?Gi7l?0+<&~!!Eb0AVO&tJ8*G6 zY`nAHtUJ>3%CF?cl2+cQFNV%CrOfHtjf+5QTrgZcRKQedDyy=Nj3hz-->MpH*LOx$bm)pdpmq zGU^4EC3{oM($L|5IX{tA{TWd!ADxHrr@hXeu6Tif%ph2WeO?wrTCCi@w8O;yIbSC zgP-)2gTFkzhjhP+IPVMES>N5`?>>gv7b-%*Q7sz9jl!(WrOLKOPoHlN307N{5rtux zNbtUp0#mEBlEZ=Hvg>{DOc#CVnlHw}%N=uH&ulR0AKE<{P}&>Yz<}|r7_+6E_3p6o*`M2n znAFM?VlmV)wUr16U=(ZrEB7-_i{*+Wu!qN7LzuS|U41kW8ONZj8{gyQ^fDnTY4R5y zO277o>OweGJ2*oMG7jJNI1PC7o1{~W|4^2({+8tifc4sd1Lb-=KgIz*2e~9QPpi>@ zV@Ol(SrIIF??Ws9v#aRS^Lbh8PGS+WzcG_25S8^({M?+=U4YA7p4g#ncNw~eL8+UX z%*2D9%rZmwAvYEZ=jpTS|IzgL|AidS3E%JkFa=AofY{DAA|@y8+E&>P%;enRRH{_r%C;)M-?L+f_>+qs*AtpmjITz;%x+HX) z{n02LIc;ZZ>PD5Xa)Df#1=a4GLw3FJ$B4Iff5kIJJi8PAL)eG@8}>iO?||}Bo1N2! zUA1$|uX!#HO*G04tL>UC_H(0wNM4oxxys$A<(E#6bJEjtKfCO5uqj0}6p)j7T|(Kc zX2DK2##pgQl_R8aTA7xNkY1s+H9=F#a;#VB(07pFKZ+xL3=YkZ>l*A0N7ka<| zJgr+#W2sJS0@4RAsxs{)N@CX5#!{ZAtH$Gx{<1kzm^qcTm_Nm(s3J?|uSZIG)6kMf zwcVY#8p_l^N+z+DG?nzonL{KpM4~m|4?5Y`i2uWzO#b&x?h620_odeM-1gJACI>D= zrd1j>rw*lXi=}xug@=CN)^B#1lr-9Ov(rEPfxJ`XEjB^9IT(`yzpNmjVRE;B+`99V z9(e>XJU87?lFzmWMQT-x^n8u~uc{hO30SZWeZpf`u~M1+B?{IzkB3{VX5xkEl-o_! zLwUIbI)8gdyB}|C9ojXrMtYo9;BY?3YR{p~+ie!iSE2hP+@q3FRYV*@^ zOoFo!tdarKfJ!0%g))VrLS1Yl%Nsvnif+=B)tr!6D4Cim5pO8MD3Dz2SExiG69-s1 z9lp4l*6`j@+7rGX+f@JcI>CNi=@~V+=)}6;DJ*&tgFfHR+PA9Xqj{tIuSot|_}6@Y z?T#;x|IVxJ-5yBrKHD6`Ei1NepnUO?j-GOb_$wy%zv&2MkThzPbmw}+BuRY?ARO+G zualP@R!|-v&gn1i@jb?Xp#*{nWrGxC%k?{f7XbQi$nvLUlp%eHm`ZlPe?-1h9my%} zwZ>RfV4yduC;Ta5u8>V8oF6*tH7??~uB?ZQ%O^w*cBN%j$hD=E#9v9$9kAOOUCW`~ z@x{{3Oy;!A%=>|y3wX!yhNN$PNJN!9|tfnR=^iE~Q+ zvJFUFfkwn;^K!qxbl6)iV^=OwkY5bkKTd0l1nK}94e=}+C~-D>Q>aP!B_0t!qRDoW z7fl!EJ-tfH#usVKA~EPP`?4R4XuF#emoN3u5Bp9~Z~iL92N9QGEQUsYKs>pZHq!0> zdK}P3^gop=gg8sZdJkWHb>~54;B?34bp9~tVa%vpq!5sb$LKSr^S%Yr&&(|0dcIKa zVbbrXX;jF*V2z7GBS%02qw~!OFUv<@(tdo{h;k(yslAR0c*l zR-C-FC&q&Q;MrO!&WfBXzU5u1n~>&|)K#-!0Xa})16YKq6!8w}%9?C@@2c)L8C87f zZ*Wp0<${=W_@l0FCDAoi6A?kbgwPf@l)9tFadla>zV!AIPBU-+6I+GojPmyrXg-;BM#eWJ8nY{4^S=3U^-RxwQ|TxvS*=KRVQziI5kdP zCW(!LA8J-U3^+BeM;GK==^DgL3cqH?xoS);e2&Q zi|hq_uatsEdw7T9E*yi~-fqOyJB8-Y<1X*VzYV`p%=>-7AJc*0^~ELovRp-`ugR(x zXV;nx7?t1&I(8>?@6YwHSHe_GM;ObqGu< zV}$C({@Y!nCgbS5N4Uwh`fW^hlQ-(u${;-*x8z|Q&I0{yUu}?uC!J>3(P%l?R@-lPvCT&5C^y#lKIPj@J$~cSf_vIVX9kr8ge2+}3 zNi+0fW}#Tx1hLwM52mQSN~toP>$N>=tg7GugSU!xBJX8UiX-oDdJj_)R-(U4j~ruV z(-3sw@usDd_N(;uSQW{~B(l^?sW}hI0EQlW426!8j%lfqvWOfOR_x63Ta9te z05xpTPBf$KH>k;xcQ}cT`f6Wa&b~Z5lcJY9)TmAJpA!FYXL-VHSp-%j3u8vJQV{u6 z2_{ia<;0DAB+#^8i~I#kq6>QHqK^oMz6Nad619rJLZwpi7^)cA7J4^AWA`8$oo}1b zKvnKE44jv;4IaPKDnN~cY`xrD35-^kJ)U!HFs#s320b#(O6wDA+-_2f_NSw#wz9Xb zCW=x?wwlHJq&{sCb9NsBXeipEE*t+YKQUxn_BnpSsOw-pCwV3^F*xEw_$iJHxgU=W zo&p_|gm#`+u$h6#xK$~h*~0@5akv&LL+3dIhHTSUAd{u3^}FodpRkQ{Sv>{;8@NPaymq|$4WUn1dl~o;O48WybOfGL z$6Aq~IQg=rN29TJKg4X6YaLtCw_y*$iCZ2SVBtUsN!YXr$y`UhHq7KTPp&)@_?NAx zQl@waceIr!TC+-&=V9cWVQEb^!@qY!+MPd~m`b&Tme`Vj?IFaC5wck~?I2Eht?W9rH z`P2jWq#J2Lx|eRGq#LC{ zxF(}^rIvdCUiWj&JoDV|*PYp!{jlc|zdF{tjo9qJBc2+#u$KS3^;RZ{cm%+LAe zSUdH8=;*Qur#D=+$+19u^Zn1WnD4NeoyT|KiPrn)v3ZcEgEI!d=T>!v?ceSbVCTB` z(q!rwWc*@BK};pM^h%Rsz@$ZN|3?a5LeIr&E}yAF=IkvxvyS(}ie>W#^w0Eu1(f5s zPw{MO(S>+_d$5u=imm-T9Zx!cqa3lfOXTHI{4ZkR^Kdg2wncxicvMjk0Dn1E-jD=w z$-*1__T`K|YobEGAn^DSZMtr%=uE~soCXS_sm-evv$9Lo3DaE&j+joR9c^` z{Q`_ZgAgi)((gSGH-1kDA&Yi%TvQe>r9NL0K5r7Meq7C(b9|zCC2YzE>CC^8Y)~Ml z?JvysW^5G@16#NA*PUh8CL0`HHy*VW3~NpR6CU+T2gd_CQ+m`+2GuhOUybf<`Wv*p z(I@$%)fG_zx8tPGoJB}}N9DA?)WbF_)+WNw!LL`bckYLM43QfG>GyBf{EC=)^pp$a z60?`=OQkHQ3hP9u#_~K6`@h=|WE_rJs=TbOU1IytEDAU7mN74}~-# zcZ8L7F)xlbLcRfsVyjHAO-cC6i*Mt;y0o(I6BmniO+XaN@mPj{e1U{sL%1)6VY{I7 zAZ_yeZ_$bA?kX7))_spkgYq~FOq8G?M<{FcEE~=Gv@|eAD%5-I=K%(&SYL5mpCS9) zs50QLWKOsA!QBP!rNHMA$3SM00y#X)E=7@_04wIv{IL<{Q4BJOH9fajcN)`wcl9yt zLGMV9`6)2*Tlee5y8AQcB)8Vf*9NUJPEIR%529&X3rKdrxrctKH|Si$tnqN zKIF=*?hl4mSxzC=2oF48m%I#eHyEreQl8{I>v6h(D_f5)9)ev;hz(5PgzVY`bA1M5 zpXi3zpqp}dLlZ8!FVPxt@~hmot5kg?py`+=3LfGQX;Uq#sr)+E+o z_J3eW<3bip0=c(jkLa;X0`Y1G=NDVd$}E9IyoBh6J=Sct_IYXZ z9>VqKS(XC;tcLUpMrUjadYzJ-VB0zJ_*DhBB-+LUUwlK|+AqKux%<6RyJlef!y zHet~d<vz=AkLlC(BROXsYE_w9QRM`-JmH3lz6hw7k8O=KMR- zmEx9D26TOQ9>p;h@xw_?GYpL`Bh@3}xnq&r50nm-HSZtl_)JdE8}>+u&VrkQ;$#*{ z9<&6k>&t2?nG9NyJ`bmANFF>8|EQp4Se%L$UQ}9gJx<_f-^aGY@%#}-8?MSV!M38qlbA7ImeyZAU;P5r+3%?_8=<;pZb)JV{ghbwz%Wt2F_DlgbM-X*(=b2NqH>%7 zwC4#d;xKbI>hg^gV~HO^t({{NhGoD~x*-|R8jI_h3byin(xeXsS9C$*fh zKqbu6fb1zI&6`h*y4>n7<&(YksT9O7u^8nMrCtC(f3l^DO-U zA*GDSGvxPli!j8<+ACr99-wU|0ax`7AH*_cxM}c@b|rS-+Rq-XmII+5q=AyUm88^H zO@wE{BJBN&6G=`T)eWSoksg{A;N}T-=%+F-{p5T8AqGgFi}hUoQLx>Te0m@MK6wuD zJoW5~@WZ2g({7T0CPsDx(#3C`tm93j7Rtves zr4i;2zkw-{M*83G9Qj|}xgPWRC#a zm{lp;t>GrD_UaWsO~;04Vwy~t{h1GgtTe0@%g?eQIDN&NZZX%K>m^HMO|S>{znaV$ z%=Z^w4$Bfv-^vU%`vHD}Ply8F=6L7TZGTjj#8zKNSpfg@;&4pcV`Jh>&z%wl20Hmg ztGdzr2h`lfUwc`XBs8X%<8}R8L1B&Y*%(4XChs*W5c8zYd>HHb@l%3^D73XtopC{>tO} z0rN)57Qkd4I@QDoq-W-krU-tP(i(rfznom=F^rXvRv20mnf$T(vff?wRr|MDj(HZD zoJMazuzrA2dd@45w(rxA24q}6VYOf? zR3cM803eg9A=v5$wnlsiaq7NW%TzS4o3FCMVC|u-gssIg>1*!D5ukp2g z7HLK!V~?H`zk%%b+~eNVn95z$pb zjz;@=t=E_cq=AUYr9+8C9RrbJA`_ADE1RkgC2JPL8J-4DeLDD$i+F75PnmVe$hfv~ z#m}@!zHjG`&Y~`K_eM#7X-6PhygWxZMJ)!rItNy*0p)^F`{iH$`{Yu;lnF2H4l4Rs zwjYtMaxl3oa{ida$0gS`-hyPOJGD%^0r9Q!CZ6!57YJF24yqtz@9TmTG-ND}@NreJ zYq85>*gSs=@CRmBO|MP)KBkN(zuP1^g|~HT?OXhgcy(x=Nw&+Y)bAt1+?>>7qYRQD z;&2YC>!PA;Ab8mw%t%9koRQ{wHM;Y(3g3FPBD$Piz_A!M>0r`Dt3-!j5@UVuB)X~U z4%S*cw-|~L$Jb+H!6KGGK=&BZFkEO<5%)YC^p8pt3P_TtA>*`R{ye{$Ok2M?mzf!# zIUXd8v7M=u;e|cWtQ0U3%@*k!V%L1pSG5GX{6&UBG&{YlU2t?wEbkTMe$77gJNM-G zQD{KB6Z6R!?o)`&)2qwh-Uou3FdTh@rK~obL96z&&gN}(6}PXN4w}qEC=ap13JgLe zww>bAq25zlg<8W+bG)LVc&4l7S$_WG%Dm;o1$b#L?1dzP`^f!GIGdBhe7oKh{l=uP z(oT;9uHznbxQghZ5Xj0g2@}GNH+)7!-{kJX+T4lYY#m%rX#@HA5AP;>kvfm_*qEkz zIr6mUHFw))HjhQ;4Ap}rzw67lM3cifYVg`h8*)sfd!1sCZTBQ~8`Z(yYf_ z1-VcKy4d|0=UZaq_(o%O>vesmAEhMH8u zwPLlRo9U)u)0*^eJs~yGjxX`^PkC@HeK}tpo%P#OWr`D=Wc7ynLU(cT~_S^ry%KdL8>8 zIS}U=1vTV8Z;1U+cPaF=C^PyxUfIY2^4PI=P<0nYK>`wGPUZI?aXDHIovm1-SbA;r z8m&qVL4b*Z0=ZnfN#>Xty-qk!S>ZT>;SzL#xLqvG@0#2qI?WYRSYzIyGCh(7_euJ* zuhB@cQxV2=(~&7h3VE01pI=1>{-I>Cwj7Hlt!oa|4;uE#Mk@BZwaMaT-zC&AHQL*J zuP=AgjS?CAA|?^x7R@WaGg&}4c9;xWBTd%4?E{^oc-NzaQmWj|K=B%Qm>;vz5r(1>Z_~oT=?KSB=#-&$-H=^=4zsGD?T08D|{5s>r=LF zR{~opo@8!ug!8*AAWtjSjuhAIM9*aI#J=O1l{uL#{)wNg|2E_CQmbEo)l<7`v%m2o zzWmIiuF5%_Pf`;!?+$?UWPHc~3(e=dEW=)(N+)Y-h4v~!lTcIi{Y@$ZzBy|PGfN%z z33DG8&zE7k#a`lVLUjDK1fw#}Y(sk&tEWSs&?Q1W_#0H9!TCgCk7$jvuqF~b_UW9e z>(LDb7`AK%X8;*ZKi^N#{X8@@IZ5^hwxASZlgzN&;B3qesp$?5unEH|G$rW}6INWb9^IwDLi z#C{?zgEBN7*9>om1e9nnmdMXm7GJ(SxV>UTrO4|Fub3lg9)Z18x%%R!dxo7q;vr>Y z_S3a%l95dPsjsr*nQlpstD}!<$&B63VRE9k#9FrO#<{wRaBVLP%Ow)SG?#!gLEOqxtbR(8W-JHs0`tRxEEn=sQ<;^5Vi~X0Np@MJ$ znR%_@DxenzuWBlfSw@mx!hdkB63M_XoC_%OcL8R*8|lh8B>`^_>{4v%yVxR!0v;qK@G*F#Lj;8~bg zYW-@{J~yE{sr^Z;K_;DSFbPCuE@$$2M+Ef; zyP+49ia}>eHFIPGvd_L0FT){twg-$#n%Z8ysrT*h6cU@4nZr4cT=iwvyO@%cdx$^_*LbFgvy3>EF2=4kW8nPh+1p)y3g`LOPOibd$CI{0z!tNH~g5EvdQ!$?Ko;q#KkdS>Jc= zo-Up^fYA78wtQS>-^{(nWG*2QaMfe|WBUG(&hy@Zu{7)=WmEksmit8hhX(~E@ZywbcN|yCG*u4G; z8kHA{Dk)g&kIhWVwM$y*hV;u${^z-iYHR_x?Wl{mSfEUwFOANLd5h*Lv%Y#-(2sB* zTGejtkWZR2-AEA2LvQxdNB?PtHGat!M+B44>w(ex&xis8Hrdghjuw8+Ef@NfK5yP< zy~Fnd$*Iw36bUm9@NiBPDP*yyW%|0-{`oMpd^INgc2YQ(wT5Lr56`QP;{ zdWo&9gc#J1M7TF%SS3UEnun2;Uyw#$)!!LSFDHgfc$3dMxh*SBWvujx*T4gIp>j6k z2eT^s8xg>zZ_=b5QsZz?lQ?bP`EdfjkifKFWT%mZ0a89u-J{BD@~a zAO}3%Kc4YXr>u+#woVY2md$M$J2zLDjajbvs-5sn=_A7$YzzoCp>JD?LC%4lsTyYn z`a9KAMZj7fwPhgB=h2>a(=oe0%-IhYzqAjj^z~P#&U}Oou2>>q(NQurNT)D_y-LO#-V4>{d9N=V_?JKHGD{0&e|v}l zN0VqSg_daQ#MRKx>)i7_cG4CFtX`}K(pl)O#))snk3U0*utX1*{4k>_>h?+~Mkhg% zRGVUzT!LyJd%R%;8!FxMAoi4I^cXZ^z>>NquHWKZrFTn-uBFjSg5g5!g9|~NXx8JN z3E3i+3@C;c8gxMC5m>_zF^}bYgC9O~Wl3k8PPR3CuJ$^P7z@>&y*Wa?hCG2&-*tN$ zCcW`A6cTF5#M+_%k|$b@~Yfg8SL7HY@x5=3(^)OH)q#%27VTm+IA*q}b7E$2(wkzPSgI26jsBULz+o zIMX&DYbw~Lue&c6aMYS=n}B!c7k<-)`rx#s2OcBg@a}#Q54}OHK1LYB@CiN@i|tu) zRxNx#euZboYLz>J3fsRCP+SMs_9U7GbYP<1FU6N8zh4jh!ot!gKUx`^lMVUF*&sj@%ZdI=6G#3RP26@B0`{z@ zh&CFxAwXNVo^6+}>8_r?z8GbhmCz2J*{V@SsJ3u!Nf{{abx@dmukD1>IQ%VwhE^$} z-&3c0SGQp68g){|x@$SaXm5El*j`5t_ld97a0RNEPlX`ImlCfj_b!N->x*EUAcOH` zVA!~Ir9zzytEV7E^|FsLQ!%8A!fGbm4UCVJAohB!D~l)P#fnlPTAoxIhfKii?AOXi zRu~iZ!x6sUVr3q`c-de3scXe{Q4nlO`~g}Si?H|I26r{Dh@1|&BbDbb=c=m3vK`pV zT94m!VqH8ED=7Eq&ELzC$-T5CgTk&es*r;0Pazo}cU5jENy&Ua#?5RG28+zEMDxRS z*-fcVV1=ZN4!S;jj;`heA1yA;`Rv$bT{fnevZq?H{Hvu^&Ii^uwGrN z9C9oTO?z4-0i~ufEWKs>EQ~~#XGq9K9^6Y=q@NPo8kmXLV5$-2KrV-j6+8_mEffef zELEbmHs4=V5$n7BH^L4zZl>->3WFmLQNtElYJOzX7dbfqR3-C z_NON9j;wXYizmfWT1$D>Le5p?t&rEqwzoAq4@~s{}o+Q zITqX?6Nig0m|#RtGJtf}(IKg2J0x%4aUD(_i<3XSB+-IaDndb2Vd&*>oVy9OALrlD zT)pOf0w20`+ef|<%+?CdM#~wKUb83AaxvmxIo?(_3~Zw^lGO;EjPdO?rTJ?Mgxgw%O&`dy z`ZvIM0qpU1e6XmQctK#2lef&HCd~UZReC{RJIPQ>?NLyF$D%+zBiBSrdTx7p$48Dm zZ^?Yhe!wcRGD19~IbOMGSKSar>^tO=rnBziDn+nSjOW>jzJGq*rG$+OfjsqD_?mtd zV(kF3LitA6v6N{{K^=_>2?Ft-yX@09{<8^vuT$gzn!w@TCh*pOxHl*8fp4_m!4UI| zlTJD%+w@J)408Nmr<_?V%55TNb4$T8=j*oNQ1UW*E%gs9Kx9U|58b7&{ZW0bi?!I& z4_}NePWso_@Wgx+N)b|Y!}OC#(5HX|cr{xLDA~S=k;+?JA!gXQUj}#!=-3d&cT80u z!<(D!M&htCVrBIsiru_t!(o%|&cw_N$FUDu#e_lvJ!T?I6`UU!Ho$^;7T0 z96`y~yK5Bd60{>9ZPk6BYsj&u7%r4IgJQ3+9|j<2oC}nNU&OS8Dp`3LvwfK^hVoGI ze1TG)Do_e(^;N8s{%4iB^fFgFi9Q@-r(3PIHJwjC{wpX%Ho1Z+8xYNAbJx;JJ-U?u zBcKN3=@}=p>MD+M*_Enj9KM|qdQAN_=jm{xNjr(_CN z1RQdKwzy@1ff6lZe-hINYUA|SxOW&-ahOrX+{z*hMjc35b*nLg7293!0fxKS^6H`d zI*%*2verwTJ=j09lEO&^*EaLXCu3BxoOb^Nkr~wSd9J)s5*(5AO8}gYQfDf{?(8_5iK1zEmB9}#1biuqI*R`h>2D3e6}kJp z&VnfZwb0&~@f_E$dU!SUfdiXskXynxnq`13<-rUK6??yxq6+2p5tdzmQj>8W{;b&V zsmE{%ENYh%Ui#`>y$lgek;|9rCw_-&5uwFE7 z|DD0d7?bK>yC+;0m-*F{AoXUu2u!5M8GmuHx5+}0?{lTqTg@b!hVhd)SFr*W=#l@Y zY%BgbgL6Z??qT(!f8CX9GNa@84e?e)>(!Jby+_xf#7lehK|{G|%xKSz#IQ+>3w1!C zOoA;Gj@~|VxmN0U)-4_gR38WgGLjq_dhi0OMz^45YTXe`{%}puc+oZ=H zU->n62yd5N4K^}$VmjIB_>_WFmep`wnCX0>gWOGNZ{{eWJw2A2Wi2X!E-YFLj<{M$ zEu|N2_yyS;3_V3TqnPSNBOrZWK(&RlQpMlZSIMJpb-}Xm$Xg?Rynnf@vQn@@^0#lSwQ-xmZ4sgx4&+^ zOr;;i(tP>bcO2AEe5@H!JEwR}Loc;c3XNO9k>Ro;L7eAzrL|}kNpLi6%lsPw2V=P) z=adf_N%AY|V}F+4cc{KhlmO)G*W2^OBl|&l3Y6qGNvp>J>_TNFVxkp@gTku*V3+y$ zv7!ESd~V2HRZCk%6(pRSbat-Zw~^AUOQQ02fbEjBbl+js(;9gL<8!@fLKRY2HEMo{ z);B=qy{ZI5E6^u~Rm7e;N4QLuWvHc5vAjzN2+Z0xIW$fclHfZWLI_-a#|}SJ&2kSq zzu;JRlywM{AsMSoMhNDIx)H_a{Xuw}_p7GOd&kQ1C*dB2c|g7;LE|2{j{}L@!3bNB zy(ePbta+@5^+d-7XZ0WlQFp5IIohior_7)e3Xg=|;?_!zGi!D)(F5u*jHxjy8F|e| zK%*}ceVUP9C+o3!ZXW%W4HGpDON(V@u$8e9`Ja-zTx<6fH1cu3zX~M2W|n{1C>4@Wys2^sEx~H%xJ6ed8hMIF<2#P0s*-qD1enjNK#Ih~kV>X}_H)?WJd6jB9#oqr7$?IC~>=2Ad8M zH#oDYd79Wm8uu3`s6gc=%Jm7+;6vg@?hJnk|BRdludcPSk}>r~1@~H~VCN<+{oygO zbXIk*0=fCjqEC@-xM_8d_Xl#U@-_jC4;3Wb4raw2m3_7g8W^AG(w1)8in8Jen5J4J zrJHw@wdg{EO=#)Z|DA=^&NgnXO;|woBBEGC-1!~WHC9ou0i&xcx?;}Jg+~e2bhX5u z_nhqg>=QS61RL~mvO@NSwHD{ zKlgXw&GPij|7stT9h60jPyLwE^p)v#0A|m>AdZCKvyr2Ay_kzE|3ETN^fKQr_VP8p zYuWOxD462IcN5IO@y!B8IqQ!~J=j21{PwE7wYqyso(dV0NAc@i5U<7hUi3q2JAjQyqzL_lN=~pfcF5QP}`is-vBLgHO0e0+bzgNovzA z+1?aBSnKOxMS_%^{!M1Mq19{fVRU>V>063Ek=S=j5M-s_g+GEzYmF<4yB$>qIgnvB z>4)JeFcESJQ{|Wf@7RPf%z<#zjY5~K0|;zl$rN%!-bZ)%$ipl1afC~ z6_Z7|H`+?YlGpvG@}W)bXZbuA=uP(q)TRZfO0hn_+d&z9LQ=A85z^4>mKL=CsjvoF zm6JYV0UhOpPaSEp87iSLU+ZH@GhfS}Iek&^8gvVQZ}>Wt7^6nMbB(Te&0x^DaCY## zPajDx>2fq(gNWSPXOc9~H;*$X$HN69R{iiVVRdj0wW=%8wAwo*ErJ_M+Vjj zzcZmubl&o8eQur)2wL5zzN>@eO_qkB@=F!*#QO0|>65hR48Q&&Z8diM5X=^rT1EWT$x5GQ>c|5{AkM z1RUAW`=i=%LBY$9T*jSEB9>lO;rXNMi6mw@K3&=At0k$ z`dVmUOFH*#LVvQ&a}Tx;oTNi#}8+uJo}g$Ts?^WrWxMbF{UoMDyrV0 zL3SMOGs+X0UNhpTTIw!N8H0@*q-__OJ*AUk&FCNp`xtW#!pioK!73dF#)^r?ZEW&3 zZEUOZ%3~GuoH^0DK)jbRE_}Yt!sKM#Pe;mkYzTK-iP8^S2F6=lZuNT;zNd3!2R`58 zeFu&*6zfH*1Ow1ZC^Y8vk#oxuD&1RZy(_rgx>JSBh6}mi+-BsAVu^{hjB=1}A=UW~R5V;pmmuS~$fmtc5g+ZHv!#t=xgZ_uSh!#gt^SRq5uZZ6%TV1rxKZu6Q> zz^boNpnrS=QeuU$LI3S6vIS7Gb>=cM;m;_bp zlFzbNMc{5qa-jY@x0Rg*tpv?$?9(<|<7mOWJ~Bv_COc8Gp0kfhHUjOfCf$Bi%*kfb z$^ZJj$@S@{MHID4{Q)k?v0YUfUqg|f)5c`QZt{uo%>BWc(o{7ry`;xos_?@xs!o$5 z3n`zc`f5EB?(_#>!}jbu(oEbc)2`4iEgC`o-jk4@_XIUC?$P_50qTvJNOSG)Xg z|5=xX)(rYVm)s9CJQnlk5k5BX#wTO)fafAX@4qbeiv!pNzbO) z1TN(az;pLtC#=6wB+hI1D6_3QJ#-x@>5XEYd_xBE-AILuZ*;5XJU)Asij{$UfJ_Zb zJd86`7@F@f6W5N{N=*;W6?50a$%s$o2K2w1t$M|p1y??2 zF8JDSJ-{(}VQG!cu6U+0W--G4J8G${A>d1h;3T;#$a6O1tDO}#6uAmTPI(1|lUaBUC6v`~e0K)b`VqEw{pM~)vo_W>$&t-lN<5n_(It~s2xmkKBrvMX|eOwGfE zIH1BsJbv~1%e80@f4XsRg@R?kjAbS)-B4bN>g#rQ5h0kV-ZUZ`6sakmFaz8gS!+PzjI+LYU7(UySUS- zX#rM)<^a!|L#FGdrG|ETwq%CnD&*eN+&Y~f&8la%^aZ~ffs%%o{7*@XHD5VtoNHu| zpCz`q3K_poAN;$*x(bz7Z2ewy>ego~browd7!B;x+`91;({vHjp?h+lL;O_pcjDSn ze<_n-BT2EiI(Cq9BsSjvlLSoO6|qT%u1 zhI6R#-b+vhTt)d3|4LEN{y$O_`y%A@XHjM#>^eR`w>a``We%w>&*20=*%9k6Kvbf* zyTlH=D7i8u4g~IExjsu%LzN_WGzo9q$^A^QSQ^()N89%+Fy9!t^eP90^kamHNe)9L z_dLnvh5@CW)@@*ZgzO25agE<~>h)h|WO*+)6V3@mH2``(DjC3`iN2DQX~D=ixMI7p z{2r6unZT+hm)cpb+2Cpf|4D{sQI=+?UWT3*1ccGnv*u+ez46ad!jE;Bx z({S|P8u?oYjJx3i;s=AZarLZRbL8EzYSG_Ld#sCqexb=pp`((#UkvEQBK!B3{1`QN zoK}C_Jv}BlD=qZSW(ZDp8>=@WsOC@F9Ht&>oe$)o63j#-!j2L;td0@yXmQd3jkOqN zz0dSXqF|+=+e%%n%WzAkak?GCTY`1Icn>e%?Fcv~@fgXoR)+6g!Q0hKJmX=MR?U1!&kJU-(x8xFW5kl*(;i|Qjpa)upX7K=fDQRrX4x?JG^Es8zHkU=>jv>K!#=ex^ zem=}9HmV71irIm~ISS-}C?yF||LndGzGl*ljWh=qFqKnR z>Al6v#HYhJ5krF#^=d(Yv({}E59r1E&z7|UPG#-}oOnVazECSK_adhbh{FDsFw{c8 z<#aa9_su6?XMc?>6q75N41TTx$%tV#H&^QYWN9I#%BnDSfS-5$%gr5k5k39Sd6+Kjdaa{6)BMHW?d3h{ zwB{N8N{9IP#4qtX%cWz;6BkX56tfh`p;rXVxxzk(0Mv=V+?L+m%-R>^n`^)L=sfmm zc(Owl4dQ)&z@qL8(C+GjAYW5``Ex6g;{z{%Nt$EuT2$II*!i${UR^I_mHrLFzSh|& zjl9enFh2-J%9r~0{YU=S`;W@Mx=DNQ+3RBdL>KVutEj65Xn?o5X4>*%o{8l_9=fwc zw(5&yeJj;i@8VwMWLax2X-S z5@pcl*N6FbQ#VUlnH|L^F^7zu5aM6%7gPZd_wvsy6!-!0 z2hX6LnKPb&&G=QNgnS3bcX;}Zq15xN5ngxwe&a% zk|5bIIQllh#kCIu_x?WQh{*UjefE0UFlm3%41|w^91=618dU3a;_EOupej8*B0K*7v|1QeC_-GBt4-@HC6*oXAZp%xIx zy)L8wor29X?LgwjVd(6yr>NnLY?Y`vJ%=>^u4DL`FN;O`3vk9N%+b7=+5N|d^B;h{ z^*_&AI_vimOTxY_4p@ru=DAm*0`j(qLM3AvB&h_?mESg9@zKDi7 zWRVAbVe9+#@;}guy34y%LARsD*k-pjnI3KpVGV<=w^J}w9WOjfn8U6JFYuOl_J+A=gZ0-yOuG6DsybN|hL`AoO#$ANgeYxcUsIm_HE&~5^Pyx;EUfD;u|#M z_{r|EJaJU$4*%p*?^MYyo$T=hwk&ycJ6)7#(Q%0>5+^@U9wE0WwCnYFMIvPN3+(}L z;yPw~o%Yjl+DO4J^mLv*I57}O9$YQEAI{m9)4E@id;0RuULPxAD)H}fw=@e=KNoA=MJgCb#@yU=`%(8=mk>YtYZ{|x8glq!!{vZA?K_v6Im_)5d3jDw85DsKB5qqsWn zhkBM`zmEL*g;8hUy?6=2hA=;(&w>ws6d3iJgVw003v}%lYwu})jy8QQfeWS>LVS;u zO4ibL86J%1&L+xR!S}oSgy;`rg2uddB8n(+*PhXGJ0x4%YzXRM&;0OVkjcLpAHow^ z!*UGQKiqi`n-e`G3o{fJ+>^?6nu7z@Vl`~{PGA{Ubl%a>9AfOO6MU19?KIMfjN7n; zwX5u11Tt+os{XyuIcT#!z5P5pDTA~sS~cKz(Ys3;Uo#Qr=h?#ZL&YFoUKfx*m7WV0 z2sZ1}wqp{u;jwy$evH~FOtWu}HKIY$X)ylHN!01^29|zl3w8)MD1cd5F`YAmPSHcB zMm(h;z=Q-@E;)=H6_q8t%4q4MJMmpKQXL!sH@9x#vFd#ck!{=PBhHV-Jw13za!T*> z`DhjNOMHLU^`u4ixp2kFz_ikPf6UM8_O$3Bts&7IKzj0BDcjUN8BDIdAzKm&iuUAV ze|&U8PR56~1!u)`CAhv$3Z4B#Q@eidfv{fJc)Z2kKdVveub&*(_M3GsTZKR6Lr=H4 zVQ0*u;3!RI&UY0N=MHLU8)j{rRxkfJN3oS>*bY7Pn>4^iLE7-Y#tK$3djXn+I3W{W zB8SUg=l(*Rx%cqRbG4~@+ZWY&nW9MC!t%Q>5c7a+L0sc?P9680Zv=8NgHu(8EhyAu zo){Wf+*^dVB zlv~~FI3dY)IacZ3x#?bSc^Ak&B&PqkY_D?lpr{lovHRHrqhF}-qFUs@;hw;*#*PMk z4yc7E_K8~6l(0qQc46T@MEIlO7vA)oR@s4XA^xcvXD7HJ%|0Kk{qq#j^`Y5-tNRk% z&+Sqzun1-@A@9LiZ=ZV6t`28kfaREzQNzBtLN?JwvV z$#Yu2in^;WC(ja2oXc|SjVr7FXlMTNG(1O*C!ds_>3?0y69igphebSy89z8#{GIFz zJut6IG7G1`DvUI6OjsS0&JZZQ^n6-8jD9qG!{%*sBT`a9*s$SK2f#IO(Ur~gQY#j+ zJRv&2J;NYaXj}M%O+gEdv2kZQi8Cjf_Nd&ugWUbz8edl8TV?GyM&>Tc3>to&EX!kz zyLg}Xz(?!()*5Z#5D9NO{QOg#HwBE}YZja3t)In;?oV|QP1BrgOtO=Ql@UH9veWux zhDBecR>VitB(2Z*@m@hVzD}|?TqyEgac6kdDH9*4P2Rry33IRH7 zbyic_mqdJz#+IFyf7JD+m$GG>-qemw5)`e7T{k6C7R=$Q)B{HO;gyJYPBG`U^Ti0T zVllGDC83PP8hNRY6=o*gA{oh+uUY@_mUOTcHM|k}^SaM>;FUp8FC-vXeh^+P--(EK z;n)=Lf+Tdy;~!4^7HO;U#f_U7`aC9=C+>l>u?CzcMG`OoSw%`T%Dbz;+v;@aFm7+l zqC6MtFTWQPs`U$HLtLS_&NE7;z>8F*#%Vi?7FqI|px4if@lpgwz(SvQXx-&)8&ikv_aYhzK+c z{m}{RNmhw<^VL6NK&{p0OiJxM{hCkj@e*$9mosN3c({!}kbnmnQ7)C#bskIj6k@5| z#={FxA-euiGhWr~nbh6yQxF)lm8smP{j4Ty=Mm(BtovFu>37<2q=HlC8nLx)qAPAm5S+3_B?G9$) z8`@o72n#+~{^YmKgW%pKp7s=7vUMJ78iWY#4RV9$^K0h~$=q)~`Y?5R!d7B-sp)}u zI+Mp#XLn6uOd{3dzenbNE{hC*U-5KTxW{goND5UBV(NXIq2U3nfW><@|I8_K7eIib1=0e~@y}WYzcpjJ4WWcKT5$Pa~ z&&ImHPx>(sFfcHVG+%R7)k{BNRO7FfyY@}_xt=JMMOe63(N*Bj`U#IQWHY!INr6t( zcBwWdB6twxY7ZW%OslpRgE|Js;H8ztuk~_rYGb}~c@#40lz`ax#?bpYcv+~?d{5pX zL~0a*Rgo|k3Pf!Q?jlPVx|gW6ASe>@EZ$oz-#37rB&y?fknvAFkL?m=kWq?_*Ymxi zP!5&U5{K`PK8>#VJ}45^8ilI*KYCb>9J=p>kM5<}Ka0C1l~K^&7u&8U4BX~3`DIYn zIt#W&nUDI0^hGMBD!^Z_uWu7bE#SKbY<*dG{&=iqJzX2j9BxQ{u{m6++OH&Jw&*m` zi5wF5z6W+BmG37Ia7Z7m_<}WAp}M2vm>cn6b@560=;p_v<#ctsliZ1*+U>OcM|&>D zwz*2f2NX5e$}`mJUUHvfWbK50GE%W0#51>y_y@)r3Z@Bj+mo{?94?E@L@(-b$=n^T z-Cq483fS{Ptx3S7Kp309D{`g#`8LdsFxa$N<;074 zV(F}v-d>vPEBI9T7zY0SI?$@0elj&d9@K$8Kv-~E!YpnHk3R8TFz+MPP>37wH$G*+#*F5X)IJF96@`e{>SO5n8;2iA$i`I6qls?}C@G`C!+#p-$@7Tn4uJ#Ce(Mi* zijV1Y&ajKghTi@)5$8z;kom1GBuXYxq>QkPAK{OE5{ayT>_2AKqoo5lr~DHFTC605 zsO5YWg`kD$%eQ11SbE))(J#8^aUWh)%;^Er(QI3o`+X?|r9Anxt%!1mh04=k@k*dc z#8?&Q%)>^9sU+1k|7W7B6YOXzgD4|s5qKPwJ`iYC$%?m&G-@23@rWCw*)MY7?F6d=(_cJY)6 zVmht}WG-1^Qr_vt>=W8{Oe<{Ya_>#etjA{F2D@y8tb5DOO8aAFCq ziA2KQX4?~SxQDub;m9yk^GHfYxiQB9xlOI%r{B>8p&;49e?C2s$5RtNAq(FQ|Gt6q zq#3|EK1iO+gNW<4sneXyxaenG(w^eiaMrJoZGig2b^=#e^7UvMo1Cp^pvHOV?C($*r#6xCvyxf+m z>Wn~$?&5thBc<9XZR2ssURfO;pwF0qEd7=0xI!aquYY(~J5b|ggFY-dAmPMrD7F4z zxx{Djz1`-<+)cw^(?rgSPDOE|M2B0~ZGW+R*AoXbIk4fKuT}oYb zMy}6M90_Q0IS>{glD?>1Pj=n9|C-1*XW&pf8wkfNoYJJ6ACaaxVATZCyB*ynr)Y1C zL)l6sv}nf#2kC*M`)NG$>rgVCn1|zwnn7YOgTb+CfDtBwv~Of|(+wKa8>*chfW}o` z@Z?qBB(hOL{b9!ILVKpWeAQ;jZHmt*-gUX5U-zqASnTJO_UJFJFGhPzSg*RjoRSvz z38-0Im0C*XQRwW1Q7RjqQ9vT3@GQGi_F$y3qz)7Y%3_H4b@ZcXH1mDqzk z4|!AF`0dF%ditN{E2O&=NH^169W?uH`}7qD%N0IQK1}N;S3ZV;P{vTGuBEoR^|ZEy zc`fIALf*!5J>M~0%L9SPvbH_dm-5|gwJ2?(O>ai7S!wd9n7DCz)!8!<}BKW}Uy|r}13?qc~nBcVjkQ79G&k3#dpmORX4| z^>Xb2wano9ULWkK7I@EUHo$PbDP&#~pWNJmmBht4;w;sni7o~6J3JnkWqpA9H7@i* zW|x{MW7rl=+AWEO(Ux7D&_5|Kdz#c=;HS74jrWw-QTpyHlaA!6JV*XzQ&Lg2T9Imy zyq~uuUNKX;5Z#!)I1>}%NXeWfi9??Y@XN;Zye4-Un1*}z8St}mOUY`>6G)CehT%nt z2^U+eLq4C0L&l1e4E%`GT6s@g5!VM)BsH;cmt(E>mh_uR^eS3-Ex8rGBJM;TZkA-; zS(qg=WYnuaG)hEOD1|JfDJy7l)xd8MvN9M6zj58N226?CBGL9KqWIB^sq>ue0qMHhz+8`{!{>0JYX z`ZhD_&C^F1KN)TZaRI5jcC#SBt_T~rT+8(wZS#1^PlqwAM;PyX6z@LlMH;Wfxej(U z&Xh?R1lE7ERHK`&rP<0DEXxAAEK;^K<6R(}(r`tIP%=16z1F5~IV)lwDu^Iu-t@%o&1-De){CrY<93#t2bJ4Cl5a)9}PO`pkuj3e_ekGZ7_WlhX6Ji?(Gjxq| zYOfl|H?4XFwMgU!zgsd!d{2Xl$M1(*AEO#}dNq3Zcu96+{M-)@GBa7iGI5f(!!a-D zwlyUpM!Gjmc(Ls)C zWqxdXN7Zi6eQekq-ABd=?bQC}bnX)trI@46U-cB2ck%GjIkUhx%?eKcDU8hyc#U~K z@BTDo4?h+5*z^xu>bAw%4x_hc?$fhnu|eQMz1DFhs=IxM z$Hi$c4aPqkTvHdT>FqCU7_<$;ZF!=M@@)YRUH~eTq*Tr(AX-8l@jghu?oftc?u~|u z7Lrt~*r$srx*gguV|orm_VQm_w8@$UA=BeWN?U1x1Q|3eCRoSm0s|WA)N)$C0keq4l6Q>9X?bcA7bImfr?@{#OO@q(Yz;~ zbCN#CzvtoAY`F&t=}fx3{#~8rQfi&u8U=PCCqhsloy09Nl}P-AG^{`hb6^W%x@QEc ztfF#Y3ozgNnI+yzRUskJI&#BhCC=>07T?xzU`GAPQ(sTzG{KPj@*;A4*xvjw^BLbX zFlcWXG2NMK53{G9L&w{=fRGjen<3YbT(Tb#9{86yjR{_oGGdRP#!p&4o%6#8bDyR9 zCjevfo!t58AdqhSoVZxN62cMpxZ8uxOtS`8&Imo04!!$Lt1-LVYE3YquYGRC{d6s} zJ^Vf4kEY*Nh85W$C=YiZg9bp2L!sEYr%K12s=>RQ0po=Ei{dA|+{e|NXqKIUBPlpT zNa3e@e(`yz{dYKxR|j0&znC`L(gG6Nj)v*BlSa=wwe*3?8(O|;>IiU+$Jn6S#YeJV zUnsS0HJy4u#)yKA4@fzS0Nl}J>|ao{e;VcJMgb)v6IKhc`D)Cn&~-r5yH;(z!h%EPvTjBZ)A!oQZy8bBbWU zhp@|-PyiN*($*zV!26%WiY*dCJ--O|htuNU>w;TJz>WAJuNN>ySDoKtVdnU&Z<_*T zGk0L&XzDa&-+9pTRL-Rs_?zZ{Vj_U(*|k0~&X%mH(nd_wC5PL+iX4=mGI4Yn|0je# z-CM3il76yr=jj^L%cz{6XdWjKZQzPq8#Bk@ud`lJDrlvUa^>~i%-cSw!`GxlyhSky zkS)89@b&Rh40A2}B7hVmvF>~vgtF5b26L6K&X92L{&n6?R`sCk5qigBxC8F#eWz&t zC%-p5C8w=Uh*#`JO0jR1DN{B}+i;KJJ(u>nP?NysE zw0vD3w$4_y!HN=9MqnNatJxJLxVtCWie51Ua&)cVi^w;27ic{n zKz>~;xUrEUteK@gzB3E^2{GL|HuatWb3@%!J}adZBss}^V>d-_LH?Rw#uf{@2lgp< z!1Muf{I7=8xN>uLkNY9w+D1p&j}(O$uG)P1L1$PFQIlY?=F+6gsF7@`DQRY3fdv${ zY(eJ!(5;K5n(rv#>FKHLZvT~YfKSQd9DlZB@J|1G)E>1$%5@-E9}K8-D_U&05M9>F zFtv(Xf6v@zWu@+ayH`5q*#c8si5G2u!R_k4uC8 z?pXIhgETMMx=sAp`7e`4MdNz*gie)j$OsZviMM$Bo{Wsx&{uQFm|7#tK-xPyS84kF z-mN41mm6g`?M=(U_V&z`X!|MU+iD{Y-XNn(Y}B9(Bm~l z)hEA&Lv2v9qM{)mI?0H~StRAQq81??u9xxhu;mvlVTV)O5zPl{Yn_AGsM+vC0S*a= zJyE?>_5wnw!Cu6$MP2WhKtSPibDHYVkc!(DTArG`YpnP>nA1S6wy}9A@%tqcch)sA zw`_B8h!DR@N6G7!&;CG;2>d0&dU3Zm8*?Q!J=6i$JdU~u5m_0L{m7(ALY#qW1I3pw z(&)G&Kv4uUh{Ce(Q1r4(ddOi>IGr;crUdGp!LGpB2`NU+w1f0{)D!7_ zseJBclbzQ)&=xB#BEjqVer&%#M>0IqXJhwYbfi(7AG!VS4?kIXPk49|9@z{j1k9HE zaCw>>xeg*r$^!%O_)Jgu<;-M3^tIj`u`r9NL|hiPfr1oJlgvShgl35((qTkZ@jJ|s zECn1IZ|c4Rv98N5yeeu}+HxVs%k{kEH7je;9G4yRv2!_%Qq36VzSA|o&#s^ki>Y(v zdNQ7to9yqxt=A!mJiWyAIjqzH^}W+Pqh5G~ouvxV!Dm4^v<4b-BJc{t-@2htJLpx1L@o+y$!1pzg%EhmykW*hahxQAyaE8k;Xz_j46|C@n3dDFDpo5S&#)dAx#rqk$u&jv6^QXFBvQ z>RvFP!^$A;0>AHq4}O;lTtiYAkT5DpIl^~yR8qyQSY|4)V3xZar?iL=M2hq9Od7>& zxD923dvC3nxy)nHdWQOXVKDk5qNhfPB_s-FgluH-q5k|@u3eiL65cA42{(A|>hKrZ zleEB45<{f55cLr|nyadh0brz5KoPl1`Ox;**HgJ8dY{d21JuL{D-*BxQjD|5gdWex zPdse0j*lF7TSDL??a*`hg$pEpketvK(eL*Ut2fuoVm)pnRy4+})rc%Xu101Bpij;3 zciTh{`?^SZxWIoCEPBkl?5kQ`+aS>Yd~@tQXnq^?Fj1)5(u~!qNQ9A}b-`v|EnvEd zA1I~do&#z4SgcyylE_hwkO8~P#MIuR7SnBcKKJR^q}k^?y~wITTpeoP$KHzZfYKFB zcYjq)VMe=Od9MM5XuW~Emx8&F@=uQ{D*1CvGdXu}J!3eP9daB_ zFDMAcn}o!>Kh%{`yPfLfw{3KzC=9szyCI1-^o<^Apd;Ek17If@tb~-_F|1CEtZ%r8 zF!c~mq>s$&@9ARoA-06au2o{7HftKlD74*XBC)SnJ&J=nDZ2Wojji^-kkkk#1q#A% zG?2WAW(rVpdI;}2>NM0&hPL*xQW0^2+w|`iCi)LHb|%X4xqbjL{9og$j{zk`85opS z#_i5j{XLLI+bstWOd-q9gjZ3`p{!u&ky+z#iH?mv(Ike^!O`P7=Mzd+lX=yo_Ex&Jsc(H`7# z8y^QXSx9ft#!E?;vQ62$1P!E>KpgIc!%J1dWnW-ydkc(hpMTd6?KA>*(++Xh#X@_@ ztCDM5R7c5h>R4s1fg&0^>VV`la66NftQI=d3h?vbjO8 z3Y^z3d*-F0n?2(MKWArOBBc?Hce!vZ?}(8eNStZ1+*cWXg6G@+%H)~?gk@PAmohh# zSg7JEi19+Rk*|{W`q1)cTE*j;068R<6Ef5rT-%kpoiW?+aY;v{K!uXqpp6>$_Y zSmkNcz(r1lq9S#7%k>GfDh|3SP4)0)`-iY%J8S;(*)D{I}&SA~2u0w$O!@f#J~UjZY5!JjZIeaec~pWU)81_cyXAdj!m8nN`n{oC54ivy#sTMrAlkoe}%H}}r7qb>v{5^64CAm$sk z!wImbuo<+V9DjW2e|aJ|g|$^?47uY}C7KS2zixn2iFyi`4qE#ycNctatbEKt7H=YU zoAs;9>mAmoW!OxFWQGi>jArncK1JZixf3A>*etq4KbShV%~!1gNPH}l;Wd!$zr~}Chpym3QSSEVhZ=XA?cBo?c8h@9@-sPfJNCV#g z@kE_iz2;{EGY_;n^Q)ndTKf%FdYlYxzQ*b^kor}aA)G+8jly?Yptai?3SzU5yXB(Oh0Ql zVL@M0Y(`!&?nAeg=nGGW>bLkToyFSA$jzsEL={4#e9*($g9y*Xz0Q5u`FDK&STT{;Jv@c zz`rqv_DAdeEpRmK81MdGidhT}=lh}Kdl!PGFS#AVzm{=ZS_b}1aBf@05+@~Ja_2BN z(oqZK4=fBTA-XZqMTk5dy}t;`d3wIxIFKa@nXRd7v+5l`;t~Kt*v08W4O0O^Un>l; z`6lkEl7L-q1`JR}PePMY%JFKe5j~2=!YcSv8Y7$W2|WxJ8Zp>3J-aWnxFZttzCY~W zvNfmTWVTN#$(e`Um*TwVZ|AozqkudRQq=;UZ`{~5SN;X(6SPg%p;kMX^FEWLnlVcL zh|8S>O9Xn-7*!%P9Bp9WKSI=hGp0f8MmR+_(5x3ZRuyCvh+cM-aUDv2ai!I(Zm5+G zyPy-m<9;nnZ#tU7qwsv_Z$#zY32|4f06Lx83iKWKKooHUoSq%SS{KX8ri3>X#d@|* zN$4+IH^-sTFfbGzL**(mLvA4G8B6t=1ydc7av)380XhAL8%I5U_%BRYPT2g-@9H8& zb}|^O)J0Q%e(91?Qrl?V@7|)~7n|AOp&t?R*RS_g8F}B?7({8VBX@Lz@kt2<8YhLh zvm4SCk~)x7^BORsa@@Iag!#ONjCGYh$Hb2dCgulhsQ5@$_Q-79suUj-Z{(}`yX5VG zNYQ~I3SRH*-a*Ec?Qo`5X?hp^_GIHrBvTUOt^+V^gu4lvd42Fm{hkjxwwB#q*+lXd zrS>LP@Lm(=*W-JS%iZ`;sd*%7JzN9@-Ju2w2E=s5a+0+i+V(#a9-t(ZiUT1WFh6&6 zqquI0coYGCUW~c-SD<{6HxCmP?T>hDnxBo~078jWge35aW0mmP)w?0$A%E9`{CL~HqTyGs&P^) znT~pfVlnDtnGwL=l{g=~1#hQF!9ipFNCz?uH8ECE7UqTu3cNxPv+0|SJI=1k$_!@v zu0UsPJnlh=&wgj#Y$`J9oBgdM8Nwz+yT?l1Bjg|5aCaeFh|L;9TYmX!nX z+Hf9)368h3#Hs{qG+1|fVAd;g&8^}R!IL1d5Ex`7B1x-Jzudr9I1_uj^Vh-pPe8yn z!;hDEKFAgapkKg~Awb*Y)-CGI)U6jLUU!ZC87<92bJ&Y&a3Pzo27nlH{87L6kel8?Kv?*2y?;`ZWM!Kc5=LJIRNCqR&e=cd0Q$4S-Tb>$jCA=#XNy$$G`=bB){uDk#T0X)xb}es@`)cAj6o2q79z{vcB>S zJ*ls`7;FF>`#a)U{vLEgdW2CM%slsN^#kmu{8q1<;bEpyJ;nz!GraU8BAT4N(9GUg zYhV&}x9;nSf0ib0#k`I=KKRsX@j%4c=-4 z9m2T0?I+82IH)y)Wu#C#=uUX>3CqqeEu#etyBi<%Ji^YBM5bX+;YjQt8TGAFgQo zt^<_RIIEv-5dAjPed!Cb1bsr=o=*92e?c-XR$!1yOyPIVLq?>tnT=0{>=}@_PVUHJ z#V}f?RwY0i@-kxP7+q`+Vh+AR8w$MWLM4?+N>d^DAva6Y+De53=j**o??dM5_;~cq zYQL_f(FB%UO+T$&vCWSfD=XW1jJA^fgt;n}*qu}tQ8%}ps*B7}guPx740+zMt{x*k z(^*Du;X^1%)GZYgzaYqF*2J=HVwp6!>z49u$N8d!F%L`(ErP^SL`~G0pn*`MtV|JW6yXw>#s=Al$z;b$MJWIR{3Ojkz6zsQ98s#FLC897t(`F41Ln= z)uM2~9ec zzRueFSiw=(TM9zbjT}}<^zqwt_*aMz&(_}xY9Tdz0Vh2kQ|V13XQ!8TpyIF}^!1fdkx=A9Rh z(OkM9eCVOb`e@GMUZFCOAwk5C*r4NZK~%}|1@W|T3Qp3G-4dGBG*SyGd@wZaqtDdu zeQhnAJs4YWggDy);D6?-{UIew{e)nRlk>apb?;u{hy4@My^qr=#2&bBJ9Q@0>{`W= z4Los8#^Yx-H1?7-v<`KO`J$PenDk!+`3c<_QTwkXg^R^pnkG+(Y1${%&-rN0C zZQNG=TSIhfqZtX4s@g5*KCb`flSdd5F7-+bWy_iFt3S4EQ)m#H4an90g17PUwayFD zS{HIh8qKGl<)zfNzW;~qu0=$L`0J-J-hZ?JHj0HCpWP5TzZF1jjsSLH>vpjnZ9Xu? z2XD$u1Ylre2hQM!dj7>**Z!1j-DKVQ^m`YF*>AH*5@V8&!(-i8S>m~U0E^Z4jo{sq zrFq@$Ee2Okz0d>L>5y&RkXF4GR$!(q*GlUn-F3t8|E2!4F3Gu-$rF)Jdo=f zWm79Z^O8B;do^z!Bb<=mlrsaqvv~{<6CUj5U=l))ry_7q&DLym;=!VGah>#e)%2@U)K%gRAf@u^&bMC=cv+lO!WU4PUdMqu2Z%R0gC59kDt&d z9Rv-|F_puoq&v0PNNTftkpd)`gD!}sq6nVttDXPTeVuy__4afLv~Og17I{Y2EfrNj z$anJ|c6a2PDCSVIz}))!Ju5_%FdoSnU@RN1sPK}z_VR!)m&uv1JLaXnyVzy}2$<8!Yx)Ab9FXgs8(-huzPHxJ4f%k zrT*hu7vMw~0398&mq;!}fq0e&0pPpr0zmlX-}U?RWalNFv zLNa+6DYrR*89c#NRnBMYXyXv7rHV}b?acD@Od7-#~dVS8w*BF|r_W6!n|sodlbLh^JMSP7CCS6Ubt zVSj&PfNP3oH@fHo22MYWnq3%7)#9h;#Ib{wnQl9rkN}U(IHG?xbu_%?__t&IJIL3; zg*umyp7$5S75k%7YU_*Z_~s04JE&1ILB(8A)jk7dyi;ZWtp9z<{E& z@y)Bb#DDor9q0ilg8U+M8iG{QGet_#A5q8ujyZ8GsDBe$l#KtC&@x>A?>K{89z#rko`3c3fIgShy~*1Cs>u_fZy>_;qX^dBOAnj_1yiB zB=o=i%fES%cESHzLo%-Yf3qj{NBjKwuNOq7PIa+my^ev~HDmJeH)lJc>8Hu*|Moop z&B_1THv!`t{?FC?pR4&lSM&e=YxMues5vjS4isv+Zfks!g+k9Mta?sJE1678daeig z*K9^H|MnsMDIl#EJmxozi?wM?yP@T2LK$s{YJXQxAJVj)82|a-!hMqK0Q=?O0pyn* z$=DQkO0WK|TXh#u?k&6#;$|H7dN}a_%)2Y(Z%oFiHQTSQo7tUjsqZ%)h%F{M1XZug zmE$V~94;_2crL@Yl}gPGn{GEF$FZYLQcU3M6n|0qY>az;-_tyYlDZwU;Uz#45#+f z-o5K-_)vS;xL3Tamw!;DBzEw4IVLlF%)^8L`v|!CIBf!C z$zp5r^NrAVdeL9QNgR=>a0W=x9WRhY=#=vL3r<)QOzn?vz1pOABEVp9q0ydq2W+|w z?EJ**^A?b+s1s8L~k5A8I_AAX8xEXK`1%%0`;>o3o<&I#AT}oiK z+@BtfDh^KS*AnI&_gX?&4U&0I1&7@n00&B- zUUm$4H$6oY9`m-BIS3sn%n%mA!jiaQQ?N0J3DIQ^>}B^m;;*a_a>Ww`7%si{R&I*b zGeWEA3*~SX02kDlj7yO|;dSsqY_aIhx9ZB1#oFxGPhR=jolk9L{Poxr?zo4`#SB-c z3kmSrP~`V7b=Ld5W4LFH-z6b;nKWg|i3rI- z3|idD5!g(Yve?_3Y?s)R14_mwe+B@qHfy4(tWR{;`z1PVic8meJ9k^LNQI$Dhkd72 z-q=BbC_rsMlJQUX*WLB6-CwdD8^j(x1PlsMPOpr{zCVki6$n4bAr}7kiV>;<8AkE7v=`RZ%TcnbMWc8T-GtU7L!a=n;Z*0)Mz^j zCH#QVHhxoyw^|?!$;3Uj`jm}FmzSVy$%Uh!Ddc>|VASy1&fz(FVG&^PTg{tVJ}NGU zKU+A|<|B27uq>~k>wCAB|3eo=k^}>S(^=R;05rVs^rFZ_-5HL~(qC(_UJJ~E zET&$60}&WM=NtkQ{O0DzKj)A)Cb;@I$>AqfvQY0G0wjF<0>mD+N)2g5Xo)DXoSa zR-VMADf_nJgFf3^0~y8uBuLgnqe7&eB&5f#vjJ+&?vLAz98428LXRYFA*o$2A-FQ` zlPqmtKmB>e7XS-+vU+DcJdT1(WfX6A6T|tuDDQFP&!f(|czcl`57`AK9y7b+=YfDV z@n=DEc7H)0pMb?RA&}M|c6x{ybzo-z?s^MD8V<$xM~l?GNv^e=Xnd||fOw*GFhw!N zYBe_e7U1M0>9TX}#}1<+++Xx%X`G5SqJ_^qHps zK&OK#eGa~sf$xv(q!oPXSHB1#qXI0_ci;q2*S^Ch>n{4AGF;jJuZ*Lt2vrhW8IYjH zeZp4rJ6Km4i0ux*J_?||M+(TIvOcJ=IqRW#YE1xZe~)UKw-9Um`U~_f*T|-eW<;|q zGUgoC6nGx3(M+6(>6>e)S^1n0hWD{EjKgaILipTIgsb1(2;B*;`e58}$^D|Zd)HuI zbK2iOw8*H)K_e4Mo{hl*Ml}pP*o|`IXB?92y22#nhyJYh%R=NoOT7`rXyLx^?N1h&XAjB%_^HxvtPe1Vf>CuDWsy~J0qW`nX zqYvuGH-+uv&Fr)9MjC8AGXb+_KMhFM@5t1-<4Of7Mzw0d76 zFzM)HU7m@Fgg)R0M}lI71ZDOz+=u`(^#ocqdMT~yV zztH~u5vxkbg-NmfMqv5)M&IJz-1MG-_lHyo>s7C#nM4br`+fG-N?UXv@7*XwDCOGs zq`ysBSG=d?GF!lDW(VvP7448H>$jG*D}@Y z>QHSzVwfpbX3qtJ2nLxj33Ll&T~vPCoE8teT)s%mb5Ge&-=*3elhobJA73e++n=t} zdNu~)x%Nvcp5*|hm*{U!4+P-Q+KX-QX`9Px%-h2!8kP@y>eavP04(o)Ms@jWe4w@z zt9QoD?Y!^8>wf!L_H{`0cOat_i#3TQ7P_B*uX5UZZ6t2D3?|Fqze4l5+tEne*-fK1 zE7vJ@9^3>oo!bU|IkBT=SnLJdokjW;(gE9O&$QngZud9s?MZ8zwrIN*-`vOdBHP|k z)Y^8N-ft%;>=09GM}Tj>17Zw&j}ndgShf!1={zzocjNWWThUwtoa$wVD}cu-HeW)R zAeT>MmftlGA?nEi(niFM%w$1L#B+`P24B&!UOxYwSmV*7;F%mfuz0oVbK`RlWZ;8N zm~V>t@H!T_&$9FU>6L0^yMhqGrhLksgi2* zJ8=Wdcfq{~M_hxxZ&5fb+==xZb}+|&KneDH=`!{GN-(m1JZ%OYT(@`>W$&xNW71Oh zE&2`E^D90FnvSQ7V$vmEZD$bO<8ze?AUyBM4UpNjUuN<**Q>h>(QiYd`<&ll*Y1WO zUHoz(lOv4(70qeut2D&e@FHB(R|V2}DF;?%82U=ZBjl>8=q5~-cw@>> zp&pmmgWnIfE&_tWERO0jz~OP1<;~B$G;~-r&Mf%@)#?+b5?f z(|Rp3+h15-zUrPSP?k}*_cG`g4Kj0xdkMCiM&^djfjRVuU*MP3LSKqc4e6p-Aq3jYFDK=sQ_53*MHi2>edTUF z4bMq&sU%#6F#(Mq&iNuIGTfSmza8E&w-+r@P6}o6yL=VAAi2ljYGB|?^Cc*u8sXh?DJ%5~bc1b+-~WbpmW2V{qo z_2>4&Xi=uQgY7^e*qlQiYoPWQ$&g-d9dU4f{pvSz)jMV?-Wsdgob{%7DiDcdeLdZGB7_a-aL)2Myte!9%k=ry?lRkSBvsodu3QberL z?8dtFHI6=hVq1^6;?Y}AE_Ti#-c`KLt%PZW)h zv@-;C4c;^(HmmG3C6WV;Wbpzt^IF*IanIm1xHTo0$HYs8&=?BAJ^QfW#;=$|KP>{h zXa#utb&5F>Jpnz(gq%@2t+SLbE!DPoVay=7rRf@23;pb{PeE!&l&jFoohaa?-dX5B zxhnMY%Q_Qquq{f{6qro1bRyWl(~|=&^LxgguXVsWYY9kBA*5+5}axlHK2UY3V&W&I}-@rHF zx&@#!;djNK8%JjqJ5om(VlVKcvGJosBy}s(YR;T^@jKs1b~c+NyC5VtQ1R3STrkmV zb)p7SYQDFMw!;%<$*)J&A$ef==wp$N?=}lCsWmnmeN62(ZI5gc?kA9}|1l>;cy#iv zNVha|*3U)Ia4Bt!YQRByPu~`_7FzmI2j)|k2(eXq6LKDxPrc!s*3{Cn5XG94^n;?_ zn5q*glTY4rj4~~qWU4XQ3$+zIKn0K5Fn?qB*`@EMw*}wn!-$aOOpzo`_X1ajcx$6p0(y11hgcHH`(V3ayiA*rUxIkS)yb; z`DY|(@xQD*UP53jBV_8y*6az=QI38H3}Ukx)3M7LOz3N3)TgH47^*S7dK>tW!*NEl z80>Zt_(0>EoN%MJr)JAenmB76KQF3`iFp^^ zZG`5YaIX$UZAE;W7>v)`-Rr6iW%W@77xd9tu zFw+20fyZvpNzcER?3qwh#g}ivc6~dbIQ?q z&!EC|-r%$zXe%_W8p#IXmPx-l-S_z}`4hilY^Ffd1hD5HN*aWp-^Z1RbpnxaZ7Xp9 z7DJITkY&?2H&hYVP!%`no4|bE*NV*@Q^bpf!mR0aQu6IUE-d7XV5r>d6Ky4!bL^1x zwM|pQj=5$K$^AiLRzg+C>n-#13=t7ivb`IJR?R%S<-Cye>>pq;qEiKc+oe)?F((Nq zdx$aV&a{p3WR{^zO6^CwXuotxH_72l2dD~ER@OOrO&QUY8;ew`rZ(?zr^w(C*=o1r zCF{XK(@{vPC9bh!q*k=G2quylu=EA)q*S4s1B-#dg7Y8tov4Jaf>ooSS?r}_pQ6#N z%n0T7#^JVR1fPhK>O%#mvc3u7tQE|%#nFYv$|@%U%b4~ruDO#Ch=dWzKi=)2v{aN; zyu+0Uh<(-*c%HYnoX;*dQUEd#n6pu>HZsfiIs6;D6$28ranuXHKiWAU1d<0a;K8)t zXgMv-88Owr@4={7_k~jSo4h79n^O4OUh|m&(nNBvi*x>X;{w}P!48cEYeZnh*ymb9 zl`n6lD-gC@;=5s}bk4-TB}Orgx!JKk$I^NoUX}dS8&kJR2gI~4^5bwXsag)fd-|NU z-QOI2*rJrpIZznq`~}sQVJsPY#*WyZg50}1Q_OZS4>P#q$lOgxOqh-!Tv6 zkiA%q7sgx%Cgu*VLSr9MuAEpzRHx5uVYi>+PgsdGkzEb)^(h73l zD7l`wv-X#Gg=Ob zJ`q)1I`UorNPd=`X>8oY{At6zRJ%sdmnxXg`#N!L0i}p{gNtMF)hDkO;(_Y|J2avo7aUa&aJ7Hq?3 zjl*iSZL$%?R!C}m*=RFyMG1Ea~nRfF~2T3QW-TEyvMfEb+>jOBDw z<3SyW!z}G5DHo}VkO$T0v7rSbuNy-8sL)@%ht9KV8dEt>eA|s}?}=pAx%~+)R+nH1 zH9p{xN?7>GBQ`r*C?Mf;#ZK?2|E1SJa_L$CS}DO7p#GHc^pKE@R2_#NYUI72=oIUbV{dn269Y3E~E#$(AO5d?+Gr&MdP-HP}Z2V|`cbRTzS(z!< zhA?hy$)hy3P+kvaino0!v9uT~vv2Zk|Y41WQO3FhF1e;p}NHkToM zWXABU^x(UfKc)>z&KXH%9VD!`JSWsg|5dL0PT<`~G^|W@$L#^eukpkYgmjwQYV2DF z;m&UP@Q3Wzks+aVgiVv`InHX3`TUSMfYd{VAO@fJ#w1j5Mi+B~!>MH%>y%Ui8b#_;&)$z2|Z(D-fNSn&CM`P&6chx9-}_U0;<{PzjPcYuDEoi zUeIo4$~BgI$W~hxrt_3F1S(!c_YW6a-K-s6K^rX8nB>a@pO1W~GFHHKmh-lxAI*G< z(rr+brp}1v{k4WtZs-`*d#51uK-Pya#|EdMel$m)(9S*gp&!0E)AO2$+Rk|6+0XA$ zA@K;qP#G%li_Nd74>M0WzThc3ot8)m zdx6g=Y+BgDRowD5{>_+1ndY`tYDK0_%cl@UWyXX;UWE+eVC};1@M~)QV*+^^^t}kp zO-k7GO$P|l`O=t$P)3GOE-uQJ-eMM$j4^F@$S(WK86Q8s5!D$3I4pM=#t?Z}W9Os% zY&RALC(Y1nerWK~9<17)*@HlLS+_$cl&(nGl7?;ItA=IQ-Sl;nPmVQ0d1uzIN~cI8OXOloY=u|p#+LAEDIjLD-}QJZ zq0VLrR|ds+D$TA<34FCSL9?L-|HO{CiZ|1)!B|9=vqbjZ^TaM4K1|a@09o+`(8(BT zYxBNqHa{&b?wirX>2k|4Klt7#38JiMKc`%oTlY;ZHHmyUi0^!wAm;Nr-5}v;8%2wa3D-#pV1aebcsYBD zb;NFB?d!(jH@+7?P*Y;kChM2gQ-Y=dgSr~zXi7;E9)W^C;{3+l`mr|!A;^6PS9O-0 zVk`3bw2HcdDKDr3X+fnT>jPTR5a_BTmclpUbS4QCe82AVM7oRUw96RR+poI;O`7a? zx6|Is<-|+*@OsW{Z=8J$^K3mz>(cxfU7?qx!##~$O1%mEMMk{cQ=5@{gR<}HU^@Jxp zeJdsScnX6^5i=iTNA$oU0waEvJg*Tn3}Gb=-UJ)IeV^RT(x&S&i+SM*fEpHn*u+lN zec`=S9{$U=@l1;v3=Me)36|6M*bpf$z1<+HUmi7u&0{DNZnc&xH0AR9Fm@Q3;fjnQ zxECPCPeEbIB>3xEPxu-92XK?}X2YZ8KbFsU!yVoRyYkzOId2T!dRvkXK`%gnfCW6( z;cMhr*oZR$6ry>l3}9cfZ5I5jUUf&htuz9&;Q_m<97`;O_5jW2YA)vEjsjE!xP^*I+qR||qh41oV>WEyO2LH0J+R8+I!L@tY&{@mb`^AbE@~S&5XYsQJNYE)D{D{r~dofFn za)T}zFE^Vz=u0K%@r@GR*i(YpKY$47p6aaYq4TFjLjUyB=+e-qILzfJdYf9H?<$E- z@p`;odr`JA9GJ>}Hb^5=(#BCb{ zqFDa@px9s-%@y-ThunMceC~%!?ir2v)%tH>ye9hnS=WsFzDiF5_l*)CJ8~_0@f3f- z)kxYEJ)M}nlpD~>Q^ejb8k3-RERbKH_5UH-?f);dejV(BY@GJZ3p(?uo~kN)az#%R zLKf{N+TL!(p$^m{$KH#n=FjkjP==qrpIPQ&q-b)pdQ^3YLbqo5 z!1tD2#08xh5BCJtCd>rmN}s4k?xhozQ3 zB*E%59C;J?dODXpw8sN8kSkUsD6}7|xfvb`qves4K zkzFK}TIqxbX`F*}q&{f8*FN+410iLBd;4oK^V!3n)ta#5b8S9!qTF!wc(3TF86{)z z&~-{3E>*v1uPIvAA=GNfzMU}5nZYAzVg2}8%OI-z0T+A*d(cR!F{ab&rE8&j#=TZ@ zNpJ!#cbwfeac2s|Cg%zyRf^U|+75=a$WG}Yi`Fj}@VXv^lB6aTY4vpex)#b0ccU0-TV(nUYo?>V9xs@)7{bqRVu`CH@YPCE+4a#1)DPTkc=A?-} z?%{gLJF9lmI8d&VaB?^8d=NEw({wrZ9dfHQG0t+D^B!kAb|v@?3`(*@z&WNUzk0DQ zi#q$j=Y-xE4wK-(Hgux#<(BHatCV88A%CH5cIw$C7aL-P@L;`eb94m+5BR4yh>)%z z@4XF;;?OxM?HGk(o94{pz7+)LxpEMM9g$1SAb1;{mV?!ntfxVKv(&Di4zChG% zub<|Ksg~ilrZet^KO$F7qtogVBXe<3LQx2*`#lKg76Ac{86BJ=GP@{z_Z!L}{DCc; z5@F_F0rj{8pa@joY=ak4$1R9*QdOjT5-O+@p>^D%E`a(e8w`RYA{gM@^*|o}%{N0M#?Pp$B-WvtXCsI% zeAgxQ9^Q8Ckz%jH z*L-v(pH0ef?$?A6%j=hlO?1-V;lSLIHP~Q5&BPK;`G4qo>!7OtwfkF;ZX}fkrMtTn zk?!smN$K1)NTYOjcXvy7_ue2Ko9=u*e$P2`pZosJ^Di?DGYqV~u63>J^+pt$ZC7DW zAawjTXNg-XnHzjnWc1|GvtkXi?{}qR^tIDw*Vf|`VUf`p#?vtH)Ty#YZVJpHOlAAp+AM+zgo1^m~m`DA^A~X3dPiErVFrTCwL$B9Hy6>1wiQIY= z*;#fLb3Y4F^b%5z0qZyE2`ygrEo+U=PlS@r`r{z26Cdjcf(BEapn!!+jrP;D^5*99 zhU~_Daz{cTA+Q{7y6*v^?0FHC`xNL5C^vKo`10_$bnz@~)4kinsOGmf(AXhZ3ww-Q5)f?TY zv#Q#~Reu*>w)45fg+5c2^yr0Zu({2k+~CMOv2wj+MB`#olE6awG=~N7T%>lCN=k-! zrOD*?HT8!Xwog8R4G3$pBlwturyFiOEzhC~(QI$;zwCvi@YubJCBqKGtig*|ojk4_ zNz*!g^7|a+7Wi$UjYeJ1?fHa`yOTEgb<6iUymaMUNaS*Kj#tB$u?|#jU=+G|y*lzw z=D4Nq>a0hmAlHI%wN~=|{V+f3x$M@e$kB~;F6P`o1o{1!)EeVSK5B4yIT!p|msI3f z1{V{n)G-{^+RRA9CDzwuxV%)Z_0X4_%#L)7!E z{Qhq=f?T?cd;6~#Ooma@Em5n9Ty`tk&!;O$bn=X%S}ARv*9*s=Ba*WX)5@FX)^<~kok($Z%$E7m>9v(<$B2@+4i6WmQ*`jeM2Wj zB}UG!*EqPUm2H>)B6*U>Vkbu7;SMOK6bB*P#oUHSImdsy>34xNBU@;Wmc1W_XVK9N95C*P#o~tD zpc19-R?83hg%TzbV+lM+W|&R*iY2yHp>E=U>u_Y!Suu%>uZ|(P8uuKD>XgbG3x(h4 z;}`Cx&+yq}o8Bq-lU0swVw?EJN+%Izy#V89d)!JbVfH>)xm7RiP596F-#4#&D6#N) zWjS?@PBR3}f_2}&y&mjJQVOFhU#VxsNu(K2hzjbU(>j-5ismKw@)HxS_F&GAx)ak& zAeBcr(OPVQa*2Q_vnIGG|yw@3=$%)<%H( zxe3&2``p3=_1ArXViQ=Wv3Rl1i``|Z4(V6M-%p+DVK?H@H{P$oEJ&o_H2s&$AZ)kf zu0MP(J^Zfsf9?<>N#E1PRWG)@)Y~|0yj>8xcQsSKr;tmk{$v-cxRWb~&lrPg{RatE z#8so5%Ipncdk5&Ng0&J2bM{2&w;6h^IAMfM!FwCV@mzzzv>?#*C(Nro0*;2@Y+6C~ zB---7vsdtW7K``MRQ3>>kv*ov+bxdCG%b}P@w=LH%@+c5uPI2oBgI<6QSrh=9@6Z{ zvCGlI&J}n3a*e+}Ssa0M->y0nyT|0m)-t9Y>TgO&r0!`MjrfeZl-#hRu@JoFyqQ-2 z9yrtjAW}>yq!wls_7ZiYg6&?sz2lOJ9gDZeG%MD>5N?hM?>i{3EQ@~Yur5sd^(C8# zAZ0l2w_M;STSz3>YgccBtFcWJi`3eBg}!{B;IlE4j%!>atF0|=D%|W2o>05)DUKV1 z-Ro1Q;AVe$kJOOFNQsH@P>widO9~wIQhkLe_dU!^oL~{U{4CV$^~y%GL4;G%{J0a3 z=TrKyx|n9{ch_$>u!p2#1C`!$U;9=OhVq4lvXVAh$8Jf{Wu6|rqM{+h52AP?>Z^4O zL}c;`m+zs5ar^m7Oua7{)0<5%|HkSBAZHV#{>+Hm_^`@d>r23G&S^cZThnLU|FwGR z3(cTp(nwD!{g$d568BYUOy+UR-#qxV-(W5ntKkA?@?Wv~rC6rx%G0w(W zb}g#?AjracY|N$}yF*~Kg<=WMjLoPhY=jaAr-ZEnY?^vt5|WAUz0_&Z#cQU3cLrIGXm8#|^eF{? zWKb<4zrqd9TeN-~oMyvavTACD5z4;@t*$B}YZFPitRKHeAX%`f7oiD8y5P4S6y1Cf z*Gk`CL(86$pVG%`9?Cv5v2+m9`gYl>^q{2UP(5~CVDIuE7uig39!ANqI1kf2r`K5# zT}Z7L^1Sw0%R(tLDnQdMAl$YU zP~~p1QcAQrZ*$o(m$!hl>>_tga*mceFy$Owz{hgyx85dZbLnWCCM(_j0`%F}btV4Y z@c?1|KX9FScvc=eM7J+wP1uQq{o>TqOx@T48j$>g>E2i*YQm+U5lbHb?11IaZ+~)8 z!S)2YAKjSPuyLvLht+>lAg<@bh=+u(vtRg#|BTclv<;Mib3^g33cfvQnGEJ%)2eAp zJMG~*Y1vI25ada|HL59%Vb!T;VZW_YkJP;ZK2);UNE31~u|P#966< zZyiXU=E`AILUAK@9am9axZ&sp5}K=@WX+B^1Ajf0^V|&39aQU703E4clBEfllf41= z!zPVBktq!^wyoJ!!VdNty6k$sL6VtrMCt=;3kptchK7dAZwXb`7w+y!5Ywx!_LegB zv4Xu9C2GtrQh0ksEJ#_g0{zI4vZ>aM!BP#K*|W9;Eq@Ry{1MUjpk==VL*8gwpZ#!0 z=b%1sCUpAaPjd2?+Pb#gGYLbQ%lg(K7r6ST2;6~dcK`0^sjj!sDG^xZ=V{N=W840Y z(uw*Yfc5=@!OA-e)e;{vBprmlW}>Bv(lL)ci0y2LJlO<)ot+#&Fxx^1ZL2ug;6^6p ze`M2cqNOALa0)*zuiavue|Z{osJYrpX`nexK=dSzYq){8F;_*7#LncOxBJ|WZY45` zl>&<(O0J?w;+1;)x%$Vc;!TutHrh6{qrI}9$xAM+Y{fs3^bHok}Cx8;W7$VO*`|2C6{(Tux3t;QT=Hc842m9|2tcXJL@U1F10_szbZ)- zZjeY%EoJ{?L*$Wqo!U#TAmpw3Z`(Y^ofV=Jz@z1c=R)Q7s*H-8)zA?h5X@GgQUeX= z-uhY6N~N6x;0>6gGmPr*QzLFF?yY^V<)iIdGG3Wr^;7cP>$_F;_RA0=_{su^o*oY-#V#oZ4u6IYY+JBQbsLwS zozOP$&eory$~VUJL=}89A45XN=;Zw=K-rX1&IlQmWMJqwPG`xD>A(h?cQSYYG17Jx zG@GDnUE5BV_6V{WZ;#Qy)e=j*vf1-}JcxR@cl$Q&yay?dbWZp*{?rSa=%V>&djh)u zvht6+JwUFhF32L{r8w^j-qrB1RSvG^h{zDLM|w5zD(t^y5-E3}yZ?fHyL6b(uGzsX z&#!{M@IELC#&TVYY{LSeDO1=+D&}t*0L?yLp8s%fA+Z0?Lfsh#O+d9;W$sf8JO|oW z!!IV3Zp3~y#9Q^7iZ3&%5XcjlLW%xZ%5NO&9mPC%Q{+{uVb%AFjJYacWvQ$#%L4Ci30g4-JO}PqEJIx~~LCVlZf;tmd8)ZqL)5 zWbVDwomn^pm&kiWqGGtunv?XJO>vi__A*gNmpMp>*a@?ZKXxDS-~2QOf@*9q)!@F; zrx)QWtmr-}o*F$M0;yVB*x%HOA>k|`wT(3l-ET#TnIGd>M0{Q=m-D*`-Y1RmG8ycy|cV#!7Q3_X| z!yzpEtF@b`A^5xsy=9e9<@h}888~cG=f2^n-O>ei3T}}dY_Tq zQNF|}#sn$$Y1Esgovd~&%@lcW6;)TC`Uo`Ggk2PP0A2jI_gG%5;@E#ODA(Zw=wcjY zl`Y#qS9Y^jQNYiJ-_$iQNk99D}s6t*^xTk(0 z&?`_;n2w@dPlDt>gMUsG0k_!-d6N6w$j++Z7p<87N}D~KWjl=5AdmYZewaDTq%EaX zn^)OO5YVnr^P=3U~^&#yY}E8u=9Yy#X;0k!lPALd~RxUs`b3 zw^-QYv2>w$ER==t%g99FA1BoYe7JH>Pj$RTCCE=1@eavI4;wtky>ElxmOCidup3h% z)D!b|qfhR$D7Bz}#3A=eW|R?@-Kd9$TY>+s8P8u95a$nnBM!8*H6-w!QN#&8T^E$^ z$s-BGBQ5exhPh!4?8KNuikR*x5s#os}G z1i?Xt=JIpDjV{2dOfGR`4o>QZL=)#<5php7QWC!#v|Ng`0!D3lucMSxN+Bl{L6+`c zkZ9Ho{--l7e$Q*hbbe=T=JRq$(}(LM6{XTm4uB3ep{Qi&a2=~qWPRuPUL@8}1*nqb z^$_+f0ZUHfyA~2z-0U}4QJ5&@jZw$prMW#Q6i94R->KHRhENNE%^3w!4f$?fq2q#n zo%_hhId~`1YA*+sC1$|(r2*LwTVGjJvu&yuvIooP;%CCsMC9%E_+G7jSKoa_u1OZ% zPh>C<+I|zQgubx^3n$htd6OrB;oiM%aM+-?B`%C#pTL}Q`0Jo@M5*JiRM8fbUXTj> zN#!n9a1EE(dc3o3Z23yy^e0E5UpDrkbMoo%i?>YEIq)X)LU`cn9JF-2s_{PjERIBr1%55a z1Akok-ln<(}%xTwRztghGe8C&*0tn zPJ{ znS#W_+Iar8CV?X5n!vYX{z&zTj)NZd>#Y4GX zw1G5Dd8<|%^C@69QG`)OimR>pH;}dK3G4>k3w1urV`^<};S^wq;fv#m%q0Yjl3eLs zS*f#NQz+PJ0EmU!aLe1l_d$Xk$-jgtf;#RdTyyN6XH1kAqIyQMUIfV#l6faPm8`EP z+llt~onF&43KQ;ce6Ot9gNT8`)7YmnU$82ks5ZE^Ba(>72*By2b5_0XGET}5Y^@UwUK!h27m<90);f`NQ}UWlZaX&HNQ zuJ(evoU~kL+SLGm=>#zum`EGkEA2*9xu2F--6y7trli0r**;|&jrem|s9mo34DS{f zB>~9l{7PA|Tx!PDN(wX2QpU%UYr;C+k}|P==Tu)FJ_f^gBP_Y z-`cT5OZzhJgWI>GdlM4-uDO@FK#;fs1c}t6-{YD!RJFPbFlfJ~U!BczK9`(u|0p9i zP5pD7AwrMS%3-!Pi>Mzb{A~>IRaDMfi+k@(5C@RShnh}fZ_Rut}<{54pO|GC4S zviQHrp^GAnEYJFx@~1-tdHq(#jdEDnMS$K36to7mk2g$%9xnWWHw;owb1qOQ-As0$ zuSr_JuC>Tr?xfny45<|6A(Es=gN``u#d+J8D9cEZvMbKF#`9@-NK26;XS((m41Dt2 z?oK#`akePY!X1G6Zi1?u@?({Bs9&rjqEkRW=JZCd-S2Y@M^@uS!$N>Y=HWF(1-37(X{q(-S$m3Qymf$>Eegd)C?y=f2JyV?`YR|_<;v9+Hp_L9j_G*i)Yr z%M3QrXAwTR((t=G0mToju)%021-@~Km(p=cSN71s= ztT4hbcqNF3DkfFO)`CJ90KyGxlu@wo6d5G;N01)Nz8MUN9pgSKq;0Il@c z8_|Td{|_y0!HXNT{`G9WC)vHWTlXw7LgR-Z7_a(ot|}7PVMjUP%+q#=!cnEENc&-m zD4@K1ojga66x12Whv}}NN~T;}0EMLxl_@)unGtE4O*eRKPNULkr<$)2>Zj6+;Ear_ zOWpHlzh6!#54*SZgfvzlQ*PgW%yK<%#C3ZFBH?Hu=j=IqVi_%*}g4dY8wUrqS9G?Vx{p&jil#UitAJg z$%0O4K>*a!KjtQ;Lqztz3PBz?dSOIINKaZ9WO~r61aZo{M357#*xFH#$}UWM9B)(q zO{bC%zP0S|WCY*Y29(vpu>n6FzsD=HOHzTX9uvMZ%jEO7!7!tpBw7aAgA{+;^)0+< zR=UQ#mz!BEFuKSstGq&pz8Ll)9L*?*P$5A8bmL4FcJBUCkj!>FdFx_~ zqMS<_g8UCu{WevJA%+hx35;nnec=R=Lhqb@8Gl?-EB6q(U+>+lZtErHWqN*s(@bw% zc(0o;6AJV$ogdKIjRJOAeitjydEY)CevdA=;xWzDD?%}(3|Wf+uI3VB-#P0&{qeM- z5cj*NGASB3vHQF;!Y& zx00yun2^p%f$;5X9#ma2a?&)^moo~4NGp;TFc^-AxIlU+yh62x)zdNTidOL)8>v)q z@1)apl}j`#=`YlLNUw4@=(0ce?!fskb847%agUMaGg3smr0+pjHbJ-q zaXb(11kidq4KdiaE@`}W({1_7tG6v#(6y%H_(iCuCN|d+=zfX2SdV99t2Wa&7MTTK z>8-Z08_Q*d?oxdlhWc{}wcB;HrWcA-`5sz^_MdY@or!)-(~4gx;M-%gf|XwSLC z(;bD87!r95tS=n$YBa|{*r2X?^>YEZjJZj(B|*VZT8T*8W1!~fcd9QiMCEno3Gffe zv;vKxdkCGb2Vnf-(I3$^2oQPDyPfOXPgpbE`7YZc=NVW(ti~e3kYo}N;1D>^<1)Jo zT28-2f2M$WvsjAU^y|VNPrkqa0X~iW2|@gitJ&9G{E?p#Eu_hRSx(I|s z;ATfQ^@v{M*t-5!^Rv8Q_7_<_4j2CHUbMtmSb4%WE+y>{!(jLd)6hTK;j+yR7W+fX zo8iQjV;vhA<^IzQ>zp1(RP2Ac%hrl9qdgT5=w@`p#xriuQFz7PwUueV|82~y8ljMq zv|6!{^KExR43UH|T=;;zug^4bzl`!6{vs1&P7y~YkDsor4i9)6RrN?&6OnsIDIE^I zC$iI&|8Xa|$|MeEWx4)vJ0-!!NilNPUvK5T>x+pIxJlOQBlB>ORHyf&VsfKqCn&&L zafeC0LZ5;4t%)vE{!hbcDlKit(fXo+qSpku8}J*#!rRz>pA%UZie+52S{P-BA{sG} zJn?VF8h^mNaVauZB-lDylhqS4m-}@mth+m>Cak;QthB<>yShbXz0uL<%!9&diA(Pk zSf*yKUUlJ;%0A-{;ED87)u%768E493S#BR{vTL?~6hvL2WshO;?(8}e%DJ-JuwLzIJa^nkKzk};>2@ZetY!0x z_vvybe^H8N(C8a6EMje-+l@WACcK5r=9MnLb32CUlT^Vo>%{sGyYomM@GoZ~?qTXV zJMDV&@<#yB$Hr3pHL>PmzS%@z-$kZS(=AlKK$uMr0lh;Kave3{KitM6=*cGXg`?_* z@5GMdEaEVAq4Gz&N~OqSly<_#f-OrM^#SQSKKaj4AVM%4_2xMLA(}Vo_C)tLuVDBk zQ{6?LZ#@nl*5P8&8eLBCa8Ls8PPMtL2KneS`{L*xa-;dQC;Pkf+?1%e5LQy;{z}a_ z|AkYsfM<%7`SgJ4u$rd8JKO}lms_SXH=K;3cKfx)tNz0q0a^p*fgR>yglV#R+0-`b zO%ZE(@*+}uH3V{9&jZVCc(ZOR|Gupu&pMtF_lIl!&QK8I_86#N!qfYzo@cPdndoXK zs!`vEYq~!VCEn`bxkZC~I!@^;H|V0iUK^)0UH_z6kMt2PP1engdF`*_jHFgnL6QzN zYi$E;do?QExolO)xkngvcm=A^FHj-Owfonpx=|EA=Zm(dQY&epEhB%WM)`(j6B6XB zLVd87AW;?~7X&qN>64B8W>-7AaV;v;OKrhX9NzR-U@#GE0 z6LB4nef!BCGY@!DIrFBez0xoikK3#ltua|;lIJc2?u#!RE{ui)X+_V_rA)T3N&!Ul z%VPyfN~yj0uT-eZEJhM4Od$}nNL}EUJHN~N^L=#i0iL?)k=Ui!NhP+_9*pdyD`Vb8 zJbOmUW!@%uJE;j*T+erT;9mNd;C|nDqj1s@uw7heaX6yC5v*0ai)_u-yXN31xj#*3 zvK0X-#pQHoN+ zw}RH?8>jThOHy}G5N!912SE4`v{;|Gy?%eadwyV4VPu_ykJTG$0mR}c?`Y*kp2VwD zxKV~7pO1btIf(dcipLEUT?+Y>!uo>gkJoq^&Gu1@1l!7`JU=zD-tL^4uSRM~+vB-3bRus+ zFb;A^6dV0%z-*#}oJ0UpV)qgCvsk_qtpAKXi(sQ@7(--uzq5)r(&Y!>MKDSL4wq=I z(Wnd?pBj(6q&ysFt*&GHm|_y=LX+I^KAHV(1FO|Ep%bUsJJ58w34L!`rgnrM>pk@c zUaTGK?#t9x7c3+*550fM2KX}hp0<#SxB0-6I#t9DbSCiMdf}HPEZh7}45hPfQ@a%) zR|F;M&o@FrG@N1SDjhA{etNfiJ?DGBt39%pSctz$m1yT?bZq0<-@6?aDpx+6-RnnQ zlnloWY56mOHa;~c$Ub4?lHN+?sXWv+Bqe)wn6^L>kAE zXyT}0M)fV3`%bREa}D!&F+0-fWWEnBH(JalWrSTZcSizNu1)jNqnS19IQOXf7C51B466y?BP?XfVIUQ!uZR5BH4gb$_QMD{b?%-Vmqth>H5 zkxpTera!-WE)8#El;Ul*L;XGkjDa78{LR+{=(1Gt4XNS-bE?Sm(>mTI_U-6UzuJOJwPShlf% zQVfl`2z)~v4f34XlmLr0oe|Xc+C570 zeby@LJ5DRvojY60$NGS=qVR92_&9k-g;O}#eq)xLeqgUcWK+UC*a>L|l;Uel2_)+$ z#-AD)mIUnHlWSCF7n{WCbMV+3iw#O0Jp~77&0+&ZG#Qtr%*h9ywMJ_ z4}tzF7CG8|lMY#|sAV+B_~47Q!NABZ;XLW20ewVu2D6U4?e|G!ho2@IOzEaR1SZ!w zTO|xwMxfJQfH{}M#qZ<-V#~{!OSG`b0BZ~dx9gxHW=yxQ5Qp=b|Aj0rv^~7`XmWH-}(k9kS9mP*2NDYj=IYylj=^9?~q|)Ovix zGM1umDftsh>_3S0CAhUY{Eh#~c_xRQbsz37r0^x||c$2<`8MP~LKsk0%Ra(%1jaS1JV_6Z1nHM;0ju51;B z4oVD;TH2P(22TV57m?NB4=6^b;Da9OKZ>D;%-hU}I^>nPZt^g2;&h-Wm?@jO(Nan1 z9I@{p*Nt!Q|K_zd2-%^*lBZI!@`?HJ;E&UYE&Bzp3#nl-JigAlqA=L=Bw#Ukr0O}H z1Y<(V7w(!Fwqi?AHz?Nf9@l>Ok_agjq9*2uC;!Evb}ecd#*t#(^dtpIo3Z!Pz61f! zJpIYf^)y_2b^+roe2=lP-I_Btf{yz)u(AaGvsa$*Y3mb7Y(32oQ`CxHv8r?k$Z09=Aty~%6eAo4t9e|?3b zHFr%osV!=G4_TKIys9&*RzveNYwQnv0tp$Rre{f$OKg18!(Wiu0Uqm3XpP@I?5~T8UQm}N^ zD1+4ma6>RVWeoL)Cu~4a^yZR%=lgl*1@1(@C(k=~)HZV~Sh@W477gP;KYjHx>fc~obKIr-%EMDu9r zh*wkDVNQcZaPE-gVvo>c>X1|18VzY?5sl4v5@(KsJtyK>qK{*2#Ihgcy`8+o%Rjj2 zpuK>?$?uvw#4>u)rVQcbWD)5&qKkMQT_TJ^2wgrKQ{udcq)Ov z;o4|nfpqAfKKrGG?p4-NB9n+gz>_QO}Eu)-{g3 zct*9`>kw|!gWTxfa9DCFV^i1R9U}d&`ah^t^ej(hRUO?A*ctWZ4G;$`qb7Rm;22D@ z(Hd0r?=X*io@KW=Pm6e|;`7V$=rVg-HI|Z<;bsm-gP>b!F19}hU$d244%Qk#z*R{6 z2*4yVF5wv|>=J)af8~W=r@*MsDuW8Ma+c(1E#I_NQ*gdyv2yuAXN8j48dCjnymC$y z!k~LgtCdq%W^CIxkv#!-f`X6ia48r0=8@}$20@{YOKkE0n`O^gtb#}^|5Xk5 zac|m#=Lv{hCPApMe5y>HDARDwFY52u(gvqHx z*4ke5u5Jb&=bVJ!d1i=E*c#X&a$0&ouXwhHsTU418*iGltPEIhMzI4PNgHpb%*{sY z0fl<^GH9Cr?w*2V?kHK|Yc?@XC7-e6em&F5FrtnyXJTCgjcnKAad{(T?ygs!h?mh< z*F!2GX(%ldbg!je>8^HvNyMmVb+=&9dS#rI>5GFib0;^I{8_^{R$ z@oE%#k@=tegy2t5v2c=;f2r1mpu6enN9~UZ9fnVSlQI|hS~#jzoo&`bx_+VPn?3wh z{bW%_0rt1m0tJb>x(2A@8y@@>srg2`OZgXb7(0JDx^14eR4m;rah|~0C7@}l+$+kE z1&+FKr{h(jg0b!KZtv}b9XVt0zjW+c^dtpe4jxtL{y+kj6spY|96korf+*=8FS%GUh}6)-*A4G=C$p@xLaC)X31U zW)5AeHBu&lmYK}%>C=LhmAVj%B>s73AZFM~M@YBeD4DSacA5=j?fPa%=Y$@n%c`%A zt!cLr=~w0hBl-=&$pE^4dQuIG^M~(Tv(-=0(z!Fa+Ru7W~gKxVM?3=|Xz) zBGB0+p3A~vjUtN4E{#%k^kiO8+==VU?^uPQu5Q->Zq{35UYnKcX5HDUYUaZt0qUYE zWomAELyL>^STm;QkP~<$@2-c6XvNz|u3LeDs#BQq5pSrkxA_bxu z*bcy(F8|Q;3Ufi({P2-0KZ-ome3s*SeVZzDA=NI{drhFaW8=fyczBleJ^ycZlY|ix zgR>eV=3S4#TuFAZ!ivr?a!N+MZptE^u9){@H0>IS*rjJ9Ye)jx*fx$3HT~fe}g>;Rvyfq_pel~j1N{o zFrv1yV+WzV8adT?zRXJuLx|JI%HCnAv0m!~Kjy>r(U}_me@$^di)o^YLWYgN^4p49 zXoqm#!dJ=+DZDq^`v7D`P^?-lZhb*;a6A+(%v`%Wy0GILc`ee}(H_dNM+oG|d<0Og zp4xiV?DW3J;LO&isAvA@CXLE*`18{y#9H^>M~WTn4Z|GLf`fL99Yba{f8`Zn2x;p! z&3a&J#jw|6Bc9BU-QUc$g&E4)so;Qqx60a;U#!{6C$+aWiCT^#b0C}ySFYesno1*7 zkIT$qzBrGq^gi6Zv=DFh=NUYEabXUnli<1Biu*q15h6|aX4%Mg|7GO34~dSVVs|_@ zM=kpqtq4-FrSFn;?vqAci6pv;bx6q`Rv@)7KjleM{@b51tiy%VhR;%iH=~5+*CR*} z(OP6!kAnps$~a~<&qJH4oBvuiV*)dt13`JlsTUsywc0~C6S6WEug(ikA<~ZQE9dD# z4a~}gMLQ%r5mw`fcdtPd>2c0s@3>0(j?V(&z8Fr>tV@sgs_`aM%fl4PcKXHK>fv!} zYy`gNS%DUjuJ(P_PWVSitOl6R-EOB?;YG>9PykL4ZZE=Oc z%_uG&i>@mVM8}7(WGA}ul*4D`)DSy8uIL@+-xiiiv`3rv_RF1Lj=D?p?x>Wu2>Yj;bM&Q%x``N zuipxJcv-W(6?g*PCcrO*r^126ggf|kt9yUw^1F%b>!#Mwqu8b6pjVrVI`(D{Vy36O^Ys()=<> zvhEf7bjUr-JBzjy6(CT=!bmG^YZ*nH-Y%&KU(TSA=Z%e~4Nu4F5UN!7F7N(uxc@u$ zatFjduxp^p47F>~67iRqERZ8QH5+DQ=o9mSD_;HI10+yU3_so_0gdSP*HNY*>-HR~}gVbDE}Is67rYN{~M&?$F>B_1b?m*#L(5OR93*{cBXPCT?)ADceECH#G*UjaS%2K-%Y#Vw!P?ZjTtGUYe>n%>%fHgN0nd0($>H#vJGOt|D+6q)~o%J-1nm%lc7nme#zl=o6=0tohn$YA9@+ULupGcsv&PA$Wp2`E8GZ~S!?!AHBg zbqh$Cu)oonJCt3#&Vn#VsaIF(Et$-Q)5-?c$mkqTX*t|{?7xmAFr`ohMKhQAtoTee zb}A-ENhaCv7T+~-^ECWo4;#yDeVKk5V>9VLFsNrqJLzq(k-wf}FEzdHw+P&Se&_y}Br`h3$ z_Kh$3=InsQUcGRfSe@&|&Uj2sVM6yrQtjWw|0w5Z87i~LFUDwe@SC(TK67 zb48z}S8Kn?!BEzfbD;no9aye(SjpxF@W}1(xQUj32g0ne{6)5ZiE<8F&&Eq~kj$1# ze&_(B=*ByIV;&n#3e8!mc_W&QNE?Bn&sC$Bj z?Cd4(rzO-OKUO7y>o+{W*g5kjzHgk|*`fJOsJ;WWhlc*iisZQF2pzZK@a&I9ZXMD2AR@c3bgs1M3gDV}*=7_g$$*~od6CIa zBwIOk1kCphVwS8o{^w?oO@yhf^h8P{byA-Hswf=zrP02|s_Ra}KS1uh$6J2xk0NGf zq3hP;xWSuk@w=7C5Me;{)%Dq&Oxk_yxIKy4B5eL?0#q$j6L0a0EU)6K( zn3c0zgvhBm0Sf|8#DA|3Ih|2>&F;|yf1i|OHKWkHZ(g#Mg!%%(j5uu8SpI;G_aDx8 z8~oIql{Qnd9+!J11iF((!6TM1G{61Er?V(-_Cm29xiQ39gj!tZS_h1%hyyU2$N^ai zc{TRt!{k}Za&yba;ygiZsH(u`<;>+c*jQQyO&1r2<~HCC=`>TsN2y_Gqjy&PPteuD zm&%OB`;_^f3nfH4W|~$k76gvHQ~69`hJrEVg^emC$Kz&;5#J8i_ndQB^L}F}bcc4c zV!U1-2`$&xTQ4a;)YWte@j2(HDlG>pe|KBZ4Ad?fO&4oL(d`S{HVd9M-q>YLO5(LF z0bfD!Onvc-4&&((h~{8_L5$OpS%8EvaT^vVQS(zk0ky!1CM}(QS<-B=ii!vA)f2M| zcUQm5gz#&#wH9>kAjU;!s8{UjYJEUAp@4CZ`~1QsnNmH+mC4Lil-4Js0h+N_Muw`!z)HaFpXR2}Zogh*|o{k2OQ9O7lvt zpdXH3uIa0%Je`Vkv_h3-PteB)j^m$F+9t8C6dgIsLE*y4b@J2>O2cjL^ZC9wJ)hj% zSTv6UUuW`gS4MFt5!W+r-T{ldQ$%@IHqNUv%Ox;&9bO`%mUv=lOTAUMr!+GVBl@+x zv1i|iR9;pt5Ed3HXdcmJQLihS5B%{RN4C`K@=Vl?tdIr=lwul_X%Nr`9%pD1HzhOsjt@{EfV)di2ty8&u36Fz7R0yh42 zIzsG5Vcb(8=Vk(oPU?!rClW$841enmva9=16agJvI3bY#kG;1Fsw3*! z27?6+A&@`_9)i0&1P=uF;1b;34-O%?ySv-L9RdV*_u%e&kZF>9Q}cc`7jLTmi@BJp zE^g>LUA^}1z2te;vly_U$hgxiZqD?!<7jPZupfc2LZYeu%TrQJD#_lk3)K3S_L<>Q zvUqw2E0iZ%y!SZLg-|-?#=o}A17`jts)-MQxDW44WE9k3(*j`;0Bue^L7Rp#ED=$k zF{g37)18NuOgyVw8thFFSxsG7*b*vtuV@E&a5Z8U$pQ4DX|qW_?yy2~^oam56aQ-c zSkkevBrcKdRzy|PvD#$M^||vF4?3Y2Pw-!q^Bvv)Vp?(A;~>8{YC20z4eo}epRmm{ ziR@GCCQLuwRI@aV>CqRDi$(7nfFjMe?T9I}YdJiR;o723xUjoMH9zinPi%H`*m?vq zfmbgu+uXK!pR(hRNR&BTQ($NLD3GvntJu1gNtP0=r=FFI6{MNwcAAiAMzwBA%xS`2 z!`DP0(a=tNSN;6KhfYzUzs#^su?7bJau8;l~i0%vksrR;%2-wZT-b0U(W|;SrTno^#m~#)8PZ|47!Gm z&aicNebl!GV*TW+ga{aUyXg_D^3^i;2ae0Vy?yJ{q;@{ACFork)ej#bJ|JcZ=QmVJ z3Q5&|&m5%Ke|QOBm$lreA{FtzyWQ73o91{`7DOL$Sr#35hN19ClX8_Lu38=*;t1O= z7YqY19N){Nv&OJ9o}l=4m(SRS7oJJWdOuu)6z{UscWiRJ4D27(&Bo;Vi_9N2|BW@gbB!`DE| zYE)on*V{4Qm?+V>4LY*%1dAEx>LvS0x}lQOSmhqg(k;(18q3nOVN<=wg?zy%o$bme zY%h>~x!22{1Hruc>H(8OEoCBJ+@irWKEOKiE*Z_ZC1C*?_2MySHHpB zW7DS-1H{8dguleY9r^#n{5KBN37>_$(!ro#&hJg6+?t#$FqUae4aT?kpv?}J4S$CC zHS**8RUe65ikP~qXtxhZR{%i5Q_S9;OG9&rXLB0zoJNOF+W?_C*A?^x#+otp+n$hQBFWfiAf#64YhVWi(r_+tZW0 zpY4rE$P9Z}cPvIS@~-t8o1wKl%g%e-f)Z1dj-7YLLIq6mEP{y4`@1_6GxE-#{$eeU!Ebzhwwev%UDaM>2}5rb)q)0 zP4)JHPXGi1TsHB~B&sGF^CR}30C%d3maG79U4vMSN-c%0MeW5t+1|9}#MO%7&-9kr zyAx={>sQ*XyzvFdT!oLr^I5qRKsRFBjD$PGjh&jTgE9c6r*p}Fomef zkT)Xo%WZQA(b;EcSFZozszK8KldIN0&+%yma^xt!MmpVBW1hCBf`rNaQHtw8f#V{Y zv5(hMXSQMu`AZ&Y8T~3;07Nk0JWeD|E7RlbC6E!PN@`2L;+>H{eWf&4YT}U6bUBwa zH{bgBP!z^uq7X$9RXH%>$1@DrVg9!rYU>$W7M|>|usWdf*AA&bAya;%)YZS|J?G93 zjBK-6`SobVoQ-K<#%UPe5+#*VaqObUW68Zjft;)e53@ulwF3Diiu5PKZ_gl-{B}-Z zPv-PSEJuB@Syj8#)jg%ew(tQC=}wx8^cj4zw5x8V0=brjWmkt=6tG$T=J1i9Ttc04 zJmo{R?XmKUyf-Ov9Xa$q6`)<;7M1DGz+`=0y7N07=6n2rk|}iZi_zfx-o=|LgKP=d zq%&Cv`xG7UuHwAjNYg_pe4la(v3^XhDYLey1G#&apS(}%1`qnr*Iv!5^8>Hdx|7ae z*?W`Eg0W$qc5C>@!&goQtLXKQ4-u8_J)yo4hgHM8W2?D18?mJ+=g_JGVNLwd_|W@o z&Ys6)zxOx4r@aw+`Q`gELH{K(d3;yK*tHJ7u*eAF!>GO(svo6Kg=oM6e-9rSRogXu zxMUk2n|IcAPxT2gVk)-@BCpzIXK76*oqDsy+d#AXb&Q*xJgq0tS8g*}PEO=tu1N8W zn`3>&2e12QI`DJA7u=I86-C!_EmzRMqrhozmNXreMDGr~HoAe#qNj6%^DuR!W$V3^ zlmb`jF!z~VPDq*E2yv4rG&Wf|n_CigzyR#}TQ6Yt(P#f6NocFQH0wn5x#LXSpdct< zA%2^}0?qRah8f89YVqZH(@0*LJFO_*mruipDTSN)aDP&n?dNSs(z*xaPI^e@zRJchjU)iy!Kg)y4?`UKL%8$*kS*T- zFS3Pm(v!*l;Ubj$`%?xEW1wVcDP!V+*=OocrT%p@3vABcuk+`n{+qx5!wn1hU)-<_ z^QTkrA1*KEzg+jkSuuz;5_?Vvy&;U%6Wcwis$PHlk7xMr|Nfz-?63*^CzcUc3mobH zU@TAAu@8S>5n}_ggC2^NXx5J87j3_ZE<>BK8Go1mzp>Qsd;WUWWrY6>GS!zz|H)WT z=%1qgSUx);1x%3VDL;o`%lhoM7vM-z^Z(7;`|pkQ_Z1UB#Q*>1`fsfE|MAWBNdYQv zvC**~;44I0w%ka8Zf^`ZPJa>u9Lmrb3~BQBK*_fC?rfN^(t3Q4lk4_2Q26|YS1#Q7M(i0)vQH9F~)DK{GmIb)zJ5`6I7R+XVG40EllW zMwMlFddQQv&qt3zYL>QI`=|Rs%S+str<# zzcflM_oR8lA@@Z9tydn8XJu}!wt06K(^{2}AV@!pvKC30R7&o>Y#W6?)jh`Vk8Nk` z8`0DG|1THpD=U(%Rm$Xa@Md>2(Cwf}YdZ5$JdPF{!$hD7&FXf2eE92BvY{R06BaX3 zr1n{g%C&Ip8XVU0n1^nm3fX;r;nDmtYCNpb!MOFc)!5sGJL{)&a*b3O#qVE@^6HOK z#G;258c?C~zHjvR-T;s=B0JGTU{oGuMoyZY-0o!a$03frV3o>Xe=>EIrgaYIy%a~i zEAo}sf)NjazxTzt?GLnnI#+?4|HIG2CVx-0%Ve%UJ!OZ=JK$;w$Ce%*6(iRu%w}#V z)={@HgCdJBNMf}2-@ntzyA*`VV^`R@|KlGtEv}jWZ3)7Gu={t%1L9{(spSCAaVU6P z!W1rWAcmF8&PGh7@!1~dZP7a|QW7^}$(sMs0{D&<^!Yo@iZ2ep{u9^KXqANk-4eKk z=b%5q2_2hT;=h{zZ`Hv!hb`|&unn z9>j3Sn9;O&^9G;&mD>rkmq7MpRwviG^7~mA$;V><`YiH4JS{TuMB6dx_fUBh z@_GG^%N}$AcrM?-mh+ry=sT_BLRxH1sJxdoZscdFUyAR6$OTo=h3|I=lpTk|EBQAK z{ck+M(>-V+f6lj#v83@U3`BMsjZ5#Q`$JdG+f7NFelfCCw7A!0p5uV3=GTqg>baM2 z$-iWBt~!6m5H9trd5?$oN10ici$)?vONUkB5u6{j3L#(fo{L^-3XX4&cFslVyE{EM?t? zPv_+O|5N8Afa;)W&w+1VI9uOPabm;cT6s>CXvm94laoFbf$LiL^3vAtuWEmu#wqtg z!MzM$_IT8Q+a8^%x3B;4knnu@XCs}6>6U6Dp<8+H)WZeE_5yA-I0tbq zIh3cS9Pkdyr;At1dwYTby`xyo^?AwpsT-IySO%17{K@`L8JiwBYcqWy z006hKc*PrT*$R;Y0kubtN*N-K!#cfemC>m6HW=i-Fup^e03p>x|0nQ_5cL=P$8YDD z;E#!K==2jf8eO;17z89XzeUH}+MeeQx!%towZ!$xr;aQYDio4$P3IR5Utx3FZMQB3 z3|h{#1?{5WuuKoWT1~LUb-mcZV)@R!GyMCrO7;CA^i*E>@2@s|p}$c|2&+UJ;?5Zo z(~X9YLmsf;yZaMGwqDU(TN3V4P-G!Q7kt0VNOHr7X&?S65`seiCh+GWnb{m<_ppfU z_rM&>N~fJOo3TLQY}H9|YoJvlHJ#jeY5}g&evfQU`YbmjXIqSH*A|JrOXu@Lz;9o> zZreuouf8^t=uaBco6SRZM+)rak+G~W$*2^*yYws?$IAp8{*rfq=YhJPM_|7{*-ER_ z?J}7?pToIzSB3X2iIDu9xr%&@*aG5FfK5;L3L!ON276ltt73Fz~e6Ch~OzK+&=ZU*=G>#83VY! zZ@*7S{W-B?k$Hvxao3^6{Q10uBps!0kO6oySqo?ZZK`wgGh03p)<&12{UShHe=WeV z;tBKX4mih9MNj|uhYx8my!`+8v~ZL^dt8k4fMbTn>^=_l|z=|7$Es_}&YM z)D@RNdm1;ZJp>zdKuR$#O+^J%spx)Zs*Ux`aRd;GnO5v86v}8eERIoJ zawSB)tTBs#@;MB3{d!%V@zfW#R=SZx!1s6>1U6qLiOnulq!gPTV(S8JT}5qaI*D;Q zG$=mYY!RfyK;0jO16^?`%T+NBx6{R6$UGkTq7-1Ll+J}XcCs<_!aT|*P+HE0Ah(VQ zawHq^dSQW!Ylw*JZ=ivSf7nxgK(-aH7aGZ!^LZ;K`zBYf2TSWGK;SAoO0n-YwwYNC zxVc>uM*nFHy-hcm3JFg3&}6AM-F`M(`_U^o)TRve@xXCO-n}p5w12o&{7920yW3gu z3@*$NlQ!bj%g_h_2qJVu-Jv5(aS~|xYwdit$%ekDWD+eL%Ki(1<1#*LgFHu* zV^=EM7wYsm;(i8f4d}GMOxdJ&l`@)+3=SDT{3LmE+J&{Y9cTd`wwXV0f9mrL4JnB5 zZ)vuSnkp$BJXrrq-BMmr+RrGpiS0V*O#AC@p$rLPaEiOnj1C5Mk^&>m`8L7@QDbd~ zc$q@NueVb_YFko5Tst5(F#&h#T8+Blah<)NFYHwCBH}ieF3(TKQd3)-?j$#R!1aIh z3b5t=6E+BGbl=^1vtHIZhGn(;yEF4L?f5ew34VK$Q8|0PSi9V%OQ-dKJL|aYuGbpH zV?173)dY4E-mkd7oRqg@LSbT}B!=Y=LHd)AWQHzz50^6^*CI&Pz4@3DKnO-bz~zM@@&qR^;q_8dC?RkL0Cf$ z1c7U4j*s`X3I$4{^NOM)fDi`}%~Clhg+)eKrUmv3_;SNkAyY73f@$idsl~I=3*N%BwjyeinBb(E04w-_or5FUDgZ zXen&ZREg+69T#GdRO%Hw)37DR#{^*pT%eyS4(O|T4N2MW4G|~Qt0!6U-Ev)8lANnH zj!}>SSdEPe#d7ji;KfRkv+RjplKs!lyLSNupy<<>p|G3+9Uo-l|K&-)flMz_^SuD@je zw04#vXDK7f0{Abk#1hh^i0LoszZ%8LoB>I0??xQUx#SQ~_x|b3LKh7LjDpKrVC~ps z7xASj&}LGEo88G0MC2#XFuMI(Fdjuf|E68TS|8bYZLro}j%WdvDDQ4R@|6|i4Yfjk zc<@`0W6NlR{@5Tf;`#?D+k zwb0$kN>-G*?^I<07~r>l;6}=|n}nP0kg{H8MS~Sp4UZD^OY)-O8!~%n=Mt?q_WPc= z0q#r%^1ZIoo0}cZ6RfmHvNmy<67<@$j#m$-YM*4c)Xr4hQ|AWCQc?Rq%Ru6% zHmjCCj?-QPUWX*e94I$tVut6S_J-5td%GKKG( zdQXHyW7fjBXdUSCeelns1EdePc|IBpCFaT3Ip2SG(?=>+O7X8WpWjkldFBmo7l;Vt zZ2(p_3qjD%V<2@|DYcvxxV@4rHiw@>12`F^TZ(0rlZA>)5bX_{P+8@{l-iHl;&0!* z3R4hZx@mvngr81~+bS-v3K{2vy#<&|Kn^vY@-IkWXR60N5~(7}S|68ajy#StFn588 zG*xc5vxc0FPkG0HGS~IzI9?7u!!%Ep{z12p~(j>RXAxitg z=+bUScSC8P6Uczq@>>uF-Xiq-1*EZaJNm)Kg) z*bU(j{1aN*kj9M)0&tL>686c?@66|oFRg}URM+RU5IFdYdYP;wEp-b7cCW6isO~=C zGZ5{TYdp-U5Pkt6LSmm~X0&~%MPGLYW^HE!kV@s_1H_Q@)eX6lgt)E^IxRL0o(|r5 zQz#)u)Q8e|p@X`{?{s_tKJCbo6B}D>Kf#S3!Uu#(`kk;b5!bPx0ei-Ok&lPBbd{fu z4|f7oXL^hC@Q3NX=yQq@iXzsbKYWY)dAE=k7TLORZp**;dA(}5Mr<_;xqmKZn5-I# znr%D*$J#H%UL)q@ZUDe4>SfWHa6}pP0?I8(t2nH2TrFsmt3oh(VPRt3YnLTJNFiq> zH#_DeQl=JD%15%#UeJ$B zwNaeiOM^oM8^Do2_0GYYS@z)8Ixf0Mg&4Dk#H99K@TEaQ(df}u5Z$`k1t^s3;Vtl2 z(pb*+>ack60CK4I(G-SKh5j0^2Yt=fyS(P;S|Gjh?8p0_yc92`%s4wjPN>b=hk=CF%^TH?Xw4@^g+z`PZ5I=Z}|3mii0CQTmWA zuIv)V9F>+k4&#k!2t8b~P=0%_8s%m3@@uO(+a(g)NM5h6q$cBGAw=OHABe+)Juz$o zTC}1JrhB=LTFRJk^l`%<^Xa=jgadFja2Dzg{pc`^w%ZWpQ^&nND)FNcsed}`IN*{W zsOVgGzx-ZP18LppB;+l#HS!fOnLF1F5#3F_#?VVQTdI)yFVX=U#{czdri6Nq&mT3B zz2sq5(sbI{sJjkT5+GEJo-M=YY3K_iOGa}XKI%E6qmBE*wAI@;X}S(bm4rEa5u(DH zHQ4VJ!|F(V*U+l42>~|{EqxH$&$XAkoM5axPDq%FIUWa!5uf_M$HjFgtcbs!8?0TR z1?n$n&3NE)FX3>q)MO$IQfWm2q2=z>0FE$uX$lgWQVU9DjoDAgb3ZId+1nejxh%Y? zH7?zf!vtUQ2$@hyDgb=*OGS#uCn8KT(pFb0)Fao?KHVCA>IZ3V20_rx#q><`lg3jK zcI(Baluab4H{ zUU510Mfape=X_3!3*%ifq1_|yyBbWH{~SdcqQ49c_v_FZ?nn3(z_5>o1cXJxA!AXE zlYB8)IhHE3SQ2WmKR{!0-@M|%15{GvY7gPYYOl;~e!L3k>Cy+C<5=ZMB&8}2h4CD| zb>7IynH|%=%eAos$mMsf$743L;1iY`)tD0?AST~Rb*+g88lAFXiHSyPzV6tn9`CD> zpg4^Ml6|7kzsymM5m$Xi6XW&^TVjMpJ^JE6S8lE@6mRaQP5@r62Y!|VwHD}$pe@z> zt9LlDJXe{NTWa8u*$8pMQi&Zo%Cs$CB(^y({*)5{haf&`x=$T@UMq=aS)of-z$W7p@Q&s|1~w~RaPG}5x4X~3v|2}*Ew>1Iu#qLp2n}&SP)fw;K7GRQ z8uT!Tpw+r;bkXmkbbj#L5dU_G`~+cdw7M5=zk96|r3t27 zOwAwlC0Vu-189ygD%ZIT$ExpIPAX%+cC9{W*k8ldY?m3^%}{ABkB{L?tMr7I)1zh6 zO>C)Fn4E#?=M;4TTBQrN)7`Dj0e6uWh{qXtT4P+j16IZD`+U+hAr@SL8z$3B7lkD}O2L#2EYsslWohiLj!Zx0d<2;J`TO$1l>T?{+B2P^M3 zVov>TQGwm3GI+6II}|J67g|p6l3O&TRVAT&{8$v%ZmY~lP6#}+jW%3%*3 z{l~NSG%Q0!yBGq$J{L3*_!?{#ZQe{yaM9pl z1!^T?OuAMX3`eDy$V1=Km#AsrV1?epY?=~*Un)zTl*qIsyK^1hjZ#Q&w`dSeD)Vbi z*4gi|_U#o~c$+Sri}UkM4BrD#-RToY++5Kz^TGFKA49QOYcqOGPRkvB=^<9@ZC$LC zdpNlS#HAA8zF+dQD7^!Om{1M0A2n|w)t%sBjSI^WEf4-`XW+z;nC9Qs(py8=wJ?fA zvgwT`v1ISP)My%BV_RCS8g9*=8-8iWhm<=mi)k{EU)t)}Y6x^i7rRQ}@{3yDa2;N+ zDcsvNB->BzM0AIfnnnBM6t>JSYLEb~^CgL9|rIKjIoA94mHZoQW|BhsvfV;q_Mfn9XnX*Cu?Z2SHl$gY$@Gcl2WvI_BK+7Tu;|RAJ(`h z1G@YHrr5BOd)8^;5>Jjq!KXEk?BOo9?ZGWzZ03iw{03Os(YI?qFoHe-iZxAPyUROn z>c*T!{f^$o2nPUte2Im!Ma3()0)`p6w)m)} z;_k|sc%$yhH z8t2MmJ6hM1aruo~gIXWrtFAZaEpv2lY~gI%tqKNe=iHhvSUQ-^6L_{(bZL>!d{mG- z)Ys!oaJ}k3kLUfkOV?V8NLA1%bcg$4$Yi7OTw`inb~bx1;MkPlfL-#-$sX~|g=_hy zzcW>F6)0;;vrw^!T9Hb*ucp+ux(p6E_D45jc`0kzQ<#2I}RQOp4DD zyaOw-IUu+_XOmV8dQV>v6Wl0kY^8db)sCoRw>zU`(VSzeZ56*=ELY-?{4DJL<9iHy zl~2uwx0k9LjR(uFq1=nE!)RKoAJ!ZPW)M%?e6j|mmx(dITPS+#9ZW(hQj|Z!Bdaz} zd@WM`*#pXz$q+bQU&EMKH@(wMI$`?N&cJAK_aojatp1SaeJKNRKo4f!pnI7o1||mI zomCO02m`38Cz9lQzx;UK&j=HpX$JseYKMXRWT8o%(*zm?t8L_alGc0=esSMDhMg5E zM~D@w&jnd%g5uk50k(ULe?ba}A7Sj~dO~!)3rj14-{(k5#Uebcmy(w6EMgjLM!YSA zQ)|*yrP>&KlX5r;XjS?=zsT-lVMBneZ$f?7hyNB7hG}%ZROO-f%ys$xC52aMUbPY| z1b>JKZR#La%BBQ;=9BOq_H$)rEu2S5r!`SqCB0K{Vwn%}4h<%97$w z!g4L#=~~jp&Kk&NSof}gkm=WvaOw;klz6sMk1od!9GGQsMbo__C;{VstpW7!l`vgj zJ3PGv^`cfUd6MJEHPvs=4~eiQ#HqE%^G_yKe|9!eHKGusAQ)PxwnFE*Kv8@T3 zgLWnd2=x*kx^guQNUOk*6oCLVl?%@3UVzO<>4JC zA)wpU6@-A5p-VK`08JPD)2y(){w&eOG%MLY?f@5DfgG#OL%jDutJA=O_-@@zHCwiF znpfq!%NO3&OO8g6s^Rhb&88F2qwtKY9X|vv{A9uMlLyElipJN4d{}mP z)D|jI-0c_eQOE%CQ})7x;fUsbK6N4f`lk_x6e2W>Dabw9)B3T6XX7^ZWSQ7k79W>` z)d000hn-osM!{PaWdAW}&*C-0o97(aT@QySTnLzDZX}=SC%7$}{$vtG>a$AE77NzG?WVf;W_8u*SA; zbZW4Rc8=GuJT+gd^$cpN@#tMXEd4MF3Tw)hkLQ7n^{|FdcCX?i;;Ypbwn6?2&0$R# zSOF<3pnN4h{S;L>ON1qXv9~p(!Vg5aV0X0WbIx>)q3?y&f%Vkbr-SWlwW=Wq zL0+;B|6I$9fHc2Qb@t96bBNm~gfKsE)xyLhYhj9KZ(*TlRPpioGXC6pU=)G$%oR1z zs*KHY+F~YPL&^jAq%(;^cdWA}jv*(L*xJRIZB)ih4lx{mO}lcGG|SV@J5Ah2Iw5ML^Gh>|x}>Sw(8iADE$3;5w#gpq4bZV(Qd_= zsG!{pcs3xM@DlFv5nsmge3wTC5!xgE;Fa0JJKlX-ns@@?Hj3JL_NW!RJzfl{T*y%Y z_>A;m2lpzw(sG)N_V&GOUBhl!_I&H{R9p&YwY;jMTKAy5Mm{TfC=Qc=Ayv$cQc{@~ zUmIB8V5q%IePZCt+1Hulfb-`ZN07F7m%GlFJ0YqDsLZn*x?BTx5F~_tgY`m`hT^<~ zyXPr+^8v>-VS$@2cekG|tG21)C~Pa4Eo?MuPMB{!VQv>QramX`6t8d$TzxDb-YYe8 zt4G=P2sm;Y4;{8_T5nJnjJF?{33Pct(xWfybo+I)r5qP9LAoh8=@NJ#cVl788&$H!I1mJ)8;Y0-Jvca0b(A|)W7v$J`a0aJO? zwqK6&P6opl?U^uenLOZbYyXh8LVRH7uD8Q1=?q z9=$F!J-4_J8lIgV)gDgZ8FaK}o^|wrlcSD)xAlgUak?{fB%NLv(S4?a*NIfzumcqb znja0foCTbnNJ{~HfSj+B?CMk z*q;MgGtG_K4+Cc2TqBgW{Uwe@{NVf`bJ78FQP+qGOD|&Cw+?w|!Kw0zu))WKr%gb; zp}@@peNQ55KN$P$2_^4m`GM)@7Yh+Lt;Zi=$a- zy7OPpF^|gU^F1kXMxfzsBdnat?r)b$eOXGUTAMxgp%h`NlVC31+r_?h@- z7*N>yA(D9nk8xM)&?;!_$(n*aPHKqp{>KWaExi$Z?1!py&moau>UWt|7So2q%$Lxv zl7ehv5|!5*EA4BF%-(Lc8%}gZ!36Mv+7uAGkP?AHNun$Ls*!qCu5zD-I7Zq*dcxjV zf0Zv+2A=-@HvInl?2zenvqBsnxXTD0gQnQPb)>cj;1%ZJSPRD5hvyu1!IJ;=B_+=K<#P5XZu|N3*XG<1c zdv;l_edw2Y3-@){QmxURrO|_;%Pt8T4}H^r{*+z2;Oed-mseRi%p_eg^Ag;bXaKq& z8Hm0}lxL-O9qJny++MGhFc9hSK_SPpI-L#H#;hDGpJp*CnkH*_px};hNkO7@w%B@g zJ(TnF9+bSjMAA{Y0nmxQTKm$vn0iIc^H}Bmfz!<3L`&zOw}w-9y96z>NP6n6LPCV! zv)8xpQ=5c>{62qz@>n@8Vzz*?U+HM+fSkI!BK zxZS}ia7g}@l{f2;SaB<;AxrdG!1)}=0D~@c-D!RVz$qYf94_70x@ChqJ~CnAUJn7f zH>GLkaC>1lH4L-(O0mZ(obFguswF9C=E>G7+yR??TA&9-(@G(pX92g)cXa`7EZ^%( zT_sdILLI4A`vV`#Dqum9a~=;jSc-h+LDh)cf}zUhjanhsMR(La9b5x*wWFg>e4awcw`-7 z)imIFdtaXi=|FUoWa|~L2C2?uMBMJ9HHDPTNy@onKy%2wv1!G7u`HSAte5>oW$wYr zp7L&6F&{rjo`Q*Xm-)j>91TDf3%%%UbUEN72KaSzUbm*v(dx+a^cS>%gxNh;0^cRq zWo*%c8H_wO*UV9O*Bx5jEcY?jBx9IiCoVWahhatq?--~5Jt%k-9<@nrdR4@#9mj!m zjy=0wcrn20RN#>(#sS}0dip?2fB6A4b`T`Ax?9-#E7L<~Ixe2L%MEdtoc+8DRe9}& z(OUu*oA{R4KDVO5X*Dfkneo_+R4WE7ebW~r*E2JL95e%hx|+z&-To6h;VSHCpkloA*CWNr1Y zFIPq^T@}n#0kI7CjawYf?FXdpq#itv;KOBpwFYZ8aEsLK`K}JAJJi9A-`HNZlmTo0 zc#Kq4kOtk;S#ZD>Kd-xouh3OwkLFN^_;q*BrJEw(Q*yhU`^go6}yB>Myy~d#qo_`3BXLM0~>43RP)L~TP7Tq@q`JI^o zDe%;|Y$kSrQ!0LNi21ylnG-=BLSEJ*&rYtRMF*ta&t2Kjz8g<$%hA9Bw3j?p$$|uS zwwJ?6OoG;wMINN+qyq4nQU{4PrHSb`I$L3?$Ue$vm`g8~BCKZuzlS0`KkDY!x2-BF zl`(LUir`Zu!^!P4;O60CMTGTOg6kDB7Fg15DM@CugsGAT8e%7rf05GSHbd*{Kuguy z?NK044C$|xqsi7<(Bv@eF{doOj0x|UAj>t?dB0}3lgZIA{%tvM45_^&DCT*VUMR0> zs=NVPs_bUdz*OK6@Meek| zQ9mG6{L~_kr%#=&X}_^ylb!g%Pp}~)qSsCD+JMzk1)FWwc+3rgcdvh$d~-{-8|iSM zYWk)~MWs6wCi;;`@F+yG3Emi&QRg`xgEbU|OWmM}>=iu=yTm$uf^!o3fLCXxYuKSi z^ojrrk8$8>*saOkwHU*AOdzbdS%&m}yeFY|$p##T7#^4XIG2DdU-Nl~fL0V4t@&p* z%TZOj8WPoo=a5sASGYd5(eoD+NV~Wumx@d?g{M#V|PH;mhqW7kL6MB9+DAD2Uq)<)#VTWCn3C| zTJk}rZv@`haW*}QmPfIm#hd*g57 z_~&-toMH7aQwtnx_x_rl_3he+V!4HBrPN6*?b|o;38J0Ig_m_WU<~G!?d(tNcKE*5 zSf69s!+tSt6gIld)B)b+OD9;hSCQxn)nhMv2A(ea9~b<1@<3!XonzamlHwQ0I)V7I zsNvO3yJM@%##xJjNWcq%f_tJDBk2yq8Zc`E=DKm*nCK&xu*wMc?amO4TU}p1QcE=Q z&JTTn&|)b|S^8tZvK9GDdM;M6gg5>^BVD$iw+zQ|cXsb*Hr^Y(rx0%r;Cy|-3M*2m zP(Us%m6Cs^{pgi$K3?>Qk=ey#wB;F#TE@hcARA|s}RHtT@ zXXz~)dCR zrUFGeYQ|K)4OFm`J`G&?GkQPNDrbf4HS}yoZsWSkHwm&&U^~9V%a?RYvO$`B1X>!# zb9+|EB65?|T+gR+DOcbPU;|Kg^WZd&aZsbiVPa~OnOOZrstV-04Hv-=bN0f~tUn$^ zh<*fLBCmy*C{p$?E6$mR7S#-ePOGsIa+UTmb$p-8tDU`>tU~CYbgeuKb2znR+4PVx zPtPxw5X%aC836LQ!f)0!XM;^*XgT4;T25yh5Ngc|Wl(qjlO1B@%(tYf9aP?j-uKm| z81I4yYROEL~T|IT6IkU&b13(P&+QK|E_FivPaEW1&= zCu^|K5SRyaFMhC60annVcCIza%HzJw9ztd2 zdWjwFH15kbyy+-Y-zS6 zz2Poqpm2!LL-d(OyXy-fr@D_kCR_yPp;67XWYO-8ykfO<=zeKI58sO({4{`8N+|h= zw|lh>52_r3C1L4w23H;jcf#gz<}IN6XXCN;@j;a|DcTaKFJ?#ZEvaz`J=O>kikVj> zz9rF#DYo@0YPB0gTVUI1NY5W}kJcCQ`Y!U`hVV-~RVinOb%XEv=~~H=u^o|phesWK z!Y{Jq@v+Q)dX3xudc33RDP&-~OR;vDav|*F+T_hRz_GQDPjhvcbz^ySkW;@rxb_~~ z%~Y>d#<5KGu&xg!fR+GiMFdG&UBbatu2=ZktP`!-_jXFQ1LeD`TbZER`h?bn2aEf< zV^x>G#Kq=?>_=v}0%Ugg_N?`_)44v@nuNo;d!67%Ux zBNOeCAAWp2V> zhY-_|<34u9UV02~;g40}mUkhz_vPH#1AK1{O|#q5YXM@^W-PB}F{UDc%kk1oW@>$) zn1h;A5&yjlw7e(~$YW&Zg!nvOn zI5vrQ)k?X@%4MKQ?m|pIuAhS@Iy3BQ2vF0&z>gtCXSO-2V^mEy2YxfTK3}inGC)oA zZUR|RrQG}MGZi!ON0m+okO8a??eQf(+x!w3Ny?5kgzIXdYSN;$VAHS^7tqWi%9zQ4 z4^ajUDRuJg?exSD_IQdh_p&k=VNV$^L?tJ|^D%favgg zqW)NpXV#C|*|9!pMV8l6?%k)6FaBp3k@*fF9-4BwCKU$fhuItJjd@=f9gKV_UQveK zT;I1Ny;lwSc#cQ5pjI;S5ZJeN9sI7(C4}aN9`Fef5TTF|Bm}I+_DW4$Qrd*DF`rGh z-d2{5holuH7PuxYkMLP4Ez`hq-r?@Po}jT>Io}_1uHbJxv3%}v9a8D5fg4cBexJ-! zu?mD@>+W&1JYXvfU(#IZr09`k&qd63b*$&9ngL>doLHKqeyDuvN!`qaUd<2QAlG@R zCJ%(51V07Hmqzz?AqX&1CJ-aHVwuJ_mau7kt6bXxGp54YQ!p@{nTWd zX}HZSR%7c%5~=ZSbHCEti|2GpS;@>cB+tw)u7#eeWqU|j+qHUo>mZ*e$mi;@@FZ)MG#>R8&NIAl0 zo|Y*FNuQILb-)va*vs(knV|cicRNRD&|yHM_N(bD24~@9W%XHsqhE~EV{ZU0l#ej) zV8MWRh%k+!z_!&-GG83Ct-qcJ(K=w#^;qw=F}}&~i!3;yhRJ{;qZ$hJ-P9a;`wvZV z*YpFB5{A(RRcaZQfKCV6fVrWxstcI|pQ7rZ1i^e*3Zh}ANq|TQewm;B^UJ+A7!7Ui zaTi{JbldTOAm?pxt|jLl#bl3c2UU0d?y~m4MD1frrO~Jy-nqOulWz${=Z^ql@Ja_R zQ*zoDy)*MQP;_Oq#WGyfC^`IynHK}{t`5S|O?{GOJr61MAxi+lBkdl3>vi*VIS!`# z$6!1eE$O#>jB0rPhJ@-4H*W?n-oA5oTdjw)j_Mkbn%?y<1`W%KpDjiDRk{1|T>J_` zH)44Qbw6j(So&=YfX2vwr53_!W5nRCcPXkF_Ux75sK**+bC{$p8Eh#JH@HbVJXCZ( z^!AW-A%ck(t@6p9l&tBPKYD^n$A0bG?|^)kj=e#m*m`w$xd)EuIo$9$iwpfF4zr-1 zB}R)e^o_TE_5C$yN%h@f9txJJ-c!^Q)gX&v^mshp_okqiXN(|Kb|zqUOwK@E;rPDr zl@NDZg?S_GDNq`wD4ArrP>*&j9SW;wt=Dva zPATtXTg_%Bj>|;(Ieh`LAa8O;3VRlEhSP6Q7-FknPJ5pR24Czi_dB+9B-N51Fs4=jvU4!|49c9gvib7#v5a$Xh`5o7RQaaY zE?Z29FT7_;6yTE8FW#7m?8H7LVkj~{1OFd>BFVKXt*c?@1IyzE_Po|EPkDiyq*C(=8d59+Ji zRBhX7uC0OvhMwpw1P#*#Cq=VPGccGZNw*LE0=m3_fKz<9I_AixpqTI7BCDI!Vj_ z>B<%Fmk*ks^^)H?pvGlKE5%XbCQ&7qi8qCSlX9G;co##Z96c`0#lnhknyZLxQ>;;% zvniQq^}`!s20K7_{z*OQV1{Mdn%Y&QE8d@que)YBncWc>5D>@E$~<@U!bjYQup(u& z*bi>TFmABl0Ds^zNSI5n1ASK^Ly6@LmVM;c+u#Hj*;m#d@#;&g+i>WfkD_IET6_ak zmy97>HEHtFW$8z&kQ6?)M$hn*pheG-;_#os2X$mFOD(91-S#musL5&H=Uk6)h{7b2 zGyOs=4qc+`_EK3WfQr&*>yCW#vzS87^wzR|T4OZB%_AQgwP=x6>eX6+0WDunAH+%= z!GwHPca#^{jq+ecu&IQe5Yln~sCUG<0xDd(mNzyfr}+9t{Ob`9BSz#K0fPj;d^3Sr zSA8nRzS!5;f%~K2+D~d{Ww&WZ&Rl$3%||J%u1h&vCjh;)@oM0R2jVt-wvCL#*>1de z8#3mLG4~nVOUTx4&>WTjLN7Udw!zsPw*5{=qx)2iiNX|hDs zY7s9ql}rhe*~<5XTYP2eD$32LL{l3RWsK9i->Bs!B({F!8 zF38RSTcP%v^~=_ICA9qz;rSe*m8ZXmw*LtN>xzAQZ^`!U-V71RqYl)toYe52eWTxr`YU4o71r@-MQ@L{A1_+k3^Al&lM=ZtqP5)JV#G z-W?yL6^9w6%3D~yfViLAtX7AwPaYESSOWO>VEzi|G<(!~pwpYWMmnBG;Rh|XKY!J1 zBb6Rz*UHPxq{gcG{4p$=668kC9aOSaEl|a0iS*c3U@j~%*t%QJ%Ak^`T*wr{MBXmO zrtusjuk9n&_!P(`B$E8W?%5q49kmYW*8R$8K2)cXc%6I>_L_YO?6Xx(&qt;Yw zg*^6ElqO$%(jhX#A&=E@_y~0BAVc5?y;70GK=PtXzJ#;5nRk?O0>f}<6+ zy_f}=&X%{u24x21<2v2X1^%*Oo#_CHIOj&ooK$`;%Z&H>h}ob4S0x&)?ZU;zW_Ch2 znh~_d`eG!u3<&!h%b6NFiqPfz%QGYxjA^Qj{k;RRnTl@&5}l#S5}|bGm*}QW?8iR! zO#~<7bi}xLT8-A8yjw_BnAPqODU*@>dI?Tu5&^x;D8l8i$znR@PW(8jOcM6?s=x~G zP}z9Kw^|2aj8G>F8nN4(98N1%BPQ2u@ev3GFwWT4gkxu(+DgTJkub_CX_x>U2op!5h- zw0vXW#^U2iks457HLy;3D-*x+n#Te0*SG>d)0@x&33wv*9Z~_b*2QO%(pX@~$Nn(% z{LuhT_ir&q&?SV#{x2LBp2_Q|l;t8y(WutuT|Anov=d_C5f5zI?Vk&LFfl3y04NP@ z{2aU4LL1M3Efa>W?+K;tVtf~Y=8OE0bIxNOYp=ET-hpxK^N4?LpFhkNr!A|l^XZja z18Zq?v&Hg@sPNxppZ3qoG3raCtyL}MTtG2DA~y1*OJU_pZdaXS=A(-Q9EI4F23 zqdhkgLq?P1sXLM4Qw=cbOl%NwYCJ=TQl4{+l6qwRN;mQfOm{MWRn5qTy#GNbvc~im zQys^?Rq5Md_f9nu2?rK8Zs^Gq#?n&|5;q>>?CZD&lZhElndBzAawx0CbeoC5p}Mss@7<{v(P&$Aa7@Ar&#?yDZ1do>T7?1 zt=7B0t7hSv=4cpMH1y}rmPU%@UxmO3PpH6^TfRT4sN@5|;anmF$*c$+$^qH6NzdqLbgzA62lJNRDH(^bK zw8=i5Sc;^Ab^Zq2+7*Ty$4D^lt(06!EP`_=sDRI(^rC) z^LO@YV3Cmc4e|ttEBNMQS(?JHmb0bjzx)5k}l1itYxqWGdXNC_yQVKrX&poJL z{TCIzE;+HVgc9~u1QqXFE09L{GFj)Z_ni2aOQCS5-vmtqIRX(Qqac<$!JqY6tY|o9 z#PS6XmitI%yG!D{b=M0q#}A8=5~bq&GRQ0I?0yfnKd+Pubd|5NwMJ*uWpF-aTs}5i z`H0@TZRxVB`Q7dJsOAE|{kZ`J#Q0BNi`-wS9R|UF;^HUplriP(Q1y4D-!O5Yxr|(iDWB zf8tw2TbM0(()6q1)|$R|zXzq0^%pf`)7r2?rx^g(6@4oaQTl8=VO z?k&d7zBvs^!Kgch4LiNV+fQ3vO6`%3uD-{;aMr(&8jNOg^@%vtlED$Yuen_oBpB37 zcwnxt-DL1EIa}dnoWq<1eyvd8S(x>B2X2> zQ?8@y4X2&sI?sSjvf~+wK0|73M39>k zm9}#avilJbMrg$#m7L?muH0a(3qIV%$Ey7+0@k`+Qk3~B( z7vIu#)#Y9)s;3~2r#q^mF z!}n*6(W6aBl&hQ$^AH^8vW1@lg4P%S;$^pGM>4#P#1A??-m8wwlQXRK$U0udKYV?C zN#tU=wnhjnT~dmfErCpzXy9yDij-aVav2`ytOhji^jv9Vy$Cc!&L!J=Y`jn@ML$VM z452;XqM0*kNF7uuiT%QY_2vp)K`$<2xtn7B5cW)|)VASlvT_m>zPP1?iF|Y4`Ern8 zxV8>jKxt%z32*#z!JMVT6g%Q^3=tX5iohs=L0f>;Y4#m+#+s=|YVQGoo7`7v31RaF zr;;?3I#siBn@`?pN36$w6-d;f)wHhQi+>E+QL3fY zxQGDXe2e5wfdCxvxXSiIf%e{T*_()ScwKOQ4dg9>wd}m@+dtczJ#a0uy?R^%Y#(tc zY)9a&ECst8t12V~5w2z{@%imAW$^lGq&BX!S_qeatV@O7?^~q2E;Sbxc%_rr8Lb!Z zW!YRW(I>xr=++_E!a09kbh}i+aLl88A2zl^7`10M7|mfpP@rFFK9Wp2^eP}aW;ype zr+^?+GhazF_xvnUm}dT))bm&E0MoOdOL>NPKWkR%D9qQa6xI;!mZVwKLmhZ9bdfhD z_RfEfl8R5G;kqpxr~ROMU!qy?VD?1fGySds@u5TD&3z}@>c{N3!qI-AZI>KD->U=6 zAW#o3`@TkQ?V!?M^ji>#vm?e}(J0Z8w{Deu&hcu7wl1hue6-Rm4@YHe*KBZ381n;9 zkeZ%k&+f7wiidK)85GzE_^HOPP!Gfix&{drvf_-NgT2wJ(}t`eCd<7Z6a6w(YY#-ej>@OjQmTYF z*8IwG^SE!*Xl;s#XkBW%x0%>z+oatVElv<)3&trF zmxdnNiEXpcPwNJg83a-qhVL3Nq_6R^9tMIS=4Tw_sl_mJ22xd`S7MGTseHLnBswHN zX*6=v`aXg-tcGxyGS{Vsd2rm zP{x%$7*f7UYt!(0$jG1a|7oL=|NSas*?|RM=&8VE!1|#?ZQH34WSYb!5a?cUAcl*| zfxb_L-ZF^{yB$F+Bz(u@9;mLKSa4X%5ctn9H8o`)H z=&i`J6DmBb->Q}CP#1e)5)XU%W>uq9gVGPtxB{?>x-^Q{xPFQ?2Ywd zlqnCmJKrggMBAZr2&mwL^PCQstk_{z`w6Zx`u4Ml&bH|n58=rRYUj!&Vo-`_cD5h^7h%2Sp(_BnkzsUG|UVU zBLe*wqjle|?R1BHo?_Z8ts1J(E|P$ffYK6w8i@wpaSX288OyI)#2jz(IDpf)d6jRV z32EbCC{6@zL;s;}+zeeun{(E_pURgVc{82+bmg z;n@H!f015a1$NXSDn;2S9-dAHgCu8(m~ zzYUYIo5KzNuxHQZ?ORKq;@!Sd10VhNQQ)wC=sm)vK3gY~wWBx54b4SC7vJ@QI15H|j zjsM|SYEpq%>4S-tG+H%<-=Y_y-?TW62Q+gKO37o?WEa)jwS)I|^e!dMTWscY>075--`fvFtp$xx+P)3O&X>#;>mkM_8PtG%i2bdIlg^5= zPjlYjIzI@O1>4*Q7SO-y0D*#cs=4g=!=iFpEA~Ir6ut?3P~KOeetQB`wgH`S)Bf{j z!x;c@DEKZ*_u07xlZMgT+dS+J`K{voIEB!);BWpKJhVv$TaL%yga-wpsM8-jlebCK9{^)LztIY zeLP+M&fk%q!8|xcJ^a%MVJT+HZcJiExmngX7L?FdWkkSImC5H-#xjBTD2{aD>NJQ; zEbE#ZkNVKYlw!3Pt4XD-B8cWZ4T5Nl2t(HC-ZI>e`iV7XsjF!tpIKjg=2YPaFB#(8 zxkl7aG-kNd+VDWy-7fsx5s~@*!+#s=z2?%}g~j8kuYA#;zA33KWiNvwH|lUYdA%c? z3vRW&W2g>T&Rh_xfADyr$fSs`KBfAL}s|FV?{iSV!9-I~!kYV%<7h z7tLv{6(gcE+0*@(jff7`$fZGM_NTv*aM~=$v@yVol@Q(>m_`h!7EHz(eGBH@P=oQb zK=6%t*SQ<`CC13ry<%Ndzy(`InnNsD+)K>A(CPQU?CvkliU&&+g(P7I5r+VfWcg|lOtt3v%==_tb z1B`!UN^x2ZzE7^@ElH{(Cjg^6DtUoAPhrZc!C`}Y#Qg*(*8t-@>Ug3k6RG3l9mu)v zm$VbB+0UPEIDCEyc**n-VXWRIIM%9Ruu264xZz#_AiA&1j8g#D35w7l<%BpRwswN|Yy zAI>K5s6qXFN$TXd-(tj zn8lNoEy7NlmO~AVmPg?KCJn>o&<7GV(QQH=5Hm{W9%AJJ1)_ZX+sbw!X&!->vdoBu z-9-)9zi)d9;R`&;3p&{ntLYGoVXE;zspGj-kw(;b3eL%b)(|iF0kJ2WvSZ)?tWqlF zrmG${_fx0N8ol*T2VV!@pLFQs8SSFsF;^5$_*xL6;Kh;4<@3VOf+7A$miXQIcA|OM zYvwudXAmvtz8fXaJ{JsIC{*w!Pccw;mEC9O_1!yQ8H8ZGSY*X!UZg-c@Y&=wP3gt;T?Sa)Qmly zQ#0j@`*YQsPQcO(vf5UWD2Cd85uCU+L>$!N>|}ggnuE^tF@$H<4ivcY>V%%|2u-MH zRhZNx)7MYkN#eFOa6ZYH>0EBU*+uNAVxx-O%X@psB*VJ($Ju(FWmA2+LOJ3gF%nA~ zRC_**EA`XNTOOTcpy9r#i9NV-V*Un)cr)}Z=DFP&S$RI914+@#?jWzu?qUw#EiEFn zVQ>fOS7y2QC<0I=Giq9ei|AnZV(!-! z0%n6I@CV&=9m`btjQd|G#!v#fc;sZ%~ewx?A%N@fGD@%>tKi&pXi7}KLC6r9z zpy^2CPD$tyeuX<24rDc8SkU3ejX*_sD#1PI#7Y0OxUx*q_h$C+Y+Zl0d$RiJ{gmwc zeCvY;&x8oDnYNNJ$GQk5q*$fLYNs|#B~Imv*exG7qc#uMC$Z(dQl<=Y2fPGYX&ArY zBRC@QH*)!h#fDQXyB+fTz0LBkmie8Gy7jLW)7T+T@!M~O9@b!9Kp13_3>QfO7aUNOe`Op7kDa?{S-&m?yFE?ugg^L9f7@++N}UiI_k zmquCzv~TFaT8qXB&$sWD3BHwg+P~+P`KzH93zRY*bP83wH`(<;c#~FJ%oqW-aeBb% z=Cj_&Q(!~MqG!JZh)mZ33}IoS=d5}=>iX#w-})5t*fwbfX5SYoeJ$9_SlZy7fy<5} zydXUMh<}P(8uNbo`VFRr?9o~CDs(0FgtfggsOEGgQ>z-phhKf)@IYqTV(jgUUE4Y} zc{#$My&QUjW9-c;Rt_H%3jBu`R+IS&Hi!LL z`s8tF9rSDmD*$$72MNFT{s;p;llJZLj^ zx@$R&<7$N+0g3xhfRJm%j`=Vr95%TncQES% zBjzq);ijK{@1~B6_z1B@hcGDKUrpr8Urq;z*J@f8rf_)iiK3U{Bp<@&nPrTz(uE!# z7*T6BChJX!L!7^W_lZrx}7N(Vb5; zlzK}v{!RQ#hiD|)>=p%_r3CWMgVmd~dByxij>rQ>8$n+(yGd>4?oR`eF(-@Ia4oNV z*vucY7swe-0>dGW))Tj~2a^wyv*7G66(7cHCeUeM;2j;LW4HP97`d;3Izhpxzu$`4 z5Gc&5mw)^M>=`h>+IhIiZAkiXlXNe0GwNn7Pqxrwnc;{kvMQmrKfat=_8-)G{xvWa zl8M~l(~kCLim)iY^>WqZp}N*r(PDOW6ymQU zwTtZ@Hm2BJ^fBSAVK+t~2+C%Lq=O2%D837>Ufh@It?c5{O?Bp_wh-Ng2M}&2u+^11 zS$>{c!TA@?&4)Al!dAHqsXnc`yGf-1Cvq;UY{8d0rbkr$x%tz=ouBI`*D3pCk5fH` z&CUr+K2#`l7Y|&6af*hbWMOdwXCv;SMRsGqtx}wmNyvoF35`A))9z(u5j=5p%SCs5 zw;aNGTnL+;leBt;K#yiNChrQv?+*gwF|YyZ4(sw?b?0}L!;JNHk@^`#d zFOQoWG5Z&Z=1qBLzI79}KOFCxP;+d>01jVs9j%5Zpm&$8&p%xgS06B!qKSJ%Y?1wG=<>5$H(7m_>v$J#IwD6pLW0V2@<0MTr_tZyY; zQvzPQpN%sm&=M@1A{CVhUJt=T(HeAG2!u9T_@}mm%`4wDVx{9Wt?(t7K*=%SR0S|q z*Mo)>Jy%6RkU0lo@bwTFI4I8Q@NnFk=%hTXQf@Dx+K6eo(4K*E?}`c(rRB|gl^DkY zAS}buwv@tP32?6jB}O)>jKO*||A{TDiLF#`i)GUfqtjPg1FDzP9pY*&me`B%;bO1e z@s6Q>xyV6IXwaHX^1h6abgs!vmX(3w4m$1EsBcS=;Dn9T*L3=hr{1uz{8Y%~NF}E? zz^}~zft?=v2R>MCsE3H>`ulp+$k{`1af+?>X++rCYP)K|ue{mMk&yxyeAxAd>&DfD z*xxp=O*vyr!38LgrhBcR2so~n3-0hyEMDgXNfHPxO13aunvyX~6;(^L8r-5@Cgl>9 ztpo_M%q!UB_v6f%c+uqSg~#y+SNXyCF94Lu0{sOh4f}K%t~fgZeC|J5 z8pA^D0qbJbm$y#T^>g%3->g7oNkin#Y z0dH$;Vlt`d`rZW+#B&?+nc7kZR))KJF~ou?fnrywL?pjGl$dy$dl!fvicl z*GT90jP0zVq2&W?h9z`4w`zcoAg^o>jc`ViS}(1a)d+@#xM$gO3Xjy;#{*BI%!B>x zSV3C5mC-=@P4W*$4N050^IM*)qo;6=`LIZ@1WC(33_9dx_}0=v8`DPnURF610e{YUQTC(H-0o=qcLFW zbKOJMskdiY4eUo1E6qUEuC2X@vA8F^nYYl9Q*|3G2L$QheRob-M|R#eBH?v6Vm9k^ zg(giwFSD@^scaYM`&@5OAff$sNAT~vS^pqa;}mR*4`qo~V$jB-z5#zEpjTnt=qy)? z;hC_XdTGT$bAS~I8Pjh)C$sAOh$b}it`{enLbQS)h>lZ3I_=}fNOZ@a$i$z7KQ>In z^@nwF^3WmR<}KEj#c+;ZTQeF{%RQo~NGo2u*`K8{jrLapo*apLL3^C;bNOn&op|5D z%zlF|nd`a9&C)1omuZIk1C9Isg?D1?U%luk(H=-49H4oi9*13J+A_pVs#vb~n178E z`OeFcF0O!;9$#-5gnXPbMr=z9C|%Ua3?V`PVAs)v^-WnYA%D36h4@Zgh2279gOCeQ z_AMj7wF{FxXI0-Rzg>9hU9ai0o{6ip7SM>2!l1C-WjcQq;#wVz4hYmVQWO?7#}TGe zn>26wPN5H8F;vLje$?@P4cFbMpTnv{ITe~$?Kkn_HY2i+^D{MspV*gt)=5mCPwMf5 zyKZ6NviP}_)Hl>mcp@JjoU6cYI z2LNTmkZvC=Se(!&=@;`g%SuB@!ycw%C9*I1zQH=v_1c95=? z)3Fk8$%WQ)*$@fbm%|Z1heL5(A8lrt8pvL(@xzLOgLOh7Us)v`Ve<^hu`2L2W1&WY zQ=?k54LdBkgHB6_KfS{>Ta5M$nM7?of1N-wB*RhL=37OxLqhYi z15{#`V~h{0Y4AJrUOZ(clQ^TT20>(M?3H{T=hoHo&ncsMJ6?|e?C2``F=zWuwIpat zY<+KTI~WT8!W^YMPiz}6%{kdpe5qDzP4Iw}>+rqBFFJ+yiP^|C7F{Ovw2<`I=6u3b zmX-Tp8GbP19EeNeVvB_agl-vFonwxdpT8Z=uQG6M=tpsUx>LR|{u!Mj%+&%f&5C>c zMDU>>5`hIN>SQ9O868`Jzd*#IRitw?sFm;%TX8)1Oc>+KPdl2JM9WpbBECZL6g`3r z{!o0rta3c9tID|%)kbQ(JeG9m)^_}(ZCRrGj^ksF65~g@c1F=s?Fb7H?Id_nSepge z={#gq7zlXj(?pVpCNQd3T}4gK&LOJXEx^3K7l{P4)Z6D&mx%Qcs$dMC`P~JN<pRDyt3*}<`>(_Pe%$IDVQPOB%@;Jp@c0@9? zwN7_e4E*JFuy`{;%(>q`J<%5qtcCRh6Vc*A2zTeM>cI0R;-Kw1>x6KXiaM0|5N& z+5L;3Vgd!d-1i`9?yp#bRteU1`NeYG6hQrBe3WYw8l=(7+xb{8lqANNV7QK>gZ)7& zXZ&U}%Eq|yuZzdCTu}n~i9R9Z)B9Q)zVZa$jL=LwY$EUTs(ZKHHil-#Yx~0aDWK}? z>TJ~|t83z%pKa{1sS%*W1dKfWxs*qU{awSxNqvqZobP^*xQfRh>`L zKj3Jaq;WNjxl?d2ruRK^HV3y=pBFp*vf83uh(+FRg~4Eku~s%WTC~c3%XRM64sUfa z3J6f0rVEUIMTp1f_Iw1|tDd7dAzwh#&wC+KId2G3odZL48|C(du=X+7Jy3MAmN|G* z7%rgIEV^yP83JxP(Y!0kSKds{b>uH+5K1*F#d%(X{s$@L!#6!z1)X)9C)?HM9{xk` ze3yFPY5-l+ijjXqTz_@3%*Tob@BhMN%WJ$;(rOLV7vnxRYjpZZzUNfhP z(ReMa6Ps5Ifi!@+u3zY*YJw7UIxP&ayZ@A#d9@hEe_VzNk_A~32%fx8Y)FgBUn|#a zNI0JBI<)?FeZco_a?FZ9DCe|$TzvH(i7c*kPaGuXc&~hvub6iBdG1f61YiS-7y4-m zaw<4ZjE8Ql{4GHtueGbLaDvu?Ubxz^-J+x8bbWr6?K4@}r$nE;7DL9(n|qFfnUm`q z_o3qz`g>1*N@hBm&X_C*5^l&aF=GqKdh%;-fvY`hr;jO z^1Gct2xA#K1_Y`e-vBfa1x<&#%IGEBIx0yNV0|UFPgB@l3QwN!7QfUJlxKqBhMI%Z z>ETk0W`<=EH&St!UR~lvSj$~eXCCfQ;RLR^uC0lt6d)jDL@0i-Ft$rBat&#uK?0cT zpKmeK)RsR*$x>(rRn9gfH%!qO4gs-8j7J$YU|%E|D^lurRP_gVqOr3jKOXb z&O>P#DsTv}v|-^FeJk?8@OfE}`A3G6pyGNUk{R3uCRq`eg!+t%R`(4AhXZr=v5N_>Q{0I)gvswNAAd zZz_kA8mWaS4XHjF$=oaZD}}y(S>4#Ms=+LP7 z&tKKHIb|c_iv20HKoCn^P1N(?6$pJhutzYLdnKL4zq5O05p6Xz!G0T}cZl+^qg563 zGVHYN52SvGReX9gzhY2!O4Bm+(>^Gc9?YtKlMH5Locz4UD711O25!Dix_b2g(@X{Q zWJnsRx?$|2sAw$F6r!M+-x@|^sz_OSq_aEK+L68NMJMKpGc}8kyIIj=8VQt**@#)e zZ{aJVm@Ze9Z>?!9X`?VZ5Hy0v(P1lx@c;`_ZS$1e-|}>1v5S9puDX6Qs~M^L9I^^* zc${Xm@w7*z{SoY?z+@>VB5*R%7zoD2kBsdX;R6QBMpH$av=sM70os2kZl2DRPwA(n zP{mhReQ#QQ+-ZwOXL_do%}aS8~jT3 z4XRsdACY+9EQNlEJ^U~qrr%ITQbDtb(EOu(Q~@_#0k9=!knNcAbY~%m=(-wkpemaN z@`<1_Z-T_Nyrdz^KIT0F)@VL(?Xy^V?HH^!!WsH|%xJ%2cZg+da>V$Yi6T_EV38cJ zVsbk5dgfi4@{RpypA*K2q7s)ibC?;p-TuzfuIxToG0Qq;`irT0a$}mXN20(dV{{aA zbepLX+6RjtK0fPd&KGJPh5oq0exdr9vz2=EVX5ApGd`_CpZ5=!HaN92zd+kfyjhME zpn&|R%QoA$`rTZi;43w=&G6(TsU2Bu@9y;?=6tUXOXrl8pg(l*yY9ktbI?Ijpsz^+ zS92-Fu;Nv4(;fG4UqN^smEWCrn!tIvUyHd+uw+NT%6v!0@}C~y~^tfJ9fU*xLv zKi3K8cVLco;iO}IF6L5<>cZMd9v258s+AgEH0nRm1XUw;N5$G+l|q{nN`p=i)KedS zk-=sxxM6X{?3L$hZqvJKbRMS_S`rOQ-twdlP3H71wE&CNps*-8o*Xh%A^KZ-Pa}5` zmuT(e*mT1wlM+X2eD#^1ZO@-29J6EEKo)L&mgd)I<84D(&_D3V`K`zB5K+d~x3)Bi>8 znM>+?ZyQCDKV1JeO$)GPDXM15AVH|sbQf7N_d_y zm15_21Ftl(*W&HgJCW!-D$IFDhTb7cJZBYRGvIZ+5V zEl19*z}dtCibR_)(o}=XV#nYPbxjy|N2I+S2T?G*vVCcX$AFuoCqlrD>o%BS&mbn) zWzGqKx4_xSV)lJ__fEU!E&h(yrS{FUvn-)1v}#XBJAJU%)BsfbApt*df$7GqK(G0x zf5b+z07hcz88E|$^^c5wU+hZs;}${eqki=`^jqlZ&wxPZE)^xrC4lULNN8DuujA&b z)L`r{9&12RoP|Ij>YbQ4TJ6YQu0uA!b!C<0eU}cm{qo1k0tbpH&WK;@bUEt&gz4(} z5Dy-Hi_s{^l|+?2b_CYA6l&24h-@NOtJuZnw0c^MFKN*}lsh#&&2Y}-kaH5GS*8RF z&iS0kis#EGgl^+Ss9h0rgjr_5?U!p+cdAL$TFilkeD{h3oGL>#W&S*fr7bSMWugX8 z#QW!Y2i=B^pxKsm9Vsy=D=B3Pyc@vfX^aUITp&$W_&&-0GD`N=9;?_97G% zP{AqJb9{peqr!KJ|LFg8RjX!ZOZnl;h1 zJsB_f@B};ZLMIdQRZE=`ZG?TAyT}~V53V0$0016}&-J0w2DKA^K}$wcwBl0=m7eQ& zR3Hk)q7bj1q|+@DFW~Vk;0LaC>9~xu76v9JFmpopjcyxfX^-79i>QYXW!R4w>+s#) z2N-C88^DslcS1`(w;J1+rGF6q$bww30oo zZb*2A+o;G~*a3*1?}xiC+O=sCqmkf-MZ53T8A0L>J;3YK+>mq>Yc2j``kv4EGWq;% zUmDmMK{GuwF#CklRx8P|i6ZxA^gLixn8AT?G6y5@wSLY=BH4nU#a|8l9~7;tNVuVL zREl(^V=5NY0w#YgAG8E>VYECALRWM&&t3E9ngFns?J6|o5%WD6c#$t`5wN0#7&VKj zL8@~i<+{604B9n>c(|LZ{?E@L7^u2O;EIlqbciM0%FMlt&Z!uL|-#@)`aX_T6tTlJ@w!}mxViUSqRUi+yktA_N=#6Na#CYLQjj@ zq(j=DV(r#u)iSOx6w@aTE(LP#x52A(^yQu>1TLz6Z-##r^D^Fl88p_{t&w1pbFxEa zBVSNV<90~6z$DdI_6x%W4!H2ixDY3}{3Ep#Md;s!!iY~?`iJ5LgDqJOpX%%D$G~Ce#{GWU>_c}R`c9uNc1RpvlW<6H zkz$(6h#5nmkjU>x1*^We8b%-$Dj%jv#4!&kV-1qTUF4_P~ z2k0*H1Xfs#bG9yhz{1dpopto-XlD(zWORG{Mr;#6ru?_{lJL8z`W3_qZULGl=lUS@rbC~GgKWQ38eIX zsbo>AY7~eJ<)Ax%4qQJb)X@NgmL!<#4Yh)!iVWha;hD})v3@Y1WbrcNi*7haCY?HO z55F1KZll4^WY>cd_GH+Hi+MHZ=5_L0QXc+ZLc>OZ#h zu>QVsp!p<4Xd;cC(5-T%(EjXu_O=ddXUh>sBgvAs(wnf(nmaZP-%~@*R$??=NYKnV>3?r+NeYZ znuC6o_L;4<*xE@wm7I!l)b8b{?Z) zN1)+nJcvJ(bL2cHM%a4=e>jW2A5BTRKgoTCZABvOQ~f+cL0gPwGS(tjNSE$e35My6F~^Qg7ij3@cXmKmr)+U(e} zfGSA3hs7oKI_$EQzB4$`9L)J(DSR?<9x7Pj=MsigsJj_hy8 z3w0*A0mH>;8x^{h#hb}!TQj`h3qk`W6{*L&5j#L6{TKf^hQKVE$pzJ;RL1O^07D@-J3} zw9xjPHD^J|t!6ogB;`3Icw)Eoua_6%l1cTS4>|E0k0$PZO?mco|COV;|OWFQwiX#h~jhXV7_Z_=C7{6NouUc{Z zH+1J}^5nC&mh+{nih<=G&xC<)P~`3t4|e+Fj4q47a$9gF^Ac?xzu98jc0Fpyz*taqb#QH^B0e#ypI%OTEBY|b$m=*4iedhOY4#8 zvoY?{GDtRt#fj0~(zIqFF$G2UWEW4GPc~hVY+{XPmls)W()JYJ6sUBKZ%;}1hlrkU z(Z5I8B0*UQ_G=cd=Ufa@^=1`saNQr|N&EN#;bXBe5=>Sm{lFCm3!$(Dn8&BnnWsyT z9{(0AWDk!iq#ej>FUhe$q+k?i9<|rMAV)et4m)aVfM1f9Qzt4 z^z~H`=WdjjiqF@Hjc@wKaKe4!m>(LLN9R*BluufK&Rp3m+Rbh&!}R_s?z-_mf+{&o zZb!e%3U_D(A$sLdt+-0>ANPWTJI%1ED$UDzi(l#a%|ZATA9*RhK4I75i#=kcuQA*H zMZQM`$QPP;Bi}-=3;Z4;${dVOMa$Bc?T~}ttu#}iU69`BET+MgmQZ)h?U#NGwS}UL zrqns_c7K2Qwk72V6Ta+Q_PNA+dWa4M%|{<+0ckx%bcw8U ztZ5PpW>;|`Z>vlTvcp{J1ZMk1tOfm6jC827npC%RoT}S+V|xjrE7faukEfEWwZOtf zqzy8$K63sT`Awu7(P)wnW0dYbq%e)^HOc;!?TG(dva|VXurGH<0s0`@L^Kov8hpp` zhRa?p3y_w@1%VNH=K~n2JZ7m=bEedMqKt?OB)42iZ?s}i3H(CyIN zi^X#KT<%FzlD~#r-WYyt+mST?t*;OzP(AN<&eT0m*O_UMbG|wb7>63;@v&8>?l;4 zR>x#Y9U%A^HjjLt@|W#Ff#%$Jul7|0ok9Zw8%~Ge3<{f*sJgL@o`|ukS^Y1EuS+8N zu+y;&#BQ7zVvH7r@E|Ir?LJo*GW;@5BRbu~RvE*5r2_YjG`s(#iY`tpfE1+3&Ek|N z_0eDKv>J)Xy3%0rRp&R7o~i5I^>~N+T=m@S+ryBvy%Q=R;FDFg_eIq*jak38>PTF) zA5f?NmIA2ygm1{Hp+ngN?m9ikZs`15PYoe zU8$3o=B2X}El9mi@C>*VWQX)VzXD6#cy}H1EFa8%@|3%)Juu+Nk}|@h8eRL? z+Ug3NfJf|Hf^s#YpGI6n3NA35JjBcyd2w!deKghR^1wsewIv@StQE`SeiY>EGd-;- zi;m_FsV)K)yxG#psaUkGkh$wHPqw$NblSXk4f#;MM`i8g+s-=c@|=cdEV|DLPGJvrN#J5}iL<_iU(Qc|*9 zM&u>d+u;Wm>5OVlSwdtcDuW;v+Yf8aW!m4P+iuT4O$`Ph`2Z0-1%zR*k@PF`ZVpnJ zkBVs=k==M4ig7ebxWqEXT=)}tvWn;jvtdDQGRciPbq6)(fzI11(Y64#!>s3-1DglFtdZ`+YxlN5$>p+xY`>l}?9E}!u#`OSh6o2(B zaEm`?wV<*S-;o+h9$%fy~c0v%DZe*)*f%N|v0Q)Iz<^laO@ zNR1MQQ$ETXLiRmnZOxgqZljr8(zshG+)`XnS5GhCgXM=LH+{v9|KuXM2(+dYxnzDp zbp3=*I%RbErTzK3lq2i&GwA~Erm9G95N}~xP-74>VhRzO`8ev}`FfYMY3mT(FbGEV zWeS+#*Jup4#hEgkA|_8gT#ATSD2bNt`?6!Vg|O7v*QZ^J=~8@YYLolU@6xd{$cL#n z6X$qm{<}d$T)xLxAa4yaUtTt=O!#qtTv`^RBMN^T+)Kds$)~bdrRCV~^feK?{1ZCc zJ%@xgnH&><==05Th}~3%Wk-DA8+s@FjT0MGbm0`=)YVeFd&l=f#}8bIUoNZZg;Tjg z*?-0qNs!S2;!G=HV9_LiOf>1Uo;cMOSIoDUaGHG@=bX`|@?aJpTy>#5bs7i^P}A&u zEzlb`k0jc6oE@#%_|u$j;hO(cfqLYZcVpku%Tnt7iIvSp@;$u7)tiC^j^yD^(ie6T zG;ATkVJ5`9YgR;%%+RNKNU$K?~ns5eU+x0P+;2H*cojv~@;5Lk`C*{m4NX#6?6x^CBr zR@hy9J=z5XtuF{T1b||$Ea9D}PuQ!bNUh2~q2^gAQSU2*&^tQZbQlyn!snk7GI8rn zN$L}}AtE8%KOebLFKB+sC6qIbAZU*`NRbuwQ18c{eyOq?kj_)Wvh4#Bb)aDX zX9Dcss_viUWAqXB%I_kmBuX{gHjC(o(1E_AxHJlA9||jcAf1S+>Oh{wEz{Wvm5UWC}$afqC z6XP?!BGwLiI^(0v|NHCyeZ~QKQ6gQ1I#^8XKbKD92iR;dWchp2`J{f5z?J#C`c=~c zN>LV{n`iGhLjUU#-MVj4#$ANxy=Ak-faDo{a)$&w&Tb0$j}Dudi$SRK|9sh?1cyRT z;3~my`>tSlDG1RF^Cw$uz?QhMu^VXg2SwmSBvs^5O22jOd{&^mH%yR_) z_sfy^ko7pJBs>n-)bTs!Mh&DiKh9BH3LIE`yWBq!y|V{SRtP?GxU#6u4@Wp+A=hkH z?k`*#jxw0Tnb$|l1_kKm|IB4c0!3)Soy9Ps!X?udYvtp2FtPW`v9!-65DRH@L&%j- zzsIt;o=O6CT>R;pxk!7a)6$*SPefW_x@waD`yT%1b^MA6kA#|3Kmon|@)BcD5{R-U z@5>1l@5ZykJ7ItF>Sj>4nH)Fi+ZyVPI+&DSZtiAN!g#*`h@JVo4lUHn9YJ@^jH_Ht zCj{JXk3qoqafG$zZ4)AQ?qmK>(`2{Q7(mx9qtbX1pum!;CQ5Y%>oZrA6jOe52!-8> z+}Jf@BanSAw%$f9>yXXG-iY~I;-q6~8&&@AtNoio+Hg`@w@CZKB5&C_HUj9_6S|4o z-`LU=`nKnLvwWl0Y;VHnbsqof<}*n}VFnoS>lzj-7BG|u<>d=hp->ce=eE|i|92KZuTw=j+o9bo>MfNM4euwxU&v6U0U3WjX?&vQjY{(qdkby(Ex);6p(A|(O}lF~>@Ns1sy zcXxM}#0=6M(v9TM-8D1<(%ncmk^@7%KfU*Up1r?)KkxCq&!3~m;f&X{uC>k;=egRS zWf20^D;OS)Ol;fHNKCS&(0x)%{j)#xjuc)H*4+h6&v|HSW>hI+Pa%cFjjf~K*KIX3 z|4B-rJ%ZU6fq!)#zgQ3uQLwU-tCz!9zw%zBbC_=6Fev8gZzT7po^20L$U7KX_mATv z@eRnvP&=7c&az9#sHH9bI(r~R7*uVM>I#g=0%+Sjrlhgiz;Z+35f^J(!ZH#Pj~0F3 zSk(8oDSz`)js4s2<)Git@gVYB(k}}=>$hVS2Hq8hf8j>Bx%3t*e@gU$&+W*;MZ3(J zo`}o%+i0*uYB3+UA8>akQIBR8uKIWcjaYlmSn*}Owx;A}b&()IOV%fU#7hJv{w(X2 za_PY9(`89}pw*0S2W+mpF4Im{R4&l}LE*)vu)WRL(D3J-uE!{Yd;Z0Reqlt&41T!L2-9JC2>0D z8}wTFRCz7erg#btFVtQjux6q8E)wd!iZO;Tvw=q?|Z>2-bj@?8Lr)ulGC z=cScF+X;c?Ox+ZIAf;K*e;(9d##Y9G^m|0o^=Ou9S}HP(vwq$AVoGfc@q-f_ajsgD zX%^_==vO11;lIF+4WxSCbg@!@UZ zqcrLb@?iIJyfkK=Cehboa?vBz%id&DWhw$O+B7^LzXc!l{+SeUZq(QNfB4FOi_c)( zKlU2A;@p?W$KpHDuLQBU@9bfR`(B?ZHdf_rw94AUFJ@NE7rSRR&w$sTTNKYI8Z(Fl zL3=*7l=OS>IE0z-AnboKl7Bzx|M=9-^4HOyq#j*4Zhy3gZ*|slSPk>Pyz|30&gmm9 zr95`=o%p@Zwo5!B2N0nT8izt}qB?kjMDM5tO%~YD&HcyX?ziol&La5=+^fad z>4E35mdCien$>lxvrizid%0j2=D=;g+U}k93_D)#-zUR$!8xB|j{fwdsU)I#6#jSS zp)LGRp_Qx)N*z)$+^d>B`BUay!8Bz@hRDwMGAA_4%I z!Jk8+sP0<+o5=J(Ji~uC4t|`!3{%{3{jQeGcVga!*kUH5m7PhB+u~E9(h7aQl*iid zhw-~tk_o=o5M2j|k7?3iAN-I0>)*fcKR&e!{lhdfPp=$ygI3Rb z5>d^2ieSbF0}0F$0Z*Q+#f%Oc$BT$Zj`M{j4)&?(xo~dwr^f=^-e})@u0Mc_Ve z;eW>Jzc~@|e^4sX6Fkn!?r_q;cZb|59PTK;V`!X4QfBc@`SrWz$%8o zq0i|HjGCuU_u`^sA+i%&QT1aXv9ua~v3Q@rnSajwB-PgSZ}WivA8!GNb&21xB(^~c6>yD=9jSGJg)15HK$?@K;^>i4k8Q?37{`}C>!hmWuZ zcfzqH?co4z(EKonIDi0BT&Niz>U=iD-ALnuE?f5NGnC`cm8BcHcKmOz4CNnJ)*prr z1qN*xhKtLZB$t|WlUFyMM8Tx~xSeHsQ^Nn9!(&A-_{;6cCbcB}8bqSSqw6(#@dsb5 zu@%4K_}Ap=|KS_Hy#LF=hukS7t;WsjwZ8`c`Jtr!n=^m@nFdPx-EnI8X|bR4=_I`* zsE_1ae)yI1naok0tPbKv;{WZt)Ho3OUgLh~FpF6vU_E7T zkbu1;376jR#vjVP;V*aYZ`k=vqi04)vVQ(`O1EZzVE|b}Z!{Vn>Py03!7G1Akm8Oa zP(k}&Zz&LXJPvq(!lJ5}O2vD@rk{cPugC6h&pvvbYzo$x zjesr3Zs{%W&y~Kqd7Z6@kT)hcKv*u8U@H^Az%R9)*@y>OUL#*b>BR4NE*42Drvw?w z7rMWCy_Rnn;K*UIR`A`fO+p=MalXu7rhjwx54mx|j{MH?M9t$muS~hfxN?x?dYBMF ztKJqO<$d>YM$elBpCM@_k<8WkrB7B@-{?YEqQY)WkHu)aVBA`Y6*F|B8R_HucoWE{ z%EqrWGM17~GdgjUAD_^g-SVTqUcHD&kI~KkjC*SJc8UIJ^Y-ry;d(01S6t{ z5+DR4o^}V|px;%*Yd)D2K6=kO>#uIYw#pfH&4;QUJToG~Emc~?NI*`BJzwcm?-dIa zUI${{#YQlKpA$TW8r@f%bk{!!c)})n!l1ui^dp?dpMsd?o}8ckn5%Z)`0lVvTW_9f zhi`KpPni!dXGD(5jgHFY!Ipc{z>{Y%bhE#+nQ^!4S$Utk0fhb>8?EWSg(kKc;(&AS z{gmf@=q+Liw}Y!4BH{%`5W=1I4v|XsYOh@c54t|Y%8Y9bA-Mwf4lAQdLCl9g&WU{N zpSOZh8wW~~xHn)Vf1d0pqeP9zD?zQ236UE5JaYY#R|sKM>;Xe{TNTD|V}zJZnN zWtp@fBXyn`6?7)?1|qsyeO3adpdU~?w8!;zj6N|qv_!2X{J+_xnkTNWxNqA@d~O3I z$D)MpMJQ;Wh)88D3 z0uoGO`J*5wk9XTJ3AD%1mD!Yya={ggv`G|0^RYF2%2YF%49ia-50@y>yZv(L*l)S6 zm0C+x$WIYev1Q~&1$Sn}T=Jpxw@krmu2z2S{jrb&>j3f1R8^*w@vtSng@!e81KeMH zTiC>MUN_|`HxllS8GWBBzqTp1*&S#j-*5t%)E-dv8lPz4&Q}6P0DY08{dH3JpfUCY zu{z^1YyA9$zQLSRYABf0*6@AB?GVlL`BgCdbq-A1F1$0@8UdJQN78*Q#nsctmBkZ0 zR;A(zzh3K6KX!;Vy=zjb%=j5}-+GYhg#;ML+FYJA#jR-6#uSDUDv(IqA-zIv*^JuJWhD-B=| zHI*R_o7jA}_O|IkdA7ecoDW1L6!ud*k`xjeZ}pvA9yOx`hNOa%lFEF4`*zdgilT|y zL=v)k?au#tb=oyK(xZah_XCMjlQT>{ahs0^m;$$8v}PDM+d&()-yk-VB4 zjp=6MtHg7A3%`831tU(da1-w3kZ@QZ=OATh0ZL75j-y4+g=@;yVo!19TW888u5b2s z!$&-6c#=hCj$1~w!~0os{nLVIg`5o`4M0_NC-rRQyUQf4B497 zj3&c3bmi2-a*>b&bRg(}`{}!b?@HBoE8`F?n$F()#j_=g>wziuBu5h|g~NGHI;pK6 z=YU^S%vfwRL_T)A3nau=iK47m4pERbYcFPcr}gROSpked2~ z8T39yGNJ5^pT&GG#l?}k#_M#>gWOwCsYOKj3ZP&VRI=ip+4v$>?-TkvY&r0bwrpjY zaPEgjb`F!%q(PRPmNP9unTVTgX&O%{Dx_zQX*juWPKh)q@q?qs?(bkL%7BVc7R42+ zulwnt7I#C+)Oj;FLw-NM-^J$5_eJszTGQh{5t@qbFVrP5`eQ+~7m1cwK4$!{sjZU; z8AobwZHq8GY5P!UOa|Z`@hHIrU3z52Fb}Xq9Z&NO*WPy&ubQyeOv&X3yONSub=3r>nF)5Zf z@D?xahnE6ixE*7R&xMS@<*aM_plF$J|HX#z@>@7&BDyU2Yp6pX5=8%VW8;>4CNVGD z@iRx#qx6*AeZ$?V_{qloL3W=}-o;=2gZbS#GPQr1vkmW@-Wqvy^g0sMgv}v(Rb7Vs zOV?X!NtGC36~Z7hu_&T7qBrH4&7JEvCee*P$M5GT{xaT2|1@5zzcpU9`86bq+&$S++b&LzhJlde`?{==V8z{1EWK{8qo!XCCaC z=Pl<^C441=BjDA>9rb?H&RP)pUWB7|EbsIO679ombEC0uEHUra^7YHJ)G#LXPr_S7 z=#369-w?K2F9=SNO-8EKcwgDdEdn! z$vP2k4kiW-Br?j-E9O>UZe3VZZ=6wg@=qq}k;K8S&$e-C_83GLg_df*=6Le}vx0nw zv$d(R>AXyfI*b*~QNYwi^b`CaWawR!L-%q*gez9qBtj0MZ8w*Ek}l#5xGQ((=2HXC zgQfL6Vtx_PQ&!UhFck)sZ%6();B z88dciD37r+y)X0He^TW&GIE*>#Q7Ad6|aT*a4wodQzuOOAekUxtvcJ-v2S2HiyLdH zri*E<%@He6;?dh($RwkYTERWY?c&Dy;0hgn*;yZ_;w9zPx7bMf#5fS+X9*blXIk#> zqH=oSac+30vCNcY)TmCjU}Jxs4WGsFy<8riikUnY0gEoh#6?g=?K1Ru(Af@tJ}x5< z^H`kbTE{XN0PtW*oRGJ!oQ;O9O(H4k(MXzrR4Nzf&B8H2-BcWi;|@EAiOstPEyS0nNtA0r3XsN!Sy?shrL(0AmyK zA;1uBz0|rS&_;cG^l;z=mpr$qUv|XElgmtCcv4-Q%x+q7Mw!_dK~|E%9L!}J&uZp& zJ}c4m`Gis+ZvF9?*>-k|p|ul`EA%mp6am;u-eI>P6UXNI6u)>E*h4tZuliUa5##xH zVha}M2Upk&#BaJ@G^Y`_-ay~vS@S_Bzp=^Kc3#645ucrcz}u73B}!L@L0NE}^_*Bl zYF$iuu}Yz+QQuP+M@eXK<=A}feos9gMca}`&GpN#tk9FHbum)%nhJp4wC*Hw8-z9DDKJKT|Y zUxxh5`#R7hj|qt;J$-E(sZFR*l!$d0Mu1}A)zMUmTsT@^CCA`mYA` zSUZ6o^95Ofr;B8Qj^Wdw2I1*)J@9pW)pBM)RY+S+9P5MolFN)nI^X^MdX!N2=L5vK zNN=WIqvW?{xh?Lb3F~Qb+$(PuG^}xiHEYc!;V-(_u@cgjYj!>Q>U17%C>zjd1hf_qheXaR`QTGe zVfl~}Ls_*Ja|7QHMZOSMX`_0F5VoV!&Eo;KBRy}OmnRirG+3?1K3gXu2ZD0;9xW_O910?XFkrbc!T<2$bzA!)I2Rarpdj zzD&S#vwA!)G^|m7k+XyKu55u{`&hhX8pp5Uqp5u4`9&|e6>8<9)9cqMjl5gTb}YIy zM)q2%&ilaJYgoP)?-X$fp?Afl$XIVi?7C=4^K%?f=E{X;E>jlNPq)6aKn>@+BS$)z$eZMu&(WSg?wHJR;+adbs*t zAhB)q4F&f0z{}6lmw<|-`OUD7V>-_I)VbksDz_Xn1JhQknvP4n8Lu;{N6PJy(+{u_ z>D@_RW|S?O-=ZiQtXdi#QXHhDNy>4!=v=GJm=6Y@9>a)Chu3MbT3;0j-mI}egWYkbO9b~$nH&i=Wjnpt2-(E_yH zEUvyg0E#b@>L-O{n*#~iGbk`OZG9K5w9)k^(wM!c79=cJ|dz^1{T|}4avOgK)u!GUi9L(`Vs$9)aIuJ(Bon#oPx7UL!B$Tf+9csByjGd_=^B-eoWfHT%#4XF*{D75 zah^@KKItnRY7cy>+~5Xk^INBhB|V1MGTMYlCQ(IZvc?m%wu&-$7dA}cham!i<=dE{ z1Nx>c;gWIOy*9MoR*W~EehK?%1O&?PI6q?H#u&TQK+Z4J46{4DY`wT$=L5N{l5^Tj zi#;PN`RT*K1&gpKBmN468_uI|(5bD)_*H-Of!K)DW|W{h+kU!j#e8njW%%{4!|;qidzqudJ}(-r{UEt_2u(wL-tuMg#xUMTE*MBid|tY!|Z&#AM~$Pa&6XqQR2 ziQ;#~xykZ9J;6ydbh7*gzTl z0v6!MKI%TXI`->XlYg=`5-n(`J-)Dj+OA(dP9h zvHh8z=xm&}Wfc7%dbO4xEoTeffiABTKa)tYZjrLxsyM7>JCs1DBa*e#7m$^8jS)gv_q_1xqfn%mEioHEn zG6=H3P=291LsYvrXcuz1yB=|@Wa&`QC9@|qCv8h$Sk??)8c;{Yls<2k<)M`E5Hf_d zo;#yD5W60}kDA#R{Gm+}#6$)viBx-D^Q&|&MtOf%o+mipYu@0-apa!EyJMkFfy?5S z>SFgU`;hO>-~$-7ylK-d;>{13vir2&*p^9PI>>t}_0^^3riiL`pAVz%@#n?w&m%5evVj+Se_(7 zwMRTu#ofLb+x|t}vc^J?`Lt@i==?}G$)FcvaRg}82#)Auyt2y2`;m< z+A-+)CBEBdIAmRXK)>{;>&Tl@9?yt8e!Id7n`U|=I)T+iXA^mb)kGA#?~rt5vfCqH z>(ysSqR9Ciq=%wZ)rR+bmx(HrCwdalVJJii4EQGfF<)()PQ+fKv>^l2 z(=?&1C6I$DO5gG2Tm(g`r?+5$kL|!g^fT9Po0&y?9L-JrIaCtojs3VtKTsNDK)Ow=}3)*FpKl<3tAo!C9CZLYYISW z=TAoefL2p;d%gYOpOC)()t+@V6Sl?C7a<#K!Y^xRL%Mpdl3JI(@kqK=Hj%58-`!#7 zmqJ$muBfpHIu*-PkHhY6w~5M0blJgf`V&BechLOe)L&xEroP8~rs;D2x$h@jr)iOY9|!qV4WV)uYJYL=5PE83ypY>kgHAc+?kJoJIl=h?BnZY>)Z5*b7-_*MH0Nv4mzQAW;*{Y z?Y2RD{+vfyQ$YN3 z7bKRkptM9Ix3W()MiixzljZx|5K=2PB7|lphd#u@!xqZFdXs^tZg?VzPT>Cjb9Z2> ziu9C}Rm0e=Y8d_WM>Od=-RG&FaUdQcf~Fh6YKY=i**0RiNvoYdlJ?FP` z%rj7y3ZLH$C6=sDy?Y&=4cuIcD$?_)E%m6CJH5bDDER);{~*Vns7z;Jda_J}aCf*V zIXH-cIj*$MMoXdjQpcFs4~!}ro6{FL=utrF+s}zpeP9ff&w|^sjB<7LMNl?h5N%gH zE)8ZHmE>`P=N51>rog9N-75s07zS0{_PE<*fE8s__m&E(tzGz=IeAHOZimpT(vkbo;?9t1+6kkOVZcOEOsF~EK zCo}gxue+}8xHJDT+i&>wwdvC+-ayv|KXrX4N_BJLo+eCqgpqsQEgBz|n7s#7|B&!5 za+j$;^h{~<<(+BIlg*uYVYZasZy;Jfb8+zPPujvmTnty7H!p^gxV}0Hp6jPo{Aw3% zlc`DngdyuExO6+G4x$}U;m!I=a2H7=(qYqwHKGW?8d;rI?F~;A%|WJ@L+7euHyfan z%g(mxhP79t8;Q&3(6B+2OLHwYXO5>tb;btK$IlX9cpH{I7WQUg>Dq}W_2SuA*ew6MLYzW%5+Lhi|mTHf5ql=4cM9jl=ikA6GO+(?$TY5$vP zW&i0eybn$_vJJpWu{%r;YEx?)nc_n5vU9U>J@d*;qg%L zK$`^Qb>qD)2T#Ly*gKr9$nx9J?qg!w)8W{h2xY3N$0TM`MG0~L;K5e+?#(orL;DBx zPf_Gtd=fyQY;M9IBGQ;WF02>@SQtZ67c)r{D)^dRQ<~hp2yS^2OnbCr2{OmUtD7oO zh}hZm+YGwL3^%?>g zEZf6{3}QY#a^aBLMo@0{otLF+mcLrvs7a-yzpE-#lMI10y~hUur@l0Mz)uPE#S`yB z7W?)5pNwjnf8h?RS+;O{q^VawVq~avnZ}Y<-Bkn{ZZuss;PZl94Wv z>jRzzCmRZT<)yx%4AwQe)(kERmG#RI^q#x&!go=gZ`kZE92-2Q42vF&mEM(i>k>^D zRoVa7an=2Q%BH>U5rHrX zhbpD7$nULpSL(&zI1V2z-RvN>z#Ps0@|c55WVHW$|18FMculuI1$$Z{v5aL9 z8BhP_M4W8ab`rJH2^+tMT+D!t*ujM$rGY_EQ82p@oyHYl2Y!4l77^RuF1FTOynM-z zUOZl+=lv{wD~EIMV~t1q^8qK4g#j$tM#Mp=NFU+G2gHn_Ibkt6CtFbSL-c!+$D`am_A;bu?)1jdieECAD!CpWflY@#=Ef%3FbgI(dKjvcpxuE*m%#A!T>gMe-?)WB zJle|4jtFZql7R?W!f*w-8WPNYsR?&TRCvNQK#=A=WG7^|l5t0K!$ZmpW*(vUH}c)QeD`3S7R=`YRPpfo+9q z<&O2-7*O=C+b_G$MaraQ5vmk%=NvwoSe(y%RQwEckHZ&cqWLGaIylLy!QM;EpZw{?d-&gi9SQ zq9v9(jegi>dKg%BDDr+vL_NRTv~ii4wQhZq?bi@^o5WB$l`O?z#&}LAR-}$BKc(rm7*^fY457dzW;(l z)fYUo2QVypEe0}LQNM49_@Epq_(V|BT}Gy}Qs?eO@m72EbfQ_gc&t1%IAMX_Gdx!& z%`57}C+G5#*Sm0#Ie|5yFd5N;^S3(S4f-U6cRzO1$00&$?!ChPnE}biRsj|F0oy3( zYChB=9pChQnw|&?&}mJ-sWF^uK`VC3Kefp|?$-Pkm;^DZ#zyi1`FXemtW@~GSmtXp^nq^RwED5rd9OV2^xLl}fA#uzq zneFnGkufO5{>fI}`wHI^CF$6QTuPH}D_x$b*9+fD?ci7+%aU3^rt-)MDLIUS1iW1X zZyDm1^}*g6=PB{HwUfiID4MU{R#F6q!^poEdz>E-@~)bg;mb{-LN4`-mUiz%N~$zA zhFBue_CL~W#ChD8u%lSr+8SIqpj)AogecTu*ME*a-5R3aob^~K2;#1Mu&~53{&0y? zEREHWYPnCd7hFp^d&gC~Rk>~h%AXN=#o9~y&KPahPjm_KC;4f1k5oX5M?u`%_2ebN z5?$=ONv&8%>fGHAua!JV^RK9gS3WcHpy!2C(93!Q>f*8|H{9l<-<~%^eGpOeSMA!r zJkEmN-`zk~%k`jOdgz_QQOiBw-CRBWRTw8({uPo;U)r^lBD-n->xa<9vb57p?v=T% z&Gu%pT;o9vy5|HdUS~^DxCY1NWA6j$6#I zx@#$c({4aqTnGz;+ho3&J|Jv~ab+Y$KoFgm&V>#iX2NQiOUYiRgut4LpZPwW=L6M^yIN~m^EN>5Gen4D$vm{VY{e zGO^j*KVOj3nTC=GI(nf6eKn#9*u)0i$1cGKZV?gd!Ho*l(-eX`7#BSDWLBPX%I@iR z2ZPVU@d}8Oj=$ZOxGO*x1n!OLM$$9(NPzQCWur}L*0&IS(Zl$i)BGjh2^S~!bRY75 z!wx6JO)~YOap3I$vU*#qK^Rj)Qjp;f1)Bv;NXLwFo6LE7msmJP-bDzl=iC;$MQ_XV zJKEC#cY!TNUa_PF!^q;(2>!j}i<>@PL6RpyzvzVST30+{Z9bOMe!_zrL#}z}QwAUt zM5Bz@MjxSNpY@roFrl_xd|5D)<5G&_<8lQd%2#d03Xl-5Oh)gMV8^kWN8Q}(Q{2|> zU$A>$#OK@a!NmMUNaT;%Se#>U)9yDW;tcZ?XMVnOB$RmvK6OT7=_UliNK|w1&B3+xBi6op5GM0 zf-kZ|J?DzJMGVde-r!Tvva$^W8@#GwB^o0@3?qAcv?VJpnY(T8^Tu}PTQ=9L7FlU| zUDx7lntSV$Vso(GyZ)N0UkM6$%X!rFvmolV>t^2e2xE!FoDXUFL~DO7?;09qssW~VMtNk4b8J&zQp~i9|AO~eTr59oCMlDrG}fn5!b_#a zwJI)5SS%$bNoY8nX7{eP85SldS+qEQo=fzRzjbS=G{35OLcMVIrd9|b- zQ(8)n)R6Ox;}Mx%_oTI0YyK?6+;bmi5{f)-V~CQ!8TcaR9(j|7mWkOs4o(fJ*lTR^ z&1~wcH9JVSpAf?7*Afqd!9+qxp9-=Rws4n-eAO_P7Zw?re@Yk!IV?*D5PrD z{|XK}r02X*S)AT)yF)ExD|w9`;jNzKYeksuT&h}3Q&*N}=>~sd>#U#rc(QR~y|Y!E zFWqJKPT0VblvYES2R9;ij5li-=m#sS6AzRLVKoPF#_(cuB#fkKi(Cm%Bg@27kkrvvmbo&(9#Tj!=n3R#e$H>ey1c1dVpTo#*a))))M>(DVT!P5N5r`<%hg$$ARw z0y){l$U5kmt3N(b&qzej0zK!F$+I^U?UcTF9gf>SNi^P@9Q3c9JZh5?y}0SPKj=Hy6LJ*;`+>{lFuEI z)QcM5GT-YV?FZO*=~KrEEZTNX#pB7~)53o8zCfy{ns%U`zRV~%O?hXj{XpuChBvvcv zGfsH}Dqj;m8wl7frzMx&J$=(0nHfDF#e{F`>;m)jxM|mZ>cmb9mr!r_ouX3@M*3*g z(Qm9ZwvRquBUM8UX(}D*Z=Lp8H4(RObkN(>JXI-9z*acU(CPC1+0C~lQlYxI8zNpf zs3fpic#*y^6I3$3ikNf_g@_{Wid7_3$Q!t*6~i7^&|Pk&DDO)de7(CvJllX{DwXce zl~zxsJ)z~9Dbsl6nzA^6U4=9$hjSXpuSx$sO)TPMmCbVFOt~6ENxSxtSqXMLg%otv$s=T>cppQmIqFibROUm^h5Bv0NK=nGE(r zw=I9K8uraYh~Tj0jqY7cy7Yj}LG{*>oH}%0a02ARSvtLR8Lx7L1TjI(c%irpV%~MP#$g<;>R6ZeKrY_}C zDOC*VjQ~-+fr};}QgTteeyfw=kq%3Fpu~oM#!PgTwe825rfOr*<<=MocmtHhP|GU6 z)Bb8D6wa5=vm|->G=pI^hS=%@pYO%YB}qm$ppvLkLhRTNdWDbWm{>HmKbV;$K7?z# zWzf?f;yg2?YW13C+bvk=yY|zccf%GJ?x`vSQ7#(&Rq=TTXc>wLVYLaibHpb*{sK`^ zSY!!R!+;E$e~B(dcoDYiVTW<^#7&U8N%g0Cy;r!?Vbb5}FK#tVo=#sab?&V{GxqRw zwV`a7bEOpE@0=|iJhyAT6vCbEenqD`hd7Q~EIn+j#jPM7O)F$V|uT zTo}44AsItB81oq^v({?UW+)>w>v;p(o`Gn-kC;BA>Al8Ch^8qx4k0>475QblptoL`ZaYa$Y+5zIFi>Z{J8oI`6{X^UD|y*v0(ka`>^F}Z!4-~5<*KON!SY8S{Z7{?|Z&0qv= zA;n|C`}u9j;W5KEJ_ z{c>?`z1$Md8igSI73B$MXxs-Zuv{u{#y~t%sqUQa&jIYMdEqezU`Y|9hk(yJ*hRLG zE2VYtR=-=6Hp@celrwx7-l7;kKaL!*r2V3-Tj+KtSo!|))d^)M~ zPA%8PE-b!(Gb<5BurVxPBbBbR)?6>56Ego^5v{d2M68D4=2{=P7CG~~>Oo|z#Ke?n zhj#wLuBF#ir=!qgYJkJ^+|CrX8-v`lh^Gps-Zx({~) zBIe#zU&2!sTJ{j2o*|R47Ke3rea!$kL~&iiuW~!;JWKm`r78-uyTr}gzy><1UkL6b z2p?bQ;6f$VQ8+&9lkq1QJAzRIg|`gRj@4u#BMT)}pFw0f7O^NC=Iuh)BCmu${TD|4 zQ=AxKi3C@$Qa~op%;MS9Csqkm$r4v}aZyi}K4C}MR~~sc_2fa%QV@F>Jl|_q4s1x* zx!}2o_OVQRmUBjt-Fpa}^S(g`6LZ^SsYcZ#mMjRKRyflauZ`_zskL&_tNE67{EUh$ zGUm}`X*_@C>v(YeA^)6T`b2`uwq!N#;Ms4!)yKCzeXko0%2zPW6D|0%GaV)RnktX@9XZ6=njPc(Tvx3Wb$(mg)`$X(u<=;P)iU>BY?tXeUcfEn`UhB0w;+jC;ro!K>h2qWg z6k=8-aAyEMu%qjFp+k?2*VzQKk&N7E+sm-1fi)zu=cK|F6*C1t5tXw(XsB87dGKPl z-n1#N1vnC;r9J~@yPnU)L=teEA3K!qg_D4YZtI=x~_ z)KgB>o&NSXRNpC4+u4$H$Kj>=ZCS2Quq{#`#-y|#xhMTEw; zB(}l#tQeGcY&(6vntxJzvQT4yZ}KBrxaxZCj3xI%=vICjdyn(8JVT(Md_4iCH&~vo zscZ?JJ52bowT5&{n7kGHm!=Sf*YILM-dcjo(f!!Z*UB>LVTf5nZqoz0f;HEu2D>Lw zayW6BE1zvNMT5zmh&X$1UbUT~>1C={^Tf_$Ik=-ma*+&sPWgu+y%aKUQCgVctNRg+wqOT&2vH=q_Tth>XjdL$<-pPkHl^s#xQD??^VJH3&f=Ngym zGdfPzi|HmCPj^RkIV5RSLT1J2)jjlq3fZ!lbGw!!nYdcDSD(xGsxm9&d9AMh4|`wP z71z3@8-ZW}f+xWtxCM6zPH=Z81PH<1g9nEI!6_uT6cF6qDg@WU-LS?=A+E%&1lMsPN>PBxx5ERyWLk@E(ksA_(R&k(B`P;GE>9^ zu31_*h&a+eZo3A%DkDkBZd8$Phw#;>GE}PoPZyi-8c4m!-4}q~N=CM!ya27^8D>e! zyOE*~{U!*m#2ToxNIY~r)(e%9BsOOS3C-X|?m0Wb{ka+;iqA2;kz+4O+tmbyr)YrF zCJxTu8{GuL%O1*$A=s=8CSisuc>M)6xXHn zZv)SA-eE^SAcp&>5ReiKLC{#a{C26cvt!?KNAm3&Y77gHa>;5LVU$P@R&9Bn8Ipol z!6q<XRE4^S*41?%9ehGyh`pmK(c2i{Gk!D|+CQy=6^c)EX*{jkp|T>0vZfgn`T zM{?v1)QK@`!AU-Ufva4tsXujM`KG>OUC-|(10-UPJ7A!pr~ounDT>MQ3>E%yp0QM7 zhFyXZBh6-r`CE3f)0%noDRe?2Y#8ArN7v|;Hh28T?v++7I+adAUg@-mG<9$MuVh*L z5pmN#15xd~fyZGm1>vv$I5#efs@eV@R1f`TIri%P!pxQ!5XRo-2%k>Os7PCz9r<=t zYcemZplJB?AZKyW0`HApeRmY_87o!3g70st{UFZCg=mDK`1<+}56Ni#GJ4vmR)}5o zp&zwA=4-ycs=!Vir=m^kvdj}9uNz@75Oct-s2QT?plcinP?E=>$&Ic zuVV;u74LmJ7UB9?Mjo%0aQPH+q{R3@Se+r4l@t~0>H8{QgD&BMW9ccv;~covx}X=R zKjtsd!56P)H(PAYhf*Xj=2Urb7Z?oEN~|Xu1ObNbW|r+Zy4S4vZ?8syW5A6UgcCTP z=|<19O=HtM7gpf36StZG*;M^aG&8|qNE4%OBphxL7+IrV)ScN_zhJs3%xEEK-QxUIyfoxd z2R2<+%6TheyYfwVX)kCHazvq8M-&Y+AOd2SCEnN>3Uy7LwQvTWVD8 zR2I8u7gn_m;BL$73Nq`Pfal4N#*-g93l;RoSFaB7KxCTgJvO%qqz?;on^lt+?Gd^u z>KW-JBU-H!`hl>$yEpCPbrh;s8aDvL_v{!5gz;peE)i_L+Y!d6n-G|2SYGNnTq_)N zI`l4_Y|U1-W1H9TpTxmIRp6<2kG2y;T3;GSILHNfv@&($J#l4g1-1uZEiTZ(KytcM zah3PAQ{us1i5t@X(=^;P7;h~nhHz^%vCs=F#=f*3%ZO+rdXhRio%yL>#f zH3l~m*JP=mM8;zJPSmgL8&Y0Foe^nTC-V{r>x;{pzF{0F>Z5IMN4TUZI<>BGoo`>Y zH-pN3aMM;s2(-p_V7m2ZRSkBe=N%R)0%j>Uzxw?H3+<=Sd&yS6>S{HU@ER|l&(ZVS zp@qA`_1e00hy_>>9*RQ1LXAo6!_ZaB7hPHNZSi=Pqs8FZV2h0ne6zd$VH{Zwk%;ycSq;Bw35W;#JkRj+T6C=5JQjR*oO+z73}k!K(5{T*BjBC}0y!yH9c<3%6hPxa43_c!VZAvD*@G>U#C1TvxaV zZd#_bmu5zX3NMv!i$Ik{`>sL1ofqcR`=m8LX1o@$V$DWS7>Y!&fcwdViQM+=Zdt;i z9UFnalTrz)f=nio0p$CyGiRA6#vc%&E&O*c_wzB$s`EYcXYnP3Z84jZ*N4BrdPv=S% z%U8F#1r);8u~Ubp`x@3aQn%L&SNErRg&GCNtOLiC!{%0k&MKdE$UrX^@5uth$Pp$D zQSpDV=nF5TENg6^W3Zgv(PW@14D38w33Ps}_39}5oEVfnVWq-^t2MUq-Iu`-C^)yN znfogDT$7%FE|L3=Zh;z_TiBwVE3eNgNrD?5n+{|X-R-AaL$?7x4^{Ab;U|`2>+*&w z&kKWDK$f#VS07@u{v_PtMZEXNRxW+|?VzMELQaB;m5{>&e+cmi#=c(cmcV%Pfvyl1 zssg=%UNmZADS*`ByDumL;?Ts8ITP0>9jKEK)t*973;3AcDj(sp5TdZC12T=n`fhB; zN9vTGL||;MU%oVN5+IB;|D9E=Z;(Onv8WMR<@w=B);cKk+@_%k#``h01JUyL-mC9w zQg;GgYsl)Q8J~#z_@rR$3sMd+ACsbpGkec4a-PlN_c zs2C0RXR6l16!Wehy% zuZN`uWad_S_^VhypIj;TJ!kt((_rX8cwO>R;QA&U0WFuTcuzLyEC=Y0Ti$gaI!{1r z99sR|g-WzK=>1Y7h99^u?lJb_wQsG`QTYxDpfoJOg8@m%irVN4>K zBWvIVf4*~S?Y8yBdSXw%H_lO1@ZS~Zh%)so+>|3poHGCUH{n`$jhL&)CwU>Rsn9K+ zx_#f2l&9Or%R&Wv?QlUKq=f!tn_Z8T>_STTh06tCn(1;=It1ghqJh)|_qMNiA%;P2 z>32nhnn`pvi!_9&*JU&o2nKBbVwq@a*vYt{xOLeUWhbw!#?F;by;xE!)R7YU314`Q zc6ot`jC(@BNRDJEP^oI?PiL9NcQ*T?O}yAHmTi&y?|Vt?B~pbs0$M#8ox-}~R&E*YH`89# zlomLNw5so0Ds6y(5CH0doAiW*u9%&`jl#LAtf3114&!~BD0ge=1tz&VqY2-gEBuda z&>k}ga}zip7pzWR=3Jr2jdvQnlsNP~ioGE^AYZ0diS2hd4KNcR!0#0ndY@4AL&ut4 zwE+QOcfU7dA6DPY0eQKaW-IOi^Yv@vC;8GKWIhxhbuGPsKutnp1sgu67K00${GLNRNmmb*_Fj@2jx4uvM#YD~VZ@BuLSv1Df{ctK zZm}vZfou?@!+i|ZE+?uLG7`XH0$O~uRL;vOu@UMxF=b{f77a}io)v(nMEeCFbE zgDHK{ky87y?`$vOyT_&mWUw3;1O|2eKvB-VX#cnbsgnJ9HTObZ>iGKegW1pWAG>Vz zD~iW(qTf$3(Z}d6mUM_aaw_*wZvx$l|5Vz7?VWlCmd{00B*JdI&A4rwQp(vZ(JO>g zqF=7>s85+*HQb!;@Sm}=;d75mKVf{-34bJQEdU|KxTkLANCzT*sTL)3mI5*5YZ5Zg z#0Pg#@fuP$npwxl6BnYVO1x7^0BYrPvnC8PdCrA@>Q+rn}K09#Egp`&zv!mosep zB&zU?x$)ElMNZmE|eCM)1h_iorV8czD9?3x{S%BJVPm4Da%7M%~jon{i3H|m{ z)-73$v?a==VQJw-99m$WJLbhb)~KH;EDD{U>}!)U%TkDL(_uE zm%(3O@RWsfk|vwTX-0#QjcsgwTVvBNHo-q)iu%|&{K6zAwtjt4M-r|_Tj2GiiO+SL zxfHxaVoKXHY9YM|m?;SiIBQ~Pa>3heiPCz=T_ij*Owt$q%wy_bc3h{AH4H*wHE59e zw|p;{>L-xm_L-&cGr{~gmhJs0V!{qGs;4EHtKUe3CT|tv9LZ=HCmdO30#G+y&b~bq z9vjPJRjIj_Uj8^M5^&v-fo+`K%${?dRxPr%{_5`8uuWM}zLf1^#Xy2n3&H5$?F`d@ zYiBHq@fngmKH;X85Ke&|)E7qP(Z0l{`^WJmhg}RG^rEQqk9vQewLaP*W@x!!s}5<3 zFIG}lkgGtcFx`chH*KZ2sQ2acnoWi7;v7#f;H#YPxlWtC7Ycavj0QpE9t%}@O3%Bx zX_2yoo#XSjow6r344Ge-#BE!eGf2EketHoX`^VX|_jqK_(cM{qK!D{0y?wifR7+ER z;6&*Wx@G#I>!P>f3$B$w2LZga73JPDuPz5n-zMi>&IHDiCz^5V8#M(g+ZT=RRF?zv zMyNB(FV+bRWx9KX)F6=!tVM7}37P7Z4Y5UMu$mdV_s_*${xg!)XSw2@A?j%SXyi7c zM6}Z)MhG6316mEWUfSJUE2xPs7aAoZ!gsJZuS7XSnuIz!RrBG~P8EHEtLn1C7I{hS z>p}>m8+xT@Tg{uy^Q|cFZ0tAqiv_G+JE;$EYjB zUDbq7!GDz=j+gpQfHgxhd|kerIWD%2oDV+}5ktDj1tL}JeTo}-GzlllNT-`PW*(-a zYTg&c{x}LP`1Sc|PU^E|)(pJ8-6(y~pakeyM@f@{9SY`LW=W&vybQg5Sy+|A^xZKR z?87F8$MCTCuu$9k$Io{20m=uw#sf#@3q&43PoOVtBRk~V=|*wJ~vLf@(cCxPH@A8tRnWp*Zi3Xi|?@M+Md!wNvDZ zgAu33($B#;uk?y`Ex; zJ!Y6liErGKJ|=%Bs+0S`TwbRIn#V-@6%e!t|g{6Kl? zGmdTJv;Nky?p@C0DW*~YUfceT_6)Wwfyrg{gND)x6QZib?B^iu*LXU@XutD#fJA{q zJ&EmJgHd`QfleX_mEcH?Y~*Swo{}ptb7?JiIx}Aw7)hi-+k9o4GY5C$g=4S{-CdNA z`TSu)FXNl{ke_rmq9w$*iCAI`IEDfHdT6#xH^E_0(89thd z(^iD^X(h8m=?x6AN}Zk9cpi?x0M1-Co_O@cR_Qb(WdJAdBxF#yN}l$^hkghC1P0wO zhmA{`R^)^43RJM5)K2Tv#j52D_}1xAOQy;?eaM^-FC7nE@uw+6kb%uGA1ka#b>h=_ z4#34o(70I2`#q>w*}`g+e{~3S#PB)%5o(~?v}pMNd-=v=#p8)y{|N>H<#Ya@g_?e4 zdGL?OCQ~i-Q8nQ$OjPT>fvmqnspWl=gQH(k8j5K>%|b`M>(fL-!w1sXvOUA_9p-U; z>&{FAaRf8<0pbiw-&`P%og-4Q92uem|%Y8 zbKE9fN>YeN-q+pRu;9Y{G4gj#ov$n>`_;5}1)2|T$IVzscAA9}9aYaoTHSrCk+!Cf zqu28_5NUAs1+mtHx_0sP-}Z`wP;!O%{U@Zf9(|#Bt9NHJU&X~(qPh6FcWj6q$LzBM z$sQN8+u-)?8iBD|hJCdA+)e$@wP1Q5-S-5+^j7@z5rtl zziCr(t~9tqxGdjeI^xu&PGFWAi4hPsl4Ue2*5k(WR{91PCVTB z?GuEGb`?y4v+6~0oeLa`tWTUfkTvelN{LIK#-m^y%gyTNxh=1KXzE+x0s@4Oy;@g1(kHczd!rwiqoO)e7@g(YE|e8x5l<@RWNu!HN9 ziu9(DhmAdSdtnXO*msGw_uJk+=gVl+wv%z<-EshHJtz;xiht|ws>K}7zG<;JHh9j) z282>W0GO19Xo;D*Nyq3mlJ}m?Sex)fir;^(LNxa^LM~GAD8@#0x9mIvCRc~d?10%R zl`y}M(ng_jF}blkW~GiL8*l5=`wz4{RA9L3(&s981{J}c2rJ^wu`mM^tl18` z?qJm+*M(ShZ;GI*BlThh&}z{9Gw?!1QM9o?(Gw$vb}9o(a*i)cL^^xn_`mW5>JL%T z@?7B>QKwgdy<&OQX9@#+iTX8?k0nV zT-Sdw>$|pfL?t-+rKgJ~6*`QKoz4wE9?$e4wo>r|y$ih@-V7Ma7tXpa7ADi@=$KfO z&A!uovEL^1;)c^=T+e1f^bxv9l|mAkX0f&*eNa~Ekozs3E>8d3e8XY|u~(!WOi33h z)0w|Yf1coK1ta1M&%s|vtn>|C`jZC^V~3{)X^3s@S(DkFAw2zNb+ zx9pf$ws_^$-lSdQ%?-~*O07jXGeq9|u{6GJ>q?8fx&d!1224jQbW$pxk$!e4)#C&0 zmB_Xe6O`Qfy~QXosDAtTDc;chN%RF@JyZ4>NAD=}=5O6Mj1q$h2N=sFGK)tOV=K=8 z7%|UEeHu)LzX$~$lj*Qca5F5UjIshe8vm2b1hSYE6KE-9=ZcoRx;U##AmaWVrp zmWRI0WEHtc)=(6zWl@%V_P`UM^7ldyl0glhFVg4Pe{C9}AU#+y_y^%Ih_IGg)h>)W z$9b5m+p&!-;HVo+^5(@^T7DkS2DBy~avvKN!TE*jorPL-L@+P@$P4<zxZ0ja8~HWE@dzWsIXH!}t-E=1$w zvwiMFnPPpH#b@$j6p$(A=AGoi3gR1JSk*Et(c+4>slWP&r@vLyh!LbhHYwu)9qIjM zv1t~%mpD9mLO@FneRJ}-TddfDfQzkBXkPdCPrMhC_9}vhD~|K#4J=B>*Y(-p;BI%l zpJs!i02)FroXX?#Mx?(#N-nZ-sdNg9pXe+OUM?58-@848BznZ?*&V4ql zuwbF$lDT5vn4(uqnc+qY1ObCYdL6m{!0~jilKW3!MW>Su4yH^NZN$rYL6gKgC3JeQRa8K=51(; z^il~5j6M!?`HPk>Vp!Wn5y1Vx%a0u&wZF4<7+fWS2;0KALi1xvS#jP~F*}j+;)i@}%X3PwMFD+isefq)LC)O4$eRYrH{R(&)9>mYz`A z&p=Z#5`vz4B8s)q%Dq5~00o62?y|Z2;|%b`7E3EUjO+89s8(CS zV7~?vY4(C(;{a^?9;UEJ4|qvA>~uL((H{kg%8d1#(OFG0!V;9ofVh!NyJ9w>A1GF? z*aSA4@kh0C$=}o@kj-Qg=|#4gWod4z4su)+pA$T(gAm%ZT{- zKqMJ07ZAA{QM6$A9{GoDb{vi37$ANa34E`b>wwd-rD1ra8{1MBfYvuznmX2uH!xCK zyC7$X8g8Z=*zNZ|Tk(qvUtfQBoGvJHWfUyZnKNWDU01J0f=DVyX53#eEhzD0Wwj!Q z6yl?PhN2m&dShE=fL$@Nh0jvr-gZWWjJ6Ui>RuXOV0(4CfrTpHEA9!Eb=+ja_da~@ zJ}Py_pn$5-V22&{elV5uJN~N!{K-Wph>Fine9JM)6vIBLr#Nj6UgkZH>2of=L`HqO z`LFHRD1orCd<8OCxP}9ym9qp&F)Q9E-Y`@?j2dzVwL674iudjb>7EAO)H`=C2XFE|MDxEGn4|P%A17aVA%&l zF6@p^iUap5YjenstkqjxABp2T<5LyzLZ&9vv^~3X79PNWk|{iEv;!;SS(1UmGm-+P z>!6hjTs+58u*Jo}ZI#2Bj-|6Kh(Q}KO|I%fJb~aXFYEwGtySb_Ot#R*AiBhDUm7ot zi17nx z$diqk5fSyG4ifhD`LLQKI`~Z}Q{c9rq zg`Xy5MkkD0=IXZ`N4O8>8>0cA4@Uh!ikPukRMb}q#giy>uEs6b(z_vaAzPaFJQw*T zwX3~~-&Y?`QV*EeRpF<_;Hz3wSqQQgE+Nci4re2qj%B2$ilWx-@Rr#b0*SakhAAWJ zmAss@AMTSU&sT;w50dgVzy=a!q~W-9v=~@7*iRTtzd^(9~J5IVGcu`Qo8U}RKSYhXa~3GN&JgKsXbiAa{Wbfql;-p*Op{4S$q z`kvbg#>aU1wf3}6KO}hfhpd1Wok=@V_a0ygaQRZ8Jg!9fJLJR*EkfQ{V2Nwp;#WsL z6=))~@rmr$ywyaB%qzAz`rfud(H6fm?#YOO(&E&nk-)uo)Lp&8s#eb8xl>f!>+Gcc zsE8x(4gFO9QJa7U{_9-)bDDUpXu@XeZVF0v^#|M{C7o=QvxPJ0zg#m8af(`j?-hxJI$AK!&Ei#r&V4nI(c~LcSX(;%Vu~hqMrIU+yA}OS zETB?4Dp5qPeeOdQrlb&W6acP{;V#v#h_Q29ka>`f6%Q%va@fiM3-Ve?f^%I^?LM@6 z=G=%|=!o5RY|2c@$#-vL(nMb)zN{He4_DIa?!82+ogn+wty!u^cHYU>x3~k+^nN?> zz%#L_)K0J=zR+4v2js-kiJ|{!{cTMiE-XNQ5P*_QG|;7{%dIXILrwG0`iF`6mesy5 zL0PhYd~55u%!^og=$qddff4@no2p?3`T1N8MA2T_bHAt?%Tv)q&?%*O$9~n|u+=2? zVH80AdlZoSmr=mlP^>UF#7;dQ4Wr~SxmP-e@f+*;h5#eL94>z;a8v1r<#4%##b}s0 zpK+VdSppro>Yxi@{A5(MOX(-r)~fzmC}(=2PjqC6dB`~?`$XXaz++SIGmW(BIwIfs zS-zV$c&HQ-r7!j*qC9anda^V{HQGj>BCxRyFNNN2Zo7DKGD%WBgfn zT8^$+j?Su6;HCjO-$t+Pp(tN3l?G{*)GavwEeW+xZ`5mM0vEWWvu*a=4P(&_^RQY& zc5@U?7$#*9MJ8bu*xg_pQ;Xe!f-z(H4)vqtJaVc`4xwNTU#@}+_tIc zfn)*FR-;kaq>RC_=&ETM`)f91cZ&?K{U1vRvS_8yDjKbovHpfv z+xZIZvwfNXr?w~9QoWv%Ri>D<(2&oYL1QNyJo`a)m=c`f(ebbclRGA>Gp#1q$DjJ%B*jt4{r>ud&qNTt? z6YImJ*smv|0%XWUd;-Vun{JOKWe#b z=3!rK0f&o1(5Qanyq+@R5xcOevHh8fooG_%%2E5**wxRy%1xT_fP?94fT7zl5cU%) z*6R3L*t97OsG}&$*wq;-mZ%;Th{> z40D{Xu{M|H0z{~D+^K-%V1tl;pQml6K|bVda+*g71LHfFy(8LnhL}(NUDjiOrW61* z_7}zhP-AOAIu92Jx*y71txxpFIf(-t6}$jPg}i|=#dN6@Jj7Sv32y46v#<-%#8~%r zL*vL|6hxY%cD*9A-E+~<|luT znTZe5MnkrIHeOme^X^fT_^&6F0eJfnXz)UKyXAfY1jX%+a^OB7sYN!h$24 zDXsPRC2NxmJSV-%|CGYE2g+frn*iuV0i|>_VY1F?U3b>Pw)M#%r)+ood=|L`HJ)Va zN=%z$|=!HG~MHqyIAg}H)0X&DB_To_(@@@B{ zXYI6eHD8wS%j?RKjj51P%7;@0J$?F8;!vMb1&9V6t&b@whjn)uUAh^#)yL-d>z&fj=V%e5D4>Y=zS`%#3r-5#Y%YE=iDrz9iisvyOYrzDB$QlGC^u#`8pHAn1 zPp>J4rRo)Oz`K{W7zfautq$>L4wnR0izU;{`ZX%?yT68?m3l-~Eq5XB=mAzI1-)3B zuu$;t6eh)K)5m2vJJw}$yoFHJbZY_wI!U#sH>{!&Sl>{Uw(=?MONBN)||>E;IpjSn74-VzVaXs%HBM!DP-e;fWrY#i(`_ z4oAa{S@q}(BHCx|fa{vo6wO+wH1cN=fb;7{V&#bx4Ya5$d~`i%w7#ZMDB%Ts?brqo zeEFai&ramTNAzinE!sgBo}Ah})DNsa(HP!lU7#!`5L%4CuhVbS5w`^t<-?1;e*~XzoRVyZo}PB zXG0ca?ThHedt~c_o}wj$yNZYPc~_M`P6=B(A5P!k>mNR-xM8o$VS${fIEedCZveoO z8TnIwi-4-m-RQ3HLVKko+OlX%HILdOSzd!Rb6`gvaRDsiIjBqo*ArZ8FUk?7nq7To z>m6&IDk94jk#mvj#5A0Q;=wY-^AlYUd>cB~StX1wXEOS%uTO)n>A&aG@Co4?*Y~Gz z0p2s&a4MJ~&(=?CD8@F1erdq+0TST(;yh)YHhy|&KA#`LJwveNHGeR&8P8;>Zu5?E zj>7orb^d5`A0SmF=tU>w9xDb9W(juC^TS`ih#%BJ+dZ)k^v8);1pbxfMlCa>rR7Aq zk_jwL0o=59fa+*9qH8+eXNxLNRkkU7|Wav_56)AqNtE6B->ZN5!_@% zYY-F5g^w>wighXj%Q3*+y}`JND+I*$6^ZP=?GzsFv&9b42t&B@Zv;qi2~MVknj3jn zmTQH&W*_=LpASFjqkSfy0aFmC1~1iRFW(R-12 z_z9Yo3@OUk-SpCLa<0I11)j_0(0)iLPJUM)cQZ%i{yF-JkTaYko%fXP_GFi3RAOj9 zneDynTbo6RQw+kBq$GX!ZoGKaG>&%$aqM z0Z74ela;i{Sko0Q`-R@XCFMTYkJ1&}nFmJ>SfFfo=e?nB6Pf>@z1sDRc8>k-Xf=BD zv+$g$0mIw`Rk(>|qMfe`V0_;lXw7jYsei9m$JMqRrd2hW(zMJ1N8Xo*(Xod=#( zQ;Om(!nZK?1`(LP>-qrsU=%bcay_}QGKva)#3$SZes@he07nf%tHX`RV+qS`F0KN! zqc2F|cE7gY_BM(B0wcpO}_Ktl5{4?>$ws?m(_}E3$g|B+e|aWo@rlD6_stRXpvsZsg+TU$9(vf1}dfG9mmIJ{9OE1x?BF)Rk~`!*`^Ziqr8~Z zyAvolp2gK>mCdv(@vBSzCgc$CClM)ekO9mAPZ_(%JR&#m@(X!mP-dR)=vMazs-mIiu9!E{uQPg3q}SY2z7zX z{6h^d8HShkP5&X7t;izp((TPfll$P)8+TQ;G^M}88p#I99uZI_9Dv0(f2YR{$TWL6 z<>tx17XNWr_4fYO>B8Ism_$y9JD$HDrlBAb{_nArfB&Gr@<7{gfxpBOwLd%>lYu99 zp6#y}5o_|_Uc~7m-vwNtWBNOwQd$+cNx+&K-P=Z-RzBGVB3d_|iVdYs z3ABN@CAv}mBlC2W=*h!aOknU*%=~wTvmCVb?u^9)z!}D`LeHJHa}#-jSO0xZWYW7Y z7sB$HZuTcei1}5z2Y)GdRsKb}yE(Wke18)s*~Fuf$E~<9_h3-5DL>2mUk|Dm!rkR6 zrcTobUhkuJWB6qk3ZMH9PQx-B7B8md(7>>nu2ya&GLXovd4e-Q_ix^nU-*;93-?X5 zVZ6-AdbP$l`*Rh#CpJ?xcn{JN`_Pc&XHso{j#0a#(K#Bh&kN7DxA`**^sx6m|2N;o z|1}-9_W#{8jnI|K_}dHM z@6Y@{PUZieg#Z1z{y#jk|9ejVIq&*^R9pSRas3v3uG1CL1gIPz#LV!E)_d<*iuRe0 z|KkAhatfmd-XN(J9Q!sEFAY1=t?g3ofjRhu8Wh0(A2-97r%yKT;NhPpY@|tn0BmAt zZO-p>5?AE@=>PeZ#LR{8mt6ejUm*D_P9Slx>y@7>5KX?|jkNVmmN_ehvuG5!`N2HB z78$qoj+&3*|@efQj}uSt}Yb|hrwroA?w;7V2vHz)VQvlHQT zroqoQMT&Ss|FhrxuNmU+ZeCpO6oa>c=t~H)O{*Zz!zgt_Vfs7b-M1G;uPncP*|hlk z|MEM1{`jwHQs`f%NuPbcHh=>$n#>LwP?A6;h$oyS0qMnnFGzXBA4Z$oP2u~zYzfrd zWPud9DFTqMNC-d{a)+ez(*VTP=2$N-i@CIiP85A-Dr++s<7><5>M~`|u)l|*2`l}Wy%=oSB21*yA!qc`k@V9uf+atirZ+P{Au@C zQ|f7xD*RP(GpSB+9@q;Y@r_&2h~4^4!ykzTJ`x|TJE07;s(Dc6uFoDD@2xi$Js?a~ zW@E31jzUq5*z`{~Sn^1k;r%}p)7hfv<5_SVxxO>&0!dgq?=|)8c>zW+BXv-q+m9iJ z0Dw{>YPG0GfVCH798hSJo`N+r))DX!B6osE%N(bY?}GlC)k|7OT4RahO-|EOn$cmT z>vPgKuGt_4f<4p8qo znT7qd8%3KlYsMCv?VpFl{2XC)1rCJlVX8*zpo>0Ie0f2kUZ}?kOSOIJCr+*exYxj8 zKZj@mpN%TF<-Kx?F%u|?>=NS1)7?v{S*J`=UpO73lv3MFxsBO0`yCeXpXv~nSRxJ7 zqtXwr??Q-S*ER;e4)AU8U#)MZQUPVHUakA%@?pq@W*wX=A?;_uYl$ zr_pi}b=>c%KJa6yfm9Z5eaEv21IBMB3)Wd+Y}_B1N{escDO)G!1BOj+p*7Yt)t|;P zfH5Fi3A?O*z{#hc91~769m~%KHKFYYqEG{aad|WM-N;?-JbwxhAx4t|U-Z zEME;;{EkTR;ODi&s``8>w9hukU(_Y-tjxKGTlD^(ube|d-xAX_aPLOcMPklt>QVW6 ztku~4jpUAt3&9^|UeVoU*hs!)?!9{M9pmZ)xd9$b&geQlgqQ#FAHwHt0auJ~&oCg! z2qItb()=^=K;`L;PWfgG8nPQaw3ls9t7)yiGde{2y52P%hr%e|&3w>x1g zDSUqt9?_yb7Od84Ks=C2lc+{C+hC1!MtH=S&gZPt65Tzx8m4YlsJdx`5Vw-3=YUd; zho5D%%m5g&!@#~+#JnC68^83Wl_;J(%xG0c3ycF^p43;+1HM=)?MBf{yt58})b){$v0X zy#7Zeeb3eMFN3q?owES7uSlRv6H1DX#OMxnZNF2p25sTu7Wr)LA=$jU6qnA=Ps+cn z2c6(e*yBH4ef_JS09Ui@R@7(hHE*5!P2hdO10LR&?O*R}?-)aIG^N@WQjjV<9SA?$ z$BK=x4MaKDsV0=IeR_|bA`MCWjpoI7Ggt2jDE&FNf6e?GhSQ=5QF=PC&`@Sd(YJv=_^4Hg*+aF;CYG^n8@_4?f0$iLdajKOjvJMnaisqP znn=I->yds~*k3V0q~TW^9cKS?IdLG!tGy5HU1Rw=90eTz@WCGx%4&G+%alNED^z*{ z;keK&W>B5q4(;6v#fauxYr8j%+djqFy=sWvG|wXafq4|5F3|C&p{e=A==IMB=J^gM zp>((mBHBiN{ZGT#0V$X5=>}&cL71KH)y^IL#mHp^<(@yUkmp4+W`WCm&xv>PD3wL9 zlAQyERvf)r9Y^~0E=vQGspM>(Pp+!cY_%r~r_HQNi(+6ZpN5?SBkC1}H1T?cARgoc z^4Xm1afh2H>qPgdbJ+`d`NpJ^*XTk!slda;wQZp$@guv{UQu;D?%Fc2q%n()?D)D-+BZcnv|T=CW`3^XTj({{fBs zrj!3g2vn8lzm^b9v|4z@Vu>uUGzPeJ~ZmD!kO}xJklix1~Q9yuu5p&zW?p(tzGwqV@h5 z34L!-)}{#ZZnpISkD6?%kHSF9{lX4P%iRr<{XsNyh`Y`5qOVe+%Is6VwA1;RqUB7b zy3-Eq6?_-$s8;^YhyIOSomf1J-n*;Zhy5oIvby~%`KSNTy*M%C9pb@nSu;pHc3Do~ zAuU0rY@+HBxb-er0~rTSqmY(O>U$IF5bgj0v9RF1*bYXqjzguQl#0x^$I0va(Ke_K zNOyRt7+`Zw@a#f%i$iG5VZ5d2ye90@G8Ah8Qwb4E3yCa-adiyWum`b3c!YM*+DPQ-Yt|JG{*B|aia~tKV zIzPuVOzEptNt0VCO)cg@pqy;S*m}&U8x}1B6e6fZE>nC988dJa#7OHQX{{>4-YZMrEc>x zOMZDnuRFwfyH#|jM!Q<&>Jx96AuCsXtU>R%&ZE=b+iT2G5!NA}`+(>*A`00Jxeo8% z+xJ6O9H*=a_ip39-O&9o-cG<2F0AF!dxpUGbg;IP<##MQzdF_~OGLCr*vI{kiNdva z@1k+)%Dkl5!rr1Bsxgp)FS@ko(mh z<~?HEx+(DDn<=Rzq|Kp3sj}G0@EWAi&GJcS^OrfTzDY^0l4F|^W1yMb#ogqLRtQV0 z-bw8>0li?iOnvd6;Wu~X)%{qwS}+zYwmJtsy$bo&p$^7BsL>}qC-UB+&Gn_zjA7UXl&05;>A!4Esb<{=TJk}(2X<Z>L2S^WedWUH@j@YFH-Qe>775B=@CZIcWYAR5TF@pP|e4;bI8~@i(*;kJWkx4 z2>c^1;Xk?lduC5$?L9?6eoCGv5y{O@+v((L2{aqbXapv0nc9<#VWVjk-M=63+DOuNRVoHVKxu}~%4M->+uHPb? zyS$-r?phy@x!y)+?DJWeH!dEZ<{Lb9)h1mlU>3}}NqMEZt;;qYe@&y!kfXD^T$V0j zB+9P+Mdw%;1a~4)&$Adg{_1ram-WQYCYPPRG}Iya(&ed(VF5AfR~cUUUt%~ELGHTrf0<*q$A5d=L74zOd)GxmCv@ zO;AN9;4)=b_2&|A^(!W|*;4hX)y(}HzYam2CMPGeN6uN7w^8S!TBc4&sFoR1wi`R- z+2`5Vc(cHSZF+&MM|>MDW6%Dpird~uf~^LYiux!F+e$lJW?)O0*57ZF$VDxmL$V(` zvK`iy1tu#$N#S814!bLX%aSkp-dvqfAN>-qMSg8%#3JB*j{^+UBwcQ0YMtwho$tuj zSt!G|S7EMI`gonv9^)VhG=zdg9T$Rw#iq!LRz?+HS+utwa6uKz!*&<> zm{#{z@K%yqUq9CCE}>pqql*jm>v)5Xr0??@$kX3JBT^9Tb!MiO#ZoPuzOM95sbnsD3VtvZyBD)o`dyzanLSqhH4 z4jY+ZO%3XJIU*redX%c+O0lT)uSUHEO$O*jC2)`?g;u)(o4^FvAKjqbW5M)vhOVqc z>E?8+pa;!#5~xZh*ofd}?tRq*xz3tg#Fnl8!F-89V8T}!0GUGTdf7~?@l9*&3yMP{ zivqq;vS~@4L9WFdTJFxpyrcE@2{7Ex;k>h$L1HzdKI3?i4x@XqB|~Dh`CdD|lH>Kn zWm-~fD~oBH^#XHWky7)^3$KFab$-Q4qohjV&RSsT(E9GfxzI=oXWrpb(;V%l`lKnG zSW35(Zs8_LGI-~<^14ZttGm-dB& zol$nHYih+oks0U0hxB4M(^@ZGc6f4zccn+={n97S3M!d*n@uF2iK;6816n#FLxTEF zktlqxjbE!GID6w91diU*LAI@%cnj4uDkmIq9}XxP!1LB~!ofQk#K>f5o*EdMP!%7x`$yo@Mw~1%$GP2YYnyBrx0@LCR;~H00@1|nS1igS+t-pF zz_$9F+lvPc{BJMbVr8WXc|(*_+Vxos7%e`Ptq&$Ji01Z?p~d~iPNq&Z>od-i=T+kCXWStosh3|KV-s>2<(&q{8bJi}2Eacsm|$@ca+3 z6Q%K|0!){Vo@vGD9{)?{7HG~%2@MhojH&>WPgYzpD>{-{)v`93iRz#E7d!m+Wqk3F1a=KMXCv2)azB8`Lx-B5T(D#ICsDr91i#<%GBi_ zSu*-6-`l&&_CJIWgkKpXa6)Ft7n~!ZyZa$#(YK!Mh9^YO2z6}Z61fCDPXlTW+j-dn z;siXqk2lz4PUE;B{XoOpsSpG8pr5a~qymx3Y?d-nHc1ENHL4dGITsjl3;kpz(WG8s zAyt^(I$@`-%bbsiT)8*x2b8D>2raasS>sS8;4C?ujW|KA30`=^{zw?`poO`l(U$5} zf|dJ48r1PNK_BHob~H4gp?^0);w&(}_EhSv|87&?9iA|zGY7^hSwr0UI{kK=_7oXF z+AbQrRWxw94Ec7FHwaW-BFLz(>P_1sM$_-YB8mBAxo@!iApI>|y1!K^HV_XoGi*vh zVrZ1MOOPgp^V}W8*Q+HQwiB^(7pT~~9dCQvzrq7Py%Ck4aWg+v_akKI3;pH>?a37a z#%P=xlM6EPOW(PqmeDNm@WfYsPhJtz&7BbtNWIIC6jcHh8d6N4Yhm`Q{l?8V<){&6Ya*(uEU z40;8jR01x-1unc>c#D%Vp6N|Jxbe+i5htJu7ueDTSqRH9J3X8Umd<1edx=HNIDd8L zb2*B5IJ5&_R|D@en7_;anwOXTC8i+!i$b>WCyZP&h5>5!I2YEO{O|k7qnlj(o7Y#_cPQjEuqU^WB@}^~G79ZLiF`;XUndG#a&S?2gqv0bzRR39}`R zU{$biD;M0i<#h5A+u15nE$QzwCTWCDVHP9OX&VUbodWsUn*UxSxRNAAP^s>O#J30? z$*B3jez{SGjZ$|%gvMui(yvU&-diRE5}=1zr};h>ZdY`K#e4m^NAdf7n4H^N>&JB7 zy2Mc(_gzl>g*bWL };M{QU4U4Accj{UV6Z|KD{9yKf6SY}-)*Zgfa2W6LPW9mXk z%koKb=?lLMOzSqtx4Nq&7vM~E3PyBwOII#)?R`X)B;c~C*6$#=PzNUEE<(@8;_H;k z{&G~vrYmjg=2O2j;hDKUdo)Q^?XMalNHd)ba!r1G6U==z13i-Hlhr|rm9qQAwv4o3 zD}T?9Dw;&p=TZs27__>zak-6&L;f|Hm1;Yb#2pmFS!Q^DKWSq93DFvzJMPVTZ{#?t z-L2n6->69LH;mW;r9-s@vgL6YLR-9^4_rnAUu%0>FPa11DUmICYla54f&Q^nGGLJp z>5?jHvhXpUH;he5TPhL>t|ng4vqzP!SV(V-bDNJd#arta@wpwx;{U?%`m^_Bg3DvY zS@yAeJ?%IBX|Ddpzj6iT!IQV&S{%#k$+|GP%syg!qvnp-wyhQo)R!NmIx^`SCAaZe)e;P1F^tZRZX%mhbj0#Um}&4I+eTM6<6QLXFqDix(!M6}lm3V8=f* z^+Su$AWY@Z%k>tAH%1Zfw3;<(ZlKmS>RLNf`N8)&z5OeG6F*rI*MB3vM4z@k&0?px zarpp4@(5>g-gk=7?M6(tGu!T180pLOEZHp@<45wMzyJ9*6t<~|b!Xg@zI|(L_#3(G z=4-?}7})-z$u7jjNK0^a;OhHu5olq6gO>#HV_SIWXR@5;&S?F6o>HMEG0T{2G$=-W zixQE-ezg#2c9qlaUS#3pU3utL)>|4ax03ZC;*#ptz6Lpg2-Zha8k;qCKuod+NF^OPqH$El^jl z-K+SgvZ2L`s**U=HDZbUa>B^t%ds>gjCWpGU1EqdZDtVdpY< zoBsk5)KPaJzih;W{)OAe*(T;hZ^iMutL@367KC=jK_;IheyZnhKRkacF4hsB5lqK3 zV5no)Rlk-h&L+WlZQ`ILAwi)vTiN%@oU4mDYDS zor6qJgSYMWlX#~gDMkJw!Qh*l?LnZ!`NF|lK7DdKv5g)+^_`N^aGWfe?T)P2&WcAX zT#)e>JY>71JQU=IiK+TVpO|cTJ{_Sv;N0hts%t8AK*>`1D$Z5kdhJ%1YoOYguJ)Z( z?ionY;gnAck9YfTnL)m0ghk;GN(jqJknekLC+|un$Qt%1z z4aR^+5rpPg4Tt~o*(a@PwAh%eQR->U!~!o|L4h%evY zW0t;3rA8x=kLzmgvoJ}b;EI)%=T2cW&B*BJ_=)#JT8;le0*8FEo35Y3@I7Adic|EpUzqQ} z2mT&18k4VTKy2ox?5&b-9*#+D+MT02QhwX3YdDV0g=F+ZgUVa>kGh1^jlL(tsN3y{ zz52zKUUsl^Qo6}ktfn6PX$_m=J3}8+#0HC}aM`vky18b2M2~C68_py)bAw#maq>gF zSysb*zqKHCd+9$eyE~IdMrhri*G!x(2)fH-ZB7gg8Y+yJ^sokx%YWEo%4q1BAo_x&R-|B{6pn z_PA7iVbY~jxk-_N877L`)J0>AtJB1VIMIxS zlAt=*)}}&_0sps4)ODil3k7z+?jTg#8_=MhQIYFIP`SmUck2exbQ`l4a_2{-BSAzb z(&yD#*X)lW^W4LBD=$mquE8t%&ka{(TG{ZLJ7Q4rT*?&ia>ixZQRRg3(zNUNbmHC2 z2W!xo$*5dzSCD1|7Qp74(>H%fWmOX7^u6${s*EqH^T#%zk-aKDtkwc)Gwr4B(5JVe z2DEU$MTofL9HSgz}4y#8@(KRZ3 zdeu!W9}c{lSWS@9*EW*RhFd z_xc%E%Bhoec;EdGmO#|1fgk3T)l_6kf;t9v68 z7u|2rm&Toe45MmzUA#YT(4Fb&=N3zQfB&Wo?=!qdP5|8>My`_A32FBoqScoO^5UJp z3ubv1o-cT|ct&8IYSpKB0-WIU^!@ziH74AGKC&8`{m!v3hkCv6TRmfmr=@Mq2E2g5 zrxn==P$H(+!|tH1(9qrv;1PUGDh_vT)H}2G=T)|V0xzyWJBO2{rgq?(s1BszCIc% zj@ix+T=%@#kGD|4cp0(cM|trN=EdIqVgp;u>1p_#@Rc0~>7l0??9~^?p4$35b_OO6 zT2{6eleUU!96?r8qq_!CtzuuPH7V0*wAk4Q7EtO66SrOL38^6(jnsa`eHm>s(x9iW zFmM!|IcDxsRXFwYZr)|^gdU98TmZe`j$5`c)@}Y7@7A!*>&r*G?3^107D%+16z?lAedjy3LOCbPEyeUNzepg2D%&wT!>YPAt4_1#T*i`67GMn?-};kH)9~)3!6aZ#Br2;ex1sW` zS`{qy`^I%}r>{g;p6ywf=AH*bN+y*^yhG^0{_{)FJ@VrLpEiSS`&FBv_BR=etBW|T z@G^Ec_7ZeoeyhG9cU){nz>E+QF^2>M>(z*o?ZVZjRKq|PNxk!%xEUfI7C%$oafOb0 zYcN99qSKw^$1;)}fDyQ3x^#7g${0>mq=qfmf{Z@rsa32+u1=-TAh_Y)qGfyxhjy@w zb=^dq7wrFy)HymBXIFMbVSilE71euox6v4_$mPd-zW#wAIo1aHg3iC{$W5qroTBx7NftWL;&Q~tzlW-xDuu@O@_tlJ$b{0%kjYc&DF+`B2-hm1~7y$o4tzWw1 zYl@W&*Krn73Dout2cJ&tCu*Cl{sHw{lut0WxM~_Ayh3zh*3YDOkJ8Ga>A7N2lY=FQ zw-=Xq=I?yWp9uZhAfogSa&L7af7Z9g&&xi)) zF|+Z%p5;j|@+*NQ2Lq|v9NyXTuX?_4OFt9jJX-B&dWOm~GTJ^@Q-ts+{ExX;47RXu z5=`K1Gg>$|mT4!?R*3PT9h84S&R4MtK@4vBTmNjWL?{_dBD)7uX|)a}WdE(9ZtZ=B zfqc8N3=?@NT^vXaFfg%l!;spY&hewWGV7VQvS3U1Tc z%AxPU3sVQZ-2&AuhbKqvLQQ9dlnx0v_{aCi*-R|+wH(=~G3;THQkNVSvy*6B%q|z_ zYi_;=Zt6kZE>lz`oOZ5t%rEf#!yQ$y2;B_Y8QWz7eitlbg~fU;9Tu*-PpwhLc_AV% zm5^esMV@F-i?yFmuo~Cxh$o-)d-+-$(Z_J>)UoI1m!q2SqG$#j!7Wawx{%W@HDP^#Fnes@NAiq zp4RuDAIFEk6_^i^dHZZFW)OKxTUTw>uf-Ldn11ht7Z4QH@gXTdGd@Ubbv`c*ta460 zu+mUKZ2Dnx+r`#;Frn`P)lIh=>Sp2Pq)OM8ay@2ljFfPJZr;k04Y4QKd)6-`wn^mH z*uVRmReQM>##Gu;lUb(OM$t7EOzM?KiOM)W3t}L`F}#mn`>FMK@wlvyIyop>+njZFBIO|B4t4ROXQDHQa)gK zGKYm}rTWE6Ab_>0RZi^$I?O6=eCRUNb+o7;A09-mSydKmrPuJ1SUaE87MDxX7kKSwYLQ0lEz5tw# zL{!NlrfX9@QD))OXk_jjBk6^$a^iH$!zFO)48WI%?#syjTKt$y%8?k$%-afh% zx%>pV_#cRac!m0RbJIyrmiSBX#PvPGnN77co56k44%6`mkQOmrxfIpJ0~WO#Ayxw) zSZ%yQZd?@0e0~Bv`0{Atd};r@Cmz|wE35}m{p;&3&_t7OHNsCY&21^2u0~9D()VpL z^&CxZAK9d$a6ye|<381L=nezfZ~Uqx1nhl-Q6YjOH9OiOm+gcSt*!~{iriLtJ&gok zzB&_*`=gC4~e7! zKac2p#z0-UZDON6oLgOh?|V=<+2QIfysdLa12J(<2UrZn$Rg4y8YPS&Ld%a1G@EIN+-4kDzoWAVjY!e=} zv2|)Lhy_QR+!692aI6jIUV~j+O1;Sjl~TP)t&gh&Nom0}tJKJ%@2I;wmYlcCCu;5e(T~jE`w(D> zrG4i+35YlO%N@XDb>XB^czEkF9_rZa?;FA;-4WO*1$Bu|)YGjf7Y0tA4E>h;9PK;3 zlk^%w>r#eSqf>E&Q!Q;o&}Wd)YKLV5$gS4$09{XSgCLb>PBDKpg2I#*{Ydjr%>B_) zuGvmdcaIHJu*$0Sr^iw&8PK%nE0G@Ied=o+=A7vc7}%?eB|Wj3tCI@(YbO{nDcZYe6LKxYcnjrmF*CVKryvDxEZpzEADg zydXvXPE>R1hcD0PT+r|Y@~NooFtl0(xvdQaExo{xp!NT_mS=aP=17xfc746IMV)?{Z`|fOJsdo+ABVLJciI&)Pxv~P$zUi*xxZd`m|J3OuDCEw0 zX^RVHC$Hyed2iC$2M{A7(?u?;m^j;tT@0%#=S`I*r*lv`KG-$XK`O&Xp~5VaYZ*Wm zoQpCqriv$QnX|w|#Wfg>#K!B?XOWKUx>P6T-OB%u6471$TAz;?#a7?*@4{=Iv*3yV zI8a~U&MEY(y7?}=p<%UcNiIyuEa-7tPn|ZgehBKv-5*MKPhp=}e?&hflsW68UW{k4 zfVZau0Lex5NUxBI$+olA!Bv_?Umy>4sTy}%iS9~BPwl(ZI$|)rH_(5pjlo}_6%I?a zel#gVn*~55Fp%hceIsQhVNVO;JeB+>UJ{ILPn(DWKmF-*D3?Lv(I@Z;h5hb#icX9SAqw4ZbW zSZWS2eJ%Q)ik-#!4ZN|tw7HEjv783X$>>s?f0;X`dFz^%-{;H}b9_x%LyJzZ z-%wr|xr?dRVS2%x^&N5Y4wmdGY}7T4nMrA@%10@t+k& zRgaVTOx=4Ww@P_k6MN(YN=dI=`~ANg<{Sn{a-8Sk0OrhhLq6=vEKJgjuEF-V!T5=v z#zyN&(fAMDbo9!v{Zi+h1|#I}1jdrQGnFema~k7M0J{&6gCA3*wuv|Uf+F<#dolF> zGT7;Qz!}^e9=?>*Kbsehea~6n(DG)Jt0Fh*I5y#Vdeix9>h%r=O=dNhC1X=BqRZ_Y zeBGD_#Iv%7&k|kByge6pm|-F@doWLUL>UaE(4Gr4$Mx_SJvEbc9;JmAPS>_rf2>5@ zbW<279CiLglV9%EXVsqKjJ(WMkJmk%8(fffj~2y(+Qt|;ovM|)%RF44VQHKVXu9bo z^Y-P91w{~EtOt`g`~nBz*Ouoqz;vH`%pPa|hBC~J89TO=Q*(EA3xOXIhd2f`5ml!k z)dY(Q-$=6S3F{`K9B%22{}@ar>g^1H3hWKS&Nj{`Ex&07))=pF%aBCSii9t=!Cww_ zMGa3BhT)GY*)v(&?7(W`<{#u{JWI+qJ|}Yxql)&5w8c_v#=CVp-=R9h{vR;;?OMmG z-px*(Ni2|Ff|b?XFv4hi0TLvpJ7o)YV@Y(`LMHX5z=-d8&yDWjdRABHzZN!-#gSPt zjjD$w#Ld$$wO>bJ9e5>lE&iV4hRQ~MBh>MdYgg?G!^%$*bO8~_%ii7G4Ph(nZkfIPEpGwQ;TPh-qj}MY0j5Cx7B)>x9Z}Ei%({cR_9_~ zRPIAk%PS#oB&x&%Sv6zblv8@zr3TY0#7xceq?9;%+$-@FFD+T^%;yL>x3T8C2K|!j zFyq#AO6~5+20xp#Z(dzm>s)AzjJ;y$up$^W=fgHYd%!&q3V2L}o+nw|Zd!RRb{^42 zR3VQX^eItkaM{Qkn4H3!bJLnYVtCJP&)mMJl8Jbr(U8U7?ZsCyM}7~4M~P%2!K?E1 z!eiDm>hC$ZrDQ7V1YI0G?!6M^~|6aa9cu4}7JpWfG|#63}q(d4p|< zI|Sh|Q#P^p@tA<(x(Eh8@fyZK&3*}9JcM>(Mr)l{1m=T7MOQ~ra%nK_Ats+6|Cf5m z{x9_~(q$GbUQuS&PM2_gN<9WB4U`H5v(!fZGQ6UH{*BZ*Z>q>Ul-=q_!ZPbeGa)Mv znAP5&ufY+@+ddn=hGo@Ldi8QX#}%?8)+wYT`1}AoX8nZbdwZ(89xB1$kG+4s5&vEl zN_0gPwvRC80|CeXjLcE-2=!HH1_JC3Cg1-JIsG7-5Bkw5`8hDu4X-)|(_xOTwPq)d z@TI>^o6NXya4OrLFi-ZH{Z_&i04YenO($HI4H(d1Ge8}YlGLk3Ov~cp3%{rdTJ}*;> zUCu%>6Qd5LHBdTx#7FU$J5mDatFLMvP}Z@+tSTS!V8{!J~exqb_sr3moDl z{mPM8Gzh_)C>auyu2Lz2(qBnR=`lK{D|D0!q=|BYuiCTCz<-LTHT%+`^JZiNoQ)`O zlKLqIh6ycM2;WA-Y@02REPq+-Ip4dzWVnP;&9Ph5V>r8vVui^V) zg@zT(kdXtX<~m^xv-6Ad56{$w-kJ|;t9u5iU~^lY;nmN*kVKU4{p`a>jiK;l&-`Kn z&+q=JN#tkoc_S-;9C!z;+E!&CDW{Q%>$#=gl*PXQ;oEii?%84s=M`Te=ENPv08+>J ztGal&BPFTPK8BHPE=3+o@gUA3BWL5-ha7WD1-;?^I2&KWrkr9Z36j z7FT)ChfKlXvkp0i$xYycG6$T)*=dOtj0x93ty{QxBcd8;JOCeA!DpOpdd)f!m#A<= zT5W~D{baZV_Zj`?WN-Y>$>!feS_hWQA$3hBE2|NV#HU$dzm&p@o!+G>TkFyHg>|Cu z6W7DiD%b5Nf0CW9h@5E2U2m@5;}{?8WZ$us$=D}chocA*y=30xY1ZXp5*k56Z*?00#!Cj zKu}M^gD|V!>8ub9J4!yhLZhZ>IQ1%zWEe%s4pO6MeJj};nfdVypMFB?&F!WEl*J0m z>A}q{f&KmpEyqR(KHPgcZ@Smh4&OWI66y-F17jX=lQ#v zQx>e<2Tem=G!NFI#T|8Ez-!#aQpT(h1geGK~zA5K3)LM==Ur`={PE6~Ce_b=oG zJ;`U}qJt@{k@#%vp~uP032o=0wT)%Df){j=b8c3DqaP-6Zrrx2#`QdmdEy$grs8Kl zNtEiq3gfXJA1;&j?lUQ5la6}^|2CE^KKxy8DR--z@g@1;Seirdw8WGE)}~iDTI>Lgeog?Lhh90B2U}}A81aoT?ke@mDIMLdV>{VI zr5ZLbDq1wxahsNr=NvVfb>il{>JbwWOpTNg6w@iPy zz8$mvWZ3@2bmDZDiS3YILhl(XkX?!8lOKB7R=V|ge*;Vm;B$Xn5Z3hC;s3t&#O|lu z3NyCvNqc-o7++9Vr{kEx0y-HKH z+39H};f6W^&#kT(ud?$3-AnaK7H@dFN6~cJyP{g1l-^9Z_LO~! z??RERuTXeV2-UR=Em~>VVCi$WGO;dBiP$)mX!eRc!(Z`L28;)MfoT*G#@M=Xev>_k z+7)MiBKZzK*tS>k+ZlHEv$Q{D?Pczs_ziUm>Mfw0T7z02gY{o-g9v@cbU&<;~=AIRUW_(9*I%2Gr1-;>g14H7Gf zWq9)25;7C}Mhk)C1M;8?fxkjq?{5ih2X01~{ygY&yxJ5_^8ez?Vj2bPAT+PYTAV3Z zIiJHuo1)!Z)=hn;#hRg7n2=Vq>xKTc3n8KE+zjjpKqc$qacfI)T$x&#CPO7Hog@>3 zhp6|>=Hsw+<3p3pL>1h4xsgZwR;R07XZddEH91y0IFPX>lqmKW2o-@21vjelL)7SK z&77%F;yJgC{zSq$V6@HyM(c#M^gPT05Vq>A)`io8vgQ1)x+$i{#ZrZs6Rld?b6NWb zLU8Msd_v&%oCtlbzuczu>m7!}cSmKiElVj3ToRf(^?j(ckI`wRFz!YCG3p)Caovs} zUgEp7DJ}%Ri*Nvc6#19>c&D(=KG+n)dym~wJA~dPEcjRB&Yv&AlYGbeu=QqugfQ#7#iF?fh1-jt* z>pPHw+fw)Z0|xNG)GGs0*Y9Xuz3v_6y1xCWYK8Ra=>6%wXwH=!F%ySH+S*mjQ47$n zGG5Ioyl{TJlhl0J!hY{aCZX4{`OaLKcWwFVGDXc#R{?Bn{a<>smhmqZ_r@Fp@cmfxcK*3S7kj`5?`DV^-`C z)HPu{Q?JCxAWMq!KVZN7*EJkLco0*F_o5a4?vYoW}YMvQ^)J*7BPb zgMOaLsjj71zcD_6xLI@*voj*|-su>#fsER7tF9J5nNANW9iAYY$R%tW<|8@iwA=b< z!?!`&c7~fN3kWb@Wa6TF$IfRe(OtBV-MZZ!tp|zE>q0ba>q3WDp_n?~2V(FqD9&~_ znfi4HdFA5v0Q;mB>Er$9>=&V*kT#jQ-60Yj=M<`GUu2iuW$HrifUy-dRR*87tem_} z8*5bA#9DKvp@i~r`{ug}XdVzyiA>sZ_ne@=J%&Ac(3@V`Zeq2Xj(6uT zI?0b*)qvV{6H+6sH&#pt_<7u~RcxUS!}jkVU6P3lSH5qg!t!p0ucK}mezp8M21Ez= zt=M?Nn8L}(rfv)vk|H-N(RwT6%v@xvaMa7=5rcxQVrKNtR4xvJ3fVw8L3=LCUtDm1 zJt@}|p%@K4A&X>$8fno!5`q=llR%xSAG(g-GhFMg^hOaD*GJT4N&zAxwoS=mW*Svh zzQD(MGt4F0w-0Ox=g_H~(295;N23hUzLe5R@Vq^hGPYF++P0cq^$jvJR6$;^mo|E`SxcIkqu8!D3faJ!D-re zPkuu;$3L~9+z>dw6?Gd_-)qY$;Vs3=8S_;Zhw$;NVmCsgW*e0r58@0Y7TQZZV{dbW z_#JKNJ|NQ|H(Kl!(l{Dmkj!YW+@9~!sA(@ReC0aEPiO$GpfWyT1!jhQ4@vbJ5qQ7V zYi{ZE(1X;hQk-xiY3fsX@cKsMJ?9(M6n7y;n+DsfNeCWz&SBYmTYzVKvE?DD9fN;2 z!`|WQSBteZwAO}3ox1L9ln)x^(dT7DL=iQdR6YhrWy>gxM#Mj&lmP680TZI($hLte z;&H50Iyzr{62^4yH&ws&VvVH&P#&mZrd-Q$rq4_CV0-!H)j03BhY@J$ zTfC3bRS3e~0G_D*>fBZAxrcV-6Bntlg6f`y!UZ9bvH#eJ_H^Jc+mzF{VT@JRZ_!K@ zC;=OmT(v?`%KP?ayBk#&Opm{r$Po2EdXoG2~P0GhS8N<`S`RTDk|*@+Oc| zIB7a9u{oV$8#AvQuU!z=%xb<#;QKx}g?!%n0jcmmoozF5mHXX5u}Cpq>EZ<< zzbP9iwiX@4Adar2<&Xg_3yIX@^a1>e2JSM|89YtxMKKS{LV9g|W~Sj)KH>Wn$f=d0 ziGjg&B~_hdXp$+# z#?ug1qgyX9_t87NqG)^t?`{qEQF|Cr;OcK z+is5-3+A6apzD|AEi$#GBT&oG*J80^difu8>;-%JOkCH;BTmqAC+n}DCZFJ6Xvw>= zl%$8KtW3+5m*O@XNMPb^f&-+0LQ1`!$OU2B!>%PynL3Nx;XIYmke#%$0R&Y`r$QY} z(ZyOUpnxh>I5!AU4?~kYTX4>lSaNr}a8*Pq{lFtJ5xT7n?gKkLU`f_LE?kEwZGxd^ zaZ?pd)#=M6qnAK_l0fMN?&)rD5m<8sNt z4fTz3{Hs5bPmXKF!vX;a{PeEb_lyp?(dr3CvhOE1((rg?Wm4JWnuD;Wf#l6(IE}pZ z4s)s=Zr;&TWQ%;IQx#lBaX3=H{QIfH(Mnmzewaw8hII$O8`JSljZQCEzUuq*ck!?= zJ+~}onP)%YgY7a0^(SE-l+j9}$tLS>ZW0<}VCrTI??^H*LNlTXdtDHu66U`g4TPFa`Xzg^)^tx{HxErw+gkR(#T z;8=qn)#5FjYekn!2EXutcLC{eI_u4N$o>B(8-rjVOzCbVcNCJ)Nc~RVZLIUFA)kh+ zvN=0bSKDt}{|3h>TX>X&ydw$sws)Qze^ZTf^_1Yr%a~h6`6q4)L2gQ&#UR*c>Fa;k z!SZLMA;K#MpV3=PSEEg_B*RlN?^FUxSMk=Yr>k7MjlyWW%G=+u?#5WXHtD9HohQes zINWX@d*2?jrW&|RG-YB2$KZrki+9zG#VC*;fk?-8=fD(fARp4kyy?Jfha6s{hv0p& zPvD`Sb;!{C-+>HK%*W4WZzmEa+F=;rqHrT=qHILXCMSuJlcZ|nmb;2&bLGB4wEZK0 zzymFlEKxl}UwRYhTykS!Nn0lN%0b@5Y_T@b^K#zZi_h$nqjC#L`dx#z2R!0dBZ)f$ zP?djQ%mqkXnIo}dKbaw08tSjOCCW28Jc>QsiCdeN7HfO}FGMAS#a?>CtzNBLREE>H zXC7nhEKb@FCr$ABnB07h+m`M(-pClPl|8`*UNLKlQld|sb2Y)7gyzeRDHeJ{vWIsR z-k&vIcYU`^N-aA5Q*XgqAX}c|5Th90X~DX#-EK<`cyhWtc!kntm8$4Uc9Mhi^nwq2 z`dWoHBGv}GPpUnn&-CwraJx_%l=xtXsag5G}>n{iNRw+6s?@_p!|EDvFGz6&; z$ha93!3SBk7k_6&d0hvUtk6&Q)->NDQOzFFl!f=uGFh>`LMm(!7Uz=%vH|K~2Jc;s z=|lB=oSp}CV+`N|cPfP|I%%;&BngM~`>D!A-)YP~<0TupH2vfAU7Oh*YvR336}8^* zHa2ohb2-31g`aAJD?dQhXBK=O;ompur0zS7)6^<7wV3C_UAZ4aG~9DObouk%I80KO zizDQJZTQgR3*yfweRUn%>yzXcJLHM?y2mpSm6oZuj?LiY8TO;yRrq^GpzFn^w>XSM zO^89jD_z%Gy2HB{Ez%XwI?N{A?gFo}d&pJX`b+NLGSJakZ{G8HF`O>gdK2kU5|E5? zaclfVBN2OcKd#E-Cx{0v8?@u%V+*E@J%&XeW>Mshk!N-?d<=?P3MmtVvD!a5y@uD# z#^W~QxvB2^?e+swi5-wiKA_!D25HbZWxs*nTRZ{%`IpfAwtL7Ldiy_8WR(mIv&^Ls zYteRgq9W{~Nax)Ww>463>y+WbhXvaGVoO~Lho01u$pr2t+ECMU+$YA`(m#(L+8Vx? zO0S^jn`#d3t-T1M*i%AmFK%g-tw~^NrDamN|xqb5cNq;}~G(Dbv6)n>0$+*U2tf3#E=wMcogcB*ouZb`R_;np5@ucNKF+!@I?WnUnjKs%U4RXB7RuS142}vtC84#*s0Ba-%0L zKqpzimkb%G%f0w*0uiK;BuG54R|+rW=l3d6=B+XeB6ak;%@STebB3YM-iL`qY~oIF z4C#6f@+TbOJIjvnn(nxe+uY)&Ie2kZ>>8)aaIWjSKmbEswM?A>H25Oe9^SVlScXqC zH@5!nrg3DGSfF&YA>lyXLga!(==#`5z$A>5gx)BKM@xdqZ2K4crmgxr;o180e;Vo^ ztTj+ujx@`#;;mIPer+(nl~7j>7Gx*F5y(C!*m#NAqyiF0U=;79nLVYWmewWzZzARY zE1Qvs1=+U4Wi5&ZJlR)50b^W7e*~`#;i@@GP&fBwgw}4SUshy}z;C^KP}ldZt}BW0 zZWL8_cE0+2tGNaRI)sjHe%>2zwOkhg;b832woJEO7Hi7cBdU?c)|h7du3Pu!GSGG| zfNZ8~lo=EK5I!@9(LhZF^i`*$8I2~V!Bp|9U`x4Pqp%NFxZl%CUEfbvu^Ir=8Aw#u zRceUoE}9I2#zM0d)1 z<7;nin_pdzfR!RmSEaX!{GZREelkPo1*p|#atlroN|zwqc&n`O_MD>*w}06}>UFz+ zqju6rVDrO*Wu8%Ri|}hG=}Uo%*0xH0zwU+)NK{GF0sKe##AsXekGjf4&Qk(-E}EvbIhE^jcv`$^;%7 z&9bKuG^U+_0nw7y*UDlm%?)KsDhQTpb^uOlvl*=X*!acw0u{p>IdB#aADr-%`od5Z zi=<;O*)LeuBA@F^SChDXHMcHC{Qb`L=#_BWNKX%j^sRBTjh?d_G$^H-$Hr2=2I0Lu zB;kE)#ZiVry@6CBI8kp?hP6{KdhMqhOxWM{NAJWaf6MYbo)PK6w+yy zea~TDaYj3+SX-I3a^PA-+T^seqtV#nHXh+{B02(Jewz+0-O2?sb?ZBIQR4qaX(f*5 zNccw5#AN3|OC8H@J@bXG9Rk}=#@a=Z;$yx1N>gD}`a9X8xio9wtNO9;CsOVWp17UA z>)+R7{(D?lf*xG)lCRHw02$0uTSp4 z^MI8~xN*1pVOPw1I-0w_2S+DMB_|f}i>TX=9wI{|Da$Nxa3LLK+jU+qpy&{pR9;yN zT4QPGyo1BeeIuKf$YR?dYynj9LTz46iHc>_p z{+KEIQ{e@eM>|m4Xd9+`V1Y#w>?y}CDbFkz7EL=4#5$cp$X6$=O>eaJLK#EKC)yc285XgAs`cbayRmvb?wOmaJ6T6rOo{n+ zDeLSwK+gY~ie7dV!md*RkD5hl&VqPS7Kxm zt?I~&OW~Y=ZuliYtTzyqKw<@Buf}p-Nw$x%kVKo!%D!aMct)2bU#ml=$hC@Ad{La* zDwaHes{%6v2yJ0C6D{{+@0&pkeDeXfaI^3Ddf%|KP|JW$v^Q>YE5vWwF5R$@h|6=Y zI)T_~b-p|HG=u+ztM4=~Io$pH0f7VV{GD4_@!dNrPULwD`fu}az`$JJy}m(|)f>?b z?beP-d_9x(ft;~qT8oqIG0>z2#n+xmBr0cz7nI^Cchol&;V|7l)jY>i=`5Di>Wlfl zF!W7Gpk{CFM6a`x)}Ug3SVcP`EV*^nY-<+O!QeOL|A}OoHs}$(0-Voy^RVV*@k>oR4d!BCN3e5xppH(h!%t#Q<{ZyYm{tP3iJp&K}y;8 z1`zG7No!GOA4p*W0Yp*hL_MA9QcJ%QL~L_#RN&1ZZk;FJgqrpb6So|&Xnpufs-!xi z1tjW5Wg;DQO$~g5F&}q{vC1>TFC-q66BFkHij5~^`PFx|O7S!UiI&?>mlK}SNIqt{ zl5>2BfvdNbkg-n%qj+OT!U6|ZxTzuvRZjKR)4qvwo!F8IsyN=YEUv412H@5DmJx-5 zH4lHW0-rsR(k!}8uB%fKpmm_R#Hx{_Vxxm9pq$@?3gVut6v*_lUN~ilVr=gfw1_t7 zRB7vo%J1m!;6+j1`EHH6e8n=)dv->3NM}UyN$9<55*yStAyhj6Tjq4m{Db2=qIa*WTqX{a`z8w49*7m^HdUQyN4NS+-%1 zpEm3LKAbJ^vwRD)6x)lsf*_YlyWCnWV5}vT+cHzxMziXO9pT!T^^N2kcDBWahweQ< zm3N;9cJcgO+9~_W68zwO1y5_Vm4n}6RGr|Ev-5hcy+Kq~%(nt$zq8pTAw^R%8UXXt3o-_AM(Jes10UAC2%sCAAh~J1^?^sjGW{Lcz~z=+ zIm-A=dM#f@8F_J)BNtOrjAYa)-~q^=k(tPu)z55Zf1+;D*@|~EswZ|HMb#YQ+znSR zs}xJ$L@_Arf!+1rLClcQZp#SPz`kvI6p=(2KsMoJQB|T}`dfugUogn5-d5u?5$=J! zF{fR`&GmzxZGo1XQg2bBQJ(jq``Hg>$hm5cC}qlvgB~EVB2fe*`?C;ck#GB6#<8z| zzi+^SfHVpor4V(Lc1G!n8W~g^%0?v_k7mVDHM2B*Gj(bv%w4>1uv(;A_-4Umm`gP) zr*b-*_J8uAsFKgjjRw$8o#bBcEAs|_vL>0tAKl{;`4I}t0S?Pk$Ef%N4l$5cdS4laz`9E_dd)++ z!I35-#M1{XYn$Z|%I{hU49e<@{PQ%7xKWbOMH?J*}b?$6J3@N8GE6&uWIdKt&WPh8$NXoce%Qn_9 z%G0f5(UballV!>?0Aa*~=P-o#w?R;_YXNWwVr;{Z&}9$l56R zYvX7$`r*?c2~mx47$B7(29%qlKmpJ9xsW{UeaUB(Kg^c zJu~d?G0Jfh95Gt~9y;+NajoExJX6deP`M) zIb_N`-HzYS)665t!Nah!i}O0SpK{E8p~M)4z{2=~IGf2~_Bz=fUsmo@Vjh$0cCE~iuRU&ooM|KSGs`F49zaCk14q(tL{8gcnbjSGH zZhK$K``O(ib+pXQMcQ>{1|UN3JcD&K{&a*kE$n=|<9aY_96EColUW@vg*hA(R0JtC zh`o+OfnEx_pxbhhs}Hl%b5KaTBizT#NY_Hr_`LP`gls`c`I|7Z>dF`NNg|7&Iw@%S zE(JiLspajF=r>o~24MX3LJ3Rf6U}6HTF!2P^W$=iMI}ea_^;geXCx+$DC|rv1Pw1x zq}hG}w{8^-Q^VFiTeW!=^m(}~+F$w)0E-2BD!OQfyL23A-Bu5a%9ydTH5yp^A+W@D zMJLV&d(djHPXKF5^e8$ONcLM-Y7}w{Kbjgy>Gk$HE0G*cEOdR!gh7(Mwnv#wami1R z+68^vfNpuN%G|dyf^c%&Mqo=Tvb%*{0hT8v!C%(wt@DM zGsc#V)-IN4$U~x!4R?C4`>i(y8T^_<@Y{ZCJ6Cn7D=?EGvZFPxfye1a!m{?@DZT{N z%CXCSw|f`OHnFR6NxJ$AK^btarl4X&yOO$~bezc(e+QuV*Zg|*c~?W@`BQ9}FmQtN zM}v%SFww~}J^iZpA(vFJL$&}45!e;6o_+FG?HksE{TV<}e!h4&q3VQ_s*I8lnh5Ah z<)IZc$*h|)n30oolY9()Z1fjDum$v2$P#svt{kDsAr+l0e2nYV?1yFm;Ag3H;Hntu zlyesd1;|Hs;!REsLf*WZ1{@VA-`rhRlW*c7yT!TOw>a0k)uwauyXRo^9MqQn zh6Ej_MtQJ!K73g-xaB@p{7!e_eY5SUUP#!wvmNHcLV^ zT%v^u?mR_;f4c&CA1xSxY#}x(+z}wk*Nh{{H!DmS(NbE>!v`i-65P)M!W|)7k*;a}H zNYJkLs%sm(2ob7d0kP1V^AZuM=@Ab!z}pjf7KjK&gS$o)h-K>b9*I@tj<<{ z6)CP)x`+o4;qe6>qIY1-@om?znvA%tMIp+u4e~2(Wg_xav_YcunACFSDHC`W-d|-~ z8M0!u92kl&n$Ne~G%Zq7&P$6N94tq&-$gKRQ%&bhD}=KVNw5xXC2gr`KA)6`VgilT z-m_(ZVItn{Vgb4U_Av%Cz&h%50yEaTfq^4-u;r~3R5JW1gN<73Wyir+V@z!Y6&*M_ zz2CyxK%|PDDlmjBd(iU!9AAdzAjde-#mc}@RZRfhO7^XK!3xN6zuTUF-Pl2Js>;3} z-acWw3aDDV56U|Xg-kC-HI|#p?b66zMuyOVzozA8evOI-xHQ{RIZT2)Su-Y89LU@= zsu$)L1QziuE@c+AM^>}20ebw_;G`60&Qm(c_x^X3>(>}M!8$DTAq|h_$n1kVxH3QQ zrJF1>mJ*IXb2Q3N7dOewI$dU*HraEQHmvd5=BP}_cXoPEQz)Fuj&bBe1f&pHN1FY>j!-=Ci z?v@TX(ly_+Ma>DWKcm->8P*EEKKB6h161nad-r$2Yi_de18L=GN($K#sahrY2&tdT zT@&WR$YCtgesUF&#fe+0)M6mYq$YuSvhm|S9m2N^W|%7x?jpp2}eX zGy4Yjs0F?irhEf>;CKL;vXr@;Yt(r0@+!a1fTzn}CKsAjHbX=mU5DSa7aBagEQT?D z9=U$8(35(E=!(C4-(L~ZXw5P$&2ng^NWm#Cpp!1(?-{WhVphNLk0wB<>Ax{xoA0T> zx3+EjnhFR!PWZq?JzCkm&MA`D0cvMS;OS(~G&|V-`^MviXt=9Q`ui(&; zNb@j?M0?Jjs3Lc{-!QfbU_I?NaZ#KIp>I1DkdiKG2teWm=6Y*!vLHIG<2RU#&&{&+ z-u9IJ%qTuo_Puw-+P7TEUkkD4&jN#Tw*rV+6F(<8vTRy>W&5imG$~hQ<~u_#=T(tf zdZE5Qo>O3L$y}w|jNIJQ>4H~>9utZ8kOjs#X(T{w5=2|6Iz)7xlcJBgvi8gp9&vkx zKaKn{VEQJ28D~&cq!B~Nbo8M35x?NcH?K&NgC_hazE~-N`7YW@evjrn^HCc%@p}m2 z@xXR|Q!^NirciZy!AVm%5-bvw1qqUZL#~UZA3r!$v6{U+ z&$e^*f4e!6;V(awEo(Pbyh1L2_BkANB2Ok_gn5eGde<$?{-ug*J!_;-4)V}BNTt?P z*{slWr7G?~XOGmCTWWGUZ!HmzCV+39;_EYwE4#IUrees&G@KDR>aWdbnm3Lwd7bQ4 z6!+Fi2tx!}l;FJuMT-PB?gQR&oB5qieDqlA*75Ep=O4;i3UT+rWr{U&6~<<{{3tYC zYV1F9Gb@eP;{h?0@C;^0+C2(aPE+CF_9;&*ma#W?Tkdn3U&u2WM+3)rG3`Z51_}8en;*hdJ5$y z|50j8iyr><$2A;qH;Lyq!+SKK*WXmG+reUBO7S@FW~yPZfC8qOY+d3Pat}b3xWBn-ykb>Wc(Hnl*?Qc^V&tMxF0ZWdW9u=AQgqS z1((xV#fuhyL~fuim53LPJh_LS^5hk2x?(v-7yASg*FmlL-J|W~GVlACD|N>4<1Nc2 zed>lf=|01ZTC+$6&s5>7wo6f)2hD71w*@WuRxsQu{=*9AHER0fSu=l8&1*vj*xRop zcLBqOXC4&pCvspFWoWHZ!e!M~wbEBkbDWJZ72rt&T0qz%qHTIXPgH4?Wgev_A6_7J)5!lqzP#H7E>yqde z2F-2o>@T#$-$f5DBs|#f?Pl#W(`0{AkB|465_r3N-0CYtu!byqJN8q=@l0uve3I6v zL}y_9GA5bw`nBL7N2B81j8{?vvPIW&2r*0>sf!W5U1iX5t#SN>GKm98U7nK&j#o5b z^}zs-7sH>(R9^3M)3CSfIDkozEK-O-;j0h5iHv9zE(Mcr>y#k*6tN8LX<&B(KlCn0 zO?1HwDb0Fh_O2|F^jst6qa@!p75R;;1`-l* z(LtG&*1n`Z;;grHd&&;36f6LPMk*K#c%YHflXAIHF-Ec>(WcHp+}md3Q^t9M4tcIZ zU+Lg~^Hk8O^QhjTlTM^9o8+Agl=5#1lsw&Q&EN!+j9+xJ7JcYW-vdlBI=9^~z9ST~ z?TWrUvh>@?hs~y*+vC31I<>E{NQ=JJg_7NtCnN>SYf!U{k7p`HK(=duV(&GW0K)_oIARMm&Ik zE~*`^R}`E#D*%rf-dBJGfOyT1>-srSsvuoYSZv95oE1DXBslcaCzq=%F`mI1#K06f zb__zn@B&^B!^;U0tD3`OIwm$=V5wJtW=n|4onN_!u{u&(yiq=n_5IR8@0_;2y2!Oo zI}JFd*RSk+8xq=I(wHZvct(&0qtZ5CTLGlk<4H8!9-r&~7Lei9d-qk^t{VVGhhZ3m z9i4y}J0ih;+L2ER6!8`bPVNyRc&hlJ+euNhTBqp@h)=wDV~_5+>>qNmi4IdFxm7A7 zwCS|nD#RtCnWzkO`o-1wGvt!9-jnk2UN6}g9?G=b$b6#X2XU>q61&`sllc!I>w_@) zqu6An3PllcwsqDC9TViZa8*j zkMlqO>wkIqSHZuGVJ2w#5GS-g5u(e{POCSf>t`ft&D% z^fUU*jpqYxWwrFpRQJtr#s7Ka|HB_cpWdEzfJztC{W!Aj;`s2}K&-4D-CN5z9y@z$!`B!Fjc`*_n4bGrd zZB%Ci!_-Akjl~ofx%={vcLzhodw^-(^1dW$x{KJr?fZ!nJD<`1ofrRb--`X-R7UgL z6LoSB08G>XvyWJjR~N^8r@$pN3#^ASDW9Rr=ZdKdilC>eb)JKF&bJv^nKaW7*J+~Y zc-pz${c=C|Vte$e-&5gk1rzwjgR4jkiGu2%ei)(K9S-`TMX4o|ZCweKKcp^Yo^qSE zvl%NgpVr%7Y)x_tC$k@uo6xk(e08UTTfM<3sLeh%fdM|>Dh{#vINCDVar}RC<6Tkt zk+Tp6Gcy{5ynKfxSY5tRZJpRmn>?oPwh1vqO}8@W6dR;{umLA&1;R^x*yYcggzAcM zg7QyS^wIAX#Xt7kaqeqqE*X2WwvENNLGMTIR1nFcB!4}VcQoT(r-}&%J`rosqanGZlO+*HZVal_## zf#b<0je$BCDjfs2!8c3yU)S$9|I#N?>>s}0d;HsCgRHEPo19m_z;yZ|P#LJ}cQMtD z$CvkJEUwP&6$3slrggq63kTr+a3DD$l%D|A{Wgj3fIi&&XJ=(+oA!4R??x&MVUVZp zHH|pctlDCC)LCUaJzKZ1ImlAXqfQ^?84hIlzxP%^d*kz9nzVL1@}S|`3YqeTtdXsy zHEWh4WKWjh;`ZP-q{mYIPiO4klW=`q{NG8)%p1gwIE-n_aBgY0sJmi8+o-G4n1)3W zD=~08(qmlK-yrx|g)W?blrY6+XQXc--Ai@prlr- zcJlc^`JCF>L-&l?h41ty$T!&5Q|9@bKMd`^f6c?4>vs1M9Kkc>Q?nwOts^5D)Lz@M zR#si?AOI;C-*!8xMJ==YgT@QY@#Y{m^= zCSROvL(T>pq;>HO(6}>?TlWg=3x{EVPeTbNQEyjK5=8S)LGP!(H%0D!@-gk&t2xor z-Bt<%KQg;euVG~yVfz`)0D^V}N4C042Vc$JWs9~@D)&>?ulseE7&Qjf4s|+i-M0II zXkE|UoR2q`+sBBoZxSU;*!K(jKMP}~l7AOI8@^a1k2AXT31YkQ1rv2f_fL~OBPhfx zt#7`jO<}slKSfOs<*l^;_5vvRT@-O%<{=as+BxJ}_ijG&W4e%E|7^p5LUFl$t@*%d zBEeM8K_byamRJ9;Kji=4KN#!7na-H}aQhi&?ME*%=m6g-u?CdanZTVe7pruRI}M`x z4UOpw$swa{4QLe9QQ#c=RO)IMR zx19G*eOVXIw`*g(u=r)MSS@3r4zGnXm5M7B^R+k>dveRBK=`y%>D$|t>v?;9$Pq)< zt$t_AL6hj$ZUcF%1S)OAdJ-mNT-*U>se=I z?}$m*HO~1`v&<6nYrow|_wtG3&Uev0%`(;vH5+UmF;F6*U{0+3C7)z4|BpylzWI|( z!^8+cP|%j-ewmqnT$dSBQoHc$zF-CP0$>scYUhU&KbLGzIEP+SQh)Tb!%FjaLVchJ zVc6ip!LPlEkWw~pEb4&K`qW7_)sZ}P;Dd~{lyhRMPcB~f!izib5se|))5#hkm9wRCdG@O?{NXi`| z;Y;`xsN577l=47=7SkL49!Qu3YAYDNuZ|OI77JUK4e$BcrU0YTqHc&o6-U5r@r$z@ zoki_oFx1&ivRCrmVu@Zy(!wH@sr1P}V4#4SMfE96{-^O`!u($W7Wa}VHJVzMI{^o^ z9N{WuEQ|mPpz;qbKHsWpFE4zW@*&8WM3>cKMlI3ZEz^E;B{?`!7?b?4$6DY7to9*W zqZCF*yIFNazVOGq*M;QYQ)}AD01tWo;HUTYHAuwApK>X?I^Qg_5}9g2T9K$&b#sCyvH*8sm=32{r37EUHX3&c~w?;Kk0UC=h<@idh9e zHa|q-o8{or9&oLl6WewYua@x;cQc=iwJf0i{@lu3h7}_ok-kdLVPpfF#iAsE^>U&9%Zd5fp2;=ya++hdSvs2d4!0Zj`KVtsX@z-V#JyGsz}JE# z?8fEST>X8=3Mjw>5<9v4F@^E=H<7jd--s;pS~xJ*1p?-~7BmF`<1efAzB}?TDU$84 zJPy0fF{yxU0Xx%+DZos$;^fCtOW@^H^(irch@GxkfNQqBMIf=9Mz(lFZErD6`rXQ0 zb71FwufaQO3H*0LjNDW#SEs%12E|1?sEg&EHa@wtF965DfxTZ{+eoWEO1Y!j?vlRK zoEr7;^0Xk*Hck6!WAGs>(~p{~2EXe$E(74|!uIl*b#!2Ah^B4%598|P=HIz~+v~rD z)mpgiI!9FL`pD`!u;oIST-^$%IT@2$>sY&-H^f&h%_z2yu65gGGrVnly-Tw|?2BNn zourmsp;uoLyQV#gaU*u`Gbxyj?f7iK;O6p`ivIT*6-daRg5555>6(|=%?wsTy|D32 zZIy8_)McH7&gxFj1dEc2MyvsOu2OZx^9;GFL{q?rBOxF;nEy2qW8k~-O6Kpgk+kOS z0C09e43F+Lr{(MC*Bph*(@3@j^U)Tc3}v*VO7N^aC3_50(74|4BV+LooIgw zL@2#(0UX;Cm3wqROhs8S^1NZ5#qp1?(^IfOUVt)Fqcw+#VuX=Gt8l={%+)0w9aK@_ z+JtB`Up_T}Z9VL~#_%q1EZ;rX6o2?Ke5HE(i-RMMoYYTbpsy-|AJkd(~Q^j>3Vi_M+Yzt09>~vI_-Guy3MPv z3}ml~vMMQXu39&c)c$IwZ__B>F*?=nPAAdAl9vFF+N{!d$0yOTMCyL~#BMdDs#bCm z_XPs6Ix{Lwu};U5U;PdT~C!DChr(3&ziK&#W9KDJyb)=0Y!1@uD4xSaRz1kK` zus3qxXV|gw9KWQQ&e{UtP*TMse62Kl^NL)W8LqF0ssNfz6_Pz!_P+M!7E^u&T8^tE zj`}U8B+lFVK>d2Rmx3VBIO*M+>)~ihR>O5*Je%sV>Kpa)6(iO*o?>zWRavh$i3Nki zY0sj}+z@l;6e$DkrPT$bTsf>jOKysf3r&`7eaik*7xXaXmq!1Os!qhQbSSz07%U(Tq^6a;Uw47|#e~lNsRFw3t+S}IkwEZ%>#^! z+CTmER{&E;cLLR7FeqZq1hGttLgdS7aY-}pG}eAAP&~%0B6D<&EqcdsH0>Yewh@0> z07JtBT<^ewmj@BRd`79uK|h_{9zGdh`e%v>_D7+Y0S-_+)4QDv0>O=fln{r@78xi* z*_2z1J$Xv#!qmn2Umec#EmwLvCTl5$9ZqTwT)Du&;N->Ew+*Xlcit>C_s}ZvH7TvTd$``dBG*)H=gg`O4v(H`!O^?t72k7Ha8a2C_Vv+ne*FXwMnn2Ts2U z`^7A^zb8KIadtk8f$wyo!)k{Lt$l4czjG)P`yOnbJx|W7SRa_*>&m>`96|VA8E#9> z+S6ZhUouB+t;~pjf8!TPU9qjf!x+FKZ<8I$uEO)0N|D}hq05yH6gBx4P2yNcf`_-? z{pEhKNg`si`26wy1^EF9Yn@@p_;y#dp`D%mQSnSzT=8-Ryngmddt2I*iHC=a`RrW^Ul7O4d zaW{QOl`L09p*eGnS+efIdhpD_-kWaqb}WuuawKmaS0{nonB8T9#fpsj#hA)<9ZWSI z;e@36Or>=?gSC=4b56M!VBS9TR~A{K&4753GWsqpI5Df^Je#yt?X&Q#&DknXs#9r) zm04J$l?}}FT9nE>>mk;Pz7vjODPK*6iw%CT{qYGhR_&(9dTPxvKscMa#!CwnNp9du=~_(AN8S5`WA1|1+S6q~I2$?iU!) zO}T&|67~U(r^DIF(fOCp{+?z`&=fp-e`Ne#>%YF zs>o#2oD1Zj?+IzYj|1f&&y4?)^lW^RX&SVaxHcjw66Dq$ge=t6WrWYRi2pkJZW)*5 zyg_fG=188C%CiTF_<5jrB5vnq=~@E&IqU}go^e*c{OP!>4!6&@StIyJxva|^QB}5r z=`qibHeVwoJ>M&`AlE~};jO%Z-2}mpd_nP(te-9J=GV7Fe_-wQMfY=3p0RnU zh($haH7c?ihg)Kv|03k>PPp>rzYt^OVl;$0`c_}{kRoC0-J5oC6nG6hLoXDtYa%J7 zm*0Ks{kxWfZF63w&$^~QGS!og1LL?T%l7$+nu}?qCZx_4e@v{3(%w<$#a6c)kd8x2 zw;6pUb_RR&m6?xf{BR1f#qB6JnSTr7S~91jS}cApN=( z+n1tKD@xgl;lyZ%wxivs3z@$TOBc>5+#pUko+p&Y2UzusM zH72rExzA}or7?CfP9%`D!41X5Hq^y*Va`&G|{u$Q*Bsj>YPs|B^%zUeP4>6zvn#feEaO01# z4R3v}v?*;@L{Q?{CL*;{e}AdhHN+!B_cI2CM~n9?*=$D4&;hb7%El&`a#7S@)^gT@=5i{t4X;$QxNmp#v$8w?{~K5 zVIhL!8Pe|=ia4^TJ6|k9jVKbGFo;3SRmU6s@l;IChPxW<5)$bdfnxMMQ4ajvn`67# ziIsigPujeq@t4(Tel`t#1n+?53}K~E9_uWlWg{41p&2TuB}x;cE< znlUFrGwN51B=*~qO&_rzzE-xM3rQCC@RPhSq&sllvDnY60kexTo$1bAO-vsWm)XDQFE zR%y^`rW}Mt_m{yDfvlCUC=_acJ=p6mn^`pwyz|B6q`ndGaE?m<>a_JCA1(b`$)pz+ zlQ%l_@@$A2EkFojL~o3N~faHITSPwnHTy>vevbjQS3TE8ee!D zbSPOzWU8m9?ct0}X&$RTM68*-lUm-`cY?5?g0(ENrs+2gnS$FGJQNU4wu_l8 zWIrt`TW^-rcC<&+>aa95;fdjy@HzBk;vUb{#Xj5S{M|q2qA{S=W}lGgSi|0!PDMX! zk4{kh<-wh9TLmt1Ic4CAt1LBXv+g~DjC#ElCCtyK0rqHVa6Fd`xmpP%%^5n=INA}R z=(BE8S=GqT9jD81{0$jX2Z`?O`C<&2)r?n2$J+3EwV!BCQ-s2HqGqaToIEbPNm-UC z5}isWaepz_r??R-EPj24DmDx2&X26pj`awVU=)wR5cM*NOg5!S-6{Di+ST!z;;OB( zXO;tsM!@h}6=R#10xt@fup=fcJCA8)2^({p$x&t`2hlg>%bkCpY#}fH`a}GbPOEVe z;Eprm>pW#uR#>hf=dn#J{(C0yKZA)KZMyf8%Q~Il%5|@Q%|+5q$20C%%my7Q5u0%L z5g1tDkU*_?Pjl2fw0~z0Y z73!9DCGu&%JbCZqa=NIJr}Bgj%PmCkWmQ)^yHbxJYgq*kWSr@$oi2fTeWaS=Aq8Kj zQhCE6gY6vZ=|a%y;L~6pJQT3=@m5r4Ub%8T`FE>S3^rA1_lE?&_sLjll3b^Q!p*V4 zZ?s83HBQIY`^ddgWixhxRlz>2it1VNxq0u`gy`Px6A=d1qjTb!<=0)F#S>KINjG&^;DbhCPKA@lt+V@s#{(&I%K1p_dm~N4iq({@ zu>NH1+!ikpu&?ktEX@k{iu78sm^xO|mm>)&m_|DLFZl;9(Bs%GJgWGn$f3Hp+*HM8 z6SmRq`7+k~nmYlV24_RNTb0E)+4b%I+ZQbpT0pX)OpJe_Q{n=^?15QW3+|6vV0ck% z(D@OM_|6}vTCrDuqt&R_9txEvl+>>354xihLN`P<%tyaxOF>^WPaSC_ZONDQG=SzN z$r61kxi~@(NniG>7aQ=sdpTYpu@eX+sz*G=Is8*KWx;-e#~d~4?*w;atISz8;APsO zJ*`QmHe6ihHdQtoeG{745}iRd@2R3eYi=jIq5=plGLFqe?HzEca`}!Dn`2|Po1AM| z4V!7?!G>)gDOQ?JlZ|PAqo+^cf(IIfy}jT`ea*n$uIQvK(_;`)IvTl?22pik8#8=I zMJv?GK(cCmsd0p&OguHKbRk6W(uRwI1&MB<=Jg`rC#LOQ;kZ4MA2XCC*R9$5nh7ag zeg1mI(d5P(2N%wh8TYsfu8QI=%ERV5HEDnVG=)vax#nApvVdL;}N20 zeZUFkS$eoSvrUqzRce)Xa3*;5ynEqmfzYu^FBb*Wx z1yClU7(@(POEc<6QfLmGuR`T3O_l6TruiN9dve;HD*5c^J$!1;j{uL(vZ~i?ripyc zMH=CP?r(>l3eO%asar@EpMR|~;5&yF>ksWvgJ0wF*hr1v~QTy>5ZP>89Pebc8NGoSlK(#!)|ZF=CmFK3XRVTbQ%mw zMdcdWnNYl9jhto+cEjGl9j3*HGQ$M6`7&`Vxo|tH@Gv05%Zda&e*9c><~l@cCOV1G zO1Vr0siay7JFfb;Z89ZDvXW}`$P zCj6Elo6!UxV6jmO^_t-EZ2?i)2|wBkTgkKe1BOF&M>Vs3II4~|qUWA!MHzG|S6V`S zn#5;IOpeV#$=>ECwn+RDLiNkulpYa5M~`SBubuOq72XI4BuAu6vR0_3Nt4jElG#Ix z#jo;I3nlK-KiBOU1V^-zJFBu(Q(F9V%ll*AC`{~k>}5}@2TgI{gxBW!e&_^?*|BfB ziAcvC742#1-z`6{q`ndke`<(z>e@?AV|AiI>9FHFDANnF=!$3A+4|-E%mbevf4FWD zd z(1x;>tel{ZYMCHgHS%C!yCShCi-=meyf)1xxd+#~BZ{GIHs-AAyyB1duPDRsqO<$I z6`kEBqzZ0R{TXrfn3DW+C-cROA+M8B$Uk@C0MBz$Q$h@2*bzlrO%Im7$3yBRxuKEg?vRwUA*E0P{5TKNoPq}lj=-nhedjNz_akWJOv-6{*gePr?)QrmjY*h_su`)1hKWLcJK3=;;8{l5iKteeU>4poB8=*xOlHcD&d4 z`MF*zD_MfirHu zj?On|63u%CQ?^V}alCI5H-@AsU-}cxPGv1UP)Z$TmofS-UE%;-#DJK*rmq0@L^KE^ z|0N>HS)?NpM8Xjaf;0{}0Wd%45tk`iT1!)jKKK<+()J8o?E#_&sO&x7W6e!k_W(l~ zaTJza4>&CB()s!3zg)E!9TN+0*eknDRw@mTs%lbb<*S`W85S;|9fWe2i~yIy2teN) zD0;-Q_1%18 z*F5EyIWz5M4FfJ#QUe&3({ahPnk_}GA`D`I6<4N=Jil5td8fmzh3n-5YjYa(St*xBJs2;>5NT%q0Vq+n3cXt1J-ec6GAQ`-^W0aF}_-E`EBs z+@7{om}36BaZPOXhbN6Vr}kHMyc6woNz~F?wd}t_3-hIIP;KZktiHeWY6NuSjDne* zCx{UBMb=>~nU{yVFXGvccXcMdj2n-n_a0iUBNtO3PM4GRHthx5uq70%6py#2*a^^i zB^D^|-Pe84|BDx%pgm?Cy$jEK)2bT5O+7z;O-eit59N<*JW$1HDsi~b;aUBOFW`)w zSNqbSvcw$TKDDENPPeetVqdXY7ane7{@kljYo>6D)5-QHs7(^6w)g&*nj zPjZCuxn0kvj@fL_OizK^fY?lBX9*gZw+L^*bU9{dGg|fTlKMdH861rop=M11c^#z_nCqAT2zu^%GOxsWKe9Q=@J z-k_GhbB0bf(m$zz|f z_X|p(eS%y%R-{f#mmcP5IeDGU(?KdFyR|ruO2vz0{8Ni4Vl=^m0%!unkq0Wh8csc0(6_PktY8sBS_fY(e$A=dm6k~hfqew+DMyYtFK+SDFLzn7yqg?<@T$B5@@CVQi zx3sTIs$Y$akcqx7-BJS1BxkfJ$CN6M1hvhicZ#=wOoS!>&$#?I@#cmEa&go6Tu z^jPb1zPiN3l#r*Kp=5`H$NagEiSw<|iO2mJZWT1)}H$J*Yx9j`gX zrd@aIOb+^^#~`fF^9F~#_G$a&b@%=wCf4d_ij>Can631LTFc8sn{N`O|le z0mb~b6HKQX-ILyj(IEm0=JrSNNa_1l%i(v+CYeMjTqEPjFrXR){>)?7C!bn5AV<)6 zKCOFCu}o{P0gcfGU+`#5FI6fNZvbW{c@5ucEz5V?s0(I;g!iQK-S&n@wa!^Tb}W_n_qat4l$*tb{%=jrWX(e1M9_>tf^QZWHzG?2#xj37%E+QQygEx%N;+ni+9b^}O5d};{q zSw?b7vmUm=*KvQObn5J$oA28GYX|aAzbnpdECp!u#VRTYuYe&|1@%>PIA6@WmjdZf ztB&??W_8Ylqed>`&IxR;l|@Pq{3ny}1r{?&(U&`ZNOTb*NS!x*C3FH0zdxL#zKQ~9hedL&_z_tnQ-h;_Y^Kabm-bh-gj?#VjkK9R#Y%9 z_b=_j>yYhVhr;jq$9w8_(u|V)_8b3+4x~P4KO(iuOi74N&vQPeB?7d}JR1e!NE)CLxqLPA0i9<-2NGYix-Q6WQq?CZP zbeDj1cMOemcegYQ3Ujd*8M0wbnhqF`=X|dU68% zDw%`3@)U?mmDwMSpW=FxPVaw|Iu50L8M_&F?%% z;LnGArog7wJL(8>D=MsY8+?5=U-}vPo}uqh2(q^UKoiGH=;X=V-mkW|hgiEpv{_8k zz?ee3^qTg69zt(iV9$TbfJgtQaPrpXFS~E4Z|{n8T(jm&xQPB(d8qn}eKtI z-WOe-A=NeOb#q=+C?(H(uep}RL0%!=dd3HuWq*2uB2vL`|LKRO^3R8sbo8mKKmAs5 zGv9n$R6F_sFQto>Gm}z-zn1pTjHdeE2MHMRL~oFlE~Nv>kPqCpOSC}al!KLL#PT<+ zXQ9Wg|7lq3{rzq9ZEl!5PysaVbfvQmPRZB6>i6^3liO$tzR&NtH|ud$e`e#w%S$?E zS-WKIFGF+{T%-u)38K)w|LiIsyu>8@>(l+;ZLU!s1V#O=KI|p}#-lWLpF5?8Ko1<^ zQgv-Hv#f_nO9AZg){TdM@Aih=DW0SY8TM#;_V9Y`p6L_F0l%Z8HAVQ}Z$6MtY%B3U zd*%cTfB)?t$&~*No%b*P2aU_0jQswe*{^r)xhF4%{ChgtVw68+aUw2m>c78e`-}Sj zvuz9ezqV})|FO3E`9CeyOZl-trInc1?GsP`exKIn=KuZT`Zp2&*OmeMe;(EU>nr=8 zNA>@DLjPOE{XdWDe;(DJ-Z%1pa@Bvg=KTNvQLWLJxg6hMfOi8CU706hxB?*D@vDNNDIhZK1Q1)`9Z|>QYUc&t7kI@R2!~a%jx3y#YXtM z^ZlOpqn4%bXUkZuNP+sbd?50wgsL`+aJllrkD8-M#l8sVm<^2;Y3- zcR6_F?tJ^|V7~f0M~+Nq2p#^O{Q|;*o8Ve?06Uy8h zc#ZQTq4f?$MW%U+>OA4rxMR%lRR7odWl!rJrH@#7y1@6NX+_oOz_`u9bnB?C2;pZy zuly*gJQ48g^XUhL&3ui;uzmUcTtS##z}ZM}*+BDd2?@;!q1UX*rY}5nFdax@L0&;< zz(d2Vi)n|?x6|Xv)%9t7=zLp&jof7Qxh*SA6M#i0_DyWpqd>1Fo+#EhzTX*buB3=& z)y(C+4Zy!6T;@=}oR$#UE4uBazdWTb?-C|GLK#C5DC{0xV@)3Pu%2&_1fTek5IqxvcC$;2h>eXH&75IRFA4-6Z7^a#SIIV5u1gVn7giar=!9Wei&3C z`m)M(G$qJdtxd?PmdX>|SBi6@vA*PNcUIZ9?;Ao((2xwI-uqB}(P`Uytlp5EQ%38B zPe;<5oeJg6Kx*$87TiaX=1ScXF&NZr6GZ?a_k1m*1>oJ&yC5M|$A}~EU$LsD9-SX+ z<>kO73UIobi@-IZH~f00le`*D#+8YscXFr*g{VN)uLR%kZ}xZohi|CM78K<^kx@=U z^VrR{ms|x>RtclA-hzd44oA9>4M!VB2u<}iz;rb)Ytq<_*;v$!lrP<3hYSFPCZAn@ zg#;}l@GUlp9S}JL20Q>2NxjA+CmiZH&9K=?qHr9R_VA#eN8d)42_@NYNPkkOY4{3aE8GTrFJ?LW#r>1gxgbX?~ZQ`W+ z9L5i|8-J|CVr&3t=uO1;-Z=wPV-d1Fp}|#GNL+QbvsvEx%(}JF);^d&+SSKxT&_A+ zbY|wh>)1&G?%gNAh98aa57UEVwD0LjA z2LNR6`g^E@?_b%G@t&wNSWHSTMk%kGedHuE4jlCOxI&7-!Sk;JuMy7Q_P^rC2ud)3 z%-`GoFk^fdI1Y?#dej*2w9hc8@XCsV|B4!axWUhQN6F^*y0w7>2j>F;(bc#TW7Qdv2xvrghr0q!$ zWi}?pRz#De5zxIW3<+_kfoIrwR~3tvI3Uw>9m2w;;Y={l&-vwL0Y|iPK#g_b2ARhKtdB zDv*7s9jTV4kT;yb%BoR4eX^Q=R`V4r?-&2Ixsf)GtBo<`UI$+eo6q>FR*f1|-bcwq zS{+On$p;fN?a$;7H{sMJ<)!352;=)dUN=6sw0}RXF>NeT-ke6vw_~+;xCc(}p?{VD;{;Gz-7NPxX3jPr|D2^DdbZA{9v9XptME0gvp~?%W)Lr^Vz5rA{X~akxnvf z5&_lq(Et6~4d6|`bcoSB1Aul(=c6T7j8~9PfZi-at9Yz0?!fFo(PPP%r@GlfnHe}x zlTH>@QtTo1IIO}Tt~j!9-D&z^6wnB00x`wPqFNl4USk`dB~cQEMO~+=$+jL?K2z-w zZ*}pTME!QOr^0^Sgbx;@^xWmpA*Uiqn#nfpG<)M+XX&+96c2#g+d@uTmYmjb7~TO} zUkrKb)oPSXeVfG0~pQZGO#ec$2%-q?Ms+Q-%RXegKsMM|xu+1$`Atp0Q< z5(K%@@KUipHJJQb`pL;txefJW-nY}anN!3ObSpVTIr&DT1$i#udB-_dztUpVx?r|- zYz~hXt28l{6$=)XV#r`kb+hOub$X$>ACc$zOt_ERR=Dpx%u4CytxNaq^N6JIY`56G z&4J1x@tB`&8^2Qm-+AAVhZr5q5pDwQ6>-&xWeDk1Q@Dt*`zK9(ffo>AIp^!EWT1yh z1{#X+pQF=pB({CW-%D11(g?pl5hfrwQ$A-t_6K6^31zrzV zlPvG&Gwm|(7+=>njvfH!mpyfRUf1kg+ePXQ#_PGm#@m1qfblNBvbZ)b?xQBHaDhW! z_)J;AaIFVZ$da2T9M}E)`W0EBrTLA_t2hn|jY!x1iyXrmmhCHl*JP1-TFa@TyHfQ$ zL6UL!vOdS4t0M8ru5@s-Ulpj<fok+iRJUx;2jaij{{~I%w2oolk2#PGXz#b1`}@FXn9p=v7M`V9jdE zPP#7bI6XZUl0BTKq4^YA_9KCf+sM3M zNw1TIN_^#n5T`F#8`nMjtw*Iu=r+}o6(PP&J>S`SrueWxWf+4SDD~GV>I^T+aB%|d zG2J96+qV1lm%8cZlHXj4Rc6EtNwK<>!UoO4#p*lOfDs3bUm)gpr)|lHCw(O4>C=Mlfe@zC)IZ2Ef#RnZkrABNqND+_WufI(y;t9 zFT@(dWkOIHM3@6dxc^|WV%TzT)h{K3>6kwiWMWa<7HP`dHQ{#)4knc>-<1F~%&;B^ zH|?|eCBG)ZF{Me|Rqg}+9Wu40IwJGXr|jbL{3qf^B|rVdmV$`L^;*sMElH~1{c5B} zOY%!Q%Eru_>TMpRnrhU2!vBsr21G(UsnS9hDyhzn)snO{{*KD+XW|3|_G-52(g7aC zEFF0UbU2$`aH)qK6-o z$QDwW|MX%xQ`fbL!zhvzq%#4sn6S2G)mfYB00^DZNVh0jVg_ne=4T^SOzUSM;dzbl zDeGxVQdGvO%`!4cqJ}$qA8Xpl+?SDq&1Yzv7yCUpTElc1@&i6HR=E8oee#I)w-hBA&_)Ixn|=q z!%uEtt;;wM=r4LNaH2i?okJhEWOzF7&UC|5B;8&}r33~{;2vm_zXsSn9y>-oNTJP* z=*`JdP+`0i#)9?9%PR9}^$&1t2NO9J5E)&UV|ei$x%`wch`K+9iHR<-d1L?EML=1O zWU-qYd8b^nn{(Jq#KW>3o)!ApoAGm5L|LX7I{-D!m7jYE6}kYOYh$*%9J81Ug=m&T$}i`^DT+L+B9|Vj3kPjSj>MPnPiOAjC0vZU36y6Y5hCZ2 z3-<1Qb?AWv0!JY6h15nzmjK@-Bt~$PY01-r?QrVD8_%f1TnP+L((!q2zwQ|91CrNq zFV$rmoDXTpbXL3vImWfGxAtYJ30ogu#S}H=u@Cb=5VCn+xOaPBid^w5Fz{G-=!y&R z?TFeT<`VS3bD9#EDz!uOJrxP)+Ygs(<71e#Y1zVxMQox8Qq#V7*+TZ*)A0ocCy*8+ zLWR+9)Y5!Q6sAd8*WIlht=4OE8or7Z*)D}7BAWXF&fnBqBP*U#OVQzv`p9Q#IlhOh zgXcsC9&fz93mpw`r~Phrq#efnCD49_Pu=fFSWVQ_3dg;`&RKR4q6ZkUUX6xJk5s5d zrx_LkRZDbKotfKKfotB(xMTy7;|^PAa)@wAIpdu%ypy8k$0!8%EtkvF9D1jk)xTuy zDGS0|E=IDo5w9u$!gMaU8P8|k!PD)XGR$5rp|uzMHBfJR(Pc|&ref5D4zJ*0(BfRx z(aVs@8DHI2A*($Gs^}%ps}Q=SlchogrKyd(*NWZ~wvpTVk>G7gp5r~+Ig!g4*@TM* zhj*Pv+AgnS0M8#ut5eN#0`T%D)m-fF_r*{w4q4tzpQw)Cwca z7B=y#!gvTaT_B+|R+DCpMJa@%dm3uH+U4(-nPSC8{R1I9FLu~~5JsZn>wG?@U#wle zB|3TVbSXr#3HsetsQLn*0DSs2@EAem`suhUlVpewVa2BazSo9(S{ehV5Ug6q{SioI z_2P~SE_Le8SQ~50KrIR}nLRZB91jRA10H#;Jod2nOmfS#D~9f}+r6Q-?hZZTVT&MJ zOl>?LK_iiG$SWYbJa(*-Yd&87;ji@gx#Z7JnJIW)`Q9%+Myh)qWjqh}gNM%_HA#i| ze^3PxQZc%oJhWxQ;zjBR@?C!Ux!H!5+$Tm@JF)bZ%?|J5uR8z^!dxmS-%6H6WG#V< zhKV1y9lO$r^LS$@o+siaiM*!u(Yy}DjH(TUrNc1JCF4QQ*G+PJXQqhz!2Q^SbN??+ z=&p(U+khURumH>B3E?mU_9Wm*vDBP4boQOp34G)kA&xh-h9R2Pw4E<_6g?a(bL+XrVde@ydR9wkOh>OqbG#dm96E;r;HVM8Z z{BDSl0i8!PPpgIo`H~&wZleULK(~Z+&-ZV%kDcl`&4M`a!8>4Gg^%SE0nuh^S`Hb! zb0OTC+kw%5RfW>ZYh$QdeYL*fB9H#JPbtf0V!Dh94~Uj%TM;V5RN5PWJXwm%I*uf2 zg!@_CrTv})F=`uUjB4HN_zP^Wb^P$MLBTw|4${pgVRNfar-6a#PW$YQ@<-ON{x{fc zcH#RA^_nf0=bG@QHZmwlH-YNX$hFWxCSglsFB)nouKXL#zMQ6h8r`AGKu>_p@P^8B z!;3*f3u4mvR<}+U_8uUzO>vKqt)?uuAjx9kwhO1EZ%0+7@WXzGq=?HA(0y%Y96a=% zXr{`1b)#socd{-iS@t@@CsllRqT7kse165IakVmcFY4Iu3nR-m84=Eb1U@N%q*}t0 zxSa*|6abiBF}!X~V6}mMp;)z>ShS@%yU6iCKqW7=sNl8!?YBKXW!7Q#_0Z%eEtY4w zlx%iE9&jYQ;bk2}OavPApl8|1)Wk=rXnS_y()L|Q+`;1KHr_O}aK-CYzHuBt;866) z7ZGec;4pc_A06)g^wi_Daag?GYBE~g3!Cfc(_5T{>{T5q*Va9x zc+&tM?d&5pDGe)RWsqd8)4{X7`Cik}&GGsac4jm<%-V7Kpl|*-tfAEPqV+UomXJn{ z^QKZd#)tn6TdAej@W-I_Y3gowLfaZF$SKOs4DY|>cMnKG8|1!3H#oh4NHXfsG=I>`E#ti_=f2-v|V<^^-Rp2#h)m zNiGV2Ip?3&(0@+%Utdy(rmct(dR%<{vP~n%ke-2#&}A;2x%i1w!|ptiBcrgG4hHUg zVikxCae5@QGL3TY{qWP(%05&yQ1s{wymzt+_z;!O&F{>KeRpVXLe^-aQ{|C7WU7WE%s`Blc5&$a+a9bLoRYf2 z6LEE!(20*V%*Akz0fe z=WizlAg5mrNa|`rpN!2K|^GRK>C^X4_nNmmB{qzEa74N$TqRB zI~~Wy$DF$oyy!~ut?@aJL*13+HvPydhxFo?#p;Va1kACJNyZ(x!c7B#s$aUp4o8$8 z>X$wCV}1NXlx-b>KhS}|!(<>{iEAE@wx0aNjIrp}-5(c7ZJnd7MK^Scbl*EG+egp4 z_VW9l(2%uy;IapzKN&Omy@CSpT^&_#!{s2l4j-0|x}u{Ao31ZSk3@EjyXeQnjq7uTZAWg24(k3P^cnv+Yb6oo5dS{V`o(y6{sj{v;luLjx@yz*>No8LNHy?$E;MP*vNj6e>I3={B7_vcGx{s!BU~c>TJXs z_B`7|U7#`R`%ZsksH}j&?zTbHY<0s974zyg{tEHYm5;3kg?X<2>fkO$ESWi9Pd&+c35(l;yrsyJC60#F&1ilS|-X|C5EZ4L2QJnP&vuAd=DMmF!qc@@w$|%~_>U%cr$_ z$@!2B)u-23Prv>8b+?P;N(kLGgPhB)D8pMyB*+p+hXMg6YPZSRmIF=``;C|bX`fZxXZH#Fwym^dX zrtU4HR`#7)hZ{j_)vNW`U%ljP_p~d|S+4@H&sXVfk;%3baN@d>!%s^0n~{fzv5(D; zoid_L9?8%v%g+)%{Y|-Vrw5%e$~$VM6B0qTIy>3uiCYs1uJ!(2GsKTONH%z5UGmbM zA8_!s?&0fXb)3rfu5~#=LR`0CLn^6rb+6{u{^ouZm#XIJJlUSC^Nrb}*+s>y8K(wY zvDs(Bo(=9G)tB??;B5zVuB!~*n^rO7;)Yrady+W@oelkOTq(=gM9K0W!9E{IJ$dGc z=aj0L#tfQUx{@U9?h&G62e~;PWewX(jRfSsZge@y7P`1mW3_wCa!5Y*m` zt7}NbW{rmjJNO<0r!SUH@!(FbiNS8&e6wpOfjJcq4)~ETbJ5d;J;ToqMBb4R!$%yp zo6>p(vJ*>|onBDFUvFJ*xugf9&K59KUjN>PGd*+jrl$errb(n&&RyZBzw1{yShykI zCK}2SN8K%yEWRq4mJt-FvYb}by`^w>-7+r}RO2^r9gmJ8!J3{Wx0o6q{yKwM8lzBU zMfFV5M~WO<{rEa)WUV}KvWHdXehNXF{LU{Ew*`*M=+>%WUT zBL7i3{KyJDnfcMmM5SRK^hf*|{Sx&^bv3uyJvPwa*_xx1Y0Fu&A2_JU2(xKcO1511 z?YuLSS#9|gEs9=7x3n4hhNjAFRn|o}3wOxxb$HLT^@ZZ>F#WX@wRZvJtYal{x$48# zO60QMYxIZ8lNrzZzO?yIGlY%OGTXN3Z{KaWhGVwL87_ok=KE@+B}YotxjP%z&oy&G zo+g~OJa)E9I6DLKx>E+^>cLbTU&0(+ZSXlFkwgH93XW&0c%I$l_^Q4lrz z2LaXKu;V>^jh}hqkHfdIWT}Ig*a!8dMCJt0wY-+VG#dqgCxsAOSo{M^RY?3!SAjXt z97k@Bc#`XAokf>Gq7^TiyKA}67f>wnYU56T5tU_Jy$zD>4G;$AkGhhdIu~!$d^s`) zukb(6-DF2pJ9lw=jV}aD21LK`0x~dK*Akg^%`hRSu4sR5E6&L=?Pj?Yr+CfMdIRh? z9iZ?VOFw^I&xv63`u4FEo~=w7{&S%v!z|XogH0LwbiqaHMP<{^f`4tZuro>#5BH!H%V@NDb| zWEUbe0?Tzy}T`9L(C93O}m3Ek{7;>VlL z;pgEV@dcg1Y+Kls+?(uab9Xg@1w}Uqk7!5V3T{8K(qq;siN@HJ%DNKWghx1 zrN_LAr0gw(yHp{!3_C4%<>NTc)|uh)!p3yZ9RIe}Wb+{6xh$C=uq_YE)|lMe4DcD| zGo$rluYQ`V=J2)xumG4ZVtUtoF&VnEj*-^hjPf?BfrlsbSTb=E5U=BSvI(xNY8>Kk zO*9fEE_v&AdnS?$n|KOvcX*A%iVV-w1LseD`MhEWE*N&Zw~!UlprZ?_F(0+kil;J8 zB}dT=El;1_lb}EkJ(|d${{U9FRxDkQSXL`&jQ)2yWhCl94BU($`$QmPz0y*rp8X?E zc!GMoF!2_L(f6$`*Uva+1$ozx{$!Hak3aOZHauw%UX+o9B(aYi!edLH@WdH!i$1kS z!1w6y&}_HII-Tw7xSbd*@vPxNrY%M=5?F!~%PKB_Ig+!sI>=kY^bTdmx8NUaSe8ej zdNw@$U&%;_bT04*_w_%D}OnNLWJk>2l6{1upigs zVpXlL$UzcBfRp(ckKk*)y;yh#-&)9HYEdEgd?bT8VylnuD{ZfAhCVq~Hbr8(?8rUr zj|Vs+CUswV)_MDhN?Qf)_k~0Twc>!jAHT)%i@!h}KYPCQWBqz9WN@0$)ophy-}42C zd8nA_E>6tjVYhO=eGdO#Dk-QMivMujODi(>e)uI?cC3 zzCq@MgXcc^oU#$>OM+h}xg`33Ti%R4A35_>Ffvx_I2EG=#f;ejje7ZAsfQuQI$g zdjI%Q)Dp5|i-3C|i{soa|4-U1ZFsad8AQY2uVgoO1e2{p8;?N2Sz$43l0jiXy(WVcvA9_S2aNQ+_q; zH|d`$tR8wUhnR<(==V*2*@4g`EZ3Pw-(aJUW727(FQmsZYQOhhK(#dn@bmMB8+*TY zheuQ;rAwEPXS;x#Lv&9p*#m}|eFl9_ZBzzLWWg*|)3)E>>5$br_hF9Vfx_y^dA-+@ zVI{FxqAJCJ&+Vp+S$0yZBVCwA25fgMh(?{2;MO8z)W^&$H6a$w1tWP(rd4 zt63oHOf?s~>n<3Qv8w1C4$FDoA zPcAwAR8wLJe8xGrxY3Rzy-?so0U{WF?#+gu#}8kAwHI88AR3^Ng0L)vw>}eYdOq2rDG@b*U7d6zZ_>YxQv$u zVMinL9XjnM{4=J0r!-;OqfIz`fkJ@&%U-uv@BArk6KVoT9VsqUy}Lk5?OpK^SDH=j zpo=vqQ%gCf?M>1mV1gYKxC= z+bG2r@StU`+fnvdE82<;Yx56y`G$W~kc1Cxhsk^5Yckm9k~(-r1~NQW#MHlF0r}sr zWMiesRTK?UUGZ-_$?EFgY3mJn)%3vXq>%8J`+h0F{l7W2ue3?v6 z2z)Uo1Ex*xV8R(h*5LW&Dsa%yydZ8c!Bq$1{?(yb8|esGD7{##S~3`ud`}FS zTYWa?Z}xbb=B)^qvf@F~Wb;Q4<`5$p5lUkZRFSIw33#OkF!~XsS^UsEiCxD|cf1i+RlWpG-b=WunDrcl<=(|{mAdQeh7xhGr-X*2^G)%g^2LNL2=2!|ccN7j}b+E(IS=A#gG(SsstG;H7<@xZSW}r+ZXzv8Dswp zCQ;^}JF#rVq`9hJuxPefmc<`9QT%KUHVTf z1QA1$U4ODovU932bZ`931QPgJ^YT1*ua-)M`-oltWZL#Z4WidAgN^Pw+B%_UEFC{F zo&!F(EEd@8x%aCRhprlEkQvZLhIzcW4Q5RU`rLH|S6vyX%d4mz8cM^cOTS$_rSdC# z94c}AZTn1eIQ@=r$z)r`r{go8N6sS?rtxHnX{8)5>tQkq>}nA%@(aF<%*eyhsB~@~ zNaIv+K3vF&2Df3W@eEl!bZ0mh@+xeC*@k1=9{NNJF2d5pq zN;9QQcHg$Z169Olv^Bq+jA3VvE5Is(?UZfVSXRO}pLLaQD;w5NcZIu*=E41f|F?+k z(5GJSlK|mIoZN|2)lIsi?H#m&th5=!FLn0kLjAI?k+53r0_wrzx{2#m#OCPM6N5a) zQBG^_dF3MOxuzERbbeujZ+;ohzvUK8QeU`hSM@FmZb$FH2X1&1Ee8R|!A;D~_XH@& zD?|3Gs26bIA9;9k;*qv#@THbt^O$7>tIB|V84R}SgQr zTiUiF^7(0Lh&#P4)jHoWl}BB`fNtc+p-YFn_vh>3u%r+*Z$v8mhr=|L8#y}u7HQGR zkG;ooe`-8vu;WZ}ZHcpF_Mc!AV(#Tb` zI3T*q;t71BXJ35ZQH}Dh6mmuICvT_Dc1PFe(hOq*A%1+Wl;fs7d$rE(Ng;Rwba8Tu znQv#`lveV84)s=a@@yx~KiWo}=rClmINzYqE+uqPhtkO-D236&RYoMu2`sw(h#;3t00a~I4APP={jmJW4=&IW$br)H+(FJ8XkRV!)7=hpxXGo=A1;6<^u9rE@yOnlOL zVUYz?|Cb49wM4B3G`x)0(0+o=hdE8*s0!l8@BHr_`PbFFczUdOyB6{)2yrZqg=_}# zH{)R8q@*1Pe9(D%{S6+OFlG6vlolF!l)J- z|G>QaI&m*c?y%;nIp~Hr*mCf@OT6^@3wZK{q0Y_nt$h5Z%Fj3LU|4{M8ox;+i)u*~o;P1d<;NNOB_y#^bl0@>2L042p8eipLDOu0JTe%) zbh}w?e6Y_wDdTj|S*T0Ht4v;H#mH?ei340j+EAcJtIU{7DeD=;MAYHy0P%O-l0xG~ zyq}Q|Y-P%ldp=r9dnTGTg|QhDV< zol>D2;BfWg0_Evv=28VBi|{7`ZG*VEV_(TO&4#c0=%L5qWXv6?}=j|SXzT~WSn z3!;~@Ee=2T&wm{9rKb`onqI@VCP|{!&ZBECKg=sb33d2gtms@8!O7sCM;vSGy#7Ux zc)>ko*Bp=KR*WiJkYiyrtdzt2zhpk#ns0GE)cOtvO6J-V$o;3cRh|$ER$bnR(mH0t zYgDB$gTB!+J-3W=av6MqG{nDhli$%i`5RKtl^&7zD6Sa!8Qi!2Yc`?u$>A!?%=y73 zt#A^nY2&5A%btU-$8&YfuKV&Yu%|GV z$c+)ps$^9CZ0aI`)~xP|)?qcPtHt-REbC~V3(s==6ZuQzSm2CVchyOz0XoF6dA=QU zGu7ici(P>RQu1>Dl$G5(_BU`cB$k$l|je*wWPP+FKqc;KL z+{X8J>Sv1r|NCyuUd;`29pR0feeD3{k1k$2YBivTWh4})6}VGYCyLfpGaSJvF<3QA z%&RXQ9l=wCm#RYGJBW%=I;r&W(@R|y9V-40^RMeorJsMrFL#ZAaz=-X-pXq?^v_JU z=Y2ivw`Jcv*r`_SXx2>ZNjfNyKX~eE9^Z~CXjpB5{1J{R`p{ol9@`>Np5i*UK%S)e z)!c?n;W8&;-Kc~sm$H(;K>NmHyg@B{*VV?Gq~jB7xW4P$mt&Ar__)y|G?XlQNcg(@ z^zD-1P@`2%W3p`l-SJZVJGMj*OSzUUhj+A8DFebgsa+0t{F&U#gm^gteWe{&5ljTotonLjhsw6qoIY%3P9Mk^c+Je(9pvC zEdKfRn%Yt*`T{_rB$vO`!BWLir8v+ z$)9rYpu=BLUVVTNxGU%rJ$U7;)(K6%D@80nZ(pCLGQSEI2GQ7_gz931dGmWpyi_9R zm-Ri2&))1e`i-3FE8jz=@|u6E+KW8-7zb1%9^vVK3!T<~JkhP;zs;& zl6ar{d)#y=>JjTaM$&JDmO{P~OGmbpUMH_}(l}d z17G*c{`;iFNGJj-W4v{C91qb>+hr;Ze^Oex`e!`!+B~ztWlrXKHMJf_NT>?rz&Pc? zX$#xe0K0qi)ule}IA<8Sz+1M1kS8fnL(Q9OvfkgTemN0P3RFTKY*%7nBRKLw)Hj^s zn8E79Qm<46Dua|7wgU0v0l}M#6;SY?M;D7-`}9EVA~@oeQ>8n(O!8C&7Ew?OuI}^p z${~hyo3SW1X=-b{zef!xG1Oq>m&co-k;q zyGcr&ZVhfwaI;bN%OsXH@gi2)anhnGB3|ohRXTe%&y%ty8Zes9$!_;+9Fe;4x&m|y z>7q~K$Hts3=z)1nWo724CiyyahHgLSpH9W+HJV_H*=Y%GH|+Ed%}cebjJzp7I7Qp( zdO+brvi2&0NBdO+o9>E!F8rm#y+@w6o6^rQ54Tf1Yw+CZg-cn$F(ar-qMhVp?7wdT zJ9Z<$mYu&)sL|ykYR^owbIOmtAW^8R?<{6^b0>6;Q`d`uKR~%)@AcB(N+nb&BEO|! zD34TA4W3Er-fSH2&1}P9;L!DGLk_B1Kg~J1y zz(w^|=c}sP-~iV=8m@Ir)?}yXNuXY_{Di=_=67z(DdEHAt`ncFQZfZyQ|-HNQzly- z$DI^N99w2f@?7f}SnQr=3x-I)&~)O2f*t(7}2n=`8L2I)~zrqAL(FU<6K-g+{P~YUt7!;U8DnXob!~>w*%6 z-w}7dkDh`j0m&k+m#I81-_Eu6?;`F`d#JKF-=x4Xyco5cNIgD}g|Q&Ur_>|a$b;wh zwRX~kJ=fl{qnp3bg{MUfcgGhR<~eDx@I0BvyCj=Z9+e>(M%n#qAsvXpAJh}xNtZHQ z@vcBSSesE}yG7YAx}IKhbQgY2a8`iHpD4JhY|G7AoD5s%iA&nKXY=p25h4G$DydcZ z%YIymA*v}7`1J=Ip8XErpHyX&NYbr?r%8^L$l*e?-&;FSndxtQgmqP4zHJ{qE?X__ zi`VtcadC312z%YYvang3PRX#ShR!TXc(Em;0*QwMF5uC3CYTULO~|3&ht)7ny)S}@ zKU*tmlB1fk3VQNocR@_p6`|{>C*McFkhogFjWbQj%gV17-BqsjOtbd2*>iH?D{(37 z&+yc-Ol_9avby*VOx{$F&M~h65C@M}lXS>w3SL|GgBDcDpe~s3!dZv5$lE5=yubQ0 z6Ibi@d+eCsNN`wU7>nDiUG~IQCl3NphVSR1L2)@_?4T!T?e9y2eg#tlkZ;DhSzIw7 znSuhuVT#j_-U7~txp@Yn{wIckxP5E>Wmhh1t2}Y>946Npn}Cw+8KW2t zNzZi8A$91DfUP*iu1Dqqu$RTKzDde5K8rC+z(k?SZ)!V1#}IvL6^Nmfa&C5t_2aSw zt64umX=$=4v*1ywL}egI?pcu7@gCiv&$}a1bD;E@fzNDsFyuja9`XhG`Le$4DPY`< zsI4bUqf~XbI_8tKh8p)Nl!}Zb4)|!Im@0k8@aE#fMZHgR>Rj9aE0Y%u!~29Zs2wo| zqq(;BV`k2}4&c+`T37)e#(i$|YO!tCZ621r*ZU1-(T&Ak!}}4@^-~F+mbBxK`C$+D zjtnDe?@qPs{H`>S{SG)>K0p58iY}`8=I$EHo7da;|7URXAE7R}d_&XujVIx6SaPME z4m7zy(@Ia`1Q-_`7st_4ZENK4d#_~kJgeB-mx7d=&<@( zDd(%G;))$dbuv&FJ4)D_BwA^uk3F;Fm>1_jGGAa+VMRCCZmU@4o|6LR3T678e$OJ!h zM2|Buq`w4ql-m+1=FRWN12+z8mnY|u-HM!)t@!^oqUrr(*zzbOE|Tef%7gYeC#*lV zg)@wNf_!1*BI;Acnx%senH_UpTYVUA*naVN892L<@7xi_pQ?ssxoXq$slLFi?5NK- zzQ2m;$x?V#6z>hX&^z}*bfJ}G072DRGK{1~i@;8(!5$}*z)ct{F~~^*w6HdmT}PP< zDlOTYZ3_VO!ucw0J?uz`YL7@hhFMd>UHtgeg0%Sh&Y*mvzC2Ty!1)U1DQPiX6yD|k zb|%B4%Hx9Ggy69|hWtOyr@uRZ16I##mp)gfT@T-(ne+)b9+F zA3bPC(}b3aLd-4j_9s-hCzIo!euW$Vn08sBLqi~{`r}`PAIFfW+%~+yx=8@>Wi`jJuVb-7;5S&GRo6q!PbKl( zKhl_WsIJvcKt8TK?7;)9ZK8ssbEN?hdK4mwE6P1tsK_OPicefR_N9&FWPh&O??E0D z>6RuiM;ANKnd=r`Ta^oTf-v6F=)i{>52rRT*%VME=IDKz`O ztt^as4R2)0gV-E@-D@K)kY^K;?&NR&>R^=x*?D}%BjlB+Q}lpUY{q^IhG|d6nnDjg zO`(Rdd0=RjymdMN)0qxY^fmTq5~D@|APZO0q|;N!JCH>MV!Ef{$ItSy;k#G4Jz%D2 z3>d;M6_hpDw7d)3bi*sB5O=^P`pv_WEAGHyvKJt@p9qTC8cfS-R^#7s_5O(E`piF4 z8z<}70Q*_$wDb?^VX%ptx`&FKribC823ynzY-t`~AuGC_eBgGU8AA4x2~%2p#pu3$ zfokzfKTP30Ia&is9mbmrBw;iD_#uD0&^V`o^`=l=I+T`z)VTmLX&DkfPQ_B%SQ4a{ zRCrtILVe6V?M=}mqV+l^G_LvK=nJNNqtaiD_n#Q&iK+){eP^U|_a6uqSAQb$BXlZV z`k91@%rmW<<>8}q#Nsj^ylC&H#GSD8Xua*S_>_Ejx!x4Hw0&Bg96uc0Ff@V93x+in zAgFgwRyLQlfA-WJ1~y7Sb>{lFl@^_4PFujR-HTT5=aW7Qh~x*Q3RlhViuv&+M7+&o z*CcX3Yg^q;iw_5LAAD04Q@9x{gz$len^ljZYitlh9HWNe8{pPOA!Q596i06s<$OAz z4piuP$Frt^RCOJy}a2!X2jOTR~hw1rzZhuwQxpuT8`Jv3^2gE8Q^8*S4 zevsJ*@njW{W|NuCh%S-slqvv{j>vQ5f)5vJ61i@azwZg@q= zaynhtw{A@WZ->w(cE=x1=^u1o?_e|(Oz!b5zC_36bvl|=1;iYd#h{60ho3$1OS2_j zKRVi8aqkIQAA+_}kx>4H%LrYW>)R-;7<&-4g_?JLp*9sIAGCI&P;>rjvqtAP_9uyV zjryY?`D#Vc>Vnv!KDu~aum8o~dxkZ&ZDFIECISK~0)iqc3er`2$AU;#I-#k6p@Vb? zp;$l#MQZ3xdhaEnih%Uq38A+@Kp+VrK<;Aiv(Gu-Y3}FVU-x;wKY7ARvY2a*Imdj* zJ6fFm^me#b^lb=4tp{a}VEC!GUorXKgK%D8HjDK_#d?GzF)zRP?#S1en;`wMOmuhb zov*8`cmnxbwTgWD!Xg}2!nhE5dhasZddQ1nlDvJ0b2TO)GG55XscOyVj%O2MO=(=( z{Zhxz1(NErnlLl8chL4Xk0DwIcA+7b0(eY)!EulEtt!PpU+uYM#3K9!7=O4_lsKC0 z>FHge!UbXa;P`+*6t45$g$>TmWsNSfsw*i&8CC^N}?klTtWxD*yK45|uYm%3LC+Z%&8%ll!-;$L-5zxJ*_=heeqDhCl?U zI2ul&LXeSh+y27bIl9uC&=V#&BhFm5fpJQD^NK!Hg4x=iw4jBaM)^$j$X{D|GUcuy zk=TV7ojabXlhsQ5j8d|u=a!hw;Lp`xxJK|JGL901RG1!hAU$s~n4(>L?r0HK2ayq5q)$Au^^xz0->&UK(;|MM4m)tcfj%1k2Ou>jTie5HPzi6tO_16$JO%? z1-23E1gd`+nW~~!0#!$@RQNTektsqOljI=ZR$f5->-Q6Occvj&d6h6N{%bZ$H;%f~ zqrV*bJ3qf<{i77rbD&vst&&1>(xF^`anJ)@C}EV=b%yQkI08Mr4uOHNS4w+JYQGQD?XLR^~>0wC68ImT8h}stl;Uv#R$=^{!0Ty()>;Sf`?+_gIB}2eG|uInX<2 zKAR|lH}!5XcCwptNVexSP5wlpf@*tk4^!Kaux}3nI6GhuJkBbPK^}ZSFBuJ4J%}5u zmSMHs3poM~<<4nux`&9cvixPrgD02~KO(0^>d(iYd!UU^x=sq!YWM0g4mQ2P^<)uUJSU#s!!aY<47#?h$%bikHxQq;Nh83w3as~$P|4a&>us4zW6 z$Lu_~H2s{m@GzbE*uLna|$2&hqiUC(Z~{M*F?s)LhWUh+emf&N2^ zs&5`?6b5uTEZ?8oV2eyrYPdauykfhZBQ}Gr%`RBK*V|1xC0c?v8M~5hEp`7EF7HP(CYBJRWmV%zDAmXJ7$$7cUtok;IyZ zkfhfd$s76sU(M!Uwm+(2I1r^@UtNF@85KZ75Ra|vVG>~qC=3-+Pq+f~Z9M$J z(vAR3CS>(kZUVzQ@l5FMgkV4G>RiJu+|E|ZOcuY?{%cX_-Z{u{?Z`^OBW~;LxhHbLyDPmVW+6#b-HO_m>!h>I>VcZ3bh2usQGeGx(?Lxq3IChmI2}_$1^8Gu|-^jVksZRPy_p zDQKfBwLL|7{lW0MhOsgH;Uu4Kjd9y&cJ|v%2&)j90Ey3x0cc29CO73QLwwMrqiuaQ z4IaO|5GL209Qmq$c_=F_6hD)6Zo8c$GFu(ia5QheGFfdWIJs1Ws2%%i ziuK!`xtwoji@z%^>vRkmTq2l_T5d}Fbj*8ykk|3S%I02DDGZ1-=*yqG!f1Acdqdfp?ttP3>@^gbRX5S?yH`Ma7WL5gR1)?UF=c=om+0Nj*Mk+7%pth}5QGf9@+ zNFpfP4%=UC-1tD+DB40NB=qS>#K4@xd*5ntHpN!4Sc-%8d{$%g3-91zq`D`Z8AmXBESW-taPXdi<|p-~ zB#tcgTIAlh5WnWiA(Qu`=IXZmpo|hx$LXencSWD6f7fB?gO*A_h1Gv{?n1Xx*Pl+) z2&=@r5i3-sQ=(;!IHG3w5zJXs24B<`cBqHQIqAukJ~}1}M=W^yH|Mf`Z%a6A7kcOwsb7c zF6OreZ50K(VU1MkOE9~h)&fW;$_i;fqQmbGT1}h3wu8w6wTq*P8rA~fUc1Fqf{d|U zUR$X6HA&)~QJ_(>cP}Nwk@1bsIOz4-TNc8XItq!bxp3dL!Rn4~8yeHXnz5fHweNv? z{|SjIlkY-2=FS!vnIDT;&r`~+%NrWAj~4S2c2}6|$AKu`@GQFfg>|W$xGBeQwwF_Z z$9`2#qybv^8c<*Z(j)eAYUG9ifU0~(W81@%A@q+Ykhf1AkjqW8wwfJ74Q1Gy{XLpg z9$0zC2wLb-^cVd$B|CAysIeh7cl$gKY3iWR%E};nkX8!-s_3DSw0hop=Yf7*`N7$g zs?KOkH4seyblHR7YkPrFrjSp}gT}9rvU1uas-F%Hp28FwKG=JaBMw!cetzGYty{Pq zEA<`ujevmBvpNr287mAt)h{|8hbIIj=IP~I9!)mrewZ&q9nw3_)F}mhy13xRGdcgW zJwv&AWjf)5_4tu*YM2fbI)Xo?!7dR+jUJAet}Tjw6U@-D9ySU>QcpI|J)2Vd_y&^- zAia<=B_yZtbiLn<8piI^6__~QXV2;DqfdO=xX*(w#Lep1)O8xyzyk;yP%#O&8{?56 z<o`Ftw1KfPGN;rkIy>+ zF=VT$_ek_;4awgOFrknKHRUZn(nbDLpEM)}spwd47dJM>VKN)LKqqXH9IB8ybOI*Vr_sk=0Gv$u zifyOF5jXEs(LD~vY5f4QC2qwD*M2{|Y7vZOPerz^L$F87<-w;O>d4GyoJtUfUK2E~ zuI?@TY&XD@ zJH8g1-TtI^w32Jr&61q(;w|*qZQvy5zLaC)Msx^k3Twp9kshF`u@~!o8!tOH%A3@J zlI?e+l=2lqZN^KIu}3-pr?OZ=n0eBodVq{R)m$NCK_s>*_C;e#%8Q)S)@Ump{zMnJ z?{faBv;ocC!TbmOjDW~?)GLr>FVbdhAN>J8FUXr{8EhSYxw87=Va+G^)%`>PGj?;N zGD7h5UdZti^lALPNhsA3)Pd zq2DW`{Uu6bdWZe3^+aAh&<&s*3Ja+AIoF_!Um@3@f4{horGo{pE3E6P57_d8#BTBm zPQ!M>H-hWy0gz*=na^0^I5beZ>Vz7D<~L`bdte@!;M6!3?^{L2Q@?w9xAhpkeBC-H z>sno4@gDP0Yp(y$6O9C%mhm+nk~hw)G+pUo3ngwr3P6xBPtVpDC=@AV%{fe%=m%<5 zLnNNw^L#1%wR$~UFO>Y9=eHz(ybqu)7tG|ROJdT-9M?BOD%qO`uj;3YclyTRC%^1F z))uI0WkIApnKKpf14Wl*+%RepQi z&ft(S++$U!5-2pTw4+pN;LnNkdx!YlG+7K*Ly3@$yzG5fqKNOYYj_5=o4>{TMZr9E zVpMd~qugRe=sD6of#RUi(!G|&&FWmYm(#$DHQW1{AgfNt14C5Zt54SZeU9P#Q(ceF zI}g&@_Ia#1){jo$={)jBGGACgc=9UF*?Z&NNGoroEA!P7JhY6P!FU(U^G)1S2dqI#WZ{gZvv8qb>F?v7f!|@etgv14j5y4ZcX|J0R$`xa58~XWz6}HHM z+!w6&QuL2QIVHN+8pQft%%~IHE^!tU9`4DNa0t;?m?n*uDd~7O>w3}dv1jl0VeK*< zckPdRMRO(&Iy9g61)sW7A@i`1auM~-K|*Y1=-~l=xz(H2^UzVam|XK9F+(qL@0(># zmcKHnR@4wS2-4Fw*QuCleWHOXXYi>sZ?cuc9o2ay54+dC5nqu8OinG#uI-I5YT;wd zb!aI1j?%&J7AIVOvq(3eE7*=*V2!Blnex$T4zr-3P+P2pgJ_y=HIYHjQw_KuXmOW( zMfOi9U}$wSyH6;r-g->AxG;d&d_*gz()DD}?NriNa%{f1jCQLG79EM`U+g5gWCpQz)2#3}dd-T>icDr{1>Uw3xBbgQv1t$|bl!>%8>+#qId0lh@m zgxS5$j7q!P+Z_c|j4=Hixk>?rt;-9UhUtGF>hu!*8{KundDZ>R!mN|bS}b@IpyRUF z+YJ3cnuE9R?-Y6wS#UJ)6{OHO4~o7UjJAv;r0y2e-X{9#{Lz5)m+PSq2kaj9Ce z@kwtu7otLRu4U{DQmw~>>SxWw58r5v)bW1N{CZK2eF_%!LP5p`H+He}ZE8?rg7c`y zijyuU*MVUvz^&$l-I&clk2Fdhz1xV&9qLiaa=>L37#AMs=RCQGCbTwoe(Zbw{1Lev zr!`esk?>vwXLm6H>5r$PyYPjl8&&5)OfRMm1ZiKM#^dJ`x$*7;GbxMk?BQKA7jwIf zg`u;Z`!qTQw9k)OB`Z|#i44g-S7S1vzwoSl?A%(R+0{hJHHTfe=u~cyCCY%L&neD) z4;1I{j)>6u5D(z{O#R;vJ|9$Jn$)?B9&Yv~G_KAPtCm>Hbf{rY}gkG~@_t83@kHC~8O zy#{7CqlgPo;BdwcZ?jK(Kfg>jA~Iy#A_0nPdW8x*kd>0ntr;J&3VkoHWd`P3wQg00 zGOj4eOpRq*b@kxcmZ`ZW_jW(%YtZB5t!i?YiEaJiqh$w|Q2;S%H-402;P2WYhsQv7 z=AjWqTK-Ld^<%t)+cT0^C>dD!?I>-^j)G{^cM}KRt2pmmCMZv|k1RO$D%`jey>d6U zJ@r&sCGCjda!M7r2W$%%$8aI2dFnNVqh6K#N$cI>)0bDF6_wAvkgn56u9?Qc3&V5D!}>z-UEvs511mdj*?L9hJcP62qb=+1o7?mS zaC1+9+sb>cLjRlJZlgoG!tEr{=TX*JIre`78)m1yqIb7#<;*VYxQyOw;+67QnQbtW z>#qa#Nl99YhkqmYxo7B~l(#9k8)%F;mgY>*&nC3`HdllD%+U8q*w&99_3zQ(GBE_} zvm3F;2fME-d57ydzDKjDAySO*Z89WN-PYCn@qA}88X#V6|H|<)8&TD+cKEf zY*(3)tu0h_V9!&b(-0Q083Wdu-zS>T3lgT*lWDUV8|D>t`s?U!OiGVywzsB~&qfZ` z?}7Q4nBVQG8G53(R*#Hww@M^pA7-Y9^a`+xDd6 zq4{0Q*C5LWW^%&r>QVzmp}qY}y9ufZN+B0M-fH5sN2Fz?bd6;L zj*SsiB5Y+Q1h3-Leei&S{`(r^i@n{TAfLWvci|{=+ittDIkxPBwJIlh{6;MB$+4m; z>J87zuD7;Eoir-kQ9?`s{SY1oPh=G`^D}$g*$@oEnKYP>I)-=xU4{HPkJ@Qz%>X+! z%|mdu?#(hHE5E3LZ&sH3`ah@)=PGiuC0SriPFx8#z;n$mMqAPrw*TQGN^7yKymH#Q zGVt+zJ~^Ly0TFOdX;sbQk@r3SPnc$*QrbVx-E{4%x=p-%jPH{k{oV5k9I6VJSzXfR+DEgofh=^2-#Nh}quDMxf`QUxNj1iUiy7&~)yvI|PI@vO-}aUjvh6+8+FMVH{R zC^dPKZ={$O%)3^r?_t*+vft_NKswWafT4HCr)T&7^v?xW@NQ#XEA80Bu8j6B^+v0m zVqE9RR71sBgC!g%uX7}?U!Sy!sY-LI;}gp_0NU6w{o&O6yL)f2>d(0Y{inGDRDAex z@fnY*WC<-tUb92#+9JcJzygom7%$ZsEimHAo}VwU0%DN`TY?!PHnJW{(f^I9_oX^5 zN$|NH^mo4Ni#TNkyAA@^_hW}SzkowEMZTVZ$$oF*X?mw4MWOhzw3}g-7I zo_O(?WF+FfklKU4FNfd9{9fp9-pESqX~|3++v##>@fB;pJK2r7E5l4RV_acq262ep zm_SIZnq{cLG$*TUc#~u*VX+q*#%XC=&zvl`KspdGkyOwFDX%|K-&GB|L*iEB-BGuv z5+}z+-@w;P$U_JSp-m#mC7`T?O#P~{HKTo|6RmZflNG_?_>B;Rhy8BT07Wb z1(CHQTrS~)NOWRRW6}F*mo7o=Jjg4<@l1jYS}=|XHXPsE#dLcVD{vB zSN~?SY8yuc=-8%^RsHcF|1O(97wx}1;O~Fl0|ZHx`TW$M+u?tG_mAIaRQ*PlJuFXQ z&-o9p&9C0QrakLStJ?Crcz*TWKj!1H1KIDhJ}&>)M@Gqfb@oEkzX|d`J@((OeVHWx zeP5J7|6D2m=X(76M*4@4&YtE|Y~sB|FY-^9{?pg~>DtvwV3L8={EYZNhVm~He^xgj zK<5ej%l}gs{;%IF+R~kMR^1e7!u_c}em&Dya{yVUigy3gV1F?Ifx|3UXQL*;0h?U^ zGUEUKr{WVpmiKI81OL*B{QVho1%XMf?E7s>`4{8)$1?_|0Ugme|5ts4%22t*3?K!L6<`l!`-eg#{h6)0JZQM}rAd zv9TA+2X^sWVF5!@KI)}xe?Bf_EydhY$Bz*x7eeE69Gl%8n@SvC`oR%dY|rd)HxDo&&;+F<$pYDH$q zr%&_;XFR!SwEUW5Dzm4m;Rd*p84vUTe>#WNwZ9_Gv!`z;VzU(a)E`_0QT7J&X@;Ll zWVj-G!$~9Qc~boO)?!%v2*0vh{Ao$0@89qLl_$y22&|FRsb6#JpX9c2&7R<=JnKyP zVlaVE^Ui>2YX>@D2=G;PDkz7k#B>Qq?s%ejDH9We=_vlny&tADb?Cgx;>lc{a>89C zqnxKw7j}=h^FUqK)r-TKUB<8M5a{ybGNNbDxu9{0$@$~DGPTTKz+%qLbcvGgpvRJ? z+l&;yr5iQfNW5#6`Z|($;k>hp-R;o7aBM0@0k>ddT~cnsa0Mu4h3W!6724C1wPF|l zVQ~HTrFgXy04PcX8cY1S+rT0E5Vt|l!yUu>d}Ip_tUvA6)D++MhUd77Ui|O3_{UPc zN_%p%ITZ`oxoxJavp04+;Pg8k2lRe-e|}wiVPVlnnP|bCEXCxepN^n39!+*ct#fQX zuiqT~YLM>!H(Tge2lsX0#uwt4)UJJr0sh$sd%_OTXyb6mV|(`6ikqqXNHp;L|F1^7 zNO#xc(2}swr=ZlL5hmfF-qZmLzVla>a=IFTCESvPaV&v~z#RJO{fmM0ca8CHh)z)# z0B#@GhuV9&p3$6j-ho~PUa-&3{Av;aMHiR@ps@=h*c zoWh zz0%#&h9|<|JPo|aUHlj8^FOZct9d8O8M|h&a;}6Dkg5FH{||xGG^u>ki{CPkO@`#<=bCasYxs7Ctisa6Pa|J#E3Px?U75_qobt}l2g z2g^?8wY+@wjTO+jzIyb#zF_=$=ba@fQk%FF0VHoysoy}FQ=Vvm!VU>zDcU)TejN!s z*%t&shU~-nLSrUSEWvYjW&b_5W->3Qfa#pQ`>GujNE%G+mHf4uR8*zOhz`&}`Mp`( z?W>jymRj!_yP=hA0{oVjzr=Y!+Z!@z59X4CEAUIi@XdwtSj_@Wfy0&KD?BjT#~7JR zJJ{!5FVJN#yPY#eb&C(cV`iL}`D3oP>CAMOR%tltyYKioqfN0rF;0=0HH4zQGP&040^zMTRfoe_2bG=ba1 z>pH7GlTru*Rd_!dEkv!l-d-EZhRLX+t`l~a!}*J9O#F_$!!{1qSSN5ukCde{t3JM6 z@MoA^aaxx&XYA=mz|IQze*IT^1E?ebSm|%YS`F4Y0XMD+hRY%&Q_<+dj8eQj|QIT9buGuKvQaU&9&4hB>%{KYx znK*}dwe~8(`ycK1?L(|bcm&Z_)oE{ts?pS#O64~*R{-J+w}BKMGjYu{amDI^IX(`i zP6fW9wn8;Lh~QEd?}*-IAIQ^DacB=C@e^^JUlfk_yC!)GA2Qnv_SFkQ_Pttig=o|Zhq5jl6dgds~nhsGFksvXs~P_{?`&&408 zGS@N34|m2AiffnoT#uw65yfn1?0kJACQ^a4uS3f!bjzsBBzf_uWGi*-TV-%2@X46! zZS6eWR~?B`<#NV#43eKXB#d5A-*9`1$3+bQd%tWHr65^}QW^89oQv#K zZ&e7q@mTtWX*G%XL%{_*V$fN+eBd7x8~dB_X$2MhJ(1s(!>&a75_TSc-k7*Q^C3Xz zaq@)}l_uWS4}D?uq3kkBK@CzfQn4~!GeVJVXI`~Epbt&jt~$deK5DesK)Yl)Tw;*; z3ZnSn38#S3wGbBnTjqY1ddIi5xIKVY6?=ltp0QL;0RP!FHrWcIUGg$g^w;CzXCyRZ%*c&|rPL{jh z)DbiLPXW^ZS!wlFDppV?9@NP`27GJF+rCGw>;&Hh7PiBK#{?9q!9A<}dj`zKGbWrT z#TrBDmKf(p!5$hf`O^Un!i6bd_Tuu%5`aDhkH0>0Zwaz@BR-h9c$2eeyB<@3z6ELD z-0_8JM(}6}rC_JHj@z*#F=1BCkPxf=goTG8(c=ztuh+B>57M<4;e)u0FV*XjQ328Z zO*WImcOp=fZw?xFf&`4Kv^8sKQ!WU6(MrZlkFK8=enJ+JaVin$A0h}8a0j|S%X+ui%dzi+ATqGi5}Pc~yP2Z=s*!SPMfEt_TzxE1M7%+-DPZ(Z6^bc$ zA^7rVEUWK3J)5f9%ADg8RBAEw$T!W;tGkoYxN0F>3Caouvr?)R9$`^JWIX=o+9+s; zy|(WF{xYrJI;)T@H6T_f@#hl?(xIzvr-5&O?r{|MqrlEudx$S)r{!}>#*q$kDI2Lk zRL64;r0P@)aV(wNrCdTy18FBhOTEy-0#r$d{~hpQS=ZgJ@pHwUh1F3mSp>$=yUkf? zT<2e#9`Cx#wq@Lt@wC2KXe>%2Iu$SObExZWv0szA>mDTkb6t8;VM2Bw5->i$xOM-^ z^b5Rj;r(d95EtWEB%8>}yD0Cq)}kG+-8M3qQ8`Y>=ExE(<+0(_M8G{Pe6cs0A@L?C z@lohy;WuWNo(o^u)-PtQcG#_9j@@Bw-lw+1@OcLNmN`C1$67b79iTrx9zb{9h-bBa zeJZo)_ShDe!!a3;dxDHpn%0+ge^Wg^6c2EQMXAT#C3Jrjopqc4PO~;#ix4o|N$Wd0 zQg%Dy>cNs!8K=<-U2G(vqhcnseZ`6poS$S9c&f|_aX*NWBlgxIiwYGFee;N2|McD& zYyJGUrIxSaO*-UJ(;m}lG}Blc zL*)1Q0Lahz_UnIElmF@HT6uC6M4=ge)@F6_=$(kh($^RAA$DUacLIYymgZCDc?$37JZ>9r*YIP2J)lWnu** zne5`WC%t0htj$ts4r;ty_w)6Ycp!c?0YgP8elb3bV$m&)M~*0^LDd=+&-Ve5t{`#- zVD^8UH)TS_A)fePITknEVA)fuUCUvT3mqI@a5X2RI{WBycBpo~N$G1K`jjRA?KFc9 z(q`s0<05fWCeiBTw5$`6OPi8OI(KWc`mzUX%|H)PFjYa|PqA9=z9@WC?MB@;6-cY& zQc>i(9-nI3eS#H2DyvW2@0y+bPmM9n1kwAB2<#&;k}~l!?Wtd(q1nrH4_%v7___27 zT^gf!T@y~uR(+Uzq`HI$1~Y6snq&XYXu5zS#pm9Pan1KQH`CJV`tGUPnIF48F!A(Y z(VSL0QE~DkAu+;9`~VR1zIpp4~mEIa$HgE1|FfTG0Q~i31u~S1LItof;_>mVj8ZU&_muj_fs}<^V z|H7~dn(zz6Fd@9E&HsD22Mi73Q>Z4l7-pr%M&~}_Xj%PxvXWv3?uJT zC5le#HqI?%HL2Z)1%fneyK}C6@Nq61^DlCuk$lI(Gkal@hj017vsm1w`)qBw{kKRn z7ls$bx$C-_`b-0hnN?~+4TdwmUmLye@0j|VOxxsIMSKP_AU(vZT{`RmWCH@56BHvLv zmGG6PYx^lvqROZBv2dUUjeFY>K6i{RDYJtMn@GSl#r|~aLT80(A&?L)DK@!7W{55T zzb~Bmm9^C*#qvHTAkyE!k=^m&*?@XVfB2w{Qfd%mwc7oiF?;;%5w`;Ou9i zMNYTUtsLZcF~2<(h%Ttg`noV~V?NZNaOO!Blvpn^gU@8mFd7O@dIg!)aP1+C+-uo< zn(IZ9yF1`cKcz665t!wNyaW*q{_75g)#^p?byYi1O+X|Iop-A0w@P<8fs)KFy6 z#)MG4ZZPt5o&&UNJiJ3vaX#{}qDr{u)S zJnKIlcZjnGo3E&I@_^p`$oBx?V^-cEtx<*7J6(8Vov6d$)!97d&v61RkGS7{;10WF ztLI(DieqG==a@Fgi+z~kPsF2`T*i*v7Xe$b^ySiza8r3BLQ(nns8h9?+>li=(((4q z!wOIo!X#M^s8g(?U3Fg&Y#JYAS~u%phZBzkYSxsztIrX2)HN{MIIa-ET!Ch~0As{}EJ8BO^EuOty0 zz>?D(Ng2x4GDg;|6;|zd!lJ9QLp@^L8NV}lA!!wDtTHaCYd!QB2rV>jjI=)2HSUvM^Sorp>if6Np8zPRvj6c3a`VNVy% zclOoG=Y;9z`=*rZkphOsE*ZS%cA=zV3W7Bn|57p>XDwnb|i)m&?7>l~-~dF`g@ zv+jgqRjS?B%mp^L)`}IvB+I*^K5K;@6oUywk(*8chv0HdG4!%G5=LD&r|x0Ud3|=V z~Xs3gwkY%jOZpykE5Nwa0FmnjDP*$n5~|4v)+f%ISYYDDj1fp)~^9Bej4z^ zGYVb~XWJ)@RdqPq`|oa%j`j}{3bo5UEWkVU%$SId$eD_^7OF%{(UX>JM2(aBEE=$> zTt}A_ZrkqN@Ln28>$NP*$YidpkhAs)nJio4w=?8dfz}chzMv0u5&I%(A`S*7;YuG9 z(r-;3zIh3{nb$=pOxOT3M?SwuShlmtbBV0nT~;91ta>n)9-PZ>LZHiMk~Vsg4Xa}swERd%T6S#MgTo=l!Z{$Fe3L3~QNvRPUB47_X$e_N)A}Tlttha#VTP}A z9Vi9o*62p6kk+4jQLSQP;wvUZUuqM(X&>+3q=KM5TnO8@7{clmxfO@9$5p|B{+T>v zJk^ip0qgsNM!I9?hSXL^h;ZZbMzw#*kGSwqz@TST8%r)NJ0ALUuGs1cDOtNK`inbt zVp#%fjC(!{rKv$qfKFj>d2a*HO#$~MO8*t*bRLH2^3N$wuL4M9GQC~C$PDZ${lh`TW)O>^t#2^lEW>3 z6sTONe((GQuE)ntQnyKk?(4L-fgaf1-PR!Txzi2ZYL}Nar$isR?>aCi^6KSH3tK4D zmwwa3tc4eJ8GS%0rogB>zN-u!W^spBIZbK9J=S7xU;9LV&Y2c^|Fq;g+=q{~`SKth z2bucW4Lgw`32VR}OZ>UWU~TYfGZ`b~e+Zilb%XLUJLhTVe*j{0z;+TlOFdj&EE-{M zFy9zRst*^zs;^jnDB|v5)`-Pmt%V(02uRYq-d^^f5dA-2o6T%`chOn95x~ttg(<57 z0J>(2(4hmLVIObL+prW~MYYDu$qh~c7)%AWGGE-49UU+<{wngOhvNMv8(cSyi*nps z$0KqMaBq3r$T9V_X=~K9AWP$U1*1n zvNADo0o<&R1j_4;Wv_4UBB0jt!#YUVU&AZrHz0a9)$`NRX%&E7n9RF9G|gG)N>C-C7N6 zxH$m5uaN`3f0un@AV!~D^O!lY?X#b0o^FHi@bkx+e2h}2TT7kZg#{$b50$<-0g(qR zLh)x4^Y)tR@A?%uc?KR|LRx$bi{dx$iJSi0L6|7*-s$$#6Cq-XH^39^<4NA9D?S6( zZ^xljWTs`ueBjIMby0H7Gy;HmaUbGqdyG9}w0a_|kldl{0-n+Q4q-ucBYclQJwF_& z=oTYe>1L`P@ifXApE|?-yszxw*5*sqD!toQ8$$JqFTijH0mGVy(`4CEd9%ZW+L~WK@7&5{keMl{c7ea2C*vV3RAG zNyiCNcFD4+!eSAd?!y6h~f z*x`Br`-EzDp`098*6?kOYW)*(atIFCFWeU^#pN$rX`J3h0fmS&v5`qhQb17-F0BOr zNyqvl`FwU5!%7{Fo#2Ml=1_OpEuOtVbYUEDohifWl)YGL?>%t41f)a?PL!F>sjJaZ z04qI(3XlOb{Uzq_y448??6n-IIG5~KYI_|rMixFLy&?rl?V?~% zA|oi$;>Doa)Tl<`bJ7uq*amZL@qH;0LS{iBp~1MCRHM@wf8OtSPKaESf$5ii6X(xV zLc;+=Gv@e?Ao_&Y9#YGw=*hzB8S6Z}dTH{ zxGDmWBSvMpI{qwRxi zVnc!x{N78=4+>^Uo=!^#y`~>ogs$$s)aX1%=lEwWfYb4vn*)&4ZYUnw@#Wf^ktAdY zG9~=ma)&nHPY4eG74knT*-p+10<^fyR%h(1$faQ5EFR<(NaU56c7|SeFI0CsqW1(L zQ`frhFCSp?nCS|ZL@&FqnCc}FFYtp5R-^V&p;(vvpUpV7~pKeWrEb zd55jo*8oib9bg|Nb;i`2zSO+Lvo@INnl(W|$08J?spVd-oC#DZ8ywlYpD8%(y+3e4 zB=;%t96%}7xS~c2<5jg>kqU(;ICc6;tPb-aHXnP_8RS!<#%n6&V=9WeAo|B9hs z(+`{=ffwo7ZIrHV`gA+;XeNGPHTO|^*{yzS^=ak~^sdJq8h3r?;S|_%Z7{lbpCd~m z<-~0ERu5D0ka1lGY6{t()D^0bQz^RLDlzI`_VBK`Pp+7tZb@6MP5>E+RyCR*$#3z9 zUJ9LeKagA#V1u5&d-su5czGl6RHF~}%B;oL=+F8B2`g_uegpWHTW;p9(&>6-8(B!S zJGAKQVi&9Vlbb@AiWByAZ&Tv&-a$Pwd3`TKcBr;c&Q{M5zA_W9Dd8@?P|8ie)N0nJ zlBG^lukSEC=HfilLcI$5Fm);tz%XLgQo95~CADgNT_Yjmi%8xv(<_bhZ| zH&pM^eu35sOxPH-v((2$I)>a|ZmfRDz#+Xc-G1e9M?j>0UY=2PZldHl6l;#7h(%kd zOLZk4fc+5;e5T zzGJgO8bd65CbgE(B&c0ql_NqHsSgtO{~q_KsUz*#e=#@yfsuPlo8GZF52`9=cFAA$ zKXU&bKH90ujN#>J{m2}X;y=TM9`naA#STyb-D`yKD1({Wupn1H`6mX)Qa;0RSz{qv zH4EakrMfg|pP$N{(Cg-YSOstcgPZ@HjrY!nm%SCv&FtWNjVCC-BgxZ%iT*O78ePoK zr^ed2pdhvAl^PHUrQVFt$T#7*%@$CJ7XLBV(l}9GFl8jd{n*HL+Jg?z-AjiQARN>Q zf#n`h9GP$BXE4Db@$5(uBSzd1BQ`H^;(-IT~fha)A7ye$|>-1 zeb(I(OP#b%$MD2pz&{-#J+T>~CKLmZOUzi!%|}hVdu#w_1F&@~MaGR!rDL_7l}$58 zzH$7ZU+%D4EID@R88P4ix%H#JT7EMYIj_MeaqT9M;raarEfIiTajP+ zIQ~?E=~cK{F8TU%|6 z6+p2WyzEuYsaZ;1Q7E7?sql$J*c#kexyLEYDB&o0Nq%av;I2WDoz=v=lnY@ACN-pO z0i1N!?!Uti4-BJ>Uqo6=pjyz|B_4iMC98(p)Qp>V^@;I)02tp;KH&)w`MV?0W~=A; z8%aO2<#-1+-MQ1IzM^%qiu)6oddhjME3>nc(GGfUW!Y44LH{w1yZh0sO!e;pHb0~F zUor;%{hX6zB_CjS2G#mDSUsyuIj8yk=h{DyAsbK89-nK2ko91Nq22yQ z`5`11a9%KNN^e#g5Dl5*rgt~ydJ-BVJUHO{Zq+aQTv0Q|b3$ymw%_tYq=EG6XBs&R zE-0;FtdDQ?l_T;~kP7K=ecq211nR`)?qo}6Go3nfsPkzuXgASMqnev`a&^-c5 zwk`2!#qk0pHVxZ(M3LdSxZ937i#-WbGY^}LH));s-@kI<6B3&BP1;v%r5!{Vux9hi zLYiR*4SbO^@s44tCA-_f9jv{ z703L+fj|oC!a*m05|cTo_;DI}mMkjC1(s{QSoHA1dO$U)q{U6ST6eEwE11*s#8$eA zP%#`WG}Vh&n0&%OIu7=&*|7jKM|X#W1chvfTi_6y1Sxv`O6!>Z)`a>{DDgOgS$=6L z*`8qCe>AMBH}tZ*vZ}ssuQFGwB2$$G@Xt;QEos!kBW{vsD{MMURA*}@MNoyrbk+f0 zEmXrBm(}g(=y$m#rF`Z^p}M)c&Fjm3W;0O4!GgO!%nMiXQiFZp4GY#<`K7z~>^A5s zV8So=8sCypzr%p2v>y1LH)#=8EUqL}BBMVcw)TY zr`nzR+FR&RDekjjX;ghIG9ZLv;5;@3vi($ftz$-n#_|mEZzz4FWOypYe}}@S-fQMi zX{R?4Z@E=W?1nR_0a-;E*5UWlo3m{7IKRkxyVP#YpW#`mNt7YbTf=;GW|9RPk62Ld zCb*4>r8NL_8xFj+!}xish{A1lVLLzLvAug)%hGnT{4daH(xviGyeXvcE`V1t8{Rjtwj`E8ew0Bmz z_`~ET74!6SFTV16j9|DdWt$$Y(eG|EzV`WUwb2-fUk`5J5_iNnte=3PtoHp`DZ)Rz zV22*vvRvEI>n)RhlK&|B{S|(B&mHHtz+zHAQ3J6dzj6`+l!9VB%VU1BY|&5Od#`HDK=4{p$a}5^dn_&DjgAtUSRCByJ+(<)|YfqhWh8V0J~wcnSgG zf}YnyfRL-3@hsNlWZ5V2+ka*4r~>O$spG>7iF2o~Im=TfGF+Cuu_b6Cp#1srx&Mc~ z_lkvCaz=6!S|q1t zfBtixz2{tKt+D>G*Uh;&SAEfA^w(YezE3?>^;Fg20DEsNAbHWS)vSD>17uYcNS$4u z{w=}r$~DmU<&MtQkJ;-Y3p;&>0OmeE8`ALu6|F&aS=pcXN3HW{WsGJDJpv>;NdYRE zRfX1VKkwR%HwGH`3_P9!V-(%n;vS9v8-J@o2XKNoRuaeZas)AdFgg#W@m6N?tQNjf zaxPK)eDzrp-J0-F8u!N?1R#Xl_!KJ>wqQBY$MpJpWYZe_u@fKaHFJ*-LXriwI2Mm)$Ms`fqv09~{epA?gYwUKDkb0lfINm{A3ek*vY1HurOYQ4o(pF7q&awzmw>{8=%x#$f_4}7m5Z& z{hhWmQZYT{d_l*vXK*jjvo_rj7PQKO2sgWAcd6=ZHgi}niUS<4XsJ7 zd5yfWccH{M|7{NH&q>H!#&juCqAcS5!4L!#xh8702y{RfOS8$R=Q-9R>HM8T3-Xo-P9s^#m*#H zn9$Xb)W6YV@lNtSex|s;9JDtpCP>VZnoLp>hX>Ha9q+8QI7NEK1N9y;0V3g%G`K3T zgUyn`HaL_q3G?Cz7$L!TPV#pg@;wFSQS8j0O5ekQ%Vnd_+=rcErqr?HX9Elw;SP0` zJAk7YbG|1SjzF9Q{pum&MlGYl>%w)V5MU6KUow%U6i$(<@bKRP(n0Lv_4}5rQHyOW z*rP>*%BKS@NMz=eNyP)Ljb=01y#eUEo%Iq8AcZF4y8YeM&S;lP`LW&9^URA;h9E+~ z^v`1>#gPG&kWAZfO=%3PhC&} z?dSbx(4Fh`xyx*eSB_t^=_h(&dM3wbz<@&0*DcK+KKpl7R!8|Uikx2#|DCQ*W?LMk z?`#}UXhtrA7Gmv6xn*3%Vyi+V0z=N(>g%7P@NDbGy;=Bet_T9e zy~F$MqmEUIzGJ%@FuZ-$EdUcLP9OE(DMtPe6_ixOf73BQ{G8$9%1Ov|-RWX0=Tic^ zzIGFjyl0VF;%)u@)w#+UQ&CVN9DnUUvbO*FF{IV+L{`>`*=M^?D{k)wE34ER*xTEG zCl(k3<)KDQhUO6a`imI8uYuIT_6A(#Z$aok+=67%9}kc~(d+m(oZ~;fpz}kZ*Snc? zOt}B=^ez2moB#hQ|DUy(jB*D=Y%EhKF>Es1YPmTDVe_Xc#hvNzHKqLH-*v~&RPJCi zY<4GOGH9NendvDhxIH)^ek}X7PjYA;v8Qhs3pBd`-7E2yf2#%bAJn`MAl0?rMl(Qz z-~TH(`Y+)6PjA~>n*aJ9{YNS!nGO)kBr5=N{vKWW>&9*Y7E{?s+36ux!y{35ed6*qTK-rO!iW#30-X9 zbVsp4W>??QS+V?mXv=iN7{O>8hvy{TR=Y`_Rvg6n&Mb9GBKEw|4Q!0g<3!C^%ILK$ zVF`91IkgEl{x`VjxShX3-j+qwFXpwhh!nBMbuvZ@p-J2hG+MXh#wRfvgAr)S$8=#!FSV`E53{`W7;DUY1b z4j4`(ZqIH(3Q9@_uHsp}CSFBt`(7~=nQ}1?Dv%Q;c0(C61eihZn3?hM|1Vx7IpkK( z56x}oCfGOyI!h2`Sz6PMi*AS3apnJmPkl!eSZAkQrc5W;I)h#Gbz`t3R^nnp8&*y6 zH-GwnVx2&PsE)0i8-2qTpOfAvV2oP~EW8q{k@T&fDi3x42cL^U5qNWTR)TZ1Xi1Sr0W_GMUc zc~k#S>=y>?zrHtlU?{@h6h5Fh%KCN0(~8XAv#{l8F>wWZOi9KS*Zs=j$Z77dWq)>O zas*+2+Lh`V0*(rCDXCTKek>XG)>gAt*3BZzHSbqKWE?`C zWQc);PIhlU)H#n14>|pvi}9bFjz1Ub8zXkdt?t*i@yQDz_tT2k&)hd-ZbbGMVO!u% z0$S-+WlYUk*A2~->E=@4+_Y@y@8SHdd-$I|jyZ#L1|E>32Y0M!_B^pAuO0wBKWsVL z8D?qu>yb3TBY7ao*YmYlKT05pg{9j2bERgz5v8xIe-gY~xC=V{pBx==%oWr^o3A|J zUm#hcE-x8G-*$YQ>9KeWW6^?b74fn|PTY9&2Rg^e9ZND62h-|4|!B4hG=r#VU?Yq)IHi|uam{(_dx^-e#^ zOOV&L{r4q=OkFO1Q+lj#6KH+%(bCF7Tbns9HdnOYntq7M*NU_%MLcKV2|rL_@Z2oF zK#qTWUMh2FX1ljdCiI~MKx3AUGYPr;u5DIGY?C9RE9Uw9>hpPv>q&aLxZ9Wd zO2e{3f1XF$3SAW+`-3vm**v*f@HBJHXFC^1mIyN_P%oVa2__yZO98V9(BH6&p2m{~X}G`fVGxx|Q9SKWGNemYOL;LE zt#p)oRL7B}SyLGOY1A>Mmo?++hjeWB=K&eQ5F6FX802D^?{WBgK20#==4Sp2R?9!O zUGUk?!q%g@!B_XJkvM$>wp0_Sd7%dnZl(6fj<0ob;=~o6d4!cy+6{Zy8e!zdYPm)jY8_B}# zPoXWqGM6U_Rt>o-*cR|nub2sU;HO*HlV!${J)u5C)bcLRF7EGaP&-1q55fDN&Ka~8 z8c@BD!IYE)P&Gg5aB0$S1lXth5(?y*0=oqc4eP7UTO17P0ZbwVjz2O%V2?ot^3436 zArQFk2?4Y+47;dyHA^T11deyF$V^^I^29K}d?LX0JPy0)A!jx=Yl<1GE)#nHDTGkB zZ2vO8Cr+4=Tsi?0jRCu+y=>Y;69xz<0YJ_7l^!hO0>@tW4aD zJ?dmrm^EHUpRbqjf;w%Q(5LlHJCVX;_;cDdh-JWnUn%+}%=M)Ddzfzwp(`o#eLa~S z%z{NtuKW;ZVfw_C;~ya|&jQfbO+qbA!cOYbh+EOwI*AoQ&bxY;55YLHir4?{`wJo^ zg$r_WZM=I&97kDydd@Z72;8Kj_Lw{iU1|hxz{_qSDIhes7F~a4*S^{#96EBemaM~6tYrLHAlBb&O_Z&dnzrf zV0lxpMB(I%?d0oQS9F)``;1mE$7QYF2x={@XF7%GuPi0h5y{Rms08Q|_ZLzB%VD=Y z$$)SHsy?m*tHm}iRJL#cnx#a$wfsP1zT<63v+gIKIxaOkNZC#jr9bZ6=p?SVA+}^GIqx(S)@~JTTQH8gP^Pe+DMrd`H((rR6hU)-_&=ZUql#$2 zC?-Bhm)Mp@RX*7<69m$etMVMK*i$>-z%Gwieyk&2WLTC!fRwa|p_B>Qx)eel67=d64ZbfHbSLVS{soiJ6O^xVJncz>xD-k0-C>-mlo<2}kf zr)Gy)THj`ez84~ABIm@+7?~0QQj2W?##)~bG2Jj|+}Z=txHrTZTqlHX6Kkx{om|Sn z6s^h9i?ZlV7+zqyfQ^D1U(2PzXzT^wkj+(Jp|PL-Foe^vNzv}zZYk*q&a3a4ZP2Xr zQq6&cLY?WemRiT}Zm@_g+>sv5Z|-jEbCX}z)UUl1H_{l(*dc8|~cB||TL9@hwJNKLF#Mg8$2l74!mcp6^c z&icIn(UJb=8sgytkBl_qR^v|>qMFvr&Hk_AUt(y*Z$_mGx=P?sR)pFZVjlg#Llu%{ zDz)D9t?f=beSdILdhq-JG1p=dZq|HTsozk)d)c`s;obH@*z?!4bghK5vY6ksLfxAw z%`v(-c?7%${=UWL`ucZj_KqL83<-z|&TOuMVNRLCUg*{XUv3pn$fsc+QB{SG!-x+){N1dZh6ErYll8&x5UBGzFx|4nWn6JuOH04x(xHrsEDH1&i+@& z*7^-U9jrT-I%B?7nb_ZGb*E0(Ju`YWUap@jB8alBwOTVzb4 zVodN7{spKan|d654H3pHX!9KtpDPY9If;-qDmGk=Rw^pm;gu?AQnCp9iCnB63+ICd z;01`4*W1s^^~7C4S^ZVRdDe@hM^S~Mtl+k}9ii|c!Vs*O$FBnNl^a9~J$^ETp0@dT zB0`EnEFW+(TR0Soq9mME#r%KsJ;t>kgfHY}*`BIX2q%ps{v4%J`)LqJe7(ddpjz>| zNuW9}%Ka``$jQJ#80?^N_T^ zoqg8|b3ATdVw=mum6zvWaX7AB$Vs|w6JEIm?xUdjj0^{uJ2r`oj%wFiX^T-S|?l;D?@qI)qO)-tlP z>hR42`>2t$hVQkC-d@RA9vk<9mAoDLJbl4!Qi$c8g2pL2rnMAAG^u%&K&1u+oXVe0KRHI%=w^T5T|8Qt$6Bvl0 z+u*QVQa>(u%5&~n&%MDxMmAkOUkuF#S8OAt74L&3DoTU}PNzJuzygNuNRV zoQ~|9dO!EX>;stszcv+;-Oi7?;-s04Dy~0QPUh+HobN;kjr{!ld=eFN{jON#%?e$3 zyZgNph=}K&Vz@99LsZ*BdzS5Dsdoh3i{azeJ_ZT6k>%}lK>OE12fh6ArxbLKBCw;+ zN>}Dk&P0Lt;9-~VLW1vGbi7-uq+cHaver1GO-_TC>s|fMP@ZjhToRw|6w!1W{Y{xq z_PpC(a}7?P01A=ACjj`RS0}!04bbs>TL2y&orfpFzVDaXr|}@O;-#AurdX@cm-DET z>v|B`S%1`n_!*_IxX!OH{moC&yLrpSv5O{I^5zt!WGBxTo7)v-5gWx)}&6b>7qNo2>51h8zn4E?4^Q|Ha? z#g*6fGCBpFpz9R#Gnu2QZ(eoVZ0-4e^$~q}*&(oa;^|pAPg8j%LmuQ_n8<2g0IA=y z&Jd{31XfBw^6hy#2k4JhrhDccMCJ5`k}KU8QBT$K8<9&DkstY-PIF#Yu9}V6>s4a7 z0nDYilB+OazOm0V=zN2l3vuDBJ?UYM7Wk3f_r=h6m<8#zcS6W;O=Z~5R=o8)lB!bG zdNf=4P$!C`pb*o6{Xd$TtSGfR-;e73xcZ7?o6Bk z2E^!>^=)X=>%g9zUYXavzd`l1S+V%aaR*w=0T>>^NW9+>(&w-n-!hL~0h=HS=eNG+ z%#V_{D&--_*{?ZMPmuTqwISpQEAEj$&4ZhdHt8zu2-xlWRqTt|4Qe%S_s)D&>LLp~ zR!)U}eduALcU}a2Mbv-?9-Me>=VCx)BuNjozZjLRH9km5f6p}2@m#Ypx1LiTufgBP zYZ`SPOmr~@zt+BWoWQpTa=c^!ztBBRgbA(%y`QB%KA1K9CWh}f z?WV-ev76LjeGF)Ae~~V+HmEi`WxslK8u3;l2z6!!7uO3S+zQXx6qJI8s1U0^B}8hi zzIpfsTi@zvgI77`3ePj-)_t|l3L+TV2@B%?AzN|EWWt6 zB=PRwlx1usbJUuZ*amel#5kp3Gsmc9`N;~y$bcu=Rz8kv#J;D$cePHL3dhFPQcsAe ztg`6DElkL>7d*Le8%I)xubJDo{I!3fX;yh0G{U;+@v_`B8ht_vt{e2Y{JNd{a;&blepY21 zd-i?-mb*e_*V+35`6q@`g`QCtIG0Zmxj8}UG79Zx!KWI5pqQ16<@^gArd;!p6_D8F zM*Lb2Z`d+GTho$+3`cb)g6C+$m8+|2o z7M0U$#xmfkn+FYiN!}%f$+`EQy%scs5|lJ@Z7)~pIEPJbaxn#f#lt7=iBLGrzm2LJ zH`8;HZYCm}@}-ei+hQIG=QJ{L06R-AC*2U&-0sbODtj{Fg8sg|{)ixRf4=Tv5aN|E z3;}mAKxF?609Y+e>OWX*nuOnTgVA@LpZO<-GC)lk^Q<7DHYvAK%OM?`o$nXnnTGZ5 zHYZ78+tW6LYwZuP>wlmKr;ok?*hDg%Mg77c{XF}%ow-v9m!Ydvfd6!(La>KA=cN^N zAdB8~;tES`nBp3Vy|s8c>~75b%@1=&PAs;1DIm@7USnJ%4>( zmL;m(q5G|&mb>%AZ#NA8r9KiUL4i5yPq*dN*HtUHygCZcQbCN+l3ZE(!kTAmTK2=a zgyfuGei+8W+j8<;W2Jr_7=z%*M$k z$8Nc-S;oZ#0;mC|{k6Y{cPV-{-ET_G&nG0<||!g-)nwxIn&{^?#M*@N`N z`yam;ir6(&1!#ZKD?$mYUTF5G%*Q{y)+2x@H#jZ-ur;f`}60jG2$1*E_R3%w_6`4cdS7D;Ci~hpn!}HF^nyjXYV^-P~-q8D3zizxv z*mW}`6gn^VhMa0OK7Amn7*{{eTOXXsv$$<{?@-xr<25?DW6G=>4SD@kM}_6})FX(l zJu{iv%?EZFfiMQ6iDIKHVw%h98rVzr+49+YZxXNB;>muOID%-)Ac=C}1J6XI7MERx z6>!3dWZs|k$JR4yhVE=#58|3JI)}2!`QHP{OmKfA1sBXKk{5aQ@3MZ2`y@xK5>IXh zqnN-smR9W#$D+{w5mb-KrY7QbFcn(itz~q=_>pWq#9c|#TYk=zizq>-kKqL89%e1% zxVK!G!bQ~QIJIgqt)reazN0V7M_(!;1tFuW-6M8!#Z>oKKFd#v0fY$TAQoMsBP2 zoobrY`^335XSM0hoE=Gm?Z7_1cVK}KqegD>lxF9kO~%e_AH!*AQOYv2<(R___uzf4 z44!_e`Gtux^o6PQ&lnix{^7}z7+-igi3lg6^Xzxd6lm^Eq!r)UeuD5>bSXPlOYyu? z;580pap}=yu94VeLII{`xhls7*EEUs`tvtpBA!tN?i~6t%zRD`6-IeNaT&S_>d&NF zG4t;9U~y`FtOsAWSEo@I)9sCKXjRNbY~7~bvK!)-uvbT7SDqb|Opc`4q#bF+X-b7~ zUvW*4DLiiTKZ_ZZKZ>n3mk0G(_N=AI^Hwnb#lQ79c$Z-)seYN7>+K>b1Do%;?aAZj zQ_)y>Fs6sxNzANrZ8r^2T?f)SmtN=y}b-UUr2v&(;Yg}g4Sn$<#@$Y_|jd9(QDz+@`$9yMoKj`o+l zNK8R~$c3ZdK6)S&^@Co*Tg>vO16}OHtkc2CvSy|r5NCN!#CI>#;C7Ou`{g$e(;V-U z2`ZbP=4?BuaZ1Tm$|qP$jk`VggXz3CR~G zjQIvi-+wLjx$63W~OHZ=| zCJtv@%=F7+1Wq#Q7s!ZcX|uA=bbDy{$LY4g=KQfzj6-hi9B4B zdn_C1-k!k|EKu$E=4$#4MV{p~u4RJ!Wwm`TvS;6kQ2 z2tpa;>*sjbO@sI?vapD{tVHAH_(Z60Bt;%`)#xQr9wQD z3xWTexQmua&fJ_1nXQENb38B@Zd-Qgv0DuDH-DzDB=Pq4J+WqLdC)R5IGP#44CTF- z@q5cW)N;nJce8Xv(h^Y!qzw`)#8&%Of@ELSiV)xW#Fv8Yw}puWl!~soG@L?AU5^j( zN*Yqg`d)0wQ}nfvXF3Njfxss_lSG|~gpP`wg80xU57`DD-TW%Wi(}^U9*mV)XC}~1 zI>WfRD%GRXN_o1*XS=1dhDU0yFv(oNTRmHYqE_DP#BHSBWL9!S3(383vxA#n3RQkJ zHTUC*f5;>TD<)CKO`I#Uoh)@tPRs_ITOt?i6&GO5@qG=@#mU`$n8b4KlY2slH*6drA{)E=Susg-tuIFA*Zttb+g+#j=^l%?^U#DIL{%S( zF5}3joaDQT5Yf-`m`y+lO!<&pDKRM$b6F`pAJV){(ig+%*h}g*msY^hLJsLg&g3bN zzeWK7DgpZgpv)smLzbKKD6#F-H_1#c0=AOsGB(7vWeuCnbsGJ|>2Iwzv0hgcUEFRw z*LicW@J(?lG(f>SUC26{c&-*gk^ab?`_m7UXvj-Z#jG--W?1C!G-R{F8{9jO(`GVUnX)oA%%hXw2n#S3o7A75GwbI*AQ&LK7*<*SGOt zXpTbWcwkvYBh`FnGXnn}Mt_MnWrIQmYfu1{4ItqgnNa8*ZQVsd&2g$WxWZt?h z()XezHIK02!hx-L+*TJs2ltqb;1@F5!{+vG0X2r!!l7$Ew)#V^h|+<+8wfx24- zTC}(u6-ARUZrpkJ;Kbn@BC#>adS>=gU?!66UvnI?G9C4JD_YQ^fGwvG{{3Mu#E1=Q z3)hQAzcDrP$~c5T#`)#_*E6B~j;2b!M~i7t8o|*FUcz~i9H|?jZp%rOz8BMW@hhrV zZIn4$Hj1r59XUi8VY}j9;+}~=%Jv!$r!t9oy3)l8(6$aiD-+Sgks1)%nd7bxV5K3d z5ltQ87XuS3^&A2(hv&bKg1rvr6622f+`sFyHMqMRkv~ILu^le8d`b*5U-RkPYnuB4 ztl7{ob{|ga>ZP zS)*LW)cc80ioQbow_Rhvupo%*7+I?=8|{-4S$*8|8q+naIyWz+AViD#5`+c3%6t%6 z`NVDqw^#x&doqhyzmggEH}V1)F)nsdyb}p`Rvj`5RoEvmm=}Sp8(ji*q}6PxwY2AQ zpRa#9gt!DQ^a8S_0y)Rw;#J7OW~_e!%<^RJWX^Wm#P$&tKWs+ToP}C|CojLx%DLk> zhx3Uot36K!a(6~zg70S^`!F4-v&5)PcrWSqF@TKp4OWIPmK&w)G~m5tMy+qd6;d;f zRp?)W3D{HVR;#)_C9}eUL1Nt12OFpR3hGMwo{FN~$8JWJ^ED}#6DM>Yt@Z7oc;>d1 zTrXrT?IkI{1VrJ=q^q%@n;t;L1w2%&o0dPp3A+K_4Hy5BNxOWSR}USo1}~E%DoSKw z7mp-vcOFtAlb8}~=_|eO5J$LFcS1@f zukF!C7lJ+vs#yhu#MKISi2}-Jc6M{?u^70HIZA;b185@YdHc7m4 zAjtF~r+BwH4KC%>CWT#%RCv;xik2e{r%6z9uv2khrPl&@BaJL@KnsiG522)_}N(h>d#;^2GH8j2dcYbHYhaE&7ws4(xx+&t! z6Oe&|IYw0j$mJ_0q!dG%8OQxXbtV-B72I+vGHw9REcrFR=f7E>xT+9w3DR(R2;P_w zjB_{k!`_VJVZ-RD?x{Sn-^WF>I-u?7+Cu+SlV(-by<2 zpmT0)lAmv|nN;_1DPFzW|E7kMaF1|epOBXCodBCcifRaM#&=`#olUZm`<-$tD}hcv zeKA;pV_|W>qxuRrIr+83=*bmDvFKu?Ts}lJrzYiI%WY5SRH87K;|gnBj>d#n=x&UP zqGv4w@6=R|Y?g)YpVpXkcmg zBR#%4qR&kLHD9Wxpcu;%QCXi)oxoSfdDkq(>Gpd9Fk2mGi5kymJUK&&E6{ah$@DpHbqvx6M5tlom_H_#EAtQS;1TRml_K!E| z{I%MI>l~IQoaAGE7??D9?#{R1`Nw{Mp#NYL6o>>?Hv|=C-Sh+KE&nYZCZ2Q`APB#% zV774>8soOH`xJ(6*vU*IH=8Wf#ypnmP^kB@6jZ&(KKAs+S#E{vWbgWYr0wOfkno6h zt3<|Fmwn^7$y`!5IJLf7L}(gW3&$DxY#=EUHr0c!pBOv4TzXs)8NgW} zW^o`r;uL~y^>9Q8Ek%~81AL^NRiy8cKBK)SY`0_Yb%Rvh3U%#iH+{W&ucCWANXslc z&C3A#^A&B??eQ7`8#e17m?G3#Rv;-PHEi_uTKHCQ5$gHt0*F=_wtt7D=s{&HFB_um ztXyF*Z$g<=rLFF;N66;Ui=kH*b+HmYoIAg9$IW>II?0t?`%x-k1)Bnl??hWzmx_XG z^%j`9vV`2ABk6Sd_oNs7beJG(Gsd@LTNEea(rM$L6DNq|+ayRgL^~^?T!pYP_fwpp zVb<$${>9+Ye=;$iP-$3W_S_WDlclu1e#yszr*&1vI!)@9377|MmG2KpPma7U@yvJU zYlE-L|D>^yD6OLgcNkrItN+PZ>O}P8~cVWs-yAoX>xeox}-nnO?Rnz2dhwc?9kSf!id+ zy@P!*N+Ph>a5nuD4lGKY5XmtTaQb)K+x&A>>N=U&to8y*XtX@L4NmspNA^$YZ4;?s z8g-(dRd?{Q&`pYF2soh8JdMU`h@oQR9WCBue_j4%xv>PI(~Vht5%$2^8um-1Uf1H5 z!#Q;QOL$^?QMF3R2N}KH=?3a@3S9E?6`0hnS_oh)x4^H{$g`{dUrCr(KxDJVw$;*^@~}a7{sLp!rG~YFBw62<#pZ%r3gduIzBI*HLfsi2sO zDQKm~fdY2N7HFk-C_<+FJvJ)~SSo$y4RuhW+i-G*nqF66K6yJ^poLIVKw7E8Bh%SFyp5%ZWTer)+K>;0wEaV;51r$M~hAV zC~pcM*Y@$^H$wSX2G)}j?jVgllh1*)%9*tmOkN41<0O@>Y0IavIkF*P5UcT zKjFHy`=f-t@=Ac%yVAs=&;y4xZ_Kq2S)Vfk&g@)G11~ z=B$;Fq?`Lr@CTFW^>2I$`_^l@wmB{ATrfCQ}gwk{UktbR+YwVtzO@(#TJj=nSsIP;n76d8 z(O7e1P_WxqMmF6Lo<8{%Z6baH)(T2wT2g%;z}K#$ntJ(uvvw)OLa!K`*fpF@;npVk zVN`$wTojiDR@vOM<2w?%%K8VAf}iyfpdK5)ToYXt3V2ACxot)=yex5}dRBKCx<&$8 z`>D>T6t!|}pq1IwwzubpjXZx4-jgH8u zcTFklI#5A)VcW&%VVdV=%ShJ#IiS^ZbW|axNL>jD8-IxH{l2;hcuR!3$BLTI$ z1Bpd!HZ?f-(C7zKTRE5EBSDb`K6f+7M@MRGK#MzFpI=?}G0UV~(T0`Ya>I7%OKkwC z2?^~Rs}dac=N^^medUg;Wi>wSJJ% z^7|WU%`-N2CUxcjSQZM9I9DZheK?9YXVpaeT`0U~SUlj=e7Z`ti9=^XDot{<)M!Bf zjr&#SZ>`idT})PRIv}kH?lOOYp0+mL5x!gf6QxRnxCWcY1g!_TE5$limQao4l%@21 zco1gC<>-)r!003h4d-H*3tUNh7H-=+xJ8~R>HJ$=4Ua&Q3#^KoFA9YT-kJno3w5$- zT_*2LS1?Lm-`-xGrOZ7Cl@a-r!D(Mqc#)>TZcYT>3%H*kwUo16Bk{3WGiqzNZZh>} ztHQ5T)a|e|A_oP)lCzTM?*@5j=z&pXlgjWl65^Zdc1Ra*o8RGN0T5O7IHTrKR^V3o zEUz9~T#fTH-YAhvqT}6GF6Vn*;eB}GTz1oiln~q9;x;Q;U9v;LL^JZt!4y^-QdEIr z@H-}j%HTnSyHYo<>d-VR8Q(uqqbeGnDWyOej9n@n8uKQ9HJVQFn8?c0OZ>-j#`r>4 z@)#`>%r_dVFaSW|bm7#8!3&@3=!NbEA$ayluvuA!1m5_BbqqPhecdWVgOLIog2X

K44c3Q00aF`?4*b$AIfN5145lzkQ z&9!mnby0?J3x?b7E${D~MUxTBWS_`b1e{*g6ucQsW|PM5vjtHK$Z5OM$>*NFp)H9P zmnrZ?yxR-$X8s-H0ik8-pHYevbp><6Q>SH`{uM0_pTXI)X5GxkqotfLcUhTcTH`DV zqye-OeQHnxGj`SRFwEA)(EZaGz+JcL8zhD%vWFEQvFK3g8qn39wUQ8;qg!vrAMFrYD@a;+&SLPVKspK(NJEU%X2k~mF=7yYRwtr; zm4+GKhw0s&IXYrQ!41$pEuu{AieMYi)$)U)S`P^3R@CqiBn{}0#l=H!TexA=L zgFLR@9T1&Xw6BOItgKr*F0__>cIwEX>m+b&^lx{#ay*%LlAkancS7;S#USd2%49U( zth@idA>w}c`b7a+q#(}o>x>pffYS|JBtrBSZ?iVX_ETfImD3t$`Z1q zTGIfDdG0M;5e@i&ruYl$uv@SA6$?hyweO6fXHaukgWy$>ymNEFZ6LZY?CLX6@IWms zW({DYqnp0Zon+j1{Py*!7cbX0gU0Y^x|)nN+Yet8fZas`ERm-Nvm?VXCyOAzN!A_G zxTxMI-vmwJZ($H@*5g-N4?EdFoovX~M-sXoCP8$oOxSTiu}dnB3KNlk@v2nbz*R>( ziPAdd&D?c!=B1uS4SKG$_``$Dd!N zsB4vKs{fHavjWAKr&8nmB;?)6V`;>H-RoO-;-st79f-wuO#-N6j~=fWycLJd*AW@} z$x#EgmaB7f)XniGt3&X&Fn#1#_D2<&Z~pfs4tBmCY0x-HB|I4r^KHeobI5&mdj0;D z%MR2W69QSS+C-q}YK#qnW*`=)pZZl^bE-*-{n(Y+>`U(+p)b5ZFZtuW8F5`|mcY&z z=<_KzRHERE#b4VHMAl(!aWA^Mod3hq|GB#i5Ov6jNY#@I>L<(9 zq|1zZ8qZirGws*ZGUB{e9dw_o9V6lmg-jkEM^cn8-+Mg@q zG%#$V(twmain{Ie`>a*AiJa6C1{GXa`Np9WZ+_DD!E=lmv3Rpc+|XUax8SJh#G@1j zY*x1yo(VY%zZ+eSwCBcVOR^r~;3w#qCj8_HQlIwum~`+no)d{>peqc7mN z7fxkC5zmDq6T@D=dY{BmQaXA`NcdOBEwCxBv|ojhwbje}gmUtWLs<`snX`y3Ce3Zr zu2LC4v1&DtDPylI4Z)l9$D6O#3lxlq=36fc|CK+frO+&?R%BCvmnb(_Blvg2&b#nN zR5jj|7SFd@X3Tx{Xo>Op)*)CoM_%b3Hx_WaW#@0AQ(`8|Z)@=toArxR7*-Fi+xSmP z&V`8k>N+7NtP= z+VVIMF~qYEJBs15zls8zGBKdQ(*LUk9Gfk@=EQS0ap+37j3iSZ>H_fz5!YOEL9bXkd^gelvqMK z<|JJsQ5Z}lp#9pm>&iT#Kn>gt4#>!O1SZ54*h$~(noX9d9AQD+0yC{AFvv-AIGw~R z3W@QgRRizXFoM_7D>p7S%LnqpMsy~=r{j4W>8W#RUkr2ac4kE3^3XY6bk5*IBX_Kp z9yV`~937;_UwSs86^VG*MD-rEu^VG*laBGG$!zqz01_ssqJ*O;BzG3RxipK@G`9zIie zjPV%Et@^TD69`b>%vk79XZ(42aS(aFF11C9TF^mg{A6H${iqE>Yy?%KrXnP&Ze%HN zK#YMC&*&2dh;b<*U=Cf(=T@>VvY#)F^9ID1Nkcg=FX;Q2uBD?7*a!h&Xgb8N7xhu* zp5^d;-iPrKO177<^7-Au{1Si=mpWyktyRwAy?Uw@`Vp3ON3r8 zUWcLwpIU|}%7#QkGltt_2;Jo>i}*Vx){B`HXwcxtMF5vN%G>&JR3 zc+Sq^m@#XyXuk(gSu_ISb7sG44C(0%2#Jo|1s(hPP-R#nv(&jN#_)bki-=?qB9CXq1w3$8$E7ouQ%f}3{{MGmOWnq0ml@E}`C+S}1QCC{Mq#GW35Pu-bqC|re=cf0~_?z4bcB)%9gr>*2+jQ~~4YAtH<&0AL zJ1Fe1gS{|azuGAu-Dn^ChG4dzv(s+qjzBGbQ9Ohk*OG1$%-Ge9=`#ZG-G6M`dR>Wh8Amt;YeD$5|b2jb0jRxi3*4-cO&gTNI<}TXP(ignoT;yN7tv7oB<&Q zmb9aFmv8-PCvS?iZK$bB&kwiCw{eLW#%w3W5}Z5iOz|AV;cG!@P-g0k=amVPapz83 z9#(U+DqtLYW5o=V+NQ)m3Pi+szr zNWsxbyp%s8zT6!Q*HBqn9E1Me4VTQHRaSi(ntCNxV6Ou$m{%~?v$SL_?rxURdJa~- zLwG728tN(8<6CS;qjh_BVVyh975ct^IJ1Hmn=?3Fe(X{izX z;8-cfY}U@fZ|eJaL-vf&?q#U!!h2P~DaPOcU+>4_Gt!jTff|!5i5PtI{;GSWHO)U5 zH)gAC;NjJyd|U(dNYyJo62L#$2NAYzWwX4I*~T9V>#j^HonF=z-Bw6+6fHtWvy49S zO7$VQfojAYouwqySZXUlcAmLEQGztM9dUi?dHkL6_1h7P&w*EMu5q?9(zttPl1Sc) z*~1fcDHToEHQF%+)^_zYVQt?~h`~m~#gWM3C!O}WHY?$cU!!L)67BKY_bZxTHO^!& z3Jj!1z`6p|DKpgK!T|Q?6hzT%>-S#b`5FfVz4m4&2L~bug@I*d zTbd_f(H%sUoJ&nwh*wsvGN>hUt}YOuRB2H=7_fc6e1*u_kaVAp&V9^5G8&dwzMUa8 z$4G^(?DV09bzS~gl*)C^O394Rio?{{pO`oj7Uve?bHuH*#V89@6K!!|5?1drt9T%8F>MbE>+X!$5VD&sShpQ$R*5gUG;9vPCd)PPC@Y& zItEk|t{G8q`=drL{Jbo^bLRc=Sl@JgW%|>9LnBVBSuqf@S(n}YmGyP&vAv7NYN?G$ zdY(BVwSfqh*#YDNGK+L$WAr*?wev7N^AAwL(@{@+qW4HVA*lwSz<_1(Lnt$T|7=xG z|Cr!W5(Cyp+-%!zbj-UPiCsQp5lzz}G>Rz=B2sqY7uf%^wXXjJvt>e}Ww&6eN-(RJqa(dKk-?d4qEq~d1ZjN)dHh{+4`tsm}3_U;|zJrcer z_8f>?_8NL9vkmM2Z1PJl z&@1aQJS#}qr=JH-@j;NOMZAu9r~h*OA{G5oo#YH|B|dep!}?P9GY_bU>H<5bG*mcb zQbLRLeT+9Mi7V-`T^`8*%!wc@`Qn77M6A{x!R7JD+G&$n%Jl=p_X$eIF&%B45deC< z+id<2QqY5HkjYbOtg;$;MEwW$>mobQKQ~b1-G`M7rs&&w4JI-8(q=Y5;<6jZaiOz?Xfk-D7uhL0@mctz^9eh*LielBvv3{%b>^I zEFf%y#Cf8Voop`>$do)hBi~o0W{ET5e+W}Z^I9c!1d28s+fS^Ye^h`(i~n@B1&H@b zx(JQvT>|mRjE8o?lW+qt8iJZ=kqU^$eg(o4tT*q4E51zwNR)mm_XStQ)3H#CU8uE=~5yLrJ)uCip8K)7R}5z0$#u8 zr;hMexX7S*ZydUkW4qJT+g*dM^3(}tDXr-i%FFfh#l#;<nnyl^p zRR*NfM!NY7TO&%iH+24nCg{S7ZFa{9EN$cRN$x*Fyr0C`Ph_#SE&6TT{M{;Y{TFvx z(GL-V#;slFeiZ%Z>KCPYy$2!PzZywtv`~^)bXrlip#d*DI| z&0Kx-khWA zM7ssv+EWlxS1M|I;9tb|qX%dMr@8`@sISyG{j9Tjk#Y2&#lTt0dU$#VLhe*L$~^Gh z`3}g^@nUAK+Xzds(^^$|nB7oVp)^5-_~PxXTT_$o1_;O>?HnV zE|XQhN4mZCdo9_nZGLDarO>dU;#k&&Br5fk7T=8pR|c3@LKMZWFJod#PL#MJ)0^nq zsk7RpLGN63UuhUlmOE}If??2?b>?3r>CE~o00#gsIAls4RcaV6KYX&XNj>%qQoDY^ zIVm{)jyn^dW9EQ>=;tCNgjQ!j6lsBuke>x~+LFu`BBRlYOHlaAqQ8`}g!g$EboAT% zICL3(ya0KuJd(Y?HJDv{>>vuG~vBys- z0{(iJTzZI{zUZM}Z={xw=WYpXOYW0s>2@}FfWvGoo_`6-jKSdhx?3=Bz0_D#S*E`1 zFxB?=$7IUsLUNCtjBg?Y!erm%>7Lx3L#MNw3)RGh|_X*0hMPjA2L$njUpS1xp zW7z5VoqVY`42&+Jmbm+n=pAW^ttwU~$?yFar>^-%QwcIVgMf{>q?*K~FB?NA35m=p zVcNVy(OEH0P7Kz_=X-NF2cHk zUff)syzn9GmjOi#9c}-Bb4fy4xb^F_B)+(2sfytjfdoyKDeo4#${{+oy2rl&Q8az* zVZRdeKYL12|2N}PB)Q7&B}ISU>tWF;>q{4c`(|5dkxT<=)$1ca^faa-@H)U+3nQ@k zpRH|t1@0Iv0yM1>Y3f{Vw&&+k$dScXO`3rZWdUcZr3wwL(ke)fy2#-`^x#77b9#bC zwI>L*?>G*Z(;CRBKYxf6V3F|o`q!-H)o{6Xz%K%NS(-^v2U{m(V(dQt>V^zX_#CJ2 zNGfPJ)Lw^hL&DHX-&4cK9=OeQKdT(}0e;y->!A|%N{R-U5+RtoYhvG@ywEG3dv5|Y*lZ+mJ??B zK6Sp`jzzgKc#p70xlIQzu&%+IEKFR5{b1=O+4!{QX!fFsv<`3|zWv_AlrAp2kIMV2 zMlMRZ!8$vpB^a{;sZTq4{NV~m015boSd_?h-MYYEUEn963#2;!5cLv12=!_g&G^Rz zVCf{*BB|u*^`R(q%|DkvbSl;@%fGU?_Pf5k-TFp$k-SQ1ItW_VMxN3l>^z(M;Vvga z%+pUlX-^}(jO~-uzrg{k{P-h68anz1QIo?9!{lcN?k8NK+ztkSNf6Jo$Dst{l>*ZC zUEzD%Wa$g-3JS(=*m$tSxZ`X=s*K5>4}6>vi0aZy{p(3C*5^G=y36ZS@okGK#(2gOlEtEXMeS^m~k779_c$FQo`@COTs1pq&CN<{$oKc9Z5*b1^ zT`aqp6UVDIKV8Q&?`Gf`b@e3pLjNN%KR&{dUp=$>itT^Q&% zf8AeiOxC$8`9c-SuAoZ?#qGrRBqAVmVG^&=PAb)zpe8J*-m}~&Z}o_}X(%zNEoqV~ zX?LQ;DdX2c=@Y#KLbH+bs|ST1YtNH(T77DdxF;qGnC}qkXe7Od=^M;QYfZk)9`Gs* zR`N*3tWL30tus&sy2vpTaBYdTqcz*p1Hni#LP6-6lGWV`Zo|5oWIFs}s6)kBavTO# zP-(2dL?v)PL5Co<{^zd^PP373%%*tG=-6tSm(twIQR}Kl>`dpl4bBi;MxFA`nm(jm zKTc1}ZDndmI~vkUMm+}he(2oRvh{2BpovAM`Mn3^x3WQP(1#|c2>xFD9iRjCW^cmE zxPCWtd`!p@{qlrQ;|1Y9vNtNF_|NPo+in1a^B{g0Qy?5Ya{G+k`hg^y=i{$=6g~0( z={;F5gyM4?tow?2Tg*Q=BO;8qu??kNodW8Tah6q847?7MD}_&l)6Kq{Jp1$AQIJB$^Wy1?-~1syBaQ`o}^%w0R@gcAA8{&?%zUB_GTgY4>Onl|?qOX!Q2OLa?|eh!3(<+}@xT1W;pwy+1mJ5H2kstchWalOb@^ZK%L&-o~c7DGa!g0=v9knQn`KC?7l{SBPJ)mbg34BoAuP<}`w2o!;f)W)?!cC1j3<@DfXSBBD zZt;)-Q>PGbUL9zspl4cOn!gUaedwd1Bpsz#7N8f^{s*gq8!!7(fr%ne74P0!^usoZ z^In4j9>yMArIk)Gcabvxy%-q-yg}k^*|>N|cAyAAx?MwxK1p3fmo3W7o2+Oi0%a&h zKq`_Gvqv!fiYP7W_i2`E7WRnDP0ms_GY$+wb>uCW{lRg6oV*<7gMpO1UGmRdkC$lHMhdd=v6k<`7)%-h{=Q8AIqpwGD8=-;Szzc5e zFCz&xkP z=r0SCgOE#S^hejXmSfDb;?(?C_|w$g9h0$bLG;dVkQOcI+BBAy3*^#2eo^SJ{){h` zHlyt>4z+LwEfWEQV>}Kw3kFJ^JK@#y#n8bUC=NTc=$E-Qd z#bx?MG&_^c9NZ-!LxCVU_LMV+ zlC{U?T0A-Tzfr*-LCGP8K?OQpYaJ8;uF76OUwfcaYaffu7sQ1szgU|=y&Fa82b zG6wpyE-#xu3LAU`f@EV%cp|3RE@HrFg}agwPdKp!A;>>R+DU)46H$|0+^n zMPpO_M5ZkY8@MINT4EgWDSH2Vj4`7dEhn`$jrlq0McYRua~LEjZTIKP(${gmI=t4} zt!?sQcSSDLY8q%cT33GSaVM%s2%5daIS!+X_ z*p(z;GFzuPU2CoxVVn0~7}gAqW@;&MX{ta`!pHUc|!{u5z_w?*1i#Q4)n;VPnp+|;SkddFw%h^BAoQeM%r_G@*GVDft*!ZaQ#HSavpw|9K=7-I7b2*rj!>23=bgzmYE`b$a9dlkkk#<@g$ z9~4g((O7IsT_3v+HO<{|jx)PI!LWML@WIE>paSpH>svcCL93RZbFPZ$3H~$`)Nrbq zxHXN-hC^D9erz*+tIZMpJAyt|m8YM%r@QC;N`1@}n{ahA9Ez#=yq?g!8A3S{&eD>; z&utnTfUmWZGd(+;b1luCtI8k}9tAL?K6uME9mNb=O6H?!p({PUZ$>OGff2*;-#en$ z_bWO|`3H7nmeBN0W`z?c<3qYPal~pL+sHk!)x2%0zjzZLEFb&+K@&IC3HGS{+oG1@ z!gaEI031^9_S~P%%Um|Bb#gnPfg52WvDqV~8E^nUG8|;8PydZu^#N>FgGyWm6Z=jY z*mIwIrU;*6)2jGZ1$j?H0#*rgo7M=<|atYZBVmjY7WS>fiGO{0*LrWazomZSK`2; zhyuQ+=fhIoS0K)S#-`CK#0P0V z|8)?FMn+-q@5r2smXn&SKJU3b$YogHJ7iswA7aSaWIX@AfT0!=spaGd8xWB(a>!$a zJ!n%o_z3&EAmY&=L__8I=HH539LsvJ)!lQX_Oe}Me|I;`PgM1;)f&vTp(&Z-h))k% zS;qj!fcX3HZ!=ewUvi1k*7-| zXI%nH5&~B%Or9+9+-4*Rg7vXj)6W}>Xnv#=kT4zZx&sL?&0NWd%Dqz}J&~NlP|pR| zauB7I&!=)F@MZR}0a7ST#ptF&o#`f9Y3p^FBZI|mVCP#u~KV^v3GQJ7ejM`;5eHvWMLpBSV7S9d*3ML zcwfY_NNJ|y}bL*PRVUN1iL&=@5kM5JsAKRG`JVhUSw z#U6d15W^(*9MrR)PA#H&rDj(98r))00(nl-R)w$Vcmnk#@?Yc8b%U=c92t9<=Y>bH z0nk!tbuTIW*2(t`t%Wd#&tLh$Kk~9UWv`gezXM#ME50Ho&-CTD?gs)N>N3*jR8oE3 zOYW2c|0ho-79$hYUq5GhZC$-0p2yOyK)8C_oo|HuxuzkRz)e3RgFMnu@Xxcw+ir=) zY67;R<_C%P1tz6&SvKznLj_{~sw#=YX(IKDW_|MYurBn34@J} zh57fxyhGYGF=+F&y0D+2C@-2_dD;-2?m{nsjI2pgqG;KSE}PN{FdB;@0B zSi&B8bAEkAz@~D^$j*_)+R4eGks|nyUGuh*GF6RBS}j{t4k(gNy;73Mbfnqn<*H0n z1P;Ur8EKYHBi0b@J?bF{>a?!t|93j$- z!C-eTj+bU#VSC#u;eo!rpQt*CiRER$B%o(6Z^*(9=yjitzMVYd#xL^FNPzxV$cety z6F8RDEPdm!Q1V4@q0i|*gtQKMX$lV&!PS3|Il=Q$w+cl3&W$$N5n?*2XoTni;wQ4?LB)po@94<{6aTf=%FyG`@q^JM;%k=SlBk5~&wEp=>iI1+9 z>-8~4>EOYqf75_>pUHGYf$tWQnXM3SFDBEr!>*+aybeO9Ht~sW0Y7jBHTiN`(~$oM DOcpoM diff --git a/site/notebooks/quickstart/quickstart_model_documentation.ipynb b/site/notebooks/quickstart/quickstart_model_documentation.ipynb index e88d9a8036..0128601870 100644 --- a/site/notebooks/quickstart/quickstart_model_documentation.ipynb +++ b/site/notebooks/quickstart/quickstart_model_documentation.ipynb @@ -152,7 +152,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -226,8 +226,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/quickstart/quickstart_model_validation.ipynb b/site/notebooks/quickstart/quickstart_model_validation.ipynb index ffa1bc42ae..4621eb0d42 100644 --- a/site/notebooks/quickstart/quickstart_model_validation.ipynb +++ b/site/notebooks/quickstart/quickstart_model_validation.ipynb @@ -159,7 +159,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -195,8 +195,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down — don’t worry, we’ll adjust these permissions next for validation.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/tutorials/model_development/1-set_up_validmind.ipynb b/site/notebooks/tutorials/model_development/1-set_up_validmind.ipynb index f85b592d8a..61ee21e2b5 100644 --- a/site/notebooks/tutorials/model_development/1-set_up_validmind.ipynb +++ b/site/notebooks/tutorials/model_development/1-set_up_validmind.ipynb @@ -139,7 +139,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -215,8 +215,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/tutorials/model_development/2-start_development_process.ipynb b/site/notebooks/tutorials/model_development/2-start_development_process.ipynb index 1d62e9c085..23442b3d30 100644 --- a/site/notebooks/tutorials/model_development/2-start_development_process.ipynb +++ b/site/notebooks/tutorials/model_development/2-start_development_process.ipynb @@ -221,7 +221,7 @@ "source": [ "

Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests

YZyjL;K~85P!VjS6^SO*(3aJ2cBMiC5`W9f1jLtB}fK4tpy_$LL6elVI0*t9Hmg? z76C@2+#riaU;i}OkY?wO;JDo3yeaoxC`((0j15UNJDF&y?L454?yTE+Tzr+LJ#yY) z-Vwy2?%}}UHxRxv*6X&I1wM`tu5^q74P?$ep2+y=CsV-4WCq7q+B^}Jb8Os@6rPh$ zQw8M%O6YulbamV%xm>TRXD?D0_*C*dh%Yx9tP4WEPQovws};+VIt&s&qjX3K*TS-^ zbrj2WzUg3A5@GDD0Wkv52C#c%^8`rsvF0(6=<4sRnc$jgamzF7-x*WE8?a)lq1cna z_{o(U>*(*Z1>^u2j8A-&cj_o4?$ws_hx5D(q4}^pc4LLBvU{eo?*=zPZWC_;gE7sHZa??4m^=CVer#lArK}BcI7+u=1zGK7`okP5fAXv8z29_FcrF}K{8G;!y5G<^ zs6hh?$EcxhB&?I9#O#}OUjVPXZe>LEdTwPS{6shwSVW}N{khSO3kKG$8`I%L%S&E43N>)^N75ujXz9Fet(98SA^jjH8WkmdiZ;3@X-QF zHZ59ybycm$fb#Wmc&7f~3++^nD3f|JK{nD=pKdMmZP>|GS}D~Kn_*e`wT7S8B>~ce zt$^m{3b?pLj{2J;uvyC7SESB;sW7H-T^^6+;G@L)tZ@=2ki!hrgxDbgB6eqvZm1+-&df zz&=rk<41{>*B>pSp~;JT%%w9ddbixMR^A|?8!M8la(nwc=+PW($yB;VEkN;42RS2F z|CIYp{hnoG!}@l0^(4`wh!_cy ziFBes^OMZZYBJlvWn4q!hpT12L-+-3<%>1+C2 zc3zC*<%_ETB=re|@7DY(8LJDsY@U}*C`ocEHuwn=J6KZI+l4k38v)}~-!5Rs2wpKO zGLSPAL+~GTRqg$gQ;mm)I!eYr8zmW^GlO48j+}pi;^yT>n&m=cVk|tX#spy3S@$shmVxRPgWUoI-B&+vazo8JCBYb-rh|f$-|cBKbJ+NFSks!PWAyIKYkIwk`4V z=VP6Xan6&>x-$6u4@26=Y&Ka{&(xqMNk?&* zVCm(Xy113cR4du~q#os{X|rc^ZjGfhxwccl5KP%z&^5j-lDlzwE+)7eZ|_+#-=Yay z1r7&lOmaOq@z?t2;HW6j*1LKKeT~oG7R>P93e_8YCNDahC!CXNRn@ox0b+1Wud5Re zOYZu=pO`e)py_vvT#Hp~=T|%{xnZ`zC#vT$`dh?G>YwjRS%p{7HBK$?xrL8JS$;s8u&2?b@ z)QNrR*i(4uF*PaE2+)sg$|UiGjQcmcoj1Nv`d8cp36Wfwr|MpAS(wLK%2ZvDYAW6g7?{brC$*KIpQt1XvbCnSsT$8>_Z%Z_SBiHU^CX z>;j`0|Hb|6k4aV(h%x@!1_4+xJ_!I+9{;Go;ffR&73ZM=dX7IzHk7}n|FfmGe_wL- zzmjA9_iO*3e{Sj&$543uwe@pA92gSzFTx`70C`z?Ks5Dtd&U2P+b5FPzOQfZN6tJ+ zhtIoBM`{BBRxIc{{3`wJ`qY5KszFTz3+u56RJi*o&**SP52gDMN~utw#w@ftXr0DS ztSLAzrB`H~x|EDvL0Z+JT*pW>-GQSOI4^6=936dZuDAO#st|w9z_;YR@x!B4-%s${ z>sFmP`ioShpE2>>Rd%cA&w;z-KY(n|9!>Kn66$bm%<0q6dGb^ZTbQ~0*0acKM1Rm= zl_;L%?*1!#ekQVC_N3#IIy$_rNb{H>n!7$Bw|i**I07`?tZzKsydd;rytY2=P zNjwsJb$wF0?`!Jjvi-Bk|6qU`@~R}CJfmcTUo73^yb)-kC)DnB?nD1V<6Ka1Y(xMO zi0dw%GyQ_#O`MXak+7k(aEAjDSHzgw%K3t*Z#6>md0?I_0;aREUMetsO{>3 ztBAPwqA~2XwCX`2tjbf)Y z@TGpHAgGI-VT$9?b{M}dWj=VJyczwHb#V~KTn=?$bzqJ9CGtBf&BzNWSJ@Vnw5bj~ zq8Mq)#%t%C*w)I=WZcB>`ZuWa1JV08L)-} z9O$R#9HcdpbrZ2%P|w#fK1o*#Dgd1mp{lp^=YQ^HcKua>9t9iG72{p4leMH>NpCZf zkC7(6YdF9iftVW%r_{;Ci7a?{4Aq1qF}^iKvrVF--_GG9UNYXVo>`fVmMk2|h>>Ku z-3!)~1RtVuI>Rs9UGNZvXFvxK56o65_kiwUGpMU2Ftu3*gaQD9Y=94^vZQacc%Zds zaV|AP;Ia59Tj2CUZ-{7&ThA_HUEOBw?k~_~Wob#6<`#%k0Y|?#(CZ5STh>#KZw-hE zi2MehMbhw7dW)OLnQcC zA<88Da6>jtBMqcX50I-w?<<2-AK!X?TBBDy?|rT|icgnb^fD5Y~Rso(hx8=`I(2xuK!YoIfB<>`fM1<^#K0$nzn|f(o@i8 zu|P0~y@zIdYg0WQ$7{EkoQTGCy%|-R|HszH|3)nQU$#&F@4xY1x1Rowo}EHat8$6x za-PIqLDb(}r-88MbID4}3S#}242|Ni*%98x#IiIUvG_}%^d6?{;cXg57l}+U4<(e_ zXNEMsb77#4Pw{%xSgg6h`1w1C$&7{j5{AV29yO@u=AP+bxsR-=>ChJ!E(Hs9L{SyvJ9Q8agvnSV)uVa0Zmgzp*&}@Hl`kvw$h( z68dF=+kfR_`~%(a-#`C-EXV&JpUwXc%47aNRmV}+O(m#%4kDpJAZM#8V;a0a6t_wxRD;QJpK4<_NCf{W7q5@2h&AD-jLLMgux? zcX+?x?v17sj-9GJsepZsx$=eo0IHX;TTaeQ*CEpj7snq@^giG|EX6!(a{B~*AuE`F zr87=~{5;?ZaDXT|?TVB5m1eNsK+e`0l0?VxW_*0y+tFQvgam zOwcjl2&*BaK%)G}Bs!{C43XI^Msy{6fwYEZL zYi;!GqogZn)&^dWhiFNY=V1jFX^Y$VIQfB^8XN|LfCD8&(iI?Fq`?DKlkI~;gPTZ1 z!R*@g%f18ZL3Ct8lq9eol3TN{uN1v(ldCIb3bghzcF!vbhgPJEDexqAsyeL&W%_w6 z+$RFr^%(s+he!8WOQd&?+ zJHr8c0Ke!EZVg{q(==pOQ&ju4L`B?X)+90$3@y9=)!EVcwVe5D^XW<=S=$eq4}|Zf zNBuBCdw+kW)KR7N2m676Rz$u{V~f4lOT45*eR>x%96?RMPDz{)96Wa#Lq!M**m*{?OOdPV*#ICn^M_A|Ng0{pg+=?N!CrL>`S}XVTOKLEmpbVc#mPTrHK}lxq z*hA(MW47>#vd0udyIY5%5)9Yz_V-wuw<_qfgi6D=bC2%?X0I96uBcBdQ?QuWxo8`h z%bmm|-FVLpY+Dkij+Xkcx&-4IUZ3gS zMA-+Zj&ZmX&3#gW^T_2@1?=VhXNdCczUVg<3}s0RBRgM@n4Aw_J>}!PAwuEsa?7&N zEA1hp3aKaej7a&CDjhRb_1oC)cV6x&NkK>|x^}(+h>i~|T1vw`7^lFvF=cOGR)02f zfPN;!D8EJ0aNs`TIW~2AP3U=--_ME2!~QhfE<>=t-KID<@You)+}hFj?rCo?`&*v#!D1A*dGa#Z_ahX~J|1OJSN}vh;aK zw#nY#WVK)OrC&aVg;HaEVy65ZEwAsJH$|2RD!KaX{+pst8qL5{KPy62`*^c-U;P;peqTZxd6+& z;N`s3`WCw&E`vEA>D3YJD;b0kt#6blk>uEb3TLNw{p@Wr*C$c75?>->ifl^Wy^=a6 zMo&tHKSBAselJX}WAadJ(^Lq-X@8Tg#fLBmO)vMXCv6EjWX2+a^6z|9pNY(%A*~@_PK57#Z z>tZs)y8dNZi&4}cv=%2S0IxbR=?Fb~ZuQY+a6ryPzu-G9@U#fRMcEnzQVl|Y_E58x zoo7v-uZ{tTJR_1YVX{v({EEW|zr*5jA1gCwK&@^@1yZ@|qynrEz#^L0b=OfUfN^%% zXoysVzA;UWkZ37MSxm$hv=2O{`@cP=QRbYO;rDZ?M-Lsj_+!p8Qg2+*t&xuia^*_x zf}FeoXc+yFLzPQE;NXUCSP(VRF~`Dng*W~ei2o6dgbS%$(n@E)K_E_R;e2EkKqZEW zTmxLr1a9cjg!oKu9>J4#MV88$5M83ot85>(#@MLlJurK2vFh2o<)*L_K*OkHfhs$8;S zgT8a%wGfAkf&s=jP{7U;2gQlQ<#m4;>DsjTyOCBQX=>T$nWHq*dYe@y z2$v%@OG=_n8GNV{`ApJWnplhO~$QTAUoyv*@p^RGdpqAIQecdo~Q|Y)}JD zfrY^)Dj?=GC8^og7T%BIBH|f?gKkZnY`ma9NR??JE!r)EabD@Mn@KyK&e}8N8!rE) zN#WNR6(8lgW9~Bww?Wk5K5C-xjgWX7_kCP z?LC;90m6eCN?_Edf1S8YsaLh8rEGEQk8byv9M&jmQGG!8vHmc;^=U)MJV99@zPrZI zSbw2jmYs4Gh{Ww6gMzQewfTuH3`6%!&jg#Cs^@lAQ|vdX$@uatE0}S1d-V*fN{C!X z+dZ*D^t95`=BIq-z+?ffet57DqOz)Tk8BZ)4BCK_RiviNJ&bCO5d&yV@%bB9n!_q! zE`sb;RTYJJrbtRnBHLxFbnX?4T0gI7kfRX3u?F6Gu}_ebIjA`+mnDItP}nz+(D`{6 z;hwJwyLE-u{MEIaA}VS!5^TvO7ceFQNdQw>$};;C<*_#*!%q^zFfMNODoUiu#R#xpWHj}|#HpIj^@EZU| zG7(!vc@u5W4YLi3%o&3xT8G^F ztREw!LBg+u8?NtLiqwAaPmmmpVvV7)V4DifU&MJR>jwZgD2`F)Eg!;VpJ=KkDT6S3jPc>D^l3DdRXBMVltFQfvm<+`uOAfRa=b2W_kj!Dx)X>zD<3XIWk9hTqO0N^EuW+LnO|**(XxvRsa29-&N5cM8MA@|0BYr#mjNl0 zVo{|7R-g3iPuY-0tUs&x{hw7#@FXpYne1|mhdZ?>phNw}!aS)5unVh|v{(RAkNX^j zw;ytsO+$Zc#neeUkz{^+62(tn3<8E2z7g=IXrV1+l3k6cVa0m-g-cce_%L3K6Y!dr zG#S|te#@{KFzWv^A(;&c_2QkPh$siV8)ec9*r8uqSJ^50DYm7EzfN_VGH?KP zXxo3Cu>kB694Pw22A48^lnECpDggpvIqN|&V86kW#}@{TffOZ8Z}BfS5p3Z}S;?h8=rIR^2<5P`>hwITrdf zV$*3VJc2*74O`_d5Z2_dD?u1mQP3AY$&TVm3=TwYHhO0?nm)J6zq2 zK4-_|Jou$Cn_a0FJtknyp4`ZMNg5UZp^X=2!70ibTYYH+c&tS+#U2-0l~VeZF6M-m zi;6Lj_(t{D&(|*weZXth7A6GJJHWL_qHMs*lD3mdRwb-!6Ix?y#$DL?Bh9I0ckx(n)tySg$&`qFw4a-i;%v%goKHzP65) z;+K?xqU&O>XC{Yxn8a^2c;>aP6TTc}36O8zpBaRQl`p_ycHh!|+AP|JBzYUnbRF$g zEn;g8s0~8=e_X1uN!q9p|N8+7uT4fZxvLPaBSJ&D=SH|i_I2Q{yZ$Jf>U$KhhICSr ze37C+&5~{dy|*&N${{5i32MRytxenm|kp^D` znX#W^Y>3zCCF-7pe#PF(xqf} z0^{ef9`#xJW`_zgjEgl)6HhKJ$+YHDfQ$u_&cuFX8W#;!YVVXZQ| z00V&wxEM#$P4JEff=IO68%Kg4Yp=u{r!<}3QGRda%M6g> zAR)$pAf)RB>kH577L2%=IAs_r(>8{_!sBSBY>Fq|dI9|trz@EL?K|#OegVg(8{5JO z`4jmjRW;u_nbbGj8APz34PVTcup9J;f{QX5TfKZ}(zgc32{uuHTsxKEpImSJ^K$sH zFx>NQN?z<~lwk*=UIp3Vy3nbel>q6l|2!Twz6K~EkZ(Rr-S=sqcHV~++%`UbwjFP& zf=PDg4Z-**R8_ofNVvRHf2yagoh?D}291pRg~Yk8oxg=erf3}UoC>;^?8CAr<^vi@Q-OU@N@5?&@+)Q z*EifZqJSi^8E%4pT=Nmb4>VusO@C~7oS1RIq6Bf|xx_t0Rv(!hu<7*CCLg;Ge1k|S zHxa`T;J`a02d&8k+6;~c46sNjLGbbnRb6~f(la6APyP`#sCP3RB$|)?A>TPK07)@2 zH`Xh|;otV~qen*)@F7#VmBupY zoCJK;G#{n>TLE9O^y5m?=7VlNXccSuwQ)T>UD_r%{T8VwkT50^mSih1S{Picz9_oZ zraAHA*975S8$_#jU$Y@@&AhR!I{iD35s9BlbI>~$jO%#gkz&wCNx{Y6g6{rA!;fgE zL2?t*mwxikRA2t`?-%v?y^DYwqGy;m330I$y(Vcir_RSt)b-SxUHS4H3VHTX#28bW zG>A$h1qtU1p9tlu6J9uDD0X%P$<;B_L_}V^+Kl7L@r)odcHTMqA#f+j#;h*YJogT~ z)zXa4Y~Wuey&cD?%kgXi+g5E!8ZT|mNoi%J7XVoAozyc6E)+JAgv;EFnqb{p#m!FG zeJ>vP!ulq`Qdb4K zr&?RNp^j=@g8`+)*NV9O{tZEG@qA{nTV2^eH3&dgR^j_x42EQQh7H0IrTgsCb{o-B zjVd`GyIls$*w_xxob7 zY{A;yr=nLMz7M|Qm}WyzGhqqgiK5qL~GAfjfj`SWT{ck+n7_SA%H!ftBp3ge!Jl7BI?n8 z@d^95*;<3~SQ}@jV{TcqI{^i@+UuhsvKDPcG189M^MhBJyen^u0{DSlLtE8grP$O5 zdw>5{2b=P>Mf}|eaNX}@^y4sE>Dw5B@`xf%Km`LJlhxxT!kVj}J-*i~eej?dtC))| zRkqt%F@Emoh@($N;7Ldni#lcdDta&`>iknt46=x03euAqF285ZVJXU7$ALi%M7#Sz z#j2mych9&uzIirWP*H}jH6U%p_aI7bE)1-x^()~J-y^}tmgKBAMXu^NI>=t1Xb74M>l0{0zokiB{ z1J2X@=9yO-7P$P(i7_!bsmZ1f#U;zj%MaOG;j%g$VK*(O!-7$u_KnCihBq?>Hy^_# ze_kP~u?-Qa;fx;Q38f5e=YgR$=T*_6UK=C@!D6R+Qj1H2Y)XQdt<{T5tjpkHR4Ern zrM&?o-{+zewxIlq{ZH3QX~b0}3yp%yo)w$zda2Rg%B%o{_a&lFJgCfQ-&1XIL3cQ~ zW^}x1%~M4ndAhpo1<>fie&zDoCr9Qr`-K$%3^YUf=r8)DlTu=<sOQ2NzMTbXrGWfQ{*RU>F|-cP9W@TTqFeerzv z0IshWv=}&Cs?~Pg@Q8qyGZE2~d9k+6)#dOTp3}EAf>EV^=ox;d|xS#N$tH z0q0&aUysLMkj5O|Oaf)j5%DBt;r3d!Ra@tkh4xCDYa#H<>mc_lYPdEenz^Q4yy6#w zPxWuchiz_@gY-Hnv%1<0DTwcMNV+NU0#Y1YKtxFz~v3tEOw zkFh?t`F$=LfkAy)WeoJ+4{`&6tS;z{w)?ZE^$)@(O?O*ql%5l|BU7*w*^K(;>LLpt zSzQfFXP0?^@zc_*sEBslX=TdGRzZS!MW1^L1~(s>4Idmmw{3XVVi$-RuJ1GP(fBliYdz#f@MaSh%Ar8G~8olj1TB4g99=6EnUus#`7WoVGSdpWT zSL*GmZ?ILRpmUcAO7HQk6)*0iLTXWMjz7N-Q)91AkD_{xUFkuxQdlD8GqA?gkXmL4 zZ}C^1aG;suIF8*Q=@O1F!8U|epS<(ntTO61)ig}V-v``hg*R}Y8z{Rt(GDabOW}Dr zT=%`|JGqKM8T!#AyJ<4GBVLQm(;II|t)a5qm_woj)@Bep ztxfQo7RK3+E941| z3+8tzv%Y%i_KYFNX6gP>g>?epr^*$g^-OG7Tk;$^?0gW%{%MagL=gc86iH2!AJW`l z@^1}(3=xl?$oVsuE&cwqa=Woy5I^!pN@VY6|K{Wipnnyzdh_(}OmE#YniKRy`t+av z^lJc@iM-vxb!YBsj+0q$s3~t>noN`_Q{!=Yx+u2LnxqL?{1YbPo&kktM5D0|B5z%P zmz^&r+InyzzkYrY;92{VhUnaM%FuJ|`U;p5>eZKi2YhMMx5FP3(I=;3!UyE$rnhEF zb?G|A@oOKJ0=lmAu!vVSB|2NuHWqfGyq0KVEJF5z2CHLvHf1`R{6<)w4zJ&l1_)>> zA1TUpZ^+af@J8@62hEaP{FK z@>Va>z|>#!)Rp?S*kExFOA~VV zyNALmP~S%L7D%k}id$!EYAEfP1sRic(WecaD{zef;#zBoF1AQNvzV)jYrq6uF#F-9 zVFwwn!}YrIyGK;O1{wT25PFztVcKz@+vZdmbXjXKHk;D9C@rJ>`?@L7iu#w8=q?GXqZaU;s!Y$#CAj6S8p)%gXU>~OJ+%}wBe0i89U~5T+Whb?yp=A8J|>S#oY z<5Tz$McCVKUTy_q?sw6QhN*F{Xb0_Q(iqF+H&Fb%p2ixzc0U1shdt3A29jb7h)7>exQpb8duf57n!mZrIz z_Yxx}r~7mFzj})M@I zDGhBn(g1+4H;!Q4;MJa|LEbRSyT!Fl*OH7dfI3oxgklBk6roA2 z;T*zcbCj8e@`~P58Rm)6&p8PE+W}l9uTgeT1FOSs>`vH=(ep`*|u3TIPzBWAG-)9`QLUC;eHZkN5F z+a=9stZFOcv#bSBz>5;}cv92uWa~+-TAKvrV-Zo`la}jkoIR=9(3d}8%>3|LLW<)x z83R%W#oNYtJ5qdcCpt31c}LtJz&l>B$ECW*(AG6(_&jwYk9bY1A)T&RV2JznNoxG@ z<}pUQ*}zph^0~uzp{rz`v+bBvQRV62*Bj9~pVJy`L2i^1$AW>HcD%?wF%1b++qC(w2Lpgk65ApjWTn`;I(r1~F~h zm-;z9$tYa^^c@4r8kM6ctv=(hGR;gMlJUgL*%_UGv1$;T!HLa)i*Xb@FeA(5B6ga-};&XLYQY6_umdkNsi*)-^R;T(=`TgNxhaQ zNn4BY9|==G$OJ*ZDa6*7K%XbI@{~I{YM)mv^QxcUi#XEWT~(M<-qv7>raI{s_j*nh zLvNBoE_juKu(V&4>5e`8fEgG__`;cy_BdN_A5)N~ygg?YZOYDrs&cPmemK6#>%jkRp>rF#HKpFL<776w&wVkRisWbHh!KORYejnptTs~@M z1mI|7Bh;9RoQ$cjasM;d5bx|fH z{y$t6V|xJx5S^U^`N!yO8dT!Lrvv9@Z%s1VWNicriRb9hf^Vd;SBq`;dF+BLY4iD3 z_(VRt=i#_1&m0|AZ8k=AZ@!k2i9eyx^ZaaJn{K25EJdaWO%E6f7Rr$lWY($vx7PUW zxVCrytabldmou}re7btm1$~D&Ci}xuP%lP1$Sl;Yz}8EfQ=jY1fM=ZwE0sKV7jPj_ zYMy>~oOwKHw(0mz2gqweiP^h-$L;u=5eMSA8*o&Xx--=3rszEykKETi#%al7%7C&oOP4 zVS1qIsu8YRhp_q#3D>DFoIm$CL{Yu>$$?1gA%^hl_ zmaTZ(eQT8Lj+UuI2$E2wMeAblb0W$B0f4>mF6Ve>*wl{ZHJu8%`2R)TTSrCNz5Sy@ zi6S9Nw}c=KN=PFD0+P~0$^s|6*LQ4<`XR>q$ zcX+;Ep#Rp|CibE%!1p4R?RDl?t>!ageX~w$k*myc%)yQV&Ti~fNFG(K{rXL3i9!2( zJUevGJ+AKwt~IK!Ax^`f!_V%T>JEEPsE*G9%S8q8h*_^>k~(M_n-KMB{Mrf8pDuc6 zKYKg?3jKWUXp#!eatCUD9J>vRj(JYm!B)D@4ck>%5&H8I)(5>?!u{_~S~72O;XUuY zRR&frQ9c;KtC~4JEW2WFBH@T8joyE>^d$>S(?wwdUw&&HInTFrQ`dR&;Vb;Lno4zt z?ts@3J3~xC@Fg1NL|KNwutB=_cDk_+4_h5ZN5!9?{{^#0^G!bz^Bk(+#byUFZS4^O5utTRoxrUogdv{6@kK4ruLBz$wWTQEI5 zkT+;uFUz5O)0OoMc`X1EPqt5{3MzIl!z^G#%C}D>ulGrH0CF81@0tMROEcd+z2t&O zMVtsyp@pj@IW`o>7cdz^A&Pj5hW z$C%^GS{(Q~-&R+y;%Ma~2&JHGf*=1H)0>$5YBy$3XknL86kEWE0@?H@vVd{>dl5L{*`#uRP{@2sTC7%)FX&G)qAN{6^Kg+J83xex*|tTT^%s zPwIXbw%1ObaWk%DB*6Jb_RGBAJH`N%v-_U^YXUI5_h3Pa#m1_H5 z=?M;jL3XFSAv(6KG2OGK$66@v#c3K)73^FwfMXnAiBw*Y`>oEuvqS2TN|1Gzne7;~ zBxfRK^gW8S_H1d^7(xJ}BP~4ZW*GAg_Q+reT~W|Gn|0KQS^}Kg-gJDbSv5qpc znvT_^g|EB%dW9lrSY9Rc>NZRZH2kC*JRUVj6oMOz4AwAm;dzg#BWe;25QFbxRe^Ls zrU!1}tFd&~Rb=$qkNYNPAVSLOLPk}^#5L_xsh;zurc>&JvTWKg#{k;fo^}JjPceN2 zAbIa7!gm$x(I9IjlGvZuTGF>%Kc^k-(k9U4Ut8#M6PMO>_)>zGJTr^+&x+09JS*au z^Irq%U#7&zuoKa~NnIu*nl3=2@|nIyPKa-eTj{IJ&6HpAXsi{r3t8D2k(?lK}*}pg!e%h-4;N$OR27t#k>8-wqF)%;+2t795-py(Z zv)<<$-k?z$520tiFAI2{<`Q{R?W?VV=S^@SHyh{TAC(3^jiNU)Ex~<~B`34!ex`fo zdFNg;l&64OX0bvEXRWs2lNJDD>8ALv8a?4;hy?f0Oy~L?px0_L;-Dw}c(JO&2BC2? z4{ta2KF&3ukjhdHw36WO3cM4RM-> z_FC#pl_dkL!sT^Xb_=HV`=NwyCm^sstlVwhod>^<4GOD%olsjx*7{GwMiUYUp%CtG zZ*$Y_uur(g5MG>iql7CuyWZ5)JS5XFsdJl!+!H1MO{9~ek4RrYcVxZXNQmh$$J^F# z6FCkox?V+@)YmKL{;DyIo9tirL69P62Adm$P}A#^ckOJZBDNG#XA_}!kz0nwOU1`nkqO=+T^d#u zYCAJ!%?)FT2C-WX-e(P38Lwkhg$1-{<7Mdz%gmM?5VsXRE5ropJ~bZsEUpIUrkM+wIi0QVNj-i?}8}cD0QIXa}jA5J*5IzJzHUFZ{@jp^>_+M-E{(nLMLyu4C z2EWt2#;9)+l$73)pn@WKj>^}f2jBZc7yv8ye=*nom-pqS8HJ4ytFp=c1~I_x1klBL zVV~|bY#_WN5yfMoDjL%1rRz-zpzN*#1hBu+F#&tz|Hw+G9DNwfQlGBiqO-^O2xoN0 z+@%}P-X$OTSiD1>vcJ4jcuW4OihV0b$MM6Tl(PT!6D=KKV$>6Iu+MiG%C@B32%IMr z|BWe0xfe5tX|5-lT-zX-tn*xyz`CqY!SCs3^-Tl9njO>afBs3@S$LRv-zy*3zM!X@ z1i1lH!`|-CK}lnU#Q@Tf#el-!8D-Kjod?DQ#BkeptI|pP-@onuk_TyNFDgdmLtXa_ zbVt=xX!||bQ|mAP;r}>?bjbS%DFBqbaMY?K>qK#iZ`xozW5t70U4*gM*A&1*wWUBH zwH)zKBa$aCKQ#;xKNF59fL!K*5Vl<5p-!Z^-pA!-Ws&69aHBzzI2!r?uaVClovzM0 zS7ir?g5Lr?ewb_Da9{{H!~W2|ZnGX5`qob3M-VIL#^j$F?nSvWUg4^H%2rTwdTlE2Y~|M|)3 zQ&pClphpOjBYDPT+*@;TL+Kmz)C3Txc-V6(fDl0V%)4AUH2oSMrAFTw$z!qx@YWxQ z{>saID7-JN1~RnVDqF-wUJekM1^-Kv2$Ww34|=i<8<5Bi8ofh{)Pi=y?S6|Df$~cY zmp_)U4Y+2qQ9Pnq*F*nt<9`hRSDF6FcT-J~2BSq6Ij7i1hZkj7U}5QsNcI%Z4S1J# z4j~yO;sG)Mr9Mymv5M~+sE?h+d+ni-X-|+-BVfpGOS_iW3o_1+EOdtwz(0vj+6Jz_ z?plikm=JpHO`@>Gh0xX^ow@DF#J=?fg^>#S+afkJin z#!Nyn`?(~4J()jz%Xzqmw8ab>^}}(aEJ2TUbAEyJ9LE2dLI1q7MO5EokzjEndaY1q z6bd(1^iWzB!h@h#LwJD)4=e?pj?PyaZ`Iwi3d5pFw|;j#sqH>gv@>fo_s$$kBJt0qB80|Fte~o(>AyXNvZV!jH5C`R#f?%Y&utZm zhb!-D3s2*)mix9`6hOcJ)6@Nb#K$c2IapM%J22|wNP;+|11L-dm&rErw%xkN1s<`; z$)bMGJw9WlicH7oO@L>uv-Va5QBZ#-#uj-uQ4RyG2>}za%if2~97B1YOG`MDzWq z7hi`I0wDfB#kc;P(|`RQA78*3!jpzNVl;oAWFnWKY~RT&>T^qCoUIs`kzP=Oj2?#h zVQj6Ns^8|O|wI;8Vtwrd()2<<9Q`jY~!uS@qXL$?Wq0I}#v z*3YtdUpVUjH!-oIQd3*PuaPM9VjDlmJ$jr?9~iVhRV55s674F{YQ7iqpws8#BQJd< zHY;tZb6#4I6Vfip7UD__@}#S4ngo5YU%ro&M5>OrPN~B$E92!9`~1e9J^u}ChGWJk z>jxaNts-Ugjy?#FQ0kAWy>o^nZsU6-{NTdsm;A#Bjx=XqN%pfgzLpZWs{I9Om7~-B zJqUV|r!IL7e!Oe92ZiA3>i;zDLhk*?2l3Yf0bT*=gklu)-uPW$K>s$~rOsf9un?;M zippfaQ2#6Dt+t}btOq57m1W2y!{RCUz`CPK$B^S4K!Tm)Z8^+VzoCQ5$7-Q#jYrVL zS>_F;d3PQ6Tzf%aUC&)&K$>ODVKD$>x0g7OaWX1Qfj1ufs!)igu`y-VKkir~y_+v8 z&AQM1D|h?Abc0}{vzrU{06($^A8+CL;W5~i-0Y(OiJ-own>2u$X$=4qr-&&kX2JBPb! z6#zQJTn%6@1Fq!&&TZWa1=yNG*X?j`&Uv*9V%uq`5e+HJRZ@ z(lcx8dUc-S1N+wO1C5E!bxKFz^fe0viS11_Uo$^(!>QUcq7CeG!x_lAeY0qS&?TC6 zEAIIu;>b7KnSWLMYF{A~Fde_Knxpb+h;3c#wLiHW<-^_GIakcnskc6umPeqzl`Cj6 z#B6SfBp59KN1Redb}TnN`b;#L1>5Qs-6fN?jsC`lkY78ZIRC4!I)&5eNL&N1c>dJJ zdS@LXLV4LXOLb-1)^}~UiFkk8z`pWCb25HZpnE2T&nbn+W9Zh))Y^}0GdY2?Y@fH{ulVv|suH;gf|LB71x6^mNsZ8@%-l#G?RJMcsj`Znu|0_a!Fwa?Y;@-{cx4%6609 zi5ZPdz^KUN1xV#lB;I@h<<99|J|K((=A|c2!b-VhkwG$l+U-aER!EW%*(^`G_~}BJ zWRNVm(sJx!-vUoU*{gqj5T`M&GDj`r!%WLF_X4BOIh!KO|V?;M_7v>(p+_VR)p4bSHxGSL+Sd5Lz7te@Xt5ujkWa#ejDlf7SzK|`^&g3i%;yw zoDgU4a5z}_eYX%?czlPK{`luls5?9ha(GLKg2L^WcL+669kargKoWgffUe~~#EdY_ z?=_qwZ`>M7;bdca)wr#_TQ>k*`63{SJx!h?T_C^sf;(w{cCmk%5B|ojBg{r53ak(A zr3}B9CFn6np#zt^d!p8iWR~1*#;lrjBUE`F{XH!2B=utGMhDSqZyK8H=ilCTVfkDY zRDy@6R?{A&bS(ajmQKucnEKn%MR|Lbxzd{9)YQ>Z=S+)1qWJ!%peYiMllGVbbk?)a zc@gR&`&stvR^{M7y!n9t96aKl+kH2+2SLDzS290Y>T5Z<_0?@LsIaX%<3xO|lo4t? z_u5uLPIhY!`h^nQgU?8HOVZ^xU0+aaH)6(VuM>_XL}#2-Etxa#oOvEqZ8itl#+a*d z79BI6kzBu-QZwY5x1cChWJBIKs>H+d3=Y)szR*eL)O{+4Q!1$-U!o2wv9?$82C_#$ zyM$Z)&Xv6A^3kvsfi&I0eii@2MY%VyWtta#K))|RYhlrWF6JWLJA)-mTxe4*Iy1XF zaqLPd>f707&VVELWS3@zIe5UGul3tfz9^>YwkFV$qPwkD`o-hZ$@9yy8Kvu=qJ=jr ze8Y}6`_Uq`vJacM#TUd^%a-h5+WD177w0CQ2?N8R+!>$n?%4NSWmE%3_04s&knnH% zCnu3&e)3AAl4L+O>9p;WgQg-kjmL5CXDWz<16UQN18560*%A=^lA8zupN{#8em7DE@9K8Q{2KMkvsPk*ahg0tXYEf#(ed4C0yrQRBpx(Em-{m!);b%{ zoP0HCZVFGiFK7apTdgC*Fn>ykH@X%3rTHaL>~0I)MYq%kI5j$2f=`B9!$L8qL_NTc zoREHg)m(gF461fR7rkMBF!J&^!^bTaXp^7mk5-q6W;xMuEI*9CEpj4~wK{f65O?;a zvwup4JC1eKhJ#JTZMfY)d_71=?lvH_>$oZ_3wT|2^aZvT*rI@tgA$SVDTXhsK@<{L zz7oZ{|I`!8yQZ}Z_0Wxewdo!iyVN)B8Sn{Z;~4i)jD2epsSy^!3cs42nOeH}${t5N zM%OA{GJOJNHOdB_#;Gp+C7XPk_^jvt>-0mWws$ zyG%P5UN*`ma#?rUCM}FrI8k#=zrvEC_4;PE9V`pfa1R$zbQw7y;ZmR33 z%QAV(c+pJVC{1>HSQSZ#omnXZIj`wWr@-uo$d!)y20vjVH;qrcH zjr!0}eHE95*8YsTpc`w4MRUdujWf3-OoLz|F?y5+kGlF))~bsO9UY(6+P&|4MN}I? zSkr4b30A`{lxb(1Jg^L#V;_GjY90jPw#F*H$Yykc;EKo6-4^=&_xW$aSpwJE~xDN z*uKJ9?JXC-v zc_%7elhsJVaoCi0RF1;t?_#x!q&`_AnE+Rcj)aHii23Hn7@~w(lE`wed~gq*^x|b# zRNlhHOJt!KID$(`^lo@#{#T-<0yYX9mqCnPb0W`qMh+k8^-yJc_b$#@C=t+2haQpt z0s&rt>Q6~Mt`5$Ce%d+bvlsN-d$7lV(lLyk_LE`4EJK|9qM z{h>p4em8Eg&e_0l<~g=#9sLA(Vi%Q7Sgra>{oUsG%mn1g+d#6s%644e;z^5x{!@y( zc(oLuMSfFt8`zMx>6Q6?K^kINm?NHs+HTb1WAvyz?&*44>HtF~h^IRC7!5A2kE!Q{ z@Pa-EOY@KcZs_|brO?@rfE_xUiWeWJPT0&IwBh5+gHkqp<{yzeyQ#CCYv<1Z$N06= zO3y=__Z1GD8gR1P!0`Q!QyG_8)o(i{j=wFh$(T(vZf~FI_`W~Ug!uNHxDAU1=d75D zr7@>$?$HZn9`}MQ0QX9hpTU0o`GP|mHP&p5($Zx?LWSz2*FwbrLrb$X>C(no?-Mk~ z@G;T@kifpaeN^V?$MGhDt@*X}j0oPi)oKWpw5(}U+*nh);8Q>HvY2wL!9&T@Zx)G? z6@D{&5K9RLpE9>EKTNvnmzSSml7y~fbghE*2*U#l&+Xn^8m4Rj{=k{RNg%LFsbiu# zyMeiNBJNmqQVm!qvqq-3v4mtnJaSl( zpBc``^=wd{i#nTXQIFE^k7K>xuI0F2QBuZ?T*4|IMl}bI$7UqghRJj$18CNmQW9e$ zNF!>W!ue!FT>KdC;)=2!FdLkmZE2pY+!USo5HOzA_^ge>AB!zyF0Fvp>3I}ScK6=SpB)uQIxAH@m zO4O~Z<{!JPwf{)Fwi%PM!Q#!lt``)1%PZ;;M?zlARMDY0cW`@4WGnycguY2Ns0X={ zia;Nrvpdz$A=cutgelsz?$^i>Tw~6%QhZUh$eH8i>%4FnL-E2v?)x^~&5>f+%t+za zKc?0Y61w_m^JPNrDD#mQ~8h&2}e>?l>017Vl@hLI}8!NKd7rkIk zUn_1G`s%f-#o5?;t`;Vy)J4FCnQSgaazuSlKAI=m`3E_eDUC8?Rp1oQ zBna?o=E6nf`M6WRAy{zHH|J+-?a)}5(gr(9gtwSpR)0NDqKGk&g2W(~(#oLyC7nlu z%$xA>c_=cI`Dg?hpG&+*dbN5$%i4y_a@iC`2I_CTq$9h4?m&;0&)`5Pc;!P?@8fF` zdNt%!aB1<;Ms~8a0v++b1cP1}J5F`c7&m;8v=3Nu7v1fc5!lN+R|6|JVQ`T>%z;;w zkB|G7kou^#SOb_+Jk<|+cRV15%SJjEUrsIg9W3eT#WCDRYB?6yBAVbaiL0L%ZhPFI z={l@1r`UINLgr>c69D%TqKymg$J_M}-0(*(!o4W_ef>pK6kFa&)b|(cYjNR>%am79 zHHSg#z<6)Fqw68tsdlRNfB(j>HuiD`gAcixS+;+?L+nL9h`yKFyHYM_D-@QQN5omCt@v>ch92l!yhD zg1A-IJEq`%4PS~yC**Roa{THDtakmq+T-RsgV^h(3N9hG8%)c|AW!4XJn1C29>?&+ zm19DesBd~^NHaKYKtN|h6Pu?KNJqMzNAtP5=*sOeEf&Axbw<*la-w5EV?_ZW1~Z7A zK*lW&_4m0We}TXhI6j};H2Kp}yhR*}r)vxm>+U%YDe?`?5i_)f&Gu}*DNa@v0(dlM zo&~z83qSqUq9McYfEn_!9IZmA21i~TXs&bNDb-X%W?ZUe#++8sbHr)aZg7s1PPdvA z?PAS!n&IFGM8sTpw;OW*f(sAGU!A>><#^ZE5)o=eZd`^7x~^5l3>7 zPGBFP%nBTIa*0iSv{E!03|`T%yM<&&W&Z*d>Hz*#p)wN$-?F&2hPZKaUKqEeqOx&Z zOX6?q;BLhRSk_eiha;LCz0X-F>hK$UIQ2aUAMkuQuK0jDI$b+|(zvAi!Agn3nQ3`l z&(zyU;6j>UbFovn6$cY0n`IjHL}kq=e~RCT#)_QN0aRPGWU?b#j~))pC{c+?Q&yOZ z?3(;4V6NRBE7_+MGy9U@{1VE&JW=tr@{JfQSVT+MDkK~c=uX;U2hJA~#qyeVY?uu+e?lm*{s$Q3~ICPgM!SG$BaYFI2 zKr#ozYvq;T${mxSgzKl|*Ivd7Ycihu>n{}pEo@Bn@08^1Mp-Zu&;=Ih1T@$l$dP|B zXcW64UAaxdSO1k)@)s!cxhGF~)O(+;W+=a9d3Mia?s(@qM;UbX#RWQ!LGU+JaA5I;D)P?qe%3&#V;Spu&)`y*TvDU*bO4pIDlPLfKc>kN%FOWIO zML1o`MKqny1hDUDzVvd-z6*yNuyV${Gy4>>yrSpoa>R_CfY|&bqozGCznYN_SXbh4w+gPJV3pPC? z%>}@xoRP9qXy~lhl@db4{FGo3`iF9|9d69XMdBBdlE>bUqQw|v?6&KqT7uf@80wWi zP$2+;Pxt9lJa1XGMj)w%o%_G7D2Je^(YGFmKourS4G?j(R1rd7lR2 zTFTY|+E)B_T=i~aseQirEwHDWcpa9DbsjtfsztM+SLz!3<`-d_=eu%@B9L9wh$CQq z)av8~Ot~doc*@qkI?bgclnQpgHE4(BTiUK<4DIS(9c+fAJCwC#v>=d7~57y3a7zWFS!UqkGAD#giRbI;9n6fft|Kt z3s`%hMDVx{;EheUM^kNcuHJyH5u&knnNBI(?hYt#wdc19GhF9%a&Basr7UpIpI6%F z*f?CTC*w2yI7Ftwg&X93d@u?4`;i9ry{ONH9E9eK=C#6Sc|Qjnwf~?KYA!(Ie)j5B ztAzH0|EoHRp6Xm@dU>Vv}e<1V-lu59JBU%B}3*YrF{GNn3~E$oxQHD{HnJ8AdCwye+EvA zrQ!jz067pk?ThqIY5mOCf^(gvpuG36cPTh;+{j!VqKscSmK&O@U6ral4j={!6O%3U z86|v`>6=RQ6t=sExx`R1vmy7Q_7Th;8awbcm#EamE69z$X-{UrHIYq^mYr1CUsnR( zS?rCAW!yK})PfydT{@=JlotOMGfn#YcN3U#{0;$?zxMFOBShZi`%=NAx^IzkLh+K! zgDn!9&WEi$K{~>-i9_x*9NKeC*O7&uQ#hgJJJi}^q#4B!?42J?hk8e4K2$ABSvT4u zzc2Vf_~-@jNt+5=tP%UtZ78?I-ddIIY~L&MELVp+z0TGTC*vvZ0&$cGG@86@FqybI z32|*!V<0}nr^GJR1q#`FI$O3-^Rf~zHzS#{92Ml>t&USe>67Pz$W+a5Li?FyPcnd- zj-rY3TW9{@WdO}<=cFch9RP6Z`>y-Xx|(#p^=(F_9#Z3-M&6ZG`_F#K3JL?_*8!1$v-x!gWD_MX~p|An*?t_=eh*9G0J)KTmUV#wh>k^h@{<0 z|Kh~wQmNfB;WqoIPZ-q4!`W-`3uG0<=;-g)mZ99F-D+hWa?~{fdcjfD8aTo?f}EQb ze>#qbY^LOY@8>PRIa!^m`W-a*8N%(3*{4+P9{gc|0Aq&c0na&hWjhdqt^}`^{3?Bk zb7~8_;lQIur!U85Lip!^=Eex1v~7YqM;l;QnOO$pVNyBmAbmu3mKBIteg#xPYmQa z4xnO+Ggz}!jJ?>rv!!!hl2t9G$MDWfO84m3j8q0|o0~Dk#K2(ewkw}=s=uJ>9j;pwKc+NW-$B!W5lv#=r?`qQ z!f^KPp9SzWzOA-MiAd$)YyyY|aFc!ZiQW%`anS?e+B+Vy$)E@3?J!1mQ=^W_Ubmic z)K{#RO0n8`dCXhC0}3}B23%i&Yf<^>`%jzeR_Fa}gx@J=s~4D6K^rv5hj$DGq5b*L-6-e-IcfuK*mBQqiR?Fa z@>gRXzGZIny9WA0{w)!VQxdt__1Qq;Iz-(NC--_O{;m*cetSgss@u``;*x29#l%IA z6S(S~T#|FD)&O0x{@YH0T@dC%W31sB;#<%G%9q+LO;HB5VPZTC8I1yv$XBMnVTu5l$h5O;0zl&R z`v6LIp<1{OCyg^_l@(OBGL{K7mkWpZml9KAHM{;+gj(QbS(}>$TzEi+u6yzebRca1 z3$)jHAW~Qf{Q14DtLAjjE1h=e*LxiZN!kS3r`SwAQ`NmqW-?{3JIP5OY~4=!9B$^E zQYc@#Yi`~baXqEX2MSSc01XrPUL9$A%ZRhHJX~EsfjUUNSEI@HyV{)(R z-n76<@(}y8Ji=n*FyxgIP@p}zbINxwX^5%4qp%mKoMj(vscBpS!>9Un`luP zyB5nW?SH&C8gSPk_&nYq*Cmw_{OLiT>*21O(!Ovl(<>ID470zF-8eq=#IC<5p^jLX zX{rFzI5q7T?ibf;OS=RLTFyoHTf~#Ba4?+9+Xfc^E|kSF?nL0_*)@VNwPCZDCTC{Trtc-~AV2D)oPal=|PhrbPMWp%*oe zOBS)4(#C9!vc|9;hf0&ZxY}OmE;B7=HjM1Blxpd&3!kA& zG}kkx&&0ZiS`RV+Qi{#AiXmq<7p9EI{i57va1f&z0!CL&U)fGgw{YFkmp+bi4B~2D z_&zMYO#87CFuKc;p|ACgR6US{5nEZ_bRYMS(67WMw_jXa+VCh&*dZ$^2W$+cmxhcM0IoX%CXs8tHiJ(G$Jpy6At%xEFJgPx3LTdW!RDCD^bL>pcKPG-BrQVm z4tj*R(N?zI><6c1femmUO|+~nmsESJ_gb`PA2;B8`+s|{cvL?g!&{rr4gKz>ggmpQ zMXo&Gq0>VYS&9+7N%9cQ^mhkj>?lBbY8dfqtl6aHW0?ny)%eWY`d@Obw7P5L;yHQf z!f6ljD3cxY(DSC3W%KLuV|zaiJcLkycPK7VUWq*do%EO|B=~tU&qrp}PDQ*25EPrb zbi-!CBF!zMw~McS>Xm~%8yeN@4)lc@Y%2}(Ln&BME0I^GifVrExpc%ap`k!mgEX%I z?Z8{7(#&p#m)E0TNr4mYoNL@8u`t4(a+`Be7$JA2w>u(3zZJg0OH9h>y;gFxT0J30 z;zb%xrbJT)qY|>88FIwoFc4pi{&9#~H6hmxbhUZF??$FHIogWLpLlN~zT_&IMkLr5 zKTx6CN;qrHgsV~I5Q|Oj5P^qN3D{!VEZmWBU@~$!&;?2)H#KI~l~^qQUU(ulzIU77 zcm3!PoH9*~PXcy4Er5<{zt2dtqDuM9kf6&u%lS z;NCUmync~lMUd6{vajZ)CP=_8J5(r+S4LVZiAqe8}>TCDKJ1|Qg%wZ=FOh*p8Ln|NtSuJ!eYKP}qQPj9+rg1Xl5m@;*#&qRuHS8W#=%Da3=Nr1^4 zEopjfX3481dA%t)cspr}SrygV5O#bG(PUBp&E_5f)US%|Yqv`#knFLQ(S(osq{S&N zB{knXYdn8Q5V<8Q`e1CFs8bQ7^Y=%CO7d>8e90vn%mVPP zqf-xC)m3ygfMl(zO|XpI*JIksh~FBvWxHrZ53JcQT)B#B7@mNiB<+J6n$Nrlr}Uo% zGa&H&ey9rzEGj#dTmUHGko$&)J;mUghBqm7G)I2Vr+SX>*z#v6uo@o%NR?Qf@O~P| z19wn^a`v1X1e1bckRj$uD*r%H9Zhz{cUdP0_m9V2ze8r8PjOEG?;TY`dY`3M_vroE zIEPN)G9!Q)75?Xpa>TnL&oJ>b3%t_2__25ZoJ+RtauvmD8y(q-a}QQ|>Tk{k=qSH% zfhlOd=wN9TTtu!!wgO(fjw=~@r^G)}p8*Stqb-8lKLhqb}e^x{P0EZXo;$mPT1po-$yQfxg5laY^aCk1cvS zgM<~OX#|0~V#4UpYm@MjTwC!a+PXyH`T^e9xyv2;YA7w4;i z8b7c%78|G9ZBC4??&r(_I@sCZ4)bD-K@&KWz96B)HK-GrDPAT87>&dmk_QC@lvi_ zBC7O2(dU{E72h>EV>2z7T?i}90nGNNw(-iUDw$b^MDU=0gEDQA@(PA#J<@rSmNn@7 z9e(XvHKRB-aw$KQ(v|mfXQSvW9#@cL5I3WwrnBXWv+ zin_<;49kU}8H^gzphI9geL%D;*O2|v>*jzfo0kxSO@*P|ae6?*K#K}EY|mrvccu~t z&BvIl^7}|Vz=_$@N^e42j=W8KfINJP;zdKXlV`7u{nbaaRWb!n0smMh=2r+c~}b0E*FEa$s@Sa}8vnh!!0{hHaf z*`!)~?iJSP=1MChJlk(G#$RdV0QklUIdU3H!OP6tgJUcvRtQB?LH1M0SvNV%Z#AYR zzY(a^sB+;JCMP3uK#g$I@{6(3GU{~7%Ed8s&j4{7E<}7(mIw`8Qy#Rqr=%}I!|1WN zdm~=LPrWlPT8i(LpU7M;cq4V!#-nxh^%LeW93P$}M_brdZ$|pS){qOA9_P# zdSaRw0B6#L$0>p;E~htHlf`$#gnHE#L3;3p$jNCZ6dxT)rzoj%JCLGv3l!hDp`Q%l zyeGIUa8Q*P2^y_O@m1KS`wA1!4emEMZ-)>_pfn>!{kGoLIz}wa_Ys<9222Cr_qxSX zc)7Qv)>W92I~U?Ys$AU8pI6*d3cgJD7;Glo=#mPi^p6C|2?`mS{UMVAn2!9`NjXxG zZW9xt;neyv*AJY0$B?<0F?^36Uc}}TomJ89Y-t|L9!)qZ@4z_)uavVC5aQ5Ny#5MD z33N=c0~+X@s1zZ*BlBg$46o14f&O2&p#@->n%aL~TmUch|2>f{yQI*1fD0uv9(bVX z$_@iSS^ zTS17UPcbemMELNM6{TU5$T!Az=VR37GkhDBQcey$nT`Nn9?<6_%vm$AD8%&PPG zAv_Bh1wTBn3IDo#7wh}QDze3@*GybJRo+6QlIMW zO4E&YZ&W$g6~^CNUg<(20)U+LdftCZC?l_FnF7+0^uUevsi&^Fh#w!9$I+=qJOWw2 z2qSw2n97s70x5tVheft;flEmxKqt=N;c=G1adl`lIL&zuJ zsSvIms#`ny0Pi-lu1fyYjf^?m6c0qEYRpCZ$B}jH_=8WF3zodkL1+bY>kc2AL{oe- zOFjAco!I1dMInmn3O+bZU{rGd&h^hgz*SpmwbV7yeE6mKwf8BM`%B3O0{TUnrL*z* zfn*J##%M^9q+XQo?HT;WxV^qSr62+z2vMqDbu%i>p+}8$gq6L0fy_gTAlK zJHV388+cYPejc+uxVe@eg3v z7905&AW~jy=%^SO`lu!Wh8o`o4aATfGww?YoX5^GwAC*>82@Nqew zwBmhbk2J0Yjzu}o!qg?t1z`ta;bZhQvc%qpADU2nd!9$A%$k_TC2X{({@zGe)ye!& zV)F3lClWZ3rNwg!MVQ}NDE^ZU4H_4ZsW>R*p}e7ps8UypQaTU4XiZMnN{DHnod{wnqS+bcKi z9t%S5gp^Ik?Z#S;DkvSm(fK(-Xkgo)95g0!%)5*2*`5wpb+OQnIced2sc4mkXm+OM z!h?njWIaPek0>1gJ-{PkSQVo#R{SVevFx|2Ipy`HkWuGXaY9O}CwSY>0bbQ_%NFI$ z(Un_1Qbh5$^}V`RuGpyN-Z>zP3HwEy?GJOgeB2%B6LyIB1mjjopWj$(>< zoClO1Hwv1QTLiK+pI8hQ)-!jPy@FC4zdfKY#ttD)(D54Q*$D$~~b>!^>SoJ`m^cS0HQx zZkPYt)r&JK@D{tS!y2KS^hlnLI46bSoT%2ivgf7y@uSrM``Y+Kz{bERn<`TFu(?L; zv=EA;NsdKl&QiuT#ny%5F{hvTyV zUJhfUoV~tR_sTJjwJL6iMz;Y+w#zX=3uX%kSR$44OYLkkGtuiEJqgHlp4{eoL@N>h zEzzFl0E;UM(9rI=vjmyU`e#p&7w0$;l8NmJ_pEol{2fnkGw#@W-~^62en%k#O<_8U zW5M;iGGtCRZfTiusa!|Ae|Bo6)459>2VnI9TK^)_(dHy|>1}J%o^(0O&u~-9O1{EO zXM#xL@cz*UJKW`p?yf5aVcXvQZnyD*xn-KPzsmdezm(m@< zW<=S!X9k>1ZW z*yymCsg0oAX~6n1vJwFLcPrLDLCtG25dQ2hKZ7sEeBg3r`w=C|rkkOjQSXLqB`6I z*~OS^Di!h+?mrt(EJ*fU#E*LQ?Dvpw_pHJ&SOL0l?I!+-(EbsSlYDib%~JuE#HoSq zG{S2wL~CB6Ga@`n@E}PSHwC{)H|5h_FYgZ9dCMOW%$9phKQ}|}?;K#{iwF1?%$yvf zi&qKMk24${DO&;+i2%wQv77isx#T|xT6#@~PBlD%E5M`e45Wntm80eF2`f@R{!e>f z9T(NQ{yTs$ij+#HN=gYL-69|j0z-#%2ty1lErNiQ0s_(zm0QUiVF{k93(hh2 zZ>bU4=EU(u;|Z_iMm@LXL|jC4WG%TWM+>>qEn#NT~UX(0)*cZx6B1F`(I~zsKAxejWzQ~2WJ-%+8gnC z?}+%W&QtII*!L;Bs)cs=$Gj z+*$7E#=Iya9k|IAwUtk|87DQ^c!Q}+*@b}CbXUAs{*&E%1s@oIlF&EHAJ_8Z3g>fx z3gGRqY3u`SF1Qpm3VrB)7{R#qzWTWJQ<27T?gfg&|x}H8$a21;p3ji{qs1pc| zyMr%(J9o3P`tvgWme83mdvZH%8n?^LvsgS61PGq$$hJ6yxD@b2z3h+?*aiM#8E{g% zM^*-SEm>74f7Bqk?kj{)JwT~lZtEYrEXf$C`8^hG3mIhV@{p2Oo`GdtmE4ZIgQvT( zYy&>4yxPIxKDu^woE&g%Wb}Urxjntk7I;IcH89{(M1Ul_n52W(Q^$}m5)U4o4!T^X z>igw^AJP$fCi9eD7s6_59@9}?&?6ukxN1b`BEb;OSGcB0BKT4<+ElOk6JU5lW8{7Z zA)7jU*MY-#@aD`p$Ac5hq-yXc1?$xTtl@NN-82qaOrJqiob2s4;sC`J*tn}6enwI1 zZ`~Dt(9sYAvRoKF=6vAShUFFu}ilJgDO|iT2SEEiksH5V#R+ zb}Qlux^E^1v_F?9yd8S}9#e_sBJS(R^pcBf;f=)9M)so0_)P5TSNIxrQ&VsW&9CV{ zJ^QwNuE8ZrJbuZSR7HRyR8`Xt&}#=fN5|H^O02-GVbHp-8~rSo<4oMh!f8q;7=^H- zekbDmTv;gT&^p7!b<+@0C6Rw`Qhf3KWY(7kKhxgyPgS=oalzFMpyZfV-57_H4~+OQPUY=#E# zUdB~&iUB);)}`sKG#>9DQHaB61F0b1&?ptRrCK8+Eq-TV=R$5eq}D8Evu-zMJZ3yXD{NLn39gh^k0d{)VIEP_UFg(FY>6suyl zV*Rg8iLr}k2${mq={>o6WBv=wxd2K8x>ku*V+(WSIy=s?e`d{oMgsQklwg#%I;ATa zyH<)j=y}HOfBn(U3i9c^sLnQ%HVsFw{?$iW!79gOc$Z`4JlBy;Ajfe#a#W;*TltKW z^i1%IJ=YoFME~TGu>uf3*Sxr*&R}FjvA8E|HG(?i)7DXLV46Uh7vhrC(9l7NsX~S5 zlzv8uiThZC`pQS-tM_ByCWW$10S4URGQ!d1Ktr7agSvZAm^(JiM`NIexGfWWn>aCA zO+U1~c^4mW_Ki5V&!`R;;+hY?iOn9syQ=&UA~=eAjy*TN=uP4sykf^CsiCyTBg?T+ zCzJNQ#DN(?=%Zd88nz7&9H5u%NS|$P^1VQeN>PuAgc}$eOx%leoZ&=HPxj8K^~Ad6aWHzjo)YnyrH^9Pyq~@&DJofbJx*A84#qs7 z&Odj@z?H|c72H&~Xf2grb)N4jKP;W&k2rddQjMXHBWppt5N1r^Pq>HA)5Vy_UY%8R z;@eu*wVQIRBS!>Zn8N7JLp{6zruVU-@7^C?@ zj*{Iz9m##5;B28mFw z+;1(;HQ=Mg!nej8lPR^8%dpZx1X@V$`G#jB{*)5hIHOpc<`l1*{p#Vp zBkeU4^QqDuGjg-hY8?T*^UbjbtBz@(<)2Vui#_W`zSA*rD>Ts6XYTD{qae@wA?KjWvqdlqt_b~FamKr)EZDxg(t z8_a>s4uwF2MBCRDu>+EP3Q6uBba~i*g??-=#3krmNjAXyj;=ft3b4?7RsB z(-mN00N_&0v*v?|<`}pypSp_bRfgRR9PYDsUF`yZA3z+wUjm>l2;vg}elu4roZmptSQmE{TG?nvCO6X{ z2WGsFRgTw8m2|T)NJ*0w--4EUh2z!mr0y`BdSrjI?jzySosSI5WFx5Kbjyf7O5ccD zpZR1^aZxvvD4jD=ce}ztHeh_VO!TVkxPXxgyU9jMNwd0AR_)e_?=A z=;=qJC92y_Ww_sk9dH`}FGOEiP7!Wn*h4}o4^DKXf3a0;^0j@gie~>N)N>-OBJYI*#Gc$!Vh5P*JZisfm zkCjy~l#(>RUGBurDQ~W*uS{NQ7Z6#f-vn)SKGPA+n6l&1bF3=g`&7=NG6QT54hnGc zkOw|KM>YbLe5|@?)494$bB$Uf{b8=gj0YP4;w>#zVpFX`yM6=?HSX!!#pv7`UQ}Bg zDp_aPr5ySJnuY=%chw%muArnj8wCt_CQ=iTZ*T9O#AvHHzo44`&Kbg%!xuWh&k@Kp znBrxqAjm1g;=_;&J9&}ske5|C!FhI@ty@yDdwWc28ekcj74^^6+;ClZfKS|WW*e6J z>_OOC&FB!Sc$==X$A|T5ils=K6k?G z=9GgjQ_d5G6^`)cp?T!i{t@?3FXy~Z%`-B@VI7p@2dHx?35dVWJ$({tk)wo`& zPw&`IZ(y)9koz2$pAK^0(qpphsUUcluBO(Sv19f+*8-+Tnx3z?dM$qV-ag~)0O40O zZZGm(vn}4f@rt!B6PiEBORNetsdIRgRs8US#%s#SD5?G?iM#3TE}N^gAcmoOs=4Y| zLBp(j`I17H2z+gD;*oddQnR`xhGZ05r^H>4q~J{~A2)B=IufnF>|3RoMy_y>7JkpG>3keJfpx(;1pWaoiOQ<_P%W_Jq-^cSfKR{tD~7Sp(5r~hVKZ{hw_JJ<0e&u zuDlp~2#QwwTne<^?RHx+W9lM#3767L+sbc5&e!Hl!q;47m-ELnWa@%G=#@`KjUY9Y z3B*lQ(iZ2M%1tkL=^o;j_luyebE0p1>-9MaJnum%FA~~10tk%Grpmh4ay~p?VeP!H z17Dxu*W}AGk(0=7NtZHOSuNsXP^Z%Wp85qu6jQs$jaD>mqCfb43C0B&>ltKKQna7C=V;E$N^W<%2rAUx=*&R8B>;)cC4@^^ ztS3}_wRM@C((B3xJNuAXZ*wL4-Zy zeNilOgFVJc8ibvmDxDU7b9y$F@7ZzXTi`hUYC%>2Pj*F}35mip%_A1N!1p3M9U?>a z?mB`FHV#OScpj?-0;~i0+MPfd%@<0qBs_0Xc|0J66~(N6a1Tp%7#M$6mY?~NPQ)8B zLK?7W3jA%72tKy5~tXNz2aE%NbveD&XI1l(BU6H7i5q9bV@!zv+Z^ z(5`|&^?T(hL7z6HR?h%>g6o0$`;W0{ENnLCfi-COXJ~6i;F&BZGI# zaH;?_0+`YIydkHRld;m^`2%+$(l!$5QJpWW$5X zM5^m5!+ljO**4f4(iKck!dlqnHQFhZ5Nq9stTp7Z*Tp)~G=bx0^`F^rz&gmB*K()0 z;f@?-Y+bb6;?9t}`-26xS-pK7Sc>U#*i{8N;Jb9wFK?+;G$ntTF?~b}^1T=yNOcnK zhQRXa1Cx6Y;rXmn{GqJ3xK7?KcGuP@j=VjulcmPbvC_Oyv)K*L(~I7IWV9c33Bm}Hy+-! zs6M?nc}Fu&_xd>I+G=toCwKEK?I~XWw3mX<1^5f8fUbV(Xi^)J*zTw5?xvMJ`n1nB}TB9 zB^sASZ^TY=b*aTOqrr4F0I0ek50B+lUa!S_v{ChGL~vc#ia4${e_ma%VkP`mfUX^mJWYNQ6hlOUIx zlCO_#htbl4F5irF1E7zxx89XikKCV9^xBx!o2@tPjKRnF>(>=dpwV^$H(FET$eH`3 zTISD0A)BOaIt_Ahqq+l@Nt9SyvR1aqK8Ubn-?dF7Y9#XZi1K?QJ7o(0tN6R*)i!xphD(`GdU_9Ny z>N5ERUnhs(XyF=^d*FDffhf2PyE?mUuhMh>G_&%U{ndsPL)R#rv}KWNx{VrU_GOZ5 zC<7Hmd6%wYAFjepr}C`(fR!dklGFZ39nu7VKcSW`$>+W6EVttLwVy_1Lmit6>ziJo z6Uk3r1?VJ9OENmwQcnZ$K#u;+&Bp+@y@IQLKKS zF3=IBjn6-0sxCvMm32HIEWX2t<>S1$U#w4_>XzT{br={Q8we2s(9upJI`RM_&L+V_ zz-oQ-O3jzI?{|shXSKdgY9XtHw>EC+&DDKmyh$QI2IP&4M%W{3I?K!{nPMwxOz73h zSQn9M9^SF&RRdI0;BX23$YqESyP9%1GV5)Wm1ya;dsm_5InYS2+!ur20v~_(Ze3Rk z_uP7MRUuxOHrn_Cc*xTzY3XWiwPVE6aW4Vq=1$uT4p@xt4OO8d$Zde99Uz;oHNd} zOv3?mgs^qp4#75m;d^a&8s(6VubzDz=8EN*zVaxf`&^S+9h~}3AZ@|W!>(6lF;Y(^ zzt$D7?vN$>CL3tLZ1u@^utKtJktnYt!IWp~dFang6r~Js_3$ysAQP-C z%n;S!s11nN+O4zY;SnaU3sMBF9gz{oV z9g;A0IHSa^!X_E@nY>gBEIQT?0OX?>Gvbo=*M$7iKk`3jL1mMYf$Y`uKif5 zB)!H~CyD`c{z96Pu=lKTpy*75d7B_KA7y|S#S#4O2K#c1NMIU=n$}3x5FDaIlluK3 zt@RI32^)@a_?(W>*b@ym4=ag2?->%QGe8=}#s<|pnrZFqd@sYsZ)-qmHC>+*tRm}_ z@Re!wu6n1fT`8MbJlKY=;o!_(>Cu>VU4w{wCQ0}cBrR++cvCiD%?4quJ;qwyK<%DR zoGMJaoG~xda#&Q2I?RN6&><>7VM?uKl$~pG02kPuR6|d{x|k^5kvYcEef9=W8!x#- zFp^d4itv2FbZR(izu!q0pVg1 z#tBTrb**oV?VdBZL4(K8K_ka0`of^^7OyyzNhNyQ*{j-iojZd z>4;~pI}|1c&0cct942}WTb(2u7{Va}8fE23gha0sgA?d1!#usCSbnaFdHfD{+D4;a zQc0@xbv1ISsLVGZU5rsL>gkkJkop}~p5`hAk4!VJ1#T)|9iu2S(5u=J&DNgU-HV+h zqfaXMR5JaBJP#T_BA(-x^ei*yF7-s6$ z0=VoFMP;U0)C4ZS-s;GW5(gpp-Fp2NB}Qy*KBzffnek(k41n{GFpj`D3g0}nd^wUh zun9<6ls@st%f1lnt`YTn5*6=VM@%#&ZJEj6{>ees9rF&qF2_zkp^#D4{1IHbb%CyP z+%+0-YZ2jKL24^CtZqDfnnye*P7DP~HTI7$^r`>JUE{x-9De@Z|NVsgk7NE{C1?Oo z$IQh=3VH*u^ca4pBDkhrJZXaysl_?2NKAkna$rY_(p7H-wVQ-TTS|ir;z(&mI*JJc z%|Nwx8)(vEe?uo(JrS0QT-9BhAsiZ@3)$g3d2^>-@RY)6#KL6L%Uv#iE7C8}Gw+s| zCQlu*1Oe(y+A~M86LJH1ht)agfeEW|x=%VFIDmQW?CWreT(LHyT&Sf3R}G;%F@?Hq zq*zuT!)U(d3E0^Sa9{Gbj!`-ipqp0KA)Adp#xwDiJk(TkU@+o&dlyP-z}DlU_E{23 zK7`Ts{(76q+_C)mo)mc_HF%sw{OpTgQ1O)W+;?6y@%g3E=1|yrs=(lE9f(h2HaLw3 zSH?MaN@4DD6&X&q6xMPYU_yD>&EfX`8jj}BJI?V%Rje8F$Q@)joW7<&58aMe_cYtN zwF}gBG|#wOc)-Q@I2x#G$ISZ>C#g!7QWo`OuZ!dkYuXbO)5-(aFLf7iw}Bc;hDkZN ze6$3f#@(ICZ?5MYwIxROkBlH__SCq}&%m5D6u^(Z^`P`sj_HY6Z9m>}Pd(0-qoH5( zx!uc>GlTmkX0a`i+Xft6C(+6EsT``&N|oeY9`q$&^8$oR@%9@|@(&PS>(YrHSq5NO){UO)_{^?Q>HS}$|OgLqi`+jp7D{^M1_rRFmgrP5U$@8^d@23TPg4XpHl zMx;2%tEwXfFR^sOU^2ndoaWg$=1>%CRbPMlEYh9fupbaVgA*z#Hq?*8p;)ZBE97{7nGqPjj?as=&@cSG&d- ztjA|!)+SGv`-DYR&^!$t>CXSMV70z~Jk1EQa$h~NDEM8C^tk$^0spQPR$yARCTpI} zz>B~G8%>{FD@g@P$^G~#Hd52KL|t2E0nx6J0h`hH6Sifj{(#kY^Z884pr!q3X1?$w7n6&aL z^ro1q^YBg-8WTTaxT@pwJ%~#dv>QXP*N-z)?MA~i%(+nIKS#^(K!h6fC5`5hm@vDZ ze-Td{F+yHb3#Rn&!}dg_Yo+j$bYRw|NeF(&GNnW$8}Ux7Yzf_mxWD%Zn8l_X+a`IPp#I&-1U7^BJhw z(pG!!h8;`IQ*K7veIDGesTo4Go@;$Gj}fL6DymmaDzp`hzHX)@DA5=#6Fo2h)|R!H zn6%?HF;sEo@nksH_xCvC5rW&vT=boOzWzXCV1ADB<;84bGyIL=4NiZd?HwB|n_qjb zKYq6!#V_>>h;d=gq+MCq(y^YzI??nwK=ld`ik-DzHes=!d&2Okn<2s3NF^lXL#iVe zyz{d#(*f%!Vo+K7Um1Y?I_+VF6x*oVJy)h|zZZ!||26A9MeCR*8O0I&_TBBs^d8pe z&kc%q!d_gyT^+(TV^k0f(i+xy^cVM!KRE>d&p<`^*EEBJi#i+Y2dAXoj>CF~!Q=D#7a{O4S(pPB3DSp6ji!0C^N9uHs- z{-q#{_3z+6KZotl^ymo01rlc^$pQpaOTfS8&KrPnnWrK3``%*900W8iuK+y2!O@Wz z&P@yliG=@p!bSR4qS6!)U;cMZt)Id7Ckehs-j9dq-&}Se)lkO%C(iRf7=`{9pM!yn z_JZ_Ri$Je`Sg`$@X4Y?Uh6gDG5dRu@iA(OM1%S9@K*=qHeaQzGUCer^T%Tlr)q?_l z1~Q?PHvvvf9}LY8gb{OZBYj!*YDUzb(=e$uh8{=TjhEY5Jsi-r@_68Y4Up5~`y>En73%Kvdt3NS#?2S65K8wZ zGWX3B>6z3vf~YsTM5Ey|r6q1tA~*#!9WJ$KEFp%#ndsJftPXIBc!@~%uV+L`pAmzEGtN=a{xvL&oZ{7N0Ye zg^{%b1suj_CR;y18!s9ZFOaoAclH0jS2bjY5@PlBV#8T2EgtQOMQidb=Q$ zax(20Ah53#zt|2vd-nBQ=eWvxT%e}V5Up=8C2~{P%8C_43lf(kfaRNtvxLt)NzocO zuz9+*%=d{GCr(W>>YP`f{?_na#)e?!3V){F$Y#}ikadkB1P75->yA8=%~PN~+^hPk zN_9o(+{b4$uc=tgk`!_n5fKpun%Tr8ZLpQmF_X1+=X`%LisM;RGQ@_~bQTwN{di)A zK)|_ZV{ShDiJPxz?PKc9 zUmtWXD$O-0g(W#!ylZrn(migCshzKZD!StgXar|CqpebPlet2s;`j{FRV^{x0~5#2 zt8w;@F|>!TihB#t=#T8|VXAo3g#enJdUr}hTyb8#t`eEETT{+`&VvPDhwza}2mXZp zv=v_*9ndpD3K(DFTv4nhgx%h%ivbf`2@4Y)z&(9?GA3gv7(Qi@s0LfqEoV?ms63Mf zu6V?#P-y7QEl?V0*G=XiHq6$}>P9 zM4z1}QCR|(k#!=kKG|x_`;JcW>~6(nA`%B}e$%h#>MJIkUrmm&#g0xeq;>v=HjBWr zmK|9dUak$-(k6+)o9+ZO0T+;Z0A2pq`p7@7F8@te_qW6gCItAe6ZFidT4))e^$j`V1%JiO zhvi;=PHq#CTU{OOS0`!gzXKXB=%yz5p}O?%GBZ={pEmYe03cs3FU8Y?I_LqUvRYPP z<;rxQ6L@iKzeM-Lbm??I14{W?dyDpdp9p~;|Jb-~b^TLg?f3KIx(AZ_T_3i;Pzd6! zeT4*$FU(k2Pv+eT{4%||ZHr2+`5^%u|ukrKOYQhLuS+N?jJFlNQdoq|e``Lv2Aw#lX&fZAHs06Ej#n9Jkalkl-dq z)k*)3`t$2nb+6|iopw5=G^x(`+M%{6OGMJOC^SRLH{LZHoZvy&+XE&yK|Csw^A=I> z12W%z`>O^krpX75L(H(^r546rC~+ zh(8PDd#Yh1zJ16(u7cX?-}lyftw0-cw&6zE!H2nDZxv0R#5vTlc(!424N3I_L~Uy7 znTUCS@*nnJc~DtZ061WTm2l_qqW!UO@_gvWi|b!RlV`f(_S9*DrNG9nDkGJm4!Rt| zyj{V71z{$jr_y6QX=hBWyHwM-M^Zf%`UCW`n-Z`Zd*gLPF6d$02Jb1clo0V=gpD-id%p=%c{ zp=sdSF5HrkCprrGx_PNK31RI))~~l*`eU7GNC-svDrHzC?F=x zm`w@G8!~&y2-Abmzxx3SF^+&#q#aj&e`$LI0jBjCrR#j|@YS3-rx&|M8;5a4m=@BJ zn5}Qi{w%Vh1~;V?Ij#;))3?387H3RJiT8YgTjEe*elpkHk8b3{81cBHgx(Bgr*dGY zn&W|w^57hjIawFooEjMeR)d8MI8H8(q_B8-m26mx;L+b)| zm5TbH-`wXV|6OzXw|VC0@ejVC{6D>s{5M^!pYinvkFURJf#EnSr~#sPS?RK%d*QyM zpwD;q)p6MJ2c>?`TW|2)e(8JsZMp3465BttnpE=!z!`u-@Xc!jLjymf?avc!|C{gi z^AIX=T#(l}*UfxD?(Z+d2>yG`G+cpH1%O&S-BZmdb<&m5)RrI z;pp2@0FL?`edTg&IDfo$FW4CZ#Df$RmuTi#k zkT)0EvTpXzkN*_I!hYo@!Q}8JR6~jp$yUm{i;!Hd!HBT^cBD&mMHc+?+gG`l^EHgk z7Y+UUp=U6SUythKM*XT;jh&wXfw#2j?&w#rc7kPz-e2SR@~k-e7_ecF3K#qB z1iPnhby(-^u@1xD5OxIZu8;#T~;1Am$j@=b%xDJakfn3b@HU}>4+V|VR) z|1tqw`(x+(L}13U2%!?gZyl~;JF9>?dXUe=4&X19tCE87-F_wMT~y8e|GuG)%M9SB zQFm-I@a7@`6g6QxWxmF*UkPwtD}`$e4-Mcr%iS4P|M~Uj2>cv@|9k|n%`P4&T+&K) zET#WRZp8X~NWn$j=;C!nmW%lp9__^=dUMUz7OV-7Yk&Rl=kYIt^m82k)i^x#g5^f9sjGX6DYl_ndRj-skT9+ONIugsQ8`;p33uprD}OE6B@eprBw&p`f5` zVPOD!)Ff;dQBZISZKS2u6{MwU)Lk5{Z0syiP~=175-_ziH}1bh`Z@|>v){SHoEc2p zh#q-g3YVraTu~l1=)N>f-eg=Gi4iUfZ64dC!~>%^T9E7m-0)YjvI3poz7~h0lgE1o zN;{pVoz8p5@W7Efv**@pP4Em?1{C!-My1b*bueg@Vp&7?PIEfi7zg*IQSR8H5W1nf z@9DW$-qP~!4vQvYVReoYmqOr%Sl9Oy387JY@wvp|<4Zd?pM2$rRCH%@9D!c*cSca-hmPGeST0_ZS+RN@TWJmqAEY4ItJ9j z!_t;SUMqJ#fLq95s9ct;dW7;ynKVhP?24vMsQI51rtbE$u5_NR<;bFU4)muLo6?x6 zbo7SCOth1povB{R^jx>q&L0>p^**a@q8TxD80ZR*-R+a*TsfM5b}pa{#|T#8+Pedz zd2~G_Nmp{WKW{qF=|?{a8&%Wyw~guzGPKGnJ=x`h9K?fywCQ7Pb!nf|?xHM<3VYMf zAE_SEW~nAtS$@B6W|RC3%C$edf9Fkf<^+7W;`2+iw+Am!rjd&Z^xiR5`yQxz5~zxA zF?a(xo}uOZxL?$ZU2})CVo8Ds=bMB`BB2)zzQrAEJ+%11v9ClA-uJx{X2Ohn#hG_U z?g!T(o@(IX;{6(ox@P-5l+GVCUIZ_G2tUJ+B%=Bzp(Rf|{QlDu+)e^BDXK`$2Y9j) z`|9^&aYAXBEAR|(Nu=r31+~z}q?#lcGYeyepn=8Abv2E@bw1i zI}uT!54_j>W%rz#5ws{TB}VUof8KvhYe)A$K|+HF zy9||CN-@edi^@h;C^F?Eo_C<%&&6k2c$DU$&aG&#zHx~t^IM@?-OZuUd~efIVeVjd z!Syr|s)wUP4hnQ{b18A2)Y_u8ds6kl7iH#SVt+_MHL=`dd({OQaKhujtjMVhwLa6vB@$jZV>Yo8LOhiXm9%$lHsFE^z z(oFOew5d;KX+-Eh-HQutm(ew2ic&L@nw6QAv!gd*B)H!p&q9aMmG(eZk+Fy&E^1SO zQ~{hNfP;gBfy06G3rFgMHjX8Z90>>O5vv~^B)YH5&V^QAab8|bUQ7-q=P#Y9uFPZ- zi>An7`7}G;CGgQ2u2eY5sd&iqDeEgTpp$6;CTTravYZB)>QKG;LOiQ#oSG?JhGXllDD5AEt1f zctLrr#^YvHVufW@nDJ74x&6~A84q6rFB+LF89rGxpP0=sADEAaxB3k~NNlv_O%mt> zh<$kayZcK-oAR`Bby7o;EKfvV0~8;W0+M^fIW`ZO%NH&94q1ji9iPf~m4tsNt}9M( zsN3P6swDbW;>ABRkjyQZDvZd@n70eO%_aJOc$5hSAy-W z9XiS(CH0nSrKjblHVbCR?&cQB#gdN9iKxxz9nNXyDa)jc){LQ=jUy%HQLfR49PoN& zGM-Phf!NxP1f~g3FQOlG-5YmvSP-t z0@x626&7)YccpQ?dG+Fo@;c!n1NR388oC{JP0!G;z2}vwQ4;VUtAX2+D_=Xm9(?U+ zUTcnC92eY6{UK-wKl<<}Pye9&ejyHk~EYn5a8w(<#~7z-AiviJ=~cUmR7WNz#qjQ{LrQ z?bW@Ly}@i!IvV2&Q3XuWqcC;PUrQfE&2XJF_2Rda8%2F(?$ZfI-d*kOvFXarY<-zK zs&(j_%_p)s?RrQxLkaxe1Y8;eWLDYXUu9J{UQl?s~yt68g|HDN=M z>e1M|*p&o4yR`C38*;GWP*wE|sa=!Crn*7x`wHngX`P2g<7Fdn;A?3aJtN(Yx?Z}x zTf_5}{u7sTS91aG3EB@e{yf_LjUmrQVMN~5puCVCmJPR9?5fhK9&a}-@)z||L~6C!RF0(d zNA^ERJ@?>8ZmkRM*OJ0D2M8{**hr?mn7vhJ*aofd_U%VQg*qjscQsH zZRl>8_Uqd)(9AWrUUvM_#OERC&S&fin;+gDo}1>Whxg(Ski#7>W*gkXT@lmUku##6 z-*%7YRT@|ktEU8WJq@Ru{`1$Rn2?JSuqaYXe{Ejh}FvbZXIc|FD$wLqNe*Eh2lPhd99ddQ9-X z0QZ^V!R8DC-;sBL!Vj@mWzTDOG*n;cX}%jS_QpT&yz^>upZR2Dq*OLzKcfeh2)4N< zJF46@9ItjXD-IIo!lB7)VP-~Y5W_;Dr$Vt6K*_3&;W-YHh(X?Ez`9~gPpxPCiT#$0 z6XlyaCH2rrL_`GyMO`YRm0`Q2vC$}ZZ6Q=2Y!pj)SDn5b+YQg3skR2^N}2(Fw!3)) z$60xath&K{oq;LI-5BBkz}GNKJq0Tj6%-a=8w&;P4jBqMuyqG`i`^mr_x6iBOelB% z+K-BY5@Lgb_Rn)vf$!T_B=Ej%^N;Vl(ZMK~z_0tj+v_Xpzs|;%`g-?Y+h|+BF%(Hn zX$1w~ThrXd($c~8jicK&&tw9y1KUYn-xUSrA^q+9j)Df=9?<`!jh3F9o{F-NxuZSD zYYRs+OAar4r`vu|M7)H6O?ykX*EC-Cb`GvWUZS*rogoBl-|pt5rTOa=H;^c;o{Bn+ zw4;k94L`>-j%T!DI5adgA}$tILK-qJ{@EP(B})6o&CN-Olhf1Flf#pT!_mc>lS@!g zkngh8Og!*pVI;+$a#B&lZ)dS=fC>~nu^@+6;ikHvb58ev9Ska28#uRB|F*jISR)E)Z~--+q1Wn}KT^h8~Xxq5EEEJV2;r|LHAeX zzft5g$`C{sip0jh zgGMAlGlZs7e4z!KteNs@(oqv-*cdV0_1!&$5A+yHzc4_2z}s;>1KGF_f*SJxL#3}-9KkFk(|!K9ofvDwtck=w!^7Z<3c8uHh5e5;4} zroLxY-~I>btP4nDnR&Zj#m=5-!7!g$nzs;bEI+}T*s6p_&-6J+1Pem7s4^GN(# zf8>`I&j-;4d!A&n9V{{MH_Y>ytBI#XP?$-=C&Lck2RR;WFh2BoI>^>7&rFZ>92Xoe%T zQL-%zeWF>B^T_a^ZuhMLJLKPrxNjd9eM)*(|Desj{ek+bFEc-Zc;N}_vK zRsNfa17e{QxbsfEYdh?~n@9pBu!qg*Xp*be>(@8!jqWDJsKe0;oDoM%0)t)aFws0CT_HM7g+1!Th>+}qxzD9kMW-ti?oeIXlZ9DOul$|;aoV4Nuy z?urc36LUVuN`3CEp_*}{=}_~e(V{z1QX@zC@wj1{kh{TfjtcR(9>$a3UhEK?1QAOz zlnVak@AdiLN8?c>nH}{>SEx+YVTCJi*a2=G`-S8 zg(Pm-cTJp*MCCsM7_g3;kk_16_%Ci-h0-|Pw4hHlAq5tME$g`;M&7G&uw!{r0*Wi+ zn+pUdsJg;tSib$!V_H$PWu!DJMEu$aHdkfr(3PgNIa*+K6Ub)y&i{f^*t07)qbLDz z6pWQ1?F0l=DdFpGNApwSGxr7mTCe?Y$9TimFEt!ym7;mic4;5xU8Dy{R$c4?#(ja3CDu5Fry-*XDYyCEckt+$N z0hMBO0nj+F387(6A-kLFgV^&^WqU!_75NhJ+*;~c-L(dVy{c9lg?naGaF~%-@7dDsB6O7IxFzLAr!zjXe4onW+MXTCGaeyC(j~57a4-XgxSJ2=(oI>RnAw&6Z$4#o4} znDhXk8LnGhR=DvKY+j*{oDBGqD(I?X)tmZM?h6(ETfg>MOP}LDrAEi4 zLM2Nj7qoh}ZpvcgyEKG(i)@Rd42m{RFAG-CrWq#K=#}r<6~<=dW~c%0&G?w2B+>UNEd$OVYt#5V2fc@^Zb{3>eHr zwt02O<|{K5eSOd%nvn>u%7&8_P*vKMr-qX~4XN&2Ob?U>QBEDm6C);sc4)G9!xF zII9%9D<*VjODFOLAhgV--R`|ztH5~3gdYPrY}UfQ({rY1xaM`^Tgj*z^etsM^}w^6 z%gD2f%e=3Qy}@*>0E~0yRG|wXjjplw@SbN+6_;B&?5*;@E@Csh33cxTu&u%nt!HPS zpsQX$M-*ah#c=Eymz0J<-`qL5W#+qTUlSqx@8`(*{E3#SZwy$ndtKgBdM{7{-#i4A zv0-g9+^Mxz%U9Mda$!Uv;G9gsGcqyxnI@LK8T9N<8i5tOf43-dq)z~j^~mF3l=(5A z)weO|s&M5eYolo(KCwgy%&Y_-O(@;?RU2eGFS|6XRPCNnE_5wr_Y_8uq`)PNL;(#f z`D~rdXZSL651n0v=brg9$eb`QdAcc|`YI1}oQbCV6gQw7NO^;|Ne6qgwZKINjyWi& ziFSR8MRIw`4?A_cNq|?c^*vP~7x=haz3BB9a=M2__tVw8_l())>dbkFnvN=f&dZ(j zo%%hayq@Q=G+I}`4sRuWL2+g^qY5%8`t|`gOf_z6f!6tOA>nxNFN3pxeEKqTAhAf{ zC4e)ROycBkIOv=QwhkH$p9g&T2~QGaRTn+kyBkd(g}i%ucE#o_s3V2;T{ZEsa%$7d z@jM;Db!MvPXZ5?M{YF$Hn+)ePu{VPmPt0F*C2%&a)SO#&*>#(lxZ0B`JYIEuh|y?f z_s77L1{=$WyEcs{AUz%k2ct~`ZuC92FI2x!h_UTI{uoAwMMA6)jVCoVmcfgny;FRk z@l~QZQ4DkD-k*;Ui0-j9$&mrQ?8TAUFl^&{5NG#Z-2+82D(nF}1E1|!&OwaX;Gf}z zTaQVLYQSvUCfBoqC^FTz&eAO2nK@1Q%g0_Felqgi=Cc~F{^a6&GqX|vDLxQE-1@c9 znB0}DBr?a5wWf&gcOhEE>*N{8s)lH>>ton4V6(dOM*_OFbxoeMg*`kup21c4MjOk6 zlFg`7HEXPqX-0oZ;Vt`nbo@_|1ky2-xNY3u7&=rr-|xr}8DP^c7}Y5Ax|eUTy5JNe zx|gPZ2>3DoN?lRYe?{H)Dcsw}f=+v1vy>84)9sIgpbI87TarvjqF9}G1={sr!2S)i ze;heT+$ZrWm1ay50|#fRf- zNeIiRsrA4`m1TcwSG7^O4@pTOr1Xt??Q7oK9L#Ku{-z5xLE?IaplxE21hE+IaOA6(YM?gqg#0g7QqX<9 z%MCJpR#6T=Le_et!k@70+#vYf!* z&TypqOlzM1dawQ~=G=TP=bHzSBIff} zbzmmR{N^4l7?~~jk{g(QrJAG)8wqE_{$zhk75$Hpoz6ZmoRm|sYD=&vMmg;Qrz@80 z{Ytlcx!ppkrfRwfb3aWiLD^qJk%>DS3}Z{VZ_V=An6md1{OP6NL3@W&0lI9MjlA*U zVKpK%9kjX1+OSG)c#2IG?l>&QR4)lD?!o zQ~5$AOhyz`R-GZ@uK40}(wm}qwwKDbS`a%&9Iv$h%U*ykzqJABX(AT>CrqNO@6?My ztvG)?@Jsryl1%m|6`ldt#|e0pT;nX!E8=|V*Ud|;Se54xM-E4#?#IcrSeB+Hf4-_d ziro>$%C;_kod?ilF>Oi-`qXrKM4Lw&Biv?3Xdq?4lPfF-9~W9ek+|CTHV1gDmjtj@ z@%X*dK9vsyut_rI67>8nb@AWRlSz2XgH?&Fe$Q<^FD2-?!->HMt`dPx^a5Ax`k%8B%Gx9z;(OiZB`{Jm!@}d+k;Y*C#-<14U4K3pmQ46t${@uI zvBuGSeHE9r1_cJ?^n2z+isB0!Be|9vWf7}@D~rP=<;}V>a+5@}unM3m)i;-me?`hM2uItDnG~f=)&S2x$&Ro*E*ZtT6 zCttF2{Hu;BNe1MN zI-(!EW%*7$^XFvn|FD^Vv)CktNk2~0#8R4t#ZJ~{G%6FZXvnrd{ZkwgS%J31b347^ z0`<~#vH=1v*_Mr_h(=W+miw~pe+*{;DZN9_JW_s&&(~&^p@*Kig`%02+JAkOQ16DI z+W7S-;!mf8f~p|%<8jci=3hFGN_C*Sq&EOeo8K_uve5(}nDy;Coc_~Ug& zBZR(16Pw;h`KruxTWzJu;WWet-!nsoP)e{>?W7Fm0p3_u&j_*qb_Q ztm1JCCbV8N{o09H1g5)Xkx}D12Ta%!u^0BI3YBof2_g;}R(t4B(xrptDI&>a`00NE z6F*i0sR(H~**v`k@vV=4IH*$++B}tzVlV+UyZwyF)9sn6FA3%HAI;l)LO(bYy;H{s zrO|rii6i}|5lwWgM7p0W7*0(6nn@VVlG<&zxah=8tAPPQ6N{00l~O(c2xS%eQ|J$^ zl7t579wA2?JswF$N1c(wFO1BiD~&AqdnY!*!u+yOlSb4LMalVg0uJ;kiotEZ`6+8o z1ZX0`gkUm*;17)rfa}X7{Lm7YmLDlJGiAP0znE&lmcAFHh~s$&{-@S~s1nr~o4d_& zAq(RLQMb`O7AgA5ndrkm1(6?RX!rwjCbx*Y1nEyEk#b10D=Xj4*oxsP2igcSd5Dg z`V-$=ZM`H8DtoW=ix!jI@$|Au18@%df9xFSECbagbJb^}53F(Ev$WV!8~ZK!2P=LH ziR~fA_qqu<1PQ5Z1t-1g7m#r@T1+Zcf9iE)V&3X5Dhvdy?+w8shN}8_hGF)g1)mB| zqM(Zfq>S{-xw@wN+Y`};e~g5FFfLLDAJ{*`?P5pE&L~oY3KX)ad8{%Zb>Z*{OW*o{ zfRf|Hc1BxSrh1&%^g_VBp=G`^7P|lfl)$rl01qWv=CdtS%LUTOe_2#!1U}Z@cYXEc z-S6*5B8?xccyScc>S07|SlKfTQpzS^A%4AZn=4Cvz(A>Y)zXe;DPh$NE((lY{l?i--hCNBl!dTGut@vNt!% zh^b|S#S>N33@}VGIN}jPsJ&-SSSb^r6v58=3jG6XeJM@RcfR;)r3rc3a<+v({DS-V z@;44Kk`C=AT$1U-#9%Z}&5BVK=O^8a#y$jJ&OTaZ!{}0}V;dtR+jAlCU^zlO#C=qx zdL=$x)Fk>~wL6GU>7wd6S3%Y3Su<4HJei#rNyn8B;U;^c#&)>c`G4WxSGmccq&6Uz>tpXd!7&Ggd z8tFtImV2LfK(e$E9vtb|%fU)x-6Y1~!oNWlx5_25M>ToEbjd^4QK0T2%XIyT%)kXB zA+@I}@Lw}a0HARpS~g=EMQ(e&*M4-vU}*}Q$wB6XduS=N?e1wO0@~=NUm#5nr=B6X zS?SUo9^S<~ZU0Qx2|ncw-LI2beyfl(Jt&v68CkB9!L6#1UDpuTkqX$_o*^G&Vj-+b z(1b(&$X9(7=gNXOi>t%{hzvU3*Qd0B07uaoPHF10Elt$jcF)DKEi7*MVCtlpYq^6U z!B!hfs#aa^?@iT5mOB7@gOO2zPboHiDj03}s9PK>n|hRha|#M0+?`z8!mZqL$Af#l zQ&;`$WiXz`Df`8UWxpo#a%`z=mt)yW#R`*&AOa~Hrt+P+K8w?x#$u(b@gMU5^9$Cc zW-YzVLHi2N#n5QYUQLq!Ew-fD!1ns^b-LQ#KJ87Oy~1vcjK>SK%ul*55Uk^LYzg=E zm14Ya)w1J-`dL@TmaF3p)j;lIjs-rDk#HB2pnDwJ6i2U=(DebA6m=*4&h>nxl?oZg z5u=5i3qS{Jz0U5|alR*Z*{eCc-v&1ZcUi8Er8in$e9|wDC&za7^x zxDhEr5rpD5bP-}VpRVd0o7|H!Y>tE6b{5lv*`}e$Rk46~>dy730PDThG5_~SL_vur z^{eJ{^XTzhoLxICsmHl14}yj_S;s&Eu^9n3ZSI)7HnS4*>qExHjVA`WO^q%~KNW0F z&2;q}t%~L>8?Aagz84yl>J`^4;phfP6%z$1%Fq22=amD?T6!CLo4N z!Gw<&HxCVfAUBn_5&mQfwk$lcR$HE{0F<5>fw60)h*+Mu6C)?lGO?uSZ;VeDeE%AL z?T?0+rr~b}oN5fOSkb9IQ4lvr7cANzb^Qp{?9bX(&tO4B@0ouoJ*a8c(Ig15uy zIRRY;uBO&)BlpGA|dSmy7y#<~7< zD-PN;#UP$!`&#^JFDCUFzdpBhyxmlp;wp}(!v2yhJy0*>?2WqsBW6^q{sN_h=zwh- zp^o46-y9k#DI^^%Ae=z+BEw%z?BPTMmW@z;`zhP2)6N}H@kdC!XyMbo=$=_~`o433 znd@_HN)(C!_%kK(D|wHtS475isk8|r zaAk$hg)_)~c@1i@+RZoK=zg|$4+8Z$W;qyLCN%7Uh#NB+!lrF^Z5yj>t}7gUe*XI2 zj;`EyepO!7a#P%&R4s9`jqG#3x=a?}Iz9PZEC%mx7ANQm#y9OgT}!{{*D9#n>}Kt4 zj5T!V_l{W^2uOKQ`#sXN27WM4?H?-}F79Ms-c)_jpl6puG1uS?sm&$mo@%<`9xEwR zoHsm_|Bm5K5BS33>!rGH@S%CFW7n#r5eE4r?ra(9Vjq0|l-+*LrYn{UqtV^|&jMC) zWMhzEO1IBF<@y7X8GXVYTik2)eS)dnniXqtepbojWfzn@@cRil6RwSdFtMjcJ~{GH z8VJ2E|T5df#W060ZMzPbmyBgu3(KJ}gY`+2G-PhWg-pThv@_|phFbnk59 zR(%tX_S`v11|7a<-@$*ZA?R~eqnswBX*B>=^S|7F3o?a=srH<9wx0Ij+GpSkj+xl` z+bxuzVkYrg2D2GeXsKq1ZTlHcl-*3u&o>|&`^f7rq>I+!DQ5m#={%m2oddan5co_* zbTGrpA~jR8a6qQp&V_AWHTL8h$qVd94_<@>k6@%+nzfRWOMl_eqiJdwE&4udDaGw< zPXZZ_0$BJA2C^;c$J)XYjLmFlN_Ts3zDLpo$Xcajm<`2RtK-Oal;zYEr72 zrrIN)YZ6eUZbl&Y>tSBsnB%%@1n0^o1W6DkvH&Z0$4|Y>5JAlx#}eo1;mv_KZL~;Y z(V6U^`kI;W8KLWg@sf9tZ#ix1psQPZ+zospb_4yOLbO?_TOW*v4Yq4BRjYQf$qZB(LGY#dE%T0UXa4c-%Kfxjl{| zBk!=lz2bkv)qf-mJj3E{>)={iF8_J_2=EK90KcG;!qsyzG~;a^bip+t$id%3CgF3{ zWxYB9zHFVp{vwRio#p(rFpSPbSoXsF@*qMtNs-A|hL!!p%uao7euIq^8JBiQSRb4f zC~r)9#S)rTyGFLK@sr_M`9F~$%v`3QP@fO#51XB5xjBJO@5G&KT51R z@~rKwv^C4@+y(Y4VY%*NUZk@n(K#k-;cfMMJ9kf0Al0qWWI~w)Vqp z7|yEI0TNf6vt!kMNN`{FDR}^9=KQ*-1Va+b>;99z@nhKTM_s=3_RpI=hIf60J@<{$ zPv>bjMpt)FOEGyt)AATc;5m~@lVd?~uXXvyVl9*p-0=?#hlq@z0w(xmsYf^iBG)V5 zxt$LxBA>8c+VhGC`mR|(G!o`WXLO-q>OT1MX?y~O(wmLX>^dd#b+XL{_oKv8bc^&^ z2dWVi+^L&LD?)LvriKfbghv^DKBAcTR2)y`B2L9+29H^x4H_7Pn@`_EZE5vB&z>eRQ<1F>`S zWDbOUvp!>&lV|IX{9dft2c+)w_vV#re_bGqubqw0r?sMO_6Ic&;28lg{Fu^;*XKML z&fgLB-<3Hqzf?Ya+7zsBHo}WHmYYz25qJ3f;+)E7W#g;1a;o`t0v{fq`|z6~>iJ4~ zk5?~tor|{k2DH9~sUYl)b_B#q?f35m4d<69&z+X=jcI4WGc}*V1U!RR`dq`|#Rj5s zx%bL5(9_?G2=QI-A|#kcnIAW!0m9I>nJf#|z7pX60$*E2-wKt51h_%BS~{sgJOD zMfrCeeP-2eP+~|5dvz0coXR*huGM5<2kVI9$8Y-bn2(ock3@lJGmfgUu)wVUbbN3Y z&6Tn>-)wC*;&&;~C6-7=)0YcRrmMW$d!xsutEp1Zk$>d$wu$ga|cqRXxqBcug+52lu1rpPWzwa)*L=1 zto6CM^0pZ~6FUb&=R3={A!-;YD@#(J@6Povx$~<-P7&63&)?tge;;&dUawlKRqwAU z=*wU7O$`7$**~_=3yIqL)9P2kIR$)l-e5m;8nhYS$3ezo7LdLS82<&jI-i>r@n%nUqG#+whP2h}Aap+!~eyB;LIFM~LB}}gr&vr1PCf-vw zAn+06>S{(tQ?eK+>A9J^SLCbm87}hHAES}*V6|RbHTlPs=fZA0yCEtvP&NbiLjTxz zZJ`Csk2o#0nrlzjzBoGnerh&dVYAvn9l(FOlu(S4dYvxB{_8uT=qqTQ$$PL4IA^_F zGUGZRO_>HfPGVmx5_MyNp0U)TWgZ1Tw#xA*6R`+_H-3*x+)T6yEYM=(M8}=q z!NK*~Ab1~us;VS>;l-CEn!A|z?(QJ< z9L!RKY*0en1NErofN-BaAq>fQRy?8cABx(mfY~T z7N$$pZP=2@;1)9$beg)oK4gB7m;Zc)`b(0qZ*nY?&QC$!W+(*FccnA>)wvvVm7F-( zh5z~sA?54+SY@s`;gu%CbjNWr&QYzkp<>7hGI*2MORH#n?oMD;7)iH6hJYzC{u7~; zX_2Wi6$!81{SK;VT@#lD0Ta!OjZ(L;iV2eq76bQw!Rs81AOo?Pd7STp3*`#S^_J>L zfi&ZLoS3+n8-W!q(f&I4&O$R7#7lP8QsFn5KVFR;0IStkdt=yodr`9SZo2qI!>0I~ zj}1H{*A%dEW%fF=DXH51_QHP6s&o(#mE+Z<80U~2P{C>>FGnR%?^sZ9i5-&D1Ca0Fa=?yBby|PpDRLglO z7${B4QsD7CCNFa12DQ@-4XlpwmK-g9XTMv(webL)twPl9V3MKs-k2xnQ+WaN)bf>S zqdy(Bs7uLc!OK2A-pu#vWGm~Ft0YtBa~unow%O@|yQ*B;dc$WIskjD3IWN*)6n6{l z-|HVd{is5u_@0SX7rZGVyK&skdhi3>XAJHq!OyO+o$J@e-!68sw77^Z?SHNFW_vp` zK-aBI;Fhj%fC=U{Zu-nM`a(wwCVo!j#?|xg!A-B}Gog@OG?9n=c9TqmB3D_U@bp|& z!n`rZhGhSMl)1~2t*J}tA8StU)i?UwQpo6z2q9 z(`WgtXKOQW9`1k_xu$At!poI(0?J0UjqMAuU!829#)Fi<2Ok$LM)8zvQ0(d~@^CljyUTA*{RI5kFmGTDfZiD^LS9K4KL2L%v0}f=1M%VlIefe~t&F|Ftl;gg= zY{jfc)$bP=gv*ThQn_f4YbJ1tn?6=2VM(uE8UxI8e=7rdk%S9(HY&Q0!$W0R_+;k&EuHGI! zBQcTUOOq^3Bz%=*ZEcYLV27&nw8aY!Om6}!r-D});WvCLXDO$d=I-O)C0 z`vxhQo6DebDjOn2*0We9;~|pn0#j%_p0BUOC&!7*!=8irccw0@QVqeI(^lT^GLiQ-?^}m1Hr z?a#jL7BwNsGy&I!i=2EEj$B7l)loSR*;XWpl_7sTDVBcoK>*BfE`VRHm8a1`;x|$t zacHIDj&DDFB{_9*Tn!UTo8M{8-c=wYw!;})OH@}SCGK~wVvc#C)ON7yXKW&w=L8e5 zH(fUab8g?=I7HOjlN-k!ZQ1{V`KTasJFYFxFN@tQeG;tcoxwb)kevrHQKM>dx&=b(|C&7ihPpl1x_| zq^r7;YChUMHo3MuhW@c_cYgca zo0C3Zq8^{DkOd73ZvtUl_QKodgI{G4#&^=xJq>qe1yc4@8S_mlqv6kxC80Ws=e8r$ zk(y~!n49J-<sw zpqbjo1GFmwhd7k2UuOf8s&nX*R|FhN)$@Zh~|O!R18-S**?roZL7VBpGBn*8w;( z5-S{Sg`V?ylS?1%AR(_k&svBEyKap_Zb=HabH$<4c$u!{R{w55jn`UtpdU>v$SsygC#@wfA}7;1Rc71>R%+ z?#hkwu^!Z%Vpsf+uiJnyqPHIj)aSG+#&!3s>_0V@zRtM!&$Nm2=X|Zo;$!%ixyK5X ztWb$~SF&#EfUBwGwL!5e*!)OC=@H_3_q4gTueNX@LezJ5YygLpWBsz+8c2{7hPm{E zkETi;7aR|4BTV`oSeZmDoC5N7b8{dWB)`ir)?5J`cQ2oOza-P;Lt!8((91{ijd!se zU`XqQR}jk)qIA%PQRY{lkVo2{$u_GdcJN{%l9CZ|%8bGrt@fn+&#CtN@ANfF7Pn!= zi2&;9R`QI8Y?4E9BN7)Sl}Y5bC|`V{;sl9b&YD!Ryz^`HPIGJw%QrYa9B4Wl$6a_S z|FGW^Ik`LYDEz}Dnw8l;e|r5JZH}~L&~Aiy7om6}^7_zceed;DC8!!iT|`thtQ8Xq z3g^W-4#TGG^@1-jLwM-~UiX%is4oB5ZR){g=v=ciu5Vys&+r?aSmi7F<}ixGnXMQf znao$9cwJXX2O4LCfm`%H9@TAXeMZMgAb<}Ra{(-BCCk)g8YTfnMT_)F9V}S0xt9w1 zx)8`WOhLvWKqv{J9jz)NH_~jYC0NqK7Qo7`7rD}7J-3_lS(x`R<}Ws-e70)&zWp>* z^{$OT(M{!$va#o!`^FZBe#6mIo0<7*Jj4hjWQBkm|C?h+z{7OV#8APa)hUA4)m-k{ zu(TUkA*#pUfoEssClY4IavmmHE8jaT47ESRk#gtGEE;EMAW^M{wk;p5w)ep@g+8aL zM!@a3L{s?evYLgg!&gjZ%Pb+h^}F}rI|pUwJnPtp)=^cb4>_U;(o9_Jsl`2Nv>*qL zgF}^E??|9DT9le?o2{0tMsh=!QADxfLBCrm(Xl{XgR9^Tt|SxOI5+UKnhMdI^RT!w z+hc?C!vXQM6yqL8*iBdkK*2%OAXyaR{w-sq332%`9eSEyraC@ubeyCuY)AWeA`Oq* z<~cc_^_~v%i|1|#wYcA<=<;seSd1j=~b!Q z*6f`7M=Gd$j2uA$KHz8hjwtLi(%eVUWwbCt27~0fQ@?oi{U2EiI+cGVR1+F1H7)t>c!Q zH-i(4^@ud#OQz##0><%88jK}}dS)WomMbE$fNOsWrgs`@(H;Ae11J?f$BB2N6wQLyO-l*qu+IK@fWw*=yQ0~!E084MD(lgII=sbBbAGwV5_vPfykT_te1Y( z9!{Nrd6&1t@@X*f$XLF&D?4phx{p|Y(;i-$ZYc2ersT1o<}I@rN-b%LGPxWH`A;Fi z>l-n(fBhH0fA;?O7B=uCO!wJM0#MX9qm=u?w_xn>00F!1cW~A*7X)8aJmBz80Qnl} zVjTHOC-d2?3o9#&SA`hXwtD_}_E60DJag7flOG~9PTAI7d*ZdUAF{gYKzW<>59NJE zoN_w-s({i)Z|my(uykr}8V)Fcj2|q)`;)}9NPpx7_t#y-osbdVb97Bf>$VOEY85&n z&o*b#2PCok)@Itx&ko4;s(ld1&N}ofu3fWzENB-A?Lk};0VGwLC%n4{3i~CT&f%r< znN&fWaiY0bG3(M}{i>{?`iCg{vNa?@p~_CcAmk(irqL=yE_V4dd4BMGzVVVEdY?E~ zsD7y;bvb9N%xV=AP)IgS?8K;-tF>Cp{kE{Peu2>zC*nuGJ!l=Nef}KNV8EC)w#X27 zL`1U0u}p@1+^^Kj$GJ_GF4SBw&Y#As-_>NZ*NqU?=Sw$Yswwj`b|c5j!< zcDcGe3JnnYD z-z{czBy`d3Ev;@_(l4v*%5Zg$arApWNV82$7RzoiZS;!O)OekQUI=2n)JffL}m zL>82UPq(fm#rU-2#*;T_j}MXFYs!PAb(t@YS-8iQ1=kktU({VRWi-OhA<*peM#34$ zIH*ZXbd^d|fiZQB} zN((pd)1c-ReCfJzg|9LybCaZ;!jcpRr>>fE z0aZ3#To~F+UrEBBSiC|CWZNv)8-ts5js}(*&gV!9slZtu8ctkDL8MT`d6(R-($1`N z2H60@07F=98~TsLl1|PN!smt1aXMo0OWbM>w-ID~n?lq~8#|TDGS%<@u=kcxaedpjU<3(Hf&~ZxLU1Pm z0t5&Ngap^3aCdi?pb44;ch|z*2@u?ZyHi*Jg%(hS^x@uj<-O6}zx)5vA9~y|&Nl{d z&X%>;nrqIvk+UA|&>DX2c_(P>K`Qh}ac)0+XAblt5{oMPlQ#a%VbFFSv|+^ zQ^TfU)oP3ADqb6v<)q5NS+aTPR>bNP!yT*SRxLVyE#HRT17UW3>=$_Rx>ZM_*#eQdF!*=e+~ebjpL2fd~3R783Y0nK9{JBeuJAQKJ& z<#njUM;`nUf3G{OmwsJcsHa+Qrarytk8T^^g=5ly^e?&aU3Ol8;=kV!IHi6q{xK@C zAgs04;Sx30f#&iW*SxOT-R1|B`i3&43Uzj-tPF*X_AJt-*tOye8wP5~?TQruscB!r zkAM%}XqNL0j$1jOz8z5M z*BJOexD6NY2dTee97v@PLv^ERg0q#smuNEro!0ouTxoZ?DAMt$eUH7o^S3W`t*^V! z`7~NK21e9E<+4eYlYzc{L2t>wBSR4nQ5ck!)LciJ0D%r=Y=_ywj;OD>nWF zw7ygPY$?}X0Uy}@L2$DCeJ;c&4M?=Lyb=9Hyq3^!`Ds;ZCDUXPvXq9z%Hv#0w#wWE z8k15=6xoCKS|L|geF+nWX=I8@aKTwa}3-|yQ8;#@o2E z{;9Yy(SMAbbdu_oQN0)`XA&Jhpem&bx&7;){pCRe=r$@nOJ#Mhdj4thKVe%2&ue@O>NE_4+BcCsA*LH#Lg@4t!y01>9WcP zXr;rxpotc%pj`)mactG(3I+YPubl}@+`BexvRyY}-+?7os8RUt5m-IjCeQGe_q$ zZPF4e00=$b5u6yHkD-Mu85kG4d}a)QfkN*kw^0#Rm?~E$wtg=U2^I9@Ot*fi4Cs5vK84Au=?W1s9o8{f6~P8mL_~hQ zcU=I0kcIg*?oq;6Kz{1>&5W>bw)(hl-*tgmL#*FM<47D5t!3Xu;-Xfp+vEbLt=ekd zn)6L5)KOVf;XbyY!1vfIG}TshbkYI};^1VxF-?Yy4J&V1y?q@bEu1L~0xIUt;3f#d zh?hjI8ap}|zj;oH$i@8py{VcXJ3wlylqzc%B{6<|nn~iHf}rS47Iv(3Q^g!VKT3K6 z`+HAkcSA7Zf5kunr|oi;Vhy!wzz&r6HEo+ui>Dg~@!yyXA0r&`&)FG+044;Fj&km$ z+!&h*_oN3yQju_G($X^fbN7B;@ixf?TqsdCzfIX;UTh$6UZ`nM?2Q9&B3xVsv}q3fsLmf4$; zP33&KWY3C7v+ar5Q-S#{0Ilt3Epu|w91twAM3FUp^$AI_-k!hC6>FIr=71Q)fFh5n zDOnaa&t$B|>c5b5)DS#Bs989KGPobE=08}vZ6`hNrxb3S4_PPqf2F?yY8#Gd07ZG2 z4)DGNdBOBVgh&87d1%9ow-LjxEM?QQ54zLWJriu8>ZpCLKJPAi%fCB(|AD^7QZ4?4#KHH428X`MO-ZZvREp@lQ&3R`+M??6! z9wL-NJ?|Xi4rl)$eZ4`EGwW%?7j+2dGZ7ELGVV%a1YkT>pesdoG);$&<1yTPVc z>OO%oRW8u_z2DA&w_a^MuW_6(lDEGQbdn_&)~7t=x-JiD>fQ}iQ}_r;V^?n+`o>6=DO-FZe3K-4o*!3X8U!RXUSxX1zF>OtcOdBt$-(KRJHS<My4qHWILS?Ed`mGsJ&B(Zhp3^X_jBApZ_o{GW{Hmq~Z3rc_|vdTg=Xy2V!e90X;o zezAj>W-&Z_?LFuA*|0-a$9nIg4;0H-X1#j4bL6~R3Mp;u=Y6{XEI`Iv^A#;jY(wZO zCOkf^%=zTm1_jM&pP5$u@DqHWPD$;Ik)=FNO#irR(Q=FSQeH00$t+C-VF z2e9I7%Ec$!4)0Lxib4}D4WF`3${4ek$UrBb95;Hjq*uv9v?eXxqP8bu?X5r41ERw9 z9n(3F;Du_b#XLO7gEVTLbZyoiRfuz)WuruyeLM2?r#BYA^$OKnLM=bNTqWO%1SZg- z)m3JI;suon3%7d>zygwgyZ8}Qg8#hsM6h)+on7Ntf;!UX;l93CjVtwPKze1$*7r$W}vg@CB0K*%N#4VeR~J ze@deATezloHgKM#QFo8J1GWP`cq_Ua;zRC~_($8Rs?gDl`r`>NDgH+Ev2Zj{64 zdaw~iu{e_WzS7tvz0_7;C)@72Fe;EOsE~34`_H`CuVKRIov>6X2<*fCU zOY>O!;YS@BK=mQt;VxgiblNvH5iFE$4_3dXs5L+OB|KL6L34Dcw^`$*Xze_TlUo#F zY4&-9KF>tf_WbJAv0q{qb4sntqKEr+gpa-|xJW%asd=F!U_0uPn>W=*KHnU*HXxh! zVLr*x>ekD0#Cp}qqPHkJQP6J7wdpq1wOg+l&iRkBYo0Af|BtzRJ{6K#rJ zm&MY?`7CC{S@kZ}Fvs~PcSc`_=ZO)NF;k~9=B;Aj2{s=qU5*ntoJTTKj;Q;M+i}#*S2e&&e%Ym8@N94AvisF{j2R zMH7uG@>g5E?E!5)^_F5p-R@_xrAmz|SR)l0-erLFsMMOM8!x^nmbDRirDigsO{1;r zF*Lc)E0J&c4wn_4%%v$rbm~yc%PRWae0(_;z&wL-9hU&S)_if7*<3$Q0e2E8;L~j1 z7g5Abe%C38V^L)}xjFwdJyl znqWA>H-2@AS$f(GOZxNY6?=MlsXHf;)1?clO{|i#$|Sb^p=S?t{~=;hVf-yoIM{~m zOe0q&QOc%)KAyN3E!?_?JAxh6F% zS+G^UmeK)Hoy&7)UWA8FPbAd|I`TD*Qkji)L%up4Ot?e1G*TMKD~8>HmX+M=XH^F$ zoJP~zswJFT1!=Y=aia|xQ#-t>iT&RzCV`y48a`L28r1wrS z!$suhSeqGu`n)-qSZsN_l+{(a8YT-ur^6#AWwUGA7;uHCKs092>R19v=DzJ$sXsJ> z52x8)U)mj>ixx9upWRFVL>V0EWpyQu#4QDoV`+ss<{)=Z0?ppX6WmGDsvCGHtCI8- zoBGQrcZTJ~j7Z*+Otn8_KB8rCF*2fl7Lz+){?<~gV8h|$Sm3V;$bZV|%kSO`%U=Pp|LZ?Dei&s2j&m)3e}z%^ z6Njn)GXwj7B%;j#4g7F4pndPM@|WeuKR)2gHZi~~Dy0-F*lu32Mz(4IatN_>C zkNz|<-a`K)Jw1TU$4rfw89(BkXUba{G_njR0!_fQ|RkQDaNO4f&etEnh%Dp(m5AqGlr6&4JLZHl;PaNs&uJR*|CT-f``r1l1GD@I zu;8)9{2%z<$N>Yt+4wjxbT~P_>f=Ax**_=dpI`F7P313A;G14PdOutD@UQ5$|9VC7 z{JTZJ2-yB^4GpOO$Hp9bIKV7FLCf5m|MH)^C1fcB1K(hSmYH?32`&B$!OFir_P^Wf zU#ktUtN8zHd#&8=rDE0-OCS(YO^usv`a_S2Fo+51alSGMFy5#C=J)f}C&KRq{x32% z*u)&3px?iLU+m`tgN~H(-BNe$bJVcrbW+LA{Ht-teA}mPBPR><U8JEJxt>Xv>e%L%9h7eYx1obOdVS?a_Ae)@m=MCv?H2^_Qa&0%`) zjobkYQ!uG>Qj!K@I1(M6N`AO!+vW{!y+T@^6y-`0$G6>_aWc0c=(5wl=Sr(d0sbh) z8|7apmUNY|;h!2W0l|KpFiPiDSq%JV7<4U+Q)=f7(c<#X=jupO;O1gIcVf)DOK28^r&N8w5u$pC-E`9{ zhp_|FF<}hL(5IH&xdzOhp>@kX@AMau z;N<=S{53)@lF@L#qzuNhs);g2MYt<*jNbGDt^AF(Fpdqi-j5F~b!V8oQ3RwDB_KJ` z{nxWKhS58|x4=k_H30y+D$2D)gOAr2A!wt_c-W<;J0Ju%is_oxbuCG;@7^iIObVub z7O|iD`btL9J0?*sJNg%j+KVaO3z;YYKFxdK(1Fz>sr&u+bJ)#GJ^FelI0OaFq@{z3 z-d;wX!l$LBh?@l3sou)|ZU0csxf;&F<-=|u= z?*yNek7L)hUON6|DYUUcBtLjeN$qjMUZ_*0zj2e_Ey@tf3i`T7y)U}t1+y;H$o|s) zr(ykitYO`onYK{1?%T6B7mX>tH|~|zCv!0!{oj&(kx;UROa-iy0>I5`U|TX6`rGgl z_sOG(U){2teJ$3435Or?soC&g)A;OrPPj~iNV=wF(Z&9Oz0JojudUwjH8Drr^Tt(@v*=U1Y$XFFV+c| zLA$kImE@}OB8>m|iSSV@1k31k7dP-YU1XaycjGnsw9Fm%er?|*EG6gJdwhbLF^r1c z*PQB8H@#kLNoiB2DYrXv*vSqx`3}fJ2CVDiKGRm8ww#$+!E@l#f*yv$P88SslG_&- z#8iwSR!fIgabzl$2T_|zyE@~f*2eP)Dux6~9I_aCoUI&+mvF{^KU2RXpcOpIH7fw% zIFaM57Za|xSNXWtj(#U`W4SqUB^IMUhJ#wT7%SUuZ#|3)GyFTWiF!8;piIdbuMBLPv-zRq&UyS~6S?}Nx z@`xm};8B;?_a!t3C2pfaja^tKDT;EEnvaJCNTP0U&f6Ruza9IdKgdr-_*Q!MY{qEIZ>cO?rYz8Z zkG?z(<7|yFC1rmVM-TL^#B-Cakf)_x8`yQR(ac0WND(YU0ZKlrC-uiAn)$(Zk-ByI zNN?(q!ef?zd0;sEG(rU&;4EsHNnTe+j3(_bNI5nou>Luac`t~8VN`ls+{&R|tK_3L z?S`k)M{a%7%+VK)zH#QNfAz1BzG6lPM2zb-$WaNvipSDqZ{tMnM(1T>`{=@p1i3RzV2+hO^Xz|5k%2eF2Jbs|CfQZ2 z=63Qil5mWDkxa1v{gQ|+=RI&c(aZI2qxvT76LMTw@5^6a?i=3I#;j?v2g0Z%GD@EM zwuQ+zowCRu=Xvg^2mp3F@{a6A$Suhw3Q#vS7e6yH4~&&NIc{PYpC(X#+|RR#*nata z5d3TO^HP?d7GGpp40n2;ioS%Q+IF77%^IO5Zs|KCYA?zd^m2g)2=#ffJQd|KlEH2%=b+s_+ zEhn>)F5dvUb+)_3y?(d2d6ewjY=6YZ2JC4D|NR}6`4t}b=i%BY)Bh5YDnfG?{m0V1 zhqdwiKFfP%PrwIO&d*NqXwQ+e7ZV8k%U&pghV3jI`dd}CPlZpqHsHRZuk7#n>U@b^ zvb*oP@`v(xR5HC+z%-pcUdQ8`%Cze8?YVEh3AX7OptDYF?D~*zIJM{gX!C+9A`Xt^kX9aIs8n}ZK%HnPNDeK!FRoPzL-F}~58o%NY zK+G7T>vO2LN(uP9yj#M?y(>Nrqd{Z|mgT!=FxyrjI}lM%dR=52xk_ECUAHPxrMuE- z`JiD)^K9?rxS=xmwS~fTJo|79Y6-PTC8`)7DtdidrRClQ-MsEE@@#Fp-nXs)t){2H z*WgS!C@^p8o$4N znxOBfhnh29%b`N0(5*?!ZS~SVK+eU$cWZdD(sD%Sa$|gC1K@^RwyI~opO$Xm$%Q5u zMt(!F!TYF$T>1`LP#cEEJ{?DlpW6M=b5_pKH$quCqT8;ZoJ#{kK0MJv$+cFob<6J4 z3$6{DCG)nNa7vR(_fh`u_b^3Kdyccs&P{!i>E4K-6o>x4VBH!PlJ!l^ocx`t%_K(R z=1jjBCYP$dSG|mYS->mz-0dNWoNMoMWcQ0=o&ip~w))k^&XDD3Ak7)~@>RH5_PtGa zZ^MQTNZnkiA<>CWp02fSUY_MOUn^;+r8!S6^IGe^jAdP<-Lgm|m%ln@7~2yWibtJO z$9%ADrq56F2;D$cEy$|)^svNVtr5cKjEFz@++3Qvi)p81?0pasYkhOR;J8uiP}dv5 zPkQWGEx_kK$5y3XT6&~l@@{391VmUeQ#o_cG!MLq_Ec-R;*tl9>hiQ|WMlBv$b`p% z6nhUg{k{qPB(gIi{`~yT#?En5*bqParar9^E&?-LSD;>UOR0S_FWH33H|15p`*5Ovya}JJt^`!JLA_7(K1W;lB$bh z1hG|BS)F)p#+7gmEP0n%klz%hlOWO7!X?`*jQCM+QFbi>kCV&bDg=ynf$%;*y`Vk@iVSiQ}*c zjIB_awR3krj(vV3C?Xlnc%pAE?K$eA^9O8^TD`=SFnHOqZR%}{GfEDS5Jwf30ZB9S zw0ZL4{6}%5Wbz~H@>e5oJ9B)iL;mn#v-EA!1~xgP@Re`ZU56-C=`P7orCx>b3Ik}> z<{x(r_?>l)FxA<7i7uclRhD~u6j*TL!Whd}O$`Gtze)5WuP(7#ZU_E?TLk6Qt9=$Y z?@S1!=X~k*9sV<~4?5GWaMxvTMlvnCFU-+JDn%JK)VI5?nsJft^ITi0uBAIoq+@)i zK!#5CW-#1NXQKAcd>*^g*k{H>7fJs8Xv|qT6Up-I9+aU+aa^phzpqyJf%IC@pqdLlp3?%vNW{x%#z|2uatpoLdBZ zE3-4Gz~w7v8cU?FOFwsl0^hgJT#fO);y51dN$(vIy?wD|mYnsq_DIn*!{nMEJN`gD%TXB^I2lJ|&@4hW?o!8E`QWBX zR^WKG=Y^tE58XakVk2iuGT~?hNDrr)Oq`!}2C$FYfRIR^Pz503Mep>q<{DPrBB$Fn z0R6<2h&pWpApKa2{L%=Ox0$^5xnv8kNGq<+?X=)rDB2GbUpKaCl{HxqRDLn`&}Na` z=q(?s;eDSGDEGgCR9_bE4r*h0iGmp{!qnstNnzTPO-^FmVK?-L=ZlR_T+vj%*O0a+ zm6KwV zk0EpBnUp`UotN@(f-wnuoW~mruHp+E&fyDO{wB)ia1~m7)8=t@a{R7MOmVtcz2;MC zgJ7G$CjV?W(X#h>3o)m5J6E8?=$vW%$Dq|kv7T15zT-_#ZLiUwO7qUDLZCTF_tXSYzg&RSKR|{Uj zswVF1umjKaseCmx^0S#YiCpGMp6VN}?(FH$$&VD?=PLg8{!JG(;BQKbimOX_S^#JW zaVh?2a&bc$mzvd2^C)u-~h_F^a4529R4D9e6yVoG-+ zYhHC>Y5}gRoXYtlJW!oHoYQ+Tygi~v1#rDt=%pd+=h^bBev1^%a20`=Q@u(lSj->Di;5mf$Fgw{n74Re8ecX?Meo+Y zU+?|0K*f=5?4e<2;!)OdVOL!slKD&fUfYwIAR)?4jp05<;Ji2M?{wJ4t;Lm%SMuzu zAC3sYpPA1T1j(7*T2a#yAmHrmNxa|odQkR;qddnWzBjB(^kdrdOQpwE?8#i_%|PW@ zfR6oWyL$Xa9fz%fa5dzT4cs!cUYW_jt3j?GFJ zjHd)eDlI81q6#-?W9H4M-(y&yTHVLH74_#+)g3nnR%q#lIEecY`*UlMn3XBcZA+ST zclvO?>iC+bN9ZGtZ55iFzX#&Q=K7ahj}Dr=n*jsd#*4ME=J*5&TB_=65(qU;kpF5Y z#z*Gorr0l6zPMRGMZ3=raOH6$XE`gQ63g~SxD6Q}5u>hh+>VlsFTP`sPS&l7CfoDV zU6R5TC3M*2!Wd^2hqL6ljArNnE6%5eiAbc_;002iT|hSiQaL2H$x{pi8(eO^($KeLFU7U)V8e zZs){bDtE{}INz))@AmjagK5CXD!QEmW{Hmx$~(7Del23EfvUhPyQI-{TDnMg^D_QI4Xe^7MhooXD# z;ty_E@jn@|=Je}^XC_)fyR^a zRkXzpWbfk8d;&>UJ!Z7c_Db5>d0X-uF92O*C;%0epAl$JQmQ<**qq4EIlE4!!=_F6 zjgnxzCL2yqRCtU{U^QO~X)gE4c1Jiu$a<=6{BX)W8a?abhQ0sr0vKuCoRg!1M{^YE zZT7Er!atvY&XZNo5pOW&IqFy5s$RJFFpbWvr=BehQv|Cm;9p!+WE&v|^U+y*xtM}a z?^~)-dtK-oWj$LpRG=M|qJ}%}wF)ssx_Ycf)Zr1lC0O3Fs!0f#JWDS^zqeuf2ox}! z!q+bIbH;9OO0HNh=(Kjoa(}gey*TxBPnhO-D8ub1=Q6!RF^fh^S=?y30jd8R-0fxQ zq~(3l!a&P%i&o93)?ggO<0<2KZ5CgFO-_m_O#fz#_P}8xLD~b=cQ`!Tkzb&ti>RlY zddK0%-5W`)g&Pw5DdKHys4dZ4jR=g*_zsTww(FATM$>K{rb zQmwFA%h|EW-S<;;@+Vmug@!ue!lxwzJbi2fJpr>Na@-nxHcWHK409RmO?N3S!qRTC9UN zcffJ?Itv1t-46+dD93LH19}bfQ+<-+Uco=kF*5~Ztf*=-+Wm|zuKuuaDq_`pucVAm zVhlkNuWJILG&E-2m3l<{IGy4)VL@>fR0Y3|S2j-e=3pybSguG;Wc$VKHQ*Jm!@`*6 zl=X{;DF*gko2*rPEdLV;fE}KKRQdkhgXlCtW0Gno~tP ze0P(2_J)9dLDQzhYVN_OuoLGe6S3vvx@-V2qmdUN@f3IA^-VboI2koTtUPTF%4npWw2A{CGcAKhD&<$6PXX21o}W*nKTV*TJ$5KRJ@>*Fpzm zmn6yWL7g7W1yN3w_U1I($nCbNoDk0WDuhcUqdDJ%7s}6NWY?C=Ieo3Ju`_6`r4J)( zVcw*YbmdYGxe^Y zUbh0pT7RJ5@$sDHv1Q|1O)~Lja>dv0X9XRRzhjUUoz9t~6FXIO++E@v3>dz^MSI(Z z^Bx%9<$W3^ogSs>xFBP*X=EiHCXN+hq-u;cMzrWM9s7ZxVg$ht-#OPkegDYm zG|IDWg$B9#&hIqK&z)tJRTIQC@t#1UqR=1o<_nZOBB_2RTf@f1S$6xKgR9lAHl5^@ zGBYDY@Fsvjlm`md*WOcU)3M$u!zrLC1mNl<9y%pCI=0i#98GNLtJj$#aNp(4vOMoq zQ^h|+o?wMwd&*OSKm6j)1L(D@9~Ni$BtUQ~uSgeHlFH|W5DiDQOM*|W-!YNhq$8~L z_q$kLWO2ZzY67Y)=ll8V_n1uQ>zdT&oVRU|DZQ|#Wp20E1W?cYc}?@;xPcJi*S`)v z*6MHEe$qqI4c6MS3~mkUzEmI$tT+BS-&U+yXOf4Ws^2Y|KV}YE<}Kzc`|@l{7fAPQ z5-Su)%z*p342y%aQbW08eXkD#`dnR7pjtI@=Y!4Ut0#)qD}Xdt+ng} z|2S>e%bC10$g9M|9zW7z{C8ZBc@)&JgV{e}Mobt4sERXXvmOTH?vhPnE#;e3cosO| zVqkKUN44{Pb**8FY@jm6EqLy^P+<~eXkO!Sk6)hfEm6Qaq0kIGKcz}8HsfHtUFs2= z(%Ifp-xUJ2C1>E5E3^^;2bGw7Sbsur&NqRT8!T9_zAZm~p5Qf?#UX^mbL+(w(bu!- z`sv=?5`to@dl=cr02XUMFUvjPD-NSBZVZ}a0H^xD6T_C$Qh%JoEp&L_B?0`4W#z{Q z%>LNj2krf^vNl?8Bmt=vtut&QMv=f2cVW)5JfeqUOW*ksDN}hU0j>w6E5*chk(@g= z0;^Oai9dKd=RPQQmC~THccg}Z*p1Lt=<)q9jD**fq`wct$s3Q4^=r@qPw&+k3)R|u zb>RMK=0{D18n(bgKkC@vVerYc+K%Nq&f=G8yXF(T48Rh7M4ZY6-raP^EkB>CuJE}I z+e`w|yieyG`$M*<5pPpmm0$g6pz2Nu^uf67V!IQ9102X;rTy0$=XH!F1vdYmkH?Z= zjd;-<4>9nf<0oo}(V$Gk53JK%JK;s44pUTk9e(pNLhS*t6$YC>-yc67kW0`CIxbXO zX#Un5^=2W@h55neOStLtPZwQSZ|nqBKyXmkjY~SQAdl8=qPFW9^SNbB>k=rI(8JH%)RzG z-C0d3a@W95x*P{ACn^NXk~`xL^df7faXsLOMU1l56-vnw%^c zQw6KfbPU)kg%KAQz4jZ4Els+iF~SHm27dhrbP)BAkK7lq(Cbo#n9iJi^sV$V(qRmU zP?hDN_={3+36cwsqMFA--zF8;$5q#%vE{oo`~+F}9XX10&hir#-pQX(_l;TEG-Xtz z*)K$x23!l2#9z36UwSit4M4ay9+FoI)E)ie4@+#hk1)j+j}8mCe;^)Tv^IJ|${wi8 z#ae$QV<5IIQ(@)VL6?}2qdEwy_vsM7zX0#(pkKhcLe7^l?Eh8@bb!5+5pOLRiLmTTTyr#QX81{Gk?Ecz zcSU3)R;a{wJQ(Vjb2%(HYBHZyFn*}B54W`@C8ME!XNS9QH2?AIqzN7!!K%9}pSd~m z%?i4-x(>QiPQ(j|f&LL+mZJ4EmtCKyXIW}mkjLdnY_%Owht&toDX~_d%D8|IWLO)c zo*&iPP>%?ahdQaXOr;)oC)<+ezGXWuEV-tu7pV1j!A5En_>M-o`53ts`-5KdQ5sg zpx3q_q}U+8E-HOnonu~ny0O_mCR;xi{p-3}jX!cjIa=0fRM!nWj*qkg9o{^orm#UC z)xskj)l~!id6R|CC^H7d$=`sOb#KnEjZNVk?m+&{ ziXYuUf^!Xo;q00C!IjK8i5kBsbJmGM>U}M+agvv?CBdxS<{-TvBljolK>_l7@q8Sm_?J9K6zR$ZAg= zf_8VH4{a0*5r)7Vvc^|6l>8k!t60e{TTe7?+vsXwizMpgVHfV0pX~?tnGDxRq;L-U zxwY0qsrJs513Ln3j>hUBys7XA^p{S}*JxW|vHf=0Tlq{}ABQ}mR8N)&nR11T`Z0X` z;RF!HZ9g5?Mc$9(*wMY91^hP*O&Qe+^0AT4{_TDJle>w-j2O^!_#!E9!%mCHR4e3B zkmWDO2p?+PN_X^!b*h$AQMeQO9TTc@?Po{eb*~noWUQDa8Xr*2Al7w=s^dcpeBP!{ z%baiPUSvJhEBoT`o3#m%uvlw-@MB3BFh^35R+H%QLV|LMbX zoV2vZxrb{aFTpk?Dd-NK#P4 zZ|v(>vhb!pk*)g{}swt%CRRehLNo!4Q1YH>gVr+ zyL!5Cwmv>tz6vD$c7Ddn2d z2Nh=+#3O1Pz&7|fF<#p#qWbJUTv&*oQbO5NF~Qm_-Z{&5ZwhD0D^W&swpFsZWWMJ_ zF1Vvo_EN|qVP26xVuG(FT}F{rZabR1E~ALpfUNIA8|I4ZE%FvK4E!B4zW~bJhQ#x( zNOGn~4w7aP9X)S{zRR^S~dibGOnp`-&OAYHTKkY)^$OCTxs~7iOb@@SJpZ>g7x;*a^-FI%8!LyX=c5%I`OG z=ErxpnmCk+d%M_-Dc$e5B@?0X#-LMNq!?UN>fYnePyEMc?yV}j3!FO7iz~3N;g$G8W1Ai={83!p!1j=e(j}&(=f}`Q$gLZ zx%mE;y903&LpUd{#)qorZj~wkU6?Q5~B1l+q zA8~-tY1_W=dv==Dygc3rWHuo?$jBfx?%MY=q(5ooC6CTSG&iNz>Dm8Ixp`3Ydd-x# zuVo2dr}h}TbfWoSeF3$C%!4H{4y99QIO>K_J8Lk>v+Fb3*Bk-mr=e#Pg6?t{eeTV_whw7HgrT*lxXJM5eH!)q?9s z!NoeuFnCT79_t{_Krp!_tQ&9om#}G76Gc~}R(bz3NJeJ!18K)9rw3!PE>ukeCzh`k z&M%mY>ZeTtC$&0}Wf#msjqXGXaXY_!zgU*J1bjvA%Y`Epq#bq6agL6qRNgfb!8Q1G zd;+u|;W*zT-pNpTD$Qm8=KbQYcbkRKswcb1epWF|$;C>n0|PA><`UP@P}qW2_k?ks zTD>ZRe6PYx`E-!B?~L8LwZ6SkY~x{|XS|w=)z4QGK|v%;6r{|cshox$$tbS4YCYBB z#KyyCr+U<$iL;G12)=DJW(*bWie-|XG`B1LyBkM61$SL>R14;^LRA%FfMvz1e~04N zYc5mUcwM1E-erEjw=>t!Qb7TA)y^^m_$>>kJ^%o3)Y8}-O5ZzvrdoJ!O)YN2Xchyy zl^nXvViy!x`-nJ_d-SRFoa-*_NTJ00ujwyh23n27nvCBqhktGXukxBrEYvG@i+5Ri z)=Fq2QsRE>*AxeyDg~{fR4Qgz?Fv;g4P+hbK|>bD^>SGneWj;cs`bB4jx7?OGd$#v z$Sy&v%hvzIb2-SK{GrCfHISL&-L;rl=l&&^#FmXv&&P z(ch%y_g%>LuDO{rbu0Dwa%W_~Zy_;B!7} zC=Y=!fX%+VESlF5I`sJ6!bGgF&_$}CKwj)E)7x_XI=a;&l_BokPLWguGOr|9L}(17 z8@sq3`u^$PVk>xgsT=aS%G03a4jD-cdeXbpgqBtAcIyytAktWERCb*!AN)uXF}N!or%6Im$KQRk$! zX6q|7T>qnvW3Z|!g({z83!+-M9T7U)bI$5Mu+MNqKZ#8fQ?F%z7f6?VnpApTbQ;(! zzbbg40zMjva7x;13ZC7p8Ih~T148Muy1}o|sZo&;?U?(pYCHG6(0@T8nDE4c)L89( zboh)y!gXwUw_!VDq6l_r{=Z0G08@tQZxquna>68;H}OpRdDKMuVDzU0fj76M!{B{z z{U_#E+{>TXpW~_xV3w*BfugnT3Rm1?_d<#gDZlCIRsZm3ik70!d<)A`KK)7ka0ycR1hmw>J^qfl`@)u9_C z)&5$pCU^o|@!rAm z(b*ejPyKzPFYdfHehB%TkV&?l$ofDC)c|v#CUHrG2JfHGM^Fw^&(sN*{v<6e%#MCk zwl8Xiv3gXRupL5EoTY)#b3h2Hf;%$YUR3lmBxEcg#&8_c8o4XhHiy7*&-Q9&)-KU; z;$NR-EQjv6xp(1Vwypm?sf6M&F5ToXzuH)p>`9RBj-oPN;Bqe`5PaiNr_nl7O3K9p z^0tXe@D?JgtJI98H$XVdIvOY(;;VdBr+K7!p|Aj z%33xpYUUCfik)CuWnIa8Qof}f=xa&w1E*Z%_2$hVP<&&uvACVzjgAw6rlJ6@nk~~n z=Xde5aBmLcc%AtwUa7~wy;~tnZN6|9U06c>lJ@pt<UZ6~^WKEgWJtPj_i(U$sGiDz?vg??@3WBvibUVaxCdNOC#7|12@H*z0w)Eu2@ zCmA9_k7nH>3rml%SnkGtHM?Fr5`xmZ9(HbtmjRdNqA_L6AG8TcM%`BU#ba0TjqKFB#}S^UO(lY~%e9J3^Ud)3 zT}HG0FpWv)y+f<|d!b6U8@nMXY0CZA1~s04V9RGp6@X^}V9U40tS~Hx=ba*g-+4aF zeX@md-ecY0o-8Hfa?yxvD$`YWm>)T)`8UMSy%l6?oWYu}$YeifvKC$>WeZNmGyM8C zFY+-JI@#ardHZNk9K-V$3jv<)+2UD=^wHTm#e{9+Gej z-1kvsLa)+V3wf{NS%#KcjXtwwtIpRG>Ts$(SilszA1$*;iRtHohj?wgN;wiaw#JyO z`hstll)AgZh1BFL67YT_bI}8K40O>UTEnW{K~~#D15noAH|%g0i+v2H4@Hpfhzg7P z&Uj4+$8JPa7-8$?#{_1tDtCb&HnKc3gvq3W)!eLDdy>{@9_6K7AB&hcjWBdL=Q7uk zFZNVUi-H&pXqZYV31GCt>0B%Zc>4m|0JvC(G~Ijdg&WX85x1-uLBR0-r|j{}4y;`N z(fyDInvW1-ZhKl2P6zc4UBXm4gqNLrHS8bYIJ!;+>-T|E*GCCz622(Z^p2=eCnen}gfxH^xMS1d1*feFi-O6AXs zOxjQ_UW{rz`B`a_Qe9V@TJJd+IwHy3`$A#){M7|aXR+?Gw7l;A4(@(eiC-dHDzS;v zi*EvbLw>4GYiBOH?PLle@eOhSDAa+RUT{Cd0ogH-$sqQvehdpp}&x1b+EW5 zejoOFxu)}C9v|*FW%JY6*C$3)CtKnXG7x3ToTOuc^JA8o+sJ}Ob(qq_Bkf;H%{=oh z*mUSd{wwRAHm$?Li3M=1G0}%JX@_3!wIhqH)E49AX%ETI^o5?Uf3+eu>YeS4L^9j7 z1(!Ox(lAPOGGmR}bZ3=@vNb63*yyZ;JLg@(o6pSC^sFL+O3;s>T2f~k#1)=-dzVua zX9Ucv9VMh)3YvXG?&WW>2VYfmIT;}K*aGI}g9*`)C$hRxzoHV5QJJ{dQMBQlJ+m-N z-;3c=d!Dy~3r>++B~!59U!|{bt8Suk=eoZ15&=*Dh{82l{WmNM9}g#S z^DZCjpN-S4O;$c7qTS^pBNbMom6a|QGcE(yP86xd*%@8&e&2O&YbI4kid^7uj_x)0 zHX?8;sX6^@5T1&3%IRt6Vbezq?jgN1TAn^J^FE$zQgS&n<;}C4YfKQD$`LMX`q^@M zf&JIhWtg<^oMPTEmvgw7K&f9hn+9GbiVXTbjCgtnaJG1gUu$a^wCC{XeC!x%Mm*WPu=w|JE3lV51VjvBbdtil05n&&Zs`TE>>l#`|0AkbM~3sNSoH_eIliz z=^IzT276lNM5m`SC18u84e+tJgazioN1)xwc93KN`wozKc<}W8u$*Lp%hudMXfAp@ zU_rcv*6?VSl#1&F$dM=@h@`2fh*=HB;?pTrfVg6Ri)Wdu4OaE_XoiWJU}_3QP(h9o_e|RwKx4 zQbTqOzk)Nm*Rq!HVOFytTMlB((ZZhciuV+uux}@ldTXgkraw)X&d|`3))f~mVa?R7 z*qdjc8Aa!%x7690#QHg6q5Tv|8}r&zNQ1sbs${1*b1jA`$=_dlWp*Hu8=h}i@92cs zKi{h{7%=y?2|PZpq}0f{6B>+H_dMva4&~|Vc(s9qGDFrwRx;^49dzJu3rz2Mx!a`N zDSf03VY2RPyX%?b`r~1(<+@my_oNfAy!-jde%cx~>vw4XFH8tCJJAO$Q;);H>Kfd# zQ_UBkC<#TB1ZRVTYNZJ!+ImEFF6ST?=kA!-A&Ooh=++wh1eqPyh`N#Rqgkk1y-G8_TP2z8LQ&C``v_iTSl`u04f7JpyHdZJz zvg!A3w;5S;okOg{@J+k?Elc|@w_0eExFH32^>e#Ysp2lTxFBh{^?rLg1bwgBYqmAr zE>H&uQ%5Vq1>Fg;D2ZLY9QVGfs}*}pK0m^VS=Q%FwbmgwE+pp*Tb^3QqjsV7R(}1%*i~H-`8!4IvyD zA65Ps*-fnHrts8K;FPubHA@u%=Z4;xkL_E=5Jl^&q2FzH&E3BPbrAod`tnMp4 zC6YDNja-qLh*89aX5D!Tkq($pt;bvL5st2RcJRA$H=iB6VTqVd2^Oxb4*>)$`TquM``k?N*x1NhAJO<4&YO(;{O<-brQvAVDmB5soZp67d_ zVYB3jDqr}BvU(g%h>6BT=vCmWiB+#~|90vk(Z>&dZJ@%XHzu|jcLkiDuFJDBYyPf= zX>2}l;AWO+=Tzwyw2lo{5V4NXq~X&jZmUHqPlh3NoA<-=(FsjSSJ%Z6Ir+TuKNZv^ z>3OYAo7I5;M}ifc_n?6%?5H@#1ddlN2e2nTl0Mm)cj*;`py-*Yt{#Rs?X5W-o|H2N zE4(+c`Upq)Y1_$=p&B2(6&oWimXf(L%TmNiPnotf{w?2nVH6DfA1-E`gdUWCDjG(rC7aLY*^`uO) zuQ)KaFC2o6L@szILa>ahy%|-DDe!wlO%){NM$e!}l*HS(wI$0?W|K0kzYekf&>`E~ zAbDZgsNiMebj!ng$IYG>E+du5S*t9|SOzJp`AP%2DyJP~-aQw>X~Tee{qOI$X4yKs zZPW-e+2Q4-xV_v_JvS{UB>9R%MKh(7qIOk)n!P^enh|5GJ) z2g6wOrqsrWQ9_cg8YlsI7i)_A;s>=%pbtq@pRqHWmC-)Q50#|FD*Y_@1iU(=#H6{>_U zZi*IE$)@#ntm@$ATD2CexWD0n6xm&>51(;-j3QvvpxE|kN{v+zc*p+J6o{ARbB>yH z%2YcMW&(9lIg^EC?;JxNRZK*7+=c{g2_l_V8nyf7<i1yv+Ru2cOiHoNu+4-^tXatiY#p{#AJ`w77FSX{E)@0t zsk*M{>$-C!tRx+`e-Og^yl%|@h^>|LTX04irvE7V{K9xlOz1B??nWm( zS2w$`RfN8lSMrs97H0F(;p{c^r{);t@41FjgUbPlDak#c@^`Jwo4<$md8?n<bi-^2!D$jo)teb1*3kzheW4+50fjm^Fszm+Va^%yP296_hd6QFsc zzn_FCPDcm1{U)Q3Nh~4Ipqt4bI4?N2VO5&1Hqu?ARW~1K_-uEWzGFJ!WjG7R&c+UB z_p9uITDNq0X;_)~uf-dmyVzP@g|ObAsRE}TFFB*nET~ztsMBuopffjtbQfurBZ*aZ z1_l$qRkhPVojQGTivld3az?j>D6DNCeKP?`#v2a=xP;*G;E2yS)J9D1;y$M)P|B_>3YzEgUrs=sYF*-tsB!&35E~+6|#=Vpzu|O9fMa`Tn z&s&Ns@UvB?MA1pvQUy=&$!T0e6G?6Gx7=z1&a=EO2CgFZzWwYncW-i$WRK&FR63oDrk)J zm6YxWs_2Z1S=t$+m?p=b6FmMEZiqn;iV1l?q#$}3S4}z2)QFVMre%>E)xfw~{i&p+ zRtPgI)quWfaQ1F{E-m>8^RC9)6Vb#{I2ZoqFkpax=bAfIEt(?MFkmVrDq4r+GHMvj z7Q;c7lLd@MZz4bIgd8l_ggl)rB4SOVVYgeTb|rDRVt{TcYO)BPylXlbb8i-W2X_i5 z{zH|w_Dk9{?bfos&nah~1doJJ|G=?3G{_QoY!|%tj4gkVXG+x&kGH)HE=upTvtVL- z#z0YzZ)#niNO*OeHX?GaGf<#STiv;P{|*cxo{7H5DZ20MrM@JoJ>uK27+lieI6Fft{k&ulyEXivHXER)Fi!ogHNCNNyP z4=8Ju6}@+fT~@a@uQEG|F8HTSK0U40Qx&tfvP8zDIG~*s@*$g?9H?aP1+V`yc2aO| zY+#UivzUd4scL)IMSmfJp60mN|DGg4-HbeSsf1gi{9cZ7P))RU3q{^p(%fc1FylX+%}Gzo8m8DwpHYumTtla}{E zEL+Q&mh_x`BY%8qh$2AcznJfK|2Xg4J$BvxV%;_po|==&+AJEHzET)7vqYt80At!D?Q_8y~s7U0a+Eu^)+ z`&%mt`}7-LJ_%6fEAYY)Diy5Fm~^-DuSf#Er#R1Uv-6@;p)y-%v+u%o*Dl;~F_nWJ z9U@Y0T4}Q4V22hbs)BT1L-ZvnKEMG&{}D7lwt9}l1v77EZzZ<{b=+lsEp)tTTY!4!D7{}>7uZZq=Y9t}} zmlaS6kTb0tzRR$%chf82V*lJbteaZHq2jU4C-M;+2y3@ugn<|dq=T)MIlY;C%POlW ziM}n5WoDpePcSUEGL_9Hj%$2EoVgGLF?fBK6L7u1vR1(mK4AC#09E_KhVxLG&*a7N z8PEM_c6#G0%uZ+HrSzx*y-AUw?dnZ`Relis|~CX}$1bvX@z znHpCs9U2yO{vL%#gIW<{Kg%s1U_JI)#(7R_`N-nJpWz@$9$g|&_8A@FK;n_RKR=K0 zNj9g^`j2&RQHbK605IQYs+8BjwWOyz?9)i_Zh@!=8U7(2t#Z0bU@w__@$SzCkaayY zCi&yD5~o+BB~&`~N5|Vh$uqf3(6QBOt)6!*!w=J^G~b)cK1P^Vu!3B-0>c+>e8plv zn9wF~{bsm&1Ea*6Uj^-MhE)wNEx8{heU=^0 z&$VTcrKDM<-SPmZ!=E^gyEI&#&3{f!z9us0#MR_rJel(#x_4+PW0HD~9oF9rwU9e4 zf|S}SNL{`mzZ=39h5y?~duqsi?1^xp)p_?X?-oSnj9zf#aE{rVUl38uXb;85J@R(ALW4n+w?JZX#FLhIIv0oVp{C} z+q9=v)@xjOI{7${B++|w`{kn)f3kc*I8)r?_2Fbn=#D9UH1_6{j26#G`S2n!c!!E~ z;E4>;Fi~+*1dVo0%g4C;16#b!xhnhgsOYu#(mtRt_W_8(_XlD`w2zIpT{Z7*$!e>+ zW~a!HJC0(Is4rLaZ0|1T+WZJkyM6|FIZ~pvyOV~tWbmodVkq!U#hr6dC~=Zw+W^DE z$L(3e(yH%6xlgXv67?|R2+GHP3i5sV@M(dY+;fdEaj>#mLCTY@i>0{fbNO;`YDQqn zV4^AmRl=gD%$EPN0lgc?zfQE66U%_+G17f;W>&?{tcdYf*W)Sag!ISD89><@wFn%^ z_`|i+5v}!7tsfveAkVOur`g-dLM&~SF!z;JlDMVjewkkY2{i6?|0|N5(lb-Ypvc?G z5JlE>7kth>DPRI3VH`u0Mu=kRL_zz#bLk(A@Rw8W*Cn7T`MF(_y0_*4aT=(K>~()< zOg;j?UQ{$v}5D1047@ zYnEKgzuv`qRxZE?o{TGaXZ^Jn1(~mEF&-%!amFCguJ!Jn>hITA3_6<+pj{l?%j;u- zE7tk4=`294Ybr1nM=PW#B~L>-Qzo%2hOGH4MSG)=@MYbgztbC06M5c?uhQPG^#BQ2 zywV(YiL>uT$JHP8FATBPYq#X@G`S<$qFKeIlOpc=)TtnTci`1z2iA6Fth|8IadU&- zm${OoMZfPy6f<2K^uEU^12ZOrTm-oZtZ7UESRkA1r_+Jcfp0RzHCJX{A4x>jaCTr& z?w`4oDNffAG>iYO0Y!O0c`?W?&oT-q)wG zR3Sk28)n&z#ifH~fWEPn94mKFu5rRI#d7}9dn=#-hz1b*{>BBmEzq4E z8}%x14tTksX$0zI0Q5DpW?znQEucg*6n~e$Rpjc@LdmK`ORaPHD{-2(X&n0Aku#QL z=Yw#qJ|DlblviF>@y0Fz0I9+0Lt#vJIFz@VG^xuxvxm@H&tqXfz1sFT^$S_V5l{|7%qQ+!mCixwdS!D zumK6pm{tn0JZ`xeVq`ICWvtdn3FBrWmMnSc+-jNiZl2=a+;M$H4M>t+C+MXnP_Zvt zD|8MM*!s!;s{SsMx1L7h`{9YZuhkz{!1GOaHXx^NeI;jMTW8FoThkljnw5?L{HTeJ zvFXoZw>Wu=QyhPh0)>cv2;0_C)j3j;M$?jd@0tekdPzd866$A*z0`*Q5vxCMh%H&0 zu`m>puj2EablkaBv&i*(&!CN<-GQU4@i_kac#NdN`TC28Z($Qn(5L7%202COglfqq zEBH1$UbyaaVyQwOOJhHSrjavut1zFL&=iZ6XmIN3*AEfW7m8=z3CXB4U1~~hu_lO4 zwAj2qRA$kSHsxc}zJpaKE~)9?fgwBA_OLGf@wSTYcTpYn<5%qa4f49IX~m^iI(F)I zovElgA!5_RnA;;uR8msK_ZD3MP}5f_YoDX-0|^TZdRk<5FwrMABDc+Se*bi+Ajs5V zQs0Lq=*JYfl?bAi?orG&R43x>T@dQe1!rteFb@9gC9nC;K3l#MS^E$1W9g1{Ak-oOWE#E{ss z9_XUuobai`!AHjia`Xez8dL#>cFcp3Cs*AvalG(gD-v`A+0hy+8mpC_n7;v;&?N6Z z*$C@RG$lF1#1IH?EXI<$_Bja}wu%=!J6S=0HtMk;wc2xp2Hp_e|oGiEcQgk*JGSs_?w13Mo)!a7#t6xg`jCwX?yO5ry)F8S~Wt|P>(GTc1+ zyK~hPjWX-~E%LxzbX1x4^0bTo{Pg64EYhj>E1ne*xXeIe+FTb0)cG!7W$87VXw2JlqtW`&wXlso$XA@pvis}18g=lp~wNk5MZ3>u1 zsVy#<5Ve!4Cay=ZG2rPQH8`b*a~jO7+O@3peytzb>H@8C z#_BD53|TeG{3k5JtX#XpQ_@jeV2oxC(tpK7_QQ)CH7>lw&xh1?yqVfZ*i}O$RE)_uH?Z}lRz7fT59l97uk3Ox2B>hjLuJ(;04Df%07H3W?V3CR z@KNt!2oQa;%i~y~?UQM!s{^He8IvGBc>i~Ts(mDYYHH?oaZ9Q@n!@BPsT`Khi zL^Wo)J4Cu7CA6=s^^(!1=`7c)=0CkUIgtLq4AsTTTd zv>U1!pIJn}eyQMa{p5bP|3%4~b#&E;fQ$&xct^9r5E{(%APQ?sdXt#fgg)&AJEkFZ z>2<1g7jLpW{TQ=a(DYqhnBG#2y+Au3k?6_*fG-0*#I;G584p!+$4D*qGYgdBPSzo1 zLt?-1S7@K{`qXWvCCKGo-xJR2a1>k&rgimu=qQB>uUcW0m?!|zRASaTLDq|Ys)L4w zH=ZAQszDx)7f(A3=?nYrm%7!6(;0@}J(LD257_N+AJlWS+a9T^KSPb{EQ&%w>Kvb^ zm%ex6o)YPp>*!7)RS|h8O7&k|wNUcs6Ju zey$NY(9u2k%|co>Pjf9?g)kE@o_&BhU46i@?~;rcGVz$6a#_%56k=VnZ7OySbZ}DY zt}a38W*1CePE7o`2&bSUI03@d20C2^ZHJJu)`Y~Gy{O>WJPftFuMF#z><($;4tWYB zMLf`S^yEv5^--Aybm@NySJ@US35dDG(h9u*tJD#SX5S|oSPU7F_qb`?B#^B93YdP* zc!<69U_TW+&y6o+vt|68z&8en?JvkWC}?5EdzS)fi5L>ZG=F!@d6Q>VB_9D{06QrxJJ_clX* zpCF=z&0&KrjQ%`}6;|#XP=;cjbR;Y12-o7so2sGT%8F9$2XxZD z>E2k%31;DUvyoN-HB8aH_wRrjjEh^7{Z7$wT=8EpV30ty`parO=?3amPQNnF{$gJ3 z6*xjfXjFX4dZfc3!SQ_SnpE}q>&sg4xk{pm&2lS7>W*+WqGCPlRKC?JDGxU*ZrGKit-j)S;Ch7 z;|G4mKk620e+ee&0-=^kF7d&xa)QiCDqI5t3|fa`u`OE!lePy-O?^U~tuQk#j{4Mh zz|P@uRGj=Q6Xo_oNt=IC&^5AmXT`wxID!!>=Nc$b^G50pk<3(~seDJ6sp9yy+3x`5NOXpm-^wCwW|{}E5`12q1a78h={0*_?^HVN zIR9w%sr;!3{er!338C)9GVs?VYRWib)q%WXPqR1cY{1a|g&#DS&vqVc>-Kft%YCahMCuCzHo|(?_K3xACt4*7U@aH@V z<5J+YGA@HImM^Dk|B;M_3}N+%(|phq1$pjyssR|yuljG_wvCBy7R$P2XuFPkIjn)} zUl_f}Xus5=%0#q%f|xjY_u_AjB>PZrXeFCKfIIadZ7dl1Ju*DB_*DQVCu|}CzWPPj ztTu@`DMt;hrckhu!Ge`+8E|f8^GOV-XE8i?gXv5zVq@(f(fdYH?o`|wP;rZB=EfD={;fcDj z?8}Qio8JDW)6j3+K&0kH2Kzv9QuNiKKhk}?2&6hwgLtOIy5HDZT6*8z<%EK3dJ#<4 zoAL5!bF%_tP@DdaVRs9}(W3MBx|}SdXeQ*lAx6ii^EwwIpO%wlnZktv*Vi1QBt*1b zX=&X2%P7>ElJx$y4pU_Ht-9*j57)34;8c-wKe7scE9Gju1+q@@ko4HJX0#_rmSeA` z9Y|v>Cc%)-^u?I-3MHp>WJ&2^u_{x3dA+3SjBe&UR$lXAWTMR zwpZE-Np+mH*5uvO!C2jj8n%*+43J8bK_2s|LaEmbbZUGhjs}DGnj24RNbN>sW4WHyn>-F^|SWcwFyuSZiqN1E-kZPo^%jt=XdZyJ`?evHemQEc)_F_v}I z;*Ik1^ng{X5~H+r(8BJYVGxbB%^hI}NGJXHQ#b}uGy+NZTwGvRjvGLWENotaDEC=o z4Se8OVx2us8W+id6B=Q}dC@{9-fGvH)lw2HKLSEula2MgN@!sCr1m6)Q%k7*$vvX_ z_L!>l)i?Ypu_1~s8|5=C1{^_@CNfs!2&Ham^-*4_)9rWw833BfG#ER32w;EPhyu`{ zbyEWT4~ZJHcFF-vw%Fo{>*18BKSJi~<%y)(AH5vX-Wly%^+;rMZt0Q3Km z_k!bPP{!HkQ>t8a1qJ(=TaaR^`Il4n6Bheh%vV=vv6+*BeaR=|Xyua4i=%!LCh|Nl zzG!?$H-`2b!LpI+P$RtC7@%h@>)V=Dx}={){&?^pz~;_fT#>2|$+BsjRMK{+5Vx|X z1AABFRXii{(%;&N62t0Mo8k+V}f3$p* zF!NV2U?TAj&o6i_tZ?hf%|k}+aD9)wFj>24A4cND;Va~8llpe?jZw#{`5PsG?2c?d z!=zQSR<~rV=wh>B9V48~R>_=zglT2H26$&zBG7NkWggoi_oO|_32K!GK^ifi+N>-t z-Jm&AQ#dG=%=-uNdQGrj`y?jMU_fshTXyMS986g-Z@seH(Wdc%FfH*5AdqcK2Dw&x zqu(~bxHgm`tomynb{Ye6q{P!WW++?s02|;RoSOb>pCx;KZl98uE}he*w1|^Q-IYfLF%_|;FYJZfqQyUjY%N7!IR!&hJMK4yFtDT!*x* z{4KdPB0|<`#^(L3o>*;W6z?|~Zv!Y|;!mvL`+Gk_j7R)DhGfDmZPxpZHv2+u;*&ua z;R!b;0x@MmX2&Ex*TA-rFi4VLd6_cSeei-)zTHsFCO>&mfQWyUPN#TdN4(6quioG~ z_-VK6_&u)}*?PPj?fl-oI&~QY-$l`9vVwtKB86d( zxSl{(e~(qJ`(?cVnVKx|1CgI)>Kz<`Us=r@v>svVbXWl7ndc_Eqf6QMS&=Q*B1W?L z8sAgG;)RfcGYbaYqP%0~V~8)ye1RrJjwCr{+I zbRQjxnxibdyTctAS1PRe#CuTKonh~ZCt9w~NKn8YRe1{R3Gs8bKu7d{JB}0^H>^3$ zY`aUb*1P;t!gA^U5m#vBo`$sL?0t#X15EcZ`TEgm%|*{njDXIfY+)T}-#R-6-c*I*hQ6&7x{=zETL zJn>+2>^#b>cm6HHH1DP1$}!Jni?j9s_WB|tc3T}HnLc>1_M_=q6qy_hS6K6SVv?|M zxFT0`xuYkV^!qcyqcBfO*7;P^`T%Q6IIta`enUS4b&X}rSqN!z3~FuXaK{-yLvv(b zACFZ?@Hl1VC>7eRH?H|iye+5Uk70x|a{e)AXllXbhpeKmm#avLW$Cw=!6ziFv>^6V z?AlMZdXE);?c$1_y(MH8PJ;L|ax=p40CZem zPwp%n8;gjUI#A34gJ+uyYclbLnb^{#T*I$Af!6!xGY(U_v7)oIof!l^{qKJ7j5@bh zRlWxqb&&4S`a3KK(+$uN>n2%U?lqkZ!4ab@hWw`hui-K0>ogDbUK)0J+X%)A|zlCT^k6#in7PD?Dl-6C!i6psX5VrSba2r*6+seB^>J&gO?f%$Y_4Cuvx z=@X*(?ctwMWpNB3gH6u>lKSH|QJ6kdCLL6S51JBhu6j>%r?y);Z#XW0LpJ_!+(Mbz z$1*2SPU{&s(DvOW&|Rp4bJ^bVD5ki1hA#l*bu<-fj%MK|2_-?jJldr(12pkk3xuG_QfB9_?1TSPQ z-?S;04Dy;v^Bk*C#yyqRofhkfaD82*UJ|z@K=#)t6(1Ny6#UW`XnCiUP@=;UDJ%lI zIw~VyjZ>CSm^%I#YZSpx%WHd({P&=`6b3-k+X~8~99N2&IsJ0;Cp;A3R{?jT#s91a zq^6F7*ATvY zi1s&j5kMkDyCD&(0(4@jF~g*rhqE<)oIyDreX4t^cJnc63O(o0UkYT@e*!$+^XB_r z=X<=#LQV;?UpQ}iGH)bN=%*U4{%U^@Js{;MTf+F)mSLl8!Ps1o&B(Qg0Vp;PJ$0?s z+&D#8j1j~AZsJg&pP#@i&Gz}ixphw2$l(jIn~V44GN`VveF5qbwU^VO)pr}Hs-f7m z*Y`$n*>TT~D>@#nr?3p;B4ytH@ts{e$ur7Zg0ZD_&q+h?-fHH!l|lyvzOn_5KuM6!|E=iv1m! z5^L2Ba-p}lAj9JR{mhtLBLAWhm#e7<6$OVD630VNCDJS?}lpy&~vzjT5Y zUXjK2+-r<~t@i)&BLCfp|8B&865aovi2p*9*d`o#A3kK@%r9B#*(-*ZOVd?39XQlw-y zt%s36KOXL7XJ)==^g^b#275tDVL3K#2aRUePu%w$Y$wy;WJg_8$es=?HiC3do|ZZ) z69aW7FO}oY(iBn1U}bDTsGtp1vi8u z<{L*~yI=p4hW*={WrUyt#$or!E4-rIXYo^=GpNc{)p(g@lwsM z`^Cw~O2RB6q-x|;Ph)4=j4V8WNQ=D@)Ic?Egc})_o1l}`e<}a(<@{d^$7))@a^FOk z7^08vkInB2OtB^eh71L1cK+g$57!0HXKIK}>41&Cdcq~APh(^}=*YYlOW%UMrA76K z{09VRym16}H(f6Fm8Q40M1_H2MOOK{8hAw%t0}iwq(uliU=UmQDqb_CzRZaWj2{~j zI`p9OLXLNSuQD4sS&daNNHU)1w%G2=7n_4&mWXrjvo;0%(t3Q1f{^U`o z{FS+L!$OOD4?}0fU<7I==L7HCynIU*;8UQ`p-)u19Sbl1ayd(0FbO~3Z3fnN6@R%7 zUm?6v3YG%3sybS?TXGe(wEvO6($oFP$FHhp(pM22*~AthxD-cgluzfY_%vAbpB#)E zhe}Tg5NscOe+w<)Ey(`lg3Dk*Sy|-F%oo0Mv;6M;cX*7KIgPJ`4_Ey-7Ca7l>I-4& zE2#&uT)w!+WNrO z%eB+Z8K|X0V)diHN%?7zcb$-nZ(y^X4S#Cdvf0j=b zAe_ZS%M~wZXlQs+?$+9oyX(H(^=^OL(tD`*N8p_E9LKPbRS(m_n1gtMN_)^hzpB_G za2@}VYb zo9Vi{lO>iHNAVjeh@|QSI6AjOKC5m+7v+K$UxKdl*3XYW!VMuTf!``2<54WmpTFBQ zCC{Dx?4&X(q)AM2eX{WnJNqYqYY_k*;)na0U&$6IZX68&2J9FT+GnfvmQz;HrT~0nv%m|UIpgbT}~Dd*yk48pK*5Mk8>0ITeSSzP{S)a%|F%K zaoZ`ypVD)mzJB)nR)X6MK*DgY@ed!ZS}KjEt$ITxVKR)M}<$!|n7+ z^V)i@p+`^8wWZr#^shD7Ma4Tdxy{c8;&9xS2}S;2N*q&y3S#!M4Q^q(0o>6J!jQPAZN}R!S}n32$3q}Dy`$R zcQk!OChXQFevX(kTv%tlI9s1J`qoR%3pP)vL3p_D|1^aRnVC=BLhn;3Hs9>&%mzQY zYICmD`fUVpf*KH=;u;DS?)awq$G3RR)f*(#R244}N~qz#)z-w`M1W>|6(>{<)}{Xw z^88KJMM>SPqbh3VfyT`G>swr)D0lPP7FZIh%Ka%!{|K#OpTx*+l)NgT|A3Ks3jS>_ z;Kl$D3c`Y}Do0~edYg51y*b}%s(6f%xmNPGx#0g|BlLg1%Qyb9Jzl7sbIs%-9r>ds zc?9dSWmn6*bv;55n_R@h&4Jj|TBcKa%}Z-r`-2VK{bzrEf?$)^Z64@8dNtS0Mu?B; z$5E&VLS!Rf-C{4qGO$t1zItid)#i7UDr|vkcvZG=lyZ#{suOv5!8%W>>MhMvNHeu! zzRTa{ZlA=afxX3uC(RrvB__QMqN7*K+&s1B{mdB?D>wRo(zAaYD-$eWcU629O{PL4 zZnZQsxi>P+?5ns}$sA7xS9>YvBHhWn zpxZ=%DHLmf)s>(@S9NaQe0!^h(!DVt1}V`eSS^3Nx(z7d(Q5;%oB7zEu@U4*eH&~Y zWxc@+lSlp*iO{Xg+nA~7aV)^(cbQ^VbdLeFm?GpUwh+~1hE03$Zf_$c{7#u#<_FJ-c;`Y-qXSwI&dSUH&m zzlRJ1#8gJ$y4hxxf+`>p0P3aJ8PHW5@Kundx>+&@)>7oRI#8%SXJ>&)FLOl%hkgtz z0=J1-r}<-RM96qwFui|K&>^b3aGT@Lg%uZ=dG_rGmxZOJDST{c7&SA7W>5yt0=5xb zwKq66wd90INVggkfVDE>%KozeZ{9(PrJXV>>))Blb@=uJqK)RVJV~yE$GG1QxS@`` z1@ppZoM>-xf(qn) z3bDgEIA%TadDi#lvR_f%oY}-6it(h1bg2g2#MlNuL_gFD>9V0_czPx zKkZTxplXVmOo;s%M+%ogzKVqS&;D6zDHP)Y)x5^i2)w*JA@Z?*QPSw0e_n5&B0wK7 z+DxNn4mcFmB~qw#`#-pa*{O1)aa{*l(N)uwyPNN4GSa^MXE_neU<5#WK-(Eu%CC%Y zK%;PE-n|8Ms{(3Kb-ofyB7Q@~j!nEE1Oxj&E#2SlRv`Vymg)Zge9P3`u_WZa@^OH? zPUDxG)HF?UH|H9``%NWiWfyA`Cmt6Y(^go_qr3^4rdu^Q->uK|T)yXW9HgBa?jgzu z0o%}Gd~&iT#q5ycAB9Ei2r#?jYuEo2XGk+@=9|`i(~nhq!`-f=b|>w2CkAN5BeVAG z&bxIQ;7hsemKKplD@X}Xq{Fa4Hp{E(l$NqcFRGe3>%`?l9sB*$iv2ev1S6!kI1=mC zH}bapuWv@nB+~hnvfpKd>ybMbS=zdkXg7Z<%Jyt1m||SU+l}A4@v?N67dc~8t6`o? z7}BcOFaB&>%e!;P@L-i;nw?4tMl{IMC`-D4Oaw|&%#?{9W$ z%}ameD?|A)e{-A5-HeFeCqZ4ySHIyc+eZ8m1=s@Y#7D3!$>P05CGpe zS@grr0^IKekXxaK!ReY|+ovm*tE(g43(m%bmS;AjdRwn94Sw%VQ?c?PiY1WrzY)__X1h&4j=PW7 zaq{F-zo53yyBrRkOtoH-VzJig7gr8QR}2g=bxUBe{DiU0{0%t%6Z~2I@wxzt->3lK zbyc2uJmWZMIXPcg2m!iN(WSW^L<7OS!NlRy%E7XT!LmAm9q^HAVo!U}^JG$QO}y<~ z(dOr7UtQ&!uFf`1y^V&r(YhlQPL-G|y1=IOEgjO3kue^JP8qfgpL&S+ya|?!0UT5UV%Oau_{Y_QP$u9K2nMY@Y>OnMawA+ClpG^U$Lr0p2(a+2w8n5#P3XKl`ce$DuBVwN>da2S7S)AGjtCA3BTqYQ4%wDZPpQ#pys(z5*^t1@4hw#})DBQV#_ z1%%E61T=?u=JRh!HJ8qy$$ z;pBAV|6%XFqnb?Dw^3(oD1taDBHckmx=8O$ML~*GsZkJ+8j&Uh!XStuO+|VMNUxFJ z6O|?)z1L7A5L)N~Ldf@IX8i4${q0fCS!bPZopttqti{UvzE8W`bzS#!dJ+(T1lyo7 zmXVHHgqk2KsVQg0YRQC3DQzNiXD6%R_UdtzMVKtlPF|_)`qC&b@QHiMm)D`( zRHEizYn-~85rOR9IXMQp+IznT@;aIs6D=Bvn*r(5GYizBfoiQ<5kAy=wsUm0#hir4|N(>;8hmt(ouUaK1m$3H?6ik_SUodX8MtDaaM&wuT&ImnO z9(raRgrT9KDX0ytQmJ&-+X-8kII3bLedDaO#6KPWf6Wy(TH(SWy{(U>{>jHdl2YT$ zaRNj#9Z-@nUA8M?(5(a*Rx=-O7kl7?2IJiaJ)Unhg=yNl?NK!9CkDNS?YYwU%NQ7E z5}mF3kR>e1U86|>js>QDP-Kl;(0wk}{)K^O4WonsL9?CimcdZY=Zn;2_s$y~BeM2d|ssfE{*=cNJ8|`AmZVK*QhlWsI8T z0#!uYlm1acU=#$->AAUZUNg*7f|zJHIkG13lF4ybug_}@>iA-hq~8*1Os!<<(d+8e zlqPI;(!jI*i^aLNZZPVE3!ShrX_Z_V5s6fUV$#%(vPglzs{d3zSfr1uPWfm7SHW zTJwHWH~XsuZk;)c8XB>X^|MYbf@Su><;1?p4FF2XOU#u&Sy8i(5GzN|CM``T4a@HjUYqp*aJWSb$GZB=v;UCZ)LOvQXBIDuEnuU@l zz)uBYb`*G}x5MnrLlrTU+vn4|(0LTOMk?RQRvW%b zOov$1FvgQ5EP5P&*O~CH)GXD{$_uhG`g}c^{9-rK*-w?NSPTUrCd#0cR-fo@f zc@dZ_aBC+&RY`s1a3i=LO=n>^(vO2yUZRWeI1-M%K{q$0DHmAMKX@4y4301lTx>TZ zE$nOb596*`_MiEmC3iKVcVFhjeVI&W$`4fZ>wDNGmZ1qg?)0D85eqlf25>LItpWBd zlu!%=1&lNC-d1AdeiihoVaGBxx9a(P>EHeA=_s(YuqAJWY9MZJ2LAXDTZVszKS=}=yBjBLg~f>M-aDJn^=#ASbT{NOEE6XUzuD( z1MGA4ps7Pb%HF>W)jyp05B{3=d*A zDYr2jB1839x{%Xk9R*t4+Cb(b>QjMLg*u}6+nC`RZta8e(z2yvoB z-S%ir)6Y7moGKbyFcSu zgxRhJSCy;=*R#3*I6eL1HuE-Lr6Vwee?uufIg1Fu3>FsU3+SVe7{sp#B z;S8>^v&!z#&A zF>_@~9-QPcA-Ubzlm(QYfuUrdG4#*zgmh_L8fFQ0YJG>YMg3<2gM@_>N@)jCtIatD!>mMlgb-3Qvv_?S)tlQGbvU^zu4}w2`8ms3Xw8OOk;NItZk7yM>GdRuuj3?Y z^&&6%*mKXiEY=29hOgn=+h)(XF(&p2sB$@BV6-Gd!U#|pQYY5y;O^cX0I zX+9J?(E)1sV?Jck?+IP_CH0qSkw#bPjqM9E^BlMe1iZfW0w6Bk!nNJTTH1ZtB@25qg ztj)%i&@sbt(&R!^LcmEjxtaNoC`VcC&xoWUm?V>~9>-b(a|?@&)i@hwy$@vIfbTAKsz>R-r&TH%L7Vvz;||0BKz8xQ?6HhiN*54gSy*&=4(Qqw0b)VOrZ<60(M` z+wGhM)HN{x6deQTq>}zy*UlQ_C}ft&3AG2#rwwOLqH26-f!`iF^`!ub5D%hmLPEH&*ivMHsd*!k^bNs(;8 zA53`8u4^PXC5K?vnacop*5FxNS`u+Z_=kbw0bg2PtpS>BfTRj(A{|-WSf3p|=dXHk z6(oJ5XR~4b0-b4UruP%P_e0&HyL`L{Aj7HL8-pIKZhopPAKaFTGFdEt4v?&&X>O{B$VvZO8%^Z&2YNB6OM;`Q z-1KxEx22Omp1t1`T#w(56O&Ev(G1ziqB*bdI}S?_sQ5?$^{+xfXeg!T9-!LZtpI4b z4n2K6RM*qke_ZG};LAgIvsN>tJN>${vW3ur2n}gXX~naV0Q7tfFlipsagYw%XBUYvtLn+dlSIzU}bvMZi}H^_}hgOygHb z5}2PJ&g6_XSJ%!wHB*6uTF>;(P?nh0pNFp$`ePst1VJ8WDGVTeIv(V-gV_&0{&&mV z`zwWtRm`qi^0vxjA69|tD}`-sXsDQ39a*+7_+K{&*pb36K*iEt!rt1?VKe|4+XG8L z#WEVh?uQ?u#ZRGPsRAmNT16b!e)g9cg;o&(s91I$ap(5)gAxk&m={p7_@^Bc_cO2m zU5NiK#6Ruz|DK5dnx=oUAqxLBP5(dh5n_Prp~4EgN-Mkjyxz#$szdo9A{Fs@Mg_o){3doOrH_DvhsR`NWnOJTx?HTU;y9v;mB|^7eP+ z53YrFuCA^^_cq}5@KzS)4L@azRzyu3isBe`Q= zJv|HE3z>SX;_L?=sAO)>AOs@~J+sU)6o7Ph25s?j2KQ0Vz~oGveX(LeUJ1LdR0fON z81I-TtReIcP8~Nwk$OanIB8cC_K2MW5_uARA4K)TjD4KFfi^B{0Q*lK<*xx|k~f7m zKL5HOWE!^TOdsj5?vy9<>yOSLG+&7JnF!9!&j)#{`e~&h8|MEMA=JYiWTSyFAh&!j zX`ys|!wp%zZXcxl7KKHwdmI<9CbyrQ{Dgq85=Nvs>^Bynf~+a5Rmew@0GwlaQ!TgL z`Bn5OE&~+lg%;{8ES@ix3M}o2-2ddy;DPsZ>Ki;)S9dGC5Iz&{YP3*J65ChF5ww%g zqy(Db)u5tUtNcuxN#v12Y22L&FDyO|?y45=)-b3ZtLCTT%uBT~JEhrh;W_CFEWc}p z+q=W%c~+|!Ms9DfZ`jZo$GdEFTuGKRY;A!doowCd@9C*)Mekidq3Y&kyW~7GN~k*~ zzvJX))5M879VY)+(|L4Xs-@zClqmJoDzm?;9YCN|>blef2kk2yu09wxf0fyhEb>I8 z?5P&$4Rnp!txjl=BSKDd;yg8m>jd%Eh)4hDh4ndJ!O#$K6G3T5ZYtL=D_r%T1@ ztC*{tJ9z~fsTkZ$%$<|`?jwz*JYdOhW(u!O)$>N}+76HGeki)W zUX;*&KDv>;XW}_4gU#|xAQz&I$8{@4KGf8G2z9r_qA@hY7c9+j4yFWS${ zpuz&b8IB^Wf6|PCIH?E2r&@DClhoeH{9(A^LbnOsoTkr?vpssv%n56vcS=6<~Z zeQ-Z39!ep8=R(Zha#6{gR4#k;>PW?nv5mO9@CwTyS+D_W)>nB8YBgY1-eSWpDdm%0)GtZ< z;jvM@ScIUZa? zUR1hRHZ0X9OGVnmlpmdE@KnY`WF@5)TEaL@k@&-n}1j z9*H%+=Zjhy!6Poa5cW}DXYcq(NnHJZ>-i1g^~WTV!T*an%q?>wdo&jQTLfx^vf_5BRd7yObM+O2FOIiD?BteT~mP*e4ST1ENbSx zfgFr?8-T}Ae0t&3Cq(2^Edw)e?|Ao*^@e3MeQeDArrtIak%lNPjd%t-nz4&=M0+m) z2-ddCXV*D^L+1OAuvD*5DL@yp`FhXJuv9LfGtZIqniuLqwG6 zKx)|O{T%R34$1+){-;R?-rX_I5oKDvjWt&1!Q;4CU)0kZi`c0_&PAcb1V})n+7LqI z+c{~%Wrwg%gx2XrnZ9KyTek*rV92v`W|@O6;ar-5XAM~1nyR^{cuuET$1O&QMYg4_ zd#FMs#>Y9Pn(nkeP@m7M(lGH^w5)I82gn|+_}(g3g7n4trFVD#TpIGIiED2}I18fk z_c)b_kap2N0uB`NEM=stJDmOA`#`=94P{VHq=3f>B^I~OKRJyEsnF*vVU6*@I&iNW zD_#jD=yT$xK6)$FX$DgfN0MJA7)x$-9+MCDqaDZ?WCf7nMw|RtNQn6GB~YM_B#kAM zjh)FsjseHY|GUycgVYq`G{5LBN{mknNsx9ArEX(lH9j#~Ts$bWpW2V~X;4$6zX-zg zNVAZrzLC^h;lK`I*yYGx4ZIWIJm$HUf_Q_ev(jacA5E^PG=4m~G;ANnt&?7NJm<^6 zscgV>T$v#gy^Ba)4P6z*#T>dr_-~?d)`n(YBm-g0Tg`0*0RQu^b}PsqRi>JA@`KPd za+MuDUt#rUBZHjV&(j`w|Dv<4zDx2?%>AVhs>j4!VmsBUSK zu2aI8urJk+leYjz&yXl)?7)h%BN^@NZXl73MeQ|oxSVB53;=L_liW_9l3B#=zLY>o z7Qg%Mw;-Tsyez4?pTtq&-X=wNmk!kC==Z_<5_;X!pa{KymejfM>n{C1cFY0aK}z%$ zeO#$^@6**^_w@@&iYV?OwNh)-n>FM8Sg}J^Xt2cQJXE4Y>XsM#`UN%>{Nf!yl_-~5 z8+1R)ExKqDNZ$gp*n1WP`8f7d9j1#8r~IOny(gYe61|_|?f+W~@~8oqaXmvLq0m9s=ST!a)CZ|N}EO^iP5EEZ}8l#D2!2AWczF(gOu0i+;MwK6w0e>>Woy(02$h$7#-{>_-rNz}M|E zWx4?INjmp-*nULU6rnFr=0>vXGV$Cn*oV^uX`~j#+n)i7*1IvRf%_Ud6D4@zQ_K|8 z4bK0^wf^g=9!jQI+si9DUhf$uJ{P;mCmH>76wPHBdyUOjFUtrye0uwE>!Ftj{Ay>Y76gyxP1y&4 ziA5=Z?lrrz?;Jml-RUl}yG6q zzFNZI-1UNQ_U)0Rxt&dcBk!zqS{VZNu_2QDj;H>V86Y%eedyhrBwY-`foW9iZkAtL zI?jSxiIWvLS<4HVr}1t2<-;H9tq&zJW?0YElo|3@=GEA{nBAY4?a6R=$v(e&}2+iHazQer9h^Nhf80QB{=G5+8oFx_|D&G6$eTD=e>PcaqO$# z*n60t3anXoY3bHlAn6a@sSm#OOR&*$I!Suh4$!25hF6)+_RB|cD`^EwX+`hXee-K6 zt?!7N2V3}DNB1*G0|h$UP0zOcPTM`9OLAkvgO9kj&lFw)%m8oX7?<6ArVj<7o1az8 zyv4C6D7QE0cc*Ntx^wwsxxK;V3z?xinqAI6+9|%f;E=5@LjB;$u7z z9$oS~9tnT#ui$G4db)__d<_~cTLb!Rw*;NF=55snhMNx(soh>I@d_H{`nl<=(eW_7 zx(8cc^hXC@Wzv-<`g$<`$Sf~g($o3dl2ahDXS3hR;;OoBJFU?) zLu5Y#4<*IS6KSvgv@ABOc@VL=jfHCsb=H{BpE96KvVDe?&04(s@I+n(z(+uG-TB?; zclR2F+}76Ky_LPy@^u6KB+95VcG}@ka^Z!CU)g(%nG#NgH-?4@9JC5*!h72rRvYex zul`t0s&ESs3@>)V32D|-4ZS-oqh?Wt3rYh=?+afM*zrqg|=-q15wl|xgcN9$dXtKT> zyz->AFIQ;n(#U=^ar&Sn+{&}vH!6NoNl(P4XE%^tSL_8K8A1w`f3#enb4dblh;xSq z8%C*=&K_Lq;WjRCxMZAhK{?>Td}T>>+I#KJyo$=SRa+v*5pnEoRb{;t%*1s6F z|Lm8R4})0T`{TIimlNaE{s36a1MOzRJ{!!T<6TdYe5C#|J$$C;_|&7wR>Z%`}4UyvkX?(znpsHykZ?cf$|fhg%+6eknq1x zoGTCxW}i-V{N5IKQ2?g0fk9U>(|qb8mEM;utkF{~(QccY-}YD*)AzO`3y9^ig^uXa zWe-Rk``e0Zae)U*Db9V*pDP>%A+2q&;@SU4hu)t~!=)a3rky-r&p_Il7b2KA*I#Y% z;0hG;*K z*~jFoGw#`A4qDW`IBGphT-VWjk$1$~Ct%gk^igkbf4!5kMk`y2U4Q<{XEv3(-}ao_ z=NccA#eUfulp7>?V2bh1pi$Wi7{0PXz#Tk}I_0~?T9=!@?-S6B>~H&|w`StA!HxPI zCdWj=u#W2v4t?5YY`x9%nB11T+uT#i_6LKGR##=DCl+!o5#buKx?(Q z=9WX6Ia9t;R@N{TMi#|SZ&~=9uYnPE;$=1L;Hz2I^_)?r5eWP83HPZ_&h5Es=%-q) z_j|r)UVUjn*51od%R{ASAAczMrDU%ONPnyLGZj@ARX6kJo^Q&n617NAdZ zXEe#AL9d4FdV15;v9b;b&Ck!lb(SM8YYrYQ8Hgtm*jZ^6nwl>I4|cbXIkh;rH#D+^ zg#kU$uh_YMXoITgTPmgA-_pPPNrHhO7qK`27+rtcM{{G#FFrP2i*4d^Akb9F`oEmh z*iFN|&jU8}@MWrA{lcB^^>6Fg0>(B(C;%9Y20#gq)@dy{lx&-x1FOA3jqG7jc$zy8k-C`YA_}0mWV3sr@~1viASCO_oY8g`R`3o#(&%z z?k2w|upqOjSz$iDK9mcwYg{dDbvw6k z@?`f-oCcX&T*O)K7IP;aS+@)#SU}gj1y;HE&Pc7J> zQMGWMyROT{_;Dbo`D{tlbu`>r!yPwgmmg-KDPdVzn9gNd^jj}Aned=ui*QnwxoWWh z3Wu)-Ha55opWuPmd1|FP9p_jkMeY1$so5|JIWVc+D&5cXb4ZDBYF$)Iy}&F4gm*D zjysYbHkxMS46i?+h%kqt+ZYLhPz>v!^I}muYeDNtH*lQ1vPRZn7-K;JDSYm3_2Uz< zaF~q=IGS2s(a(M&ZYFTBGB~Mx=|w^+5Y2vo;2_KD?P7G`N{A&62JS!5ef-_-L<+w= z>|yUYe9r8Rx>SLRy5CmaX-!If{I1ThVK3-R6J7mB*deM7h#D$v^@%XD3@dCQaFr1~$XH#JD)C8os`XG4e z>9sD!<6jp`XW)hs_zCShU*yRK=E=|KUo?eZnjxj)c^wX7&(n!x9=Y3XBaM{kYrcwC`Ay_J?U{Mo=P5JQF!k-FVS#Dz_4o1YRp1g11=2mJ0Hb_zHX(gPBqKd24L7=cy}$cw{VCx!b$rsBU1bG_d%VApF;>(JH4M=mmgS?9t@>^4)i;FWtQPfIolohd)MCiP_BfL4u$lF5=t+pDsRaCTOWTE9#czu%ql{ zTDga0h7O-4?Fg^*SkQe0Wipe5z@H;7kh)hlx)d$7(tS$54!q^deWP^A#-jkV3Z3pv z^w>nK1o?%%h*CklLEzV?7N??9()kbS&t(+L?|x31N{@rKXd>W~6IM}McHF_*bApS; z*|0Aw84rZ%)Z;EZjLR#1fdI`!_;z}qwV0;PVC*mlr&`4^%@|zr(WU06)@zYSgCjNo z_SFlp(7rdjWf2-ogQ{|@2rG^sM+%l-6lk|nR5fM{y6=v;TDxEvfqzp?PT+s$z7A0A zB);1^MwR#)`&;Nfa)|3Q?mO4{OVs=G^UaCeM|J zPj1?C*A=I6j*Y>I30q5^^L4kBLjysW7!k#l0wAC(0z~}6eeUY@j)U$h-995-#Eh7V zuE?j4G7?-&atj)=w>@^kt(K(7?^-6~?e+7D)7}q+C?lK0;`X?A`1L{|v4cFVR;9rC zHV8=Lqe-Qx$S{xN4{wpeZ~lSn%0Y1D=<#h9|S2&|Q?PNiWmM+S56 za2H2g1u+UO^mz$V>0NsKU1FQUY2O>5A%&7wTK_|JhU_y>!QH$?N}{R^K*26zD~yTS zBNMQ;u8db7eD1*t=tmkD@8zy~PCGffs!jK9s+>r4!iG1?4bMBJ0EzQeBO6IDmO=le z^2Gq$&yt*&P9*vilS8GRoQ`I!`pikg@r#GAst(&&Co-M0y8os)uXL!{X|y*gQkgYP z%7)xN19_on3}v(7MxC)`bi(It=iZ5M_f^nP5XQ7Ws!D(tTJPdsH?FR#u(F<7qhY?AA8vJ(IE~bJQMJSx88!O0 zb+dR~TkS20Y!;T_D*>rK|jv*SlIZG56k++0y#40$l_K>vD99%Eg} z1KiOWcFCn#j*KcZkH;T;pVucCe&#&W);yh_0H%Ckx5ApKpVZv~Dt}ei`Oylb^%jSqrszM9-%>Ow$iJpOd?uXEiV8 zp8VL0FgBmzwCqdN%Zs&g0teD{D;fPl7GCMx>lq;|M^m%;G8^;gF)#t{=CADsKVVBm zih2Sw;$t9=`G`MOj`LT9=VX`=(Y{F3eIC>Ud$vaG1CT{+a`O0@+-}Waeld&)rN`00 zy%fdmTzkxRn~_sJ=S&11#E0pR_HlC48lesh*`3N-xCGEbQK0FP~Wb3g4?Z4M%P_iTW_gA1&Sh0g_RI=H4^>jB zu6wOLHuYQ)=i%JTt(8@a%@ad}Ed&B%|(=*NI;+Rv@2D@syBO4?`-0jMmS88@%Qd{Ifk#|ah z)QawP-6GRmGJV`BS+M7c$ufctEE6s8B-&lVy zU9QA72TcvzwJ}KWFq&~Q-nezLQ+t#*e^)*JQ5SA|E&aT2DPysrec=khd^$ou*sR90 zZJAY_n#WJ1_o)_dZneJGtJZZ9&{@BM!za3Q!L(bBr=1s3hU2>((I47sh1qcuqc(6J zrleO^Y(u}IM~V^HArXEgGv@gcnwQ?_;m35n)wO1JU1y6pp3riT#0|TK0ih-^{vt!(0Jx{Zb=`b;g z-Sr%5aqqNUFDu2^z|&j}{gK9;$g<5Ly9T`|^|(_xwkl+PFIj40Zq3rA?ZHyxlP`5N zq)JU{L@&g99!<6qLxJK$n}oy7)bYC|iy-GeS2hMWklEv z#UMjvh4O2b8eF2GyvTATxq01gp8;^Oh|=cw_9&IeSpW~{dPZSCY2k(_p`qLK;>#An zm*6gu0V@loaMBFePa&Y!&p znEw|f+UU~YQA!eG46X1`4^UwdzL;ONmtMWd;lnI9+GWz}`eZ0(rs3H+(?|%4kdLX? zRueGGA36otgi#f1OZS~fH0aj7pWr-*XPLTYL%LRmClplS&~i zRvts%$b-#l$08vz(l^O;v|l2l)h3E7W@32*#-2FCD{77=?`>Xe#b$pgMCWP@h4v+09Xo8;|ekD5)u3C9=X zu*3DhHgJbcEEKiU*Wzm|1L)qd>&ATk={tqspMF+9B&9^hQ`7b3lU4{Tm@zJ@h%-iG zhmwuNc1%PWK}X%yh&u&)0o_)dmkX+2&ef*kTw;SG7 z=${eqrm>5WN3;trmdl1hukP}-eAt@Sh&!rC-^8iOBLaEmsaCg>#P?7Lg(RgUO7zUDN3HDYPp9iCu9%tH zv3iNBx3F_2yORnCgwIUa+BwkSYlvoY9Nj5Cl|GdU+B&}#R}cBfZCw#u0tFgyX-wGe zR$3G!5bq#!S$>R?_YYdf_Y~t?X4tH}bIXUMUR#M7Sw@_l3a~>)FYV+WSE|`-U@F+G zjQiTY`g^o=l34H3uIEwMt+^R2FS~lzOc#KoqPva4?e$4{W$nb9!e>UkrbM@$cDg}h zBz~LdNU__)euUBUL44I(@R1S31un#?D2PxVIT>9cqi^aubk!pZ4shZGpO2PK4TCdS zz9`ejj=6+{k~of+Q<^gT`PFknyPhx_rJAj-&ja>Bi@%cSqk!)#CvGxXd=AWL?Cqy* zMGz^G(wR^mho@Sc$6((p$R3J4?g9`n&$8w#x+=#M(+I!#s~+cvf);2@-ua9nfCD7> zWWk9mO8E@Yvgi;?bSr(&h`yowz$iM1vxxnM&G3^Vhe?QFH3P0zwPh*~wZoI9CcMC_ zEYrv$Ea5f+ufolJ0m66B7tp*yfPBld)tLvuMVD&kv7m|Myi0qe#+c^}NC=N$!ZhV# zq9JyS{+^2^olY1~B5f>8Ti(RDBkwv6B3nCDcy21bpGKE>-i}_r#cF##F($GUP=sq$ zWawav()^hvL$w0Vq~mD8otQ{&1$F@$cHOL>t`ML<)UW~(DR_p6sDgFJ|}w31gSSL3V-N6>2D zwF}CFpAkf@KQ?pedHuXj(`~T8MG|8m2kmx`$v1afUHxoe;TKZ+Ip|x~Aez@T>Cw3Y zWCuA@pdkN+SoKE;m#<%fKShvbkRP7;usd`x_Y#`SJk3-9!Hg=F+1T4L+nSvSC z?#L!ygU__u?2AdtM^%g%k;UMG7u5@ylsWzqscCPst4iQr3<~~O(vLZVma@7z_Mpn) z$aTh?F8zx232dALGo>B{2RO#X2S4&`p$z`EJE1HSCIHH~{0IE3Pq9hVW|4h3(7Rnrom_zt1;R--?GPM?0idcl*3uaik zXI4sS!X;;@g*E2&65au|R5Kq^NcmzY%zPGv0K{8t<9bmLIy-Mz2yW*l*LAr%xn{RCR!spY$n3?i$9&m+h-3b_%eJIoHwl>4}{wm=~D6rP6W%>e5sN9g0Wk zqf}75ym8g=9i59e*S-{ygY(KpFI-YjWCl%!Li3I|uns#GSyBKIL*mMI7tF?{T>CCh zE33$QC2u;lTN>=_v58bq8#{0Y7=t48i5hY(23Loya`??b|KAssURVvGEKHia8&CiY zom!bq2~&yX=WWe?TM8O8Lu*KGxfPoM9togaR-bQ}Y5)2p&51GZnbCoJfQwHl6~~jX!AAo|SxJT7)1_krLbK$U&k_l9Do>V?Md zMuoZn7uB;Z##jI}!r~hm0U9(2dW=1bYKXB;5bq`|g8Fd0=c>~RNnJeKNU&9IJxDn$ z_nP9br-f7GiPDK%`dAoVygLH!$YV*6b`2ff91us3 zh)_iwR{;eJc==Sm5Vl6hBLTa2VpiUrzEYB(8hcxwAJWO*ieUBN+Z-b&ZCy5Ib+9rv z%3glw+#W3FydINmq;?^``3MCqY2SMywsAjJh=<=}V5IcxFKTu9@`FO!4+R`rK@?PC z5h~A_zumfldtc2D?oV?(&_?F zr-%#>t@7TcW(@V3z)yS9eh>AhV}At*B#cwippo)7(n0S-#mx10Hn+-y1S$%vMNDwh z)_7r<7f0NA+$+D$cB`94>AEJa@~(3|_N6JqQ!jau{O#9tvj05Tm1~;? zq_EPQN*PG@ur^gjA^HaBv^L}8+GFR`aRL3ss>$;bHO?1!w^)C!885Ntes$?MJYhev3c>43L0=(b3cGi%|aD0umTdpIc# zu5St}_QHs>Q(;rR29OqfH?Q&JqmPHNVbADI?ggYR&v|2R%n63nJCg}s)vpgn)0?j6 z6o9JI_4Y(rTlh0VLvV(w`$?r7+8m9)0Q+AXp>`*ZPUXuD3?LhTIsAGAVYoQ+khGW?`!|M)&5UU zvbV}Hr38_$TbG|j#14yd^7h&pnJreQRVD=MfwjdW(oXmQZaQ?0=(fJAEhCDEXsOEV;twWW>eG{Rf@NIc46{jN36ZNW3p4Zvl#6Kq^D85!o6CTHAT8H9g^WA7_151Qy93b7X+O2qPW&)SXW24)J&R@z$IZCBOdgCBJ<*$IDK&QSaIRxQx3N5qMIgJR|;pNWHc=+sLw0 z{>=eNfYghS6S_J!O01l%?1&e_m5{%(y`5gM+Qv8UV8w`gCsbP|(=zxhHL(@< zp3kCI-z^e9#(4m1eIgw^-^K((ns~N?Yq9uMT|F|gNtjnx7*b{a!y$aL{l?x!UG`dO zUv3fVcN%5H7kcO5Lu#P;yZuE*St`+qJ}!%eK*DV`Y|V;d-PbH3!UWTGWS)VeASPz! z-!0Q$m;2AUw?8oBi|!rLXsJsLdt0k71QOoO?|QFAe8$Q>4xcTVez{0UHhyvyj3u)Z zj3u~n!9!1uNw+=X+)K^GC;V(~_kS1&P~Livo!xUGLdNZoTLf7qw)jVyd-tY;Z;A2te8_PRa#U;WN3oViXmU_R_`3RoP@ka5 z>Lt*??54JN?5j;p#*LYenYyojz9WmSdEi1E^6Io%lIzS;o2gFRS65oVIJM;+oFt_$ z>v&y}yCf@D$o!jX%S77wB-NbI4SCC)*^05N_0kR*qKA_UeQ}pT$Qd$yKa&I6dQi*s zT&!^ACO%v+re6asAGRN<+VbJ!sRw#eg{I8A?8t#lY!C}s#aZ-nL_Bwm{t z^sKXndSBr`5K&`${MrVoK7anK zaW%dU?}EL~nWH(^ko*Ptp~mqS}BOKm@I_qyk~CdT^EQ)XXs98WP=*VFxc8GoAVpy-;0I*=kVdje^a1XZp$eEh77P3WbCIq4LgvZc^`bL1*h8A_b^hf?_-h&4CcKT zvaQadcK=UazrdG0i1866-0OoN_revA- zQOvgy{BnVr?!fW>i==r zh${ZsMk5pOPE}@r?T1C>gHl2NY}dAr@O9T>t+!y}#wo}7xMJFYo~>hu1}+4AGmWGO zbNZ**_{*33l8>z~jPZLC=FDiXxu#yBJgp5}8UUWA9=&DgB?C8?TLXoggcMCYYbOWs zRh88-bm;m|WW4pZ!s4EpKke%5ZkYYZX}e4n_VXcV=vb#T@u9TAZM#Uw@_z1j99QGK!kH%!sh+>K z_k3!NkeFvpA22s4iqOsA@J|LC;Qk_&Vx~6LT(~Azy^5hH(m6vz$Q?cS&bllglFgPI z^*tu@w|y2}@qr7W{;~k?!>m(y%-yQp3cRAm;`iq}LDxZLcCR=st7R{obF*vkw+gFs zPn-32zl%xyxf~C9UVYNSlZYu^G1ZbqXB#WsaPbBODVLV=N}DOYjP;b}6`13vOE+92fNjh^LPn zQ&0IZ4=L!liT69_8I9?%zVE|L-=3WQvrhW$5cfSi(zh-Ef0?e`-Ddn4 zM=2-R%vb6`Ap-oZUb+I&O^V;>OmeoDWJ@t>IMXhl({eFq*nF-E?O3u|8dY`fr%{MP z5~puigJv+HsL9=Gevrd+VRkjDM!!5=t zW?nd&X5g;+r3zjSi?aiNRRkD4~Vs1v2w{=G89wp~IE*`YHO3UB+rq|~4s|(!d z?#akcs@Kb~bCJzCi%hOI)gnx~J9v^4;;YP4t?TGtHuhr}-zXecO3Q;F5}SI%JW~Me&nic> zmowf4xj8+&k!jDs=yU_@IJXtz3f+(0v2VaRN;T1+|8T7o&H1YjY+vWAenhnEl=Y~T zc&G4(7c&t3Q5W&2<3eP>ux+t#iXQ3O#01f&Q80s>0!2#E9| zy`xe?i_#&4C`A+mq)YEzK)Ru-Nbfb&L~3ZELx4bX7w&z&^W1a3b2iWYbN}o=c|el5 z=2&x$@s9DnW6Xc|^nX8X|6-HBpSFKy|>?MuJg19Yhh(5uOC*pxbh@j$-6NNvJ2 zV&FKx50L8311jA8FCdhs2os?|QD4_#=ntQ$2e~Gj?72L)Ij5~5j59N>br1D**$edr zD9rmr!pzVZ(*h?)&3x7l59hw*sK;ryeSD>` zm<~9-5b26F^o5tSp0r->j+lyv9n7Z>n%eWS10jv8xWTOMc(3+=g%|$w#U}20b#+`T z_V!;!oJZn7*EOdQ`H8v%KuZ07njp3eKyVt1a+&$5F&iq33VvwOvLI8u6-Z!Mt?pd| z&KbTWz+H?x2x1HiDS=w-IyVAS9mA(RkOC@kOI*^bB$tW+J!gCs{w2_tv9rwgm*`=` z3K(O?DKzuXOW=5|?bu#>(m5!;PRl@T?(0^!@k3=$bn+!8yL2RhK*FKsQL4@k$QHr` z8_Hk7t1(qHH*J#LWe@5k8G*qX=>AQIzgl_PD}bB~4%Jlws_B=`oFNc?2rnW3aca@( z)a6OiVaC=?^VqW3&q^a%T5UHB%k*TgaNpy2e_tt`8B|)8c9T;jzvBJs(+Jj#ijT4t z29ZUYv8x2a0evzk#U7a7DwXneN3=wLtt+12#Rkcbm8v(x;trFw4|kTfr7!p6v5lHd zs}UKYZpShuQ#DxPS5)u$f0{O>%LP3qOjH4UHp8+fx_u?2>WfXU*@0UV(OARrwQVW+xt+oC zCv*u}u-$}8Ex5O7$@++~QzW~$^h{B!k;uzmuRrhm#$>6)sb8VI&qTwQL}eeRGvO;yxBEXf@l?U1ff}E7eE-=*-y*fBn+fp%T4`MxXZj`F8b8$yA9MF1QcY;{L`n*4v=8^|oc@Z+@QFbwXCwMnu*$nZx z!SFEo(uQaFrTHFlr`-`2>V{({_zukaosWQvd!!oN1sxeo*`ZW@X)WdmJ*QHprr%S$ zs5MrPD`7cg5f}4tU{79ijLX91rs{Vlf+*JB!=I2Tp$@KfT#LiPO`)g`k^E69=J^y-J!t>s<9cfc5 z<+#XcxzvdAo{nRUT*Kh2!n-TOatXrrNtPds-wO&Ekr%9M_jY`we)6^HF|>Ym&o~Q_ zIhEmLbN)07AU*=fVcEdi-V5nJM4Kf!fhuV=vR_>)Ww&?8vs^miE}D}2#r5uSu&Rp7EbB3#mIkR@Oz{|AjSK7nan!(?^Ln;rHP4|Hc;duhJKn;4Bn&22c-88 z zd^|((UW{ON44AXQrAKr^y{P(!>YLo}Ea>7(?NRL2lRnxk@VxztWnhBu!AwYr)y{;4 z6{-P>#5D*x=M5DY4jZ2=nShUg!}E31#Vc1L+k48NJwU8 zKC$3&Q~bbIaTs#i?NBJWrCShoMiJ*NiT^m+e`H-f2BDX5{kj7=OwHde&NEaqqZj=- zQA|>Z_#D$Qyy6Y5Mmk4H^y99Br?_3AUvum5Ew`#U1q;l1Sj;^2_$)=bR|W$wMV+oeGtgAAfTv@8 z_?GMx4(0;+KA0p4n2F{K_4XQ0F-hat65LX+#Y7y-tdzEBUFTHfevE3r9bjb1$tC!8 zQ%@ruIw3v}^cq&Kh*_nmC!@z5tA!#OB(KE@xj(kql70+|=O44kR0?!9;4EGX#QF`!uy`=Y76Ek>*@O=mQd*Gm3 zBx0nd*>-N0tzpUTb6)ZxZL8sIUXbenTlWl3<-p%uxXwgOx%LFOJoHU$-s5Ys^xZ?A zE(<+ziQw8!{-uE>CwPe*T`ltKSTY|NFI$@=NP8w8N^gD@ zaP+CKPKjh1YJlP%HyJ_pR%0qxd?T#5^u~8tQ9fL?9)x*%W9pM2v~AP4zWe`tjZvb+ zyK7(HJA>oaQr=6~&83p{(2=+Lsh8$zW8hR^fV>{PJTP18Tjqk`@pxxxelnq~>B*_K zn&O!gf9`bdPk$couPGLjtPVnt+p`2I7bWOo?3xQI)hKgzowO%Q=*O!gi%rG6ae<_= z^dg>pZ_8UWKFwh0tlu!{++uC{>*xUlm~Wdm0rRV;GU=E)C?wsPJ-A#7*szE+^)1(t6v0t}$$>r4HmW*cf4BR1U11Ku z3R_h;YUbSjklQOIVKH1bWQKLNVK9$cld2bLV_#pDoAP)&J&)(N*t%}9c)C4Bs5%2! z5c}tW;_Q)T-u$~TJiJL9->v5KaRThfOikJNZN-Izve1;<9)W+M&%!0E zM(DeczIF&Oa4VZZb72d@wWd*Z4MKN1#pYN$BGy8ocUTy zeC^R37IW5@DFO~L*dnch$?5!qoimg?++6FSUc2usK&=1-oIAQfX^CBHT-|G98uAMa znR}bVT?aKrc_m7WANzu*8q;F3*))#7tuPbMKmBsLppbtLAkA*G{keI^{4_iDKSx`0 zk9!V&53P@bCVmgcC70$Br_21XFXTw?H^TnWA^TXZA~gXS1n6ri-Rb2%o4mNEoyw_s*vQ@lV`vXN?(4F)iQ!Id6+dX9Q*Ypmx;u<0^;+r=*_O8q( zL`kx$EOa^l`gS({p-Y(S>e0v5$;wXB(h^|Z!|*+}1M4Neb&)}~(1=LZV93h-Q#l9V zoWt}VlIQ<`X3LLBWSd>yoZw-7g{$vrNBgfR4BHLInbvwhg4CkeLQVA2lT5T`m_;!a zgLAW@+m`LZDDMfkbVzm2K=H2h-~h;clu4(Wp?(_+m!1idGaan zF)`E20TLjzc!XS*4Ri^HM3YS37IRhHX>=6d8?~70jpAOjIHnU9_#*=nUXwG|DhT*M z`_RC7zAIw2VNg1W!%=*a7q^E|aZj2s*qQvnXE6>@P)(;6BCflC*0`m+ZA<-6gIB4O-`Wxix*;b{Z~&OYCi&M{XLP(vju- zc4Wzax@T3Hd&xv4PRU%^-z(j+otv*j3OOcdGMM{TK3=m!DOJeoI2U1FX?cQ2cEon-CO@SyMX%LbcZ-RKrszdpu~VZH zjx-uB>_~Q8-xbQeOPnpvYgVRXRMDUy1V+x#t56HL3c39-Rys70ITh$2r2M{mV7lE& zZ~@$~-H)I7ISKsJ03LKhwG$A?qPlGjkw5jDLKY!vS_r za@S0!z`$(H*0Mu4K-ZXYkpYRG74_yXvk+3Svyg=RVB}} zc<@1B77aU$pUOf4fYkS=&YK8aY1m&SY`^y`d*g4reEJ`|aITTOxO6Ij@aW8&S>_Wj zEB=Q!VkV{(rV&5o(|n@!2)Od-@*?o5*O9cQK-;%mgvDt;415%O^$(bK&rkDJRO#sw zJf_!=UIWWL6aAXUPxI`({C;_$Pwb{B(W$aS;Mz`H@5u|0-~PSW|9G$dzO(<)H2?C4 zzqj#eP|u^kpV_}1)c*po{`SWIi&6eU#(#U`f5FUuho1j0*!ZDa z*ejrXVA${?`?D!FWUed1l&9hASAg>4a3kw^`qxa{N&fN%W_VJ*diF>8DrQk859N#E z1oBgZks^S65O#=85_M7RXAaC-gv)0sPW#PGmI=`z!xVZ>LI=c?>YE(%VK8 z*O9Qb^#xtG``B=_)FRi_`%93|@vg8lBDq?rFt(dwekklv!X1NjOsQr?zT)BVDoyhB z)&ol-1DjQ1J5pOzj8i5>=Z{{Xn<4K4Zb?5M!0pj+e&T@SjBmm(Eo7iC2=Xl&F$O z>Y(t*_ffg+KtcLkZ+1Zbry$J_fVm=Tn2v|e!oOGx(txP>@aG9}P)qDm#c)6Tnh zeV25l)8#y_*#%fD1YjOHpdhyshDS1$s(OmKo^d6X9NnVyAHDED|8O_N?9&3|7D4^gM2&e1K>diu6;|FdO#l4#dP0PXgyMq##Hk^QRZJW%#c5QNJV^TnjO zua4efRWZA3SZ=A5w>=zn;5lLf-wkV;W|Bn9Z%ao4c4VLQ%9#rQ*hZV=7-OkKY($ny zqR?7X{WCwGdfXm%Pn=37FJ#jTcS*~f(;A&eN=sg;DbNN_`CP(OL}+jQI0 z#Nx*Q^AUXWvKc!u<*M^g{d=ka+Wl2cng_Od9;RY3qx98+F_El1u|5Y|9V0~(bLe4v zhwMBh?VJftJu!nO&636~Kij|ziu{8@E=iy=3UbUzP}#}p?F!j=$xp$l;4 z?1vmdQjKjuK}BS%S;pB7W0W4AettMlIn=w%yW0rdXmLHDzyM=sSi46J-|%~JQb#&j z(vPmM2i+nlifFw() zTyHOUAV-ZQ9P9aDtb^Z3Egzn*I`@O^q`ZUI*pX0$k~1cf~(E$htyF zZ`?lQRSgt`-mW6Op`Iq-2Oh|CAn%oFn=yWO>qxQq){bZ2k#235+Urc_$Xf=UGND}1Jut3Hd+=kO2!v@ea!)0hTNs` zF-_mWcZLk@vv$T+``V(j9|nOZ9Uq0heluWzH`S{=r)Afq(qZr^z#;Izzax`(_x9(1 zZr=a+Q&PZ5C36ZG5Oq>jj{&r{aSHW!BpKFidZm%y-VwbQ(99}wX>~t&xZFj@rHz!9 zo8rso?Ot*HWix2DW!I-W<4#quX&du41j@%g&bZRnI(G)Q-AliP z4B#l+3SzVZ%IaCJ`8S+F@B!!V2WTZgm=Gp{9013)*Ii7Vz$2kJ$b@(W*ve1SW^~up zx>1ii5i;RgQ=ZS*^<_n{P?sv1?}1T>&1ZG5z3zeaxe0JwXRke?-4O1qoGyE@Ysbdz zL~=pY(ER57CCh!TOia~w2vtOaU;$B>OEM0-9=F8kd%si92iygV0mLCnZ(O#NDhp!f zGiIp2K7Uh!x6jwdUE$FDb-N#%YAyT3+_(O*FAiF0@@y^HJC4&(79du+H$*lA4hN6S5w#}>hF}o(w|B=A%X%j%paY4*_K~YLwyTgR%MiHDcyW>bgzw(!6~ceK z;s5+Y(#_wu9%zC2$E{y+vSp)P?Bm#MFcDD!>b%zvjqr!|;nDEVZ*7_KpM5{tT3Q9Y z!Pz~Uf3p_e-dZko-D3*>W~?mv4raU}de38OdR@KjvrH8A+wd}IVT18qt_E{Z?)F3_ z#3S-mO~&)=SKvlwo0NF&b|KgzS340(xyE(Lvb&&ZnhKKwP_7{4*2{19-=gZX+SN#* znNd7%X$qfbwBjVVy?JD~?PKo`A7O+br^rA%&HTt za9crK;NFf7N+DZuTDHdf@wR#A2|>6tz?$`)v^x0h7;|FE_9tNl<@OuY@=jOk*RIKe zKH37bWSR5)wLJ!3__LfOccay`;cqrU9e`j00}1n;zX%%1C^A*=Sw1Pf_WgRk)l)kw z?{5*I)_A5+zmS<%f>3c;fu8IR{NbV5VM=>t8+L=&~Wr)(>LwHDv92Bu?C8gR#XJzq-wv^>lUu!Mv^Sr+wlI%2W zMAFOO_mS_9`#672CWwsw#xZyzOmgqm!S-F!^`FpRBJO?UiU%F~<1e-5l_@Q|=vZb) z#V0jMzY5s*KM}Z^t>*tEP}NphPT5{{H@Q2p+Vhg#OH)Z2fwU4J{EPsM9{$lgU?vV~ zKnozGCC=UkE!mk!%L_ep<#Qak zRm}l#hbL4ORgrjG25-Y&Scz_Fh1g&Lqi^J}{fM{BL(=o0cP*$NMClbj;-qb3Sa9)G z<3l~kwHV0iR4sx*@4RivRhU^v9M>$2u6FgB?2_yIrY8;7=Ut3IDqKz&{+JM7v-%c2 z^8xxD13(DCITmcFQ1`xsg^xI<|3Yew^muESr_(d(egHCHzJ2%CO7Do-6CGUE-L|Rc zJ?P~Ic>u9*9FT0mGt^o7mrcfZ#)DIg4Ym@{GSx+gwF94=27isxWtyqpX!AY}eYmx} zSnXE6pUO`CVgR35+{mMp*G;&+vgrgK{!a32o=^At-lvz-`yS>blmX-AHYGzfHR<)a zv^Nuv`|ommB)nwhfPx*3M3;_L~20SD1ig5Res=4>pj^bEEMf@Z89-ydFV((I_iWMd@JhO<)6GB z0aCCMM=VcU(uP;xx9=3CpFce>K8zd#2TdVTufXiR;0_US)Hi=>Gc0%-=6wKZ2Y_<6XYEJr+`2)Wy}&D-(iddav{(HWtYruPau zy<6th-UO!{-Ie1}xq-fb;mIMWu`kZFMq>Y6*S=?%WGqpZwYs#!a{cqRK!DFF&kE>}r;E2r+(3oqErbg*Q^Y?uOhfkSgb)r+UovKQy0!gD>J zQxo9O@*Q#b-l4Ttjw=V8(6Co^BVvMqi#VIIj8ngmMcjR@qIKko`7*=Mxcee=RE}zO zqClF^Mt;+C8sQPffC0!}*I@np6YZ&cWlK6RL3`mZ+6y50Y)Tbsz zaK2ZQD@p}HaO_D&q{Ej(8$4j+j9&7O6v-g@qYxiG_Zn8RpNa$E99iu; zJ_w`ZS93J7(v)o%6I^3kXrx>G6wzQud9R3QUe@3J%&Fq2z;FNCkv16&(*XL0Lq%dK zfG6@-{2jgxIQ6QbM)F3iiB`em+xj1mG?OC>RRfo|67%n7Q3h0KeQd0GI){`Cs%M5C zl75+ziSamk^Krt-kr@h+y6r^_uAh>`cK&Q1bhu|d?2$TAAdYNlq&~Shnxz9mX}U8r zGknl3>A`jRltjG&U@%ifo6~#B0PllJ;q2B-D8|<#o~fV3+PmHs6E2Of7%TuWDWX;q zd8L(ZJlnbioSN)WTtrKWF`(5O=!>kvw$yO~S_nEv+aY`!(~w7}UYqW&(-+*NccwS6dZ*RH5uQB?n{bKs$o1KHB)c6FgLNj^$>PiQ2p&6*pzPq5= z{YrgXDPaCe`$4T5X;1UH^vKV+R2(2QOl~PTNse?P01@L0F0G}T(bpsgjLo3Oj*te5 z)&{30Hzl&3EB{tK{3m$%^QUi0ECA&3i}3?<_P0ACM{TP1?W&lms3} z7-0r4rgj2q>gU%y(Dc^$^CgXYqf6}zJxP(4b<*Tpb(d=d_};|X zsx_I?ZVl{B50yGU-6RRSH@~0AIJSM$Yk{dBxG8>8o#t#WO-91hfx$U||v`@7oxm#bk z1RcTAgGsK)E&)?%@l*#}g{+aRF4hNI<>46nxl+g;hL@L=(W;g`jGlYG!|X*|(}i^A zaLNkgXL;jzK^#RbXh}bH)J{8MR_jK%Uw(H%!Y_WAcl zo34${nwb)w^gqf__sZ8cA2>zjLf!1QM*@zWI-wbrf!Vb#{z2_g=TpkjDFez<{X9p2 z{6I0h6*ty>K4q0v&B^YTV;;gBJ>HU`-2^hO-&(ml=y`DdPPtUlJ)`V{p;q4${;z$4Y#tiC&V5imsAPt&`ua%H6RaJR8zS#ZTN z{j!%@%HbQ2uI7*h#o{EzrX#VFd&% zbFDURvS-p+ED$uO%$wWJVxGE;3Le zA6|_Es?=^PSC5C(y2g35s#UC{_7W7gl9Le=R2ocopZaSWC8O6Mj30pn9fHUvp$!K7 zIdi^VWtl*pIYRHJGT*QDm^nX9+Fu;b-={g23IN~={^0#^kzDC;eObMUFx`nRFpZ%0br{{8KM0i5BW4*ekdXQi+?9osNZ$(-J)ta6gN2iBJU(<(-i zPYT;?S6nrWmwe1|zl}2Gy%ULe(r6)U@5WrDb+hmvfsZUv6k2EU67PlNVV5KCFztFw zQR8*o{m`1WmRCOa!Uu;FeeC(vRSrO^qg=)TXBq-Gf@8Gi3&uB%peju&rQ`TxTsGiZ z-3^ipOFmxby`93QA%26EI2%2Xi#kIiwn}MwlQ85_vz98hUf64+1ca@Mi7Eq`ypx)i z^}+$_%d4n~W{6sxP|;0Jop0*p)z6#Q;#wKw$xtS zJ3CE1f7o9>hi{^cPXHed7$o~!rFw_B5>QZb7*-Y73Z`#|B-`Zl(K2G)d=7?hj8#Ia zlbnOw=3}_Lao2p@^RC_!7%Fb$_YTm>_ld9O9+xe2$F}1JrOFan&a<=2lH0Yky3AF@ z1~plh7TjBscN*JD^r?9n;j`I{*lN!uu^PT6?Ii$UwOB)I<#PTWk*O6G9^!1>4_7E_ ze-ae}n4D!61?`^T9shyl(vQC)8j;P# zgC04>QRexGE5MqR$7k$Pwk+^OU4=pg|B- zWyhzgHuQjd0_?_fJBgy49NpogBC8PVf$fN5JWuRtu$Am@0V(JIc(Uk}9YD?#0~)j$ z-A)w(eTcsQRAQ&{B&+GrZ_sk}E^XWBw$p!7Kb-dzpdLT+kw7FGbN!22%dxi$^QVl-Da-Vifc7z!~@9| z?%s%^i!~uvzGSM^P(!K@^0XV9X`=j4mwcYeHMpUQzGjr$h*Y)E4<4&1b5AlCQ2MG2 zp@+0JVoIod?;By-7&QmTQ{Nt(7;Om9f!#a4L{`nBrLBF-kE2@ZpU3T{0uP;#hes^Xg!5$<-JQT0NJ6=@M{49* z*E`7P@Uzf;5P-K4K#gXZRHcGx2q;M^h3L_*9!NUKs#GI=GNU=noAg*hu1>>46nTh; zvkbWoEb-@*;&_tp8wx0tS&27c+<&eB@Mq%(=sj&f4X&Y>&-ceo$eJ9HBx_tOm!mZ7 zu+1?XOF*v0F=iy%@n0-M`mSf3Z@* z!;7YqpT|-aLq@P|fJ<_WNoh{)Rp$#$T+?1zep@z!)3qM2(|(-WT4<6l+n4T;D*bb? zME@EH`T133X)A!p*L`Qa7_B=he(<>*deWsRIXhchHMOW|HLIekfHtLC4nE@S z!CL4>f2%uZojU!5d{JGt%$IGI=(tAwxW)PVVv;HL0Jt$4Q{nOf73!4M31an!om>0$ zi^OzGWV9g!XLJv%InSr~xfFII@(!Xoo8e z`6`=Bjdy`wC#1XmFdwPHN$%9};7)NG2op&nOA39JmHbxonvm5|qSj$#WCWPK@wh?P zsRo{n&Oe`VF?}3`O9L?HFoQ+Uhjt+WkDq7N6eM~}*BI^Vh!*Td8hU)+x{W_dX71BH zH^{+NQULLngzX+Y>-fZybM2~_!r@=W!m}6DC4OU4nm;{x7E^Reo%+&94F+0ffmVlb zpD6LUN~P|`7b8;YslZ3~Qb6zgy;53lmO zxUJH8h1kcHAYACTJ~mo7nRJ{k(XF)MmKQCA;WX zlw@msnLCDeYp-^89+3OKBT5*8E&r^Zp9$fqXdQ&TxR~zRC9D#sbo^FLUhV-cRZV)o z+rUoF6RkYWvtyO^aq-^k;xq!DpQzXRLGvv@u}kTTN_f;enLMC1fqyu@_=}Wx*!XyL zy*7jQua}gcnNda)GkF@BOeKeW3SXKwaXIp*u==n+#$!@e?7;Ei_0LS1^Z_9 zqN6!5@J&XvNMI6t#ehq^fYZ)(?6`D_)hVi36xzSR4M5iX1~)*~LY)>9>I9yWaiy&` zzyJmPgk|ZXY||C0cCz7RIeIa-7;G*ydiCp|y0{t6nd(|cd z)=wgOfKrKbJA}stNwR~~l~}rs8OSmo6Q#ROhD#U4t|U(K&@L(!&jVdhX166C0TGF( z{mTQO+vxk-@$q=giLj>hs=JpqixzbPPdci0%$`&|zAC#UdYG%Jg9XZ2Q9#RMhy=7Y z z4X{d1M*!~;zeJiix{ct}G$K_ll$EdIF=Z6goopFfOItRNm^D;5MIN#Fn}@ae2Vw>c zaNQ`>wqxj;;6|0UY}LpCgCXg>KsZ%3sx*OC0|oy6L*7|#uxYv$;A=VnfxlbstDp9t0p2hT zaj-E9b(H5JoGr6%XP#2c&R^)JRRIL^FM#`r0=gIypABofE4Ml3_$5@Gv=jhMEV1@L z2&vJi*c9VFDdSzu#w7Q%WGT;gt~$?2pR_5&A?lt0&VenS&nh~8kX`Z3(f8@nHt4Br z{_wZWHU6|YkbL1u4pvGe%huv{%bqtqX^&v3O)}2<0#IV#rSGm;C7*LDiodbmEA64! z;Nz98`Tg7tAx%nb;<1W&UWj^pj&|Bog&haT`-31rW*zR*sovr!y2ekj*~t!_0T_v?POw^ysx~$^r!H{zuyuDKcU6v{;)(alg^~&>S=U4 zClMN$07K4shBYmON#~5Ro$K`gWm5UDH~)9+{*T{vf}{9>mlKykEZeNsLUsyX`1f0Y zuK4tGn&)bWWiO^>J)s6p6Vsf#b=nUD>-j@VqB4K&JAcbG-u%42UqY1r;$kB6y5{H2Khziz;uni* ze`)kpr8T5&0{>GClg>wG=xLY9UVCy^J^s|;!yiynrqAS9oK}c`Hua3mJ;o<|Apqh1 zT{J=MLVDPfaOf#=G+;;FEPeO%AF}|?UPuDIh&|$nOD1m)s#OQ9A$V!@`d@2PetXSt z0M_mZis)jN$+LVJW=$cJm)UP|3U=j>uRr;)(p`f0j|=GxK8w$lTsk<~&DG91vJ%HU z686_VRhtI@Ri3?xxal>7W|~0 zBuDJ~JQ378svc88T|mAI)@QP zDc%4w;FnvxwTa|I=&FnLEwnHbPT8gRPl(bvb&HvY3lPLLZv$3=ajze%1EXaBI-;HM zBlo4X6Mj~_p*P1CACv?b29$nPl_ zhYL!%k}IRW1;R4w4S~rKo`|?f`+3edmC9v7=4)}9D{Du9xze3T*iVs|oj7-Zy=J!~ zx9eFohXa%Em+F1g`{G3htl5`Y?fyQH^+-0uza3q>(n zYG3UzS*NSs8>zH4*MVQJzpn5XE#JRxT~0b-AIQK)w`0=$WL0IL?XQApV3R-rc!o!H z4z|l{(;FdIpi{)g%EJ>e>BbP}xF$g$p-cz#bS&G{P?mT;>yMoCD&EKtD&Dw+2n<;_gbg2|9)-#xBCX zORmq}E}tliJypJDj%RR>c)!OX2t#x*jds(z2Q%Z)yGwYQx+dHS6G(rH%AN~EcNoH5 zZtrg5i_Tw2M)ylt%q+D;whv95>uLC~zq5 z)wnf;Mlrx|TOU0H$bO|7@~5Bl=`-Ktk?`K}l-gJj?HSR({7MHA_G zOGs*M%}Ho`OOJLydQxCsNwyE}s^om&+bAVlNZ%D|Hu1FHzhzkETAE zVC*rxyRIH@g=1V9`tjB@?3U{}ch@D+$M^@&Tk(?Pfn=^(mZ=mU*~YX-9xUg1q+(^aL%uBdB zCgbY$kAQ?Q9^F`f`F<)@$BVYS&9ePkUH4Ug3^*bqa4T?}7cK+CjB%|9tR|K=RNy=VHECv&gpr~-meAFJ~!|7=)q6w7PAa52r~0mrKS9P}mb#5-K$ zmzoChk5{x}&fha_OMEW~w_sLKRE}ef8Wj&^j%!l!h)sZVT>z%*TqG4w#{u;M--o)7 zt$nrvqj{dHDpT^(A_N%<0slBu5l%fGthrlK;81AP?3TfCOO>3{w<7KJ(<6lgEcTsy z3J0`BOPttk;qo`zIP0xv5b;&rq zkX%i)PZWBa3zl$2C0@-lM+<%vONrFfR^Ai!)&EqpUWZU++28-BZ`rv%!CDd@tmXe8p5IDwi|w%5arwhV zHW6VBb+tJ48YDD?RGF2=^@F}iU(T2K$nA89dXF7sGx8AF5xfv2if`c#TRcSUAXB6M z)%%+PTk63V|692GC4KLgpJwmwx{HoX9#hTSxyN`%yp?2FF%CCb)`&u(_g{wd&)kaz%f?vHte*nAzvb#Qe_@bT&lL^HygiWujv^&qeay2E0j{xr9&zgWg z@0<`%rI<^6#m)%x99=x+KDcDa&-n1fE0(!Uq*gJv(O5%@Z+{>DoS_>qfJJPJ2tvKv zKi=8DT_M{IHm>e82lA|ja_0m)mZeoMpaUH90F9G&L9Y1>@9Fao_r&M;sr5xm*SfDq zbT@*d>fp|gB`ok14F3$_2U{H?L2k}yuC}{Ab?05yJZg#wBeJ5BE^Pcz8-7T5CjD$>>EL`+Z$m^! zTr4n%qJYBj)7M^as%`pM790%{YcNN)w*Hn-o`%#glY|~ThU{$LLy9hJ!YFm{h?;j7}#0`$_1E6pi;%G^NFC*w@!cD)0bv6 zcdX=`O2EU2U?{LL;UlT_p{b-zeVh!|wi=GT)OGyM9W{B}7t-}21ypqF$sbVXKLBR+ z$8QW|?!|eCjN6B}-$tyP?LsG=M-gezBWTTPf_RQ`_%5yc(W}d{OYJM+19*Kx*#s4U z&*6x)hvVo3Skj8hnjq_}TPj&HV_A@-9%Z`Jr{nk!$zlr1W}d~aUuO@$sTfsQ241C; zY$y82u;LRWyl`#%e4hGzwzShqcztoQm2W^X%*u+x{7@E{C;?1STDWg;KNE%1@dZfxpiyV`JlpG)R*T5d-b%$%NhBd1SW4j23T6-r7{MYFO+C87+R_9Jy zCWyt6!h(cv#d^sx?OQwF+2OsQZT6tL=RLwDmuUhDjb_dXEpNt&25k;5B)lGG+{5fN z2*8VDO)Qgk6lR}VjRGu3Dn;;sO#xHm7RZD90@XwDI z7TkVSGUX06-JCV@vYAn(`|5mrv1`b%2%OJr?pv?)wGt z#di>Fmim^bxc1O_yDoaC;r`SUk7Ey{PpHm4(@D2hAKmj%>zxL-^-+^6$u4@oq{1a6 z2k@iFDzEb#7CFwYn8Iu>tBwyER>|=T{gtlPb`R)XT>QD}f5lBuVI-zpR$wiYQXFnW zkG`#5H1E7rE=`T48L+VKPI_8k=KkJ9ALoplti^)d2-;2$+Y1SzuD$h00`gjwGM~D> zki(%VGG0JCw#W!0iJ(p3)L%Xa@^Mt#Mk?hpyR@n-p%>p#?AOZqQ(B->xl^GB$*T{Z z%ONUm^c5k@-5y(I%pI}&ZoG%Ll023kG#wX;;a4NqGip(XqGfpf`Poh-`d9h#9 z-U&FqlppFfQch(*o&*cZG%Hz;%hn^tx;-YNrY6R>U3-(l`g2f_Mu$;{^Sg&-i^=vA7Cr$zME-(dr;CjWkFg?<}TVaCgifY2 z2uDwGAk6|e244@I5(6VMUo>UCy_F+askEt@zHCR%>{yG9t2;Age2B^q4rnoiaq-Q&r7!V8=}rKdG^J5wyE1H zp$aRFl1J)JhN=D3T`1J{l~fA?8w}il_UDvs-X>f|5h~OV8#iJhA$Lpgjn{k3amwbd zDRnRKu^?h)1kV=VJ8YQY0MEP#nImMe2K{V zPD4t|ZMf%uQ&9S<4Zm~+eq-AkH~U$y353dsAZ0qZmN#saa!X>|Eaa`Vz&qE1e7^1E z{I{jOIv?R90qw409I!LZjj+Bl+*xz%@Gq5yau=nw)iKM>Eyl-kG)9=r(94^3Qfudx z?9W+w7aLVRF4S1s&PAka%gg(2uqQqX6tfc-*C4~m0UsFo#3?SmI@$)o^f=b%3k~-M z!1ogDhc*4*a#wH;ac=t>ZO&^^e25n%6n^%(r!5rnt!mom}QM{^) zXbW*OIR|P=E=xI&&Ti-=xFDO7!?Vg&NY|#~%6t3%9{xp|L1^)m7L>N*vU&5MXKUE zOpjAwrd9E{r0){F&1xG?;rS2BXIkC%%5!^N;)K+qE` z+J!g%s`Ja?(-1lUVM@xD(dG^@DGFkhidxy6351*c$8o+?yW)%ZD&Yz8Y)GdqZV}t% z!qPh4TkhLX$){~iA%(*SRSIQg>Dn{!Jr8l|>A{d$nTIcU4L}Y`FNayT12au%Y>LQ- z%akM`CfF{)2j8nfY-bXi0C;;mlNd=!o8!c?$qPF>wJM2_29kN{PNIT}UPE~t!)9Z(bX}otV~PNq5v!7%Ev=YZ zs!%fNg=G{KY*ZQ5DvABWnINxCa@40nhkZRHtz=#MIp-w#SLt0$?H`8!vleBH(Kt;+diouF-B59}`}C zSMo#=>FW>kOzc1kTccQ1hpYStGG`q#2-0%@9e?PYr4FE#_+lRHrzwM=kH4enq}peI zpvLHiQ^UIfkutsw1JpDN#m%~20i!iVO%gap+8V1{=@^2AkSIp`F-MOQZ%+Jt+HRwU z@Q#SE6jERE+HuJ-)D%$ zP6?Uk78HDvjoPFECHTu0^<>AUCz5OR-Qp${jL^*EXYhdwQA7!h#>4X|vBKl8--~~G zo4%vqS>&F{ZA7jt$3<%*GNF2$rB>)BF%HWjEm%CP)Yq*VE_5|aEObuS%_YJ%2)Bph zWZqozP^&4ye}XpV-<2uXemJZ@t$AWc?=yNJBhr&Tku2=4m zX8P<6^N9hZb*9*IHb)v?CK6Ju9y(m?`x9Kpk?k~YT%6Fn#NR5=)`$D zeCreu2S~d(&)IZEx?=&~A?0qWU9PORlD}{X6 z9vxw%l&^-s?_KNulDpshh}55M}-mT9^@9Xz^hPJ7Q@6YUJPd$@b+U_|G@`XGES zrLwl@jbsDE5WPm+?7{7&eOonunqzaH2_pLnyBM6$k|*_Q$kjI&4{Dd_TvGDy+4HDeZMKb}7O$CFv<9K3rGf$onp*jZte79}nB}8n6&3Fdn`@!aLY} zw_DO>+$%F*%R{$Pr5=M%Go3@D=x&y7wA-jS%(EmH44HF0cwx{rY_2-0)Yyq#ov>-R zR1#Ij+JlvnTi)CW0gv?!sIet^W_#tBxWI&6{||fb8PMdmt&Of_#e$%SEI^8dE(9qe zy@@C-NbeEpU25n^5flWZcOwXdj&!L}PbKwZVnWRleyH#xpw?4nyzJODq1aD5?{w6`=XmFb{GVPLd zo!x*g@ygZ?mZBq-D^6nkQ}I#28|dj^b*>}w8A<;l0*x2OFl8+$M)Qxl`!RrbJF#Cz z0!`X*`*_%(RhPs!B0I{hcUj>aczP z<9){>vaq!@`QYPHyWffP%`BVO*Yv>jtq6Q9{85IgUh|9*#p7U3M{YXetIvKjj@x1p;_ZjzUh8d) zmIA?3!}%=n$#yG8(pJ1yC6qXP*pjeO{-Ys4P9T8bXJAa58LgW&zP!)a`eSgt?*sTa8!bS zbmg7!IwxtSxR8gz4NbHpz>ZfodgXZ#MfXi()Yy-6aUE6hK#!HcJ1NC5cw*g;OSxYb zw#KRqnOU~Oi$^~tw7I2M$NNW3IR~3?By*MIrskKFG&H17nddI-i4;p<_v#`Kh{hC8 zYZUSI>Q+(q@U(nTn$i`T&~W=0!s-Dp``5jGnwCdAFVXIJdA&Mg7#yzhT!p91hP6y< zCqT|caZkdAQ@tfn7Px>0`{*Ve5vKd*=@&}XL~iL6XF;n{T<+<(!U#$hpx7G$x*6eB z;C>IvHJj(4RVlP5WhJF2n#qtCb7aEQh*8HuoLoXT1NCU9%Ik2Q)7Vj_D>mw@moiam9 zU=K8RZK_(Bant_aE3yecCFTyKdCICo-MMsQzBZSia3=EY#$IXtQoAnMi>*ert#*^- z_kuTUdIbVw0E2}DBo@_BGZ@GLW&wi(5!%i$foi2{cvppYC4*WI32qG}ZeXgeJn4LN%Dp2}k-AmO{f?=mV`yZJ-XS;9n430$xkQah(0+&Yk zzGlXVz>xV-noi5$t5mfSoG*-EPu-1}KFfGatA#zk!`B!(!SaShWJEA1RY$zGo*?KU zR`PrF^d_a((q{f{eu@dN!7HWnb+fMsveTPW3;9>eE!I7}nsf&Xj2!EAde*kLD-M($ zp1{;h+PQUJecNQ&-1nnNr*}FfG{SB20AT;%HrC^BVKtNORwZoQY%(u!vgI!FK zW-crFrs=YGuIL6wT)vzuM6p!BI&D#~7r&<(ftIGi84*rHxrjH9q8V$8Mv}3<iQ%ar%wUNQWO3t>r^hO}G;fL0jppvvV#p6d|fFMfR?x zq(zNmQ}c#8Yh9iD{LHD>Jbj~%Pf}f_u>5vhTz)Ja{qPWmaK3vc#5mW$e5Ry$bLBR0 z)W%l*c9nq4mz8Zr*W+R|DM_$K@!xA~@#Kf39D%zD=WgA-8mq6-I+z`JTk}b%(o=<; zyOF8tLdwC*Z_i}~X32FN3R-Y#JQY%u3nmBn>5f+8npV<%wwh|hL@-2&D%^xmv1G%6 zXw8Qe{|}g5NZUTppfqS3DcOz9C{(d{nsHGhfE}&%j=_@8G*`glr+M}A`d2^&qJMP_ zF~ht38w+kqe?Zzd>X8GqyVz`6@ldZw$`Rtd?zwU?y6Yvi68QbVXZ5W6TSktX2-%vx z;yB5~Rd153c{ao|-EO@R4P)nJE~`JV0+Nh*nq@W*S4P)WUM+_={>}h2X}Q*FEsa7$IFzGfx@U8gQOCExo zP#S{-)e#+wM--b($~r?iD!s!E zUz}y0myBclwCGwWyft)=wmO;K)rzj&QfSv5-3q92AOQ_~lCv zA{z@t#dxPCU?QC~><<@9gXT-%>-RxDRhs-qKnICJp+cMgGJ(}5*A^R1Jgg|)1T7Ge zDxj~kEG%|D75Rt5y;}Z*N_QAOywr416~Ndvc|qLYMX2r2nP5#25Mo@>1B4y~*Tp_= zoeLT4YdyE;*Z)JC*?D1p+N!8ePfHQXYi#E^XSXLwbe{G;vLIR-jw=Q1Af~mPhL#cy zXmbV==q4SAJhs#YlY*bjt5VjwS@3&m_`c9f7mPieZERjDyVr}X;W&LA1AJCWj0dYQ zoQ8ACIJ?)SgH|_^1eLB=Bz$bvWU*yg32;BUWd3a)?3T0|A7mB(aNMyC-xaw-E^?-I z6$){WIufO^8@CSfhwwt!{Ldpng;L-x-?}JVimRW|k1w9mbjH#({(f`kcjpV8#%&%L zqM&h;8JSgTjT7aX(u}XxhR`$A7Ke8Yj=(&>p&X<@OFb)(lk~`rhizz~NGd5((RaL8 zl1V?iWT7W=wDaY(>et6DKNucLBDHRcb*UiWnvpjiXu?C2JYj^>JCYaAXC@|7uRo!& zq}0VnR7nHF%O8_qA0ML|VV)%a87#i&KABzGH22&H4x4nO?;M$b)2lF_wd1FiAU7|c zB$Kjd;$ni-O@p(8<;(mp*r_eI)>n~_9btwXJ}QJC3=6=?ir+EQV5eAO{?6jP>$JXu zohZY9*~25}_70;h+zk720=Aq`tyiXJWI08I-$}dejm}3~G-JIyYy~p1x)IjGVfWc3 z_6O*?R26a&jq(hvq9a;b{YsZ6RnBJR$>w2H*uqV6Z@>PqK41y;T%mHp`uUer(wNt< z#;dL84%OSdnza|)d|qU6)XCC=)u*|w8kwXmn>X&%$T4Kedo4hnYqW9gx|SHN_MXrB z1SH)K#vXl8HiV+wI{GXk`NDI9fpY-W~NosZYj zKGd8v$b=Bt|Kg0E>7pObA;NFyTcDCTMd^c_xT)G=46;XISCm~))2(H|~sufS>SUoyWq*f|j;5J#S zCfS2ts+dfh|6DCM0%}g#}kbpN z#C}s{w}v>m)k-8BPi}b@x#FXLOhiFH==l)|Zs{aZby-eV8?Pdyy5YCnl~0kIZXyiL z1N>2;zo~)zv&bGx=vy>PX&B^KxN1V_y*qH2hF17Z!>?)0CCW2&8(M-uAtoYCNtjoh z@(C_!W7B7=(TdK%S%jV^eKsmbPJvU`boD?iL7}kkyMfVCBv$SesqAg@m1o>?elLO? zgB8IwL$}wkZ*~%+z>)9Wb5Ci2Gr04g`%m`xovTab+K&R2LzTX6d+NUJW~p?+n6x~a zj!O>ZgEUE)^m7mFg_SauHXgQs7GHN%G$=_LR4JZ@mgUYGSVr08sUqqPgq7jJJUW$c zh_CBVh`lZqg}l5dyssc6K-a`;!FV8_ktaEv!(gsx)%kUIys=j|m1j%`s4vM2;~Q@Q zX!z2%1+V<9b8~On=H@nFvnrNQhIV5}Y7q5&fa0?tb;ky^oV+35V!c`&DS?FVfWOQ) zXEQ(57a6DaP$zXpQ}@G70oe!1N>JT+N!?~sb*Q6AD@%wA?ToE9A7}|QVfd|mpQ_Y# zsS~YT+_eTsp26QFFMnks`PE0#q#(uKzw^tv@6LrPFD0ryU14FWvbVO~p)h>=H zAQiSpWE;i0lXhrw8*e%B9z9~1@+~&fS-$DBDg$^Osj#L6y`d7onA+-g>6YAw)v_!# zMkjUXEaZ^$ZAT7ols4-%^brFOnO_4ACi`JSJ!o{gdM03CRgwqOr~+qWo=#NCQR|D^xe1<}jn_HZ#?dP(tlS7Ur^)tE z_9^1JkO&o&*%}^?CVJ9{7srS%j~d8YbyaW1@W`RPu=uoYEek(!kq9r!41B@3S>G7s zg(aXTFHKrgD7C1UCE8Tn=-jADWbO}FH|!A~{u~@?C00|%G3qe!s^#=1EM`jaCSRYM zSO2Nn&{iqz|rqv-CBNBLV0&qV1C}I1V*VHZGBrg27T?TN-%9$?Nj1}%bY0N zkP#T#@FEwIZj)qmjiqxV@g&sYdR} zEUM27kY9H^)`~ob^iJZJ!#Wa-o7VZ_Cx0AY7W;O5Z_omAE)$3!+U3p@ zhDr8Ov|1z3Ld?Tw)SW&u6=gm+WaKh>))2XFI?)%q5I&Fd)t5&sa>mLtKHNM2Kv-V zw~b*;yb2eY?tm*D{&<)h>CWysFmQ(7?XAI8r8Zl(+r@bc12qlN0?VZizYxklmb%i~$kw}IDXl=)qBZq~3{XayW;^8Yx!7_ZcF-G8;fQZ= z?Ufi&i(HB75lIMTPYIWYr`Vc0qz=c)tA1`rtX{Wa5E#g(CDLYAMqhFm7Mu&e$Zmes z#BHMRHHHZ8eLx$fCiLxM%@^mg+;eVzvHJBNgLS?!I28h7wQwCOxB26_4~`FpT5M=#Yu+Vq7c}5H%D+}Zw~|YAk3Wyv< z8~TQuF3Z)rLS7U!2sbQ{QG&b3vETFE5f4&(t| zW1Ou;rIiY;REFAi*^T|i^)Hk-|6|ftvLVOX0{2peX>N|aAGvsUTz979Rg7j%KR~GF zy6!W0oDN+vkNjC7lsR2s_nAzAYu9&u?lRC}k^>L{lNlnTP3BRxQgiNEd#?$$K8SOf z*dTgJzSo4XNpyy0^!hcsDb$@5f?;1_0E4id1=x)hxFN57s2H3AcG8i@v6?($SpJcZ zQ(9JA7xtxCwj^Y#(`+J}=l;mx%kS&&s2&VD=u-jiOyp|3Im4K>2ad8jHnctr(=DV6 zmKGV%ndtWehZ0RG@Fiy<{wCFTw$sNy48FZb?NCrBo0FPW3nO8`=?Uy&#qp>?2 zH+AdCfp7(la(ZV!fiE2XvaCT&vs!g(sPKtmjA;!Z=taT$khGm zq?G&}frY$pmnffW)#I<#mbH$cHaExjYKqnO2BcwSfCmjQx(I5Fk{X^dq_94rr=i#Z z?VP$P#-xXlU9hL+_Jbp}8;x92e~aN`Fkuuu5@vK9E_g^K^yS;7s_)`3a_UZyx|knQ zw=3X1qn*y@88(r}!PoK3pvI5!Q--CvlMep9whDPU?5nl?q3@X;-6Ej(bn>23sd-y8 zbq5gc2oepM(f#v$f>pE6F0`!*hpHxX8JvBJpyAyrTz_5c_K|e#-FlopRqK;P`OZd9 zoer1MQP3ABiIxt?!A-xR3o)<=DqVm#@wKJpgBhnZru~t`6(+kVUV+RjnICBEh4K;@CnF<7FW)-O?%eD4A{j)8}QcAtFV z;wLiBdPJd6Z8^RqdI~U3T~>`#khgMb=>36@9wLH`Fg_+0N7go`dgz1>T|d|rf@|L z{bQfblHPcbk!Ibq$ZxCeS#<4Ait+mXKDgyPb2VHHfa`SL18EN%58o~)VfSB1O=!{F zPOf;KPg!lP4j+5k)*(iddf2Pts>4qnDDMjKs%+M>1Zu5KGh*#4FEp-de36mjvf+Ms z{M`R(@$ZzB)J*P+S0C`lPXnM%g5fe+^korcq;*w_V9(yk^rWWagY%3R`)!{XL~;US z#a`r^GuY%>4la4q@D=FSHcXEquDxc;0<=Pd;Bo_RX%8eTt@?%Dea-xZG}=N38QB0? z)b>H8Dg9=lW`xa`EijEsKU5oEq^SU(G;DH-%y2~8(xiHht91MUJo*@@N#CaBg+WH% z=lmkgL0@L!>$o0Vp(^W~h-*{J(!hDlh0FN02Y?XYp zjb8T3E9dGLQEHO9_;>TWVHYA-S1gh)gjl!3-Qx^TfBX1oWyp2y8qgJvw6#OU@mZz9 zD2r9x>1ysljy_2U+J{@eEO~X5;x>wgw=?g#+N4}iwfUl2Xe2EBjgfMwmd%7(NjjKe zd-!GUxiswF$B}Hew<3>QK*Uglx!=BzahBdQnmSI_^`4b^9-=;!U4Huy_Q9}yw z@0vAwn#OpQ4V?eZ_yBMsVi0o7_{xzwtl%b#GF^DF^6QT*Cd=iCC?QNe^R?ealP zB}&tC04n%m5M^^`KRHxJcq+%VBo1JHEN3^u0Rfz|LFyBjNClSm_!Ot{-?Rr7E_R`p zE6N9s#lKAXI%2J6%UW~X7YvvF*z+1@8(8rV@Q=ot0QVA0l(NmucbC3o=6CJ|@HYJ! z5R_SPb6y$7pq*P)PkR8)6p~pPGi?Xg^TReuiF!JX)wc>SWqVZdY#d=rd6%t-3*gAl zs}asB1;1kbIy{8S_V4irS9 z)jO&dpkFHhbSWr(+NJt}M!LEk0Vb-CO0^y8Y80;V~xvow+jM%acoenUZE zy$7fA8B~h3pLw`NHKb}-J#GtpTjFLXHtqgxcN&cNU~5|W9{!0=1c?WS6xhiX$AGamhZWpg|3 z0=F$KTk#*uH;5<`w=McCOb^gP%nZi(PH+Gi2~*bB|CIe>rNP>09;|B^%P^PISl<=T(tsO`e zwZ90ypul7%r+zl-2wG6@*_Wp+=7AY!$}6BLb<){F;y%)<0iDG-gZ_}J*rUKqSe$BF zO{dOYxm&{M8+`>MoUHBzB-_=gRp5I)NdO%`Dei z)~)7|+*66?Lc_01wVpf3`oM$w(*4kutpwHs)h--7gloQo^483l08=_$()bk}X;eDe zvxe~Kl_hl=6YeLbY0rwvj*sU;-wo9vVte~CQU6dNs2^yMR z2sCl)L^dd{lzl5P-%=p+BE!`fQim?;&gD4%BcHlStd(s`hD^TgW!zeyI5CY+vqIm*($kfK z8;Dz(0)ncCTvouAT^;ly?6k13i2zQC>yDeD#`Dr}0dc5M3Ww9On_el&Q+8lpKsNZZ~5D*mpU?P#C9$VkB zEtXRD7Jpal#*)F@{f^3gW-JxZT3WgKi=A$))HlW;vU%^$yt}rq>eel3DBx!_vn{qZ zrlK1%$k_3OmCJXbHMi}X_Gv5c zg(2~hm8ZGU^(LMf`DH1^6geVsDXh;Pw{o~Sva6vh)Kg%VHha0Xz0d5Q=DF>6@qpJj~wt;Q49mEMi9#xEi7} zrCE*xn5$y0xBX+8pTD|x;pQ7419D2S<3P0UeGz9$f} zu$#D`*>vskO1}fQoGAV1ukI54)v+YmyzY<)0f7g1{LIG9JsLg;b^)GZ$MGDigQBs3 z%ZGQ^NN?$~*4p(o12Vvop7UMmCpESI-Jl0z&j2w6x{EIB*jB6kop(kecl>Ckt#r~k zlz}do3O?5O)z&!C8%q+;TJO4ofHaF8`v3i3(6~iG%l78Fh048%d*WanwMe+g+9|m* zN3l-?0iTlYD6ku1YpQNk8tD48alZU_S^w`p^MCn_^~2Xd3OS1hzVYKb{t-oDx`}{P zrn%j^MW(c}6ff*cqQ$LlJw@WfShXxs{ANE3Vf|zEKyt~~ri*eG+F#qrmgV?#eXl0J zYauRFm}a?>jB$4z7)TJeu%&>+uY4dFjtymK-3_o*>(cwx%_b)UXerXS1ZK_O{yKO{ zjP_wU-lQDwYa3r2q)tj&3wlEGBe+f)!&rw&S}O*>CLmwGal9{(&GWmesGkSN%CQB$=z zi)@UtRtG zp{e%2`cvB9$N!Jrf)SdZXSYzz0 zYO3l3g4D-1dz=w}|JHxklK=XdjO~S+MZmt%zH)Kedj+`TlwV|kb&{&;oq%FSn1FTmUjc-FkB#!$r?ouv>V4QL5|~goaBu#%ul?(1*(G1UT@Mjc zzs2cS&viA#g+-8#&t+4~q}~UhALw*?>+{pf+<)j+4an?s3OT0s-{O5*xM)LMM3&?p zmt=(7L8E~IhdY}u!N2%i2UpvBW%INsWNJk+Vu4(gYn?~n9deV_m;} z{q&VNByi(Kh*HjW9Jv_oOWiqLbcwn4Akd_q0I(!{oH>L1#Y={EYO20vD$|;fj^cy!z zWU09eEks%VgOL7h3%6&kJ$x^lw{gQm3ET0IGg!ThdF*V`3#N$jzKm#At?QD%Hgy`1 zg@=Je+7Pj1IdE?;g4A~=^O!1wc5)6VN`Fi>G4ugYkvi!gT>pg$_g{tR&;N3)2sC`y z4)2|Q)S|@qd2y6{HB?0!m_z&~1YnVEzW>YC;e@K?>(yFdOKvj(-+p=d7bTFI;$*Pq z%SVoY8GQl_S-*X=#jtY~5dX^B*Ahe|z;$Drg*FVI_sJbgh>nxVoD3`DXm< z<*Mzcx#0;9wbi7K9*OR~8_GfPigfO@(Z0=V{$(}y4w&E6nWwiwI+OXLKD%C|{%j_@ zY6_qH2No`#Cc~8NahK)xTQ$?(pDJw!v@X)|#ea4xxM_s&5B9|osZg8o_q`y)8znp(o_i7;6YhQAV-m(vAfO8!7VQD=AA&| zgx;4`itzYFXOdX8 zeA7Y{H6$tayFlwpyWzs8GIPVyL}GNPB;Rk>-T-+)D#M4jBz+qE`>3{fgiDI&t8h!L zA0z|1CaR=6@RW=>>Y1lpj-3@2w_B*do*muiGCj7YxTK2-`VZ{K?w(bly+`?$$@$T| z9edraD#I$<$YcC^smp{7ymP>7_xnbZ56)4)aanP^i3sJ=UQ_8vxue+azwzjwuKvF< z+JV6*6xE{*kRbJ|Oq#$WY<$h!)q?$w8Zvg)7X|nrlaXqb_pRGsqIVW6Op-d0M!cUw zsjw4Bt-?EyfQqT=OLW|vWWw`I=)Hy*`x)d)BO5S5LS*QWH&L2&I3ymalQn3w#cZ?x zF{87OX$&C3OxM^wg;e=MS{q`E6~PSH;C=V^YXF|^=P8m_CaGpftp*K)2X^=9OkM)- z92IlyoTw$k_;Py1yTUu6dn*VQ)_+LThJZ3ma`Uj`mjKK+;~wbu^U-Ew`SG6!B%u$wrt!io@NIem&;j znxW0|ryVFSMzZXu8|gF^O_yq?hr^srKY|-t%gMYaQ`*j5p<)oz@W!q&j?onkH0a;4 znJn-slAouN%XoS0TpcuG1p`Hli}XEH?6}l&?u}GiV`N%aZja%MgRT@oICcy5)UmNv zt$X2D4;=jGfouMG;GzH}c(_pZLLx8?OwlCm{(x|t0ocnX^!RWAY%&vy8^PszDubpL zDFqnO`@CJVGhJYtin^U#X@kofz({PxEija)<26k`ZV4sCTH)QQdOWLruV}@hk zV6$fWF8E+IJ~u*|B4PRYJ0=EB0s({QiT6VMc!t=xCa~ARC#vGOj<(>f7%**r-I_vEVZ>Z)> zy-;VLfqutGy0qul#%11Y0FZY1@-#pPv%1Uy+%#g2ul6Oox177oz~1y3MJNz&#c zI*#hblR&em9lUC~gtkEqq2|TV``^J`?!tv$ZPR~hE`=rT&QMCmX+AVokI}|$#@J-i z!CIjxY{KoOswg+q(Z|@h9w>Milo<&bkl(~YOZviCh~Sb^pD??L{TvHWV#h*g#5u$i zFkVhuBU8!Ygd%s6&8~@{as{v#zq|$Me{R7EBhG4ar)-|uw0GBbj9DW`8nM${6qjDn z=wT^x?R8r7irO$P<@dghA;0t9@|Fz`bvliF1Cbt#4D8-V3HmPjN96~71BxroVH;xO zti#PUbjpqR$c8fSCawiV&&}<}pK__X>?$BQVANZ6&2UpwDDnz8lCG~dDBpX3vxcX^ zc?fdzPLh#tCa`AB^tk$s#)|L4OBKGg6%%#Z_UCw5|ZGkVNV2R>{gPIj{b?rPZypGS4t4maMp6;P*M0W5ZjWEFs)R zc^{84!LRRw-Pf5Ymo}l~Vk~wu3Def#IN4U$`Q?IUgX=vLe#3BD@eaF+Un46$4+qFJ zi@Zt!S5z$5g<7fbqK*5ln#;bqNyGWZL&AcMYFbNNzPng_C`MR93ewb|(P+R{o8y>~ zJhW^>22NtBq=3CpmuwAsmfc>qyW`^&3QEc~F0*AiTAp0;K%lzH$wb5Px2#$Wd>+3P z7?3#wK#r^ZFD?Tp>oi-6asJH^Z!^*AwU6*cs(r{O;o;E*o7&<>edxG_=H zB`wr*K*`=cQs~nrG!h?6DF>9E8|VX2Lb6oHU@%-6%_}DSK`jQ}1>Ay|QdL^68tM?2 zYdW9MtSqkqcQAEXaz@@k ztbI+`r;r2v^;ygL6ZkUbdYAFH!Cn%pRu+2Ufz$YHK}ZIVs|gGQ^C#ulPaYID>KvBh zp;Nad>m`}TGVG}f4pgC?=UMz;v@MISaULy49!s(AM8uF!0Iyo)2%&Ix6*^ zj&*--D-r$f*8uFBTR^1boKlD}aM~Zy?*ZXO6Q_L%jk{a(0l`u03_0nlg)O!=vWX7`;<=U(4B;=Z}*I^oy}}s7%iQ7Ub?fk zA+j+HB`v`2`6)+9R)4PH0hElNszK+5=w5@f@{OK^I578GRSV-}0NJ>g(|B~!&t^Qr zVlDfbNg8Xs; zBgVkpF|JPvvb-PQy5%PNX%YQ@cZi&Ri8$$Fy|pPV>1{pu4QT~tM}Nd{fqOA9(7J`0 zuvxQNe%OdprOiuHC}cfKT31DUBf@quu`y=s?F$*+o`lNOA?l%(0v=EW!T-pj(QS25 z(=x?o^B6#7IsQwf1?*y!FvB``8xVBHEA6IB`z^;L-<>60k5q)ir2qaw<9)k=c(L`% zrIMJQ*DlCLD$mV=1fI^9G>2l?e)&c}NBI_g5ldfewR zW*XMWG)L1U>EVo0JeD!xo z_i9BS&`$-z0!%=%;@Mz0n%E)8m@D6%qNa-o)G9zmni|*a! z#|*@#A!kPnxSjt4&f5gN%F! zj%@Jzp1b<=Lgt>S#JYt-aMBZ0gQ8Dc<=P0b$Dr-9npXyCyuUz$h{wKXv_)QKxfY~u z^JMH7PYGaCSCEOgAHu42Io{mE0?ryQ)mPJnh(r0|yV7gF8|UM03SSB!ZJywjUiZ)WbG)VZr2&+q5N3w?9MRl=j0|(*awJoy6m# zy^dLbu(u}ZHQ{Q=#fV?agavW}RxB$ohe@8AyZEVwkI|tb5)YUCbbj+oT=*i`_Be2@ za=t!6U||WlxU#=)0gOV~ojlk)&{6teBFC-$MX97AD?Nh`G)^GcO!o=c9Wd@5B|$n` zRntzQ8*!KVq!k^DG`rNe;0d3M##Y-uxhThlQoi_rFFxzkwf$_tVtNUK!G+ZNq5;hNiXZml1(eh~2JsH9%nAFxXP;$$x0x=_2LeF`sq0%)wL@p%xgEpw zf)<z5kUZ z_#YbMpCXaKCjFK3`5&6}uVl!7aLN8khWv-f^l%p0iu_-DKg%qQ39 z!r%I7MaVz4;qHkcf5LHO_SXadXOa2qg#Qxhe}Aw3y2Jl>N-N)Q#tFAIUG95y@gx?= zrT6hwy89@7tT=SbFTk|q-*5asU0m}(AntolMhFg!*BZD#60Dvk;*i;op`*$PhFw^{jj^&8f7H}lr6fK!MHatGJ4;XE5gnxS zV5WM~`Sm5b*daBZm^s(%=2!X^D-(UIZ#ixx0AoLu2eLS_n;o2~enH1`u>-+PI}>0; z*qf;Q6IPNZpA@5$#$8R4GQ`Q0=Z{$SiYj{J6M326?=FG%F})s>yezhN_EtboTxX{1 zp^59ZBTp*GGMGO^{A00O(`~57z3HMBD8|z9R}+CvrA1#zH8XrkjO!%E4cm$A7$KWM z{=xhj$7yz?R>78^zuiaypJW~Mty{@yQR}`Bk{X$E!sF%cg^iKk`;IY+lT}jhYra?g zLTR#w)}PstKmN>)+@}!s!}APQ8G;A$=VSFd6L@&~uzng*V%!APJGd6A!t7CZ_f4}K z+*$z1Mp={=kY6~qEa1y9E{I?3aMcnxL#XPTbVBf`vRI?1_tsslluC?h57s3&3L^j} zi#BFMJ?xRoVtdS-aukEG8Xmj35;+M}>B0$En&K=EqnO)+4PKgKRhej7p}wEsTP5#(?}cT-PCf;_b201^gln3 z$zTBU=+J`vl?rBD261#iB>Srs%uA`Hc6lwkFOz8`&$5`&qUxxq&LZGUW=XM$q5rXS zF%V3sQ}MG2s&~spB96;Va#elw-E13dN4QypzSXF-ygc=vAM2k0WB&CJZ%iXD$Mna7VeAqgUpC#P`! zv5>27BtSU!etJ*-yyfb5<-~glk0aC0wwkZ7YR$@8{{w#f`&ZP8fVA99zPtq#@yhZe zCAEXp5#%O$$1giNM`*bF!6+V!X4B?|EI0~oc~95DmP*c1Dj@3`4%UQ#q406@2jHsH zOyce|85h-)pff2Tuf1i}D#K&&G`NU$_*U4>+oiqfshNG+U11aL$48e=d7RWe3jGJK z>?F~5s~Uu8NMDT)+(~?g9pUOrFh{NiG{-AqA z?SZs%juP$0u)Vh*CB5Il_)59Wo+VJraiMlLXfN9ht>JPw7uF^LE*B@t<_TEv#ZpNf zwqJ4RpS@zU|1u(5t-_ypF5f68H=}wWAj(_{$*}4J-SP487N72oTj8|R&QW3`Y?L|9 z29OCV*l0cyz>VWSvOFIF!iKitZ-Y3n(H!`_KIxo!;Sq(?PdPH&vbwz;yt~U7)X{@gVGYEC`uPu& zGxve~qD!A7v_le##9^-3+x43d=Q(@K5R~aUc9(d%-TZsZ+oG5P{U`cA4_>+xrbIeE zI@NZZS^D*N$Q`TR^xcj&m~Ty*Nj$}E;-#D(rtMJ#-Ta9L1i~6 zXN77JnZu<4KB+AM$2(z08V5@C(v2_sltHRVvcX}Ut&wSi>55Qm^*qhC1C64wj{DIB zm_`3rtHpFzfFQdLEYy9&1p6V3it)A;%unY~eb52m?F@bdLP%)pB)>{)YVmUF7LYA$ z8I(lM-s2768v@_TwzIVJ`Ra?(CHBp! z7^8MskalQc8s0fp-9uLipM&+RAt8KHK1lU_EE~9{{+MXhmwJ`LLUPjS{kuQzcOKYH zt(w@3ZupaJ8aAV)@Bp(>qe!;rmS~Q}#@N;S!9l3sekHG6p7CcM*y1EqSts+pWFLG| z(k;l8AnuyrF5n{0a4KJ;DzVXjT)DG-doEZ|Kq9}1;pKn~Fb~>o;A2c^uF!0jb{J5} zwVDC<-nSk6D=jZi>@1t??peEKh;N=2CwdYQIHp9=pA?; zWoX%xdgspLs4~??P|Z!K(bdx1@u-=_r#^&L=KxmD3fa73!pK_U{OdxzR~ zFCZ%HG8)vARy&g+%)6En>E5Kt-z-jS4)9 z)4p9yyFDUZY`P^z0v4SQH9OX|Fb{tv0eL>`&+`U)Er7hg9Q7KY#hyeYI&9EXI!r2U zeIbs%0n_*-OXp=u+9FHXcnyUvX=NDTN#_Tv?7)o6{Zdp?V~-ew8x%!OM@Y+L1;$9$ z>TvDT;U3x_;C{{5VMs4c8=~celj%01m+~Flp-e zaaOH96L43^g}6<&5=2w(6x8Xo=u-R(6x%+@Pzw8Ib=V%SVqmH)Z{YZpokG}s;xc-w zww}qwey({iQ{FINeU2M?(Xes7HuDZJ&vwTHrt@3ceLtntHGR^G4hGoF_D|8(F$^Px z;c0bFL%BWGsRPZgohDr~{P8vd4r`Av2K3_;=QGzvi|+t1&XLDI1BwzAaNKAQfe&#J z45TV?703PZR1<42R?-byqu$bm3{npoUb-;;k2cR{(Vw8X($1&PUYS2G>s18DQKR?K>L3W0R9W_oAdmsfq~JnfWZaE z&Pm_B|6B%Ai!b20Ev33)O2ktMXy-m24^A?2Da-E{8oT6qsAyf4I=MD&!Lwlfn{_L; zBR6Q^DTyyo21uWISuDO;S$rpWpvJO{$s2LDMb`fKPzV?-0q16GvZ)h6=1iv0*eBYE z@D!^wQCC})2oT~}$*r#0wMWa1O31+;{@GO|1E6Q%G1o4ArdaJnp8jCiI3hVZxEymJ zVAacb@c!Izo%`k=tIJdX85P{HCMvh?S&s~h!>U$*yW^J;wG=pQLgGDtWx%RtxCn zZ{M2kZR<=gZu#h=cXEB}!RBO+_r6$GCMQv)>0k{-aBiCxQ=%@A2|OjRaTI8>&jIBl zi!}m?)Xi;AI>Ch5LXR;3n<>@0`0$w8{3v7)UVl?kbunHi*d0m{d!38Bx*`{&Jxz!8 z(33TYX{N&6C~0b!bzb+SvbKp(@LUy(*W&#Y$rX3TaD6;V&d|M0{IIVTXpuDbeCEa2 zFScGJXwkV?HfOjyVeoW)yJbi%CfS7AcK$op;HMQ-xD0gKJ*?7kq*>g`a{&nns2VD; zVSQ3sWj(hZ_*XG39b-z&(`&#|G2ZuF9CUbWlJm!>kakpTMvppS~D-1asP zZi>wKev&|C8h7s%3(VCRR5?}-T^!AOrs}=FVM$=SC|65^UfIJS$l%H~?S`E3V5$h6 z#=(zB4i-?^`pX$k**u@ET4iS2@q4}nGLz)mXn>4O+^#?W#HG=SVOSgN*8^Oz93LD? zz-CwZz`reu-m%$zYFBFB(&H|I8fikVwKg+ZzxuO-aHvXw_xX8<4 z_6s6&j*rp{y*nSjKf)qu1Qxp(s(xgcdOn}g&*ARLy z(iIe>OP5|kLJu{dAT{(7Ab|kVdneM*-)aF*Vc3Xopbh2xR_+-c}5@ixChH| z$m$+&^uQv}0w}f=9lm09h#Q6BmEc-N7e& zg5^f42*^;jVyvRWrYud{4Y_+$4d^fUMUr&h+9TRNMOBFV@~2^!4QZPHAjDcH0jR@Vr*{8Ii3-Xybxu{0{W@Mr?R0!&vGSuUZ>UtSTDI|t_>sxx3~|r=DJhnQyjnRIA*{7 zejdPt%3PIdhHuPt=Ubo7i=$7J6Q9Pr67-BN*fpziu%`KLW+&SM+l znjzQqW@a__F1hKaiD&>Eug{7+U!K2zRt#AZh2ot*3p)}vQk8@p64eGr&C^)oWBJ&I z^>v!TKS-T^4#XM~)*isbj zM@x%kaFVrV=_bh5}VY5P`4K1UZ^ix6iCbIK2o^y}KT%#E1St zC*jt%ox|Dp2+tsHQyjyi(;`9*o1O~6chGGU%0rs`qE{#bAypet%e~2_y^{fWMplZT z?bt&`uV0Z~TkW}}PK%efj)`2WTF1*HjX|adI|@-kFnyRo2C(V2fVZkR8DZu$MKX_B z=<{4fe}60pF8rmkG-d}E%ZQCB&xq1d=eSbQ^6Ir_<{Wv(L_(rKQEaMM4Gw=amo%$R z%<*v1$z2=ZI66`S{vWucX1d#R^^O1Yx!X@-b%!`(XX>`QBY!rlmJGE4+$LcvWqF_$ zr>nlat18=?oUcCR4h@zUB>GlQ@9AVi@!%6Z8|MFrkN#~4buaL551su_56zC=4Po$+ zIV%r{H_s>-XT8ZUi}K0Xr4y8o<*bm&r6CJ}mH&J_Y%8`}4J_R8`?wRgjBHSN`ZeQR zOoRl4JA}a_PiW=^N%|3}tWl%LU5MDz(nizL_N3=pF~Y=OZEAC!SN}nexQJT)$$^e% zp3gBpTX{K7zfwk$ef3SckpLaCe`BzAy_49|jPK<*z=`$~7_e4vAyl0_K)tEov*z?< zE<-$o-sMYYnvhaCE>hxa)_rmE?frNe$-H#w5X?78X($ojWnMPa)OD5_oJJIs1}nd4Zrn?PNd&<*#K|`kV^MAfMqs6Y&ckY%3+q{qEs%ryHmz**6D^n zi)OQ4C4|+jZGSbu+DN$`T1W`ZdyKPTDNNOI~sA`maBuTNEn$lngM8d*5!4wUs+ zfRX~LL!;j76ijfK9zV!Qf)N`n0ds0aT$tX%Sv#)q5Z8_ab?xHAZYBPU3*f|k?=Uwl zVL zn=I~8Q#K$ZnJ`FM>~uGqOE4GR+VJ)D)imo)UKe(pbM8)_ch%djUaO4(m(wx`EX;bV zF#LftQPBnLM84zN6)Bd5FJON)mwV$i#Ls|@s#WbrtqM34f$g?5jYD^Cn0Ft&v|V-1 zz4y%S8)z{>9ICEGCxp+D#h^-UMqx-O;j83vHn?kHb@S+c>Se|{If#Ly^9Y`yh$)8t z2MLWnC%$ssj%VOy#KE^9^Tz8cG-veyTqV#I=s}QPN-sp?cNU+R5bD-va*}G~5OtOiXk#32|7IWyF8MP{$ALm9oIqLxQcpUmq)9i^e#f`ww)DXw(V z!Plc{47`TKDy@;N%EPtuM}NSlW4CpcGEgza5r~TCF24;O~fB`me^20tz=LbXe8WpdZi@Ees~4j zzZy;;wgKheakh@n_X93vEGagVQt2{TEu)~NRq2YyRG<3Ejol+V)`Q0mIm~v?o(G$DBrob$`bLon2C<%OKdj#GY1jw_9KM|p zo>N?VmhDulc=Gb%NPaEc!ql+Hu!*4JfIQxI5C7j9oBySow=)7#6~&Zo05$@sBz0hE zgkkI_TXaQbAy&|~{Zq2AqatxFCEGD~?{O|^xT-koC)%Ni zXMEMbg=7L3GFm_3wT)0~Yd_WB8#^yoRJT=^HO!>hlLT{x2S~De+Bb*B_E%%bjr>xz zm|+Cbts_TO1>j#aV7tDdGb^Z#wjKMJr&sPwucSug`c46ItL(GGF%b4-W3< zKWy1Q4(<+ca7IJ#SyfM@g2_)`Q1d`;?kyG?X?*u5nc2e7TC${uKgPH33zhdg2b`|E z>&Ricq732SBaQ)S7@#zMQfR7MeuInh=GdIp#q<@im0!895|(s3+h5*A>^Wk(glb&; zHv>bk2%RFTFNC2b)mu`CE92oijBwF6h^0}79yB`p#mNx?liunUpoaQh0D_xtOuCvj z=H8O#xrV0KHbuG&S43JPE$^0yrFDhPm$@At7D1**54)0xDNc(XnV!pJe1*&x4jNYI zUq-b83V`zS?#@m1A%JGK)|uqOSNh?@UHyEUOwvsQw3t47RQ2h8> zw{ULaHG2L><;yX;t*F=K`(VRPDK+Zg(oO8wbLwzx^_u%e+A7|)jY_lpHhd{gz#63B&a&*FVLEwHzS=i9C~m?6}g zYz|~=MM~K-b`)X2Z7p3o5A|wx93)o@>|w6lY^m}SNydxc>VK1>%EJK@s6O8bSd`w# zGmB#BRCc;(kAR?60>~D#-1>f9ps?JT&RT53p}cd0{aqB00}dUQxpVTHB}nbc*ZTBN z8=0fh*P5gZAa>k!J@hLwFV-qd+oKm{PL@goC?Pt0xUXMZUro31dWe& zja0DLGXmOq%6~uW?TM&5+Ci8@GiS!v<(Z3t_sXc>R%P&K&R9O4RUJKG?V#(%C!o*A z+f%oP^sEjf-12a;q)NVNzyJBtZmNg-`<1cdN|RS<`Uj#TPkz?L)K?i!8yHO$N|9_U z&S|LYXD`d_6S4+__J89QlDYpridTTk`;ViL#$8`IVLJrY=?zefC%=r?nt4{(liL+% z`D9+HDa;C!?cAfhAgw&Sig@rQAk0bu%bjow*II&WlXA{%RgJDDqpq!dCJny|;`Pm$ z>F*sz0+cL@V_iyJOoPkPFFQ!l zF1Aa9>V;CE2h2Rau$%j$HowW6zqLO+ixs*7_P2}lK)`S%{-qG}mEDPyFN0U4j$P%F zwp+Uh?&`CYoyXW9%Z7pDO*8&RVUicwj(%pohA znKp*{Zizn69u0oeC=HeN83QV!BN;t(93_2n;+Y2aYG+0j$xpxYL3tkl(`}vX2~XId zB(wC69>;ZkxUkV=pov{DaKAm}aPa+5fXga@IpjKhTs^M4qeU8BThpp_IG(xM&CPDO z=I+$c2LZ`Ge|e~zwDYBK_lqlGjP7JRVsvD_JB`ov$L>k>rYP~=@qA~v?|k>T`U|eL zEV7j!NT_vFtU?bgRX4YWQ`?V2fm)10MAdzJCTa&~^yt|h<1?m1BIDnzLOA&8 z-SLm-Y%bhMr?@$BY2IV*NaU^cCnoL5+Wcms4hCfP@vao3z;e+-^iNBV1|^SX`q2)2 znyc6-yggy$9^Vzyp_-MGx2;=~^m-XS!_Xx1Bve~iQ3IRyuIJm;<0vwzgi&g-%Deno z@paXzVZeGs&vkL{Q|X(FN59sGQX7ak$#JvYv}G1JN}qd2)szPo?i#*=dqPG>u(th9 zjF^vqz22n#U@XL2!#(tOeohV;)?Z@Z@JhNFxMi3%V+pw;g_c-)vTwM_zgR$}dqSw} zA$f(esA_!aS64kt3Wb4CPWCj<58TFRO09N1guhCJl<8UCoEe_%N)f80(1A_myvC*4j55OY-u(dQ z3=j(nTWvskN8j^38GLzcbYSUjky8C-)?BbYG26kfO~2X=Fffc}`;;$MYb4#FncqP+ zlT7c+w+27gkEo(G`b0}iVULrJ80&6#;kjbc+3wHaF50S%#L>=#(`cKa-cWL~L_MFy zeTURb!Qp0Z;WR_Sas`^gN-sYqtoX~i90?2TnE9pD%xk(DKDLh73q!HsGpEStfb4%c zD@LC$GT8lhH1wO;XGTonN>#%sMx7g2ZqFx_fDyGAGM8?i?>+ zIhUpVa;yV!96Jgn&XB=shZ(BSVabV4JV)xeWz6O+rXm`GkwlHBKVFx*8Cs?fM+SZH zXJ#2zveP-;s`NM!PxzG`!w!+RUGj1s{2W*LiKtDd=a8!FAW6P%N?h;`Cy_W_>i$$& z8Wa3Iafz5MKb1A{8sZ3St|FM=u_x zb{@F+rJgc>SN7}*$KW2O(i_%SHpF}x5eL9I=g1#IEH-n~;+4ut4Fo4zMiidRB)Lm; z0GqnFG#XNA1g=zQ5qFz49ZeCbbg&U(P6i2;uKPUx=7=?UbgOKjcy&;m~)$;Ba1;F%v~K>nW%yl!*$fxUpH|8(`MB0lY9u@f6EP>_&@!bW6VaO}sfNFug-(;mcj5}1 zj!PU6F&v;*0HFvgeAw1$zzS3}Q=sTAeyk%5e{~fBiIw#kaBj#hdLtnU)9hS+VQdA8-C%Y8dN6YI-+8Cm?=rWnLrZZL8 ziZfr8dhXol79v=&FhjW{gCs?%{6$Emgb0N+fssvVS6Amn7bl(e!Ya)nH{?@Cc$slh zj=GdYgOIuh+6^!3aqsEMt^L&qiCH6Z0RgO?Mn)4-?(iOF$8p3pGtkGCr+g2y727=` zE=NJWIICfNy0Jp5ud7o(_)X0gJr`=tjwv|d)rZ4z zrL;eHp9cwYCz~8XkcM((4%)pXdYBe*)Q71P{r9;$h1bf8Ot52&Sp~sEIF!xI-uK_~{n^=Ed_a=Ex41rrm2C+= zKJm=>sGRs=q}J6ps;Oc^l57!%Zgl`+bTuI!l63CBdS7z+yg}%8d(&D6?RAIT(tY2k z&z=iP=RTSui9hLgF+3*FP8FOu!G=3Yv}_29L_bF-zC#b*32w@ zT6^>W|8T8D4UY5LXg6uk;pw#$5&>L>ej-Z~EK}i&4d~pCMzNz#EA(Va?sY7b_fV^$Mh`+qtdbOsqc9x zMZmHzv2asj<)t4JihEn>q+hnvlD7)vX<{m>w#Ca3o=VcpsBw~tf*M6_g3I2o{qj+M z=$x0O+v`;|Ytd5=zAB?40c@xi&FQV58$Fv+j3+Nb5S5`|#|DE!EJ?P`p;88WtMXLNrj2{OSOOLXHX5)je9@cBP-!l1Rv+?!7)-(x5+xR z^4iV^Q9G49im7PmDk7`q(PV2isE<2oVd6A0#4E~VG&{3`Rxid}VBm`Ce3 za1q3Q?3t&d?|bx=eVgAspx?ecUUxiN7xU#;fST)$1smTEBWqGQ!hqtK_kajHtauFj zZge*vv{2ZgBS+2$$*u;vQG5G8-w}10Ruj3Fm39kerHR_L5p_EUmw7Gjava@RJ{mdn zoVZHH>r$7auuyNiTsPHq)dyCm2o@0F9^DFM9wHWYO<*8A9p@|1o+z@~?z-d6-%_0C zvr%#Y43{vt?i7V*yxAwdYsbh$k97~5y`8QjS)&HwORf3BrPd)S%t813m({HkfP6K| zlH-ag&l0!W6eHN%JTKVCM&{u4IIg8@h;a?k|Dl0&!10(cWF?DlX6z{O*0PvO&C>BG`_bs{~5cs!Wjh9(z71?Kf z=|-!qD_4fpG<=}+3ykd~>2;q2NNhZusv#F>z?iv-5*@e@dgtH}sfXu*Oto+h{Vm-` zE{KO3@?0IiTHz2wOrwl!`Gao6aOGxE6g{wdF)cp0YO+<@)rCwaK8mEL@BBlGy){}F zD{9g&wd-{-qN?Jjs z{?VlOXo+?F$niKN(_^S>d;P<<5zZE=b0=O7+#A;&CMoUf?UN`(5Qw&?wCB^}D z$|X)Hhig6R#&(2WEwSwuu>GLmg_!-sr}RAq!acsHl!ZIf=(VDICV01@4AL144>hTL zd+|nJG>)IsAa!5{nI!T9oli5VU`e?Y(3gg3lvSGTd=jdFrzV%C3O7RAI{@!()wS;v zX`OGL{#CvSf>$0`QarIs0DKGBjADF9?9FjAo?tCEGMNEUZ?IKfpY5GStL`{Lucdhu zn@4ScEY$f`MOzgi=P(eKORZ!Vf@`rl-#3!F4vlAN-KSIf6bjVcTj}#lKUbdmt@}K{ zR8Sf8@FH|Q1=n|hUXG6(OlSSE`q$~pV)N_E)kMvT&AC>OS2G?T$|c}G_TEkmj~B45 z*K!QvxckSnirmwf!LtH2Y4Ev)io zvwmla0ojydU0nc3)n2zDRTOvckSbmV5{ne6Kqq^*SODm=%cnhIE~mwZ@>zD)h1x8T)q^ouN}!6KZc z)#BFBT33;UosZU3w6U*DKJmI^y}<)(Aseg;=uM!a*w7n-282{<|H1`AD*!f2cw!pW zDsYABo<>6=TXe$_n78_C@Y8-l7@^pTrKkQN++?8OIX0&o?)CJa@qw;VRhV;9%(eB) zdv7h6tG(UMrN&%J$1ca`p5k%mUj=YrulGVgv3+VWB$-=QcewiHd9Un-f^8x2*dz~P zl?v%h*jUk4NHgt^ySvrNjxO%ia|H)$&?vIcZhjd#V)eNBDLZOSn!wUw#qWz;d}C7=Wd;6+bGr3Z_dtc?=TIkWXA%7L;xKfsvEgC zKf#))@aT#u-m4I?KW80Q|CR+QU(^cdGN@<>_uQeY<`eQ>*0`&XErC;Kr!RO|zlR5a z76T%qk+FYBj2ugzVpNSu^U+_E^1?2aRt6XNQ?(x*P7D98ZIzV@h)K-qxKFJ}-(|_X zTa(vP3nkezeKolLI!4C6t2r5HSt|y=C{bjuE&qh`rmeYxx~H_}qzt6@UV^)H!=V_6 zL$gQ}*~Sl2bFHFw1r76d{?&N*jg6^Z9qNHNRb^UTCUXnQCJCVKA=c~qPOCXVoJf3s zhfwV84l)d-;b0f5K*xO2YPDEa9b}4`H%z-imO zE%HEQpQTf4#u0O~gz;ozNRc@bC)SM9b^tbX)15}qOt={sRV?G+tA!`M1uyp|{2Ajr z!u4u(AzKG~T@??y8VvJ*1~p&9zF*;hqK509O)d!PASb%Q7|$I|eCfkF1W#r`Bnj9# z0?L-0Cz%&N=#4&@=uLMor7|wH()m`X%^%Vk4x;RUL#XpXCJSBB?l;$WZ$gC#`=XT) zlInv{rA%cqv@PN?kFKLf`jI+>D{gc{TrP~2MhW4_88mm0wt_pZMAdsKIiw^?uhk+P zEkGIRjQH<;m(m%0Z9lJ^coxZ6)y~m>^$scQ_!8vpqlc2>&ajBZ)l|T!trBP!pKzbZ zuHR==jwztRDB|Nj)s3^bd)a^UM3Rn*;Zk`cQ2AvyEPYSNfjcW*JwOXcfs4l3^@?g0 zJK}hG%jURa>>-~DJ!T#HYuuZ9k<29Nh1pX&v?FB=jNtKtP}Ha>+RleaU-~G9C>)_F zn^zee5ozl2adB5eO8~7eR33YTyB_;{$E*Jhknhu-Xx?GU!i%Q#ZbqM)AP>!sa}c(Q z$mG$ATUykkzLQ^V2&HMZ$5xfifz ziNo^d+NK?r-)pyJ9Al$O_kUn%w{LP?3P8b4fLs;g@1sS>YLh-D>|}Qi*u9}Uj%-Lg z8A#nJSU{ZtH1)b@fmTOY`LeJe6 zK6~XImSI(AF;>>af|`L~Hkv?M0iVQ9y<#n(m^VgJBSm+Rss)#{7G8jLxs+Z?e_3!S zx*IV&m3h_E15i;ipCeiEdrxG4lHU)ae-jv_HLMi4xRwpn1~_2#haXG^CgVXQnD|b{ z2Sq%Ft=5V1L1l{7-rR$N4vZEe)!RLY2by|dzMncgr_-wBY(5WSG4XLO9?BlffEGg; zxd3GNur0J%NXx+j(;2e1GKbb%=(A=v=L5ibL4v((U`fR3WP~330(K3t8v#96`MlqJ zq{K0P$KCS%ZO+e{whfz*Q(LJXibdk#_guo~LWt$kRBcU&enfdgucCKUKU0kBM7|a> z#>rw_rq0FdyUY7Xe>G!PpG8h%JvDfNZr_5`sbSUp_71}n^>u+(pqYU7@qM{K%zbw$ z&C2?(o<=>EFse8@5d(VCd=}xi`oE(t_Iq<{!(XjS2jy~j?NWN8K>Z`LBvCRx<>cph zs7|(&r7(~A^?kFG2PG5g$IRw?HtKK>5}?nPbZSaG>P>IrcU+@DHhsdnzJFF)wI(sk zVvx$*%<({lv`(Iq^RFms8r3#LVUo{tS^Nl4c%mIda8L=188(sKn8muhb{!uhfY>)-zq;RFEu9&b zX!KXRTn-xPH{~*&Hqe>PAhUY7f6lkPr~v~6C$TqW$rK`Z5mL2fCNtiud#)JOG0)6o z2gRJ;=brrQp5*MWzRaP@-o7|^LD4M+xjJu(h${D>%Z*+=V6N_f=--(?t2yVPuJ2I53EKy~uX!o$qkU{) z0C8uiV5F3lAZ)LWG5f3ym+@=Su=Qkr2WE;J-p)e$5LgeY1DLBPL%SjdgQ3%NJX`+Y zi={yBIb5mG-`EE!*TxIOa|X4Gq^&o+Dk$X$=HS$cdd#Qk|cvbR_YDmgJP!>%dEn+ z7U_(h0a}vmI_n7&qtfsUl7fp}M=t5sM6+!mW%z{4a9G@mRj;!!JTCf-xK3Ae?MMxa zoOn21UZZ{&fiz*pnJ*mnza<_o-JE6kStq`W_!d8@l8F0F1NH^e?!`K$ zHxYEOe$rQqx(n?-6NGBd2+!!4i|O3AO28s}$r-U}TXSN{{HAT}Nl)cUjE5t=XED)A zyoDf^_V0Mu!mS*yY5kr^F*uybwBfssz9Er-#wjfGl}x-E$W~<=;);#^d1}HOwZeXe zk@-s=qANNYphjJKr_eAg*~4&skPrN+45P$#+$YgDRe>)9DnNN0U(elz9hEu-)xpzQ zt(#5!K69UzzkV_ulIE_lXKmG2aP+ix=kkhVI z<#h}9Z)8ByCqqHKw2xgvQ56T{pv0x#8cTL|XTXXU?u;)73`sZ1?LK|)m~hFV{HDHz zSLW5ek~HsGWSWR4iXufLsaT%NfqVe`LAIQJOj!@9qR!6;2=FR1J1%t-C8wBu1SNr2(_`~Bt3tuXSZzmzbf^%;oUMEYk;%_@9(pp+%6 zM{5(5z1PVx-oncw3E3~>CRgpcW@){I)yX^XJt2@AR({~%=i47MkUyh`A-(54pVaT` zizf-{Y|T6e!Br}v?j>S(Uplz-NlBo*fjHzos9;% z@jT24Q#c8sb3E!><+vSG>vO-93)bOD065iirh9Fz7TS7a^@*FIj1t?#NWQ671&{TF z@GrD|cbRo9T=#S9o+&f#B)m}BU6vhhdo*v*-YG0mMeTqk@Tt=^$xE^qeM-FxX?}FN zfgQDuIIM&n_KgZH0v13Sh>$HcIf}l5cpRy^DsE8W6c8g9qfa>GRytT1Iw*mETdf@xyvauCAD4HxwPcx2e;KvtCh!}FoAf3roP&Nhy{h0l`4+#I zf>aCzd6ybnSFn4qFo@Qj5@^^rNd6pF|i!tJ@4$%8m3JtB`>B1?N`SxZAx{IzvJU~;qa}Kww{szA| zT93-hUxh;hDUAz{M}0C9WZ*z*=Qd&OX=_+}8MYpk6{-tlU3wyC&`!yN7 z!f2yi$o%W{mAj=Ti;On{)UYo0pHX_{ld&mnPR;y0^C8*%NRb#{ zZ_%FBO5!Nap{E7t)5rPyg-b>50ekVH6lpzIO(>-95-*b9@V6y&h(&Y1_a?di}+BC%h*U#Bsyk9ndIwPflE$uFt85}@aQD1cP`L^*8 zfX0S^z9!}j`EtLTB0o%@AY>+hHA+m3>_#@d(8?0h0Q^Zp95Sw*9Z>dVfQ;sUMzBHbv6DZLxJQP|r$}x@DB9?f8-d&l>UfR5I)a zIWl&Nr5LQ_QrV*8SX}{Ck5jW{6du@y>>y?N)*ApQ&|6=Ehns++coCjRls}x_L&VoK zw>Ru{YB|=$W2gd=>Z?_e#G~W--dmWIx{WzO;LgpS;w66Sx24n`o7Y6-;>^myCb@H+ zyww7|R6WICF@oe=S|a>K$_92KWYt<~4>B2)7|5|Qq+pUtVVo(KB=N<>Is4U?Mr)#l z+iC+XwRovxqW%S5U}cd?{~9mOUTb6?Hjc&Fy2nB@x|A1}seg`SK$lj$sJTB6g9slg zB7o8a_GX_on#Q^TWg8DxEPbKYeQ$KU;>(!VsmHqGz25CK5lxj!iqnJX&9QR`RkO)->PWtp{a3RtD=MJ{MKG*>sP5>}hL zsL7+OOGbG7`woO0+vh~dJcsO1>vO?THz*bc^2_w*)1n~18VjUk>A-F+p-+#Fi9VyX zur%uWVuJ};={Kj*6?B5N^ReBWET-jOlpO;8aUsX;41qY`$#4c7ClZ)4_-mL_kG({&kn^aUB*Oq8ts^<8w{vTYa|MidI zMP~w2V4;%f)c)XxL~oYmgjKO+!*anJw8Q8_M?vT)Ynbn`bnW=ct5V)yQbDEW^40y# z{(NlO0A*F+!&S81Sy_M2++D#b0&C`1XP#Mlk^T;P@Q>t?iJUYGnAO&t zIAOxxT#wPJEoz6wK0u>t_f7aXaUrHCW!98e_NMv7p+N`;= znr7-E6WQs{@&J9kwIF zgn*%(ENtP#L*X*>=mTeKnK76Z`pa42>Qo7la_o!iX}=SAaO!mOtMJc0_66j|GlFEE z7-VsA+m}-eRHUn-c$?==q*)(d1Ey5s#Az9**1F?OI$~ZYVkIPg`edTdP$~%mMEQ-5 zq=AKu#Hh!{mtc=q0HEg`xg!9&{-Dwe`54rIQdP}2R;WtvmcZAl}=oL`Ks5L z+{jn2cZv!)29Q_@TSH!&0v+HMy1)bDA&7R>FQLi?k9kn6Xtj!R>AD4<&3L^={gIjE z$PyRrAh2GBHHs8?1NC!}dcQ*f{!0kkHi79MjxUcr{cJ{28vI$ReCC?m%(*ljTnYSU zPUiDP2oaML3JFa8@nwyBXWIAm71m1Q^3G#9RH1q`P$dwM6I1|L9`LSnfOvR#_s9dv z`Y_gbgH|%n_+1C@?=rQ2K*s$}L(c0{7J!zm$8#$`e+F4TFQiJek+C8%kBzJ=ft{X9 zc;<}KDa??}>}Lseg99<2@*x;N?G_={NfmZGOq;OUbTxbe*Gxg_Hd;)a!@z50?N;Jj zq+ScaW8<~A!^;c8Dj&qGS~o8_BRON@D@%s!URSz++LJ^aV)GDqmxaeXJ2^EQZw4;S zE45rJZy`(1YPq%kV!+%so75)s;0wA_Vi^bRfYistO#!Q&> z&+%OmaGJD6E=AG=ZZS-ct6FuIz3S{9_r)N`J#}ZZ#~@Uk8ad27Tk6KA9HXscBodg! z-J~(r#s1vrOElQeD>!{V955fE%$hI^KwB^Tu=+D|@|R!L#eQ{4PUM9X3RX)7C06Am z9`I_%%MV7l?jm(Jk%mb_V2UI3M>Vj=PlY|__n`5A+{TZy^;I#gkOk$W{S=+Z``#Xw zYU{e~i7bx&AL=!)0Bd#Bw_=#yD3*T-`ZN+mW19 zifk#j-GZ3j{PmC^>#w${_?Y5PPdpC5wr4$uV{JYIq86HUi$k-f%B0;(v0uRtct4UC zKlL77p(?t){h#jYzZ$3jerE(syM$)nIt9joHg$r!fQMVYpZ`;v;}3|-%|J36U;^&1 zo9h5p$SwOz$e$jJ5^ylwDZhq+710x%HUPN!(RJ!See0WN_r#?8_6b08{n?EMpcvt5 z7X4Gn`G=7IXaFX3j=+Xb0A?8b;(s}g-znwiufX|VoI0NbtQC3pqZX)7>T}Hfv2+1M z*;GnwC(dVNe?|DQ^Y*7MCGV4Y3_XrvQp(HfCB z26Y*q`$Mt&%TMb(T*$@|9&+7Fxj$QJ{-lg%MVQ$KX!iWyks7ga!Q+2%0sJi{FaAe@ z{$D9Uug>P&xPN2%{zIo8vY!ncR(&%A_)(cQc>yjF%~1E>Y1+S@E&*m1oBKcb0`q^r z5RGOxf!Vybe=4?2l7N6-Z6px{S@u0UK>r{4LebBpg;fVYlQIa&rjA4Vk@4FqME z$^P`Mubz!)Qt}qxJuB$e0i#GiV~rU8)U0v?Z(ZW_(*T&W^nn*AoXtRL{jLS`FGZGD z1I((wonP1gID>!V@%}d-1K72*i0A*u&fsrVZgrlT{2s8i&8K8Kv?Re|KoC{WS0`yt z<-yyFu@dh6OSt}fA$+e1h+ds~$voiES8q8(Arw;36YDTwX}Q$d%I`w$O8u=_43~Kv zCx>|)FH3ee`~AiK?68ATl^80Q>#5$G0`88zBBij2`d&jJE4vTZ7=Bq<2qQ#Cs~zLV z?0ck3-+qs9kgg2xM_Y%KJ6Onh-vgHI^)0W4^!gNl^B?A>CO#=o?q?(O{iWcN7yrqh zWJ{sC)KtAD?CE3Q&Z-Y1Rd$jbYD^6#ouBV7iGq{e&MWzxfL37ePl*Br04;Tb z#?JlvL}7fFg!1|)^JTFpd#Cd{DBD=sFkgESU}~=XjN;cOgstOPXyxm)?l0#>{?TIl zvLwCvI!M%g>4RsLCba*JWym^P-k$FWEN?#}s0mPOII4L?xbyT)OL*GDL*jbGuSHJl z=B&fDdgUcOH=iH7Y#m{GuoadxZ*hV4sroG`gX)^*SOt9v0$-h&NtNi)uYt8k0PPR^ zwz+ET5v`iY-C)p6fv2Ycd zbx$LhYG+HijrLPtjI`*Q+$&M@HqHsZV;)tplp!N2mPkk+@614C42Cp&jz-JOYV* znw-Cyy+*UZV<7c&vCA0qojxe`Q#fJ5HHm-mN)%;UtF5sVIFtu)u9MuTAuo1sr_oh-)=i$9n-eQ#Da zY3=}}ubzZ44Z82rVhTXMiu+|99W>D|LkNc!X~5u#B!Tg)OHF$|R;{_!TYxmUpT1>|wB~nPIbj;? z601=PP+AI79<2wkIyw#d08~e})FTaBzxoyyPiX+ENo zAEwJjQ;6c+;MJyW4OoLb5`k$Hn z`$X&~C6EDKR==kK#8iH_EfpbGrB-oh9bV|5?ul+0mVWNrIX=Z`4DHM3GsVMLzzBEg zI8%C-KVHUsP%o9)EEcwOzhB`yse(z{m3fAem&|io>~HT}Ov_oQ3BCQ4>58*g(O&?I z+{sC(+8F?xXWq9`4Y}+VFmiklqoESn)9UNg2{*ToxJ@&u)K#E$a;Q;X=WeoXpfkqd zvzncoEnCJ6j8;1`ECQk8!`L4N%vS^mQ#?AN>!+s6m>)&=P%{5Es9;+Cn?}bGSRVN| zv(?rqds_du_Y#x6wVD)XruDHvVbtJEzClNtFESE}JIQZb49eC`=y!0Rz7!>39rp$A zBVu09{mVWwX#njgG8_tP4Nu|s=)9n6NMWconLWz4!$ zm-jYJCg!ErU6p#)aC?jL))^mh4h5dW*VCP`Ybygi4R)CLK{zv({}E2UUDvvLds;Qj zv@auVHUb3CRy*-hFEz1;`nn_lT)^G$VfMiME-=?vrK7plsqWC;?9AM~tzS;cpQfD3 zs^qf|`oGwF^LIA)uWfiwTBWMHEm~^mNR1r~Esc~GMU^7vP*t-~Qb~#$J?`)EzWe*!&u`BU&mZ8px8hG8ca6rdjzc^F}c`azZe4F)WPcSU9@q|;? z8B?*H^ZPl0R^qUqn{uyODeND z$7V883$q@ZJzof|oXnmm?HYPxL2Vx`@4323W4V&Dj(Wct&VgpUo5*>c+y1oxr7LEw z?m69v@HOByfw}Bkr$ur~=vg?jJ#p7`l!!=tJLElKH$OyJ4Hpx5K6hr>ntTq-lLs%b z=mtdL5e~Ug5a(q@=5aw_E|RG2QDlEI9}Y)Siu~_FU}T#KPz|a9ac@56YSuQRZf#$V zPYvTB+iV4ZD6B*A5{~kIq#dJgAylG!;+?e1&=haufgudJ7V*Ft*6PNJ4@756`crXB z!<{xTq*n{dgw1%(x8*3JQ0>CPF*Xrv?Hxnwsg8kwFjz{}W;nDj&wmhMzk0u93(X;x zBk-9j-d%PjUlsyV@L3Hd{>ZQIPThQz^t2N$T#Xp+At?v5J~fDIYZ@~q;ZnIcQ4^sj zPxt4l>^?DLkTwc}2n*`0n$AYD1x_!EWAakXTs*pxJ^y}ZfDZcgfk2DOMTI*RgsLeWAvDB=ex5M16oQOzGv-1z=d*hTBHY6Va zs9+keJup~-;m2d?AwTjitS|_CEk4NP=OTIcT8vL~CBc=DZPGKxK;R&Mq@n0E4IjYZ zzTM;?_|?Od^@WfeNGIlFx6yQ1JpKxsu--a&V#O~V*I$B0z348sI(O_t-f2tKYmNs0 zcwoPO7X1kZ<`Vw~1LydIhlc2%f0G2J69m6PV{Ulp6!#X2N3=$h8*LO{S$@2AX=k?6 z@Xs3DUmY~z>+%ByDEe>wO@?&=ndd>i%km(<`6CV@c6jL-;{7Oy%C;VA1P_m3x}YA} z*mw@c-8>tz;^2(g`!%{;q%r2@i7Oo)t_ICY%zyy!+gbS-es_{DJMSm%)H%Zt8)tL1-1az5|^NE$f#F-?}_=Kh)}w$a0DeSm}l8GvQV zP9FRn1gG7J3HjbFz>p&(Vy|p}w*Bpo)cBehG?|ZV|E_GJv*-3bAnkLGa;=FLuOWTC58>B-P%f05$5+>uaQ(=0p@tay>hgY$FzAH9D&O1)lGnGN zkz%a!>}0_f4_kq-DYN%`t6+~0WIgnAikI*_&WJzb-0*!5v9fp$yTz!YA)<0elT!(d z44S>Q=Z!MgrGy8p$?xUm;}{ifk*d?*`+(!bK_&+CaJ-3!Jl=(KEh|fKRHEt;qT>^G zC8I|&1U%1;-A^7}2T;*cFbcD|h$QW!7V(RGeM$($9hqU*=w<*HUuw_f~D9_}>Y=5HePC%*V6JEO|yJEcg z8h5qhT}8GLNn;$|J_&hT5i=kt2JE&J){_B|_ag8*YC4Mw{>m{EPe0y1h6Ri=+(A7X zvnu>BH<249nn{pun~DI4+^Vy)4Et9Kcn_@JF2K6JIbz|ggK4mfX-1pSDY$lGdn1xP z=BxCV`omst^D6yd9<_Fk+XJH?+26=175rxV*C%;jTaiR?b^lAqi3(@7IOu`LoBJ{L zC)&YHGL}}LGU4m}?J!SpEOjnk?GSK|ejB!Zv`ozLi{`70C9+;OT-m#6FDZI;Rrd~aSIv&oBJH)| z)Q%_JrTHD;D_{4{&m@W~?@dpIObO`HX#`Vc*w|R61zRfonybSbF9mpnxKm#Tyb*{-reo#C&#RB8mscVGGLLoImgFcS@o}OuOggL#+pI&;+%o9v6Jlf-yE4cKLvkC z3u-Wyi3Gfqeub^DY2&e+K#&et;$%}%iHe0onY*O*l8h;iOU_}z>Je6NEM!}&D(FIm zz#aIZR23d5sBAveLZgniLTxZ`x?zLmf)dE2bl9rj5 zC*k-cB z+Oo$0pSs$p4;G?*xW4RRPl`$R)!gXV%qw3QsFGKiT8R!Da1vtqHA3w@9JUH`c-JjR zqj0=;7^vtdHF#MC<44_`WuUVrb%a;*mFo*5v&s?AAuDK`iU!QVS=6K;vhFc^NyMg&!p-}4v_d1fDZngaXo z9X+56DTCu*N@uG6K}Sim!!Oi1f`k#(_idNp&{la(O8H-iwFj+Z$gak?@oVLQUPBaq z-D~H$I{#cN^L~1KY8C=bP~hTxihpZ#y10KU)^X+Nb?z#$A7bB~KfAhL&4+}wS4(AY z?tg?Dxj(EC=wQ~QdqaI?0f`w4+t^yE<4N;`W0<*Hr#SaWJWNkVC;VaL2iRG0p8>X0isIh+W6=WL#w&5=yuImJ?d^GsQu7zEz^bR|{%!eKnW?$==IXkXKyZ2zrJd1x~YpAj<~wOpAg13^O;Q-nY%b=?f{Z z!6(rn#7&CC_*1tb+ncwO zNbgqg6(0t!Af*myNsGR#-N-VS-ncL@_TC1P+KnNPaJElke#Fr~qa@usQ-=5wN}ywW zbL14=7=c<=1>y=~3=_23 z29)!dgjNLIL&^fcXst#UCMiZqTauk) z+5ag9WusTTu?Rgcx>xmmMzH7l3YtHSAS6h?UOF^7&MlhF$|Ov1gb?lZ(_(}LMu%8FBTGTMgLyOGd)P4dznwtT8G6b_)g0GuG*_V!#3 zJIr{)&7_6vI8s;sJu7pdl9E2KqM=R(D*`M98HpG{k+%q(=4;&VI7B8f?gVqd$ZLx0 z`Kt$(IUdGwHO!q={@{8lx2Kwtf{S>m+JLaDer{%^SLQrm zrTmBM#Ebo#jvjMUz%|jIQ02{a!Q$PWZ87s6th9UL_l5R#?4i6BZ@VUomHAL?>Vth5 z&4x_YS5)?TkA^#&dLPxP6;iK$OrIOjcBgKKiD><;iP;#l$rvc55okyjT0vP5ImWKL&5I%0t`-JW~;ry0&BHlqJ~|BJCp`NLpS5bOD-#KkXkWh<&Q+j(~#L#dVaqqjG! z6uva+Kh!QQQ?Y1r27LU7e9Rw-L%%`9)M`8EOUv@f4?pfZdpiCxa6SKdQX7*p5Z$4v zwa?GGzg0ZtWSVhUfUwPS9=ct^_ZpH=yTr^YJ@UF{oaYjl_XtpKDb8&Y7K|Ga?izOc zI(8;L!?*6ru|LZHqP*ylI?(wzkdxiYbOr0z^e^f+IX68w7FP>Y+E9G8+G#rRobxEF z|IdTKO5va0FVy!qiDRXE^dG0`1Q>gT;^u{CH}L>KUoUXR(L?4dy}k3@GA4h3ir-gruCHuH$%U0-uz{bU`| znFOgI>>u=<{Vn?pX@cEVg{5dn<_ndOMA4FRFEb`bAmM})>Wl`fz8%@7PF5vpZ zYlsn2D?z;$JT$VkA`Pt3&ye=35*zLnolcZ2%KqiE{i(fX$$HC%bW^VTTgd z|J%(b;Ha%4q5c$}ddi|>x*X;#f7(5W609s5snyPm7^672onjPJ4(u9nEV^iWq05PeV~%M8;; z(M}46WikKnIT4Tw%ySi-w(b2I%jZZ4s#o^0^KvZ9ZNJkivrY8nSwfk0_%Hs;K8r;%A&~|D53TaoiHurut)G*PW@uot2U%C8o zpe-FgGOI8yU7N(t8AKia+NETRPZ#unlv?U&OQ6nvpPCp)_(IN?Y*{?oM!ioswsK-` zg}vT;21#?%@OU-v3@a0Z+2{#*W85NT5M|ONwx;5zX`Z50Xe(M@PeRnH@|W27KD)Kk zHqf&aIF}535>fA^%b3G1XY6oN_U=PU$_;tNg2(Lm(jvseDkfw0owltzu$u_Y<_-Mv zjHH%Q(~YS9@J(#Mt+wNeLaNdo^T@W@b4Q{V{avg>zh_a2HP5!^Nh`U0Im_*vKbzr7 zJst0DjTBYgl;jvd6C>1pd>4X#GepcutBosGsqh`BqA8`+oC*(wed4VdI_-*2Z}$2-A5 zg*VAkIWjiox~V+B_-iVTSB(Q6_N-7f_dFkQRlpHTlrynCN5uL;nDy5JuA=Wa!F}(G zCeZ=OUUkp!y)G5;BN-sNc6iy{-X5mmgdJmN5tz@Z{jlobH~G75#V{>ntK+2c@>sOYKvc>mxUL zR*_FPr9}Udvo9`SM@~>$k!r!ZBdREwvsS@k)Gy6Y(e=tpI|HvDh^R-M`Ol~&O-#=^ zmLw;Y1iy|KjuQ@m)$dVdSz~`EZqpaT6jAq;@EQsBDNXn8lb1q9ov1grkj&f0MZcD5 z%W|2wxIITOnlFfv#6n_?kX8@8(LxiD((#dYj5*5wlS{ZZaGHDEFY=HeeB`)RK&{Q~ zPq6@gdN_0b^NuN@TjT4d3JG3?+U0s%!{lo{PXncV1Twd0*947GVwbUj??wIOfzLO@4U18Y9U-!e*TkCDNhKv(8F( zvz98Bd)S$MH(im;s%^fHFA2yD~25gwBnQZ7BzfN zp-OF+np1%7f@WXtBy9kAX?snx)4KQZrtC514y>BM6o=gd?hU@YZQig=`Ot5r6|tyB zzDI3|3F95rcUl8!T%~G{$ySdR3ZGrAg4MZ34{BJMY{=hX=XTg@QW$J_g_tM__*p>h zliCuWZr3Q_O06*tHpRPqrCa%Y6r~XKnUsyptjNUZ|L!|jekq7#{HKNBqr~H z2*l^C-B(#kyY%%Y@{ZHo8&4s!mGXLhv7{qU zNRW4^hz9T@S=P-3UPXxXcl$1$bo+$dnLmZjrdL*-IoE&pIsdscZ|>n=9yY{Rx<6#R zehsbsYIF*rcI?e-B`&XLX@C)^HOI|DWUj2YZUjPu%L2QREA6%uOPXa%oCM{n-s;>^ zVqfebFuCWWyaXs@!HwpCP>4|tnriHQI_`5so&)ZbJ7W6Vn^g#^~^3y@YQ? zQXDsU++#t0|BS>$J%?z_qH)fS(rbgNoY*tVqfWa$!QJx%m_nR40XFVsI65Il+h0ap zKfN?o3}v1@N4y*8d(i2_l734;#`dtUS7gzKeNN7+g_mhnd)gpm+Z_a~{}@?x<4-o1 z9U3vLpnPqjo-tC_{NOpr!0O!^%1`e7+GzEH_Q=L-EraWmXf$zs@hoq_&P03%?lZcv zLgrmWgpd8f!|R$MnaiG;TGco=312hIr?4j`_B$VvyCcy-5RaF&^8rpVhL9lbFcwY# zQXo1k!)9;r<)~sQAo&|5bO#bwTzg&=`}tV6r=d=fZfCb+sx%6?nmBJlM>`dQuwNam zg2HegEN!ll`E`nx8Q-!Rk*Ux-9V3FKpf?=?N57>Fj zsw^Wd|E}Y?FZ9r^2Wm@D==%;?q|BKzw!|V#$3P zcuec!_`vYgHs3mKN1op2aSOSRdP{e?#N~{_7~jJNV%2g_%)CYm zLhh;Ls<`YQ_VY14H$7yi^5LZhxKZ=R^V(#iH50LBiwqE~&>{Eo{+wzMwI1z4F>S8h zYQw+?!%8ikq?t=$S=Ts&O?cDKC~fl?0m)IZF&W+;G0)kr!V-8 z(p;pft(`I+GEl3pi=mFC_ey2n>dLuDM|_hC4R0WBbhFc5>GiW_1#~Y|<5}Fek-?Dh z7hMGx%{FE;Y~%3M(F9*DGTOGn{B5^L{O~htv*${ApI+plpT5xI&xgFUs$Ie%ge`8! zzI=4oSm4i?MJs`!M&5AeGo|Y-0nJlZ)Syir+^+XmhN&ra3Jj*0nELNUk2K{!9;u76da9pBZz4RG1|TstuZ(W6uc+ zLHaH9KtfSwG1c4u%$|uvB>-u2-A5tH?1H4i&BI2&9)`T``%+r)X5HP@P2&1=5Mgar zYtI`ZYaWxJdqzC!IB)*AeMn|OnypdYDO=koW|P3Oq^^brEv(S{0xO;5FFr1uNUiaN zo#{E@hAA8dAK9VegGtu7T+oxN z^|g8@qWf9z{fGGpdy7kZ*827n3-KOH4P>XR`<>R9*}?(TDb-gB{97vs=@Ca?UbY(h z#kcE|&*VR=`q)fZ8>Oi&2_kk!#r6!60p~W6gU&J({eY-`0vU+vQ{u=#$iDPv}S3O7k_P3;aU`H$&LpTN07yiXEDA z=KTgXGQ(D8ngjesj1-2-Yh;!9Tjx39WN~B`$SAGR0a#jE4vZ-c%-^G(N~-E1eNnFS z(Rp=6%i%rDs+$9W2nuU&9_NV&7|5Ts^*9+ec!2wO5bz!l@Lc;hN;kZd05fQ2)C$5F)uuT61N^C~$$-hl|TgS-?kB|Z}dDpRYQ`fRDhm5rIO^>Mazt$?l| z8nR4t53$h8zbHVHIW(}~e{X+|cA(Ot9gOtEPfP?x$U|R}gpUlBJ-1^O$2bK=_ueo+ zY_wEzIN9!iSCK#}<(O~dXdB)tsSO`StS|T!({?~s_zK-Ze3YG!8sJ!;u$wAQPe-Hm z48UT?Pry44^wo&6H_W5V&M4P@({~JyacbY_k#fuMX>jga!W@2&Zl4y{)E`t{&9`>6 zfAp;mtpw)n8N00?HN8sNCsmCC%CLD_lv#VW>Nj8bDWmx1?+L%DLPJ(Eiwp7s4_m5J`K2*E+< z1xupK2F|;>i9(BIwx?9LL;lFA&eE)*ms_Y<*x>Mj*!U3gZ=l#fJM+U{WJe@D|5@3t zXq5C_q?+iF&Mm6On2&@LOa*~>d&H-|(m1X0BoK)7{|v;Olmxc)<67_`OQM<5S_)J> zB)=|TrA(z^yV9??2=@Bet3meY8Orl_GM4y5W&ft|%>l#*`5yd#6~5}l{U-zRK~ zjbEP@`zaRPKQ?O;_A>TV!ZoiCWdfoF3P4)BQ`oL28QmWet*(v!WnlPFvia zLEUn|mkZxW__Ivj%2>HBkA`BI;71E2+#`B^M3@7vW?{=&!Mlg4?qwj+(Z!DC^d;W_ z>~`tePIOyy`Wab!i&xTE1KK-;C^Nf>&a#GKC*jQo7dD+< ze=l`fNLT3Uo{a@gCtg;nv3=OdPm*_?j6x@zR}qRKuha(zx%KOk)4xUY5au15KyFR; z&Q=J@RA4d=#RO{wQ@aMyrm&OJl88bzuMwGy;^w6QdYQmM^yzo83Hv3*c4E0)2Skg~ z6`17#-I;+O|He;+YYXskJF1AlI5DV1QAZAP-~8Npp_mT^j~4X3S2dv=*G6cAyY4EMr;S z)#oaf>D^T`GB|OZtX;*Vc#lrGFgi|bc5sO^5l|qy?D9*@yo&>{bK`8`a#>kL{;sjG zsUbXcDWyWsIz{Gfl(E1?&!T{2c^kd<)>KE>+{@Y#Qg90{)s~GH)K%{DY~HXi1-F-; zKqjH;F6tBkyTH_X3EXNjE{J3HPMvJ_7iM%8i=0<9IV`Ve4ZA#~Y;lILYf_E={-0n~ z!p(TgW;yudNDNDBn;)@Ru?C!)vQF!!JIPo$Ew>C4yH%gf-K#RK(vf|?m@Ny5HdnC2 zr3_wWl;q&u%d)tkrM#@;ytiAR?`FKUO9ltaa}!U#N~`Y0oxd{4%X)j614MgA-A}M3 zvJuh__@;YlG3Q~726iL4cCqqLd@e@W=9?Uuiza2L`!-_BR`S(9tAN;#$4OJ#9G`TY zSJh27kpsG}a*w0_aS`bg*b!jdg{q&@DSDgm^E_^u1@qyo`bJRMAbPch=~IZf;qJ4e zSz*J>K-m4kdILz<_~V^o!0)kZpfpez!6FJfP#eyLBO0ufYK-jOqEy6x7^45M_$CAh zO9t%Ass1LU4x7Q0&(CtU4cfO??`)T@UiD^*;p4q03Nk*5j{C7x4KjWxOwH6F=T;X1eZeDI9gB;ku*kpO(_Gx9hpMVc7) zrF=%AHo*{2J5*8%HyoJdi(_?@k42kjK%my+aM+}B-Rkl;{L6^!ERwcoWiT~A5k+8V z6J4lQI%ZH~RVKA2QAk>I1Z=u%dI)VdQgrWk60d}Cx8dXa*$Ztlo}+c+3P6Ck-pSA& z&A*w_Mj9!c|o zkwfS1WW2pANueyP>AJ!Ubj?Fy+`>JARQ!;YeQO)| z^7)GO1d62OH4hy#ouPy0ulu87GQPKty*fDIv?60!B$YRVfznt7dU!#5;#`mw&jaYy>C3Z(kPpD$*|q6)E<{os0EyT1KIvEEt^``9o+WK7XJAY z7f-`uH{qlD#X3dvm&)~tLr`y02509CB4$v&)VKz;w~NUvphRA{eq29w1u^ngV}hwt z)({jOyku8l^y0TuTJ(DBf^=i^zvh!n!GWjR`E&>?T5x7l&Z?xR)C(cq3!CdN1;e0= zc6r44-=N|Y2BVI%&uLA8O^_vCYK0Vx4v>)jYX+l#Jz=|={zyz~Nj&ksSTX-V0AJ4F zA#)v>_;8pGdlNJ=sB4*{%cn5}7%s?74;C_X8iU_y8w=>_v4U~8EM?R^Z~ZPknw%Jf zH0RUcTTo2^&0f@6@8vtywCoCvyuqQ3wE74(w*EyyFuIK?D+f8kJfjG~; zFp`h276<$}hkeD7b*L=P#J%<`pOA5)%tOo)XC5PFc|>qPq<%Phq1ol{AUB$@Ek&00 z1}oQ_o`u%n{Fid-f1Y^U%1>3U`#(ECjoQ5E)LultliLqyJb>e3Jqydf&>4O z&bM;y|D7;T18!#rdakoRNwYWMNXk|)y}!wNwb<6%HA5X1OItx-n{Pw^X0nc&=Tc!ULc<%K1E|(9pRDy+ z%cWV$rraO7y|;Rx@XB!l2jpuuvqrwLZ+7hJ)^Vxs90?*s#) zjdZ+V`G9EXS_dYnq3!Rvy8P>j`>t6k=|L<6;*o5j zYJXF0x~#=<^wJ?oY`!@<3$M?E6iZ z{A90fD`%`dZdFNk==sJFdHhj(Go;ZgvtXgWjtC8Gjp{H9la#dgBacewhME4_no^1^ zq{yG=yOFzb4wSerM-x9Wp01bRpDEk;egoIV{UMm4xdnO=s{n!B$;o#}-cl{I>nY~T zOyJNED?TuXBG{6sJq2(%U7oQ%Kl(%%?AAlNcnNPhC!kxrYZn;O;3;=Ip%%&v=krFV}wN&01vPOOfs39fb z)ae9vU}`BJaV*prCSZ>n>xby>f-SypE_(Z^;(`W>2o!WF-+T*xxw3PcJPGD&hA-Bo z3P;ZqyCtc%7%Tfa*zlN@O-?;>Ue{cscB4do(^HYrJ=OD4I^#HAVfd3u(No?>a!cNQ zlTy8923jR{Ye*GHf#{P%_^EF7jSbs~dtO(xzJKDtGJ}T$*q_ant9L8n1E|JxAwrU3 zrgdIU!GH-wSqrMw!MnLSOwn~W007~*pCD{Ti{}Mis$e%ZqCXxCTj{5Xm(;~(%}R2Z zR0UB{#W}8K&~}}AhvXS?fwPOdBc&$ZB&2)&hb5q&w~BHpnNN6PqVAb8D$K7tLwgwh z&l2;Wz(#*VGO@oQnG%0~c8)|BQuw;!fU@j_sl9)z$nr%oFY!m;BX-VjV1FFa+WiAD zf-CFLNU?Q0LGbvT_6%aL>eMLlX0#JHs79TM_{Cyp-Pg#8Xg!nknXsKRyY54uglu+HUs^ZWdRc3VY& z%cH)%Tn8tnP`W~mTqFj~{I(IUBWIF#qoa3n?%{7o(@vZ*Jr-mewty}`8lABQ`gqMm zmeyy!qVsyQ>RNdyC#{{&I%zDU(L^JWG{q8(}zLB-2UlGkECr937E)!`NBOj2iB7X2o%}Ll? z)m`vtkd%=)3$$8ZW>@&h?Kp*3a5tjKs>l)402y>q<%fJbD4>S(qqcP)hrbc#O=r0O z=q#gf_e)N;nAH_nKrZ4^uFr1siKGH$MU^DqUKDjF$Y;8_0${rt%ql@}*LJrpe7+%F zt}tA|_sSQdHe#vyr9wH~Vum_$UAar2PsItXI~Ogx-g5G*^uHnz+{#xV5M_9iep)y{ zX^qcc?9!W1c5!!sr0V_sSg%B)O!6q-}eq2N&Xk)OYhf(tMJmCz_EFZ^7^m7 z5WN8z{+Sco|17BxiHiBV@2dZI-__-?5m;~NX`A-uJ18Bke(p8|n|L1TO zD{2+mIBaTHqYH#%?nLOir@I@PGY=AqvEN59vheCtd;DdA?}=Hu!Ko8`euGEWC9HK! zTrf&CN0}ctno7thtAIJ+a`SQobbfXyyv|u;DFN!-K6W@R z>mv4!Flh0y^#Wai>Nf$RaW$@nTc)(D0w1rUTcSP(OyTplTu{_JcF3zguR=(h$!`J4 z51(peOIKn_i8;-eEoW}GGa7v*SmQ3_@E1v22Gakf1+bcZVH)0F>|QHi&TwxI454XK z?0ku?O`*9*j{#{|Hx2!K2uncYJR0lDRC5DwR@_WZJ_O#q6~c|UQk9i%-GpsKY*ijj z0QWOk0w+DE1$7q!Gfhn&_7ld~JM*zwuQd*pH*q=lf?H5j?k zejK><-;)3#v*p3X9gySoT)9H3g$X|qkxd<|m=Qz$vD z`wjg|s;W`lum{(?Km}OtF4<^a`Pq!C>n81T=iQx=&qKd$v5c<)68|d06DOom>z;1O zL3^q&%4g-E$XgKH%TlmT(ZQpp|IEbRZvT4-Y4BPIQ`-7!u9gdnpwSfYQ#T`f~{Qf2iwQk<} zzAavSm-)aG6&K|};_PM&g$$2k3H{1}_qq#>PJ4+JGQE@ihXCEib9g3*JI*>x_T=tVRbV;P%+;w2(I{5MuMTz(ngP%18Ud_B{|Vn+8aKa z-Ko=_i|6z!U(cVipIcGQwRxi2>3D2I?)}c!!U3P%%`|3vHmY`mC$4mTxkeoDZ|?)*SPOvM$hnhM|6{q4pUMeW%}D@r_e}yIT65Lye=L3QQxwKO zF9Ki+o&$6j-ah*FKlP?Q{L}1t>NY0|cpe}rb7AtG(7$X7{OeEq@2^1Srvq}^>s>zJ zdBEtyy!n#EzuAiZ@6r=DetJE>JhLSTJP$CK&X0Cc|K|ve*gOuQddc!2kWA|NUG3 zhbqkf{hOZqZWX!R5V4g5xHJz}`4h2&_HG+JGG}Xicp1ts8PDJo%j+uK|!lK9x zI?KwXbrDaISA#tX7PrMFdn?wf+^w6Qa|#{sCHcMQtWy~g2f8%u+7FE(3f~$VYfxz} zvLl^-4K9cvKlcm*gGXrV0s6~5ci%6QBCz`2c2m2-FC3FKCO9K*$V=@h4i5~yNpw-Y zo*FsXJy}VO7yZunl-UFZs{C0}fnB@^nrfASCCQPVyPkO^cX4V%1-7_9+m2X|8QJHp z@$HU|yAC|8q9S;yHeSgIHY$+Klk~e2I(LTfa1+GEkc;fw zt(>rU+=$Qb-gw`sR=lQWnN(;;D>F&(6l~ZBru8s8Q;Ztoj_P|~M#43@JJBb6m%9%< z5vD01-h3u+K+j>P;+&i(c&kZDxWPB@nb9Xl3Kh0YAPa9^;JYe2-#X@d`%7sE5=!$a z(-{%|rJZrv-PM80gx0v(ebV@l#v%FLs@6KLUYX^AqX96bUoi)hv^hT{;2Xq5`z5S8 z9vKE-H!is|Hxv$Dn_N>FeP@tep-lC2B`lXi8x403{$ss~L}CA~!oZ6D1;9r+KN^!) z#aV05hOZu&W6KN<%}Lq+$D3uqOY^TB@j*jq zXC{fh)<%%t5P3zZyKk!^c9BhJ%d62Y|7roQPN1r@oJ&aFppQCo*p^qhyLOx*>kGA- zWwkKGmXh%9LdS_{^H2IZ8B_a~Jt>9#FI$7h-nu{s_8Rj-Md#`jpUVvXaNwOK#UK;C zmzo7q)KNg^&bUYscuOQe?Lu2p`yXrnb{Qb`Am5V$%vk$>9>iKBO$iNev5D9V28X@Y zu&{-RW4^b%yR|`@HW?`yRyJR(ODVm4xhtw5--@+*c&{T(;Z9ZU1b^g<7WS4y3FLz) zi=oJ5S3WVL;zvF+$8j!AqdkNl>ImM{LPULsXtaFox@)@3va#S%iu?e2Tfjt6LPh}& z<=2J0{Xwzodw8$$Xeqewx!H?#c1N8LKyes+X#0-`hG+RV2?A(X{7r%w>?t+rXOMbo zyh?FaB4jV)n@s|=@Upu8=OPiK3}D*z($>jVi8cCMK~b8$*8}GDL~O6g@*J7&eFs`X zl&gXu_UH8w=aq~#^hw-VE%q8PP~C!fXQU*nZc;aq=#;AEN)daVnBZXsw2=~E_~N3; z2H^Bc*I-eX21{PsfaZauwHvIt@={^%vi|$ayu0f zPD|V5%1weBY6mwIjBRg*zbI|+iP)LZun$g7f0k`fs5Be;=%w!!EH7k&=MkJqTH1G9 zf>JY-;OSWd_0qb66I4`Gucuz1-*VfGwzh!;qO$1Dy7lx8=YI*Z2FV}=8ks`F4; z4AKK;YWN0kb*cyc!`6DV#}}#vBaelx)FQjT&VA!_-_;IzOEutCSM8VfQG;SW<^35E z$DE`nkXK39JT-}_utHWu5O${#PUj~-V%xuo93Cxug1MTv(cXk$ye+7?A+WpeSKK1=bOsxNbiP0M#As73Wa0QZ#dS0B%HkO z#GrptP!{`ZrtDk~uciC$h<0CK@TxGwGb^zW>Dx(o?z!cl1&Y6Da@Dt*hC3ua3F4hz zcUm1_ICXPM!*fT1WSws#>X*qa%$Sb)Ac71RQQFUHxfS)>5qFyW7qmPx&IP zu*6Csa3kt56}-5w#12VhH@P#H7$i39*!*ZSHZU8i)#^LO1T)7*8YcCCIu2WPM#QCZ z{fEVzsFK1jYl@}Ay#(exp0u(>dfv1H3PFTUmg_qx6D$CfG#M7EJ zaibl_-X@J^Sb*10sUZuSUD&mpTWsV0RViU@Pd^gNaCS-_lCpkE7KgQ0Kazr2eKtpr zY|rkwBZmBle0;N}%jh&mN-j$uUZfsvA}}xg{j-ryxv-( z0Yc`zXW~U`e zD7CviLdtC~?(XXvhWQYo-uf{Ylk_Xt<`GuT}$c;`FW<*pELhr{ndQZ)CP zZfy^YmyIF$kdog$pAaQhRhrZGl(Eve7PE_w==$Ci-iU-R-vadiB>DV_P1}>A2^D z=N0~{Ux+n1DkjDxn0I3)YcI@>^k3Tf1nf`+s+i)X9lqQ~O{tcW>Lo`C34V#Yvjl1O z#`?EWryY&~wb9xiiVx^>xxJH`PGEI@s!MxbC3|`JeD>`<=l*<(FUo;9C4m?WqHh zS5V0vouU;wy0cN3Ll3o$$dkIc_l3H_E5AYw2apmP{vU_K;( zB{km#xJe=BOZ*2wiR=ujWDU%hQ1?NP`EJ&ivQ!WnY#q6H-CFHJar!;x^**vo_T#1% z)ZR7Ta5|>!!4{2FGr=2dWdoJe!l@SDjE6b~GjT)rfSd_*dhV*m8Z$Y}a0{Z_!GxuTJY};#yb8#&U_V zZ$Kh=`|!Z7Zyl)?t-^>QPaxYKAg?ZJ+m!=fEB-bxNN(9yhSZbi_D=5M+>b zz6?MzI=-}3GhUOZowo41p+F*Wrr892KL5t>iFQe|Q;#+?slM)x;YnS=$`sjQKT5B) z?(QcXAG;NDwhw+Hz;fcwxo2PfP##I++)vg|1W0FWytD0T5XF9e4IipsRUK%jd8ymw zn)kd9q)*oR!x=?W5nttcrfmFYVZ{SyYL11oQ6r}{+M6x1~|8P%%9$i zofl%4E5;N{J&M$fAA(*r>aGO_tfX}P=7?d8Tl+^OdHc-AeAJ%slQ+#^&C{14)$*l8}u6eK3 z9hcn;7M-=c?>rV*wo{%SCE(mJnK0E^Dm~@2vf!g@Q#BGi+HvsV70S-H-0U%MQ8AIIh#;pbu{uFOZ`nFp;7*YaFL zdWH@1=8TCR1ZPs(nKahWZ-76;Yn|NgfXY%mTZ#yd7fw=35G0H%G})k641Q0Wuf9}u z#(VR%!MvOKjauykec*>DE9OC1ecUnvgn|E47LR>!~3P64vJdH$jY4Tzjwk4mNhL*T(F z>3-Vbpr#zE)QexH?9yf`Bb)G9bbLS0r6x8-B!3M|^yKf!bcFD84>kNx55+QRrz!Co z-hEq3Ro!cDU}b%i5p*Ild0k{^lJa_EXz{$CxmERF;Rb_r?my|dz72e14?>FB@Vg`7 z^;s#Y{Ta(oqm`20*c&)3vuC{!XW#M5gxk1w+nb+s7n@&QUxIhNq1R^8kp9?TB|h6M zy+C?5$(kY-$axMBPEvfrUv4(tf^oQbq0`J?JxMQ>vo0 z)aPSYUq=Z!?JKc!Yu9PS-_y^3I^>Lqk-KBPm&oI|H&e<*ob}Qa8)j|hjlpe}h;G1G ziqq`djh_9L`!z8FJD5VKHiO#LX-2!p@&qo)b%6$sFR*!I9{hFX#DsoYZ1tlbPvx}=mt?u<2kVo$nJ*>h9|)XJJ& zef*~Qq>I6|$a{-mPezvP84Y|c!WEcL%$Kg5q$;^9^h=0=(2pE1mksT5H(ALB1;tJs zeFeA4Q2o$)qPXCvt$7_SwL;$$3d&rcZfZAE`Z9}{u8P_%a*w9#6V}n+7-ro#+|Y=Z zqGM;T`^)l9Frvt;_7y{OjojdhpIf0Ybr9@rt;6!2p_&IvwV(Iu4LVA5Bzj9jS3h6d zgJFLbtS9M?s!L%y9k_U?LtsJ1&NC1Pank`VI{(*?=l>$1_(3YkMl9+qe-C?yxhxl4`UorY&*3$bf1{bfL_U_Z`rhkT(q4{~wwG z{}nePu%CN^oPT+ON=*&D2SQqqg@KigT}P%#Elf9fvKm(&3ZK7O(Bw8Cn6;`%9XHS| z9n1r$1z7&p*TrHGoF8Fw(w9ihlKrAvNlnSg1?zF9oPC*tD?h4XH{Cs{D8QW!Mgluc ze$yduJf1(CIG`V`3MG3SytJFIa0Rw}vnVP9kh?1+WxRT_PH|%`w&q=)H~)N%4m7WZ zb^vC1YxmkpzI8|~h!)m&i!RU9E0Bz$z~DPM^_34%j>bS2BsC>~;bAEK?UVfMg+}$x zyUxoQpJ|f~+Q<-d@vzbSTgn>JVdO765t+Laotl1%4}v#tEH;=2?;iF~?ua&Oe?J@! z-r}P1@Ou0)``X#8npUoT5?@o~(wnbR7tj=wgt|;yRj(~|Jhp?YkD9Lboo_$C=a{O) zq?@_GM4@Uq`b8pb+RB%&JgE*p()ka*ekp2b+Nq&$-cZZAt<9jpRJzz#>eZMpyh=x% zt!Z1MVvp?dy9XAB9l2?pPV#PqG6PCXaiYs?*9&aq^KVg#k=&TO4h}9!hoK9E^!C(R zC@Oy#{UBm#B1SU;952`Ruu_ZAJI#I>2Ff0PJ>@9Y0$xU&(b5wRt)2GDUNXvW{}=!QcB=c`Kt5?M z1b(jXRX>k&`YiJua&^K0_Q%(zUs{Z63!2_m(S9pO@g#w%UVFp234uboGE&bCqV_gi zjTf~tM2NfuOrVnd^?dl4ghWT<9q&@9al6H~6+bVz);Z8Hyv0)6$t`L8s@qA|4X||+ z@XigR)a;M6l$kFpHiZ=!4CG)Y4H=KU)8S$VeFxMOfiTcdzo!r+`bQ{va)RgaZ+0cK zgxHB1ej4y~T+T>6nfPIhg>D>&n?30^HU3UR?&IKFP!DYyDwGjcd$o3yC>;FO%Iu$2 zg2rw&Qh&`3kc`vDQbvg2f*Yht_>bp*d+NTCrkPSLw{PC{Hw#b$A&uP!Pjc1jd&c{B zm3`M3HrP5R*V(!Xh_=*E_S-H?G&>G&@*n@Ag~$PmG1R7S{@2&&zrJ5T(SQm(fc2*W z5MyfxFrlkhjKon!4xSRC5$&!Cpk%^YIE!fePU8)x{dwpliY&`FeCq-)4HY z?XMRLS*Zo5OicTdh+b(`7SO7Mg*9es_SB8&zG}OHwyxQz8xdVFb~X%NrpbtWrD2g# zBKFXz<&&rJ+|9~rn^`O*Sly1CZTZeTY4r3DH@?0MdxSkhkqt#|?|zacRNf$st(5e+ zbN1;XP_2e(*T^S}VSJ}OXW{j?!Hi=UD#?%3NpiDo3^QPz30!>rIPHoLQGV|}l`vQ_ z*!CaBB0m^7r}?e(`#%ll7npQf1GP%H-SwgOm*OL)>S44A%?ksVBiiU2qDvQX3eMmA zqUMcCaRR#I}!zTXwti)8QT=>H#yZu!QY12@)=vso2 z6L$J@v{s}drXpUaaQ#wj2B)=UKq2$=#)oV+MY92?A~m9phY#^e8Md}b=+RauE`wSH zs=7A^H$Q{)`VUn--i>k_-CuLtuj;T_cf1`LELa>elbdAD&iebfW7ZSW!co~NJpa?A zUwgN=x$-G)m(MrXX#APc#^9BVS>cJ)POTQzBL8wYE~jD7bOGj%`iUi50F^E82b;&= z*V-GI=l{D7^f##d*|-|=+4;B{@MxshgVHG+j9}W40cvCB0CKTPXiak!$TF->uo|}% z(-{t<=tXtFbPaoY#4TPv_H)HUX6%l)i^sW)jo@WSD#Lh-+aG^Fnt0iJ!~T&HjW0Wf zw1G#xnlYKKN6q3LFIP=FrBXrdkOs2E&uPL%;jnxodFiN^1U4rgn+ihhbg^o z6P4;loJ~0CIfNDMT4bIia5=+cyfo6*d{5UR{kJzU*w}mTE65DVqnyzSj$$NsAGnD> z*n{-XL4LPi$6e&cU-8QKFQ3a@sj(4Ll(lLXIqmBf?wr(`?A5Pudu&EY8>E0OZhm?m z1!hz*M=C{^w~6wbj{95r@QQAP#ce57RG-$}$HK={skwdIzlD!h2O6<<4c5PArw&{<9y1hejnx4k`vGFt~}=H$8Lh`VCx#24|fehigfaF?dO}+lv4_= zSE`}f#z~C6k4{3IZj3b%Rp00UIY}=t`~vHi8y_sMQjm32tScCpxTP6A2ACDyS$kHX z?>notsqpWW+LlkC4X3)n2w+}MTlB~ayYS5Zhh55U3+CBE%7!0pck#rbuGDa*Ry()Q zmOX=VqMOo-1#U~4%0&%0!U9$cGFId=YreAP$ggYg#P4G6fZC8Qui=~6O5{>J; z{|bAWbZ<^FVhAF?J!?n}F?6nVJrgU_^eO@XMe0hD6{9n?O^Sd-r|zXI^|oL%1TX3<3e#w)a81CiS0*WT2e zt6f#^Kz6$~aJ$hDc>?4MWP7c{#+1D;!}~6i#0rdm<;f7?`Em6nHm%v~xSIy!z;2BP zB7dt1eoTGdnS`HpCJ!7*!5^fk-fB%UR(NT^xhr#yrYc(m7qSD{EAFwsIA{w>T%8I{ zFV%2#n5)g8Msedv`#+m@mHIOHV5j^~DR6sSU0DgLqLZX3_XqZ;JNg+Zm6xtYSKy3- zgF1VKgr3xASOmd%Y|`@*RA~-J)YDHj1JxR(&>4Mg2#?DA05$4bq$hx?exREwwt7{E z@LSPeCdAn4Grk|p3nh6{sdVGXT(UqtgTL%^LKA*08YY%gVp?&em60-80~R;%Kz)+BaZf*8#I6LZcM{unCO#Y>l^C{ay^)Mm+4B?lJ z=T>pozEZ{x4({DOh^=eHH!=?@4}UQ6fxsXu*d5O4^&xrySV0^~pLAshJB*|~+Y(Bd zW>cPnTmFXGLS2Wu$@Idz1xoxIYL6_aYpwCBq~PG?8#ddTbrLN}HV975%DIGYJ#HAl zTrn8&UDz8z4qJZSV4ssO#VaB#LXeWI@f*VlAeiy+k*}UG0ZQj3C;$4t0hJ z=Tk4*{RvydmXfTl_k1xA*z)oI>ogK3dI&ZLVbBZW`r;)wp9L1Ea}D;dcnd^oFurtN ze0fOmtl^G)6VKJaK*wvOT}dXEYUbF0|3Jb4nS~7}oleT`;vNn@#i5cO3F$ z#4mPt^_$paW_W}s*>HD{(6j`64LfB0h?84LlPeZTT<3*nGHKT|b=&3F(3J1*F2Fuc zPQyOgH;i3DLN|z78*%~8M?p@;MgCg<{MvY|-Lw2E*sxoC17!#Zy}bbf8~^OtHjoY0 zPjDIDQ4xhX=2j;8YnDF5p(xqd-Cc_$3>^r%In~(~xxlMW1BNf%$}R~ z;r?Q}6@1Ustq3hscl+f(io}JJgVZ1%$)qjXucss#XJ{>)6Xek?IdMRnA;48DLI?a;PD~! zU~Vbb!6Q5wEMV(ck5~5vs{gZz3CLIW+KJ!C7ce%zB{42es?^`d2vEr{U;+D6fLr9$ z{4`zGsomd5OEN#vh?qx9!x!Y1UH^ac(VqsLZT&ue0k`k8p}z-_Vu4 zHM0!R=(W&w3t(I&?K1#6x&sXoBa1@$k&NBjNZ)VQ%vT<)xPFt#;s7BKsph1GiIhs%4Yf5 z&Ku3D;}0nw>kxxEVQT~!e9AQ*{%i0XoiK9^TM+0_rd1cQJNaQDqJuUlOr>}sTKptz zlHVhMv#M*rEX_oWwt_KQoXmg&<^hH$81qV|4s+#!m{(^% zcsWDsg4Wt0teR*3TIgOWzwvWkjGXTpdWufCx6XX={(jY$i(lnuI4}bTGuGw#zy{Z5 z-@hMRxO+ahFmyJ!Ku8Hxdr$+U(GW#CZPS{$DP}uDIuzTcdG&{ZUOd<{0&bE8x*EEc zK7m|aFI7E!<}oQ$(Q!GliBz?6`K~`03jf6x_o6iigIkfh4yb1v1Q+lFozL#dp7E-D zr%R`iRpKRQSn?$j5LeJ<9v*U)4vK<$Fa3UKf7g@ZxzHKzMnp-!z0G>3WHx4#wnw~H z;klHo(E__l0cRS_D6UGg)xrLM~F(b`Z;x1n)^`jvD(NJ zzG-0>PHw-n&PTfB%T+jN)yzcg@$rfy!hPZf>SEtnmnpA$S-$@hpjEq2c*$qujiEi# z-_7Mzq$0P2I!MUw)EM%S*LS`Il1XS$*#_z^yhI|KPeDhDK`2{x?%Vz}QKORbIm@x| z4!qE+qWlU?lx8J$m$tV|e5qTcaj|uvl0`CrB!tqqpHA`%K-r)=tQh}=&ju|u8qyD=cU#Kf08 z?}^PmFwIzvnU;NxIK-m;;vaZA6zNu}Bt#v~=nbzuGs<&boiJR)8kk8EO%dZ&5wLp?(F){3X!qRvK4R4pDgBK{0?RG`~g&c6VWlf zhQMYt#kbOzSp}mr+z^8J0mBLRBh>)4!)Je`|LsKc@ydD5N%4=IlMKwN0l; zhd<@s#3sgc>c(l*i3STtF^n>g&{zO%%NnPF1{BpGFbz`s%!b6EtnGluQ)*9_`ng&> z!{o-e56+iA6=bB)=P*K@;I_^{)#S*rGvc~}ZPPjH6E`Vh7j zi4*dhWsvGjytKdU&}}>pTU6^jn^qeW^K86PX}>gOoucFt43BMjw z)2_RiyCx1uIwdI9S_`gp78kjuaM~x*VZVc|RBOqH?Dr$L@x4o}yS;XrpF^H0?ivq& zpqM~?-P7&=v=Un{l@e9T2X0N~N!xSZ$#Rd?D*B@?SO z_MDiiiikj@z)G&WUPw3f$4?f^a_bJuH@@d_n>oB$ML(1JiLLtjsHc^td&Z)gG>bNm zzGu-D1ZYg<^GHCg{PCzJ_jp6Jhn#4K zo!3>}#QMA?k+7FJZxcpjhkvcRaK`(4wkE>RF9M&F)qNYkvw%`qi(ebc!drr=404-Z zQCG7)Qq3iGT17AgC1hGnigu5 z&iF^THDFw0vllhL?@crqgT3>ZC<9C@dY6soF5NV5Ln82Yan219?23(rWUO|3UFj(L zuemDo9w9?lTQIn&%H1zw+a$F9n~_MRVc0XbJ_3StChW5`D~D@$k@=cHRrChNHgwb~ znFUwdp%_mv0q_fSYSigTQ*nncWe1FMS3+_>7s?N^~ajENDb zR)$L}b#F%)@jGo_OtnNEDugXfovD)ffKfNI@dV#GRo9}&RF923mS7(9AFhsp01d1= zLMSfh*20UvK)Td-v9u+3#Sks(dA+=PJDUxxTu}$qL!o_cH%K2|e(82hx6yOcec}mS z2piIP5uD3wT!*Dkt*WHMRuVHDw)$a4MuVj5S=^L|AH)Qx`G~K1^JAQ|EJ8jxCi6+= zSByWi%X<}|CIn62V7p^u`T@B^ie_DUvQ}xjW;Cp`v*L)ZLO7f*PC3&uAO}4c%gedz zS}1h}zw)P9{S^PZw*9-){PVFOK<^x2F027nbT2?>%G8MeD^e16YarWSp@^tF7%AzK zL$36Y`o`zA0O^eyBnOCRXR5|HxZ`1OJ>Qr^?ZOlEUzN-K-3R&zY^=la-lX}}vwWAz zU`7;3>pb#EEc0_z8-|zlG^xYX@~1EDi-Pi4;+`yiH7SKr9R*xX_nM;jI`(FGWdIq* zj(ned8-3-*d@ZdPEG84iXOKvWv7U8&ir$~*W4PF_ye=8K6E-=-?KZfVP7@TPJ0ztY zIpR>3TkOBHv@`asJD5d_-}cnR*q+_9TvUiM^Z9fHJ48=evM%1X@F&}fTPeShHeDpR zkxDssCAYlEf!YX?S1{bGp~$(!K0XA~Zx$o$xTYMo+%xUA0Gi*HO;iRt1^96P{Y{4x z{t@@qAhyNV(PyK$Oqj`)eJTjDjD*x0H%y8X#Q812HLXv4)`oh{Ad$V*T%0q2nWq^1 zrm$U`}0hJ#W| zl}1(`3x|of@ci2*y1BAU29VJJo>o2)Jpe~z1sX~1o=S+qKpc=XkD9X^wq5O24^z|p|r17X5W#BF{L`p z3|^!$h;qY!7j{V3z0WNCJKoZCpo7vNaWstDE2D2kKXGdIQmM9aFxXza2z{Ljq^YM% zMOhxG7H(75xnxJ(cQP8>s@0+vU%$NQw}1_Y%=a{69kA)Z3n~`0h^%4z1r$S@BevK+ zf|D7dfNP=vqS;CF6Z;Kuie|Ha$8-;hU0AN$c9-%b#Rw{+O&2a`%Lw;tH^P54MfR&N zVrf#5ecMi}(wF+iwr9`S^=C6T@~z5yp7SY>0m+m#BqtLw754se70b_rhRG(!A)MOD zM5RbX1%_~!t2%Ebf^b8V#lLx%CI##U^wTFp58OEGPd68UWu0&SZ~D@)=iR8Wvu@OJ z)zysenJ=4XxU{l6c&%=TwOu9Ea5)@vGk09AURaaIJeUK@&drUbkmY6gNSSdRa#O%Q zHlcVKY_plQELAI6hal0+FG_ZN=P^}D1!zQsz=raVqA%VtTl|sYaDAgjF;uB8=l5s& zlcO4*IGAUC=1>@-_6Mp6#YzWX$buDT_*Ys??@k3nLRi;}?s>;@*N6d>7z ztf*8)nmAK6Yvst9TkG2As|OC-n_}2l4BEc~M0s`#Xuto{CL)cgQ?KOR)!GL4352(v z(yID}cg*R~?lf=bGRVe7Ni&ug7`v0lC^mgM3_S-%?d9qXjKZ5t_?N6OUbB3IL5E4C zY6a)(ME0F53u+M9nkHD`N41R>sIKPWkE?nH{En7KiMAFjmhxC+ty=i+|7Kbq$4Sx5*J5PY42+`O zuFbU}oDqBz=IfIo6K@2^9^&|{0lV{0wl@S6jc8PQ38MWN(D#A?6SwS!Jymz8Fo1>G zCW}X0;kpC;2>)Kt>t?5ZldbLd-?GPln*!^=wbP1J$k}qtEdWNA(Qc-dZ=bfT&o`pL z+M1LU8zO3%Qh{DRK&aJ5vy=1L_3|TGZ40%wpRyPS#FR~?y*~AIR?GYcZ7%B}wp6+w zvXV0Fvpa}tKNN?FR2qGSCuO7{YpspDiSocHsMSM13x<8#p%Ggu>ZU#)&gD(~@!_+V zzuPiVGSijiM)2$wet)-j#KEk7CHCI@l9LpsYCU)e0D4E68DcQz4O`Z&*-x)n(OLDw?hg|b-gRoP6n?R$(C#|6 zDb814Mrn1Gg{R7BUs^bQg8u2j0Za*)cOM=JGl&(f^w4$I*;ob3U?HrvTvo;|vBh7s z7VCBn1)1sX_$N!L>4b9Kdj<)KC;D!+s|$d;W#11EPTO&^V#hox0*Th@s zx95zk5`zC0Ta>T>B-(-d1y<|j%%DYycj30iE?a_GezwKD(Hji#%A6F4h+-SeLqDVcwUy0bf0_a+#*T(hmPG#)Q-wLPN?u?GJ!GxO^m;Agh5P^AGx z60@xDtOVd|zcX~C{4KQcKm8$G>kM%7I*VQaE%`r8$^Y6FKR34j-&0>_sO$^>r3LW! z+k5|I>EHjuYyNv5bv}OaA71nS)?5CEsQ$mPp$nJ)Lsb7?nCbs=LI3>2f0PIQZrbbL z#`XUhqWb^tF3+#_KNtGnztHfP$TJyP;SH4IU~X>-w$cV0LW^(&6yuS^s1MSAJ5TaI zUdSBvS=y?3&OY=dx9P#Su2o`iB}>4%j4u81B9Ao#q|QvpnFjbswfZH)S09)(2wNSz zDagi}huo!bnaP&`_0`t01z6Z_n^lZ@zFk`;@>yWxs_@YP<1_04O5w$~kRm7=!qy4n z-0=aHP9+`ZLN@C8bc}P5MdcoY8r15nx#*2|p(kS4 zm)-oUUwA_ePWV@}!**XW{Vz$}zb*dX5+`hl=hXt)`7_y)FHi1^m9WbiJMN1gya)8t z96z8DpW#V-2h|VHmMl<+pXSw%ZTx(=+TZ#^S8p<0=8HA3C&6U;1dWfhEv_3=x$A*6 zjbzh?jl#L3s)V?{?m6CpwGE>>xjv`6?0mYkn)L9>Ym>yS*%U`qWw${^_bR$>Z_$!E zS_B>^trcdyaJm_4kFAgFKAERQi{aPCTSLg`j95n^2Apd+D1>z$L*0wbzPG;!-=M-q;~dUJ_g8ff&I9HELsH+mP0o9>{dh}ZVlCY zZ`)->J}aSu=@x6Pz~jALns|X-4kS&I1`E1r&?cBpk-^dkZ#T%|YKD%XJypuV1ci=8 zL5_mS1Ql1Cz;V|$?>1-D{UM%xi0izcJsi+IOyya8-{r}@8cXM~Y!RDRKS!NK9@nIH z{qNxH|7zW3Kx@e&Jlhw*?NU>;k@tt%NVuzRQX8io^4l;WPf-o_S(da04Ll2Qv zfPUw8d?x)JeE!MdLb|^}aE3$eI!H4(6FTr-M^|V!+k8y|V}mFoTjPz6PI+03-ppN_ zXVWPf&Q`ZDoy){<12f8Z|Fw0czq2psDb^B?*jut1e0u&+BjfH!>p{y=pB04t41JBj z_9O}Z^~fss>WJ`qX~-_70?-X5U@qfsj|3n7y%zhwZ}Jy#UBBq9O#x9GR$V|Nvsdgi za~BIopE40At;_1C91TK)Uz3^h(2i07S6sK=lSV829b}W;M|NSqvTmdHdWT zD}fiNCJkG10eSCy2LI!JtoyPVuciMwW@EWi)&LmzHD5{Ic{RUKUfAk@Pon2fgD#ud zj(+tY$d2D?Fppw*WkZzzWH*w!P$rU!ea?rVtHTB~>%ZSp9Cf9jXR}5xd;Y$;w`e|N zZg-}$Q~&QH`FrC&pgvcymti^=&AOZ+V+su?6;kcn>}cP)gHu5JXV z@gJ|SIiLt9d+mO?9#&+c&w9`fb;j(Po!^YI;;e_a>^f2$#|<4D%+r-p2&b#5SCgou zvBBbIZJ+PD^vJth`NVPS^0ix+V6aZ9ph2n(g}Bym?!|aVjnY z7csKH7RSlSnJm7z*n6+uc_yUmbCnu$ZL}bXA=YlJOkp!qI(sQFGDl%9>_^H>F+b

" ] }, { diff --git a/site/notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb b/site/notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb index 49df9a5b50..9a3cbe7ef7 100644 --- a/site/notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb +++ b/site/notebooks/tutorials/model_development/3-integrate_custom_tests.ipynb @@ -13,7 +13,7 @@ "- The function can be as simple or as complex as you need it to be — it can use external libraries, make API calls, or do anything else that you can do in Python.\n", "- The only requirement is that the function signature and return values can be \"understood\" and handled by the ValidMind Library. As such, custom tests offer added flexibility by extending the default tests provided by ValidMind, enabling you to document any type of model or use case.\n", "\n", - "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb) notebook.\n", + "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb) notebook.\n", "\n", "
Learn by doing\n", "

\n", @@ -817,7 +817,7 @@ "\n", "
Want to learn more about test providers?\n", "

\n", - "An extended introduction to test providers can be found in: Integrate external test providers
" + "An extended introduction to test providers can be found in: Integrate external test providers
" ] }, { diff --git a/site/notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb b/site/notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb index 0af95d90a7..04c745225e 100644 --- a/site/notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb +++ b/site/notebooks/tutorials/model_development/4-finalize_testing_documentation.ipynb @@ -930,9 +930,9 @@ "\n", "#### Use cases\n", "\n", - "- [Document an application scorecard model](../../code_samples/credit_risk/application_scorecard_full_suite.ipynb)\n", - "- [Linear regression documentation demo](../../code_samples/regression/quickstart_regression_full_suite.ipynb)\n", - "- [LLM model documentation demo](../../code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb)" + "- [Document an application scorecard model](../../use_cases/credit_risk/application_scorecard_full_suite.ipynb)\n", + "- [Linear regression documentation demo](../../use_cases/regression/quickstart_regression_full_suite.ipynb)\n", + "- [LLM model documentation demo](../../use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb)" ] }, { @@ -943,12 +943,12 @@ "\n", "#### More how-to guides and code samples\n", "\n", - "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", - "- [In-depth guide on running dataset based tests](../../how_to/run_tests/1_run_dataset_based_tests.ipynb)\n", - "- [In-depth guide for implementing custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb)\n", - "- [In-depth guide to external test providers](../../code_samples/custom_tests/integrate_external_test_providers.ipynb)\n", - "- [Configuring dataset features](../../how_to/configure_dataset_features.ipynb)\n", - "- [Introduction to unit and composite metrics](../../how_to/run_unit_metrics.ipynb)" + "- [Explore available tests in detail](../../how_to/tests/explore_tests/explore_tests.ipynb)\n", + "- [In-depth guide on running dataset based tests](../../how_to/tests/run_tests/1_run_dataset_based_tests.ipynb)\n", + "- [In-depth guide for implementing custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb)\n", + "- [In-depth guide to external test providers](../../how_to/tests/custom_tests/integrate_external_test_providers.ipynb)\n", + "- [Configuring dataset features](../../how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb)\n", + "- [Introduction to unit and composite metrics](../../how_to/metrics/run_unit_metrics.ipynb)" ] }, { @@ -961,7 +961,7 @@ "\n", "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)" ] }, diff --git a/site/notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb b/site/notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb index b9154b80c2..6a2e9e128a 100644 --- a/site/notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb +++ b/site/notebooks/tutorials/model_validation/1-set_up_validmind_for_validation.ipynb @@ -141,7 +141,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -177,8 +177,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down — don’t worry, we’ll adjust these permissions next for validation.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/tutorials/model_validation/2-start_validation_process.ipynb b/site/notebooks/tutorials/model_validation/2-start_validation_process.ipynb index 48e4dbff66..77383ce0f2 100644 --- a/site/notebooks/tutorials/model_validation/2-start_validation_process.ipynb +++ b/site/notebooks/tutorials/model_validation/2-start_validation_process.ipynb @@ -237,7 +237,7 @@ "source": [ "
Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests" ] }, { diff --git a/site/notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb b/site/notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb index fe4e221c90..6103fa2d41 100644 --- a/site/notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb +++ b/site/notebooks/tutorials/model_validation/4-finalize_validation_reporting.ipynb @@ -13,7 +13,7 @@ "- The function can be as simple or as complex as you need it to be — it can use external libraries, make API calls, or do anything else that you can do in Python.\n", "- The only requirement is that the function signature and return values can be \"understood\" and handled by the ValidMind Library. As such, custom tests offer added flexibility by extending the default tests provided by ValidMind, enabling you to document any type of model or use case.\n", "\n", - "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb) notebook.\n", + "**For a more in-depth introduction to custom tests,** refer to our [Implement custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb) notebook.\n", "\n", "
Learn by doing\n", "

\n", @@ -480,7 +480,7 @@ "\n", "
Want to learn more about custom tests?\n", "

\n", - "Refer to our in-depth introduction to custom tests: Implement custom tests
" + "Refer to our in-depth introduction to custom tests: Implement custom tests
" ] }, { @@ -856,7 +856,7 @@ "\n", "
Want to learn more about test providers?\n", "

\n", - "An extended introduction to test providers can be found in: Integrate external test providers
" + "An extended introduction to test providers can be found in: Integrate external test providers" ] }, { @@ -1176,10 +1176,10 @@ "\n", "#### More how-to guides and code samples\n", "\n", - "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", - "- [In-depth guide on running dataset based tests](../../how_to/run_tests/1_run_dataset_based_tests.ipynb)\n", - "- [In-depth guide for running comparison tests](../../how_to/run_tests/2_run_comparison_tests.ipynb)\n", - "- [In-depth guide for implementing custom tests](../../code_samples/custom_tests/implement_custom_tests.ipynb)" + "- [Explore available tests in detail](../../how_to/tests/explore_tests/explore_tests.ipynb)\n", + "- [In-depth guide on running dataset based tests](../../how_to/tests/run_tests/1_run_dataset_based_tests.ipynb)\n", + "- [In-depth guide for running comparison tests](../../how_to/tests/run_tests/2_run_comparison_tests.ipynb)\n", + "- [In-depth guide for implementing custom tests](../../how_to/tests/custom_tests/implement_custom_tests.ipynb)" ] }, { @@ -1192,7 +1192,7 @@ "\n", "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)\n", "\n", "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." diff --git a/site/notebooks/use_cases/agents/agentic_ai_template.yaml b/site/notebooks/use_cases/agents/agentic_ai_template.yaml new file mode 100644 index 0000000000..06ef71c9c2 --- /dev/null +++ b/site/notebooks/use_cases/agents/agentic_ai_template.yaml @@ -0,0 +1,343 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial + +- id: executive_summary + title: Executive Summary + guidelines: + - Provide a high-level overview of the agentic AI system, including its + purpose, scope, and intended use cases. + - Summarize the key features that make the system agentic, such as autonomy, + reasoning, memory, adaptability, and goal-directed behavior. + - Highlight the strategic benefits for the organization, such as efficiency, + scalability, cost-effectiveness, and decision-making support. + - Outline the system’s testing and validation strategy at a glance, + emphasizing safety, reliability, and regulatory compliance. + - Identify major risks, limitations, and safeguards, giving stakeholders a + concise understanding of governance and monitoring plans. + - Present the deployment vision, including expected stakeholders, + operational environments, and integration with existing workflows. + index_only: true +- id: conceptual_soundness + title: Conceptual Soundness + index_only: true + sections: + - id: model_overview + title: Model Overview + guidelines: + - Provide a concise explanation of the system’s purpose, including how + the agentic AI framework enables autonomous decision-making, + reasoning, and action-taking. + - Describe the high-level design of the agent(s), their core objectives, + and how they interact with their environment and users. + - Explain the conceptual differences between this agentic system and + traditional AI/ML models, focusing on autonomy, adaptability, and + emergent behavior. + - Highlight the role of agency, memory, feedback loops, and + goal-directedness in the system’s operation. + - Summarize the overall vision for how the system is intended to be + applied in real-world contexts, along with high-level testing goals. + parent_section: conceptual_soundness + - id: model_selection + title: Model Selection + guidelines: + - Describe the agentic AI paradigm, reasoning algorithms, or frameworks + chosen (e.g., reinforcement learning, planning, LLM-based + orchestration) and why they are suitable for the use case. + - Explain how the selected approach supports autonomy, adaptability, and + safe delegation of decision-making to the agent. + - Compare alternative paradigms (e.g., rule-based agents, purely + supervised ML models) and clarify why they were less appropriate. + - Discuss any hybrid approaches (e.g., combining symbolic reasoning with + generative models) and the rationale for customization. + - Identify potential risks and trade-offs of the chosen approach, + including known failure modes, and describe how these will be tested + and validated. + parent_section: conceptual_soundness + contents: + - content_id: model_selection + content_type: text + - id: purpose_and_scope + title: Purpose and Scope + guidelines: + - Clearly define the primary goals of the agentic AI system, including + decision-making domains and problem boundaries. + - Specify intended users, stakeholders, and environments where the agent + will operate. + - Identify the scope of autonomy granted to the agent (e.g., advisory + role, execution authority, or fully autonomous operation). + - Clarify the operational limits and scenarios where human oversight, + intervention, or escalation is required. + - Define measurable testing objectives that validate the agent’s + performance within its declared scope. + parent_section: conceptual_soundness + - id: architecture_at_glance + title: Architecture at Glance + guidelines: + - Provide a high-level diagram or description of the system + architecture, including agents, memory, reasoning modules, and + communication channels. + - Explain how the architecture supports perception, reasoning, planning, + and action loops. + - Highlight integration points with external systems, APIs, or data + sources. + - Describe the flow of information and control, showing how decisions + are formed, validated, and executed. + - Summarize testing hooks or checkpoints across components to enable + unit, integration, and system-level evaluation. + parent_section: conceptual_soundness + - id: assumptions_and_limitations + title: Assumptions and Limitations + guidelines: + - List the explicit assumptions about the environment, data, and user + behavior that underpin the system’s design. + - Identify constraints in agent reasoning, knowledge scope, or autonomy + that may affect performance. + - Discuss limitations in generalizability across contexts, domains, or + environments. + - Describe how uncertainty, incomplete information, or conflicting + objectives are handled. + - Explain how assumptions and limitations are validated through stress + tests, adversarial scenarios, and edge-case evaluations. + parent_section: conceptual_soundness + - id: regulatory_requirements + title: Regulatory Requirements + guidelines: + - Identify relevant laws, regulations, and standards applicable to + autonomous decision-making systems in the financial or operational + domain. + - Explain how the system addresses compliance needs such as + auditability, explainability, fairness, and accountability. + - Clarify how human oversight and control are integrated to meet + regulatory expectations for autonomous AI. + - Highlight any specific documentation, logging, or reporting features + built into the system for compliance purposes. + - Describe testing procedures to validate regulatory compliance, + including audit trail verification and explainability checks. + parent_section: conceptual_soundness +- id: data_preparation + title: Data Evaluation + index_only: true + sections: + - id: data_description + title: Data Description + guidelines: + - Provide an overview of data sources used by the agent(s), including + structured, unstructured, streaming, or interaction-derived data. + - Describe how contextual, environmental, or feedback data is + incorporated into the agent’s reasoning processes. + - Explain how memory structures (short-term, long-term, episodic) depend + on or interact with data inputs. + - Detail preprocessing or feature engineering tailored to enable + reasoning, planning, or adaptation. + - Include validation procedures to confirm data relevance, + representativeness, and adequacy for agent training and testing. + parent_section: data_preparation + - id: data_quality + title: Data Quality + guidelines: + - Define quality requirements for agent inputs, including accuracy, + timeliness, and consistency of real-world data streams. + - Describe methods for detecting and handling incomplete, noisy, or + adversarial data. + - Explain quality control for interaction data (e.g., user prompts, + feedback) that may shape agent behavior. + - Highlight processes for maintaining integrity of memory stores and + preventing drift due to poor input quality. + - Include testing protocols for validating data pipelines, + stress-testing with edge cases, and detecting bias leakage. + parent_section: data_preparation + contents: [] +- id: model_evaluation + title: Model Evaluation + index_only: true + sections: + - id: model_description + title: Model Description + guidelines: + - Provide a clear description of the agent’s architecture, reasoning + cycle, and interaction model. + - Explain the roles of planning, memory, and feedback in enabling + autonomy and adaptability. + - Detail how subcomponents (e.g., LLMs, planners, evaluators) integrate + to achieve end-to-end functionality. + - Clarify how emergent behaviors are monitored and managed. + - Specify test coverage for each component, including unit tests, + integration tests, and system-level tests. + parent_section: model_evaluation + - id: evaluation_methodology + title: Evaluation Methodology + guidelines: + - Describe the evaluation framework for testing autonomy, adaptability, + and goal alignment. + - Specify metrics for reasoning quality, task success, efficiency, and + safety. + - Explain simulation, sandboxing, or staged deployment approaches used + for testing. + - Include stress-testing for unexpected inputs, adversarial prompts, or + dynamic environments. + - Define reproducibility and benchmarking protocols to validate results + consistently across test cycles. + parent_section: model_evaluation + - id: prompt_evaluation + title: Prompt Evaluation + guidelines: + - Describe how the system’s responses to prompts are evaluated for + relevance, accuracy, and safety. + - Explain methods for detecting prompt injection, manipulation, or + adversarial use. + - Detail how evaluation ensures robustness against ambiguous, + conflicting, or incomplete instructions. + - Clarify criteria for determining when escalation to human oversight is + required. + - Define testing strategies for prompt templates, prompt chaining, and + stress scenarios. + contents: + - content_type: test + content_id: validmind.prompt_validation.Clarity + - content_type: test + content_id: validmind.prompt_validation.Conciseness + - content_type: test + content_id: validmind.prompt_validation.Delimitation + - content_type: test + content_id: validmind.prompt_validation.NegativeInstruction + - content_type: test + content_id: validmind.prompt_validation.Specificity + parent_section: model_evaluation + - id: agent_evaluation + title: Agent Evaluation + guidelines: + - Provide methods for assessing the agent’s ability to reason, plan, and + act autonomously. + - Define success metrics such as goal completion rate, adaptability to + change, and alignment with human intent. + - Explain how unintended or emergent behaviors are identified and + evaluated. + - Include testing for multi-agent interactions, collaboration, or + conflict resolution. + - Describe adversarial and edge-case testing to validate resilience of + autonomous decision-making. + contents: + - content_type: test + content_id: my_custom_tests.banking_accuracy_test + - content_type: test + content_id: my_custom_tests.BankingToolCallAccuracy + parent_section: model_evaluation + - id: output_quality + title: Output Quality + guidelines: + - Define quality standards for agent outputs (e.g., recommendations, + actions, reports). + - Evaluate outputs for consistency, accuracy, and contextual + appropriateness. + - Assess outputs for fairness, non-discrimination, and alignment with + ethical principles. + - Include processes for handling uncertainty or probabilistic reasoning + in outputs. + - Develop automated test suites to benchmark output quality against gold + standards or domain experts. + contents: + - content_type: test + content_id: validmind.model_validation.ragas.Faithfulness + - content_type: test + content_id: validmind.model_validation.ragas.ResponseRelevancy + - content_type: test + content_id: validmind.model_validation.ragas.ContextRecall + parent_section: model_evaluation + - id: Safety + title: Safety + guidelines: + - Describe built-in safety mechanisms to prevent harmful or unintended + actions by the agent. + - Explain escalation protocols for high-risk decisions requiring human + oversight. + - Detail adversarial robustness testing and red-teaming efforts to + uncover vulnerabilities. + - Clarify methods for ensuring alignment with ethical, legal, and + organizational safety standards. + - Include continuous validation tests for safety boundaries under + evolving data and environment conditions. + contents: + - content_type: test + content_id: validmind.model_validation.ragas.AspectCritic + - content_type: test + content_id: validmind.prompt_validation.Bias + - content_type: test + content_id: validmind.data_validation.nlp.Toxicity + parent_section: model_evaluation + - id: reliability_resilience_and_degraded_modes + title: Reliability, Resilience and Degraded Modes + guidelines: + - Explain strategies to ensure continuity of service during system or + environment disruptions. + - Describe fallback behaviors, degraded modes, or safe defaults when + full autonomy is not possible. + - Detail resilience mechanisms for handling network, data, or + computational failures. + - Provide monitoring methods for detecting and recovering from system + instability or drift. + - Define test scenarios simulating degraded conditions to validate + graceful failure and recovery. + parent_section: model_evaluation + - id: c46a7162-5fcd-4d2f-87e2-084afae70ee9 + title: Actor specific Results + parent_section: model_evaluation + contents: [] + sections: + - id: e78c8564-5af1-4ecc-b200-f131a629a01c + title: Credit Risk Analyzer + parent_section: c46a7162-5fcd-4d2f-87e2-084afae70ee9 + contents: [] + - id: df36a0c3-be44-4e16-a59a-cb635eac3ff3 + title: Customer Account Manager + parent_section: c46a7162-5fcd-4d2f-87e2-084afae70ee9 + contents: [] + - id: 67d25cc5-2569-4727-aae1-6c5b2f84e238 + title: Fraud Detection System + parent_section: c46a7162-5fcd-4d2f-87e2-084afae70ee9 + contents: [] + - id: cost_and_performance_management + title: Cost and Performance Management + guidelines: + - Provide metrics for computational efficiency, resource utilization, + and scalability of the system. + - Explain trade-offs between autonomy, performance, and resource + consumption. + - Detail monitoring of infrastructure costs, particularly in multi-agent + or large-scale deployments. + - Describe optimization strategies for balancing responsiveness with + efficiency. + - Include load testing, latency measurement, and profiling to validate + scalability and cost-effectiveness. + parent_section: model_evaluation +- id: observability_and_monitoring + title: Observability and Monitoring + index_only: true + sections: + - id: monitoring_plan + title: Monitoring Plan + guidelines: + - Describe monitoring practices for reasoning quality, autonomy + boundaries, and safety compliance. + - Define triggers or alerts for deviations in agent behavior, output + quality, or ethical alignment. + - Explain feedback mechanisms for continuous improvement, retraining, or + realignment. + - Detail governance processes overseeing the monitoring, including human + review cycles. + - Specify testing protocols for validating monitoring tools, anomaly + detection, and alert reliability. + parent_section: observability_and_monitoring + - id: remediation_plan + title: Remediation Plan + guidelines: + - Provide steps for addressing performance degradation, misalignment, or + unsafe behaviors. + - Define escalation protocols and roles for intervention when agent + behavior breaches acceptable limits. + - Describe rollback strategies to revert to prior safe versions or modes. + - Explain retraining or recalibration processes when monitoring + identifies issues. + - Include regular scenario-based testing to validate the effectiveness + of remediation and recovery procedures. + parent_section: observability_and_monitoring diff --git a/site/notebooks/code_samples/agents/banking_test_dataset.py b/site/notebooks/use_cases/agents/banking_test_dataset.py similarity index 53% rename from site/notebooks/code_samples/agents/banking_test_dataset.py rename to site/notebooks/use_cases/agents/banking_test_dataset.py index bd2793169a..895b1e97f3 100644 --- a/site/notebooks/code_samples/agents/banking_test_dataset.py +++ b/site/notebooks/use_cases/agents/banking_test_dataset.py @@ -12,14 +12,6 @@ "session_id": str(uuid.uuid4()), "category": "credit_risk" }, - { - "input": "Evaluate credit risk for a business loan of $250,000 with monthly revenue of $85,000 and existing debt of $45,000 and credit score of 650", - "expected_tools": ["credit_risk_analyzer"], - "possible_outputs": ["MEDIUM RISK", "HIGH RISK", "business loan", "debt service coverage ratio", "1.8", "annual revenue", "$1,020,000", "risk score", "650"], - "expected_output": "MEDIUM RISK", # Example, adjust as needed - "session_id": str(uuid.uuid4()), - "category": "credit_risk" - }, { "input": "Check account balance for checking account 12345", "expected_tools": ["customer_account_manager"], @@ -45,29 +37,5 @@ "expected_output": "High-Yield Savings Account (2.5% APY)", # Example, adjust as needed "session_id": str(uuid.uuid4()), "category": "account_management" - }, - { - "input": "Investigate suspicious transactions totaling $75,000 across multiple accounts in the last week", - "expected_tools": ["fraud_detection_system"], - "possible_outputs": ["Require additional verification", "Implement 24-hour delay for verification"], - "expected_output": "Require additional verification", # Example, adjust as needed - "session_id": str(uuid.uuid4()), - "category": "fraud_detection" - }, - { - "input": "Assess credit risk for a $1,000,000 commercial real estate loan with $500,000 annual business income", - "expected_tools": ["credit_risk_analyzer"], - "possible_outputs": ["HIGH RISK", "VERY HIGH RISK", "loan-to-value", "66.7%", "debt service coverage", "2.0"], - "expected_output": "HIGH RISK", # Example, adjust as needed - "session_id": str(uuid.uuid4()), - "category": "credit_risk" - }, - { - "input": "Update customer contact information and address for account holder 22334", - "expected_tools": ["customer_account_manager"], - "possible_outputs": ["not found in system", "Customer ID 22334 not found in system.", "not found"], - "expected_output": "Customer ID 22334 not found in system.", # Example, adjust as needed - "session_id": str(uuid.uuid4()), - "category": "account_management" } ]) diff --git a/site/notebooks/code_samples/agents/banking_tools.py b/site/notebooks/use_cases/agents/banking_tools.py similarity index 99% rename from site/notebooks/code_samples/agents/banking_tools.py rename to site/notebooks/use_cases/agents/banking_tools.py index eb9bb00767..a3f08c6ebf 100644 --- a/site/notebooks/code_samples/agents/banking_tools.py +++ b/site/notebooks/use_cases/agents/banking_tools.py @@ -279,8 +279,8 @@ def _handle_recommend_product(customer): def _handle_get_info(customer, customer_id): """Handle get info action.""" - credit_tier = ('Excellent' if customer['credit_score'] >= 750 else - 'Good' if customer['credit_score'] >= 700 else + credit_tier = ('Excellent' if customer['credit_score'] >= 750 else + 'Good' if customer['credit_score'] >= 700 else 'Fair' if customer['credit_score'] >= 650 else 'Poor') return f"""CUSTOMER ACCOUNT INFORMATION diff --git a/site/notebooks/use_cases/agents/document_agentic_ai.ipynb b/site/notebooks/use_cases/agents/document_agentic_ai.ipynb new file mode 100644 index 0000000000..4b1a6fb7a4 --- /dev/null +++ b/site/notebooks/use_cases/agents/document_agentic_ai.ipynb @@ -0,0 +1,2197 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "e7277c38", + "metadata": {}, + "source": [ + "# Document an agentic AI system\n", + "\n", + "Build and document an agentic AI system with the ValidMind Library. Construct a LangGraph-based banking agent, assign AI evaluation metric scores to your agent, and run accuracy, RAGAS, and safety tests, then log those test results to the ValidMind Platform.\n", + "\n", + "An _AI agent_ is an autonomous system that interprets inputs, selects from available tools or actions, and executes multi-step behaviors to achieve defined goals. In this notebook, the agent acts as a banking assistant that analyzes user requests and automatically selects and invokes the appropriate specialized banking tool to deliver accurate, compliant, and actionable responses.\n", + "\n", + "- This agent enables financial institutions to automate complex banking workflows where different customer requests require different specialized tools and knowledge bases.\n", + "- Effective validation of agentic AI systems reduces the risks of agents misinterpreting inputs, failing to extract required parameters, or producing incorrect assessments or actions — such as selecting the wrong tool.\n", + "\n", + "
For the LLM components in this notebook to function properly, you'll need access to OpenAI.\n", + "

\n", + "Before you continue, ensure that a valid OPENAI_API_KEY is set in your .env file.
" + ] + }, + { + "cell_type": "markdown", + "id": "a47dd942", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About ValidMind](#toc1__) \n", + " - [Before you begin](#toc1_1__) \n", + " - [New to ValidMind?](#toc1_2__) \n", + " - [Key concepts](#toc1_3__) \n", + "- [Setting up](#toc2__) \n", + " - [Install the ValidMind Library](#toc2_1__) \n", + " - [Initialize the ValidMind Library](#toc2_2__) \n", + " - [Register sample model](#toc2_2_1__) \n", + " - [Apply documentation template](#toc2_2_2__) \n", + " - [Get your code snippet](#toc2_2_3__) \n", + " - [Preview the documentation template](#toc2_2_4__) \n", + " - [Verify OpenAI API access](#toc2_3__) \n", + " - [Initialize the Python environment](#toc2_4__) \n", + "- [Building the LangGraph agent](#toc3__) \n", + " - [Test available banking tools](#toc3_1__) \n", + " - [Create LangGraph banking agent](#toc3_2__) \n", + " - [Define system prompt](#toc3_2_1__) \n", + " - [Initialize the LLM](#toc3_2_2__) \n", + " - [Define agent state structure](#toc3_2_3__) \n", + " - [Create agent workflow function](#toc3_2_4__) \n", + " - [Instantiate the banking agent](#toc3_2_5__) \n", + " - [Integrate agent with ValidMind](#toc3_3__) \n", + " - [Import ValidMind components](#toc3_3_1__) \n", + " - [Create agent wrapper function](#toc3_3_2__) \n", + " - [Initialize the ValidMind model object](#toc3_3_3__) \n", + " - [Store the agent reference](#toc3_3_4__) \n", + " - [Verify integration](#toc3_3_5__) \n", + " - [Validate the system prompt](#toc3_4__) \n", + "- [Initialize the ValidMind datasets](#toc4__) \n", + " - [Assign predictions](#toc4_1__) \n", + "- [Running accuracy tests](#toc5__) \n", + " - [Response accuracy test](#toc5_1__) \n", + " - [Tool selection accuracy test](#toc5_2__) \n", + "- [Assigning AI evaluation metric scores](#toc6__) \n", + " - [Identify relevant DeepEval scorers](#toc6_1__) \n", + " - [Assign reasoning scores](#toc6_2__) \n", + " - [Plan quality score](#toc6_2_1__) \n", + " - [Plan adherence score](#toc6_2_2__) \n", + " - [Assign action scores](#toc6_3__) \n", + " - [Tool correctness score](#toc6_3_1__) \n", + " - [Argument correctness score](#toc6_3_2__) \n", + " - [Assign execution scores](#toc6_4__) \n", + " - [Task completion score](#toc6_4_1__) \n", + "- [Running RAGAS tests](#toc7__) \n", + " - [Identify relevant RAGAS tests](#toc7_1__) \n", + " - [Faithfulness](#toc7_1_1__) \n", + " - [Response Relevancy](#toc7_1_2__) \n", + " - [Context Recall](#toc7_1_3__) \n", + "- [Running safety tests](#toc8__) \n", + " - [AspectCritic](#toc8_1_1__) \n", + " - [Bias](#toc8_1_2__) \n", + "- [Next steps](#toc9__) \n", + " - [Work with your model documentation](#toc9_1__) \n", + " - [Customize the banking agent for your use case](#toc9_2__) \n", + " - [Discover more learning resources](#toc9_3__) \n", + "- [Upgrade ValidMind](#toc10__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "ecaad35f", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models. \n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators." + ] + }, + { + "cell_type": "markdown", + "id": "6ff1f9ef", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html)." + ] + }, + { + "cell_type": "markdown", + "id": "d7ad8d8c", + "metadata": {}, + "source": [ + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
" + ] + }, + { + "cell_type": "markdown", + "id": "323caa59", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Key concepts\n", + "\n", + "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", + "\n", + "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", + "\n", + "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", + "\n", + "**Metrics**: A subset of tests that do not have thresholds. In the context of this notebook, metrics and tests can be thought of as interchangeable concepts.\n", + "\n", + "**Custom metrics**: Custom metrics are functions that you define to evaluate your model or dataset. These functions can be registered with the ValidMind Library to be used in the ValidMind Platform.\n", + "\n", + "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", + "\n", + " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", + " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", + " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", + "\n", + "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", + "\n", + "**Outputs**: Custom metrics can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", + "\n", + "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", + "\n", + "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." + ] + }, + { + "cell_type": "markdown", + "id": "ddba5169", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "id": "b53da99c", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "
Recommended Python versions\n", + "

\n", + "Python 3.8 <= x <= 3.11
\n", + "\n", + "Let's begin by installing the ValidMind Library with large language model (LLM) support:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1982a118", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q \"validmind[llm]\" \"langgraph==0.3.21\"" + ] + }, + { + "cell_type": "markdown", + "id": "dc9dea3a", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "id": "5848461e", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook.\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "id": "97d0b04b", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Agentic AI`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "id": "b279d5fa", + "metadata": {}, + "source": [ + "
Can't select this template?\n", + "

\n", + "Your organization administrators may need to add it to your template library:\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "3606cb8c", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "d6ccbefc", + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "2ed79cf0", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Preview the documentation template\n", + "\n", + "Let's verify that you have connected the ValidMind Library to the ValidMind Platform and that the appropriate *template* is selected for your model.\n", + "\n", + "You will upload documentation and test results unique to your model based on this template later on. For now, **take a look at the default structure that the template provides with [the `vm.preview_template()` function](https://docs.validmind.ai/validmind/validmind.html#preview_template)** from the ValidMind library and note the empty sections:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dffdaa6f", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "id": "b5c5ba68", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Verify OpenAI API access\n", + "\n", + "Verify that a valid `OPENAI_API_KEY` is set in your `.env` file:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "22cc39cb", + "metadata": {}, + "outputs": [], + "source": [ + "# Load environment variables if using .env file\n", + "try:\n", + " from dotenv import load_dotenv\n", + " load_dotenv()\n", + "except ImportError:\n", + " print(\"dotenv not installed. Make sure OPENAI_API_KEY is set in your environment.\")" + ] + }, + { + "cell_type": "markdown", + "id": "e4a9d3a9", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the Python environment\n", + "\n", + "Let's import all the necessary libraries to prepare for building our banking LangGraph agentic system:\n", + "\n", + "- **Standard libraries** for data handling and environment management.\n", + "- **pandas**, a Python library for data manipulation and analytics, as an alias. We'll also configure pandas to show all columns and all rows at full width for easier debugging and inspection.\n", + "- **LangChain** components for LLM integration and tool management.\n", + "- **LangGraph** for building stateful, multi-step agent workflows.\n", + "- **Banking tools** for specialized financial services as defined in [banking_tools.py](banking_tools.py)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2058d1ac", + "metadata": {}, + "outputs": [], + "source": [ + "# STANDARD LIBRARY IMPORTS\n", + "\n", + "# TypedDict: Defines type-safe dictionaries for the agent's state structure\n", + "# Annotated: Adds metadata to type hints\n", + "# Sequence: Type hint for sequences used in the agent\n", + "from typing import TypedDict, Annotated, Sequence\n", + "\n", + "# THIRD PARTY IMPORTS\n", + "\n", + "import pandas as pd\n", + "# Configure pandas to show all columns and all rows at full width\n", + "pd.set_option('display.max_columns', None)\n", + "pd.set_option('display.max_colwidth', None)\n", + "pd.set_option('display.width', None)\n", + "pd.set_option('display.max_rows', None)\n", + "\n", + "# BaseMessage: Represents a base message in the LangChain message system\n", + "# HumanMessage: Represents a human message in the LangChain message system\n", + "# SystemMessage: Represents a system message in the LangChain message system\n", + "from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage\n", + "\n", + "# ChatOpenAI: Represents an OpenAI chat model in the LangChain library\n", + "from langchain_openai import ChatOpenAI\n", + "\n", + "# MemorySaver: Represents a checkpoint for saving and restoring agent state\n", + "from langgraph.checkpoint.memory import MemorySaver\n", + "\n", + "# StateGraph: Represents a stateful graph in the LangGraph library\n", + "# END: Represents the end of a graph\n", + "# START: Represents the start of a graph\n", + "from langgraph.graph import StateGraph, END, START\n", + "\n", + "# add_messages: Adds messages to the state\n", + "from langgraph.graph.message import add_messages\n", + "\n", + "# ToolNode: Represents a tool node in the LangGraph library\n", + "from langgraph.prebuilt import ToolNode\n", + "\n", + "# LOCAL IMPORTS FROM banking_tools.py\n", + "\n", + "from banking_tools import AVAILABLE_TOOLS" + ] + }, + { + "cell_type": "markdown", + "id": "e109d075", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Building the LangGraph agent" + ] + }, + { + "cell_type": "markdown", + "id": "15040411", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Test available banking tools\n", + "\n", + "We'll use the demo banking tools defined in `banking_tools.py` that provide use cases of financial services:\n", + "\n", + "- **Credit Risk Analyzer** - Loan applications and credit decisions\n", + "- **Customer Account Manager** - Account services and customer support\n", + "- **Fraud Detection System** - Security and fraud prevention" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1e0a120c", + "metadata": {}, + "outputs": [], + "source": [ + "print(f\"Available tools: {len(AVAILABLE_TOOLS)}\")\n", + "print(\"\\nTool Details:\")\n", + "for i, tool in enumerate(AVAILABLE_TOOLS, 1):\n", + " print(f\" - {tool.name}\")" + ] + }, + { + "cell_type": "markdown", + "id": "04d6785a", + "metadata": {}, + "source": [ + "Let's test each banking tool individually to ensure they're working correctly before integrating them into our agent:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dc0caff2", + "metadata": {}, + "outputs": [], + "source": [ + "# Test 1: Credit Risk Analyzer\n", + "print(\"TEST 1: Credit Risk Analyzer\")\n", + "print(\"-\" * 40)\n", + "try:\n", + " # Access the underlying function using .func\n", + " credit_result = AVAILABLE_TOOLS[0].func(\n", + " customer_income=75000,\n", + " customer_debt=1200,\n", + " credit_score=720,\n", + " loan_amount=50000,\n", + " loan_type=\"personal\"\n", + " )\n", + " print(credit_result)\n", + " print(\"Credit Risk Analyzer test PASSED\")\n", + "except Exception as e:\n", + " print(f\"Credit Risk Analyzer test FAILED: {e}\")\n", + "\n", + "print(\"\" + \"=\" * 60)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6b227db", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Test 2: Customer Account Manager\n", + "print(\"TEST 2: Customer Account Manager\")\n", + "print(\"-\" * 40)\n", + "try:\n", + " # Test checking balance\n", + " account_result = AVAILABLE_TOOLS[1].func(\n", + " account_type=\"checking\",\n", + " customer_id=\"12345\",\n", + " action=\"check_balance\"\n", + " )\n", + " print(account_result)\n", + "\n", + " # Test getting account info\n", + " info_result = AVAILABLE_TOOLS[1].func(\n", + " account_type=\"all\",\n", + " customer_id=\"12345\", \n", + " action=\"get_info\"\n", + " )\n", + " print(info_result)\n", + " print(\"Customer Account Manager test PASSED\")\n", + "except Exception as e:\n", + " print(f\"Customer Account Manager test FAILED: {e}\")\n", + "\n", + "print(\"\" + \"=\" * 60)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a983b30d", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Test 3: Fraud Detection System\n", + "print(\"TEST 3: Fraud Detection System\")\n", + "print(\"-\" * 40)\n", + "try:\n", + " fraud_result = AVAILABLE_TOOLS[2].func(\n", + " transaction_id=\"TX123\",\n", + " customer_id=\"12345\",\n", + " transaction_amount=500.00,\n", + " transaction_type=\"withdrawal\",\n", + " location=\"Miami, FL\",\n", + " device_id=\"DEVICE_001\"\n", + " )\n", + " print(fraud_result)\n", + " print(\"Fraud Detection System test PASSED\")\n", + "except Exception as e:\n", + " print(f\"Fraud Detection System test FAILED: {e}\")\n", + "\n", + "print(\"\" + \"=\" * 60)" + ] + }, + { + "cell_type": "markdown", + "id": "6bf04845", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Create LangGraph banking agent\n", + "\n", + "With our tools ready to go, we'll create our intelligent banking agent with LangGraph that automatically selects and uses the appropriate banking tool based on a user request." + ] + }, + { + "cell_type": "markdown", + "id": "31df57f0", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Define system prompt\n", + "\n", + "We'll begin by defining our system prompt, which provides the LLM with context about its role as a banking assistant and guidance on when to use each available tool:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7971c427", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Enhanced banking system prompt with tool selection guidance\n", + "system_context = \"\"\"You are a professional banking AI assistant with access to specialized banking tools.\n", + " Analyze the user's banking request and directly use the most appropriate tools to help them.\n", + " \n", + " AVAILABLE BANKING TOOLS:\n", + " \n", + " credit_risk_analyzer - Analyze credit risk for loan applications and credit decisions\n", + " - Use for: loan applications, credit assessments, risk analysis, mortgage eligibility\n", + " - Examples: \"Analyze credit risk for $50k personal loan\", \"Assess mortgage eligibility for $300k home purchase\"\n", + " - Parameters: customer_income, customer_debt, credit_score, loan_amount, loan_type\n", + "\n", + " customer_account_manager - Manage customer accounts and provide banking services\n", + " - Use for: account information, transaction processing, product recommendations, customer service\n", + " - Examples: \"Check balance for checking account 12345\", \"Recommend products for customer with high balance\"\n", + " - Parameters: account_type, customer_id, action, amount, account_details\n", + "\n", + " fraud_detection_system - Analyze transactions for potential fraud and security risks\n", + " - Use for: transaction monitoring, fraud prevention, risk assessment, security alerts\n", + " - Examples: \"Analyze fraud risk for $500 ATM withdrawal in Miami\", \"Check security for $2000 online purchase\"\n", + " - Parameters: transaction_id, customer_id, transaction_amount, transaction_type, location, device_id\n", + "\n", + " BANKING INSTRUCTIONS:\n", + " - Analyze the user's banking request carefully and identify the primary need\n", + " - If they need credit analysis → use credit_risk_analyzer\n", + " - If they need financial calculations → use financial_calculator\n", + " - If they need account services → use customer_account_manager\n", + " - If they need security analysis → use fraud_detection_system\n", + " - Extract relevant parameters from the user's request\n", + " - Provide helpful, accurate banking responses based on tool outputs\n", + " - Always consider banking regulations, risk management, and best practices\n", + " - Be professional and thorough in your analysis\n", + "\n", + " Choose and use tools wisely to provide the most helpful banking assistance.\n", + " Describe the response in user friendly manner with details describing the tool output. \n", + " Provide the response in at least 500 words.\n", + " Generate a concise execution plan for the banking request.\n", + " \"\"\"" + ] + }, + { + "cell_type": "markdown", + "id": "406835c8", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Initialize the LLM\n", + "\n", + "Let's initialize the LLM that will power our banking agent:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "866066e7", + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the main LLM for banking responses\n", + "main_llm = ChatOpenAI(\n", + " model=\"gpt-5-mini\",\n", + " reasoning={\n", + " \"effort\": \"low\",\n", + " \"summary\": \"auto\"\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "cce9685c", + "metadata": {}, + "source": [ + "Then bind the available banking tools to the LLM, enabling the model to automatically recognize and invoke each tool when appropriate based on request input and the system prompt we defined above:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "906d8132", + "metadata": {}, + "outputs": [], + "source": [ + "# Bind all banking tools to the main LLM\n", + "llm_with_tools = main_llm.bind_tools(AVAILABLE_TOOLS)" + ] + }, + { + "cell_type": "markdown", + "id": "2bad8799", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Define agent state structure\n", + "\n", + "The agent state defines the data structure that flows through the LangGraph workflow. It includes:\n", + "\n", + "- **messages** — The conversation history between the user and agent\n", + "- **user_input** — The current user request\n", + "- **session_id** — A unique identifier for the conversation session\n", + "- **context** — Additional context that can be passed between nodes\n", + "\n", + "Defining this state structure maintains the structure throughout the agent's execution and allows for multi-turn conversations with memory:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "6b926ddf", + "metadata": {}, + "outputs": [], + "source": [ + "# Banking Agent State Definition\n", + "class BankingAgentState(TypedDict):\n", + " messages: Annotated[Sequence[BaseMessage], add_messages]\n", + " user_input: str\n", + " session_id: str\n", + " context: dict" + ] + }, + { + "cell_type": "markdown", + "id": "47ce81b7", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Create agent workflow function\n", + "\n", + "We'll build the LangGraph agent workflow with two main components:\n", + "\n", + "1. **LLM node** — Processes user requests, applies the system prompt, and decides whether to use tools.\n", + "2. **Tools node** — Executes the selected banking tools when the LLM determines they're needed.\n", + "\n", + "The workflow begins with the LLM analyzing the request, then uses tools if needed — or ends if the response is complete, and finally returns to the LLM to generate the final response." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2c9bf585", + "metadata": {}, + "outputs": [], + "source": [ + "def create_banking_langgraph_agent():\n", + " \"\"\"Create a comprehensive LangGraph banking agent with intelligent tool selection.\"\"\"\n", + " def llm_node(state: BankingAgentState) -> BankingAgentState:\n", + " \"\"\"Main LLM node that processes banking requests and selects appropriate tools.\"\"\"\n", + " messages = state[\"messages\"]\n", + " # Add system context to messages\n", + " enhanced_messages = [SystemMessage(content=system_context)] + list(messages)\n", + " # Get LLM response with tool selection\n", + " response = llm_with_tools.invoke(enhanced_messages)\n", + " return {\n", + " **state,\n", + " \"messages\": messages + [response]\n", + " }\n", + " \n", + " def should_continue(state: BankingAgentState) -> str:\n", + " \"\"\"Decide whether to use tools or end the conversation.\"\"\"\n", + " last_message = state[\"messages\"][-1]\n", + " # Check if the LLM wants to use tools\n", + " if hasattr(last_message, 'tool_calls') and last_message.tool_calls:\n", + " return \"tools\"\n", + " return END\n", + " \n", + " # Create the banking state graph\n", + " workflow = StateGraph(BankingAgentState)\n", + " # Add nodes\n", + " workflow.add_node(\"llm\", llm_node)\n", + " workflow.add_node(\"tools\", ToolNode(AVAILABLE_TOOLS))\n", + " # Simplified entry point - go directly to LLM\n", + " workflow.add_edge(START, \"llm\")\n", + " # From LLM, decide whether to use tools or end\n", + " workflow.add_conditional_edges(\n", + " \"llm\",\n", + " should_continue,\n", + " {\"tools\": \"tools\", END: END}\n", + " )\n", + " # Tool execution flows back to LLM for final response\n", + " workflow.add_edge(\"tools\", \"llm\")\n", + " # Set up memory\n", + " memory = MemorySaver()\n", + " # Compile the graph\n", + " agent = workflow.compile(checkpointer=memory)\n", + " return agent" + ] + }, + { + "cell_type": "markdown", + "id": "3eb40287", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Instantiate the banking agent\n", + "\n", + "Now, we'll create an instance of the banking agent by calling the workflow creation function.\n", + "\n", + "This compiled agent is ready to process banking requests and will automatically select and use the appropriate tools based on user queries:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "455b8ee4", + "metadata": {}, + "outputs": [], + "source": [ + "# Create the banking intelligent agent\n", + "banking_agent = create_banking_langgraph_agent()\n", + "\n", + "print(\"Banking LangGraph Agent Created Successfully!\")\n", + "print(\"\\nFeatures:\")\n", + "print(\" - Intelligent banking tool selection\")\n", + "print(\" - Comprehensive banking system prompt\")\n", + "print(\" - Streamlined workflow: LLM → Tools → Response\")\n", + "print(\" - Automatic tool parameter extraction\")\n", + "print(\" - Professional banking assistance\")" + ] + }, + { + "cell_type": "markdown", + "id": "12691528", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Integrate agent with ValidMind\n", + "\n", + "To integrate our LangGraph banking agent with ValidMind, we need to create a wrapper function that ValidMind can use to invoke the agent and extract the necessary information for testing and documentation, allowing ValidMind to run validation tests on the agent's behavior, tool usage, and responses." + ] + }, + { + "cell_type": "markdown", + "id": "7b78509b", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Import ValidMind components\n", + "\n", + "We'll start with importing the necessary ValidMind components for integrating our agent:\n", + "\n", + "- `Prompt` from `validmind.models` for handling prompt-based model inputs\n", + "- `extract_tool_calls_from_agent_output` and `_convert_to_tool_call_list` from `validmind.scorers.llm.deepeval` for extracting and converting tool calls from agent outputs" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9aeb8969", + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.models import Prompt\n", + "from validmind.scorers.llm.deepeval import extract_tool_calls_from_agent_output, _convert_to_tool_call_list" + ] + }, + { + "cell_type": "markdown", + "id": "f67f2955", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Create agent wrapper function\n", + "\n", + "We'll then create a wrapper function that:\n", + "\n", + "- Accepts input in ValidMind's expected format (with `input` and `session_id` fields)\n", + "- Invokes the banking agent with the proper state initialization\n", + "- Captures tool outputs and tool calls for evaluation\n", + "- Returns a standardized response format that includes the prediction, full output, tool messages, and tool call information\n", + "- Handles errors gracefully with fallback responses" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0e4d5a82", + "metadata": {}, + "outputs": [], + "source": [ + "def banking_agent_fn(input):\n", + " \"\"\"\n", + " Invoke the banking agent with the given input.\n", + " \"\"\"\n", + " try:\n", + " # Initial state for banking agent\n", + " initial_state = {\n", + " \"user_input\": input[\"input\"],\n", + " \"messages\": [HumanMessage(content=input[\"input\"])],\n", + " \"session_id\": input[\"session_id\"],\n", + " \"context\": {}\n", + " }\n", + " session_config = {\"configurable\": {\"thread_id\": input[\"session_id\"]}}\n", + " result = banking_agent.invoke(initial_state, config=session_config)\n", + "\n", + " from utils import capture_tool_output_messages\n", + "\n", + " # Capture all tool outputs and metadata\n", + " captured_data = capture_tool_output_messages(result)\n", + " \n", + " # Access specific tool outputs, this will be used for RAGAS tests\n", + " tool_message = \"\"\n", + " for output in captured_data[\"tool_outputs\"]:\n", + " tool_message += output['content']\n", + " \n", + " tool_calls_found = []\n", + " messages = result['messages']\n", + " for message in messages:\n", + " if hasattr(message, 'tool_calls') and message.tool_calls:\n", + " for tool_call in message.tool_calls:\n", + " # Handle both dictionary and object formats\n", + " if isinstance(tool_call, dict):\n", + " tool_calls_found.append(tool_call['name'])\n", + " else:\n", + " # ToolCall object - use attribute access\n", + " tool_calls_found.append(tool_call.name)\n", + "\n", + "\n", + " return {\n", + " \"prediction\": result['messages'][-1].content[0]['text'],\n", + " \"output\": result,\n", + " \"tool_messages\": [tool_message],\n", + " # \"tool_calls\": tool_calls_found,\n", + " \"tool_called\": _convert_to_tool_call_list(extract_tool_calls_from_agent_output(result))\n", + " }\n", + " except Exception as e:\n", + " # Return a fallback response if the agent fails\n", + " error_message = f\"\"\"I apologize, but I encountered an error while processing your banking request: {str(e)}.\n", + " Please try rephrasing your question or contact support if the issue persists.\"\"\"\n", + " return {\n", + " \"prediction\": error_message, \n", + " \"output\": {\n", + " \"messages\": [HumanMessage(content=input[\"input\"]), SystemMessage(content=error_message)],\n", + " \"error\": str(e)\n", + " }\n", + " }" + ] + }, + { + "cell_type": "markdown", + "id": "4bdc90d6", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Initialize the ValidMind model object\n", + "\n", + "We'll also need to register the banking agent as a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data.\n", + "\n", + "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model) that:\n", + "\n", + "- Associates the wrapper function with the model for prediction\n", + "- Stores the system prompt template for documentation\n", + "- Provides a unique `input_id` for tracking and identification\n", + "- Enables the agent to be used with ValidMind's testing and documentation features" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "60a2ce7a", + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the agent as a model\n", + "vm_banking_model = vm.init_model(\n", + " input_id=\"banking_agent_model\",\n", + " predict_fn=banking_agent_fn,\n", + " prompt=Prompt(template=system_context)\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "33ed446a", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Store the agent reference\n", + "\n", + "We'll also store a reference to the original banking agent object in the ValidMind model. This allows us to access the full agent functionality directly if needed, while still maintaining the wrapper function interface for ValidMind's testing framework." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2c653471", + "metadata": {}, + "outputs": [], + "source": [ + "# Add the banking agent to the vm model\n", + "vm_banking_model.model = banking_agent" + ] + }, + { + "cell_type": "markdown", + "id": "bf44ea16", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Verify integration\n", + "\n", + "Let's confirm that the banking agent has been successfully integrated with ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8e101b0f", + "metadata": {}, + "outputs": [], + "source": [ + "print(\"Banking Agent Successfully Integrated with ValidMind!\")\n", + "print(f\"Model ID: {vm_banking_model.input_id}\")" + ] + }, + { + "cell_type": "markdown", + "id": "0c80518d", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Validate the system prompt\n", + "\n", + "Let's get an initial sense of how well our defined system prompt meets a few best practices for prompt engineering by running a few tests — we'll run evaluation tests later on our agent's performance.\n", + "\n", + "You run individual tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module. Passing in our agentic model as an input, the tests below rate the prompt on a scale of 1-10 against the following criteria:\n", + "\n", + "- **[Clarity](https://docs.validmind.ai/tests/prompt_validation/Clarity.html)** — How clearly the prompt states the task.\n", + "- **[Conciseness](https://docs.validmind.ai/tests/prompt_validation/Conciseness.html)** — How succinctly the prompt states the task.\n", + "- **[Delimitation](https://docs.validmind.ai/tests/prompt_validation/Delimitation.html)** — When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", + "- **[NegativeInstruction](https://docs.validmind.ai/tests/prompt_validation/NegativeInstruction.html)** — Whether the prompt contains negative instructions.\n", + "- **[Specificity](https://docs.validmind.ai/tests/prompt_validation/NegativeInstruction.html)** — How specific the prompt defines the task." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f52dceb1", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.Clarity\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "70d52333", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.Conciseness\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5aa89976", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.Delimitation\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8630197e", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.NegativeInstruction\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "bba99915", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.Specificity\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "af4d6d77", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Initialize the ValidMind datasets\n", + "\n", + "After validation our system prompt, let's import our sample dataset ([banking_test_dataset.py](banking_test_dataset.py)), which we'll use in the next section to evaluate our agent's performance across different banking scenarios:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0c70ca2c", + "metadata": {}, + "outputs": [], + "source": [ + "from banking_test_dataset import banking_test_dataset" + ] + }, + { + "cell_type": "markdown", + "id": "0268ce6e", + "metadata": {}, + "source": [ + "The next step is to connect your data with a ValidMind `Dataset` object. **This step is always necessary every time you want to connect a dataset to documentation and produce test results through ValidMind,** but you only need to do it once per dataset.\n", + "\n", + "Initialize a ValidMind dataset object using the [`init_dataset` function](https://docs.validmind.ai/validmind/validmind.html#init_dataset) from the ValidMind (`vm`) module. For this example, we'll pass in the following arguments:\n", + "\n", + "- **`input_id`** — A unique identifier that allows tracking what inputs are used when running each individual test.\n", + "- **`dataset`** — The raw dataset that you want to provide as input to tests.\n", + "- **`text_column`** — The name of the column containing the text input data.\n", + "- **`target_column`** — A required argument if tests require access to true values. This is the name of the target column in the dataset." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a7e9d158", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset = vm.init_dataset(\n", + " input_id=\"banking_test_dataset\",\n", + " dataset=banking_test_dataset,\n", + " text_column=\"input\",\n", + " target_column=\"possible_outputs\",\n", + ")\n", + "\n", + "print(\"Banking Test Dataset Initialized in ValidMind!\")\n", + "print(f\"Dataset ID: {vm_test_dataset.input_id}\")\n", + "print(f\"Dataset columns: {vm_test_dataset._df.columns}\")\n", + "vm_test_dataset._df" + ] + }, + { + "cell_type": "markdown", + "id": "b9143fb6", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign predictions\n", + "\n", + "Now that both the model object and the datasets have been registered, we'll assign predictions to capture the banking agent's responses for evaluation:\n", + "\n", + "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#assign_predictions) from the `Dataset` object can link existing predictions to any number of models.\n", + "- This method links the model's class prediction values and probabilities to our `vm_train_ds` and `vm_test_ds` datasets.\n", + "\n", + "If no prediction values are passed, the method will compute predictions automatically:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1d462663", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_predictions(vm_banking_model)\n", + "\n", + "print(\"Banking Agent Predictions Generated Successfully!\")\n", + "print(f\"Predictions assigned to {len(vm_test_dataset._df)} test cases\")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "8e50467e", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Running accuracy tests\n", + "\n", + "Using [`@vm.test`](https://docs.validmind.ai/validmind/validmind.html#test), let's implement some reusable custom *inline tests* to assess the accuracy of our banking agent:\n", + "\n", + "- An inline test refers to a test written and executed within the same environment as the code being tested — in this case, right in this Jupyter Notebook — without requiring a separate test file or framework.\n", + "- You'll note that the custom test functions are just regular Python functions that can include and require any Python library as you see fit." + ] + }, + { + "cell_type": "markdown", + "id": "6d8a9b90", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Response accuracy test\n", + "\n", + "We'll create a custom test that evaluates the banking agent's ability to provide accurate responses by:\n", + "\n", + "- Testing against a dataset of predefined banking questions and expected answers.\n", + "- Checking if responses contain expected keywords and banking terminology.\n", + "- Providing detailed test results including pass/fail status.\n", + "- Helping identify any gaps in the agent's banking knowledge or response quality." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "90232066", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "@vm.test(\"my_custom_tests.banking_accuracy_test\")\n", + "def banking_accuracy_test(model, dataset, list_of_columns):\n", + " \"\"\"\n", + " The Banking Accuracy Test evaluates whether the agent’s responses include \n", + " critical domain-specific keywords and phrases that indicate accurate, compliant,\n", + " and contextually appropriate banking information. This test ensures that the agent\n", + " provides responses containing the expected banking terminology, risk classifications,\n", + " account details, or other domain-relevant information required for regulatory compliance,\n", + " customer safety, and operational accuracy.\n", + " \"\"\"\n", + " df = dataset._df\n", + " \n", + " # Pre-compute responses for all tests\n", + " y_true = dataset.y.tolist()\n", + " y_pred = dataset.y_pred(model).tolist()\n", + "\n", + " # Vectorized test results\n", + " test_results = []\n", + " for response, keywords in zip(y_pred, y_true):\n", + " # Convert keywords to list if not already a list\n", + " if not isinstance(keywords, list):\n", + " keywords = [keywords]\n", + " test_results.append(any(str(keyword).lower() in str(response).lower() for keyword in keywords))\n", + " \n", + " results = pd.DataFrame()\n", + " column_names = [col + \"_details\" for col in list_of_columns]\n", + " results[column_names] = df[list_of_columns]\n", + " results[\"actual\"] = y_pred\n", + " results[\"expected\"] = y_true\n", + " results[\"passed\"] = test_results\n", + " results[\"error\"] = None if test_results else f'Response did not contain any expected keywords: {y_true}'\n", + " \n", + " return results" + ] + }, + { + "cell_type": "markdown", + "id": "7eed5265", + "metadata": {}, + "source": [ + "Now that we've defined our custom response accuracy test, we can run the test using the same `run_test()` function we used earlier to validate the system prompt using our sample dataset and agentic model as input, and log the test results to the ValidMind Platform with the [`log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#log):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e68884d5", + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.banking_accuracy_test\",\n", + " inputs={\n", + " \"dataset\": vm_test_dataset,\n", + " \"model\": vm_banking_model\n", + " },\n", + " params={\n", + " \"list_of_columns\": [\"input\"]\n", + " }\n", + ")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "id": "4d758ddf", + "metadata": {}, + "source": [ + "Let's review the first five rows of the test dataset to inspect the results to see how well the banking agent performed. Each column in the output serves a specific purpose in evaluating agent performance:\n", + "\n", + "| Column header | Description | Importance |\n", + "|--------------|-------------|------------|\n", + "| **`input`** | Original user query or request | Essential for understanding the context of each test case and tracing which inputs led to specific agent behaviors. |\n", + "| **`expected_tools`** | Banking tools that should be invoked for this request | Enables validation of correct tool selection, which is critical for agentic AI systems where choosing the right tool is a key success metric. |\n", + "| **`expected_output`** | Expected output or keywords that should appear in the response | Defines the success criteria for each test case, enabling objective evaluation of whether the agent produced the correct result. |\n", + "| **`session_id`** | Unique identifier for each test session | Allows tracking and correlation of related test runs, debugging specific sessions, and maintaining audit trails. |\n", + "| **`category`** | Classification of the request type | Helps organize test results by domain and identify performance patterns across different banking use cases. |\n", + "| **`banking_agent_model_output`** | Complete agent response including all messages and reasoning | Allows you to examine the full output to assess response quality, completeness, and correctness beyond just keyword matching. |\n", + "| **`banking_agent_model_tool_messages`** | Messages exchanged with the banking tools | Critical for understanding how the agent interacted with tools, what parameters were passed, and what tool outputs were received. |\n", + "| **`banking_agent_model_tool_called`** | Specific tool that was invoked | Enables validation that the agent selected the correct tool for each request, which is fundamental to agentic AI validation. |\n", + "| **`possible_outputs`** | Alternative valid outputs or keywords that could appear in the response | Provides flexibility in evaluation by accounting for multiple acceptable response formats or variations. |" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "78f7edb1", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.df.head(5)" + ] + }, + { + "cell_type": "markdown", + "id": "6f233bef", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Tool selection accuracy test\n", + "\n", + "We'll also create a custom test that evaluates the banking agent's ability to select the correct tools for different requests by:\n", + "\n", + "- Testing against a dataset of predefined banking queries with expected tool selections.\n", + "- Comparing the tools actually invoked by the agent against the expected tools for each request.\n", + "- Providing quantitative accuracy scores that measure the proportion of expected tools correctly selected.\n", + "- Helping identify gaps in the agent's understanding of user needs and tool selection logic." + ] + }, + { + "cell_type": "markdown", + "id": "d0b46111", + "metadata": {}, + "source": [ + "First, we'll define a helper function that extracts tool calls from the agent's messages and compares them against the expected tools. This function handles different message formats (dictionary or object) and calculates accuracy scores:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e68798be", + "metadata": {}, + "outputs": [], + "source": [ + "def validate_tool_calls_simple(messages, expected_tools):\n", + " \"\"\"Simple validation of tool calls without RAGAS dependency issues.\"\"\"\n", + " \n", + " tool_calls_found = []\n", + " \n", + " for message in messages:\n", + " if hasattr(message, 'tool_calls') and message.tool_calls:\n", + " for tool_call in message.tool_calls:\n", + " # Handle both dictionary and object formats\n", + " if isinstance(tool_call, dict):\n", + " tool_calls_found.append(tool_call['name'])\n", + " else:\n", + " # ToolCall object - use attribute access\n", + " tool_calls_found.append(tool_call.name)\n", + " \n", + " # Check if expected tools were called\n", + " accuracy = 0.0\n", + " matches = 0\n", + " if expected_tools:\n", + " matches = sum(1 for tool in expected_tools if tool in tool_calls_found)\n", + " accuracy = matches / len(expected_tools)\n", + " \n", + " return {\n", + " 'expected_tools': expected_tools,\n", + " 'found_tools': tool_calls_found,\n", + " 'matches': matches,\n", + " 'total_expected': len(expected_tools) if expected_tools else 0,\n", + " 'accuracy': accuracy,\n", + " }" + ] + }, + { + "cell_type": "markdown", + "id": "1b45472c", + "metadata": {}, + "source": [ + "Now we'll define the main test function that uses the helper function to evaluate tool selection accuracy across all test cases in the dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "604d7313", + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_custom_tests.BankingToolCallAccuracy\")\n", + "def BankingToolCallAccuracy(dataset, agent_output_column, expected_tools_column):\n", + " \"\"\"\n", + " Evaluates the tool selection accuracy of a LangGraph-powered banking agent.\n", + "\n", + " This test measures whether the agent correctly identifies and invokes the required banking tools\n", + " for each user query scenario.\n", + " For each case, the outputs generated by the agent (including its tool calls) are compared against an\n", + " expected set of tools. The test considers both coverage and exactness: it computes the proportion of\n", + " expected tools correctly called by the agent for each instance.\n", + "\n", + " Parameters:\n", + " dataset (VMDataset): The dataset containing user queries, agent outputs, and ground-truth tool expectations.\n", + " agent_output_column (str): Dataset column name containing agent outputs (should include tool call details in 'messages').\n", + " expected_tools_column (str): Dataset column specifying the true expected tools (as lists).\n", + "\n", + " Returns:\n", + " List[dict]: Per-row dictionaries with details: expected tools, found tools, match count, total expected, and accuracy score.\n", + "\n", + " Purpose:\n", + " Provides diagnostic evidence of the banking agent's core reasoning ability—specifically, its capacity to\n", + " interpret user needs and select the correct banking actions. Useful for diagnosing gaps in tool coverage,\n", + " misclassifications, or breakdowns in agent logic.\n", + "\n", + " Interpretation:\n", + " - An accuracy of 1.0 signals perfect tool selection for that example.\n", + " - Lower scores may indicate partial or complete failures to invoke required tools.\n", + " - Review 'found_tools' vs. 'expected_tools' to understand the source of discrepancies.\n", + "\n", + " Strengths:\n", + " - Directly tests a core capability of compositional tool-use agents.\n", + " - Framework-agnostic; robust to tool call output format (object or dict).\n", + " - Supports batch validation and result logging for systematic documentation.\n", + "\n", + " Limitations:\n", + " - Does not penalize extra, unnecessary tool calls.\n", + " - Does not assess result quality—only correct invocation.\n", + "\n", + " \"\"\"\n", + " df = dataset._df\n", + " \n", + " results = []\n", + " for i, row in df.iterrows():\n", + " result = validate_tool_calls_simple(row[agent_output_column]['messages'], row[expected_tools_column])\n", + " results.append(result)\n", + " \n", + " return results" + ] + }, + { + "cell_type": "markdown", + "id": "d594c973", + "metadata": {}, + "source": [ + "Finally, we can call our function with `run_test()` and log the test results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dd14115e", + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.BankingToolCallAccuracy\",\n", + " inputs={\n", + " \"dataset\": vm_test_dataset,\n", + " },\n", + " params={\n", + " \"agent_output_column\": \"banking_agent_model_output\",\n", + " \"expected_tools_column\": \"expected_tools\"\n", + " }\n", + ")\n", + "result.log()" + ] + }, + { + "cell_type": "markdown", + "id": "f78f4107", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Assigning AI evaluation metric scores\n", + "\n", + "*AI agent evaluation metrics* are specialized measurements designed to assess how well autonomous LLM-based agents reason, plan, select and execute tools, and ultimately complete user tasks by analyzing the *full execution trace* — including reasoning steps, tool calls, intermediate decisions, and outcomes, rather than just single input–output pairs. These metrics are essential because agent failures often occur in ways traditional LLM metrics miss — for example, choosing the right tool with wrong arguments, creating a good plan but not following it, or completing a task inefficiently.\n", + "\n", + "In this section, we'll evaluate our banking agent's outputs and add scoring to our sample dataset against metrics defined in [DeepEval’s AI agent evaluation framework](https://deepeval.com/guides/guides-ai-agent-evaluation-metrics) which breaks down AI agent evaluation into three layers with corresponding subcategories: **reasoning**, **action**, and **execution**.\n", + "\n", + "Together, these three metrics enable granular diagnosis of agent behavior, help pinpoint where failures occur (reasoning, action, or execution), and support both development benchmarking and production monitoring." + ] + }, + { + "cell_type": "markdown", + "id": "3a9c853a", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Identify relevant DeepEval scorers\n", + "\n", + "*Scorers* are evaluation metrics that analyze model outputs and store their results in the dataset:\n", + "\n", + "- Each scorer adds a new column to the dataset with format: `{scorer_name}_{metric_name}`\n", + "- The column contains the numeric score (typically `0`-`1`) for each example\n", + "- Multiple scorers can be run on the same dataset, each adding their own column\n", + "- Scores are persisted in the dataset for later analysis and visualization\n", + "- Common scorer patterns include:\n", + " - Model performance metrics (accuracy, F1, etc.)\n", + " - Output quality metrics (relevance, faithfulness)\n", + " - Task-specific metrics (completion, correctness)\n", + "\n", + "Use `list_scorers()` from [`validmind.scorers`](https://docs.validmind.ai/validmind/validmind/tests.html#scorer) to discover all available scoring methods and their IDs that can be used with `assign_scores()`. We'll filter these results to return only DeepEval scorers for our desired three metrics in a formatted table with descriptions:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "730c70ec", + "metadata": {}, + "outputs": [], + "source": [ + "# Load all DeepEval scorers\n", + "llm_scorers_dict = vm.tests.load._load_tests([s for s in vm.scorer.list_scorers() if \"deepeval\" in s.lower()])\n", + "\n", + "# Categorize scorers by metric layer\n", + "reasoning_scorers = {}\n", + "action_scorers = {}\n", + "execution_scorers = {}\n", + "\n", + "for scorer_id, scorer_func in llm_scorers_dict.items():\n", + " tags = getattr(scorer_func, \"__tags__\", [])\n", + " scorer_name = scorer_id.split(\".\")[-1]\n", + "\n", + " if \"reasoning_layer\" in tags:\n", + " reasoning_scorers[scorer_id] = scorer_func\n", + " elif \"action_layer\" in tags:\n", + " action_scorers[scorer_id] = scorer_func\n", + " elif \"TaskCompletion\" in scorer_name:\n", + " execution_scorers[scorer_id] = scorer_func\n", + "\n", + "# Display scorers by category\n", + "print(\"=\" * 80)\n", + "print(\"REASONING LAYER\")\n", + "print(\"=\" * 80)\n", + "if reasoning_scorers:\n", + " reasoning_df = vm.tests.load._pretty_list_tests(reasoning_scorers, truncate=True)\n", + " display(reasoning_df)\n", + "else:\n", + " print(\"No reasoning layer scorers found.\")\n", + "\n", + "print(\"\\n\" + \"=\" * 80)\n", + "print(\"ACTION LAYER\")\n", + "print(\"=\" * 80)\n", + "if action_scorers:\n", + " action_df = vm.tests.load._pretty_list_tests(action_scorers, truncate=True)\n", + " display(action_df)\n", + "else:\n", + " print(\"No action layer scorers found.\")\n", + "\n", + "print(\"\\n\" + \"=\" * 80)\n", + "print(\"EXECUTION LAYER\")\n", + "print(\"=\" * 80)\n", + "if execution_scorers:\n", + " execution_df = vm.tests.load._pretty_list_tests(execution_scorers, truncate=True)\n", + " display(execution_df)\n", + "else:\n", + " print(\"No execution layer scorers found.\")" + ] + }, + { + "cell_type": "markdown", + "id": "4dd73d0d", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign reasoning scores\n", + "\n", + "*Reasoning* evaluates planning and strategy generation:\n", + "\n", + "- **Plan quality** – How logical, complete, and efficient the agent’s plan is.\n", + "- **Plan adherence** – Whether the agent follows its own plan during execution." + ] + }, + { + "cell_type": "markdown", + "id": "06ccae28", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Plan quality score\n", + "\n", + "Let's measure how well our banking agent generates a plan before acting. A high score means the plan is logical, complete, and efficient." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "52f362ba", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_scores(\n", + " metrics = \"validmind.scorers.llm.deepeval.PlanQuality\",\n", + " input_column = \"input\",\n", + " actual_output_column = \"banking_agent_model_prediction\",\n", + " tools_called_column = \"banking_agent_model_tool_called\",\n", + " agent_output_column = \"banking_agent_model_output\",\n", + ")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "8dcdc88f", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Plan adherence score\n", + "\n", + "Let's check whether our banking agent follows the plan it created. Deviations lower this score and indicate gaps between reasoning and execution." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4124a7c2", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_scores(\n", + " metrics = \"validmind.scorers.llm.deepeval.PlanAdherence\",\n", + " input_column = \"input\",\n", + " actual_output_column = \"banking_agent_model_prediction\",\n", + " expected_output_column = \"expected_output\",\n", + " tools_called_column = \"banking_agent_model_tool_called\",\n", + " agent_output_column = \"banking_agent_model_output\",\n", + "\n", + ")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "6da1ac95", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign action scores\n", + "\n", + "*Action* assesses tool usage and argument generation:\n", + "\n", + "- **Tool correctness** – Whether the agent selects and calls the right tools.\n", + "- **Argument correctness** – Whether the agent generates correct tool arguments." + ] + }, + { + "cell_type": "markdown", + "id": "d4db8270", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Tool correctness score\n", + "\n", + "Let's evaluate if our banking agent selects the appropriate tool for the task. Choosing the wrong tool reduces performance even if reasoning was correct." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8d2e8a25", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_scores(\n", + " metrics = \"validmind.scorers.llm.deepeval.ToolCorrectness\",\n", + " input_column = \"input\",\n", + " actual_output_column = \"banking_agent_model_prediction\",\n", + " tools_called_column = \"banking_agent_model_tool_called\",\n", + " expected_tools_column = \"expected_tools\",\n", + " agent_output_column = \"banking_agent_model_output\",\n", + "\n", + ")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "9aa50b05", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Argument correctness score\n", + "\n", + "Let's assesses whether our banking agent provides correct inputs or arguments to the selected tool. Incorrect arguments can lead to failed or unexpected results." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "04f90489", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_scores(\n", + " metrics = \"validmind.scorers.llm.deepeval.ArgumentCorrectness\",\n", + " input_column = \"input\",\n", + " actual_output_column = \"banking_agent_model_prediction\",\n", + " tools_called_column = \"banking_agent_model_tool_called\",\n", + " agent_output_column = \"banking_agent_model_output\",\n", + "\n", + ")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "c59e5595", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign execution score\n", + "\n", + "*Execution* measures end-to-end performance:\n", + "\n", + "- **Task completion** – Whether the agent successfully completes the intended task.\n" + ] + }, + { + "cell_type": "markdown", + "id": "d64600ca", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Task completion score\n", + "\n", + "Let's evaluate whether our banking agent successfully completes the requested tasks. Incomplete task execution can lead to user dissatisfaction and failed banking operations." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "05024f1f", + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_dataset.assign_scores(\n", + " metrics = \"validmind.scorers.llm.deepeval.TaskCompletion\",\n", + " input_column = \"input\",\n", + " actual_output_column = \"banking_agent_model_prediction\",\n", + " agent_output_column = \"banking_agent_model_output\",\n", + " tools_called_column = \"banking_agent_model_tool_called\",\n", + "\n", + ")\n", + "vm_test_dataset._df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "21aa9b0d", + "metadata": {}, + "source": [ + "As you recall from the beginning of this section, when we run scorers through `assign_scores()`, the return values are automatically processed and added as new columns with the format `{scorer_name}_{metric_name}`. Note that the task completion scorer has added a new column `TaskCompletion_score` to our dataset.\n", + "\n", + "We'll use this column to visualize the distribution of task completion scores across our test cases through the [BoxPlot test](https://docs.validmind.ai/validmind/validmind/tests/plots/BoxPlot.html#boxplot):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7f6d08ca", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.plots.BoxPlot\",\n", + " inputs={\"dataset\": vm_test_dataset},\n", + " params={\n", + " \"columns\": \"TaskCompletion_score\",\n", + " \"title\": \"Distribution of Task Completion Scores\",\n", + " \"ylabel\": \"Score\",\n", + " \"figsize\": (8, 6)\n", + " }\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "012bbcb8", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Running RAGAS tests\n", + "\n", + "Next, let's run some out-of-the-box *Retrieval-Augmented Generation Assessment* (RAGAS) tests available in the ValidMind Library. RAGAS provides specialized metrics for evaluating retrieval-augmented generation systems and conversational AI agents. These metrics analyze different aspects of agent performance by assessing how well systems integrate retrieved information with generated responses.\n", + "\n", + "Our banking agent uses tools to retrieve information and generates responses based on that context, making it similar to a RAG system. RAGAS metrics help evaluate the quality of this integration by analyzing the relationship between retrieved tool outputs, user queries, and generated responses.\n", + "\n", + "These tests provide insights into how well our banking agent integrates tool usage with conversational abilities, ensuring it provides accurate, relevant, and helpful responses to banking users while maintaining fidelity to retrieved information." + ] + }, + { + "cell_type": "markdown", + "id": "2036afba", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Identify relevant RAGAS tests\n", + "\n", + "Let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your development testing, and helps you ensure that your models are being documented and evaluated appropriately.\n", + "\n", + "You can pass `tasks` and `tags` as parameters to the [`vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to filter the tests based on the tags and task types:\n", + "\n", + "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `text_qa` tasks.\n", + "- **`tags`** are free-form descriptions providing more details about the test, for example, what category the test falls into. Here we'll focus on the `ragas` tag.\n", + "\n", + "We'll then run three of these tests returned as examples below." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0701f5a9", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(task=\"text_qa\", tags=[\"ragas\"])" + ] + }, + { + "cell_type": "markdown", + "id": "c1741ffc", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Faithfulness\n", + "\n", + "Let's evaluate whether the banking agent's responses accurately reflect the information retrieved from tools. Unfaithful responses can misreport credit analysis, financial calculations, and compliance results—undermining user trust in the banking agent." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "92044533", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.ragas.Faithfulness\",\n", + " inputs={\"dataset\": vm_test_dataset},\n", + " param_grid={\n", + " \"user_input_column\": [\"input\"],\n", + " \"response_column\": [\"banking_agent_model_prediction\"],\n", + " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "42b71ccc", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Response Relevancy\n", + "\n", + "Let's evaluate whether the banking agent's answers address the user's original question or request. Irrelevant or off-topic responses can frustrate users and fail to deliver the banking information they need." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "d7483bc3", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", + " inputs={\"dataset\": vm_test_dataset},\n", + " params={\n", + " \"user_input_column\": \"input\",\n", + " \"response_column\": \"banking_agent_model_prediction\",\n", + " \"retrieved_contexts_column\": \"banking_agent_model_tool_messages\",\n", + " }\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "4f4d0569", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Context Recall\n", + "\n", + "Let's evaluate how well the banking agent uses the information retrieved from tools when generating its responses. Poor context recall can lead to incomplete or underinformed answers even when the right tools were selected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e5dc00ce", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.ragas.ContextRecall\",\n", + " inputs={\"dataset\": vm_test_dataset},\n", + " param_grid={\n", + " \"user_input_column\": [\"input\"],\n", + " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", + " \"reference_column\": [\"banking_agent_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "b987b00e", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Running safety tests\n", + "\n", + "Finally, let's run some out-of-the-box *safety* tests available in the ValidMind Library. Safety tests provide specialized metrics for evaluating whether AI agents operate reliably and securely. These metrics analyze different aspects of agent behavior by assessing adherence to safety guidelines, consistency of outputs, and resistance to harmful or inappropriate requests.\n", + "\n", + "Our banking agent handles sensitive financial information and user requests, making safety and reliability essential. Safety tests help evaluate whether the agent maintains appropriate boundaries, responds consistently and correctly to inputs, and avoids generating harmful, biased, or unprofessional content.\n", + "\n", + "These tests provide insights into how well our banking agent upholds standards of fairness and professionalism, ensuring it operates reliably and securely for banking users." + ] + }, + { + "cell_type": "markdown", + "id": "a754cca3", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### AspectCritic\n", + "\n", + "Let's evaluate our banking agent's responses across multiple quality dimensions — conciseness, coherence, correctness, harmfulness, and maliciousness. Weak performance on these dimensions can degrade user experience, fall short of professional banking standards, or introduce safety risks. \n", + "\n", + "We'll use the `AspectCritic` we identified earlier:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "148daa2b", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.ragas.AspectCritic\",\n", + " inputs={\"dataset\": vm_test_dataset},\n", + " param_grid={\n", + " \"user_input_column\": [\"input\"],\n", + " \"response_column\": [\"banking_agent_model_prediction\"],\n", + " \"retrieved_contexts_column\": [\"banking_agent_model_tool_messages\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "92e5b1f6", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Bias\n", + "\n", + "Let's evaluate whether our banking agent's prompts contain unintended biases that could affect banking decisions. Biased prompts can lead to unfair or discriminatory outcomes — undermining customer trust and exposing the institution to compliance risk.\n", + "\n", + "We'll first use `list_tests()` again to filter for tests relating to `prompt_validation`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "74eba86c", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(filter=\"prompt_validation\")" + ] + }, + { + "cell_type": "markdown", + "id": "bcc66b65", + "metadata": {}, + "source": [ + "And then run the identified `Bias` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "062cf8e7", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.prompt_validation.Bias\",\n", + " inputs={\n", + " \"model\": vm_banking_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "a2832750", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Next steps\n", + "\n", + "You can look at the output produced by the ValidMind Library right in the notebook where you ran the code, as you would expect. But there is a better way — use the ValidMind Platform to work with your model documentation." + ] + }, + { + "cell_type": "markdown", + "id": "a8cb1a58", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Work with your model documentation\n", + "\n", + "1. From the **Inventory** in the ValidMind Platform, go to the model you registered earlier. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/working-with-model-inventory.html))\n", + "\n", + "2. In the left sidebar that appears for your model, click **Documentation** under Documents.\n", + "\n", + " What you see is the full draft of your model documentation in a more easily consumable version. From here, you can make qualitative edits to model documentation, view guidelines, collaborate with validators, and submit your model documentation for approval when it's ready. [Learn more ...](https://docs.validmind.ai/guide/working-with-model-documentation.html)\n", + "\n", + "3. Click into any section related to the tests we ran in this notebook, for example: **4.3. Prompt Evaluation** to review the results of the tests we logged." + ] + }, + { + "cell_type": "markdown", + "id": "94ef26be", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Customize the banking agent for your use case\n", + "\n", + "You've now built an agentic AI system designed for banking use cases that supports compliance with supervisory guidance such as SR 11-7 and SS1/23, covering credit and fraud risk assessment for both retail and commercial banking. Extend this example agent to real-world banking scenarios and production deployment by:\n", + "\n", + "- Adapting the banking tools to your organization's specific requirements\n", + "- Adding more banking scenarios and edge cases to your test set\n", + "- Connecting the agent to your banking systems and databases\n", + "- Implementing additional banking-specific tools and workflows" + ] + }, + { + "cell_type": "markdown", + "id": "a681e49c", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Discover more learning resources\n", + "\n", + "Learn more about the ValidMind Library tools we used in this notebook:\n", + "\n", + "- [Custom prompts](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.html)\n", + "- [Custom tests](https://docs.validmind.ai/notebooks/how_to/tests/custom_tests/implement_custom_tests.html)\n", + "- [ValidMind scorers](https://docs.validmind.ai/notebooks/how_to/scoring/assign_scores_complete_tutorial.html)\n", + "\n", + "We also offer many more interactive notebooks to help you document models:\n", + "\n", + "- [Run tests & test suites](https://docs.validmind.ai/guide/testing-overview.html)\n", + "- [Code samples](https://docs.validmind.ai/guide/samples-jupyter-notebooks.html)\n", + "\n", + "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." + ] + }, + { + "cell_type": "markdown", + "id": "707c1b6e", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Upgrade ValidMind\n", + "\n", + "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", + "\n", + "Retrieve the information for the currently installed version of ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9733adff", + "metadata": {}, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "id": "e4b0b646", + "metadata": {}, + "source": [ + "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", + "\n", + "```bash\n", + "%pip install --upgrade validmind\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "387fa7f1", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-de4baf0f42ba4a37946d52586dff1049", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "validmind-1QuffXMV-py3.11", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.9" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/site/notebooks/code_samples/agents/utils.py b/site/notebooks/use_cases/agents/utils.py similarity index 100% rename from site/notebooks/code_samples/agents/utils.py rename to site/notebooks/use_cases/agents/utils.py diff --git a/site/notebooks/use_cases/capital_markets/capital_markets_template.yaml b/site/notebooks/use_cases/capital_markets/capital_markets_template.yaml new file mode 100644 index 0000000000..9cb561dc27 --- /dev/null +++ b/site/notebooks/use_cases/capital_markets/capital_markets_template.yaml @@ -0,0 +1,227 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial + +- id: model_metadata + title: Model Metadata and Stakeholders + index_only: true + sections: + - id: metadata + title: Metadata + parent_section: model_metadata + guidelines: + - Provide detailed metadata to uniquely identify the model, ensuring + that each entry is traceable and consistent with internal records. + - Specify the platform to enable validation and governance teams to + understand technical dependencies. + - Metadata should be aligned with enterprise systems (e.g., Model Risk + Management System). + - Be specific about versioning and any platform customizations. + - id: stakeholders + title: Stakeholders + parent_section: model_metadata + guidelines: + - List all individuals responsible for the development, use, and + oversight of the model. + - Ensure clarity in responsibilities to avoid overlap and gaps in + accountability. + - Include specific roles, such as "Data Scientist," "Validation Lead," + or "Business Analyst," and describe their responsibilities relative to + the model. +- id: business_context + title: Business Context and Purpose + index_only: true + sections: + - id: business_problem + title: Business Problem and Objectives + parent_section: business_context + guidelines: + - Clearly describe the problem the model addresses and its alignment + with business goals. + - Include specific use cases, outputs, and highlight regulatory + expectations to demonstrate compliance. + - Specify compliance requirements, such as IFRS, Basel III or SR11-7, as + applicable. + - id: products_and_risks + title: Products and Risks + parent_section: business_context + guidelines: + - Detail how the model impacts products or processes and specify + associated risks (e.g., market, credit, or operational risks). + - Provide a description of business impacts and compliance with + applicable regulations. +- id: model_design + title: Model Methodology and Design + index_only: true + sections: + - id: theoretical_foundations + title: Theoretical Foundations + parent_section: model_design + guidelines: + - Explain the methodology, assumptions, and logic underlying the model. + - Clearly document how key risks (e.g., credit default or liquidity) are + addressed within the model. + - Highlight any simplifications or approximations and their implications. + - id: architecture + title: Model Architecture + parent_section: model_design + guidelines: + - Use diagrams to illustrate the model's structure and data flow. + - Include visual flowcharts detailing how inputs are transformed into + outputs. + - Highlight dependencies affecting validation or performance. + - id: selection + title: Model Selection and Justification + parent_section: model_design + guidelines: + - Document the decision-making process, including comparisons to + alternative approaches. + - Provide rationale for model selection using performance metrics like + R-squared or RMSE. + - Highlight reasons for rejecting alternatives. +- id: model_data + title: Model Data + index_only: true + sections: + - id: input_data + title: Input Data + parent_section: model_data + guidelines: + - Detail the provenance and quality of input data, including + preprocessing steps like imputation or outlier detection. + - Highlight known issues, such as stale data or incomplete time series. + - Suggest standard tests, e.g., null value checks, distribution + matching, or correlation analysis. + - id: dataset_characteristics + title: Development Dataset + parent_section: model_data + guidelines: + - Summarize dataset characteristics like size, representativeness, and + scope. + - Include validation metrics such as sampling error or coverage ratios. + - id: outputs + title: Outputs + parent_section: model_data + guidelines: + - Define outputs, their usage, and storage mechanisms. + - Highlight data formats (e.g., APIs, flat files) and ensure outputs are + validated for consistency and accuracy. +- id: model_testing + title: Model Testing + index_only: true + sections: + - id: diagnostic_testing + title: Diagnostic Testing + parent_section: model_testing + guidelines: + - Provide details of diagnostic tests performed to ensure model + performance and identify anomalies. + - Include standard diagnostic tests such as - Residual analysis to check + model predictions. - Comparison of predicted versus actual outcomes + for validity. + - Summarize findings and highlight any performance gaps. + - id: sensitivity_stress_testing + title: Sensitivity and Stress Testing + parent_section: model_testing + guidelines: + - Sensitivity Testing. Describe how changes in input variables affect + model outputs. + - Stress Testing. Document model performance under extreme conditions or + assumptions. + - Use tests to measure how minor changes in key parameters affect + results. + - Simulate scenarios like extreme economic downturns to evaluate + robustness. + - id: performance_testing + title: Performance Testing + parent_section: model_testing + guidelines: + - Provide details of performance metrics such as RMSE, AUC, or + precision/recall. + - Benchmark comparisons. Compare performance with industry standards or + alternative models. + - Include visual aids like ROC curves or confusion matrices to + illustrate performance. + - id: back_testing + title: Back-Testing + parent_section: model_testing + guidelines: + - Highlight alignment of predictions with observed outcomes through + historical analysis. + - Test model predictions against historical data outcomes. + - Document discrepancies and propose remediation steps. +- id: implementation + title: Model Implementation + index_only: true + sections: + - id: production_environment + title: Production Environment + parent_section: implementation + guidelines: + - Describe the implementation environment, such as cloud platforms like + AWS or GCP. + - Highlight integration points, such as database connectors or REST + APIs, ensuring consistency with design specifications. + - id: implementation_testing + title: Implementation Testing + parent_section: implementation + guidelines: + - Document verification steps, including parallel runs against legacy + systems and end-to-end pipeline testing. +- id: limitations_adjustments + title: Assumptions, Limitations, and Adjustments + index_only: true + sections: + - id: assumptions + title: Assumptions + parent_section: limitations_adjustments + guidelines: + - List assumptions critical to model functionality and provide + justifications. + - Include potential impact analyses if assumptions fail. + - id: limitations + title: Limitations + parent_section: limitations_adjustments + guidelines: + - Highlight known limitations and mitigation strategies. + - Discuss implications for model performance or reliability. + - id: adjustments + title: Adjustments + parent_section: limitations_adjustments + guidelines: + - Document overrides and their justification, including governance + processes. +- id: monitoring_controls_documentation + title: Model Monitoring and Controls + index_only: true + sections: + - id: ongoing + title: Ongoing Model Monitoring + parent_section: monitoring_controls_documentation + guidelines: + - Outline a monitoring plan, including performance metrics, monitoring + frequency, and escalation thresholds. + - Include drift analysis of input data distributions and stability + metrics for output consistency. + - id: governance + title: Governance + parent_section: monitoring_controls_documentation + guidelines: + - Define access controls, version management, and governance frameworks. + - Highlight periodic audits and role-based access controls. +- id: documentation_references + title: Documentation References + index_only: true + sections: + - id: supporting_documents + title: Supporting Documents + parent_section: documentation_references + guidelines: + - Provide references to related documentation, such as validation + reports. + - Maintain a change log for systematic traceability. + - id: appendices + title: Appendices + parent_section: documentation_references + guidelines: + - Use appendices for supplementary data, testing results, and glossaries. diff --git a/site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb b/site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models.ipynb similarity index 99% rename from site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb rename to site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models.ipynb index 6953959242..a218950c56 100644 --- a/site/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb +++ b/site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models.ipynb @@ -117,7 +117,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", diff --git a/site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models_quantlib.ipynb b/site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models_quantlib.ipynb new file mode 100644 index 0000000000..de541b448a --- /dev/null +++ b/site/notebooks/use_cases/capital_markets/quickstart_option_pricing_models_quantlib.ipynb @@ -0,0 +1,1345 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "1e2a4689", + "metadata": {}, + "source": [ + "# Quickstart for Heston option pricing model using QuantLib\n", + "\n", + "Welcome! Let's get you started with the basic process of documenting models with ValidMind.\n", + "\n", + "The Heston option pricing model is a popular stochastic volatility model used to price options. Developed by Steven Heston in 1993, the model assumes that the asset's volatility follows a mean-reverting square-root process, allowing it to capture the empirical observation of volatility \"clustering\" in financial markets. This model is particularly useful for assets where volatility is not constant, making it a favored approach in quantitative finance for pricing complex derivatives.\n", + "\n", + "Here’s an overview of the Heston model as implemented in QuantLib, a powerful library for quantitative finance:\n", + "\n", + "\n", + "\n", + "### Model Assumptions and Characteristics\n", + "1. **Stochastic Volatility**: The volatility is modeled as a stochastic process, following a mean-reverting square-root process (Cox-Ingersoll-Ross process).\n", + "2. **Correlated Asset and Volatility Processes**: The asset price and volatility are assumed to be correlated, allowing the model to capture the \"smile\" effect observed in implied volatilities.\n", + "3. **Risk-Neutral Dynamics**: The Heston model is typically calibrated under a risk-neutral measure, which allows for direct application to pricing.\n", + "\n", + "\n", + "\n", + "### Heston Model Parameters\n", + "The model is governed by a set of key parameters:\n", + "- **S0**: Initial stock price\n", + "- **v0**: Initial variance of the asset price\n", + "- **kappa**: Speed of mean reversion of the variance\n", + "- **theta**: Long-term mean level of variance\n", + "- **sigma**: Volatility of volatility (vol of vol)\n", + "- **rho**: Correlation between the asset price and variance processes\n", + "\n", + "The dynamics of the asset price \\( S \\) and the variance \\( v \\) under the Heston model are given by:\n", + "\n", + "$$\n", + "dS_t = r S_t \\, dt + \\sqrt{v_t} S_t \\, dW^S_t\n", + "$$\n", + "\n", + "$$\n", + "dv_t = \\kappa (\\theta - v_t) \\, dt + \\sigma \\sqrt{v_t} \\, dW^v_t\n", + "$$\n", + "\n", + "where \\( $dW^S$ \\) and \\( $dW^v$ \\) are Wiener processes with correlation \\( $\\rho$ \\).\n", + "\n", + "\n", + "\n", + "### Advantages and Limitations\n", + "- **Advantages**:\n", + " - Ability to capture volatility smiles and skews.\n", + " - More realistic pricing for options on assets with stochastic volatility.\n", + "- **Limitations**:\n", + " - Calibration can be complex due to the number of parameters.\n", + " - Computationally intensive compared to simpler models like Black-Scholes.\n", + "\n", + "This setup provides a robust framework for pricing and analyzing options with stochastic volatility dynamics. QuantLib’s implementation makes it easy to experiment with different parameter configurations and observe their effects on pricing.\n", + "\n", + "You will learn how to initialize the ValidMind Library, develop a option pricing model, and then write custom tests that can be used for sensitivity and stress testing to quickly generate documentation about model." + ] + }, + { + "cell_type": "markdown", + "id": "69ec219a", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + " - [Model Assumptions and Characteristics](#toc1_1__) \n", + " - [Heston Model Parameters](#toc1_2__) \n", + " - [Advantages and Limitations](#toc1_3__) \n", + "- [About ValidMind](#toc2__) \n", + " - [Before you begin](#toc2_1__) \n", + " - [New to ValidMind?](#toc2_2__) \n", + " - [Key concepts](#toc2_3__) \n", + "- [Setting up](#toc3__) \n", + " - [Install the ValidMind Library](#toc3_1__) \n", + " - [Initialize the ValidMind Library](#toc3_2__) \n", + " - [Register sample model](#toc3_2_1__) \n", + " - [Apply documentation template](#toc3_2_2__) \n", + " - [Get your code snippet](#toc3_2_3__) \n", + " - [Initialize the Python environment](#toc3_3__) \n", + " - [Preview the documentation template](#toc3_4__) \n", + "- [Data Preparation](#toc4__) \n", + " - [Helper functions](#toc4_1_1__) \n", + " - [Market Data Quality and Availability](#toc4_2__) \n", + " - [Initialize the ValidMind datasets](#toc4_3__) \n", + " - [Data Quality](#toc4_4__) \n", + " - [Isolation Forest Outliers Test](#toc4_4_1__) \n", + " - [Model parameters](#toc4_4_2__) \n", + "- [Model development - Heston Option price](#toc5__) \n", + " - [Model Calibration](#toc5_1__) \n", + " - [Model Evaluation](#toc5_2__) \n", + " - [Benchmark Testing](#toc5_2_1__) \n", + " - [Sensitivity Testing](#toc5_2_2__) \n", + " - [Stress Testing](#toc5_2_3__) \n", + "- [Next steps](#toc6__) \n", + " - [Work with your model documentation](#toc6_1__) \n", + " - [Discover more learning resources](#toc6_2__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "b9fb5d17", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", + "\n", + "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", + "\n", + "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", + "\n", + "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", + "\n", + "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", + "\n", + " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", + " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", + " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", + "\n", + "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", + "\n", + "**Outputs**: Custom tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", + "\n", + "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", + "\n", + "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." + ] + }, + { + "cell_type": "markdown", + "id": "f2dccf35", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "id": "5a5ce085", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "To install the library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "409352bf", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "id": "65e870b2", + "metadata": {}, + "source": [ + "To install the QuantLib library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3a34debf", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q QuantLib" + ] + }, + { + "cell_type": "markdown", + "id": "fb30ae07", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "id": "c6f87017", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "id": "cbb2e2c9", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Capital Markets`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "id": "41c4edca", + "metadata": {}, + "source": [ + "
Can't select this template?\n", + "

\n", + "Your organization administrators may need to add it to your template library:\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "2012eb82", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0cd3f67e", + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")\n" + ] + }, + { + "cell_type": "markdown", + "id": "6d944cc9", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the Python environment\n", + "\n", + "Next, let's import the necessary libraries and set up your Python environment for data analysis:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f8cf2746", + "metadata": {}, + "outputs": [], + "source": [ + "%matplotlib inline\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from scipy.optimize import minimize\n", + "import yfinance as yf\n", + "import QuantLib as ql\n", + "from validmind.tests import run_test" + ] + }, + { + "cell_type": "markdown", + "id": "bc431ee0", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Preview the documentation template\n", + "\n", + "Let's verify that you have connected the ValidMind Library to the ValidMind Platform and that the appropriate *template* is selected for your model.\n", + "\n", + "You will upload documentation and test results unique to your model based on this template later on. For now, **take a look at the default structure that the template provides with [the `vm.preview_template()` function](https://docs.validmind.ai/validmind/validmind.html#preview_template)** from the ValidMind library and note the empty sections:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7e844028", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "id": "0c0ee8b9", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Data Preparation" + ] + }, + { + "cell_type": "markdown", + "id": "5a4d2c36", + "metadata": {}, + "source": [ + "### Market Data Sources\n", + "\n", + "\n", + "\n", + "#### Helper functions\n", + "Let's define helper function retrieve to option data from Yahoo Finance." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b96a500f", + "metadata": {}, + "outputs": [], + "source": [ + "def get_market_data(ticker, expiration_date_str):\n", + " \"\"\"\n", + " Fetch option market data from Yahoo Finance for the given ticker and expiration date.\n", + " Returns a list of tuples: (strike, maturity, option_price).\n", + " \"\"\"\n", + " # Create a Ticker object for the specified stock\n", + " stock = yf.Ticker(ticker)\n", + "\n", + " # Get all available expiration dates for options\n", + " option_dates = stock.options\n", + "\n", + " # Check if the requested expiration date is available\n", + " if expiration_date_str not in option_dates:\n", + " raise ValueError(f\"Expiration date {expiration_date_str} not available for {ticker}. Available dates: {option_dates}\")\n", + "\n", + " # Get the option chain for the specified expiration date\n", + " option_chain = stock.option_chain(expiration_date_str)\n", + "\n", + " # Get call options (or you can use puts as well based on your requirement)\n", + " calls = option_chain.calls\n", + "\n", + " # Convert expiration_date_str to QuantLib Date\n", + " expiry_date_parts = list(map(int, expiration_date_str.split('-'))) # Split YYYY-MM-DD\n", + " maturity_date = ql.Date(expiry_date_parts[2], expiry_date_parts[1], expiry_date_parts[0]) # Convert to QuantLib Date\n", + "\n", + " # Create a list to store strike prices, maturity dates, and option prices\n", + " market_data = []\n", + " for index, row in calls.iterrows():\n", + " strike = row['strike']\n", + " option_price = row['lastPrice'] # You can also use 'bid', 'ask', 'mid', etc.\n", + " market_data.append((strike, maturity_date, option_price))\n", + " df = pd.DataFrame(market_data, columns = ['strike', 'maturity_date', 'option_price'])\n", + " return df" + ] + }, + { + "cell_type": "markdown", + "id": "c7769b73", + "metadata": {}, + "source": [ + "Let's define helper function retrieve to stock data from Yahoo Finance. This helper function to calculate spot price, dividend yield, volatility and risk free rate using the underline stock data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dc44c448", + "metadata": {}, + "outputs": [], + "source": [ + "def get_option_parameters(ticker):\n", + " # Fetch historical data for the stock\n", + " stock_data = yf.Ticker(ticker)\n", + " \n", + " # Get the current spot price\n", + " spot_price = stock_data.history(period=\"1d\")['Close'].iloc[-1]\n", + " \n", + " # Get dividend yield\n", + " dividend_rate = stock_data.dividends.mean() / spot_price if not stock_data.dividends.empty else 0.0\n", + " \n", + " # Estimate volatility (standard deviation of log returns)\n", + " hist_data = stock_data.history(period=\"1y\")['Close']\n", + " log_returns = np.log(hist_data / hist_data.shift(1)).dropna()\n", + " volatility = np.std(log_returns) * np.sqrt(252) # Annualized volatility\n", + " \n", + " # Assume a risk-free rate from some known data (can be fetched from market data, here we use 0.001)\n", + " risk_free_rate = 0.001\n", + " \n", + " # Return the calculated parameters\n", + " return {\n", + " \"spot_price\": spot_price,\n", + " \"volatility\": volatility,\n", + " \"dividend_rate\": dividend_rate,\n", + " \"risk_free_rate\": risk_free_rate\n", + " }" + ] + }, + { + "cell_type": "markdown", + "id": "c7b739d3", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Market Data Quality and Availability\n", + "Next, let's specify ticker and expiration date to get market data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "50225fde", + "metadata": {}, + "outputs": [], + "source": [ + "ticker = \"MSFT\"\n", + "expiration_date = \"2024-12-13\" # Example expiration date in 'YYYY-MM-DD' form\n", + "\n", + "market_data = get_market_data(ticker=ticker, expiration_date_str=expiration_date)" + ] + }, + { + "cell_type": "markdown", + "id": "c539b95e", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind datasets\n", + "\n", + "Before you can run tests, you must first initialize a ValidMind dataset object using the [`init_dataset`](https://docs.validmind.ai/validmind/validmind.html#init_dataset) function from the ValidMind (`vm`) module." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "113f9c17", + "metadata": {}, + "outputs": [], + "source": [ + "vm_market_data = vm.init_dataset(\n", + " dataset=market_data,\n", + " input_id=\"market_data\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "185beb24", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Data Quality\n", + "Let's check quality of the data using outliers and missing data tests." + ] + }, + { + "cell_type": "markdown", + "id": "7f14464c", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Isolation Forest Outliers Test\n", + "Let's detects anomalies in the dataset using the Isolation Forest algorithm, visualized through scatter plots." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "56c919ec", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"validmind.data_validation.IsolationForestOutliers\",\n", + " inputs={\n", + " \"dataset\": vm_market_data,\n", + " },\n", + " title=\"Outliers detection using Isolation Forest\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "e4d0e5ca", + "metadata": {}, + "source": [ + "##### Missing Values Test\n", + "Let's evaluates dataset quality by ensuring the missing value ratio across all features does not exceed a set threshold." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e95c825f", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"validmind.data_validation.MissingValues\",\n", + " inputs={\n", + " \"dataset\": vm_market_data,\n", + " },\n", + " title=\"Missing Values detection\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "829403a3", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Model parameters\n", + "Let's calculate the model parameters using from stock data " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "25936449", + "metadata": {}, + "outputs": [], + "source": [ + "option_params = get_option_parameters(ticker=ticker)" + ] + }, + { + "cell_type": "markdown", + "id": "0a0948b6", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Model development - Heston Option price" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e15b8221", + "metadata": {}, + "outputs": [], + "source": [ + "class HestonModel:\n", + "\n", + " def __init__(self, ticker, expiration_date_str, calculation_date, spot_price, dividend_rate, risk_free_rate):\n", + " self.ticker = ticker\n", + " self.expiration_date_str = expiration_date_str,\n", + " self.calculation_date = calculation_date\n", + " self.spot_price = spot_price\n", + " self.dividend_rate = dividend_rate\n", + " self.risk_free_rate = risk_free_rate\n", + " \n", + " def predict_option_price(self, strike, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", + " # Set the evaluation date\n", + " ql.Settings.instance().evaluationDate = self.calculation_date\n", + "\n", + " # Construct the European Option\n", + " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", + " exercise = ql.EuropeanExercise(maturity_date)\n", + " european_option = ql.VanillaOption(payoff, exercise)\n", + "\n", + " # Yield term structures for risk-free rate and dividend\n", + " riskFreeTS = ql.YieldTermStructureHandle(ql.FlatForward(calculation_date, self.risk_free_rate, ql.Actual365Fixed()))\n", + " dividendTS = ql.YieldTermStructureHandle(ql.FlatForward(calculation_date, self.dividend_rate, ql.Actual365Fixed()))\n", + "\n", + " # Initial stock price\n", + " initialValue = ql.QuoteHandle(ql.SimpleQuote(spot_price))\n", + "\n", + " # Heston process parameters\n", + " heston_process = ql.HestonProcess(riskFreeTS, dividendTS, initialValue, v0, kappa, theta, sigma, rho)\n", + " hestonModel = ql.HestonModel(heston_process)\n", + "\n", + " # Use the Heston analytic engine\n", + " engine = ql.AnalyticHestonEngine(hestonModel)\n", + " european_option.setPricingEngine(engine)\n", + "\n", + " # Calculate the Heston model price\n", + " h_price = european_option.NPV()\n", + "\n", + " return h_price\n", + "\n", + " def predict_american_option_price(self, strike, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", + " # Set the evaluation date\n", + " ql.Settings.instance().evaluationDate = self.calculation_date\n", + "\n", + " # Construct the American Option\n", + " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", + " exercise = ql.AmericanExercise(self.calculation_date, maturity_date)\n", + " american_option = ql.VanillaOption(payoff, exercise)\n", + "\n", + " # Yield term structures for risk-free rate and dividend\n", + " riskFreeTS = ql.YieldTermStructureHandle(ql.FlatForward(self.calculation_date, self.risk_free_rate, ql.Actual365Fixed()))\n", + " dividendTS = ql.YieldTermStructureHandle(ql.FlatForward(self.calculation_date, self.dividend_rate, ql.Actual365Fixed()))\n", + "\n", + " # Initial stock price\n", + " initialValue = ql.QuoteHandle(ql.SimpleQuote(spot_price))\n", + "\n", + " # Heston process parameters\n", + " heston_process = ql.HestonProcess(riskFreeTS, dividendTS, initialValue, v0, kappa, theta, sigma, rho)\n", + " heston_model = ql.HestonModel(heston_process)\n", + "\n", + "\n", + " payoff = ql.PlainVanillaPayoff(ql.Option.Call, strike)\n", + " exercise = ql.AmericanExercise(self.calculation_date, maturity_date)\n", + " american_option = ql.VanillaOption(payoff, exercise)\n", + " heston_fd_engine = ql.FdHestonVanillaEngine(heston_model)\n", + " american_option.setPricingEngine(heston_fd_engine)\n", + " option_price = american_option.NPV()\n", + "\n", + " return option_price\n", + "\n", + " def objective_function(self, params, market_data, spot_price, dividend_rate, risk_free_rate):\n", + " v0, theta, kappa, sigma, rho = params\n", + "\n", + " # Sum of squared differences between market prices and model prices\n", + " error = 0.0\n", + " for i, row in market_data.iterrows():\n", + " model_price = self.predict_option_price(row['strike'], row['maturity_date'], spot_price, \n", + " v0, theta, kappa, sigma, rho)\n", + " error += (model_price - row['option_price']) ** 2\n", + " \n", + " return error\n", + "\n", + " def calibrate_model(self, ticker, expiration_date_str):\n", + " # Get the option market data dynamically from Yahoo Finance\n", + " market_data = get_market_data(ticker, expiration_date_str)\n", + "\n", + " # Initial guesses for Heston parameters\n", + " initial_params = [0.04, 0.04, 0.1, 0.1, -0.75]\n", + "\n", + " # Bounds for the parameters to ensure realistic values\n", + " bounds = [(0.0001, 1.0), # v0\n", + " (0.0001, 1.0), # theta\n", + " (0.001, 2.0), # kappa\n", + " (0.001, 1.0), # sigma\n", + " (-0.75, 0.0)] # rho\n", + "\n", + " # Optimize the parameters to minimize the error between model and market prices\n", + " result = minimize(self.objective_function, initial_params, args=(market_data, self.spot_price, self.dividend_rate, self.risk_free_rate),\n", + " bounds=bounds, method='L-BFGS-B')\n", + "\n", + " # Optimized Heston parameters\n", + " v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt = result.x\n", + "\n", + " return v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt\n" + ] + }, + { + "cell_type": "markdown", + "id": "a941aa32", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Model Calibration\n", + "* The calibration process aims to optimize the Heston model parameters (v0, theta, kappa, sigma, rho) by minimizing the difference between model-predicted option prices and observed market prices.\n", + "* In this implementation, the model is calibrated to current market data, specifically using option prices from the selected ticker and expiration date.\n", + "\n", + "Let's specify `calculation_date` and `strike_price` as input parameters for the model to verify its functionality and confirm it operates as expected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1d61dfca", + "metadata": {}, + "outputs": [], + "source": [ + "calculation_date = ql.Date(26, 11, 2024)\n", + "# Convert expiration date string to QuantLib.Date\n", + "expiry_date_parts = list(map(int, expiration_date.split('-')))\n", + "maturity_date = ql.Date(expiry_date_parts[2], expiry_date_parts[1], expiry_date_parts[0])\n", + "strike_price = 460.0\n", + "\n", + "hm = HestonModel(\n", + " ticker=ticker,\n", + " expiration_date_str= expiration_date,\n", + " calculation_date= calculation_date,\n", + " spot_price= option_params['spot_price'],\n", + " dividend_rate = option_params['dividend_rate'],\n", + " risk_free_rate = option_params['risk_free_rate']\n", + ")\n", + "\n", + "# Let's calibrate model\n", + "v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt = hm.calibrate_model(ticker, expiration_date)\n", + "print(f\"Optimized Heston parameters: v0={v0_opt}, theta={theta_opt}, kappa={kappa_opt}, sigma={sigma_opt}, rho={rho_opt}\")\n", + "\n", + "\n", + "# option price\n", + "h_price = hm.predict_option_price(strike_price, maturity_date, option_params['spot_price'], v0_opt, theta_opt, kappa_opt, sigma_opt, rho_opt)\n", + "print(\"The Heston model price for the option is:\", h_price)" + ] + }, + { + "cell_type": "markdown", + "id": "75313272", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Model Evaluation" + ] + }, + { + "cell_type": "markdown", + "id": "2e6471ef", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Benchmark Testing\n", + "The benchmark testing framework provides a robust way to validate the Heston model implementation and understand the relationships between European and American option prices under stochastic volatility conditions.\n", + "Let's compares European and American option prices using the Heston model." + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "810cf887", + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_custom_tests.BenchmarkTest\")\n", + "def benchmark_test(hm_model, strikes, maturity_date, spot_price, v0=None, theta=None, kappa=None, sigma=None, rho=None):\n", + " \"\"\"\n", + " Compares European and American option prices using the Heston model.\n", + "\n", + " This test evaluates the price differences between European and American options\n", + " across multiple strike prices while keeping other parameters constant. The comparison\n", + " helps understand the early exercise premium of American options over their European\n", + " counterparts under stochastic volatility conditions.\n", + "\n", + " Args:\n", + " hm_model: HestonModel instance for option pricing calculations\n", + " strikes (list[float]): List of strike prices to test\n", + " maturity_date (ql.Date): Option expiration date in QuantLib format\n", + " spot_price (float): Current price of the underlying asset\n", + " v0 (float, optional): Initial variance. Defaults to None.\n", + " theta (float, optional): Long-term variance. Defaults to None.\n", + " kappa (float, optional): Mean reversion rate. Defaults to None.\n", + " sigma (float, optional): Volatility of variance. Defaults to None.\n", + " rho (float, optional): Correlation between asset and variance. Defaults to None.\n", + "\n", + " Returns:\n", + " dict: Contains a DataFrame with the following columns:\n", + " - Strike: Strike prices tested\n", + " - Maturity date: Expiration date for all options\n", + " - Spot price: Current underlying price\n", + " - european model price: Prices for European options\n", + " - american model price: Prices for American options\n", + "\"\"\"\n", + " american_derived_prices = []\n", + " european_derived_prices = []\n", + " for K in strikes:\n", + " european_derived_prices.append(hm_model.predict_option_price(K, maturity_date, spot_price, v0, theta, kappa, sigma, rho))\n", + " american_derived_prices.append(hm_model.predict_american_option_price(K, maturity_date, spot_price, v0, theta, kappa, sigma, rho))\n", + "\n", + " data = {\n", + " \"Strike\": strikes,\n", + " \"Maturity date\": [maturity_date] * len(strikes),\n", + " \"Spot price\": [spot_price] * len(strikes),\n", + " \"european model price\": european_derived_prices,\n", + " \"american model price\": american_derived_prices,\n", + "\n", + " }\n", + " df1 = pd.DataFrame(data)\n", + " return {\"strikes variation benchmarking\": df1}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3fdd6705", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.BenchmarkTest\",\n", + " params={\n", + " \"hm_model\": hm,\n", + " \"strikes\": [400, 425, 460, 495, 520],\n", + " \"maturity_date\": maturity_date,\n", + " \"spot_price\": option_params['spot_price'],\n", + " \"v0\":v0_opt,\n", + " \"theta\": theta_opt,\n", + " \"kappa\":kappa_opt ,\n", + " \"sigma\": sigma_opt,\n", + " \"rho\":rho_opt\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "e359b503", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Sensitivity Testing\n", + "The sensitivity testing framework provides a systematic approach to understanding how the Heston model responds to parameter changes, which is crucial for both model validation and practical application in trading and risk management." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "51922313", + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_test_provider.Sensitivity\")\n", + "def SensitivityTest(\n", + " model,\n", + " strike_price,\n", + " maturity_date,\n", + " spot_price,\n", + " v0_opt,\n", + " theta_opt,\n", + " kappa_opt,\n", + " sigma_opt,\n", + " rho_opt,\n", + "):\n", + " \"\"\"\n", + " Evaluates the sensitivity of American option prices to changes in model parameters.\n", + "\n", + " This test calculates option prices using the Heston model with optimized parameters.\n", + " It's designed to analyze how changes in various model inputs affect the option price,\n", + " which is crucial for understanding model behavior and risk management.\n", + "\n", + " Args:\n", + " model (HestonModel): Initialized Heston model instance wrapped in ValidMind model object\n", + " strike_price (float): Strike price of the option\n", + " maturity_date (ql.Date): Expiration date of the option in QuantLib format\n", + " spot_price (float): Current price of the underlying asset\n", + " v0_opt (float): Optimized initial variance parameter\n", + " theta_opt (float): Optimized long-term variance parameter\n", + " kappa_opt (float): Optimized mean reversion rate parameter\n", + " sigma_opt (float): Optimized volatility of variance parameter\n", + " rho_opt (float): Optimized correlation parameter between asset price and variance\n", + " \"\"\"\n", + " price = model.model.predict_american_option_price(\n", + " strike_price,\n", + " maturity_date,\n", + " spot_price,\n", + " v0_opt,\n", + " theta_opt,\n", + " kappa_opt,\n", + " sigma_opt,\n", + " rho_opt,\n", + " )\n", + "\n", + " return price\n" + ] + }, + { + "cell_type": "markdown", + "id": "408a05ef", + "metadata": {}, + "source": [ + "##### Common plot function" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "104ca6dd", + "metadata": {}, + "outputs": [], + "source": [ + "def plot_results(df, params: dict = None):\n", + " fig2 = plt.figure(figsize=(10, 6))\n", + " plt.plot(df[params[\"x\"]], df[params[\"y\"]], label=params[\"label\"])\n", + " plt.xlabel(params[\"xlabel\"])\n", + " plt.ylabel(params[\"ylabel\"])\n", + " \n", + " plt.title(params[\"title\"])\n", + " plt.legend()\n", + " plt.grid(True)\n", + " plt.show() # close the plot to avoid displaying it" + ] + }, + { + "cell_type": "markdown", + "id": "ca72b9e5", + "metadata": {}, + "source": [ + "Let's create ValidMind model object" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ae7093fa", + "metadata": {}, + "outputs": [], + "source": [ + "hm_model = vm.init_model(model=hm, input_id=\"HestonModel\")" + ] + }, + { + "cell_type": "markdown", + "id": "b2141640", + "metadata": {}, + "source": [ + "##### Strike sensitivity\n", + "Let's analyzes how option prices change as the strike price varies. We create a range of strike prices around the current strike (460) and observe the impact on option prices while keeping all other parameters constant." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ea7f1cbe", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_test_provider.Sensitivity:ToStrike\",\n", + " inputs = {\n", + " \"model\": hm_model\n", + " },\n", + " param_grid={\n", + " \"strike_price\": list(np.linspace(460-50, 460+50, 10)),\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": [theta_opt],\n", + " \"kappa_opt\": [kappa_opt],\n", + " \"sigma_opt\": [sigma_opt],\n", + " \"rho_opt\":[rho_opt]\n", + " },\n", + ")\n", + "result.log()\n", + "# Visualize how option prices change with different strike prices\n", + "plot_results(\n", + " pd.DataFrame(result.tables[0].data),\n", + " params={\n", + " \"x\": \"strike_price\",\n", + " \"y\":\"Value\",\n", + " \"label\":\"Strike price\",\n", + " \"xlabel\":\"Strike price\",\n", + " \"ylabel\":\"option price\",\n", + " \"title\":\"Heston option - Strike price Sensitivity\",\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "be143012", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Stress Testing\n", + "This stress testing framework provides a comprehensive view of how the Heston model behaves under different market conditions and helps identify potential risks in option pricing." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f2f01a40", + "metadata": {}, + "outputs": [], + "source": [ + "@vm.test(\"my_custom_tests.Stressing\")\n", + "def StressTest(\n", + " model,\n", + " strike_price,\n", + " maturity_date,\n", + " spot_price,\n", + " v0_opt,\n", + " theta_opt,\n", + " kappa_opt,\n", + " sigma_opt,\n", + " rho_opt,\n", + "):\n", + " \"\"\"\n", + " Performs stress testing on Heston model parameters to evaluate option price sensitivity.\n", + "\n", + " This test evaluates how the American option price responds to stressed market conditions\n", + " by varying key model parameters. It's designed to:\n", + " 1. Identify potential model vulnerabilities\n", + " 2. Understand price behavior under extreme scenarios\n", + " 3. Support risk management decisions\n", + " 4. Validate model stability across parameter ranges\n", + "\n", + " Args:\n", + " model (HestonModel): Initialized Heston model instance wrapped in ValidMind model object\n", + " strike_price (float): Option strike price\n", + " maturity_date (ql.Date): Option expiration date in QuantLib format\n", + " spot_price (float): Current price of the underlying asset\n", + " v0_opt (float): Initial variance parameter under stress testing\n", + " theta_opt (float): Long-term variance parameter under stress testing\n", + " kappa_opt (float): Mean reversion rate parameter under stress testing\n", + " sigma_opt (float): Volatility of variance parameter under stress testing\n", + " rho_opt (float): Correlation parameter under stress testing\n", + " \"\"\"\n", + " price = model.model.predict_american_option_price(\n", + " strike_price,\n", + " maturity_date,\n", + " spot_price,\n", + " v0_opt,\n", + " theta_opt,\n", + " kappa_opt,\n", + " sigma_opt,\n", + " rho_opt,\n", + " )\n", + "\n", + " return price\n" + ] + }, + { + "cell_type": "markdown", + "id": "31fcbe9c", + "metadata": {}, + "source": [ + "##### Rho (correlation) and Theta (long term vol) stress test\n", + "Next, let's evaluates the sensitivity of a model's output to changes in the correlation parameter (rho) and the long-term variance parameter (theta) within a stochastic volatility framework." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "6119b5d9", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.Stressing:TheRhoAndThetaParameters\",\n", + " inputs = {\n", + " \"model\": hm_model,\n", + " },\n", + " param_grid={\n", + " \"strike_price\": [460],\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": list(np.linspace(0.1, theta_opt+0.4, 5)),\n", + " \"kappa_opt\": [kappa_opt],\n", + " \"sigma_opt\": [sigma_opt],\n", + " \"rho_opt\":list(np.linspace(rho_opt-0.2, rho_opt+0.2, 5))\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "be39cb3a", + "metadata": {}, + "source": [ + "##### Sigma stress test\n", + "Let's evaluates the sensitivity of a model's output to changes in the volatility parameter, sigma. This test is crucial for understanding how variations in market volatility impact the model's valuation of financial instruments, particularly options." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0dc189b7", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.Stressing:TheSigmaParameter\",\n", + " inputs = {\n", + " \"model\": hm_model,\n", + " },\n", + " param_grid={\n", + " \"strike_price\": [460],\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": [theta_opt],\n", + " \"kappa_opt\": [kappa_opt],\n", + " \"sigma_opt\": list(np.linspace(0.1, sigma_opt+0.6, 5)),\n", + " \"rho_opt\": [rho_opt]\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "173a5294", + "metadata": {}, + "source": [ + "##### Stress kappa\n", + "Let's evaluates the sensitivity of a model's output to changes in the kappa parameter, which is a mean reversion rate in stochastic volatility models." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "dae9714f", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.Stressing:TheKappaParameter\",\n", + " inputs = {\n", + " \"model\": hm_model,\n", + " },\n", + " param_grid={\n", + " \"strike_price\": [460],\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": [theta_opt],\n", + " \"kappa_opt\": list(np.linspace(kappa_opt, kappa_opt+0.2, 5)),\n", + " \"sigma_opt\": [sigma_opt],\n", + " \"rho_opt\": [rho_opt]\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "b4d1d968", + "metadata": {}, + "source": [ + "##### Stress theta\n", + "Let's evaluates the sensitivity of a model's output to changes in the parameter theta, which represents the long-term variance in a stochastic volatility model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e68df3db", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.Stressing:TheThetaParameter\",\n", + " inputs = {\n", + " \"model\": hm_model,\n", + " },\n", + " param_grid={\n", + " \"strike_price\": [460],\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": list(np.linspace(0.1, theta_opt+0.9, 5)),\n", + " \"kappa_opt\": [kappa_opt],\n", + " \"sigma_opt\": [sigma_opt],\n", + " \"rho_opt\": [rho_opt]\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "32e70456", + "metadata": {}, + "source": [ + "##### Stress rho\n", + "Let's evaluates the sensitivity of a model's output to changes in the correlation parameter, rho, within a stochastic volatility (SV) model framework. This test is crucial for understanding how variations in rho, which represents the correlation between the asset price and its volatility, impact the model's valuation output." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b5ca3fc2", + "metadata": {}, + "outputs": [], + "source": [ + "result = run_test(\n", + " \"my_custom_tests.Stressing:TheRhoParameter\",\n", + " inputs = {\n", + " \"model\": hm_model,\n", + " },\n", + " param_grid={\n", + " \"strike_price\": [460],\n", + " \"maturity_date\": [maturity_date],\n", + " \"spot_price\": [option_params[\"spot_price\"]],\n", + " \"v0_opt\": [v0_opt],\n", + " \"theta_opt\": [theta_opt],\n", + " \"kappa_opt\": [kappa_opt],\n", + " \"sigma_opt\": [sigma_opt],\n", + " \"rho_opt\": list(np.linspace(rho_opt-0.2, rho_opt+0.2, 5))\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "id": "892c5347", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Next steps\n", + "\n", + "You can look at the results of this test suite right in the notebook where you ran the code, as you would expect. But there is a better way — use the ValidMind Platform to work with your model documentation.\n", + "\n", + "\n", + "\n", + "### Work with your model documentation\n", + "\n", + "1. From the **Model Inventory** in the ValidMind Platform, go to the model you registered earlier. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/working-with-model-inventory.html))\n", + "\n", + "2. Click and expand the **Model Development** section.\n", + "\n", + "What you see is the full draft of your model documentation in a more easily consumable version. From here, you can make qualitative edits to model documentation, view guidelines, collaborate with validators, and submit your model documentation for approval when it's ready. [Learn more ...](https://docs.validmind.ai/guide/model-documentation/working-with-model-documentation.html)\n", + "\n", + "\n", + "\n", + "### Discover more learning resources\n", + "\n", + "We offer many interactive notebooks to help you document models:\n", + "\n", + "- [Run tests & test suites](https://docs.validmind.ai/developer/model-testing/testing-overview.html)\n", + "- [Code samples](https://docs.validmind.ai/developer/samples-jupyter-notebooks.html)\n", + "\n", + "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-de5d1e182b09403abddabc2850f2dd05", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "validmind-1QuffXMV-py3.10", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.14" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/site/notebooks/code_samples/code_explainer/customer_churn_full_suite.py b/site/notebooks/use_cases/code_explainer/customer_churn_full_suite.py similarity index 98% rename from site/notebooks/code_samples/code_explainer/customer_churn_full_suite.py rename to site/notebooks/use_cases/code_explainer/customer_churn_full_suite.py index 6bd9c2a8ab..3f7c828092 100644 --- a/site/notebooks/code_samples/code_explainer/customer_churn_full_suite.py +++ b/site/notebooks/use_cases/code_explainer/customer_churn_full_suite.py @@ -7,7 +7,7 @@ Welcome! Let's get you started with the basic process of documenting models with ValidMind. -You will learn how to initialize the ValidMind Library, load a sample dataset to train a simple classification model, +You will learn how to initialize the ValidMind Library, load a sample dataset to train a simple classification model, and then run a ValidMind test suite to quickly generate documentation about the data and model. This script uses the Bank Customer Churn Prediction sample dataset from Kaggle to train the classification model. @@ -162,8 +162,8 @@ def load_model(self, version: str) -> Any: ) vm_test_ds = vm.init_dataset( - dataset=test_df, - input_id="test_dataset", + dataset=test_df, + input_id="test_dataset", target_column=customer_churn.target_column ) diff --git a/site/notebooks/use_cases/code_explainer/model_source_code_documentation_template.yaml b/site/notebooks/use_cases/code_explainer/model_source_code_documentation_template.yaml new file mode 100644 index 0000000000..7065b35a7d --- /dev/null +++ b/site/notebooks/use_cases/code_explainer/model_source_code_documentation_template.yaml @@ -0,0 +1,142 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial + +- id: code_overview + title: Codebase Overview + guidelines: + - Describe the overall structure of the source code repository. + - Identify main modules, folders, and scripts. + - Highlight entry points for training, inference, and evaluation. + - State the main programming languages and frameworks used. + contents: + - content_type: text + content_id: code_structure_summary +- id: model_overview + title: Model Overview + guidelines: + - Describe the overall structure of the source code repository. + - Identify main modules, folders, and scripts. + - Highlight entry points for training, inference, and evaluation. + - State the main programming languages and frameworks used. + contents: + - content_type: text + content_id: model_overview +- id: environment_setup + title: Environment and Dependencies + guidelines: + - List Python packages and system dependencies (OS, compilers, etc.). + - Reference environment files (requirements.txt, environment.yml, + Dockerfile). + - Include setup instructions using Conda, virtualenv, or containers. + contents: + - content_type: text + content_id: setup_instructions +- id: data_interface + title: Data Ingestion and Preprocessing + guidelines: + - Specify data input formats and sources. + - Document ingestion, validation, and transformation logic. + - Explain how raw data is preprocessed and features are generated. + contents: + - content_type: text + content_id: data_handling_notes +- id: model_implementation + title: Model Implementation Details + guidelines: + - Describe the core model code structure (classes, functions). + - Link code to theoretical models or equations when applicable. + - Note custom components like loss functions or feature selectors. + contents: + - content_type: text + content_id: model_code_description +- id: training_pipeline + title: Model Training Pipeline + guidelines: + - Explain the training process, optimization strategy, and hyperparameters. + - Describe logging, checkpointing, and early stopping mechanisms. + - Include references to training config files or tuning logic. + contents: + - content_type: text + content_id: training_logic_details +- id: evaluation_pipeline + title: Evaluation and Validation Code + guidelines: + - Describe how validation is implemented and metrics are calculated. + - Include plots and diagnostic tools (e.g., ROC, SHAP, confusion matrix). + - State how outputs are logged and persisted. + contents: + - content_type: text + content_id: evaluation_logic_notes +- id: inference_pipeline + title: Inference and Scoring Logic + guidelines: + - Detail how the trained model is loaded and used for predictions. + - Explain I/O formats and APIs for serving or batch scoring. + - Include any preprocessing/postprocessing logic required. + contents: + - content_type: text + content_id: inference_mechanism +- id: configuration_management + title: Configuration and Parameters + guidelines: + - Describe configuration management (files, CLI args, env vars). + - Highlight default parameters and override mechanisms. + - Reference versioning practices for config files. + contents: + - content_type: text + content_id: config_control_notes +- id: testing_and_validation + title: Unit and Integration Testing + guidelines: + - List unit and integration tests and what they cover. + - Mention testing frameworks and coverage tools used. + - Explain testing strategy for production-readiness. + contents: + - content_type: text + content_id: test_strategy_overview +- id: logging + title: Logging and Monitoring Hooks + guidelines: + - Describe logging configuration and structure. + - Highlight real-time monitoring or observability integrations. + - List key events, metrics, or alerts tracked. + contents: + - content_type: text + content_id: logging_notes +- id: version_control + title: Code and Model Versioning + guidelines: + - Describe Git usage, branching, tagging, and commit standards. + - Include model artifact versioning practices (e.g., DVC, MLflow). + - Reference any automation in CI/CD. + contents: + - content_type: text + content_id: version_tracking_description +- id: security_and_compliance + title: Security and Access Control + guidelines: + - Document access controls for source code and data. + - Include any encryption, PII handling, or compliance measures. + - Mention secure deployment practices. + contents: + - content_type: text + content_id: security_policies_notes +- id: execution_examples + title: Example Runs and Scripts + guidelines: + - Provide working script examples (e.g., `train.py`, `predict.py`). + - Include CLI usage instructions or sample notebooks. + - Link to demo datasets or test scenarios. + contents: + - content_type: text + content_id: runnable_examples +- id: known_issues_and_todos + title: Known Issues and Future Improvements + guidelines: + - List current limitations or technical debt. + - Outline proposed enhancements or refactors. + - Reference relevant tickets, GitHub issues, or roadmap items. + contents: + - content_type: text + content_id: issues_and_improvements_log diff --git a/site/notebooks/use_cases/code_explainer/quickstart_code_explainer_demo.ipynb b/site/notebooks/use_cases/code_explainer/quickstart_code_explainer_demo.ipynb new file mode 100644 index 0000000000..99ee1abf66 --- /dev/null +++ b/site/notebooks/use_cases/code_explainer/quickstart_code_explainer_demo.ipynb @@ -0,0 +1,874 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Quickstart for model code documentation\n", + "\n", + "Welcome! This notebook demonstrates how to use the ValidMind code explainer to automatically generate comprehensive documentation for your codebase. The code explainer analyzes your source code and provides detailed explanations across various aspects of your implementation.\n", + "\n", + "\n", + "\n", + "## About Code Explainer\n", + "The ValidMind code explainer is a powerful tool that automatically analyzes your source code and generates comprehensive documentation. It helps you:\n", + "\n", + "- Understand the structure and organization of your codebase\n", + "- Document dependencies and environment setup\n", + "- Explain data processing and model implementation details\n", + "- Document training, evaluation, and inference pipelines\n", + "- Track configuration, testing, and security measures\n", + "\n", + "This tool is particularly useful for:\n", + "- Onboarding new team members\n", + "- Maintaining up-to-date documentation\n", + "- Ensuring code quality and best practices\n", + "- Facilitating code reviews and audits" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About Code Explainer](#toc1__) \n", + "- [About ValidMind](#toc2__) \n", + " - [Before you begin](#toc2_1__) \n", + " - [New to ValidMind?](#toc2_2__) \n", + " - [Key concepts](#toc2_3__) \n", + "- [Setting up](#toc3__) \n", + " - [Install the ValidMind Library](#toc3_1__) \n", + " - [Initialize the ValidMind Library](#toc3_2__) \n", + " - [Register sample model](#toc3_2_1__) \n", + " - [Apply documentation template](#toc3_2_2__) \n", + " - [Get your code snippet](#toc3_2_3__) \n", + " - [Preview the documentation template](#toc3_3__) \n", + "- [Common function](#toc4__) \n", + "- [Default Behavior](#toc5__) \n", + "- [Codebase Overview](#toc6__) \n", + "- [Environment and Dependencies ('environment_setup')](#toc7__) \n", + "- [Data Ingestion and Preprocessing](#toc8__) \n", + "- [Model Implementation Details](#toc9__) \n", + "- [Model Training Pipeline](#toc10__) \n", + "- [Evaluation and Validation Code](#toc11__) \n", + "- [Inference and Scoring Logic](#toc12__) \n", + "- [Configuration and Parameters](#toc13__) \n", + "- [Unit and Integration Testing](#toc14__) \n", + "- [Logging and Monitoring Hooks](#toc15__) \n", + "- [Code and Model Versioning](#toc16__) \n", + "- [Security and Access Control](#toc17__) \n", + "- [Example Runs and Scripts](#toc18__) \n", + "- [Known Issues and Future Improvements](#toc19__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "**Model documentation**: A structured and detailed record pertaining to a model, encompassing key components such as its underlying assumptions, methodologies, data sources, inputs, performance metrics, evaluations, limitations, and intended uses. It serves to ensure transparency, adherence to regulatory requirements, and a clear understanding of potential risks associated with the model’s application.\n", + "\n", + "**Documentation template**: Functions as a test suite and lays out the structure of model documentation, segmented into various sections and sub-sections. Documentation templates define the structure of your model documentation, specifying the tests that should be run, and how the results should be displayed.\n", + "\n", + "**Tests**: A function contained in the ValidMind Library, designed to run a specific quantitative test on the dataset or model. Tests are the building blocks of ValidMind, used to evaluate and document models and datasets, and can be run individually or as part of a suite defined by your model documentation template.\n", + "\n", + "**Custom tests**: Custom tests are functions that you define to evaluate your model or dataset. These functions can be registered via the ValidMind Library to be used with the ValidMind Platform.\n", + "\n", + "**Inputs**: Objects to be evaluated and documented in the ValidMind Library. They can be any of the following:\n", + "\n", + " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", + " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", + " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", + "\n", + "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", + "\n", + "**Outputs**: Custom tests can return elements like tables or plots. Tables may be a list of dictionaries (each representing a row) or a pandas DataFrame. Plots may be matplotlib or plotly figures.\n", + "\n", + "**Test suites**: Collections of tests designed to run together to automate and generate model documentation end-to-end for specific use-cases.\n", + "\n", + "Example: the [`classifier_full_suite`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html#ClassifierFullSuite) test suite runs tests from the [`tabular_dataset`](https://docs.validmind.ai/validmind/validmind/test_suites/tabular_datasets.html) and [`classifier`](https://docs.validmind.ai/validmind/validmind/test_suites/classifier.html) test suites to fully document the data and model sections for binary classification model use-cases." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "To install the library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Model Source Code Documentation`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Can't select this template?\n", + "

\n", + "Your organization administrators may need to add it to your template library:\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Preview the documentation template\n", + "\n", + "Let's verify that you have connected the ValidMind Library to the ValidMind Platform and that the appropriate *template* is selected for your model.\n", + "\n", + "You will upload documentation and test results unique to your model based on this template later on. For now, **take a look at the default structure that the template provides with [the `vm.preview_template()` function](https://docs.validmind.ai/validmind/validmind.html#preview_template)** from the ValidMind library and note the empty sections:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Common function\n", + "The code above defines two key functions:\n", + "1. A function to read source code from 'customer_churn_full_suite.py' file\n", + "2. An 'explain_code' function that uses ValidMind's experimental agents to analyze and explain code." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "source_code=\"\"\n", + "with open(\"customer_churn_full_suite.py\", \"r\") as f:\n", + " source_code = f.read()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The `vm.experimental.agents.run_task` function is used to execute AI agent tasks.\n", + "\n", + "It requires:\n", + "- task: The type of task to run (e.g. `code_explainer`)\n", + "- input: A dictionary containing task-specific parameters\n", + " - For `code_explainer`, this includes:\n", + " - **source_code** (str): The code to be analyzed\n", + " - **user_instructions** (str): Instructions for how to analyze the code" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "def explain_code(content_id: str, user_instructions: str):\n", + " \"\"\"Run code explanation task and log the results.\n", + " By default, the code explainer includes sections for:\n", + " - Main Purpose and Overall Functionality\n", + " - Breakdown of Key Functions or Components\n", + " - Potential Risks or Failure Points \n", + " - Assumptions or Limitations\n", + " If you want default sections, specify user_instructions as an empty string.\n", + " \n", + " Args:\n", + " user_instructions (str): Instructions for how to analyze the code\n", + " content_id (str): ID to use when logging the results\n", + " \n", + " Returns:\n", + " The result object from running the code explanation task\n", + " \"\"\"\n", + " result = vm.experimental.agents.run_task(\n", + " task=\"code_explainer\",\n", + " input={\n", + " \"source_code\": source_code,\n", + " \"user_instructions\": user_instructions\n", + " }\n", + " )\n", + " result.log(content_id=content_id)\n", + " return result" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Default Behavior" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "By default, the code explainer includes sections for:\n", + "- Main Purpose and Overall Functionality\n", + "- Breakdown of Key Functions or Components\n", + "- Potential Risks or Failure Points \n", + "- Assumptions or Limitations\n", + "\n", + "If you want default sections, specify `user_instructions` as an empty string. For example:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.experimental.agents.run_task(\n", + " task=\"code_explainer\",\n", + " input={\n", + " \"source_code\": source_code,\n", + " \"user_instructions\": \"\"\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Codebase Overview\n", + "\n", + "Let's analyze your codebase structure to understand the main modules, components, entry points and their relationships. We'll also examine the technology stack and frameworks that are being utilized in the implementation." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe the overall structure of the source code repository.\n", + " - Identify main modules, folders, and scripts.\n", + " - Highlight entry points for training, inference, and evaluation.\n", + " - State the main programming languages and frameworks used.\n", + " \"\"\",\n", + " content_id=\"code_structure_summary\"\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\",\n", + " content_id=\"code_structure_summary\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Environment and Dependencies ('environment_setup')\n", + "Let's document the technical requirements and setup needed to run your code, including Python packages, system dependencies, and environment configuration files. Understanding these requirements is essential for proper development environment setup and consistent deployments across different environments." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - List Python packages and system dependencies (OS, compilers, etc.).\n", + " - Reference environment files (requirements.txt, environment.yml, Dockerfile).\n", + " - Include setup instructions using Conda, virtualenv, or containers.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"setup_instructions\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Data Ingestion and Preprocessing\n", + "Let's document how your code handles data, including data sources, validation procedures, and preprocessing steps. We'll examine the data pipeline architecture, covering everything from initial data loading through feature engineering and quality checks." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Specify data input formats and sources.\n", + " - Document ingestion, validation, and transformation logic.\n", + " - Explain how raw data is preprocessed and features are generated.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections. \"\"\",\n", + " content_id=\"data_handling_notes\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + " \n", + "\n", + "\n", + "\n", + "## Model Implementation Details\n", + "Let's document the core implementation details of your model, including its architecture, components, and key algorithms. Understanding the technical implementation is crucial for maintenance, debugging, and future improvements to the codebase. We'll examine how theoretical concepts are translated into working code." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe the core model code structure (classes, functions).\n", + " - Link code to theoretical models or equations when applicable.\n", + " - Note custom components like loss functions or feature selectors.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"model_code_description\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Model Training Pipeline\n", + "\n", + "Let's document the training pipeline implementation, including how models are trained, optimized and evaluated. We'll examine the training process workflow, hyperparameter tuning approach, and model checkpointing mechanisms. This section provides insights into how the model learns from data and achieves optimal performance." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Explain the training process, optimization strategy, and hyperparameters.\n", + " - Describe logging, checkpointing, and early stopping mechanisms.\n", + " - Include references to training config files or tuning logic.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"training_logic_details\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Evaluation and Validation Code\n", + "Let's examine how the model's validation and evaluation code is implemented, including the metrics calculation and validation processes. We'll explore the diagnostic tools and visualization methods used to assess model performance. This section will also cover how validation results are logged and stored for future reference." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe how validation is implemented and metrics are calculated.\n", + " - Include plots and diagnostic tools (e.g., ROC, SHAP, confusion matrix).\n", + " - State how outputs are logged and persisted.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"evaluation_logic_notes\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Inference and Scoring Logic\n", + "Let's examine how the model performs inference and scoring on new data. This section will cover the implementation details of loading trained models, making predictions, and any required pre/post-processing steps. We'll also look at the APIs and interfaces available for both real-time serving and batch scoring scenarios." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Detail how the trained model is loaded and used for predictions.\n", + " - Explain I/O formats and APIs for serving or batch scoring.\n", + " - Include any preprocessing/postprocessing logic required.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"inference_mechanism\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Configuration and Parameters\n", + "Let's explore how configuration and parameters are managed in the codebase. We'll examine the configuration files, command-line arguments, environment variables, and other mechanisms used to control model behavior. This section will also cover parameter versioning and how different configurations are tracked across model iterations." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe configuration management (files, CLI args, env vars).\n", + " - Highlight default parameters and override mechanisms.\n", + " - Reference versioning practices for config files.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"config_control_notes\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Unit and Integration Testing\n", + "Let's examine the testing strategy and implementation in the codebase. We'll analyze the unit tests, integration tests, and testing frameworks used to ensure code quality and reliability. This section will also cover test coverage metrics and continuous integration practices." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - List unit and integration tests and what they cover.\n", + " - Mention testing frameworks and coverage tools used.\n", + " - Explain testing strategy for production-readiness.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"test_strategy_overview\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Logging and Monitoring Hooks\n", + "Let's analyze how logging and monitoring are implemented in the codebase. We'll examine the logging configuration, monitoring hooks, and key metrics being tracked. This section will also cover any real-time observability integrations and alerting mechanisms in place." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe logging configuration and structure.\n", + " - Highlight real-time monitoring or observability integrations.\n", + " - List key events, metrics, or alerts tracked.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"logging_monitoring_notes\"\n", + ")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Code and Model Versioning\n", + "Let's examine how code and model versioning is managed in the codebase. This section will cover version control practices, including Git workflows and model artifact versioning tools like DVC or MLflow. We'll also look at how versioning integrates with the CI/CD pipeline." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Describe Git usage, branching, tagging, and commit standards.\n", + " - Include model artifact versioning practices (e.g., DVC, MLflow).\n", + " - Reference any automation in CI/CD.\n", + " Please remove the following sections: \n", + " - Potential Risks or Failure Points\n", + " - Assumptions or Limitations\n", + " - Breakdown of Key Functions or Components\n", + " Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"version_tracking_description\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Security and Access Control\n", + "Let's analyze the security and access control measures implemented in the codebase. We'll examine how sensitive data and code are protected through access controls, encryption, and compliance measures. Additionally, we'll review secure deployment practices and any specific handling of PII data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Document access controls for source code and data.\n", + " - Include any encryption, PII handling, or compliance measures.\n", + " - Mention secure deployment practices.\n", + " Please remove the following sections: \n", + " - Potential Risks or Failure Points\n", + " - Assumptions or Limitations\n", + " - Breakdown of Key Functions or Components\n", + " Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"security_policies_notes\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Example Runs and Scripts\n", + "Let's explore example runs and scripts that demonstrate how to use this codebase in practice. We'll look at working examples, command-line usage, and sample notebooks that showcase the core functionality. This section will also point to demo datasets and test scenarios that can help new users get started quickly." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - Provide working script examples.\n", + " - Include CLI usage instructions or sample notebooks.\n", + " - Link to demo datasets or test scenarios.\n", + " Please remove the following sections: \n", + " - Potential Risks or Failure Points\n", + " - Assumptions or Limitations\n", + " - Breakdown of Key Functions or Components\n", + " Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"runnable_examples\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "## Known Issues and Future Improvements\n", + "Let's examine the current limitations and areas for improvement in the codebase. This section will document known technical debt, bugs, and feature gaps that need to be addressed. We'll also outline proposed enhancements and reference any existing tickets or GitHub issues tracking these improvements." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = explain_code(\n", + " user_instructions=\"\"\"\n", + " Please provide a summary of the following bullet points only.\n", + " - List current limitations or technical debt.\n", + " - Outline proposed enhancements or refactors.\n", + " - Reference relevant tickets, GitHub issues, or roadmap items.\n", + " Please remove Potential Risks or Failure Points and Assumptions or Limitations sections. Please don't add any other sections.\n", + " \"\"\",\n", + " content_id=\"issues_and_improvements_log\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "copyright-ccbede139a26452183291a108b791513", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "validmind-1QuffXMV-py3.11", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.9" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/site/notebooks/code_samples/credit_risk/CreditRiskData.xlsx b/site/notebooks/use_cases/credit_risk/CreditRiskData.xlsx similarity index 100% rename from site/notebooks/code_samples/credit_risk/CreditRiskData.xlsx rename to site/notebooks/use_cases/credit_risk/CreditRiskData.xlsx diff --git a/site/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb b/site/notebooks/use_cases/credit_risk/application_scorecard_executive.ipynb similarity index 98% rename from site/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb rename to site/notebooks/use_cases/credit_risk/application_scorecard_executive.ipynb index 2115a88b50..6ac66e9458 100644 --- a/site/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb +++ b/site/notebooks/use_cases/credit_risk/application_scorecard_executive.ipynb @@ -98,7 +98,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -163,8 +163,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb b/site/notebooks/use_cases/credit_risk/application_scorecard_full_suite.ipynb similarity index 99% rename from site/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb rename to site/notebooks/use_cases/credit_risk/application_scorecard_full_suite.ipynb index 83a6d276c8..1666b1d25c 100644 --- a/site/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb +++ b/site/notebooks/use_cases/credit_risk/application_scorecard_full_suite.ipynb @@ -112,7 +112,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -177,8 +177,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb b/site/notebooks/use_cases/credit_risk/application_scorecard_with_bias.ipynb similarity index 99% rename from site/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb rename to site/notebooks/use_cases/credit_risk/application_scorecard_with_bias.ipynb index a915608315..f845c74e22 100644 --- a/site/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb +++ b/site/notebooks/use_cases/credit_risk/application_scorecard_with_bias.ipynb @@ -113,7 +113,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -178,8 +178,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb b/site/notebooks/use_cases/credit_risk/application_scorecard_with_ml.ipynb similarity index 99% rename from site/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb rename to site/notebooks/use_cases/credit_risk/application_scorecard_with_ml.ipynb index 357d572b79..7e0cce645e 100644 --- a/site/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb +++ b/site/notebooks/use_cases/credit_risk/application_scorecard_with_ml.ipynb @@ -125,7 +125,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -190,8 +190,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/credit_risk/custom_tests/ScoreBandDiscriminationMetrics.py b/site/notebooks/use_cases/credit_risk/custom_tests/ScoreBandDiscriminationMetrics.py similarity index 100% rename from site/notebooks/code_samples/credit_risk/custom_tests/ScoreBandDiscriminationMetrics.py rename to site/notebooks/use_cases/credit_risk/custom_tests/ScoreBandDiscriminationMetrics.py diff --git a/site/notebooks/code_samples/credit_risk/document_excel_application_scorecard.ipynb b/site/notebooks/use_cases/credit_risk/document_excel_application_scorecard.ipynb similarity index 99% rename from site/notebooks/code_samples/credit_risk/document_excel_application_scorecard.ipynb rename to site/notebooks/use_cases/credit_risk/document_excel_application_scorecard.ipynb index cc7685c9ca..6d693eb538 100644 --- a/site/notebooks/code_samples/credit_risk/document_excel_application_scorecard.ipynb +++ b/site/notebooks/use_cases/credit_risk/document_excel_application_scorecard.ipynb @@ -109,7 +109,7 @@ "- **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", "- **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", "- **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + "- **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -179,8 +179,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk - CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -488,7 +486,7 @@ "\n", "
Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests" ] }, { diff --git a/site/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/site/notebooks/use_cases/model_validation/validate_application_scorecard.ipynb similarity index 99% rename from site/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb rename to site/notebooks/use_cases/model_validation/validate_application_scorecard.ipynb index 22ab661f78..5cc55f8473 100644 --- a/site/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/site/notebooks/use_cases/model_validation/validate_application_scorecard.ipynb @@ -151,7 +151,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html))\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. (Learn more: [Run tests with multiple datasets](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html))\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -185,8 +185,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Credit Risk — CECL`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down — don’t worry, we’ll adjust these permissions next for validation.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -904,7 +902,7 @@ "source": [ "
Want to learn more about navigating ValidMind tests?\n", "

\n", - "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" + "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests" ] }, { @@ -1588,7 +1586,7 @@ "source": [ "
Want to learn more about custom tests?\n", "

\n", - "Refer to our in-depth introduction to custom tests: Implement custom tests
" + "Refer to our in-depth introduction to custom tests: Implement custom tests" ] }, { @@ -1789,7 +1787,7 @@ "\n", "All notebook samples can be found in the following directories of the ValidMind Library GitHub repository:\n", "\n", - "- [Code samples](https://github.com/validmind/validmind-library/tree/main/notebooks/code_samples)\n", + "- [Use cases](https://github.com/validmind/validmind-library/tree/main/notebooks/use_cases)\n", "- [How-to guides](https://github.com/validmind/validmind-library/tree/main/notebooks/how_to)\n", "\n", "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." diff --git a/site/notebooks/code_samples/model_validation/xgb_model_champion.pkl b/site/notebooks/use_cases/model_validation/xgb_model_champion.pkl similarity index 100% rename from site/notebooks/code_samples/model_validation/xgb_model_champion.pkl rename to site/notebooks/use_cases/model_validation/xgb_model_champion.pkl diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/bbc_text_cls.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/bbc_text_cls.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/bbc_text_cls.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/bbc_text_cls.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/bbc_text_cls_reference.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/bbc_text_cls_reference.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/bbc_text_cls_reference.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/bbc_text_cls_reference.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/cnn_dailymail_100_with_predictions.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/cnn_dailymail_100_with_predictions.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/cnn_dailymail_100_with_predictions.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/cnn_dailymail_100_with_predictions.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/cnn_dailymail_500_with_predictions.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/cnn_dailymail_500_with_predictions.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/cnn_dailymail_500_with_predictions.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/cnn_dailymail_500_with_predictions.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/sentiments.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/sentiments.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/sentiments.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/sentiments.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/datasets/sentiments_with_predictions.csv b/site/notebooks/use_cases/nlp_and_llm/datasets/sentiments_with_predictions.csv similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/datasets/sentiments_with_predictions.csv rename to site/notebooks/use_cases/nlp_and_llm/datasets/sentiments_with_predictions.csv diff --git a/site/notebooks/code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb similarity index 99% rename from site/notebooks/code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb index 633c99e0c0..346ecb6e1d 100644 --- a/site/notebooks/code_samples/nlp_and_llm/foundation_models_integration_demo.ipynb +++ b/site/notebooks/use_cases/nlp_and_llm/foundation_models_integration_demo.ipynb @@ -130,8 +130,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/nlp_and_llm/foundation_models_summarization_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/foundation_models_summarization_demo.ipynb similarity index 100% rename from site/notebooks/code_samples/nlp_and_llm/foundation_models_summarization_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/foundation_models_summarization_demo.ipynb diff --git a/site/notebooks/use_cases/nlp_and_llm/gen_ai_rag_template.yaml b/site/notebooks/use_cases/nlp_and_llm/gen_ai_rag_template.yaml new file mode 100644 index 0000000000..4f074f6a34 --- /dev/null +++ b/site/notebooks/use_cases/nlp_and_llm/gen_ai_rag_template.yaml @@ -0,0 +1,538 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial + +- id: conceptual_soundness + title: Conceptual Soundness + index_only: true + sections: + - id: model_overview + title: Model Overview + guidelines: + - Provide a clear and concise description of the model's main concept, + explaining the underlying financial theory or economic rationale. + - Justify the choice of the model, algorithm, or approach, relating it + to the financial institution's needs, objectives, and any relevant + industry standards. + - Discuss any alternative models or approaches considered during the + model development process, and explain why they were not selected. + - Describe any key assumptions made in the model and their potential + implications on the model's results and performance. + - Explain how the model's structure and design contribute to its + robustness, stability, and reliability in a financial context. + parent_section: conceptual_soundness + contents: + - content_id: model_overview + content_type: text + - id: intended_use_business_use_case + title: Intended Use and Business Use Case + index_only: true + condensed: true + parent_section: conceptual_soundness + sections: + - id: intended_use + title: Intended Use + parent_section: intended_use_business_use_case + guidelines: + - Clearly describe the specific business use case(s) for which the + model is designed, including the intended users and the financial + institution's objectives. + - Explain how the model fits into the overall business strategy and + decision-making processes of the financial institution. + - id: regulatory_requirements + title: Regulatory Requirements + parent_section: intended_use_business_use_case + guidelines: + - Detail any specific risks, regulatory requirements, or other + considerations associated with the model's intended use, and how + they have been addressed in the model development process. + - id: model_limitations + title: Model Limitations + parent_section: intended_use_business_use_case + guidelines: + - Discuss the model's intended scope, including any limitations, + boundaries, or exclusions in its applicability. + - Describe any potential external factors, such as economic or + market conditions, that could impact the model's performance, and + how they have been considered in the model development process. + - id: model_selection + title: Model Selection + guidelines: + - Provide a detailed description of the selected model, including its + algorithm, mathematical foundations, and key features that make it + suitable for the intended use case. + - Explain the rationale behind choosing the specific model, and how it + addresses the financial institution's objectives, regulatory + requirements, and risk management needs. + - Compare the selected model with alternative models or approaches that + were considered during the model development process, highlighting + their strengths and weaknesses, and explaining why the chosen model is + the most appropriate. + - Describe any model customizations or adaptations made to better align + it with the financial institution's needs, and discuss the potential + impact of these changes on the model's performance. + - Explain any potential limitations or drawbacks of the selected model + in the context of the intended use case, and how they have been + mitigated or addressed during the model development process. + parent_section: conceptual_soundness +- id: data_preparation + title: Data Preparation + index_only: true + sections: + - id: data_description + title: Data description + guidelines: + - Provide a comprehensive overview of the data sources used in the + model, including internal and external sources, and specify the data's + time period, frequency, and granularity. + - Describe the main variables and features used in the model, including + both input and output variables, and explain their relevance to the + model's purpose and intended use case. + - Detail any data transformations, preprocessing, or feature engineering + performed on the raw data to prepare it for model input, and explain + the rationale for these transformations. + - Address any data quality concerns, such as missing values, outliers, + or inconsistencies, and describe the methods used to handle these + issues. + - Discuss any potential biases, dependencies, or limitations in the data + that could impact the model's performance, and explain how they have + been considered or mitigated during the data preparation process. + contents: + - content_id: text_data_quality + content_type: text + options: + default_text: | + # Text Data Quality Metrics and Tests + - content_type: test + content_id: validmind.data_validation.Duplicates + - content_type: test + content_id: validmind.data_validation.nlp.StopWords + - content_type: test + content_id: validmind.data_validation.nlp.Punctuations + - content_type: test + content_id: validmind.data_validation.nlp.CommonWords + - content_type: test + content_id: validmind.data_validation.nlp.LanguageDetection + - content_type: test + content_id: validmind.data_validation.nlp.Toxicity + - content_type: test + content_id: validmind.data_validation.nlp.PolarityAndSubjectivity + - content_type: test + content_id: validmind.data_validation.nlp.Sentiment + parent_section: data_preparation + - id: feature_selection + title: Feature Selection and Engineering + guidelines: + - Describe the process used to select the most relevant features for the + model, including any feature selection techniques or criteria applied, + such as correlation analysis, mutual information, or forward/backward + selection. + - Explain the rationale behind including or excluding specific features, + and discuss their importance and contribution to the model's + performance and intended use case. + - Detail any feature engineering techniques applied to create new + features or transform existing ones, such as dimensionality reduction, + aggregation, or interaction terms, and explain their relevance and + purpose in the context of the model. + - Describe any data normalization or scaling techniques used to + standardize the input features, and explain the rationale for their + application in the model. + - Discuss potential multicollinearity, redundancy, or other issues among + the selected features, and describe the methods used to address these + concerns during the feature selection and engineering process. + parent_section: data_preparation + contents: + - content_id: feature_selection + content_type: text +- id: model_development + title: Model Development + index_only: true + sections: + - id: model_training + title: Model Training + guidelines: + - Describe the model training process, including the algorithm used, any + hyperparameters or settings, and the optimization techniques employed + to minimize the loss function or maximize the objective function. + - Detail the model validation and selection process, including the use + of cross-validation, holdout samples, or other techniques to assess + the model's performance and prevent overfitting. + - Provide a summary of the training results, including performance + metrics such as accuracy, precision, recall, F1 score, or other + relevant measures, depending on the model's intended use case. + - Discuss any challenges, issues, or trade-offs encountered during the + model training process, such as overfitting, underfitting, or class + imbalance, and explain how they were addressed or mitigated. + - Describe any tuning or optimization steps performed to improve the + model's performance, such as hyperparameter tuning, feature selection, + or other adjustments, and explain the rationale for these changes. + parent_section: model_development + contents: + - content_id: model_training + content_type: text + - id: prompt_validation + title: Prompt Validation + guidelines: + - Provide documentation of the prompt used for the model. Describe the + type of prompting used (e.g. few-shot, zero-shot, chain-of-thought + etc.). Explain how the prompt was constructed and how it was + validated. + parent_section: model_development + contents: + - content_type: test + content_id: validmind.prompt_validation.Bias + options: + title: Bias + - content_type: test + content_id: validmind.prompt_validation.Clarity + options: + title: Clarity + - content_type: test + content_id: validmind.prompt_validation.Conciseness + options: + title: Conciseness + - content_type: test + content_id: validmind.prompt_validation.Delimitation + options: + title: Delimitation + - content_type: test + content_id: validmind.prompt_validation.NegativeInstruction + options: + title: Negative Instruction + - content_type: test + content_id: validmind.prompt_validation.Specificity + options: + title: Specificity + - id: model_evaluation + title: Model Evaluation + parent_section: model_development + guidelines: + - Describe the process used to evaluate the model's performance on a + test or validation dataset that was not used during training, to + assess its generalizability and robustness. + - Present the key performance metrics for the model evaluation, such as + accuracy, precision, recall, F1 score, AUC-ROC, mean squared error, or + other relevant measures, depending on the model's intended use case. + - Provide graphical representations of the model's performance, such as + confusion matrices, ROC curves, or residual plots, to help visualize + its effectiveness and identify any areas for improvement. + - Discuss the model's performance in the context of its intended use + case, and compare it to any benchmarks, industry standards, or + alternative models, as appropriate. + - Identify any limitations, weaknesses, or areas for improvement in the + model's performance, and discuss potential strategies for addressing + these concerns in future iterations or updates. + sections: + - id: embedding_model + title: Embedding Model + parent_section: model_evaluation + guidelines: + - Embedding model guidelines + contents: + - content_type: test + content_id: validmind.model_validation.embeddings.StabilityAnalysisRandomNoise + - content_type: test + content_id: validmind.model_validation.embeddings.StabilityAnalysisSynonyms + - content_type: test + content_id: validmind.model_validation.embeddings.StabilityAnalysisTranslation + - content_type: test + content_id: validmind.model_validation.embeddings.CosineSimilarityHeatmap + - content_type: test + content_id: validmind.model_validation.embeddings.CosineSimilarityDistribution + - content_type: test + content_id: validmind.model_validation.embeddings.PCAComponentsPairwisePlots + - id: retrieval_model + title: Retrieval model + parent_section: model_evaluation + guidelines: + - Retrieval model guideline + contents: + - content_type: test + content_id: validmind.model_validation.ragas.ContextRecall + - content_type: test + content_id: validmind.model_validation.ragas.ContextEntityRecall + - content_type: test + content_id: validmind.model_validation.ragas.ContextPrecision + - content_type: test + content_id: validmind.model_validation.ragas.ContextPrecisionWithoutReference + - id: rag_model + title: RAG model + parent_section: model_evaluation + guidelines: + - RAG model guideline + contents: + - content_type: test + content_id: validmind.model_validation.ragas.SemanticSimilarity + - content_type: test + content_id: validmind.model_validation.ragas.Faithfulness + - content_type: test + content_id: validmind.model_validation.ragas.ResponseRelevancy + - content_type: test + content_id: validmind.model_validation.ragas.AnswerCorrectness + - content_type: test + content_id: validmind.model_validation.TokenDisparity + - content_type: test + content_id: validmind.model_validation.BleuScore + - content_type: test + content_id: validmind.model_validation.RougeScore + - content_type: test + content_id: validmind.model_validation.MeteorScore + - content_type: test + content_id: validmind.model_validation.ragas.AspectCritic + - content_type: test + content_id: validmind.model_validation.ToxicityScore + - content_type: test + content_id: validmind.model_validation.RegardScore +- id: monitoring_governance + title: Monitoring and Governance + index_only: true + sections: + - id: monitoring_plan + title: Monitoring Plan + guidelines: + - Describe the plan for ongoing monitoring of the model's performance, + including the frequency of evaluations, the performance metrics to be + assessed, and any thresholds or triggers for action. + - Explain the process for identifying and addressing any changes in the + model's performance or the underlying data that may require model + updates, recalibration, or retraining. + - Detail the procedures for model validation and backtesting, to ensure + the model remains accurate, reliable, and compliant with regulatory + requirements and industry standards. + - Discuss the governance framework in place to oversee the model's use, + including the roles and responsibilities of various stakeholders, such + as model developers, validators, and risk managers. + - Describe the model's documentation and version control procedures, to + ensure that changes, updates, and improvements are properly tracked + and recorded. + contents: + - content_id: monitoring_plan + content_type: text + options: + default_text: > + To ensure the ongoing effectiveness of the model, it will be + monitored on a regular basis. The monitoring plan includes the + following steps: + + + 1. **Performance Metrics Monitoring:** The key performance metrics + such as Accuracy, Precision, Recall, F1 Score, and ROC-AUC will be + + tracked on a regular basis. These metrics will be computed for + both the training, validation and test datasets to identify any + signs of + + overfitting or underperformance. + + 2. **Data Drift Monitoring:** Over time, the distribution of the + input data may change, a phenomenon known as data drift. This + could + + impact the model's performance. We will monitor the distributions + of the input features and the target variable to detect any + significant + + changes that may require retraining of the model. + + 3. **Outcome Monitoring:** The actual outcomes (whether a customer + exited the bank or not) will be compared with the model's + predictions + + to assess the model's performance in a real-world setting. + + 4. **Feature Importance Monitoring:** The importance of different + features for the model's predictions will be tracked. If there are + + significant shifts in feature importance, this could indicate + changes in the underlying patterns in the data. + + 5. **Periodic Model Retraining:** Depending on the findings from + the above monitoring activities, the model may need to be + retrained + + periodically. This will be done using the most recent data to + ensure that the model stays up-to-date with the latest patterns + and trends. + + + In terms of governance, a clear process will be put in place for + managing these monitoring activities. This will include clear + roles + + and responsibilities, documentation of the monitoring results, and + a process for deciding when and how to take action based on the + + monitoring results (for example, when to retrain the model). Any + major decisions about the model (such as changes to the model or + its + + features) will be made in a transparent and accountable manner, + with appropriate documentation and sign-off. + parent_section: monitoring_governance + - id: monitoring_implementation + title: Monitoring Implementation + guidelines: + - Describe the tools, systems, or platforms used to implement the + monitoring plan, including any relevant software, data pipelines, or + reporting tools. + - Detail the process for collecting and storing the data needed for + ongoing monitoring, including any data preprocessing, cleaning, or + transformation steps required. + - Explain the procedures for analyzing the model's performance metrics + and generating monitoring reports, including any statistical tests or + visualizations used to assess the model's performance and stability. + - Discuss the escalation process and communication channels for + reporting any significant deviations in the model's performance, as + well as the decision-making process for determining appropriate + actions, such as model updates or recalibration. + - Describe any training or educational programs in place to ensure that + relevant stakeholders, such as model developers, validators, and risk + managers, are equipped to understand, interpret, and act on the + monitoring results. + parent_section: monitoring_governance + contents: + - content_id: monitoring_implementation + content_type: text + options: + default_text: > + Implementing the monitoring plan requires a systematic approach + with clear steps and procedures. Here are the steps for + implementing + + the monitoring plan: + + + 1. **Establish Baseline Metrics:** At the outset, we'll establish + baseline performance metrics for the model. These will serve as + + reference points for future comparison. + + 2. **Automate Metric Calculation:** Performance metrics such as + Accuracy, Precision, Recall, F1 Score, and ROC-AUC will be + automatically + + calculated and recorded for both the training and validation + datasets after each run of the model. + + 3. **Set Up Data Drift Monitors:** We'll set up automated + monitoring of the distributions of the input features and the + target + + variable. Any significant changes in these distributions will + trigger alerts. + + 4. **Implement Outcome Monitoring:** We'll compare the model's + predictions with the actual outcomes on a regular basis. + Discrepancies + + will be investigated to understand the root cause. + + 5. **Track Feature Importance:** We'll keep a record of feature + importance as indicated by the model. Any significant shifts in + + feature importance over time will be thoroughly reviewed. + + 6. **Establish Retraining Protocols:** Based on the findings from + the above monitoring activities, we'll establish clear protocols + + for when and how the model should be retrained. This may include + criteria for triggering a retraining, procedures for carrying out + + the retraining, and protocols for testing and validating the + retrained model before it is put back into service. + + 7. **Document and Review:** All monitoring activities and their + results will be documented in a transparent and accessible manner. + + These documents will be reviewed regularly by a designated team or + individual to ensure that any issues are promptly identified and + addressed. + + + By implementing this monitoring plan, we can ensure that the model + continues to perform well and that any issues are quickly + + identified and addressed. The ultimate aim is to ensure that the + model continues to provide accurate and reliable predictions that + + can support the bank's decision-making processes. + - id: governance_plan + title: Governance Plan + guidelines: + - Describe the overall governance framework and processes established to + ensure proper oversight and management of the model, including the + roles and responsibilities of key stakeholders such as model + developers, validators, and risk managers. + - Detail the policies and procedures for model risk management, + including model risk identification, assessment, and mitigation + strategies. + - Explain the model approval process, including any internal or external + reviews, audits, or regulatory assessments that must be completed + before the model is put into production. + - Discuss the procedures for ongoing model maintenance, updates, and + improvements, including the documentation and version control + processes to track and record changes to the model. + - Describe the contingency plans in place to manage potential model + failures or performance issues, such as fallback models, alternative + data sources, or manual processes, and explain the criteria for + activating these contingency measures. + parent_section: monitoring_governance + contents: + - content_id: governance_plan + content_type: text + options: + default_text: > + Effective governance of the model is crucial to ensure its + reliability, security, and compliance with regulatory + requirements. + + Here is the plan for model governance: + + + 1. **Roles and Responsibilities:** Clear roles and + responsibilities will be assigned for model development, + validation, deployment, + + monitoring, and retraining. This will include a model owner, who + will have overall responsibility for the model, as well as others + + responsible for specific tasks. + + 2. **Model Documentation:** Comprehensive documentation will be + maintained for the model, including details of its development, + + validation, and performance, as well as any changes made to the + model or its inputs over time. This documentation will be updated + + regularly and will be accessible to all relevant stakeholders. + + 3. **Change Control:** Any changes to the model or its inputs will + be subject to a strict change control process, including + + documentation of the proposed change, review and approval by a + designated authority, testing and validation of the change, and a + + post-implementation review. + + 4. **Security and Access Control:** Measures will be put in place + to ensure the security of the model and its data, including + + access controls, data encryption, and regular security audits. + + 5. **Regulatory Compliance:** The model will be designed and + operated in compliance with all relevant regulatory requirements, + + and its compliance will be regularly reviewed and confirmed. + + 6. **Auditability:** The model and its operations will be + auditable, with clear and accessible records that can be reviewed + by + + internal or external auditors. + + 7. **Training and Awareness:** All individuals involved in the + development, operation, and oversight of the model will receive + + appropriate training and will be kept aware of their + responsibilities, the model's performance, and any relevant + developments or issues. diff --git a/site/notebooks/code_samples/nlp_and_llm/hugging_face_integration_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/hugging_face_integration_demo.ipynb similarity index 99% rename from site/notebooks/code_samples/nlp_and_llm/hugging_face_integration_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/hugging_face_integration_demo.ipynb index dd7e03d056..c424355de3 100644 --- a/site/notebooks/code_samples/nlp_and_llm/hugging_face_integration_demo.ipynb +++ b/site/notebooks/use_cases/nlp_and_llm/hugging_face_integration_demo.ipynb @@ -132,8 +132,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/nlp_and_llm/hugging_face_summarization_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/hugging_face_summarization_demo.ipynb similarity index 99% rename from site/notebooks/code_samples/nlp_and_llm/hugging_face_summarization_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/hugging_face_summarization_demo.ipynb index 8209e65c2c..14816ce332 100644 --- a/site/notebooks/code_samples/nlp_and_llm/hugging_face_summarization_demo.ipynb +++ b/site/notebooks/use_cases/nlp_and_llm/hugging_face_summarization_demo.ipynb @@ -129,8 +129,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/nlp_and_llm/llm_summarization_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/llm_summarization_demo.ipynb similarity index 99% rename from site/notebooks/code_samples/nlp_and_llm/llm_summarization_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/llm_summarization_demo.ipynb index 159a16e79d..fe6e879568 100644 --- a/site/notebooks/code_samples/nlp_and_llm/llm_summarization_demo.ipynb +++ b/site/notebooks/use_cases/nlp_and_llm/llm_summarization_demo.ipynb @@ -163,8 +163,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Sales/Prospecting`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/prompt_validation_demo.ipynb similarity index 98% rename from site/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb rename to site/notebooks/use_cases/nlp_and_llm/prompt_validation_demo.ipynb index 49633b3a99..ec70da33c8 100644 --- a/site/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb +++ b/site/notebooks/use_cases/nlp_and_llm/prompt_validation_demo.ipynb @@ -94,7 +94,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -159,8 +159,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/use_cases/nlp_and_llm/rag_benchmark_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/rag_benchmark_demo.ipynb new file mode 100644 index 0000000000..51515ad7bb --- /dev/null +++ b/site/notebooks/use_cases/nlp_and_llm/rag_benchmark_demo.ipynb @@ -0,0 +1,1869 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# RAG Model Benchmarking Demo\n", + "\n", + "In this notebook, we are going to implement a simple RAG Model for automating the process of answering RFP questions using GenAI. We will see how we can initialize an embedding model, a retrieval model and a generator model with LangChain components and use them within the ValidMind Library to run tests against them. We'll demonstrate how to set up multiple models for benchmarking at each stage of the RAG pipeline - specifically two embedding models, two retrieval models with different parameters, and two LLM models (GPT-3.5 and GPT-4o) - allowing for comparison of performance across different configurations. Finally, we will see how we can put them together in a Pipeline and run that to get e2e results and run tests against that." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About ValidMind](#toc1__) \n", + " - [Before you begin](#toc1_1__) \n", + " - [New to ValidMind?](#toc1_2__) \n", + " - [Key concepts](#toc1_3__) \n", + "- [Setting up](#toc2__) \n", + " - [Install the ValidMind Library](#toc2_1__) \n", + " - [Initialize the ValidMind Library](#toc2_2__) \n", + " - [Register sample model](#toc2_2_1__) \n", + " - [Apply documentation template](#toc2_2_2__) \n", + " - [Get your code snippet](#toc2_2_3__) \n", + "- [Read Open AI API Key](#toc3__) \n", + "- [Dataset Loader](#toc4__) \n", + "- [Data validation](#toc5__) \n", + " - [Duplicates](#toc5_1__) \n", + " - [Stop Words](#toc5_2__) \n", + " - [Punctuations](#toc5_3__) \n", + " - [Common Words](#toc5_4__) \n", + " - [Language Detection](#toc5_5__) \n", + " - [Toxicity Score](#toc5_6__) \n", + " - [Polarity and Subjectivity](#toc5_7__) \n", + " - [Sentiment](#toc5_8__) \n", + " - [Assign Predictions](#toc5_9__) \n", + " - [Run tests](#toc5_10__) \n", + " - [Generate embeddings for the Train Set](#toc5_11__) \n", + " - [Insert embeddings and questions into Vector DB](#toc5_12__) \n", + "- [Prompt Evaluation](#toc6__) \n", + "- [RAGAS evaluation](#toc7__) \n", + " - [Semantic Similarity](#toc7_1__) \n", + " - [Context Entity Recall](#toc7_2__) \n", + " - [Context Precision](#toc7_3__) \n", + " - [Context Precision Without Reference](#toc7_4__) \n", + " - [Faithfulness](#toc7_5__) \n", + " - [Response Relevancy](#toc7_6__) \n", + " - [Context Recall](#toc7_7__) \n", + " - [Answer Correctness](#toc7_8__) \n", + " - [Aspect Critic](#toc7_9__) \n", + " - [Noise Sensitivity](#toc7_10__) \n", + "- [Generation quality](#toc8__) \n", + " - [Token Disparity](#toc8_1__) \n", + " - [ROUGE Score](#toc8_2__) \n", + " - [BLEU Score](#toc8_3__) \n", + " - [BERT Score](#toc8_4__) \n", + " - [METEOR Score](#toc8_5__) \n", + "- [Bias and Toxicity](#toc9__) \n", + " - [Toxicity Score](#toc9_1__) \n", + " - [Regard Score](#toc9_2__) \n", + "- [Upgrade ValidMind](#toc10__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "- **FunctionModels**: ValidMind offers support for creating `VMModel` instances from Python functions. This enables us to support any \"model\" by simply using the provided function as the model's `predict` method.\n", + "- **PipelineModels**: ValidMind models (`VMModel` instances) of any type can be piped together to create a model pipeline. This allows model components to be created and tested/documented independently, and then combined into a single model for end-to-end testing and documentation. We use the `|` operator to pipe models together.\n", + "- **RAG**: RAG stands for Retrieval Augmented Generation and refers to a wide range of GenAI applications where some form of retrieval is used to add context to the prompt so that the LLM that generates content can refer to it when creating its output. In this notebook, we are going to implement a simple RAG setup using LangChain components." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Prerequisites\n", + "\n", + "Let's go ahead and install the `validmind` library if its not already installed... Then we can install the `qdrant-client` library for our vector store and `langchain` for everything else:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q \"validmind[llm]\"" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q qdrant-client langchain langchain-openai sentencepiece" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Install the ValidMind Library\n", + "\n", + "
Recommended Python versions\n", + "

\n", + "Python 3.8 <= x <= 3.11
\n", + "\n", + "To install the library:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Gen AI RAG`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Can't select this template?\n", + "

\n", + "Your organization administrators may need to add it to your template library:\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " api_host = \"https://api.prod.validmind.ai/api/v1/tracking\",\n", + " api_key = \"...\",\n", + " api_secret = \"...\",\n", + " model = \"...\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Read Open AI API Key\n", + "\n", + "We will need to have an OpenAI API key to be able to use their `text-embedding-3-small` and `text-embedding-3-large` models for our embeddings, `gpt-3.5-turbo` and `gpt-4o` models for our generator and `gpt-4o` model for our LLM-as-Judge tests. If you don't have an OpenAI API key, you can get one by signing up at [OpenAI](https://platform.openai.com/signup). Then you can create a `.env` file in the root of your project and the following cell will load it from there. Alternatively, you can just uncomment the line below to directly set the key (not recommended for security reasons)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# load openai api key\n", + "import os\n", + "\n", + "import dotenv\n", + "import nltk\n", + "\n", + "dotenv.load_dotenv()\n", + "nltk.download('stopwords')\n", + "nltk.download('punkt_tab')\n", + "\n", + "# os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n", + "\n", + "if not \"OPENAI_API_KEY\" in os.environ:\n", + " raise ValueError(\"OPENAI_API_KEY is not set\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Dataset Loader\n", + "\n", + "Great, now that we have all of our dependencies installed, the ValidMind Library initialized and connected to our model and our OpenAI API key setup, we can go ahead and load our datasets. We will use the synthetic `RFP` dataset included with ValidMind for this notebook. This dataset contains a variety of RFP questions and ground truth answers that we can use both as the source where our Retriever will search for similar question-answer pairs as well as our test set for evaluating the performance of our RAG model. To do this, we just have to load it and call the preprocess function to get a split of the data into train and test sets." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the sample dataset from the library\n", + "from validmind.datasets.llm.rag import rfp\n", + "\n", + "raw_df = rfp.load_data()\n", + "train_df, test_df = rfp.preprocess(raw_df)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_train_ds = vm.init_dataset(\n", + " train_df,\n", + " text_column=\"question\",\n", + " target_column=\"ground_truth\",\n", + ")\n", + "\n", + "vm_test_ds = vm.init_dataset(\n", + " test_df,\n", + " text_column=\"question\",\n", + " target_column=\"ground_truth\",\n", + ")\n", + "\n", + "vm_test_ds.df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Data validation\n", + "\n", + "Now that we have loaded our dataset, we can go ahead and run some data validation tests right away to start assessing and documenting the quality of our data. Since we are using a text dataset, we can use ValidMind's built-in array of text data quality tests to check that things like number of duplicates, missing values, and other common text data issues are not present in our dataset. We can also run some tests to check the sentiment and toxicity of our data." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Duplicates\n", + "\n", + "First, let's check for duplicates in our dataset. We can use the `validmind.data_validation.Duplicates` test and pass our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "\n", + "run_test(\n", + " test_id=\"validmind.data_validation.Duplicates\",\n", + " inputs={\"dataset\": vm_train_ds},\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Stop Words\n", + "\n", + "Next, let's check for stop words in our dataset. We can use the `validmind.data_validation.StopWords` test and pass our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.StopWords\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Punctuations\n", + "\n", + "Next, let's check for punctuations in our dataset. We can use the `validmind.data_validation.Punctuations` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.Punctuations\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Common Words\n", + "\n", + "Next, let's check for common words in our dataset. We can use the `validmind.data_validation.CommonWord` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.CommonWords\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Language Detection\n", + "\n", + "For documentation purposes, we can detect and log the languages used in the dataset with the `validmind.data_validation.LanguageDetection` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.LanguageDetection\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Toxicity Score\n", + "\n", + "Now, let's go ahead and run the `validmind.data_validation.nlp.Toxicity` test to compute a toxicity score for our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.Toxicity\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Polarity and Subjectivity\n", + "\n", + "We can also run the `validmind.data_validation.nlp.PolarityAndSubjectivity` test to compute the polarity and subjectivity of our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.PolarityAndSubjectivity\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Sentiment\n", + "\n", + "Finally, we can run the `validmind.data_validation.nlp.Sentiment` test to plot the sentiment of our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.Sentiment\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Embedding Model\n", + "\n", + "Now that we have our dataset loaded and have run some data validation tests to assess and document the quality of our data, we can go ahead and initialize our embedding model. We will use `text-embedding-3-small` and `text-embedding-3-large` models from OpenAI for this purpose wrapped in the `OpenAIEmbeddings` class from LangChain. This model will be used to \"embed\" our questions both for inserting the question-answer pairs from the \"train\" set into the vector store and for embedding the question from inputs when making predictions with our RAG model." + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_openai import OpenAIEmbeddings\n", + "\n", + "embedding_small_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", + "\n", + "\n", + "def embed_small(input):\n", + " \"\"\"Returns a text embedding for the given text\"\"\"\n", + " return embedding_small_client.embed_query(input[\"question\"])\n", + "\n", + "\n", + "vm_embedder_small = vm.init_model(input_id=\"embedding_small_model\", predict_fn=embed_small)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "embedding_large_client = OpenAIEmbeddings(model=\"text-embedding-3-large\")\n", + "\n", + "\n", + "def embed_large(input):\n", + " \"\"\"Returns a text embedding for the given text\"\"\"\n", + " return embedding_large_client.embed_query(input[\"question\"])\n", + "\n", + "\n", + "vm_embedder_large = vm.init_model(input_id=\"embedding_large_model\", predict_fn=embed_large)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "What we have done here is to initialize the `OpenAIEmbeddings` class so it uses OpenAI's `text-embedding-3-small` and `text-embedding-3-large` models. We then created an `embed` function that takes in an `input` dictionary and uses the `embed_query` method of the embedding client to compute the embeddings of the `question`. We use an `embed` function since that is how ValidMind supports any custom model. We will use this strategy for the retrieval and generator models as well but you could also use, say, a HuggingFace model directly. See the documentation for more information on which model types are directly supported - [ValidMind Documentation](https://docs.validmind.ai/validmind/validmind.html)... Finally, we use the `init_model` function from the ValidMind Library to create a `VMModel` object that can be used in ValidMind tests. This also logs the model to our model documentation and any test that uses the model will be linked to the logged model and its metadata." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign Predictions\n", + "\n", + "To precompute the embeddings for our test set, we can call the `assign_predictions` method of our `vm_test_ds` object we created above. This will compute the embeddings for each question in the test set and store them in the a special prediction column of the test set thats linked to our `vm_embedder` model. This will allow us to use these embeddings later when we run tests against our embedding model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(vm_embedder_small)\n", + "vm_test_ds.assign_predictions(vm_embedder_large)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Run tests\n", + "\n", + "Now that everything is setup for the embedding model, we can go ahead and run some tests to assess and document the quality of our embeddings. We will use the `validmind.model_validation.embeddings.*` tests to compute a variety of metrics against our model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisRandomNoise\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + " params={\n", + " \"probability\": 0.3,\n", + " \"mean_similarity_threshold\": 0.7,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisSynonyms\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + " params={\n", + " \"probability\": 0.3,\n", + " \"mean_similarity_threshold\": 0.7,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisTranslation\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + " params={\n", + " \"source_lang\": \"en\",\n", + " \"target_lang\": \"fr\",\n", + " \"mean_similarity_threshold\": 0.7,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.CosineSimilarityHeatmap\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.CosineSimilarityDistribution\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.PCAComponentsPairwisePlots\",\n", + " input_grid={\n", + " \"model\": [vm_embedder_small, vm_embedder_large],\n", + " \"dataset\": [vm_test_ds],\n", + " },\n", + " params={\n", + " \"n_components\": 3,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Setup Vector Store\n", + "\n", + "Great, so now that we have assessed our embedding model and verified that it is performing well, we can go ahead and use it to compute embeddings for our question-answer pairs in the \"train\" set. We will then use these embeddings to insert the question-answer pairs into a vector store. We will use an in-memory `qdrant` vector database for demo purposes but any option would work just as well here. We will use the `QdrantClient` class from LangChain to interact with the vector store. This class will allow us to insert and search for embeddings in the vector store." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Generate embeddings for the Train Set\n", + "\n", + "We can use the same `assign_predictions` method from earlier except this time we will use the `vm_train_ds` object to compute the embeddings for the question-answer pairs in the \"train\" set." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_train_ds.assign_predictions(vm_embedder_small)\n", + "print(vm_train_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Insert embeddings and questions into Vector DB\n", + "\n", + "Now that we have computed the embeddings for our question-answer pairs in the \"train\" set, we can go ahead and insert them into the vector store:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_community.vectorstores import Qdrant\n", + "from langchain_community.document_loaders import DataFrameLoader\n", + "\n", + "# load documents from dataframe\n", + "loader = DataFrameLoader(train_df, page_content_column=\"question\")\n", + "docs = loader.load()\n", + "\n", + "# setup vector datastore\n", + "qdrant = Qdrant.from_documents(\n", + " docs,\n", + " embedding_small_client,\n", + " location=\":memory:\", # Local mode with in-memory storage only\n", + " collection_name=\"rfp_rag_collection\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Retrieval Model\n", + "\n", + "Now that we have an embedding model and a vector database setup and loaded with our data, we need a Retrieval model that can search for similar question-answer pairs for a given input question. Once created, we can initialize this as a ValidMind model and `assign_predictions` to it just like our embedding model. In this example, we'll create two retrieval models with different `k` parameters (the number of documents retrieved) to benchmark and compare their performance. This approach allows us to evaluate how retrieval depth affects the overall system quality." + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": {}, + "outputs": [], + "source": [ + "def retrieve(input):\n", + " contexts = []\n", + "\n", + " for result in qdrant.similarity_search_with_score(input[\"question\"], k=5):\n", + " document, score = result\n", + " context = f\"Q: {document.page_content}\\n\"\n", + " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", + "\n", + " contexts.append(context)\n", + "\n", + " return contexts\n", + "\n", + "\n", + "vm_retriever_k5 = vm.init_model(input_id=\"retrieval_k5_model\", predict_fn=retrieve)" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": {}, + "outputs": [], + "source": [ + "def retrieve(input):\n", + " contexts = []\n", + "\n", + " for result in qdrant.similarity_search_with_score(input[\"question\"], k=10):\n", + " document, score = result\n", + " context = f\"Q: {document.page_content}\\n\"\n", + " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", + "\n", + " contexts.append(context)\n", + "\n", + " return contexts\n", + "\n", + "\n", + "vm_retriever_k10 = vm.init_model(input_id=\"retrieval_k10_model\", predict_fn=retrieve)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(model=vm_retriever_k5)\n", + "vm_test_ds.assign_predictions(model=vm_retriever_k10)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds._df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Generation Model\n", + "\n", + "As the final piece of this simple RAG pipeline, we can create and initialize a generation model that will use the retrieved context to generate an answer to the input question. We will use the `gpt-3.5-turbo` and `gpt-4o` models from OpenAI. Since we have two retrieval models (with different `k` values) and want to test two different LLMs, we'll create a total of four generator models - pairing each retrieval configuration with each LLM to comprehensively evaluate how both retrieval depth and model capability affect response quality." + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": {}, + "outputs": [], + "source": [ + "from openai import OpenAI\n", + "\n", + "from validmind.models import Prompt\n", + "\n", + "\n", + "system_prompt = \"\"\"\n", + "You are an expert RFP AI assistant.\n", + "You are tasked with answering new RFP questions based on existing RFP questions and answers.\n", + "You will be provided with the existing RFP questions and answer pairs that are the most relevant to the new RFP question.\n", + "After that you will be provided with a new RFP question.\n", + "You will generate an answer and respond only with the answer.\n", + "Ignore your pre-existing knowledge and answer the question based on the provided context.\n", + "\"\"\".strip()\n", + "\n", + "openai_client = OpenAI()" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "metadata": {}, + "outputs": [], + "source": [ + "def generate(input):\n", + " \n", + " response = openai_client.chat.completions.create(\n", + " model=\"gpt-3.5-turbo\",\n", + " messages=[\n", + " {\"role\": \"system\", \"content\": system_prompt},\n", + " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k5_model\"])},\n", + " {\"role\": \"user\", \"content\": input[\"question\"]},\n", + " ],\n", + " )\n", + " \n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "vm_generator_k5_gpt35 = vm.init_model(\n", + " input_id=\"generation_k5_gpt35_model\",\n", + " predict_fn=generate,\n", + " prompt=Prompt(template=system_prompt),\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": {}, + "outputs": [], + "source": [ + "def generate(input):\n", + " response = openai_client.chat.completions.create(\n", + " model=\"gpt-3.5-turbo\",\n", + " messages=[\n", + " {\"role\": \"system\", \"content\": system_prompt},\n", + " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k10_model\"])},\n", + " {\"role\": \"user\", \"content\": input[\"question\"]},\n", + " ],\n", + " )\n", + "\n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "vm_generator_k10_gpt35 = vm.init_model(\n", + " input_id=\"generation_k10_gpt35_model\",\n", + " predict_fn=generate,\n", + " prompt=Prompt(template=system_prompt),\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "metadata": {}, + "outputs": [], + "source": [ + "def generate(input):\n", + " \n", + " response = openai_client.chat.completions.create(\n", + " model=\"gpt-4o\",\n", + " messages=[\n", + " {\"role\": \"system\", \"content\": system_prompt},\n", + " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k5_model\"])},\n", + " {\"role\": \"user\", \"content\": input[\"question\"]},\n", + " ],\n", + " )\n", + " \n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "vm_generator_k5_gpt4o = vm.init_model(\n", + " input_id=\"generation_k5_gpt4o_model\",\n", + " predict_fn=generate,\n", + " prompt=Prompt(template=system_prompt),\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "def generate(input):\n", + " response = openai_client.chat.completions.create(\n", + " model=\"gpt-4o\",\n", + " messages=[\n", + " {\"role\": \"system\", \"content\": system_prompt},\n", + " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_k10_model\"])},\n", + " {\"role\": \"user\", \"content\": input[\"question\"]},\n", + " ],\n", + " )\n", + "\n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "vm_generator_k10_gpt4o = vm.init_model(\n", + " input_id=\"generation_k10_gpt4o_model\",\n", + " predict_fn=generate,\n", + " prompt=Prompt(template=system_prompt),\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's test it out real quick:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "\n", + "vm_generator_k5_gpt35.predict(\n", + " pd.DataFrame(\n", + " {\"retrieval_k5_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", + " )\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_generator_k5_gpt4o.predict(\n", + " pd.DataFrame(\n", + " {\"retrieval_k5_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", + " )\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Prompt Evaluation\n", + "\n", + "Now that we have our generator model initialized, we can run some LLM-as-Judge tests to evaluate the system prompt. This will allow us to get an initial sense of how well the prompt meets a few best practices for prompt engineering. These tests use an LLM to rate the prompt on a scale of 1-10 against the following criteria:\n", + "\n", + "- **Examplar Bias**: When using multi-shot prompting, does the prompt contain an unbiased distribution of examples?\n", + "- **Delimitation**: When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", + "- **Clarity**: How clearly the prompt states the task.\n", + "- **Conciseness**: How succinctly the prompt states the task.\n", + "- **Instruction Framing**: Whether the prompt contains negative instructions.\n", + "- **Specificity**: How specific the prompt defines the task." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Bias\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Clarity\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Conciseness\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Delimitation\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.NegativeInstruction\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Specificity\",\n", + " inputs={\n", + " \"model\": vm_generator_k5_gpt4o,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Setup RAG Pipeline Model\n", + "\n", + "Now that we have all of our individual \"component\" models setup and initialized we need some way to put them all together in a single \"pipeline\". We can use the `PipelineModel` class to do this. This ValidMind model type simply wraps any number of other ValidMind models and runs them in sequence. We can use a pipe(`|`) operator - in Python this is normally an `or` operator but we have overloaded it for easy pipeline creation - to chain together our models. We can then initialize this pipeline model and assign predictions to it just like any other model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_rag_k5_gpt35_model = vm.init_model(vm_retriever_k5 | vm_generator_k5_gpt35, input_id=\"rag_k5_gpt35_model\")\n", + "vm_rag_k10_gpt35_model = vm.init_model(vm_retriever_k10 | vm_generator_k10_gpt35, input_id=\"rag_k10_gpt35_model\")\n", + "vm_rag_k5_gpt4o_model = vm.init_model(vm_retriever_k5 | vm_generator_k5_gpt4o, input_id=\"rag_k5_gpt4o_model\")\n", + "vm_rag_k10_gpt4o_model = vm.init_model(vm_retriever_k10 | vm_generator_k10_gpt4o, input_id=\"rag_k10_gpt4o_model\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can `assign_predictions` to the pipeline model just like we did with the individual models. This will run the pipeline on the test set and store the results in the test set for later use." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(model=vm_rag_k5_gpt35_model)\n", + "vm_test_ds.assign_predictions(model=vm_rag_k10_gpt35_model)\n", + "vm_test_ds.assign_predictions(model=vm_rag_k5_gpt4o_model)\n", + "vm_test_ds.assign_predictions(model=vm_rag_k10_gpt4o_model)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds._df.head(5)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Run tests" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## RAGAS evaluation\n", + "\n", + "Let's go ahead and run some of our new RAG tests against our model...\n", + "\n", + "> Note: these tests are still being developed and are not yet in a stable state. We are using advanced tests here that use LLM-as-Judge and other strategies to assess things like the relevancy of the retrieved context to the input question and the correctness of the generated answer when compared to the ground truth. There is more to come in this area so stay tuned!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import warnings\n", + "\n", + "warnings.filterwarnings(\"ignore\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Semantic Similarity\n", + "\n", + "The concept of Answer Semantic Similarity pertains to the assessment of the semantic resemblance between the generated answer and the ground truth. This evaluation is based on the ground truth and the answer, with values falling within the range of 0 to 1. A higher score signifies a better alignment between the generated answer and the ground truth.\n", + "\n", + "Measuring the semantic similarity between answers can offer valuable insights into the quality of the generated response. This evaluation utilizes a cross-encoder model to calculate the semantic similarity score." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.SemanticSimilarity\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"reference_column\": [\"ground_truth\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Entity Recall\n", + "\n", + "This test gives the measure of recall of the retrieved context, based on the number of entities present in both ground_truths and contexts relative to the number of entities present in the ground_truths alone. Simply put, it is a measure of what fraction of entities are recalled from ground_truths. This test is useful in fact-based use cases like tourism help desk, historical QA, etc. This test can help evaluate the retrieval mechanism for entities, based on comparison with entities present in ground_truths, because in cases where entities matter, we need the contexts which cover them." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextEntityRecall\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"reference_column\": [\"ground_truth\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Precision\n", + "\n", + "Context Precision is a test that evaluates whether all of the ground-truth relevant items present in the contexts are ranked higher or not. Ideally all the relevant chunks must appear at the top ranks. This test is computed using the question, ground_truth and the contexts, with values ranging between 0 and 1, where higher scores indicate better precision." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextPrecision\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", + " \"reference_column\": [\"ground_truth\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Precision Without Reference\n", + "\n", + "This test evaluates whether retrieved contexts align well with the expected response for a given user input, without requiring a ground-truth reference. This test assesses the relevance of each retrieved context chunk by comparing it directly to the response." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid=[\n", + " {\"user_input_column\": \"question\",\n", + " \"retrieved_contexts_column\": \"retrieval_k5_model_prediction\",\n", + " \"response_column\": \"rag_k5_gpt4o_model_prediction\"\n", + " },\n", + " {\"user_input_column\": \"question\",\n", + " \"retrieved_contexts_column\": \"retrieval_k10_model_prediction\",\n", + " \"response_column\": \"rag_k10_gpt4o_model_prediction\"\n", + " },\n", + " ],\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Faithfulness\n", + "\n", + "This measures the factual consistency of the generated answer against the given context. It is calculated from answer and retrieved context. The answer is scaled to (0,1) range. Higher the better.\n", + "\n", + "The generated answer is regarded as faithful if all the claims that are made in the answer can be inferred from the given context. To calculate this a set of claims from the generated answer is first identified. Then each one of these claims are cross checked with given context to determine if it can be inferred from given context or not." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.Faithfulness\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Response Relevancy\n", + "\n", + "The Response Relevancy test, focuses on assessing how pertinent the generated answer is to the given prompt. A lower score is assigned to answers that are incomplete or contain redundant information and higher scores indicate better relevancy. This test is computed using the question, the context and the answer.\n", + "\n", + "The Response Relevancy is defined as the mean cosine similartiy of the original question to a number of artifical questions, which where generated (reverse engineered) based on the answer.\n", + "\n", + "Please note, that eventhough in practice the score will range between 0 and 1 most of the time, this is not mathematically guranteed, due to the nature of the cosine similarity ranging from -1 to 1.\n", + "\n", + "> Note: This is a reference free test. If you’re looking to compare ground truth answer with generated answer refer to Answer Correctness.\n", + "\n", + "An answer is deemed relevant when it directly and appropriately addresses the original question. Importantly, our assessment of answer relevance does not consider factuality but instead penalizes cases where the answer lacks completeness or contains redundant details. To calculate this score, the LLM is prompted to generate an appropriate question for the generated answer multiple times, and the mean cosine similarity between these generated questions and the original question is measured. The underlying idea is that if the generated answer accurately addresses the initial question, the LLM should be able to generate questions from the answer that align with the original question." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Recall\n", + "\n", + "Context recall measures the extent to which the retrieved context aligns with the annotated answer, treated as the ground truth. It is computed based on the ground truth and the retrieved context, and the values range between 0 and 1, with higher values indicating better performance.\n", + "\n", + "To estimate context recall from the ground truth answer, each sentence in the ground truth answer is analyzed to determine whether it can be attributed to the retrieved context or not. In an ideal scenario, all sentences in the ground truth answer should be attributable to the retrieved context." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextRecall\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\", \"retrieval_k10_model_prediction\"],\n", + " \"reference_column\": [\"ground_truth\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Answer Correctness\n", + "\n", + "The assessment of Answer Correctness involves gauging the accuracy of the generated answer when compared to the ground truth. This evaluation relies on the ground truth and the answer, with scores ranging from 0 to 1. A higher score indicates a closer alignment between the generated answer and the ground truth, signifying better correctness.\n", + "\n", + "Answer correctness encompasses two critical aspects: semantic similarity between the generated answer and the ground truth, as well as factual similarity. These aspects are combined using a weighted scheme to formulate the answer correctness score.\n", + "\n", + "Factual correctness quantifies the factual overlap between the generated answer and the ground truth answer. This is done using the concepts of:\n", + "\n", + "- TP (True Positive): Facts or statements that are present in both the ground truth and the generated answer.\n", + "- FP (False Positive): Facts or statements that are present in the generated answer but not in the ground truth.\n", + "- FN (False Negative): Facts or statements that are present in the ground truth but not in the generated answer." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.AnswerCorrectness\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"reference_column\": [\"ground_truth\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Aspect Critic\n", + "\n", + "This is designed to assess submissions based on predefined aspects such as harmlessness and correctness. Additionally, users have the flexibility to define their own aspects for evaluating submissions according to their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not. This evaluation is performed using the ‘answer’ as input.\n", + "\n", + "Critiques within the LLM evaluators evaluate submissions based on the provided aspect. Ragas Critiques offers a range of predefined aspects like correctness, harmfulness, etc. Users can also define their own aspects for evaluating submissions based on their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.AspectCritic\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Noise Sensitivity\n", + "\n", + "This test is designed to evaluate the robustness of the RAG pipeline model against noise in the retrieved context. It works by checking how well the \"claims\" in the generated answer match up with the \"claims\" in the ground truth answer. If the generated answer contains \"claims\" from the contexts that the ground truth answer does not contain, those claims are considered incorrect. The score for each answer is the number of incorrect claims divided by the total number of claims. This *can* be interpreted as a measure of how sensitive the LLM is to \"noise\" in the context where \"noise\" is information that is relevant but should not be included in the answer since the ground truth answer does not contain it." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.NoiseSensitivity\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " param_grid={\n", + " \"user_input_column\": [\"question\"],\n", + " \"response_column\": [\"rag_k5_gpt35_model_prediction\", \"rag_k5_gpt4o_model_prediction\"],\n", + " \"reference_column\": [\"ground_truth\"],\n", + " \"retrieved_contexts_column\": [\"retrieval_k5_model_prediction\"],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Generation quality\n", + "\n", + "In this section, we evaluate the alignment and relevance of generated responses to reference outputs within our retrieval-augmented generation (RAG) application. We use metrics that assess various quality dimensions of the generated responses, including semantic similarity, structural alignment, and phrasing overlap. Semantic similarity metrics compare embeddings of generated and reference text to capture deeper contextual alignment, while overlap and alignment measures quantify how well the phrasing and structure of generated responses match the intended outputs." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Token Disparity\n", + "\n", + "This test assesses the difference in token counts between the reference texts (ground truth) and the answers generated by the RAG model. It helps evaluate how well the model's outputs align with the expected length and level of detail in the reference texts. A significant disparity in token counts could signal issues with generation quality, such as excessive verbosity or insufficient detail. Consistently low token counts in generated answers compared to references might suggest that the model’s outputs are incomplete or overly concise, missing important contextual information." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.TokenDisparity\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### ROUGE Score\n", + "\n", + "This test evaluates the quality of answers generated by the RAG model by measuring overlaps in n-grams, word sequences, and word pairs between the model output and the reference (ground truth) text. ROUGE, short for Recall-Oriented Understudy for Gisting Evaluation, assesses both precision and recall, providing a balanced view of how well the generated response captures the reference content. ROUGE precision measures the proportion of n-grams in the generated text that match the reference, highlighting relevance and conciseness, while ROUGE recall assesses the proportion of reference n-grams present in the generated text, indicating completeness and thoroughness. \n", + "\n", + "Low precision scores might reveal that the generated text includes redundant or irrelevant information, while low recall scores suggest omissions of essential details from the reference. Consistently low ROUGE scores could indicate poor overall alignment with the ground truth, suggesting the model may be missing key content or failing to capture the intended meaning." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.RougeScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + " params={\n", + " \"metric\": \"rouge-1\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### BLEU Score\n", + "\n", + "The BLEU Score test evaluates the quality of answers generated by the RAG model by measuring n-gram overlap between the generated text and the reference (ground truth) text, with a specific focus on exact precision in phrasing. While ROUGE precision also assesses overlap, BLEU differs in two main ways: first, it applies a geometric average across multiple n-gram levels, capturing precise phrase alignment, and second, it includes a brevity penalty to prevent overly short outputs from inflating scores artificially. This added precision focus is valuable in RAG applications where strict adherence to reference language is essential, as BLEU emphasizes the match to exact phrasing. In contrast, ROUGE precision evaluates general content overlap without penalizing brevity, offering a broader sense of content alignment." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.BleuScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### BERT Score\n", + "\n", + "This test evaluates the quality of the RAG generated answers using BERT embeddings to measure precision, recall, and F1 scores based on semantic similarity, rather than exact n-gram matches as in BLEU and ROUGE. This approach captures contextual meaning, making it valuable when wording differs but the intended message closely aligns with the reference. In RAG applications, the BERT score is especially useful for ensuring that generated answers convey the reference text’s meaning, even if phrasing varies. Consistently low scores indicate a lack of semantic alignment, suggesting the model may miss or misrepresent key content. Low precision may reflect irrelevant or redundant details, while low recall can indicate omissions." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.BertScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### METEOR Score\n", + "\n", + "This test evaluates the quality of the generated answers by measuring alignment with the ground truth, emphasizing both accuracy and fluency. Unlike BLEU and ROUGE, which focus on n-gram matches, METEOR combines precision, recall, synonym matching, and word order, focusing at how well the generated text conveys meaning and reads naturally. This metric is especially useful for RAG applications where sentence structure and natural flow are crucial for clear communication. Lower scores may suggest alignment issues, indicating that the answers may lack fluency or key content. Discrepancies in word order or high fragmentation penalties can reveal problems with how the model constructs sentences, potentially affecting readability." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.MeteorScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Bias and Toxicity\n", + "\n", + "In this section, we use metrics like Toxicity Score and Regard Score to evaluate both the generated responses and the ground truth. These tests helps us detect any harmful, offensive, or inappropriate language and evaluate the level of bias and neutrality enabling us to assess and mitigate potential biases in both the model's responses and the original dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Toxicity Score\n", + "\n", + "This test measures the level of harmful or offensive content in the generated answers. The test uses a preloaded toxicity detection tool from Hugging Face, which identifies language that may be inappropriate, aggressive, or derogatory. High toxicity scores indicate potentially toxic content, while consistently elevated scores across multiple outputs may signal underlying issues in the model’s generation process that require attention to prevent the spread of harmful language." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ToxicityScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Regard Score\n", + "\n", + "This test evaluates the sentiment and perceived regard—categorized as positive, negative, neutral, or other—in answers generated by the RAG model. This is important for identifying any biases or sentiment tendencies in responses, ensuring that generated answers are balanced and appropriate for the context. The uses a preloaded regard evaluation tool from Hugging Face to compute scores for each response. High skewness in regard scores, especially if the generated responses consistently diverge from expected sentiments in the reference texts, may reveal biases in the model’s generation, such as overly positive or negative tones where neutrality is expected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.RegardScore\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_rag_k5_gpt35_model, vm_rag_k5_gpt4o_model],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Conclusion\n", + "\n", + "In this notebook, we have seen how we can use LangChain and ValidMind together to build, evaluate and document a simple RAG Model as its developed. This is a great example of the interactive development experience that ValidMind is designed to support. We can quickly iterate on our model and document as we go... We have seen how ValidMind supports non-traditional \"models\" using a functional interface and how we can build pipelines of many models to support complex GenAI workflows.\n", + "\n", + "This is still a work in progress and we are actively developing new tests to support more advanced GenAI workflows. We are also keeping an eye on the most popular GenAI models and libraries to explore direct integrations. Stay tuned for more updates and new features in this area!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Upgrade ValidMind\n", + "\n", + "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", + "\n", + "Retrieve the information for the currently installed version of ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", + "\n", + "```bash\n", + "%pip install --upgrade validmind\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-09e315440ca84258abe1aaefaca3a3d0", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "ValidMind Library", + "language": "python", + "name": "validmind" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.15" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/site/notebooks/use_cases/nlp_and_llm/rag_documentation_demo.ipynb b/site/notebooks/use_cases/nlp_and_llm/rag_documentation_demo.ipynb new file mode 100644 index 0000000000..aac1876d5d --- /dev/null +++ b/site/notebooks/use_cases/nlp_and_llm/rag_documentation_demo.ipynb @@ -0,0 +1,1692 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# RAG Model Documentation Demo\n", + "\n", + "In this notebook, we are going to implement a simple RAG Model for automating the process of answering RFP questions using GenAI. We will see how we can initialize an embedding model, a retrieval model and a generator model with LangChain components and use them within the ValidMind Library to run tests against them. Finally, we will see how we can put them together in a Pipeline and run that to get e2e results and run tests against that." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [About ValidMind](#toc1__) \n", + " - [Before you begin](#toc1_1__) \n", + " - [New to ValidMind?](#toc1_2__) \n", + " - [Key concepts](#toc1_3__) \n", + "- [Setting up](#toc2__) \n", + " - [Initialize the ValidMind Library](#toc2_1__) \n", + " - [Register sample model](#toc2_1_1__) \n", + " - [Apply documentation template](#toc2_1_2__) \n", + " - [Get your code snippet](#toc2_1_3__) \n", + "- [Read Open AI API Key](#toc3__) \n", + "- [Dataset Loader](#toc4__) \n", + "- [Data validation](#toc5__) \n", + " - [Duplicates](#toc5_1__) \n", + " - [Stop Words](#toc5_2__) \n", + " - [Punctuations](#toc5_3__) \n", + " - [Common Words](#toc5_4__) \n", + " - [Language Detection](#toc5_5__) \n", + " - [Toxicity Score](#toc5_6__) \n", + " - [Polarity and Subjectivity](#toc5_7__) \n", + " - [Sentiment](#toc5_8__) \n", + " - [Assign Predictions](#toc5_9__) \n", + " - [Run tests](#toc5_10__) \n", + " - [Generate embeddings for the Train Set](#toc5_11__) \n", + " - [Insert embeddings and questions into Vector DB](#toc5_12__) \n", + "- [Prompt Evaluation](#toc6__) \n", + "- [RAGAS evaluation](#toc7__) \n", + " - [Semantic Similarity](#toc7_1__) \n", + " - [Context Entity Recall](#toc7_2__) \n", + " - [Context Precision](#toc7_3__) \n", + " - [Context Precision Without Reference](#toc7_4__) \n", + " - [Faithfulness](#toc7_5__) \n", + " - [Response Relevancy](#toc7_6__) \n", + " - [Context Recall](#toc7_7__) \n", + " - [Answer Correctness](#toc7_8__) \n", + " - [Aspect Critic](#toc7_9__) \n", + " - [Noise Sensitivity](#toc7_10__) \n", + "- [Generation quality](#toc8__) \n", + " - [Token Disparity](#toc8_1__) \n", + " - [ROUGE Score](#toc8_2__) \n", + " - [BLEU Score](#toc8_3__) \n", + " - [BERT Score](#toc8_4__) \n", + " - [METEOR Score](#toc8_5__) \n", + "- [Bias and Toxicity](#toc9__) \n", + " - [Toxicity Score](#toc9_1__) \n", + " - [Regard Score](#toc9_2__) \n", + "- [Upgrade ValidMind](#toc10__) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## About ValidMind\n", + "\n", + "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models.\n", + "\n", + "You use the ValidMind Library to automate documentation and validation tests, and then use the ValidMind Platform to collaborate on model documentation. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model validators.\n", + "\n", + "\n", + "\n", + "### Before you begin\n", + "\n", + "This notebook assumes you have basic familiarity with Python, including an understanding of how functions work. If you are new to Python, you can still run the notebook but we recommend further familiarizing yourself with the language. \n", + "\n", + "If you encounter errors due to missing modules in your Python environment, install the modules with `pip install`, and then re-run the notebook. For more help, refer to [Installing Python Modules](https://docs.python.org/3/installing/index.html).\n", + "\n", + "\n", + "\n", + "### New to ValidMind?\n", + "\n", + "If you haven't already seen our documentation on the [ValidMind Library](https://docs.validmind.ai/developer/validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models and running tests, as well as find code samples and our Python Library API reference.\n", + "\n", + "
For access to all features available in this notebook, you'll need access to a ValidMind account.\n", + "

\n", + "Register with ValidMind
\n", + "\n", + "\n", + "\n", + "### Key concepts\n", + "\n", + "- **FunctionModels**: ValidMind offers support for creating `VMModel` instances from Python functions. This enables us to support any \"model\" by simply using the provided function as the model's `predict` method.\n", + "- **PipelineModels**: ValidMind models (`VMModel` instances) of any type can be piped together to create a model pipeline. This allows model components to be created and tested/documented independently, and then combined into a single model for end-to-end testing and documentation. We use the `|` operator to pipe models together.\n", + "- **RAG**: RAG stands for Retrieval Augmented Generation and refers to a wide range of GenAI applications where some form of retrieval is used to add context to the prompt so that the LLM that generates content can refer to it when creating its output. In this notebook, we are going to implement a simple RAG setup using LangChain components." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Prerequisites\n", + "\n", + "Let's go ahead and install the `validmind` library if its not already installed... Then we can install the `qdrant-client` library for our vector store and `langchain` for everything else:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q qdrant-client langchain langchain-openai sentencepiece" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize the ValidMind Library" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Register sample model\n", + "\n", + "Let's first register a sample model for use with this notebook:\n", + "\n", + "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", + "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "4. Select your own name under the **MODEL OWNER** drop-down.\n", + "\n", + "5. Click **Register Model** to add the model to your inventory." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Apply documentation template\n", + "\n", + "Once you've registered your model, let's select a documentation template. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", + "\n", + "2. Under **TEMPLATE**, select `Gen AI RAG`.\n", + "\n", + "3. Click **Use Template** to apply the template." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Can't select this template?\n", + "

\n", + "Your organization administrators may need to add it to your template library:\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "#### Get your code snippet\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your developer environment. You initialize the ValidMind Library with this code snippet, which ensures that your documentation and tests are uploaded to the correct model when you run the notebook.\n", + "\n", + "1. On the left sidebar that appears for your model, select **Getting Started** and click **Copy snippet to clipboard**.\n", + "2. Next, [load your model identifier credentials from an `.env` file](https://docs.validmind.ai/developer/model-documentation/store-credentials-in-env-file.html) or replace the placeholder with your own code snippet:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Load your model identifier credentials from an `.env` file\n", + "\n", + "%load_ext dotenv\n", + "%dotenv .env\n", + "\n", + "# Or replace with your code snippet\n", + "\n", + "import validmind as vm\n", + "\n", + "vm.init(\n", + " # api_host=\"...\",\n", + " # api_key=\"...\",\n", + " # api_secret=\"...\",\n", + " # model=\"...\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Read Open AI API Key\n", + "\n", + "We will need to have an OpenAI API key to be able to use their `text-embedding-3-small` model for our embeddings, `gpt-3.5-turbo` model for our generator and `gpt-4o` model for our LLM-as-Judge tests. If you don't have an OpenAI API key, you can get one by signing up at [OpenAI](https://platform.openai.com/signup). Then you can create a `.env` file in the root of your project and the following cell will load it from there. Alternatively, you can just uncomment the line below to directly set the key (not recommended for security reasons)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# load openai api key\n", + "import os\n", + "\n", + "import dotenv\n", + "import nltk\n", + "\n", + "dotenv.load_dotenv()\n", + "nltk.download('stopwords')\n", + "nltk.download('punkt_tab')\n", + "\n", + "# os.environ[\"OPENAI_API_KEY\"] = \"sk-...\"\n", + "\n", + "if not \"OPENAI_API_KEY\" in os.environ:\n", + " raise ValueError(\"OPENAI_API_KEY is not set\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Dataset Loader\n", + "\n", + "Great, now that we have all of our dependencies installed, the ValidMind Library initialized and connected to our model and our OpenAI API key setup, we can go ahead and load our datasets. We will use the synthetic `RFP` dataset included with ValidMind for this notebook. This dataset contains a variety of RFP questions and ground truth answers that we can use both as the source where our Retriever will search for similar question-answer pairs as well as our test set for evaluating the performance of our RAG model. To do this, we just have to load it and call the preprocess function to get a split of the data into train and test sets." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the sample dataset from the library\n", + "from validmind.datasets.llm.rag import rfp\n", + "\n", + "raw_df = rfp.load_data()\n", + "train_df, test_df = rfp.preprocess(raw_df)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_train_ds = vm.init_dataset(\n", + " train_df,\n", + " text_column=\"question\",\n", + " target_column=\"ground_truth\",\n", + ")\n", + "\n", + "vm_test_ds = vm.init_dataset(\n", + " test_df,\n", + " text_column=\"question\",\n", + " target_column=\"ground_truth\",\n", + ")\n", + "\n", + "vm_test_ds.df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Data validation\n", + "\n", + "Now that we have loaded our dataset, we can go ahead and run some data validation tests right away to start assessing and documenting the quality of our data. Since we are using a text dataset, we can use ValidMind's built-in array of text data quality tests to check that things like number of duplicates, missing values, and other common text data issues are not present in our dataset. We can also run some tests to check the sentiment and toxicity of our data." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Duplicates\n", + "\n", + "First, let's check for duplicates in our dataset. We can use the `validmind.data_validation.Duplicates` test and pass our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.tests import run_test\n", + "\n", + "run_test(\n", + " test_id=\"validmind.data_validation.Duplicates\",\n", + " inputs={\"dataset\": vm_train_ds},\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Stop Words\n", + "\n", + "Next, let's check for stop words in our dataset. We can use the `validmind.data_validation.StopWords` test and pass our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.StopWords\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Punctuations\n", + "\n", + "Next, let's check for punctuations in our dataset. We can use the `validmind.data_validation.Punctuations` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.Punctuations\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Common Words\n", + "\n", + "Next, let's check for common words in our dataset. We can use the `validmind.data_validation.CommonWord` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.CommonWords\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Language Detection\n", + "\n", + "For documentation purposes, we can detect and log the languages used in the dataset with the `validmind.data_validation.LanguageDetection` test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " test_id=\"validmind.data_validation.nlp.LanguageDetection\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Toxicity Score\n", + "\n", + "Now, let's go ahead and run the `validmind.data_validation.nlp.Toxicity` test to compute a toxicity score for our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.Toxicity\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Polarity and Subjectivity\n", + "\n", + "We can also run the `validmind.data_validation.nlp.PolarityAndSubjectivity` test to compute the polarity and subjectivity of our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.PolarityAndSubjectivity\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Sentiment\n", + "\n", + "Finally, we can run the `validmind.data_validation.nlp.Sentiment` test to plot the sentiment of our dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.data_validation.nlp.Sentiment\",\n", + " inputs={\n", + " \"dataset\": vm_train_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Embedding Model\n", + "\n", + "Now that we have our dataset loaded and have run some data validation tests to assess and document the quality of our data, we can go ahead and initialize our embedding model. We will use the `text-embedding-3-small` model from OpenAI for this purpose wrapped in the `OpenAIEmbeddings` class from LangChain. This model will be used to \"embed\" our questions both for inserting the question-answer pairs from the \"train\" set into the vector store and for embedding the question from inputs when making predictions with our RAG model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_openai import OpenAIEmbeddings\n", + "\n", + "embedding_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", + "\n", + "\n", + "def embed(input):\n", + " \"\"\"Returns a text embedding for the given text\"\"\"\n", + " return embedding_client.embed_query(input[\"question\"])\n", + "\n", + "\n", + "vm_embedder = vm.init_model(input_id=\"embedding_model\", predict_fn=embed)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "What we have done here is to initialize the `OpenAIEmbeddings` class so it uses OpenAI's `text-embedding-3-small` model. We then created an `embed` function that takes in an `input` dictionary and uses the `embed_query` method of the embedding client to compute the embeddings of the `question`. We use an `embed` function since that is how ValidMind supports any custom model. We will use this strategy for the retrieval and generator models as well but you could also use, say, a HuggingFace model directly. See the documentation for more information on which model types are directly supported - [ValidMind Documentation](https://docs.validmind.ai/validmind/validmind.html)... Finally, we use the `init_model` function from the ValidMind Library to create a `VMModel` object that can be used in ValidMind tests. This also logs the model to our model documentation and any test that uses the model will be linked to the logged model and its metadata." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign Predictions\n", + "\n", + "To precompute the embeddings for our test set, we can call the `assign_predictions` method of our `vm_test_ds` object we created above. This will compute the embeddings for each question in the test set and store them in the a special prediction column of the test set thats linked to our `vm_embedder` model. This will allow us to use these embeddings later when we run tests against our embedding model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(vm_embedder)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Run tests\n", + "\n", + "Now that everything is setup for the embedding model, we can go ahead and run some tests to assess and document the quality of our embeddings. We will use the `validmind.model_validation.embeddings.*` tests to compute a variety of metrics against our model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisRandomNoise\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " params={\"probability\": 0.3},\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisSynonyms\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " params={\"probability\": 0.3},\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.StabilityAnalysisTranslation\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " params={\n", + " \"source_lang\": \"en\",\n", + " \"target_lang\": \"fr\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.CosineSimilarityHeatmap\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "run_test(\n", + " \"validmind.model_validation.embeddings.CosineSimilarityDistribution\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.EuclideanDistanceHeatmap\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.PCAComponentsPairwisePlots\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " params={\"n_components\": 3},\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.embeddings.TSNEComponentsPairwisePlots\",\n", + " inputs={\n", + " \"model\": vm_embedder,\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " params={\"n_components\": 3, \"perplexity\": 20},\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Setup Vector Store\n", + "\n", + "Great, so now that we have assessed our embedding model and verified that it is performing well, we can go ahead and use it to compute embeddings for our question-answer pairs in the \"train\" set. We will then use these embeddings to insert the question-answer pairs into a vector store. We will use an in-memory `qdrant` vector database for demo purposes but any option would work just as well here. We will use the `QdrantClient` class from LangChain to interact with the vector store. This class will allow us to insert and search for embeddings in the vector store." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Generate embeddings for the Train Set\n", + "\n", + "We can use the same `assign_predictions` method from earlier except this time we will use the `vm_train_ds` object to compute the embeddings for the question-answer pairs in the \"train\" set." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_train_ds.assign_predictions(vm_embedder)\n", + "print(vm_train_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Insert embeddings and questions into Vector DB\n", + "\n", + "Now that we have computed the embeddings for our question-answer pairs in the \"train\" set, we can go ahead and insert them into the vector store:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_community.vectorstores import Qdrant\n", + "from langchain_openai import OpenAIEmbeddings\n", + "from langchain_community.document_loaders import DataFrameLoader\n", + "\n", + "# load documents from dataframe\n", + "loader = DataFrameLoader(train_df, page_content_column=\"question\")\n", + "docs = loader.load()\n", + "# choose model using embedding client\n", + "embedding_client = OpenAIEmbeddings(model=\"text-embedding-3-small\")\n", + "\n", + "# setup vector datastore\n", + "qdrant = Qdrant.from_documents(\n", + " docs,\n", + " embedding_client,\n", + " location=\":memory:\", # Local mode with in-memory storage only\n", + " collection_name=\"rfp_rag_collection\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Retrieval Model\n", + "\n", + "Now that we have an embedding model and a vector database setup and loaded with our data, we need a Retrieval model that can search for similar question-answer pairs for a given input question. Once created, we can initialize this as a ValidMind model and `assign_predictions` to it just like our embedding model." + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": {}, + "outputs": [], + "source": [ + "def retrieve(input):\n", + " contexts = []\n", + "\n", + " for result in qdrant.similarity_search_with_score(input[\"question\"]):\n", + " document, score = result\n", + " context = f\"Q: {document.page_content}\\n\"\n", + " context += f\"A: {document.metadata['ground_truth']}\\n\"\n", + "\n", + " contexts.append(context)\n", + "\n", + " return contexts\n", + "\n", + "\n", + "vm_retriever = vm.init_model(input_id=\"retrieval_model\", predict_fn=retrieve)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(model=vm_retriever)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Generation Model\n", + "\n", + "As the final piece of this simple RAG pipeline, we can create and initialize a generation model that will use the retrieved context to generate an answer to the input question. We will use the `gpt-3.5-turbo` model from OpenAI." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from openai import OpenAI\n", + "\n", + "from validmind.models import Prompt\n", + "\n", + "\n", + "system_prompt = \"\"\"\n", + "You are an expert RFP AI assistant.\n", + "You are tasked with answering new RFP questions based on existing RFP questions and answers.\n", + "You will be provided with the existing RFP questions and answer pairs that are the most relevant to the new RFP question.\n", + "After that you will be provided with a new RFP question.\n", + "You will generate an answer and respond only with the answer.\n", + "Ignore your pre-existing knowledge and answer the question based on the provided context.\n", + "\"\"\".strip()\n", + "\n", + "openai_client = OpenAI()\n", + "\n", + "\n", + "def generate(input):\n", + " response = openai_client.chat.completions.create(\n", + " model=\"gpt-3.5-turbo\",\n", + " messages=[\n", + " {\"role\": \"system\", \"content\": system_prompt},\n", + " {\"role\": \"user\", \"content\": \"\\n\\n\".join(input[\"retrieval_model\"])},\n", + " {\"role\": \"user\", \"content\": input[\"question\"]},\n", + " ],\n", + " )\n", + "\n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "vm_generator = vm.init_model(\n", + " input_id=\"generation_model\",\n", + " predict_fn=generate,\n", + " prompt=Prompt(template=system_prompt),\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's test it out real quick:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "\n", + "vm_generator.predict(\n", + " pd.DataFrame(\n", + " {\"retrieval_model\": [[\"My name is anil\"]], \"question\": [\"what is my name\"]}\n", + " )\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Prompt Evaluation\n", + "\n", + "Now that we have our generator model initialized, we can run some LLM-as-Judge tests to evaluate the system prompt. This will allow us to get an initial sense of how well the prompt meets a few best practices for prompt engineering. These tests use an LLM to rate the prompt on a scale of 1-10 against the following criteria:\n", + "\n", + "- **Examplar Bias**: When using multi-shot prompting, does the prompt contain an unbiased distribution of examples?\n", + "- **Delimitation**: When using complex prompts containing examples, contextual information, or other elements, is the prompt formatted in such a way that each element is clearly separated?\n", + "- **Clarity**: How clearly the prompt states the task.\n", + "- **Conciseness**: How succinctly the prompt states the task.\n", + "- **Instruction Framing**: Whether the prompt contains negative instructions.\n", + "- **Specificity**: How specific the prompt defines the task." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Bias\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Clarity\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Conciseness\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Delimitation\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.NegativeInstruction\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.prompt_validation.Specificity\",\n", + " inputs={\n", + " \"model\": vm_generator,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Setup RAG Pipeline Model\n", + "\n", + "Now that we have all of our individual \"component\" models setup and initialized we need some way to put them all together in a single \"pipeline\". We can use the `PipelineModel` class to do this. This ValidMind model type simply wraps any number of other ValidMind models and runs them in sequence. We can use a pipe(`|`) operator - in Python this is normally an `or` operator but we have overloaded it for easy pipeline creation - to chain together our models. We can then initialize this pipeline model and assign predictions to it just like any other model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_rag_model = vm.init_model(vm_retriever | vm_generator, input_id=\"rag_model\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can `assign_predictions` to the pipeline model just like we did with the individual models. This will run the pipeline on the test set and store the results in the test set for later use." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds.assign_predictions(model=vm_rag_model)\n", + "print(vm_test_ds)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_test_ds._df.head(5)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Run tests" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## RAGAS evaluation\n", + "\n", + "Let's go ahead and run some of our new RAG tests against our model...\n", + "\n", + "> Note: these tests are still being developed and are not yet in a stable state. We are using advanced tests here that use LLM-as-Judge and other strategies to assess things like the relevancy of the retrieved context to the input question and the correctness of the generated answer when compared to the ground truth. There is more to come in this area so stay tuned!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import warnings\n", + "\n", + "warnings.filterwarnings(\"ignore\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Semantic Similarity\n", + "\n", + "The concept of Answer Semantic Similarity pertains to the assessment of the semantic resemblance between the generated answer and the ground truth. This evaluation is based on the ground truth and the answer, with values falling within the range of 0 to 1. A higher score signifies a better alignment between the generated answer and the ground truth.\n", + "\n", + "Measuring the semantic similarity between answers can offer valuable insights into the quality of the generated response. This evaluation utilizes a cross-encoder model to calculate the semantic similarity score." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.SemanticSimilarity\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"reference_column\": \"ground_truth\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Entity Recall\n", + "\n", + "This test gives the measure of recall of the retrieved context, based on the number of entities present in both ground_truths and contexts relative to the number of entities present in the ground_truths alone. Simply put, it is a measure of what fraction of entities are recalled from ground_truths. This test is useful in fact-based use cases like tourism help desk, historical QA, etc. This test can help evaluate the retrieval mechanism for entities, based on comparison with entities present in ground_truths, because in cases where entities matter, we need the contexts which cover them." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextEntityRecall\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"reference_column\": \"ground_truth\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Precision\n", + "\n", + "Context Precision is a test that evaluates whether all of the ground-truth relevant items present in the contexts are ranked higher or not. Ideally all the relevant chunks must appear at the top ranks. This test is computed using the question, ground_truth and the contexts, with values ranging between 0 and 1, where higher scores indicate better precision." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextPrecision\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " \"reference_column\": \"ground_truth\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Precision Without Reference\n", + "\n", + "This test evaluates whether retrieved contexts align well with the expected response for a given user input, without requiring a ground-truth reference. This test assesses the relevance of each retrieved context chunk by comparing it directly to the response." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextPrecisionWithoutReference\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Faithfulness\n", + "\n", + "This measures the factual consistency of the generated answer against the given context. It is calculated from answer and retrieved context. The answer is scaled to (0,1) range. Higher the better.\n", + "\n", + "The generated answer is regarded as faithful if all the claims that are made in the answer can be inferred from the given context. To calculate this a set of claims from the generated answer is first identified. Then each one of these claims are cross checked with given context to determine if it can be inferred from given context or not." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.Faithfulness\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Response Relevancy\n", + "\n", + "The Response Relevancy test, focuses on assessing how pertinent the generated answer is to the given prompt. A lower score is assigned to answers that are incomplete or contain redundant information and higher scores indicate better relevancy. This test is computed using the question, the context and the answer.\n", + "\n", + "The Response Relevancy is defined as the mean cosine similartiy of the original question to a number of artifical questions, which where generated (reverse engineered) based on the answer.\n", + "\n", + "Please note, that eventhough in practice the score will range between 0 and 1 most of the time, this is not mathematically guranteed, due to the nature of the cosine similarity ranging from -1 to 1.\n", + "\n", + "> Note: This is a reference free test. If you’re looking to compare ground truth answer with generated answer refer to Answer Correctness.\n", + "\n", + "An answer is deemed relevant when it directly and appropriately addresses the original question. Importantly, our assessment of answer relevance does not consider factuality but instead penalizes cases where the answer lacks completeness or contains redundant details. To calculate this score, the LLM is prompted to generate an appropriate question for the generated answer multiple times, and the mean cosine similarity between these generated questions and the original question is measured. The underlying idea is that if the generated answer accurately addresses the initial question, the LLM should be able to generate questions from the answer that align with the original question." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ResponseRelevancy\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Context Recall\n", + "\n", + "Context recall measures the extent to which the retrieved context aligns with the annotated answer, treated as the ground truth. It is computed based on the ground truth and the retrieved context, and the values range between 0 and 1, with higher values indicating better performance.\n", + "\n", + "To estimate context recall from the ground truth answer, each sentence in the ground truth answer is analyzed to determine whether it can be attributed to the retrieved context or not. In an ideal scenario, all sentences in the ground truth answer should be attributable to the retrieved context." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.ContextRecall\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " \"reference_column\": \"ground_truth\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Answer Correctness\n", + "\n", + "The assessment of Answer Correctness involves gauging the accuracy of the generated answer when compared to the ground truth. This evaluation relies on the ground truth and the answer, with scores ranging from 0 to 1. A higher score indicates a closer alignment between the generated answer and the ground truth, signifying better correctness.\n", + "\n", + "Answer correctness encompasses two critical aspects: semantic similarity between the generated answer and the ground truth, as well as factual similarity. These aspects are combined using a weighted scheme to formulate the answer correctness score.\n", + "\n", + "Factual correctness quantifies the factual overlap between the generated answer and the ground truth answer. This is done using the concepts of:\n", + "\n", + "- TP (True Positive): Facts or statements that are present in both the ground truth and the generated answer.\n", + "- FP (False Positive): Facts or statements that are present in the generated answer but not in the ground truth.\n", + "- FN (False Negative): Facts or statements that are present in the ground truth but not in the generated answer." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.AnswerCorrectness\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"reference_column\": \"ground_truth\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Aspect Critic\n", + "\n", + "This is designed to assess submissions based on predefined aspects such as harmlessness and correctness. Additionally, users have the flexibility to define their own aspects for evaluating submissions according to their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not. This evaluation is performed using the ‘answer’ as input.\n", + "\n", + "Critiques within the LLM evaluators evaluate submissions based on the provided aspect. Ragas Critiques offers a range of predefined aspects like correctness, harmfulness, etc. Users can also define their own aspects for evaluating submissions based on their specific criteria. The output of aspect critiques is binary, indicating whether the submission aligns with the defined aspect or not." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.AspectCritic\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Noise Sensitivity\n", + "\n", + "This test is designed to evaluate the robustness of the RAG pipeline model against noise in the retrieved context. It works by checking how well the \"claims\" in the generated answer match up with the \"claims\" in the ground truth answer. If the generated answer contains \"claims\" from the contexts that the ground truth answer does not contain, those claims are considered incorrect. The score for each answer is the number of incorrect claims divided by the total number of claims. This *can* be interpreted as a measure of how sensitive the LLM is to \"noise\" in the context where \"noise\" is information that is relevant but should not be included in the answer since the ground truth answer does not contain it." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ragas.NoiseSensitivity\",\n", + " inputs={\"dataset\": vm_test_ds},\n", + " params={\n", + " \"user_input_column\": \"question\",\n", + " \"response_column\": \"rag_model_prediction\",\n", + " \"reference_column\": \"ground_truth\",\n", + " \"retrieved_contexts_column\": \"retrieval_model_prediction\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Generation quality\n", + "\n", + "In this section, we evaluate the alignment and relevance of generated responses to reference outputs within our retrieval-augmented generation (RAG) application. We use metrics that assess various quality dimensions of the generated responses, including semantic similarity, structural alignment, and phrasing overlap. Semantic similarity metrics compare embeddings of generated and reference text to capture deeper contextual alignment, while overlap and alignment measures quantify how well the phrasing and structure of generated responses match the intended outputs." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Token Disparity\n", + "\n", + "This test assesses the difference in token counts between the reference texts (ground truth) and the answers generated by the RAG model. It helps evaluate how well the model's outputs align with the expected length and level of detail in the reference texts. A significant disparity in token counts could signal issues with generation quality, such as excessive verbosity or insufficient detail. Consistently low token counts in generated answers compared to references might suggest that the model’s outputs are incomplete or overly concise, missing important contextual information." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.TokenDisparity\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### ROUGE Score\n", + "\n", + "This test evaluates the quality of answers generated by the RAG model by measuring overlaps in n-grams, word sequences, and word pairs between the model output and the reference (ground truth) text. ROUGE, short for Recall-Oriented Understudy for Gisting Evaluation, assesses both precision and recall, providing a balanced view of how well the generated response captures the reference content. ROUGE precision measures the proportion of n-grams in the generated text that match the reference, highlighting relevance and conciseness, while ROUGE recall assesses the proportion of reference n-grams present in the generated text, indicating completeness and thoroughness. \n", + "\n", + "Low precision scores might reveal that the generated text includes redundant or irrelevant information, while low recall scores suggest omissions of essential details from the reference. Consistently low ROUGE scores could indicate poor overall alignment with the ground truth, suggesting the model may be missing key content or failing to capture the intended meaning." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.RougeScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + " params={\n", + " \"metric\": \"rouge-1\",\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### BLEU Score\n", + "\n", + "The BLEU Score test evaluates the quality of answers generated by the RAG model by measuring n-gram overlap between the generated text and the reference (ground truth) text, with a specific focus on exact precision in phrasing. While ROUGE precision also assesses overlap, BLEU differs in two main ways: first, it applies a geometric average across multiple n-gram levels, capturing precise phrase alignment, and second, it includes a brevity penalty to prevent overly short outputs from inflating scores artificially. This added precision focus is valuable in RAG applications where strict adherence to reference language is essential, as BLEU emphasizes the match to exact phrasing. In contrast, ROUGE precision evaluates general content overlap without penalizing brevity, offering a broader sense of content alignment." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.BleuScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### BERT Score\n", + "\n", + "This test evaluates the quality of the RAG generated answers using BERT embeddings to measure precision, recall, and F1 scores based on semantic similarity, rather than exact n-gram matches as in BLEU and ROUGE. This approach captures contextual meaning, making it valuable when wording differs but the intended message closely aligns with the reference. In RAG applications, the BERT score is especially useful for ensuring that generated answers convey the reference text’s meaning, even if phrasing varies. Consistently low scores indicate a lack of semantic alignment, suggesting the model may miss or misrepresent key content. Low precision may reflect irrelevant or redundant details, while low recall can indicate omissions." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.BertScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### METEOR Score\n", + "\n", + "This test evaluates the quality of the generated answers by measuring alignment with the ground truth, emphasizing both accuracy and fluency. Unlike BLEU and ROUGE, which focus on n-gram matches, METEOR combines precision, recall, synonym matching, and word order, focusing at how well the generated text conveys meaning and reads naturally. This metric is especially useful for RAG applications where sentence structure and natural flow are crucial for clear communication. Lower scores may suggest alignment issues, indicating that the answers may lack fluency or key content. Discrepancies in word order or high fragmentation penalties can reveal problems with how the model constructs sentences, potentially affecting readability." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.MeteorScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Bias and Toxicity\n", + "\n", + "In this section, we use metrics like Toxicity Score and Regard Score to evaluate both the generated responses and the ground truth. These tests helps us detect any harmful, offensive, or inappropriate language and evaluate the level of bias and neutrality enabling us to assess and mitigate potential biases in both the model's responses and the original dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Toxicity Score\n", + "\n", + "This test measures the level of harmful or offensive content in the generated answers. The test uses a preloaded toxicity detection tool from Hugging Face, which identifies language that may be inappropriate, aggressive, or derogatory. High toxicity scores indicate potentially toxic content, while consistently elevated scores across multiple outputs may signal underlying issues in the model’s generation process that require attention to prevent the spread of harmful language." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.ToxicityScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Regard Score\n", + "\n", + "This test evaluates the sentiment and perceived regard—categorized as positive, negative, neutral, or other—in answers generated by the RAG model. This is important for identifying any biases or sentiment tendencies in responses, ensuring that generated answers are balanced and appropriate for the context. The uses a preloaded regard evaluation tool from Hugging Face to compute scores for each response. High skewness in regard scores, especially if the generated responses consistently diverge from expected sentiments in the reference texts, may reveal biases in the model’s generation, such as overly positive or negative tones where neutrality is expected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "run_test(\n", + " \"validmind.model_validation.RegardScore\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " \"model\": vm_rag_model,\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Conclusion\n", + "\n", + "In this notebook, we have seen how we can use LangChain and ValidMind together to build, evaluate and document a simple RAG Model as its developed. This is a great example of the interactive development experience that ValidMind is designed to support. We can quickly iterate on our model and document as we go... We have seen how ValidMind supports non-traditional \"models\" using a functional interface and how we can build pipelines of many models to support complex GenAI workflows.\n", + "\n", + "This is still a work in progress and we are actively developing new tests to support more advanced GenAI workflows. We are also keeping an eye on the most popular GenAI models and libraries to explore direct integrations. Stay tuned for more updates and new features in this area!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Upgrade ValidMind\n", + "\n", + "
After installing ValidMind, you’ll want to periodically make sure you are on the latest version to access any new features and other enhancements.
\n", + "\n", + "Retrieve the information for the currently installed version of ValidMind:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If the version returned is lower than the version indicated in our [production open-source code](https://github.com/validmind/validmind-library/blob/prod/validmind/__version__.py), restart your notebook and run:\n", + "\n", + "```bash\n", + "%pip install --upgrade validmind\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + }, + { + "cell_type": "markdown", + "id": "copyright-397fa35a68a34dc38f5d84d797fb5331", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "***\n", + "\n", + "Copyright © 2023-2026 ValidMind Inc. All rights reserved.
\n", + "Refer to [LICENSE](https://github.com/validmind/validmind-library/blob/main/LICENSE) for details.
\n", + "SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "validmind-py3.10", + "language": "python", + "name": "validmind-py3.10" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.13" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/site/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb b/site/notebooks/use_cases/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb similarity index 99% rename from site/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb rename to site/notebooks/use_cases/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb index 1ff6ca3e05..ebdbaeae07 100644 --- a/site/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb +++ b/site/notebooks/use_cases/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb @@ -108,7 +108,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -169,8 +169,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb b/site/notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb similarity index 99% rename from site/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb rename to site/notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb index 04f64a31ec..3cac66c032 100644 --- a/site/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb +++ b/site/notebooks/use_cases/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb @@ -106,7 +106,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom test.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom test. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a test, customize its behavior, or provide additional context.\n", "\n", @@ -167,8 +167,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -525,7 +523,7 @@ "source": [ "Next, let's run *comparison tests*, which will allow comparing differences between the training dataset and monitoring datasets. To run a test in comparison mode, you only need to pass an `input_grid` parameter to the `run_test()` method instead of `inputs`.\n", "\n", - "For more information about comparison tests, see this [notebook](../../how_to/run_tests/2_run_comparison_tests.ipynb)." + "For more information about comparison tests, see this [notebook](../../how_to/tests/run_tests/2_run_comparison_tests.ipynb)." ] }, { diff --git a/site/notebooks/code_samples/ongoing_monitoring/xgboost_model.model b/site/notebooks/use_cases/ongoing_monitoring/xgboost_model.model similarity index 100% rename from site/notebooks/code_samples/ongoing_monitoring/xgboost_model.model rename to site/notebooks/use_cases/ongoing_monitoring/xgboost_model.model diff --git a/site/notebooks/code_samples/regression/quickstart_regression_full_suite.ipynb b/site/notebooks/use_cases/regression/quickstart_regression_full_suite.ipynb similarity index 99% rename from site/notebooks/code_samples/regression/quickstart_regression_full_suite.ipynb rename to site/notebooks/use_cases/regression/quickstart_regression_full_suite.ipynb index 90992448e1..91765950e0 100644 --- a/site/notebooks/code_samples/regression/quickstart_regression_full_suite.ipynb +++ b/site/notebooks/use_cases/regression/quickstart_regression_full_suite.ipynb @@ -137,8 +137,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb b/site/notebooks/use_cases/time_series/quickstart_time_series_full_suite.ipynb similarity index 99% rename from site/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb rename to site/notebooks/use_cases/time_series/quickstart_time_series_full_suite.ipynb index 27e5727179..c154f66f6c 100644 --- a/site/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb +++ b/site/notebooks/use_cases/time_series/quickstart_time_series_full_suite.ipynb @@ -106,7 +106,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -171,8 +171,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." diff --git a/site/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb b/site/notebooks/use_cases/time_series/quickstart_time_series_high_code.ipynb similarity index 99% rename from site/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb rename to site/notebooks/use_cases/time_series/quickstart_time_series_high_code.ipynb index 33a2e7220a..da5b9051b1 100644 --- a/site/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb +++ b/site/notebooks/use_cases/time_series/quickstart_time_series_high_code.ipynb @@ -107,7 +107,7 @@ " - **model**: A single model that has been initialized in ValidMind with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model).\n", " - **dataset**: Single dataset that has been initialized in ValidMind with [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset).\n", " - **models**: A list of ValidMind models - usually this is used when you want to compare multiple models in your custom metric.\n", - " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/run_tests_that_require_multiple_datasets.html) for more information.\n", + " - **datasets**: A list of ValidMind datasets - usually this is used when you want to compare multiple datasets in your custom metric. See this [example](https://docs.validmind.ai/notebooks/how_to/tests/run_tests/configure_tests/run_tests_that_require_multiple_datasets.html) for more information.\n", "\n", "**Parameters**: Additional arguments that can be passed when running a ValidMind test, used to pass additional information to a metric, customize its behavior, or provide additional context.\n", "\n", @@ -172,8 +172,6 @@ "\n", "3. Enter the model details and click **Next >** to continue to assignment of model stakeholders. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", "\n", - " For example, to register a model for use with this notebook, select the following use case: `Marketing/Sales - Analytics`\n", - "\n", "4. Select your own name under the **MODEL OWNER** drop-down.\n", "\n", "5. Click **Register Model** to add the model to your inventory." @@ -191,7 +189,7 @@ "\n", "1. In the left sidebar that appears for your model, click **Documents** and select **Documentation**.\n", "\n", - "2. Under **TEMPLATE**, select `Time Series Forecasting`.\n", + "2. Under **TEMPLATE**, select `Time Series Forecasting with ML`.\n", "\n", "3. Click **Use Template** to apply the template." ] diff --git a/site/releases/2023/2023-nov-09/highlights.qmd b/site/releases/2023/2023-nov-09/highlights.qmd index 3c82b2c2fe..ae433a0d82 100644 --- a/site/releases/2023/2023-nov-09/highlights.qmd +++ b/site/releases/2023/2023-nov-09/highlights.qmd @@ -151,7 +151,7 @@ You can easily switch between views using the **Display Table** or **Display Car ::: {.w-30-ns} -[Test descriptions](/developer/model-testing/test-descriptions.qmd){.button} +[Test descriptions](/developer/test-descriptions.qmd){.button} ::: diff --git a/site/releases/2023/2023-oct-25/highlights.qmd b/site/releases/2023/2023-oct-25/highlights.qmd index d3ae1a598c..2103272349 100644 --- a/site/releases/2023/2023-oct-25/highlights.qmd +++ b/site/releases/2023/2023-oct-25/highlights.qmd @@ -202,7 +202,7 @@ We now programmatically embed our Jupyter Notebooks in our documentation site an ::: ::: {.w-30-ns} -[Run tests and test suites](/developer/model-testing/testing-overview.qmd){.button} +[Run tests and test suites](/developer/how-to/testing-overview.qmd){.button} ::: diff --git a/site/releases/2023/2023-sep-27/highlights.qmd b/site/releases/2023/2023-sep-27/highlights.qmd index 691aab12a5..6e1d0f074c 100644 --- a/site/releases/2023/2023-sep-27/highlights.qmd +++ b/site/releases/2023/2023-sep-27/highlights.qmd @@ -60,12 +60,12 @@ We added a new notebook in the {{< var vm.developer >}} that includes the financ :::: {.flex .flex-wrap .justify-around} ::: {.w-50-ns} -[Prompt validation for large language models (LLMs)](/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb){.button .button-green} +[Prompt validation for large language models (LLMs)](/notebooks/use_cases/nlp_and_llm/prompt_validation_demo.ipynb){.button .button-green} ::: ::: {.w-50-ns} -[Automate news summarization using LLMs](/notebooks/code_samples/nlp_and_llm/llm_summarization_demo.ipynb){.button .button-green} +[Automate news summarization using LLMs](/notebooks/use_cases/nlp_and_llm/llm_summarization_demo.ipynb){.button .button-green} ::: @@ -79,7 +79,7 @@ We added a new notebook in the {{< var vm.developer >}} that includes the financ To illustrate this new feature, we have included a **financial news sentiment analysis demo** that runs documentation tests for a Hugging Face model with text classification using the `financial_phrasebank`:[^2] ::: {.tc} -[Sentiment analysis of financial data using Hugging Face NLP models](/notebooks/code_samples/nlp_and_llm/hugging_face_integration_demo.ipynb){.button .button-green} +[Sentiment analysis of financial data using Hugging Face NLP models](/notebooks/use_cases/nlp_and_llm/hugging_face_integration_demo.ipynb){.button .button-green} ::: @@ -285,7 +285,7 @@ We made a number of changes to tests to improve the developer experience: ::: ::: {.w-20-ns} -[Test descriptions](/developer/model-testing/test-descriptions.qmd){.button} +[Test descriptions](/developer/test-descriptions.qmd){.button} ::: diff --git a/site/releases/2023/release-notes-2023-jun-22.qmd b/site/releases/2023/release-notes-2023-jun-22.qmd index 4afe9541a7..eeb5fa428a 100644 --- a/site/releases/2023/release-notes-2023-jun-22.qmd +++ b/site/releases/2023/release-notes-2023-jun-22.qmd @@ -25,7 +25,7 @@ With support for custom tests, you can now go beyond the default set of document ::: ::: {.w-30-ns} -[Implement custom tests](/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb){.button .button-green} +[Implement custom tests](/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb){.button .button-green} ::: @@ -42,7 +42,7 @@ With test providers, you can now integrate external test libraries to expand the ::: ::: {.w-40-ns .tc} -[Integrate external test providers](/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb){.button .button-green} +[Integrate external test providers](/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb){.button .button-green} ::: diff --git a/site/releases/2023/release-notes-2023-may-30.qmd b/site/releases/2023/release-notes-2023-may-30.qmd index 01c55a851d..31eecb08ec 100644 --- a/site/releases/2023/release-notes-2023-may-30.qmd +++ b/site/releases/2023/release-notes-2023-may-30.qmd @@ -34,7 +34,7 @@ Plots and visual outputs have been enhanced with the Plotly package. Users can n ::: ::: {.w-30-ns .tc} -[Test descriptions](/developer/model-testing/test-descriptions.qmd){.button .button-green} +[Test descriptions](/developer/test-descriptions.qmd){.button .button-green} ::: diff --git a/site/releases/2024/2024-aug-13/release-notes.qmd b/site/releases/2024/2024-aug-13/release-notes.qmd index 9b9b260bfb..5a1fa8d773 100644 --- a/site/releases/2024/2024-aug-13/release-notes.qmd +++ b/site/releases/2024/2024-aug-13/release-notes.qmd @@ -32,7 +32,7 @@ To make comparison tests easier to analyze, we've added support to specify an in ::: {.w-30-ns} ::: {.tc} -[Run comparison tests](/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb){.button .button-green} +[Run comparison tests](/notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb){.button .button-green} ::: ::: diff --git a/site/releases/2024/2024-dec-24/release-notes.qmd b/site/releases/2024/2024-dec-24/release-notes.qmd index 7c9a1be8a2..ca025ecd0a 100644 --- a/site/releases/2024/2024-dec-24/release-notes.qmd +++ b/site/releases/2024/2024-dec-24/release-notes.qmd @@ -45,9 +45,9 @@ We've introduced two new quickstart Jupyter Notebooks dealing with option pricin ::: ::: {.w-40-ns .tc} -[Knockout option pricing model](/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb){.button .button-green} +[Knockout option pricing model](/notebooks/use_cases/capital_markets/quickstart_option_pricing_models.ipynb){.button .button-green} -[Heston option pricing model](/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb){.button .button-green} +[Heston option pricing model](/notebooks/use_cases/capital_markets/quickstart_option_pricing_models_quantlib.ipynb){.button .button-green} ::: diff --git a/site/releases/2024/2024-feb-14/highlights.qmd b/site/releases/2024/2024-feb-14/highlights.qmd index 774544a832..e5f36cb898 100644 --- a/site/releases/2024/2024-feb-14/highlights.qmd +++ b/site/releases/2024/2024-feb-14/highlights.qmd @@ -24,7 +24,7 @@ Documentation templates have been updated to support logging each test run as a ::: ::: {.w-50-ns .tc} -[Document multiple results for the same test](/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb){.button .button-green} +[Document multiple results for the same test](/notebooks/how_to/tests/run_tests/documentation_tests/document_multiple_results_for_the_same_test.ipynb){.button .button-green} ::: @@ -148,7 +148,7 @@ To enable model developers to know what task types and tags are available to fil ::: ::: {.w-30-ns .tc} -[Explore tests](/notebooks/how_to/explore_tests.ipynb){.button .button-green} +[Explore tests](/notebooks/how_to/tests/explore_tests/explore_tests.ipynb){.button .button-green} ::: diff --git a/site/releases/2024/2024-jan-26/highlights.qmd b/site/releases/2024/2024-jan-26/highlights.qmd index b2e7d0c0e1..7d4f2e3688 100644 --- a/site/releases/2024/2024-jan-26/highlights.qmd +++ b/site/releases/2024/2024-jan-26/highlights.qmd @@ -133,7 +133,7 @@ A new notebook illustrates how you can configure these dataset features: - How `feature_columns` can be used to report by segment ::: {.tc} -[Configure dataset features](/notebooks/how_to/configure_dataset_features.ipynb){.button.button-green} +[Configure dataset features](/notebooks/how_to/data_and_datasets/dataset_inputs/configure_dataset_features.ipynb){.button.button-green} ::: ::: @@ -209,7 +209,7 @@ full_suite = vm.run_documentation_tests( ``` ::: {.tc} -[Run individual documentation sections](/notebooks/how_to/run_documentation_sections.ipynb){.button .button-green} +[Run individual documentation sections](/notebooks/how_to/tests/run_tests/documentation_tests/run_documentation_sections.ipynb){.button .button-green} ::: ::: diff --git a/site/releases/2024/2024-jul-22/release-notes.qmd b/site/releases/2024/2024-jul-22/release-notes.qmd index b37adfacc8..c48cea4673 100644 --- a/site/releases/2024/2024-jul-22/release-notes.qmd +++ b/site/releases/2024/2024-jul-22/release-notes.qmd @@ -44,7 +44,7 @@ You can now run comparison tests with the {{< var validmind.developer >}}. ::: ::: {.w-30-ns} -[Run comparison tests](/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb){.button .button-green} +[Run comparison tests](/notebooks/how_to/tests/run_tests/2_run_comparison_tests.ipynb){.button .button-green} ::: @@ -294,7 +294,7 @@ We fixed a number of missing test descriptions that were caused by a scripting i ::: {.w-20-ns} -[Test descriptions](/developer/model-testing/test-descriptions.qmd){.button} +[Test descriptions](/developer/test-descriptions.qmd){.button} ::: diff --git a/site/releases/2024/2024-may-22/release-notes.qmd b/site/releases/2024/2024-may-22/release-notes.qmd index 2e8ff71423..15ad97639d 100644 --- a/site/releases/2024/2024-may-22/release-notes.qmd +++ b/site/releases/2024/2024-may-22/release-notes.qmd @@ -1097,5 +1097,5 @@ These features provide a rich context for testing and evaluation. You can use re -[^1]: See the full list of tests at [Test descriptions](/developer/model-testing/test-descriptions.qmd) or try the [Test sandbox](/developer/model-testing/test-sandbox.qmd). +[^1]: See the full list of tests at [Test descriptions](/developer/test-descriptions.qmd) or try the [Test sandbox](/developer/how-to/test-sandbox.qmd). diff --git a/site/releases/2025/2025-jan-31/release-notes.qmd b/site/releases/2025/2025-jan-31/release-notes.qmd index c27f371fac..91932ac527 100644 --- a/site/releases/2025/2025-jan-31/release-notes.qmd +++ b/site/releases/2025/2025-jan-31/release-notes.qmd @@ -9,7 +9,7 @@ listing: max-description-length: 250 # image-height: 100% contents: - - path: /notebooks/how_to/customize_test_result_descriptions.ipynb + - path: /notebooks/how_to/tests/run_tests/configure_tests/customize_test_result_descriptions.ipynb title: "Customize test result descriptions" description: "Learn how to take control over the context that drives test description generation. {{< fa chevron-right >}}" - id: credit-risk @@ -357,7 +357,7 @@ This enhancement makes it easier to distinguish between ValidMind's standard tes ::: ::: {.w-30-ns .tr} -[Test descriptions](/developer/model-testing/test-descriptions.qmd){.button target="_blank"} +[Test descriptions](/developer/test-descriptions.qmd){.button target="_blank"} ::: :::: diff --git a/site/scripts/developer-sidebar/update_how_tos.py b/site/scripts/developer-sidebar/update_how_tos.py new file mode 100644 index 0000000000..957e066160 --- /dev/null +++ b/site/scripts/developer-sidebar/update_how_tos.py @@ -0,0 +1,282 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial +""" +Update developer/_sidebar.yaml and developer/how-to/feature-overview.qmd +so that each subdirectory of notebooks/how_to/ (excluding tests/) is listed +explicitly in alphabetical order, with fixed capitalization where needed. + +Run from the site/ directory, e.g.: + cd site && python scripts/developer-sidebar/update_how_tos.py +""" + +import os +import re +from pathlib import Path + + +# Display title for directories that need fixed capitalization (e.g. acronyms) +SPECIAL_TITLES = { + "data_and_datasets": "Data and datasets", + "tests": "Testing", +} + +# Subdirectories to exclude from the sidebar only (tests has its own +# dedicated section in the sidebar already). +SIDEBAR_EXCLUDED_DIRS = {"tests"} + + +def dir_to_title(dirname: str) -> str: + """Convert directory name to display title (sentence-style capitalization).""" + if dirname in SPECIAL_TITLES: + return SPECIAL_TITLES[dirname] + return dirname.replace("_", " ").capitalize() + + +def dir_to_listing_id(dirname: str) -> str: + """Convert a directory name to a listing id (hyphens instead of underscores).""" + return dirname.replace("_", "-") + + +def _has_notebooks(directory: Path) -> bool: + """Check if a directory directly contains .ipynb files.""" + return any( + f.endswith(".ipynb") + for f in os.listdir(directory) + if (directory / f).is_file() + ) + + +def _has_notebooks_recursive(directory: Path) -> bool: + """Check if a directory or any of its descendants contains .ipynb files.""" + for _root, _dirs, files in os.walk(directory): + if any(f.endswith(".ipynb") for f in files): + return True + return False + + +def _build_section_yaml( + how_to_base: Path, rel_path: str, dirname: str, indent: int +) -> list[str]: + """Build YAML lines for a how-to section, nesting subdirectories. + + If the directory has subdirectories that contain notebooks, each one + becomes a nested ``section`` (recursively). Subdirectories without any + notebooks are ignored. Top-level notebooks within the directory are + included via a non-recursive glob (``*.ipynb``). Leaf directories use a + recursive glob (``**/*.ipynb``) for simplicity. + """ + full_path = how_to_base / rel_path + subdirs = sorted( + d + for d in os.listdir(full_path) + if (full_path / d).is_dir() and _has_notebooks_recursive(full_path / d) + ) + + prefix = " " * indent + title = dir_to_title(dirname) + lines = [f'{prefix}- section: "{title}"'] + + if not subdirs: + # Leaf directory — simple recursive glob. + lines.append(f'{prefix} contents: "notebooks/how_to/{rel_path}/**/*.ipynb"') + else: + # Has subdirectories — build an explicit contents list. + lines.append(f"{prefix} contents:") + + # List top-level notebooks explicitly (bare glob strings in a YAML + # list are not resolved by Quarto). + for nb in sorted( + f for f in os.listdir(full_path) + if f.endswith(".ipynb") and (full_path / f).is_file() + ): + lines.append(f"{prefix} - notebooks/how_to/{rel_path}/{nb}") + + # Recurse into each subdirectory. + for subdir in subdirs: + sub_rel = f"{rel_path}/{subdir}" + lines.extend( + _build_section_yaml(how_to_base, sub_rel, subdir, indent + 4) + ) + + return lines + + +def update_sidebar(base: Path, subdirs: list) -> None: + """Update developer/_sidebar.yaml with how-to subdirectories.""" + sidebar_path = base / "developer" / "_sidebar.yaml" + + if not sidebar_path.is_file(): + raise SystemExit(f"Sidebar file not found: {sidebar_path}") + + how_to_base = base / "notebooks" / "how_to" + + # Build the new contents block (YAML). Use "section" so Quarto renders + # expandable accordion items; "text" alone does not expand. + lines = [" contents:"] + for d in subdirs: + lines.extend(_build_section_yaml(how_to_base, d, d, 12)) + + new_block = "\n".join(lines) + + text = sidebar_path.read_text() + + # Try to find and replace an existing expanded contents block under + # "Use library features" (re-run case). Match the header lines then + # consume all indented content lines (12+ spaces) that follow. + pattern = re.compile( + r'( - text: "Use library features"\n' + r" file: developer/how-to/feature-overview\.qmd\n)" + r" contents:\n" + r"(?:[ ]{12,}[^\n]*\n)*", + re.MULTILINE, + ) + match = pattern.search(text) + if match: + # Replace existing contents block, keeping the header lines. + text = ( + text[: match.start()] + + match.group(1) + + new_block + + "\n" + + text[match.end() :] + ) + else: + # First run: insert contents block right after the file: line. + insert_pattern = re.compile( + r'( - text: "Use library features"\n' + r" file: developer/how-to/feature-overview\.qmd\n)", + re.MULTILINE, + ) + insert_match = insert_pattern.search(text) + if not insert_match: + raise SystemExit( + 'Could not find "Use library features" entry in sidebar. ' + "Has the sidebar format changed?" + ) + text = ( + text[: insert_match.end()] + + new_block + + "\n" + + text[insert_match.end() :] + ) + + sidebar_path.write_text(text) + print(f"Updated {sidebar_path} with {len(subdirs)} how-to directories.") + + +def _split_frontmatter(text: str): + """Split a .qmd file into (frontmatter_lines, body) at the closing ---.""" + lines = text.split("\n") + # First line must be --- + if not lines or lines[0].strip() != "---": + raise SystemExit("File does not start with YAML front matter (---).") + # Find the closing --- + for i, line in enumerate(lines[1:], start=1): + if line.strip() == "---": + frontmatter = lines[: i + 1] # includes both --- delimiters + body = "\n".join(lines[i + 1 :]) + return frontmatter, body + raise SystemExit("Could not find closing --- for YAML front matter.") + + +def update_feature_overview(base: Path, subdirs: list) -> None: + """Update feature-overview.qmd with listings and panel tabset + for each how-to subdirectory.""" + page_path = base / "developer" / "how-to" / "feature-overview.qmd" + + if not page_path.is_file(): + raise SystemExit(f"Feature overview page not found: {page_path}") + + text = page_path.read_text() + + # --- Build new listing YAML lines --- + listing_lines = ["listing:"] + for d in subdirs: + listing_id = dir_to_listing_id(d) + listing_lines.append(f" - id: {listing_id}") + listing_lines.append(f" type: grid") + listing_lines.append(f" grid-columns: 2") + listing_lines.append(f' image-placeholder: "jupyter-logo-rectangle.svg"') + listing_lines.append(f" max-description-length: 350") + listing_lines.append(f' image-height: "100%"') + listing_lines.append(f" fields: [title, description, reading-time]") + listing_lines.append(f' contents: "../../notebooks/how_to/{d}/**/*.ipynb"') + + # --- Update YAML front matter --- + frontmatter, body = _split_frontmatter(text) + + # Remove any existing listing: block from the front matter + fm_text = "\n".join(frontmatter) + fm_text = re.sub(r"\nlisting:\n(?:[ ].*\n)*", "\n", fm_text) + + # Re-split so we can insert the listing block before the closing --- + fm_lines = fm_text.split("\n") + # The last element should be --- + closing = fm_lines.pop() # remove closing --- + fm_lines.extend(listing_lines) + fm_lines.append(closing) + + # --- Build new panel-tabset block --- + tabset_lines = ["## How-to by topic", "", ":::{.panel-tabset}", ""] + for d in subdirs: + listing_id = dir_to_listing_id(d) + title = dir_to_title(d) + tabset_lines.append(f"## {title}") + tabset_lines.append("") + tabset_lines.append(f":::{{#{listing_id}}}") + tabset_lines.append(":::") + tabset_lines.append("") + tabset_lines.append(":::") + tabset_lines.append("") + + new_tabset = "\n".join(tabset_lines) + + # Replace everything from "## How-to by topic" to end of body. + feature_pattern = re.compile( + r"## How-to by topic\n?.*", re.DOTALL + ) + if feature_pattern.search(body): + body = feature_pattern.sub(new_tabset, body) + else: + # Heading not found; append the tabset at the end + body = body.rstrip("\n") + "\n\n" + new_tabset + + # --- Reassemble and write --- + result = "\n".join(fm_lines) + "\n" + body + page_path.write_text(result) + print(f"Updated {page_path} with {len(subdirs)} how-to listings.") + + +def main() -> None: + # Run from site/ or repo root + cwd = Path.cwd() + if (cwd / "notebooks" / "how_to").is_dir(): + base = cwd + elif (cwd / "site" / "notebooks" / "how_to").is_dir(): + base = cwd / "site" + else: + raise SystemExit( + "Run from site/ or repo root " + "(e.g. cd site && python scripts/developer-sidebar/update_how_tos.py)" + ) + how_to = base / "notebooks" / "how_to" + + all_subdirs = sorted( + d for d in os.listdir(how_to) if (how_to / d).is_dir() + ) + + # Sidebar excludes tests (it has its own dedicated section). + sidebar_subdirs = [d for d in all_subdirs if d not in SIDEBAR_EXCLUDED_DIRS] + + # Feature overview includes all dirs, with tests listed first. + overview_subdirs = [d for d in all_subdirs if d in SIDEBAR_EXCLUDED_DIRS] + [ + d for d in all_subdirs if d not in SIDEBAR_EXCLUDED_DIRS + ] + + update_sidebar(base, sidebar_subdirs) + update_feature_overview(base, overview_subdirs) + + +if __name__ == "__main__": + main() diff --git a/site/scripts/developer-sidebar/update_use_cases.py b/site/scripts/developer-sidebar/update_use_cases.py new file mode 100644 index 0000000000..0e8713dc8c --- /dev/null +++ b/site/scripts/developer-sidebar/update_use_cases.py @@ -0,0 +1,228 @@ +# Copyright © 2023-2026 ValidMind Inc. All rights reserved. +# Refer to the LICENSE file in the root of this repository for details. +# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial +""" +Update developer/_sidebar.yaml and developer/samples-jupyter-notebooks.qmd +so that each subdirectory of notebooks/use_cases/ is listed explicitly +in alphabetical order, with fixed capitalization for NLP and LLM. + +Run from the site/ directory, e.g.: + cd site && python scripts/developer-sidebar/update_use_cases.py +""" + +import os +import re +from pathlib import Path + + +# Display title for directories that need fixed capitalization (e.g. acronyms) +SPECIAL_TITLES = { + "nlp_and_llm": "NLP and LLM", +} + + +def dir_to_title(dirname: str) -> str: + """Convert directory name to sidebar display title (sentence-style capitalization).""" + if dirname in SPECIAL_TITLES: + return SPECIAL_TITLES[dirname] + return dirname.replace("_", " ").capitalize() + + +def dir_to_listing_id(dirname: str) -> str: + """Convert a directory name to a listing id (hyphens instead of underscores).""" + return dirname.replace("_", "-") + + +def _has_notebooks(directory: Path) -> bool: + """Check if a directory directly contains .ipynb files.""" + return any( + f.endswith(".ipynb") + for f in os.listdir(directory) + if (directory / f).is_file() + ) + + +def _has_notebooks_recursive(directory: Path) -> bool: + """Check if a directory or any of its descendants contains .ipynb files.""" + for _root, _dirs, files in os.walk(directory): + if any(f.endswith(".ipynb") for f in files): + return True + return False + + +def _build_section_yaml( + use_cases_base: Path, rel_path: str, dirname: str, indent: int +) -> list[str]: + """Build YAML lines for a use-case section, nesting subdirectories. + + If the directory has subdirectories that contain notebooks, each one + becomes a nested ``section`` (recursively). Subdirectories without any + notebooks are ignored. Top-level notebooks within the directory are + included via a non-recursive glob (``*.ipynb``). Leaf directories use a + recursive glob (``**/*.ipynb``) for simplicity. + """ + full_path = use_cases_base / rel_path + subdirs = sorted( + d + for d in os.listdir(full_path) + if (full_path / d).is_dir() and _has_notebooks_recursive(full_path / d) + ) + + prefix = " " * indent + title = dir_to_title(dirname) + lines = [f'{prefix}- section: "{title}"'] + + if not subdirs: + # Leaf directory — simple recursive glob. + lines.append(f'{prefix} contents: "notebooks/use_cases/{rel_path}/**/*.ipynb"') + else: + # Has subdirectories — build an explicit contents list. + lines.append(f"{prefix} contents:") + + # List top-level notebooks explicitly (bare glob strings in a YAML + # list are not resolved by Quarto). + for nb in sorted( + f for f in os.listdir(full_path) + if f.endswith(".ipynb") and (full_path / f).is_file() + ): + lines.append(f"{prefix} - notebooks/use_cases/{rel_path}/{nb}") + + # Recurse into each subdirectory. + for subdir in subdirs: + sub_rel = f"{rel_path}/{subdir}" + lines.extend( + _build_section_yaml(use_cases_base, sub_rel, subdir, indent + 4) + ) + + return lines + + +def update_sidebar(base: Path, subdirs: list) -> None: + """Update developer/_sidebar.yaml with use case subdirectories.""" + sidebar_path = base / "developer" / "_sidebar.yaml" + + if not sidebar_path.is_file(): + raise SystemExit(f"Sidebar file not found: {sidebar_path}") + + use_cases_base = base / "notebooks" / "use_cases" + + # Build the new contents block (YAML). Use "section" so Quarto renders + # expandable accordion items; "text" alone does not expand. + lines = [" contents:"] + for d in subdirs: + lines.extend(_build_section_yaml(use_cases_base, d, d, 12)) + + new_block = "\n".join(lines) + + text = sidebar_path.read_text() + # Replace either the single wildcard or an existing expanded block. + # Accept both the old "code_samples" and the new "use_cases" paths so the + # script works on first migration as well as on subsequent re-runs. + old_single_use_cases = ' contents: "notebooks/use_cases/**"' + old_single_code_samples = ' contents: "notebooks/code_samples/**"' + if old_single_use_cases in text: + text = text.replace(old_single_use_cases, new_block, 1) + elif old_single_code_samples in text: + text = text.replace(old_single_code_samples, new_block, 1) + else: + # Find the contents block and replace it (multi-line). Match the + # header lines then consume all indented content lines (12+ spaces). + pattern = re.compile( + r'( - text: "(?:Code samples|Use cases)"\n' + r" file: developer/samples-jupyter-notebooks\.qmd\n)" + r" contents:\n" + r"(?:[ ]{12,}[^\n]*\n)*", + re.MULTILINE, + ) + match = pattern.search(text) + if not match: + raise SystemExit( + "Could not find Code samples / Use cases contents block in sidebar. " + "Has the sidebar format changed?" + ) + text = text[: match.start()] + match.group(1) + new_block + "\n" + text[match.end() :] + sidebar_path.write_text(text) + print(f"Updated {sidebar_path} with {len(subdirs)} use case directories.") + + +def update_notebooks_page(base: Path, subdirs: list) -> None: + """Update samples-jupyter-notebooks.qmd with listings and panel tabset + for each use case subdirectory.""" + page_path = base / "developer" / "samples-jupyter-notebooks.qmd" + + if not page_path.is_file(): + raise SystemExit(f"Notebooks page not found: {page_path}") + + text = page_path.read_text() + + # --- Build new listing YAML block --- + listing_lines = ["listing:"] + for d in subdirs: + listing_id = dir_to_listing_id(d) + listing_lines.append(f" - id: {listing_id}") + listing_lines.append(f' type: grid') + listing_lines.append(f' grid-columns: 2') + listing_lines.append(f' image-placeholder: "jupyter-logo-rectangle.svg"') + listing_lines.append(f' max-description-length: 350') + listing_lines.append(f' image-height: "100%"') + listing_lines.append(f' fields: [title, description, reading-time]') + listing_lines.append(f' contents: "../notebooks/use_cases/{d}/*.ipynb"') + + new_listing_block = "\n".join(listing_lines) + "\n" + + # Replace the listing block in the YAML frontmatter. + # Match "listing:" at the start of a line and all following indented lines. + listing_pattern = re.compile( + r'^listing:\n(?:[ ].*\n)*', + re.MULTILINE, + ) + text = listing_pattern.sub(new_listing_block, text, count=1) + + # --- Build new panel-tabset block --- + tabset_lines = ["## By use case", "", ":::{.panel-tabset}", ""] + for d in subdirs: + listing_id = dir_to_listing_id(d) + title = dir_to_title(d) + tabset_lines.append(f"## {title}") + tabset_lines.append("") + tabset_lines.append(f":::{{#{listing_id}}}") + tabset_lines.append(":::") + tabset_lines.append("") + tabset_lines.append(":::") + tabset_lines.append("") + + new_tabset = "\n".join(tabset_lines) + + # Replace everything from "## By use case" to end of file. + use_case_pattern = re.compile(r'^## By use case\n.*', re.MULTILINE | re.DOTALL) + text = use_case_pattern.sub(new_tabset, text) + + page_path.write_text(text) + print(f"Updated {page_path} with {len(subdirs)} use case listings.") + + +def main() -> None: + # Run from site/ or repo root + cwd = Path.cwd() + if (cwd / "notebooks" / "use_cases").is_dir(): + base = cwd + elif (cwd / "site" / "notebooks" / "use_cases").is_dir(): + base = cwd / "site" + else: + raise SystemExit( + "Run from site/ or repo root " + "(e.g. cd site && python scripts/developer-sidebar/update_use_cases.py)" + ) + use_cases = base / "notebooks" / "use_cases" + + subdirs = sorted( + d for d in os.listdir(use_cases) + if (use_cases / d).is_dir() + ) + + update_sidebar(base, subdirs) + update_notebooks_page(base, subdirs) + + +if __name__ == "__main__": + main() diff --git a/site/scripts/update_code_samples_sidebar.py b/site/scripts/update_code_samples_sidebar.py deleted file mode 100644 index c9d31fa5d3..0000000000 --- a/site/scripts/update_code_samples_sidebar.py +++ /dev/null @@ -1,94 +0,0 @@ -# Copyright © 2023-2026 ValidMind Inc. All rights reserved. -# Refer to the LICENSE file in the root of this repository for details. -# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial -""" -Update developer/_sidebar.yaml so the Code samples entry lists each -subdirectory of notebooks/code_samples/ explicitly (with wildcards), -in alphabetical order, with fixed capitalization for NLP and LLM. - -Run from the site/ directory, e.g.: - cd site && python scripts/update_code_samples_sidebar.py -""" - -import os -from pathlib import Path - - -# Display title for directories that need fixed capitalization (e.g. acronyms) -SPECIAL_TITLES = { - "nlp_and_llm": "NLP and LLM", -} - - -def dir_to_title(dirname: str) -> str: - """Convert directory name to sidebar display title (sentence-style capitalization).""" - if dirname in SPECIAL_TITLES: - return SPECIAL_TITLES[dirname] - return dirname.replace("_", " ").capitalize() - - -def main() -> None: - # Run from site/ or repo root - cwd = Path.cwd() - if (cwd / "notebooks" / "code_samples").is_dir(): - base = cwd - elif (cwd / "site" / "notebooks" / "code_samples").is_dir(): - base = cwd / "site" - else: - raise SystemExit("Run from site/ or repo root (e.g. cd site && python scripts/update_code_samples_sidebar.py)") - code_samples = base / "notebooks" / "code_samples" - sidebar_path = base / "developer" / "_sidebar.yaml" - - if not code_samples.is_dir(): - raise SystemExit(f"Directory not found: {code_samples}") - if not sidebar_path.is_file(): - raise SystemExit(f"Sidebar file not found: {sidebar_path}") - - subdirs = sorted( - d for d in os.listdir(code_samples) - if (code_samples / d).is_dir() - ) - - # Build the new contents block (YAML). Use "section" so Quarto renders - # expandable accordion items; "text" alone does not expand. - lines = [ - ' contents:', - ] - for d in subdirs: - title = dir_to_title(d) - lines.append(f' - section: "{title}"') - lines.append(f' contents: "notebooks/code_samples/{d}/**"') - - new_block = "\n".join(lines) - - text = sidebar_path.read_text() - # Replace either the single wildcard or an existing expanded block - old_single = ' contents: "notebooks/code_samples/**"' - if old_single in text: - text = text.replace(old_single, new_block, 1) - else: - # Find the code_samples contents block and replace it (multi-line) - import re - # Match from " contents:" through all following lines that are - # " - ..." or " contents: ..." for code_samples - pattern = re.compile( - r'( - text: "Code samples"\n' - r' file: developer/samples-jupyter-notebooks\.qmd\n)' - r' contents:\n' - r'( - (?:text|section): "[^"]+"\n' - r' contents: "notebooks/code_samples/[^"]+\*\*"\n)*', - re.MULTILINE, - ) - match = pattern.search(text) - if not match: - raise SystemExit( - "Could not find Code samples contents block in sidebar. " - "Has the sidebar format changed?" - ) - text = text[: match.start()] + match.group(1) + new_block + "\n" + text[match.end() :] - sidebar_path.write_text(text) - print(f"Updated {sidebar_path} with {len(subdirs)} code sample directories.") - - -if __name__ == "__main__": - main() diff --git a/site/training/common-slides/_validmind-test-repository.qmd b/site/training/common-slides/_validmind-test-repository.qmd index 17478fbc05..ab758f54db 100644 --- a/site/training/common-slides/_validmind-test-repository.qmd +++ b/site/training/common-slides/_validmind-test-repository.qmd @@ -2,7 +2,7 @@ Refer to the LICENSE file in the root of this repository for details. SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial --> -## {background-iframe="/developer/model-testing/test-descriptions.html" background-interactive="true" data-preload="yes"} +## {background-iframe="/developer/test-descriptions.html" background-interactive="true" data-preload="yes"} :::: {.slideover--l .three-quarters .auto-collapse-5} **{{< var vm.product >}} test repository** diff --git a/site/training/developer-fundamentals/finalizing-model-documentation.qmd b/site/training/developer-fundamentals/finalizing-model-documentation.qmd index 843d0eba13..0e53cbf143 100644 --- a/site/training/developer-fundamentals/finalizing-model-documentation.qmd +++ b/site/training/developer-fundamentals/finalizing-model-documentation.qmd @@ -200,7 +200,7 @@ For example: 2. Click on the text to edit the description for our individually inserted `HighPearsonCorrelation:balanced_raw_dataset` test: - ![Editor for a test result description](/notebooks/images/high-pearson-correlation-block.png){fig-alt="Screenshot showing the editor for a test result description" .screenshot} + ![Editor for a test result description](/notebooks/tutorials/model_development/high-pearson-correlation-block.png){fig-alt="Screenshot showing the editor for a test result description" .screenshot} ::: diff --git a/site/training/program/training-faq.qmd b/site/training/program/training-faq.qmd index 81e2951b20..85fdd0fa3c 100644 --- a/site/training/program/training-faq.qmd +++ b/site/training/program/training-faq.qmd @@ -209,7 +209,7 @@ Here's what we have been asked about {{< var vm.product >}} during training sess ::: ::: {.w-40-ns .nb3 .pl2} -[{{< fa graduation-cap >}} training](/training/developer-fundamentals/implementing-custom-tests.html#/implement-custom-tests){.button-small} [{{< fa book-open >}} docs](/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb){.button-small} +[{{< fa graduation-cap >}} training](/training/developer-fundamentals/implementing-custom-tests.html#/implement-custom-tests){.button-small} [{{< fa book-open >}} docs](/notebooks/how_to/tests/custom_tests/implement_custom_tests.ipynb){.button-small} ::: :::: @@ -221,7 +221,7 @@ Here's what we have been asked about {{< var vm.product >}} during training sess ::: ::: {.w-40-ns .nb3 .pl2} -[{{< fa graduation-cap >}} training](/training/developer-fundamentals/implementing-custom-tests.html#/section-9){.button-small} [{{< fa book-open >}} docs](/notebooks/code_samples/custom_tests/integrate_external_test_providers.ipynb){.button-small} +[{{< fa graduation-cap >}} training](/training/developer-fundamentals/implementing-custom-tests.html#/section-9){.button-small} [{{< fa book-open >}} docs](/notebooks/how_to/tests/custom_tests/integrate_external_test_providers.ipynb){.button-small} ::: :::: @@ -301,7 +301,7 @@ Here's what we have been asked about {{< var vm.product >}} during training sess ::: ::: {.w-40-ns .nb3 .pl2} -[{{< fa book-open >}} code sample](/notebooks/code_samples/nlp_and_llm/llm_summarization_demo.ipynb#initialize-the-python-environment){.button-small} +[{{< fa book-open >}} code sample](/notebooks/use_cases/nlp_and_llm/llm_summarization_demo.ipynb#initialize-the-python-environment){.button-small} ::: :::: diff --git a/site/training/what-is-validmind/what-is-validmind.qmd b/site/training/what-is-validmind/what-is-validmind.qmd index a842611b6e..2e06a521bd 100644 --- a/site/training/what-is-validmind/what-is-validmind.qmd +++ b/site/training/what-is-validmind/what-is-validmind.qmd @@ -680,7 +680,7 @@ As a validator, use the same tools used to develop models to ... ::: ::: {.column width="30%" .tc .nt4} -[[{{< fa code >}} Validation Use Case]{.button .button-light-green .shadow-5-ns}](/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb){target="_blank"}
+[[{{< fa code >}} Validation Use Case]{.button .button-light-green .shadow-5-ns}](/notebooks/use_cases/model_validation/validate_application_scorecard.ipynb){target="_blank"}
::: ::::

zB&AdDcL88O*=B3zufFa(`(rmmT81OBuKYI=B8-@>e)TTQbd z6^kP}^a6jFhFk}8jss$}22|8WzcSlkzO@YtK0|j_?g)*g-wp?miis|6P5GWW3cW_A zXV;VV;(cC=$@YvECUW{b|61L2&E4_2GsbzEd?qgtRSh3ET-ejt@F6DZ0_}QFZ}4A# zrik1mb98H=QnYlYx{NSZ!6#%gfHkRs5|WtPPX%-3aLkOX@&bB#(E6RW=H3OVY9`)W zn$LpMYiCf2T{Ptv(_r*IKf^aXrXFny2SPQ9*QcMQDA=f*u2L#cD0W4Yq4Su@Wg1*iDeQNj> z1eJc!1Vb1hK>M(k3VV66Fu;J@3wvO;FE&LII&-S@{!3h~6YR9~=n3BKKjs^#w_ELA+rlb8o|sEs8UIWmMw>G`P%k=4fMB;~!Kgf1^o$|=^yY?> z;GY^ajsjjc^^fy}pGYM#vtMZ6w^1Ey)?s(zul^{tOx09 zH7{d@Q2}Z3X24PZ`vGF)sR07-rJ2wDMvCm*j{$(m{)AZi#M2$;kDy1#BD>>#=t-O-At(M`S1|-_wZ~wqY;reG-_L+C`rt_TxQFB zRB@lGM{d~c31vsKIQ;G~3t>jRhs>wp`RJHU!Q8)t-=yiw&2ncHlJ9Pk;CkKTdDCV? zrUpQ-)yQp}wSK|Ti7{8#J!?6=!#_r6Q}pX~<|AZJ_#jGW8S%SRNA-iPYbYn}&RCkt3ZpS{bYb2mg(=3-41WNd z1o+F9iO9j6gI&DS42Lbtph~S%ou9hTDdY<{nl|x@?B*6kF2wM_MEZqT%F2C~r1BBm zy85>rFSa+Jt4?yAkiM@*0X3g_Blq4Eyi(|{HnU@&H}u)8NP4!Yf{pOgJfFvxb)0*z zxt4VU9vrCotMBZD;JRBK_FI6|9sp>EQ>i%Q*5HaSI_c^8CalnS-3+lpQwVUN=op5G zCMwS8t=dhTT+_8S8L9i4F9#=qO|~cDBs8Z@0m*#uu&GOD@d5s|bSppIr52N>PX^(Y zp^V*-;!lbMuO{}FvK0_8m;{X8E(Z}$qZjgcAakfSYWS`7Ayz$vDURXE3Jx$4GdP^9 zf_mBXm1?amhSTRrG5HLe^tz5F%vs?Yw+wL4goxY`4@}X`DK+)UkyPgaO16l&PhC)j zPE4J>(5CV6^Sg>~*kJBE#9M2_SqoJ>Sf*>_S^k^XafU^EkiZ|l2+3kIIZ(apxnm5F zP-h`hVeC$*Q%%JCtCS9R;0KVd9MVoHj9ff19hxxqncv#D01JQL3bt>(xCsyMc*o!I zTH-1C=it37n`30)Bs%>5HOuFJk%O-j{4h5H`|>!&^FSCI^*MRGNYRw4U8SyfFSxS) z_Ox*B<{c*oc&5Hk@j!;EKke{Yk_yWlLoIGaisFO){ z7Z9UWMRwrf18T#Skm&1Y-C;cjR(`ILSR3)@>fRRhk5u4#lN5q2d3mfS*r|EijQXOm zw&n!mM;bR(1ZtV2CxXCbDKqh_@TM|JPyC=Q1NdcvXx5UWB3{l4ki;mTcMT(O!G44n zjY7YC0$jaF1KRUgCIF(PS`o`0p0x|1aGG&mAwZ!121{pVTuM!@73GtldQQLd(2#+k z9f?gN7+2V|th!>$$`>Vpe${_2`RDa>&y|{Sk7+BHzlLFzywsD@h$~ub#~cSYDyvDD zQC6(G4hEi%H(NSpO9fv~)`{y|LeyZeCw19xr z{=f3Yt*8H^P9B7|5HuJDpq0p9t&pxjUymBRo(VkO*W3OXI52%DnW;OZYQvb`ONqZi zzrOW~t-v`6Yz$tYha{;JusfEn!S9*q-D3BshxWg3xqgu0_>HP2D`lxi&8DvOs}m8#AXPVbj8fBza%R@CrNc(h_H6~Fq? zQdHFe;-KEo0W0?Ld+U3EoPH;Q!90Ot*0$0~P)EdB%L%FRL*{OoRERC4nF z2{(Itd%E_^3v_aM!zRI^puZ@V;7UEMnn3=+bNkud1m%f}dI}a|L~OVJhs$Maw#5}! zDOIWTHcD|*A5)KJZGkO_{96j2?WI!=gaZ1K{!E~FURlvy*R&hpY)PV2X_dZ9HOQ@Q zi3)H`3Uw#?fpftNsi9gLtfN9D`qtV8grb}&~(k9=tz51pKn%+er+FxG$ATIl~&NH5q_ zI{D1pyE$VD6$+Xre84SRDqiyuvagW>h%h&G+4@4nBv*|KCAy*1^bsGis-q^*bLtP3 z$0>K0r6c#n#f}ObRSj5Z{%WN?nloJ^vT}u#hmp0y&llvJjKkW*(arn5HO*(qT+!e{ zcFTArM&%vrA*k|tpppfJF%wYdfT90`C>I|Qnodf+f*6G~x5Lt4`E>=XHey!-07~C4 zhSY5Zi|^)5Wl%?mohUK|KcH9mi~O z!(O0kLa`$|Nlx_-_1-+-i?7!#O6x4vtG9D}A@RO`A;=fJP2H+4YnWOpS`y2qxxq?B zMn_tw#vV!+bT1`1a=w*r@u4#wkZrFLOv7aGMv$dIye@=fg}ZIPX~(TZtv3qwxq||; z6Vt~kC<*r%8sB1js!=Uiy&bnkus$OigzI;|&Qpc2NkSj=JnuCzP14{qU=IZ;17&0o zG~r~+VjyEOFUCe(hCQ}?rLxj`HiUJHfty>SsFizEAX+#g64%6Iw$kKRMj52(1IJV$1x9C`-X#GtlpaMuBWsd0aTva5gb83 zB=jMuIgM3Z3lQ6xz2Wudy$Eo(tLp5ErmC#xVfc(=Hxh4Fi9Rh`<^xugCUJg^64h|U z1oFhH=$wPsZPRriCPoMYobQG8tN53p5^T5y_x=LFS8G76|LzcDzoJ&kwFuMkgOfCC zSRE#S4l%UQ}JTLqP2C9cvyM!JBsV!@EcW5W?UZsO(!?6TEMX z%J-QKfSl*Hzg6aLenfCUUDV8Xat9o}|CrID6v9}1uqeF*I=z3ehf0wQv}wrRVNQr< zM9V^a!dt__<+b2I2v0)ezc#=$^eu>!!mHauBM4XWp=pVJmm$`8)KycMIsbm~;j45fIpK!-rx?<{Uhz?))jaWUyEg*~@X3bS*iC>~6rpa$aw3JRG^b)|quyc|BYAR5Wioc|TQ^nUKOXf6n4a_Y5YY zE7^0;LYHd(WTF@p5}CBtrd`C7$}ZW{C?0_TJaHNoi7S;aijTf&VCDfu5N|j*@8jy6 z&R1sl9PBy;Cb;EBYn+=R7+h_0&@^jsf7i*^u~8< zuSV^m&e@`NLY>);uQ9_mOB9(96o5(t_0K5O_lB1I!}%Yjy^6sBQ{^YJIuT4OV|@jZ z>%+po>fc?pJ#D=Uqm=f-qGOxqNf7$AmO* z53$Wjan0Rfy4KCfwWBQI=_Qq2TB(ZsJ~>1)z~?fQYbBIt#z|vPvBmk)V+MFp?Jo}pBr9(Qj-}BR0kv2oFf?Fn-6hm$HGY{R#rAu_A z@V-^n;CJpR)|H#l9p}?&f4oC*_PRh+0JNB_TO#6}f`RNKOMfN6aXGr$ZM;VpzI@u) z-`F}FFF~YszOMErf4{pK_gO*~i#!;xhuCetQ$?S&lCkF6^3cemCiBA=ge1BubxfA} z&B9YJE;>Mt(oJ1mXARH*{_IEC)(g{KdcNs^22g z@4AjiG#e8m5cq=2SoT6ecqkk2|d>9->mk3TSG>4vF0t!7#JABbrqr@msp6qCavB1EsHPcX7Nqbc(Elg zt!Bv}i>dUWcZEhh>G%eiLxfl$RC-5(NY}Sht>Up4#2^e3#g`QS1SXt#0_2UH93D?Q zMtDBg?L+|pS(+XtbJ89 zByBgp1YCi3MQP#f>hVI8?(a2p|Kt&eK5$i%q4>A0CScINfbsZiZ!U^(mI0NHMp-UFF-zJ}n? z95ed_T6ytaYrt0?&J;#})f4Y*ZNE!ez>HoTp^VsLQb1HUBS;fN$?vkNcV>f;F1?h6 z-8R@M6|Zc46gtDhFkYZ6uzERhJ>Sbd{D6cDkk*&G46Z*>|GSL*yoO;N3Kt0I-4qDO z94J@#XQz*v>v%T{3ikhGG^<1kK{@_!jOHfqhBGe5e;7^8^7@2EQgj~oL}g-yf5aB5 zCKB{IYqrwz35Fb!#MbQVKfMK6s(Ha68iJ9jh8eLC;<6uXYxU|)ZhnMxv z)yn`t;YpB2>2M*X{~{e^xE*j99^zI3+`3y(Jxr`t&SwMr>HHNQuK#^|mTK@Xh0TnB ziUWlACdz`GI+dDkuwE7>n@sv}ygyfcQkFg{6h)mrg)*6LyFGA7{xkK49@b9TvS|_p z+HOczoO|w8dkza5PP>V@!MmP$=|-BMo2?E_z4RfZCq~rryp+6&TfMC{v5mbLfX=Q= z^<_0=o$pMxT!JRkhQUpziWA)r)xlUHx$YecByAIETpayVLzOa~pCaE-Un`M6_5(QY zmTQ?a8gM%VD~J%vQe)iWhtWfg7O6g;?h0m+OIbzZDK)UxNsE>^Yu+oDB%mxbMaaS! z*wr!B#$agc^_3xvuED88^Q>e8z%*bYq+t1UV3W|jsZfp70!h}DeBfsqL-BROF_CRT zh^fJpk>Lb)d+s=ko+Bt%)mG2h`mdj{!z&cC)L$P5;6msMLw{qeXVT!CxM)6AHcAr6 zXdQAZ%R;JvIs$RKM#YlalgLMtxT)VKI%41DXeJR?3KBa(-^zIBR)m5865H-QH6rNA zrG$N4II1Kci&2tREEL34k2S1?lN1OfZb=KDz!KnIBUN+eF)L8OS&-RMY75s7MOx}O z$)`ABl;-~l>kunpH4E-4tCW>9r5fw^(`yqS7$;;pcRO~P0`Z7;)E)5m$LgS8;WfW2 zq@TVPL#*nbTG2l|NoHXH3K}H|Z~yH_$}$Y85Q;T`&3m;|8j;mFqhCNN-7jnxP%`_t zdbVm;iKjXvOmc$Y;fpg)?2-LWrbj4c9AtDBfg`~)PE!-X+o!)GaUm-0n8Iw>q#ZJ_ z;vFQqqO0R+ibAof$~lX837sP+&>_9=Z>SiV4RCCjmm)?X1NJ2Z$l;;3f*keFA5!{} zN@lT=TCtPiK_S2kmR-;?iNdfFHb&?ePv7*8AEx^4@3Ah0LXdvOQc^eKPc1XV#CW0d ztQtcA=Rm>?DWP2Z$V{@8dg_%-ioi=+PBkxr+YNOnOQeozPpbWuIB09d(w!P|?&MW;Bn<-yBmd>_)oQQ;{@Py?WMn{i!b8=UdEzah z@GW);FDUnnXhFDQK-z`3``k3{fSB}t9fFu8)Rs>(t1W&8oqq5~cQNgQ{~-({Gj3wd zUJE91t$niAqXn`}(LdWPYDKtyCycTAU8$iRS|sIj4hdW-Mh)2q5Vn%Q}9zOP;>kvm;Xv%qsb&&8kqT9hM4F?yjx64~x?qcU?I1gO_ zk!FopMX@DS|0sSU_DM`V!!nTn(9Fz~A>PF)vdm}ZI7;1v1_0Uht*2Y2Ho3vt?{q;L z7*>cKkJGTvdrr?1-)T*CN0JwOd(oRLp$8e*HCBBCK&W&Dd>&;)dPG>) zA3Nt-Bqxqk^Ury6#s>c$Q`93?C`Tut{Ix!rIvqAdruA<3~zu?}?X!!jXV4 z6diC`9RXl++Uo@9!R;!42;)GXAsXBJpW$-=1htMpGK7qWGnw^q|jKjro)~# zWqeS<#>hwu%CTeiq&ANunNvc z5;lAsH>p|2czkFL2^K^x^sxN3sR#Hv8kSM0J>TUBO7^E#`oZ#YZvwu4exq|DO8T_S z>Z)TzUti(;X!2=3bjnG9FLJ6WBSRCQDb8(NG#_)Mz;EG4>`mm!Gu1*tNuECau_6`R z?ZS3R??Re`OYw>MvniN0hY$oN-G&pj+`MG)N?eZzJ#Q(D;g6H$7x?&4)X3hQ(bS8d zE!*Ca)_!S8SPql(1+`E3J+IIwrr=C=!1g*<>n>`ngM~FbDAp@(R~63*mMj6_O?+b% z*E8yc6{yrl%my8j!EcN#*$01Ywx&R$_@ki=8t?2GvqWd_5eR3jb_fO&mTj$VmYduh z?N?gTF%6Q`%;S-hp6V!wl_z|=gH4|ttIPb}>K8oNiRSllo*W4*sTRtCREhVD+~<`T zIgWgE>%%E0>w5OJd^Xv+1xyamp@7MdGjrvd34AnBEDwkwse3;p`trDGs0sC9k;|J~ z0V%?~eMx9d)q}MGS~GrWsNomzy6pwvkXx(6X>ikHNbDQA$KKS0yAuiV5JbMu!UG6; z75H<#NE2oob z$MKl7uF7^s)HU~1EHkNRJ!7xo^*|kzNMpLWvOs7E3FUN5G6rN`m-chPZIQUD{SKw? zazxnmrPU$A+a7AeC~3mnT-*U17(YnZ*8qdU=043$%%LnePcZEsq$Kg^fSmcaFKz<| z{<PSInyNEu1Q3!v7P#n240+6Ih~8MlX5h@GEP;#N+-ZD>?jQ#8v!DXOXEg5 zJE>|f>`!kSp)7lfr8(m+usj+zmHU8l#YgiF;2qKzyk%o3JcxOJmk@7yWXoH>+&c1) zDT#9k5V^BmC?cZThIGQ$)YLB^O2$f%k3aK1u+6J8L`V}ym?<}i!tOqM0s6B)c z8E4KkF0N-3U5TV8-@xlE+9`jnpV{{@kOLI67Far5Vq_9B@S zf&MfGgn1eM1UOFT-3jz+7H65jF}>Qa)feJARIBd`Qe+8Bo0F3;gtL4yl0zKX$Qc-y zTdYG?0~vq|QV0?F$B(H^JFa#t?d|$X!Jc{gZ}>p7`bL$6Wkvy9$p$^xBYg0g)%O69 zTh}+0;ucnBFY^VXGOArP#*Wu84#*bc!EgA2s=`8Wp8#VrrVn#nZ|S9Um6e@$Jf$gk zKg|p9A;7nz;a_Co-3Sipy8z7djLX%`-&a+PK2U&u58X5o<6DtmQ4J}9MZ?()CPjgD z3ppl;nLEEG$ORzzX2~T?Fq5D^*Kbo4Dw&V>HF*T|k6re~DXj1&^5ypHQeI4Ttev8( z`oPMDZ37055jZSFjbqU4GRi>!QklL1^%F;78NrI#SP=IO}xUhfJz53%ZZAI2N=OX>jrAly$BvC@pVnph9k6 zGq1!)oS<<3uD_xnusno+MlR-P0tYYNlcC;(ci0NV{j?4WC5Tn5UYB<3t^y5EXkzoY zf2b-Nmghh}992KitlD2}cDD$rJH`+S*(ZR+Wxu>{iTu%lRywIk(1#y5QXRMl_@6Ry z)3(P(Z~vbI-ahenxg>ZbLwtQo{0YZC`(Wq{pRj+Wgy%6w>4?aiqP@S`&W~ZFB8inTCM;>zDx@^n0X#cBNO6F` zCeC706lEee`3`P(N$o(_1qaE9M({pB0pQrnkJ`GEdiy0{ z{n%wXz{5FqLgFNBbTS_6!GyAb6MG*YZwOOAskrCtiF34?)j!ijg3#z+jzQkX*)Msi z=NnE`Oj%uDFhI_?BjTQ zf4MdW>-0R}K~ZDm16cWX0Pl&8PTgjH^F|D|v>94V=ii&*h z8Wu
fu7xwzhi4`fAr(KgPyc48MSv->^##^;`Hr^E65b}1P?jWAS;gsT7@7KC!* z+3>;z)NwsOqT9cuDigCsWd%^meR1(iJz51x)haRF4s}zki$MIPH4=e9ksY9chzn8! zP%wr5b38Vo!9b9ki-!mIft5r-)L}F~GMgja1L@?UKeQ4d9tp%re$;$CJ?5Y)K< zV`neV<^ziZ-LJM4o7pweC^s!t2GwFLhh#)W+)STRcF#GLye?^ItOanGONLLG2Qvyt z@t3qmORWMG-)c-^5ZeP%f!>aYSqsk-CQdqmDyQju)V~*BuMW?c*-p)%`TM!N7U%$z zj!AuOnOaJ&^B#&KjG8SMJ9&WN2dY-MPk2O?={&H4U{m>9*h>)D{Ddl?loPv;URJPh z=W=9aj6)rpZjd18B^?kfrH2?F1v-qsUpt)@z7&j_K03Pe`@o>G9j7rcqp+>v*Gj== zvGg>eha!XAz?_G)ui4%90(#4ZRej@u=Aa^oLMw*uo}m(M8&5cW5i@d1WWn$dHgB+4vg2BgVx>|rQSbHY=sW3f@4u^8YH2XSd zC=zK;egj@04FZC1J{^YC-W<+;xDYy0Gokl~Pa?r^=%3ezD6=|D$Kq>O|3zUD_s=VH@$)ZVl=}!Xeb-_!keyS&gubeeRIT#%} zGIor4aCTar9TTLAL!`TjS=M-2nRUj^a)M{A7=t}s7UGl1x-sH1Hd)aX+FMi^0%uSe zzMTn7?d03|O~i=qr3Sf8#nlKK>Q<00IoQHc>^VU1<<1*H8$YFXNI-;=7U~^D8}cU z6q1Eg8mw%3ChbT8o|B^5BGeTq3-$G@SC`nL>ujvw zECGO{Y2Nm$Hq)w|SS7MtBB`0Iz1Lt@qF7fi!?^NzH_z~ZF~8j)l}Cvsvh?_zJJUKg z_B8<(+$kqpm7Eg>`(|pjWn&{(wHV`9vgZv#D74bg`a zxDXPD$G8+?3)DqUu(%Ou#YR-phU=^LMl7Hq8uy(UiJ2)6DN>WvKC&;UXno{%;93r* zys!FKA#7Fjn7^8d&hE8ypKb0^-12HNVrIc-K5MbxMu5v)HHQh#+pvhBn0*9}p5v5u z%1Cz^${HKlL#a}8iOaaAZET~TrrIe%Q2Yh%l9!`KBJG?w{;+pscnVd;O=5YqY8_z8 z|8z(x9b%m<4bezthVzai^+)UzI8@CI*i3NnCJ^8c?A0ql{f4-zA%Zr@uzNTw#0KO0YMhf3Ff>>r z`th}~(}ts3z3yf)wN6>j_+6q+g2ot5Jv^DhI{p2@q1AAD@AVzN&N=;*a|-d)Zz%e8 zAS$nBY?J0ohTw<(kut=jjoDWKaWmm>3V-^7omQvwXC>}k$&S5MXvD$85E4K=eH{k5 z^?u;F{BL$W^Z?_ah2O26?QZLO>EoUuTG+o%EYKlm4W7+*vuu_(MOi}NttP?pVjS+G zlQVF|C{{?tSy8g!JLD#h&P>;fUIv3ftBh%5(NZ(TZ_q6F+>ooWk_5IWS znki72H@f`+X6){=ZGCpxjk2^SBV+;ip&;jNdsOEzxbb70)vn)!*V`H)YcFCBa)n-D zu=5iW%k65rRPYyehR*hy;iQ66S6e(4@?5WDMuV+g=*`U7jvU?7$TMJk7(wdUFFEl- zrMy@gb!pK@2YpeId~Xvs++rJ*|5xdC&}IM)r$5^QZ0_KFvmHso70ip${^R*S0ab24-37!{vBpt|t`W^}u*iL>sjeQ?GexY1 zgS#qGDY1VozX{*%_;Jb1yD?R7Jn9d%VY)_~ImBkJw6@0ba5ecNe)AlyGqiBdJgLe1 zNny3G<|m+AEujD`H@M9TPon>pds_M)t?!Os`krcL{Wn*vtd+?O7QHL_{)_; ze`k%wJ;Shq_R{tbCTnT=y`*{b@$Sg|#{E58JwN9~e7`OJ@ZSY+M)5r^h3;oPgt5Kv zcI*i;@3L6YwA()s>J4N7A<64;==2CK^~TLRjzRn03AYPCDaty3rj;HA8BFHMBRoXB z|Nh$?fQ3y0PG7WJk+%s|36i13)80zL$@uZH>PS%K40)Pe=?GIIwf6Z{)WX;CTh3k= z#}4q`jFzsck@SqfS}brJcSgWHiOxHr0Bx|&Hz~Lwbg@q1fmqYSWVP#nu6bV2laqZ- zFdReYdKd-JkA*gNZj;#G7xOEguw;F!ZBGKyYidihw86Ephv&BKprgmK>FveVr)nI0 zE1oTGY{VQ@^zH2SlMz$^%C1=_$#{D?#_oD4CvX?)5qY^m(g^F=5yO~#Zgeu3zAABy z4HrmV!-=m~jlSB6YVZ|gR}pFw#LTw&c%vxJD2N3_2U1IUwCs)7G-K4dUqiv6?0xq{ zhY-xbz<8)>ShCrV05N}mwg3VWE+2SYTU=sNx5;e&Xt*{-_q0r^RVkxuDJ``c=||ik z3lI07x4aP$cg{sB&tG$Mj*G>9GPZKhRW(lEvwkYbCDd`iu9q9i+CS{-*7+cr!PZgY%*4h7)US8 z+|#o8crK4_*)>r z=${0oFdb0!nN#F&_3L#z$4lyLm?RyxjY?0yF1s-{-<>nd?xYyd-gX5+*e}m5)`~Nr z1%+x)Hu#WTe@0_yuH z6+}Nl({=4-b8j#kVQkl_Ij7?w{Dl0|5?-KzWm;#EkM1PE=1iF41d_#okV$ z3fO8}_c)V$XYm#})>ioQkUnY#Oo9N6Zq}$^5rq@>zua0x5yY|AimUU)qCWefjf~cX z6d{TbJo8-I@1zzXP^Z9&fk{+r`*9J`{l!Rq7l#MSyy|5!hP+1|Th{feN&WNJA;FSg zlC_kH3#M%-`c**f)xngUBgD<_U!p*|VBwYXyp=3g067~q7+C6cVboS?6 zci?Usd=oZ{Q^z$_fja$zB=Qdx_$bO{IDyHrpDJpLo6j_@5VwJ4+5g4EE zk4X*%>H#=JzlafwpV`vKPS1-TQcpC{zzW%>2z(|}B`Bo4R>ZhHic=(TE{Kp!-7Aqp zlt{jDx)2VFF^{u>NQ3jD4+xKVX;2We8LOIlaUYNY{h1V_uSxkmhQfq(E;>KyN{e(M z+eewkCPRX5A#5&8@jyJ|y{`tZM|8m1g*?&dSONH}lYPXUKI@Y}pNyP95TBwXMNdvn zcZ^Lx00bN;IVp#OP6FN&0Y^RzqbOaE_Dt=99;r!^RVNvpRni;w8|z?k^46)^{sH0A z9IK*I}G{kbHi#+2y3nPACb5J@%RMsfk%X^l! zng*bM5V$y9{X+(1+i=!=gAqeQT9o^QC&W-*OPHtUkP4EcTXFBgS&40 zoc&|QzMjvqWH+|qAPT$nINZAwxF%_THUs49s|S)=Rfk!5XJ1_rvBC7B&${Fp1rZ?U z-pv8uOG@$rrNqV%P@C%FP-7~OjPm`-r}L>A#hGfJ{XOILs2Jv*E?OB*La%ZO7eY{-|m% zpFLo(|42H%`_5OCq99?DSli84&36R ztQEGmp3pML`%9@dy5s*rBxfYi7s+R1vgXDE+gY^V+8hbed~9fSi$@*+uVs!NI3cm#C~(kMN5 zGG*(FR;P#H<|T&U_1wcYBy~f-zn496mx4ag?JA{3%bl;{J|MR&y`aD#7>pRRn zMkOIpTUi3=W9xi@4HtYI4fVKo?B7UX9=m2!T?cjJUq~@>C`7IOo&;R&X%45u7>t>+ z%1s&BM~xpaot7;1v1DhJhejdhs^*OA_`lKmw(rFJIdwN!zU_H^SwP%h8vBX3JYfKz zj>Ox!ojg3=ZqBsRJ5kXO8vCUlBNCVtd``af+c^di@iz(#@`E!c0$rGx6o8JO3ujO) zpl<@%M_4LX*Pq(iQ$S%V!}HNZ$pCPLE=%`OhjNPLeIg8e4kZMz6^s*%wskY+8aPI&)02n9e+pIGGz6?I_ z%pvF$mJpYB7KQ76j*kdhAp~i{RJqrQU3spcFi}a{aFHmPNze%y9mXm1jJDxkSr97r z4v|Z`Vp=|2e#a}$o)Pt@3U(%80b+qP{? z-o0mLf0_9K_xJla)^)B`94o|qE)+y2iPcK)KAy^L&i1>RPz9_D(SYwmuvK9&cot zog&T0CObmOk*EQ1i~RI>TYvoW@((Z_d4K|;Z~-rZe3_eYBRnDJmmgm=cbWCE1h4e> zM*Fnl4}Rpy!^Wq#hBR67a5Y<7cgTF#jFKEt(M*(4S$hT@CV7r(yNMmq(qN2c* zN3{S1sGx|jxsk;d@tbN)Dm6Yb+68nAKf|RbqB7XZQ3+}AMQ7A z@{}&OKcvXccYd44lfQu`HrZqCA5$Qp4D62=*KwAb{jt9k#P2>p&^Q2#qf3t;*%sNx zLbpJJjecD#uspi5DS%StYPv0}th5B~m+}4Z#`I14Eu=!Uo>IzG+Q`!WF4)N}6qEXZ z%(8*~k!L%Qioxu$3W$^)a`|ODlAKK(1S=M`^{*%TrxsIn=JDElkDSm%v+$K6OBJwn zLtCrR=T0*WCxeWe0#f=fn$b{+&D6cn1h*BN`Uot|UD!x-3b{L*L5scZ7X23BKsJu$ zjVph0S`aCI%KCg_bmN7tG@HvjUti_58>8E@Y=1b=&6$+^hs>K9Y(STUQ&7OI<2o3M z&GBUMWpIH2a&_3wyg)hl>Dvg+$Wt(SD?C6x{*W3ObYK)|)VI*#TprXPT0!V#*M1YC zCrT?2#rFjj`ny5Rnjr*MiH`~#!z*Z%6n*JiV3L;C)XA6_q2b4=pTq{6{Zlky-k4$2 zSID4A;U|_E{|3|!c+uJD*hiE|hc{_KVzX38VHbqU2gDX<2UO`d`=3dbYF-?GJBxf9 zPY;VrVY47gQK`aH5aFQ_I-9k2Spiyr+3Yer8Ze4R>`%uxQdZ7WRHi!&+H!Vx^ zp-*b>Bl-Q-AxFVWT;xq&YLNq0GgZ5NUW7*X^v#a8c9>x8Ag+I6XW>5KH5lA52)SB?gQ8AUdT+(78z!Nd(h6 z@afF>@n;H~d0YLRGc$nCudFI+qcqIF$54rtTwyi=rr5~KPDEMWS59Ip!#PFMe};Z7 zJ-YOtNg)is;KSg#8Lz%OPQ~Yn{05(y5MrdLRN+rM_!*yFv@BDtCgK@$TdH`CoEnYt zJ>JMC=_GM&(>dTFzsyUa<(+T+a3!U5(R8EIL6!ZpAnI_lQLPrZ{N-pK_dBo462)nM zT6{oDq7}pQur^$znQDe(Cdo`#$$~u`gZ)kvzkXyyLDGx?o|`=^xBJc`8I*9pa`X==lJ}_`Lv68B4oWNdgLThv8S@runcFhWZkv4md zjCvLrg@3j_z)hr*px4E`m0Vqw>>32~#`d^cC>&d6UlFEFm>l;VLXM;r<9!JCv z<>%Zn4qC471$0K$&f*cl&uXW&5TH)ExmVamxOd_~L^C)C>WstzB=%$orHJ~Ngy6vy z19#r^Y<6!RC-elsr7niiM{f5hg$D$u{v^ z+;FknSI{Vdc;dV6xCohAI5wJ-w!MyTDxEyrG9-`1(U)hvS@5gj1(xB{+VW>Q9sO+q z@*nS}tkW>?$Rl=TV>DHaK4f={Dc-HJ+8%)%+)$Kp5$sho(RKw~6?yN%b>7~d4ipn# z$Di&u_x~GZ8dvEafuQrqM#_-pygTs$)8KGdda=U{uaMKvw4p2i_EQc zMSW1F3e*2lnv5n-)X)!r293STsOrI~_#tEetz}oLc{i$%HOoOY?fM3?j<@D+R=Z?Q!DE z=V7&)Q>~b5>13Ig4C`PNOfuc+!*LQ+`|2dKAbJ~qDvG9};TJ$|EGYfD=b`a_1WWgJ zjF8B}L`@7wa23N|gJ;@1!ql!u8(onQ-SYx~c?#iF--ee)C>bLwi>(h*(PAWG+Wq)c zvU|{Z&AkT~*qad~(yL{toOzzHjfcTKeP&9aDEac6+-qCSxAn>XMMW&~LBa9I4@Efv zmpTRIfLe3DnuX@)`U}c2T^5lu5(=9Z4rln-{m5B@=Of+=`l}UYVL-CVjIfhs8U;P+@DJx1Lpi!|`fI?oV!O?UHOxb6_iKsD!H`8I2st$eO`lCbERE zE*Cgg26YA@z)|xD+2s_XJ;;N93_+ZlgY6r~cntv(=Sab*I*)@f9-JZpWbHH|VFs>d zvBY@S@kqou@%B-O2N?}4Q=9uDZcH1d`(9lJR633w ziI!Bpun%fPggZ%ahD&f)C!C&dKPq0dP>kjacbHleZkA8soQcIppn>tF?TrvkEJ^kt zh{w#pF4ljLzOS+7dDPp*gjNB-kii$%^*d2mr@yVmFTk(0RL5YNEk`a@j`xOA5CEq+ZjfckcMz zIeHHKV6>MYY%a7CD04ff*vq0w|{P^oxHAO`Lp6}aR7P0`GqEAqf#3kBO(hq-qD3Z4lPW>GxE6Ls|lTBB)bAOLx zs02um)gE$D?>HAygh&-&U>}}oT9mF_Ty^1IGHYV5KIa9!hPi-*8%}!HD3One)}K-f zwZ_H5*qmzT!G=GAdOn5GKs65kqOM6ePX=J4m)WJ~!1O6m2(N6I;7S+E8q5u{+7Z;z zCBM`Em;~OUnxdgD5m-!-`f~M4zdbno**tvA`X1syQLGK>ofZnPX6xG$o)=8o3j|3u zF`HWM$qL~5=Tr*)Lq@3HTrNktkWC18xzIklR48j+T^}qkiM%|jq6h8RqqKuRpe|BI zSLMDSZ>pqm*}$xI9gTDouYvk3AQE3{iI4izN+Mk@uax-7WWCcGTeW94Cz47vu$F!P z9R|0!O^zEC@KOePXA*uA2j(^WC2>p{92L@_`y?1S&zIiqFUebkr)`_yE+W|tli~2uNJ^$bVX@yj}&Jtw@?liv*89aOp<| zMOnm@pSJxwf~wfVA88%IDAzBwy;NAaq4I1(^Be_RCBO~%b=CRlrK0EQWFk1}$34&i zr$uw!5Y!&xKS9Zfq<|kG^w%_-Z;q@2=y%>=8+5whsdvn`jnAcJxi2Wu&EJ2iDGth@ zgGr3$i<@&nJ$MjPcW=bbCj#F$Y7F7wC?q3y5Uxy#2x#>3nBGP_ZXoYb;`!h=VzCKPanEo85wC_~1CX|by zf;BkFn)TVlxcA_w!LxoiwKE4QhC81Uo20A~F*sRjR8b@`ve_+T za7Rfd{6~ZFT^$i&#NjQTx^ToO@fs9|^36un8oNvEO1(gRiESBV*9+%1YxqC~c$kk3 z-QCRTD`DJAYkLPlqZrk@bSnI>iqqZ0s`*2I7EyMf20=-tE`tV(F!%Nsee1(A5Y%!*Dreor_RwaBcXzhf%)TM zA0b&m^F8yA{db<$QHwjFDzMwZw9*>_EK6ki z>-3rz>?hDZE-vTqi7YHU6=IfEN@a1UH{V3>8BfThUBkAV;7cJd%X8&7*jQbI)_L`S zm<-cjD&JOTcd}PXyWH}*ytfW|yQt6D6{4imiC-a4>YRozxb8Eb&z>oMXx}0P7f=gm z=Lg{jieq@w4R2a^wYK@kqT)Zn_t+wNF7yD*r-qa1?<@d~cn^jytw9mwie=2LoY?SBMqX1HyPSwTc@)!8T=k$cN+{Dx2IE{$s0J za)4J1RC2a4!U=whE@O(*aHABqYluc)woGOs;--|aG-d$!hjmqEIj@xs zM+b52_!mHpv{MRTSufsH_~i3kb2?la4DaNy3LRgZ2)}uC@#e(zy!;?+8-y>kZ+T9T*dyKONi5fKC-a}ZyVPF30kUx3hu(oUr@QkeW>2*5MgcRl+dyQ z-C&czm~p$W?2=e7@8R3X!LV~5?X~#B%cJCOgc{+1Q)avC__`M`Yp*t&XGE(1M93Ki z=W6cQk00oYbu%#jh4T`9+B>@0>;9EiIU1;$TJ(~>q3w?adGFNlHm-Imd@MJu)xnm3 zBi_4JZr$3}C+Fob1`)U36G73@_!%iPlAo7S@O{yz)4q8_W9&J0vYDg%DchW9U`D5q zWUiFx*c!z-`0gHntXGiig=Ts?#6SKSccn!^Lto>zoXJlO$!ayGnw@i{SwrNi@(CbX z$h$7HP0x6@8V4u+M6t@2^jrxNtzZW2UnEXe2q2*5cwy`Nbau6CwZ#(x@p@E8_NGD> z%A*^NjUZz5J1YQNTl&RdI}O4nT5em=f}E2lf&_Vk;x!CV?s49HiQgw|8PV01cIG6e zlCb{`wf}N83uY-|B&{MF6 zt|-r5?Uw_vqJ(r&c;1nfGQIrwa9uRuf&X!aOH@z%YC2+urqG-%%v3~Eh>K{%k~<1y zeKeNKnHiEWOi&7oaxjD$$=y|cmFTSmxuEtw{LeU0V=z7}suc3G-l3c7|jDw>k> zF>|fy@HoThoq6(r6vg;GxpM2=hv{~jp0TtDTjmT%K+_K0tS^?s$5?!sZtuoslP7`Z zWf_%WCbo-l%uI9$qb+0j$=bkMx~A6SZ@+E#6jnU2A}!HQL?8pvp*Y5;d=%d!=9}9H z2L^HxRX{a}Lkf!BGB74RiE#FOd3t7up58usU!qo8PtqhT5wbNFj^`v#s_kW3m$Q{X zh$#*j(VZiCe46>jh86OjRYW~$G)pGEgL}KgPDDRkPb|Eod!Jv-mosAM-#z)$%-Ga$PzOK@wU2rb;6aGfU8h#^hXtn$`|n|bkbp-16#an5*N7vkN!uzN zrg?h%_`U~w>dvj~DYb8iWs~UkCwWFza&#D=rn|{`24?N{j$Y{q=dbReXuK2G#u`Bx zz<=8Y$`Q>Qi~L$JVlKG)UiJ2E#s z;U`<}6SxkcDW^xKbsRqHN!=ioK4~k*tW|EhxFktYP)(Sb)Gw;BoFK&xj&;^t^;)Bj zT8FU%<-^NR=7YC}HKrh;_5 zF^BGoc`t2r6ImS1eEY=+tp2_rvFRN!RUI1NP801Au82K8JIFr)d+eej|f)3`T(dLIPU30HzD_W*goGrz+nnPjuo%GcuZV5 z7Scpq7mgA4NaI2J9m^e$A%c6&6xyaDMqsv#WdtPt6P<{W|H77ac8q=fF2ajn2rbdGY;QBPKZ9myZ*f`~t0;X9SeXdBo&Qn(8ni z*7N+GeDC~xlS+5uu&)PoyT7Nr{_v+l}Ih73Z1*gO<`v$*_OYO)o+g@zSpxg)}LI}z$M47}>$ zUj=u3=npO{wPtdC6pzke@$5-+Ic04Je_I+)Nlv63iq3F;D>lZtu2O_I&%Fkb0) zwEw?d>YmQ&f7&4*0`k+O^YIMQO4szmG>PUWlEWoxH7X5lnell! zse*BRFe0QDg6O+S^FKy^to#_xl;)iJDlt2haTS+~u)*H#-yJ)4I3AAmy(E}F$!|%N zNxqNj3~?%(tz_zhUvhj1PBNI%T8qZfL>uDQB`$U3h3LdsZF(FUQz*j?1j|z8RSOvpSDO82n!sBVJ`j}?X{@m-cME2}b zQB#NJ3Q^Q*4k}8~%iyAuh>qMV)~A}iP{1Wfs4k4(sjLk(DHM>3P0PlI2V>;QP>9m8 zW}|WOOvD%%AR$2kCkW0QF_IwRBVPhSc$%QU0gWkexkOP;Bu37;I~#v9M5~p&cxAG+ zM9NEKgpcvaQ8P+?7ZwSaRY_FPGLre?m~f105YH$%T*fGAizQ0zOF>}H- zyUK_YGj6X55v>uVY>9TrX0|Loa~qXR1!dQwfsRQ?+H@qqG{u=%MK;&x4~dJtl>a;N zOgAdmnq1Eug-v`%u0QGy<^%q#6q1qVk4(lJYUtoukSG&GDLia&##~S7y#R?wuZBuh zA0X@OLSLZx1CuK|xiuaAH{mvwrkO#YGSNyTMu$17M8ToPANcPI9N|7;DN+u+guXPE zB3KxrGJ}FZX|hBz$5U+?xNvOb)UpJ{bR`q&C?9C8Iz*8Lf3>>Fl1ccDRu-RODBCY+P;yBCfnBJ~IgK>L$i3UL6LOtl64#gAZN7bbSjTQpPO^l=(=j$rPG zn4(uwLUI0Nf6@2rWO%pFBwmKq`gAHNofGexCHGMJlt<&}_dc*rnQ#G=T#PIuV#nOK$} zrVtyBU~y;5aW9~*r^4?NK@bctPD4F7St`j%KT1VEmTno?-&TObcBx$|49xh@gS&$ObrP5dORtimxJh zS`ZT~^L2?n2RywK9tAP2MsV!lpp~X z0TQ-sh%;`6+?4z99AhpgL=xYYlU^8kwf5`p#?E;EB)*e~OIZ5}=6AV&d#I3%@b)!~ zsh>Qd&)^xuQgbS1&P1p~Ak#Mt@lOaLdJQweo~#PrH{U}JtuNeIWY)@JC=Fa(B}7b; zExa!o8m6|zkj1?hXF$2Tgecy4!x030mI%mk>o~gt1wh2`8hnMq`MG>Sf>pZo-265b z@!5W7_WNh&>HeLvmH~BG20_b78!=eq&z1VOI0jBY7H=*{w?q5qODjL+969%sb0!B1 zg^(w>m`>>u#q;>w)@>0y1428=bx;uoE)XQq$x`EaqJM-ds}GCT6{A&6U1$d=_UqpK zFv++?bc#{`akl?rRpC$h{vp<$2g11fR*!4N!C;haP*ic{w_}g~EMU3{xbA07Bm~V| zs$`NrhQ^1DW95h;b>E8A>OQtbE$zFoV<9^w`#cC{=+D2;ueKCGZS)i&loT0aB?Od| z%rDT-ugWSju@ZO57srsOPeA~|imG0QjRK%Eamdh9Sxl1g1vHf>U$+y7F^R+xsO`>$ z1v-WGd`}vRJ8bv}a*++3IKLAPZB_YP)XL`k0pDEvrj@XUFFQe*$~fjnL;01*tdV;2 zo9nj#6;4l3Vyno_aTXa71UfpvS;!I=zHmx`Xl;vvk`BdfqzsJ|nP(zGQ?UjTv92jZzgW1=z{tMTna&3D(h`l^o4R`&S{mB!MTPZ;so@O`8ZG&x zkcIRpPb7QA=va1{FmtIuRzrn5>eNbKUo*M%#wMV#{#w1*gZm=)a&M_57t*s69fF%EZaB)g#gSSIY!_JWUOo zUnZR*Mm51${_YF+>8SP+7 z-({oww9704KHx&n*_eH-b(U&uaCJxYPuQe=s~KH4k6l=y0>WozuRaIv(7ioR&&Q@m z%eB8P7;C}b?Gr58O1;#0WZw{H^cB17_bmY9Bn;l5W~f}`-7zJ>Y-ACfCsk<7jS_Bm zyQIb3dW%|E_;P;%LOG?0CB9LhOJ)gE5y{Nil&1h)|@%rR*7Zw7!^ICSl>Dpc#IKL; z=+iz`r&A#;X@jtz9fR z+zyP*;K4zs>RGOvr(oH_q*LA!WozbW#)~D=99$27bjNT#Uma}G!!wLIt%#=Bna+o0F4dk{)Ff)41QMd@@-&g$O=DQ4 zjQn)^BmbwWdg+5r=+b1Sml#O{Y6**n^eC4Kh*tRVyHq6qh-NX4RFXuam_VZ1Hhe zp2ySF*n6<4q0q~hV2O-?u|wzMmD9}u;5a{u^tSKI}I-_CQJ2FuK_z8AKm%;)GtVU?ozGUfY+pz zO05v!bC}pseSBjAP9G#ttG}50)vV4MALF2Y#+#7+e(nL&6TjwW|Kxi#pZRBmR*vRD z^-Qave>#9eCZ07!@qRq7d~ZB2Xrxh3zbMvn?z^==ZFD=JcWIQ*`uUQjd}`ffV+cb-vA~z zAP!EM)>&c}-K7p<2yOMpxt6-@t-1Qnoo)u1_vP^y$9PU2b$D-B0x8rPu49b$>dp?y zzt7vx_sdTJ#k_8P=mJ>AwEq5n@uv-^c+&oIK}6LasTivo9|Z8ltJh+y`O(?hs!=R` zivOOH{?PkxsTiS7{3J?XfiqRq!VP8BDp+nuBzr0|B7Z)VuatzxO+iHt z`Oyl%@lp};Q>l$V@(hXbo%*}vQC~I551VFRr$r7)2);BG6}e;ztXo)CLqj}RMkQMk zk=kXP-%?a(+=gLj6@Z4(FsK>vht@N{{lx6zvvUNmOC4!!fd6^mTr>@{E$yu_~E*$ZZY87_eJ8{J={Lw_j|y&x^9)_=h24@Y1g1 zT8_et55AVH9a2YOy}#5$fC-^(3OACC1nOZ1v|kMlVsw>oE7Ca2LkYzU8$ccHjsM8m zT}w5kHdHLph;7>`I{@f63*KQn&b+mY&}BA`8?uMF<^hV z{2U{^+0!UQx8KwaY5~rT&)PlQ=v+3s2oHI7n=#WiufJ&Een>cF4~d%hbtv5nQ$RC% zWLo+vI(+|oG7a`rEJh!l2F*lK>dT~tKYUMpa~_Sy@~(GSRO-2TJ!s>F=j}MX^$)tr zmcC`N9*VvT5!O#DH6v(}8}d=*S;%6fj_|_&==H7Wp&5X)Siqy+u9+S%Hrj0*Z4er_ zZMUJdXlxbq43)lZ;+kCoB&%u5pAw}JywjfMcufq>wt3ecf{tL6dy0|!Izj*$vt;VA z()P!bBoe<1Xj~I@C~mva0~J>1Y+(RB5bu?%Hm{w+n(tN=#HCfL^eDn6WFx{QV z@NxGfc(V&ZY`CznqkV-6WX}hhQXUtnJXRROoIs}-%m)PAm-$)`BU`L^2Dx*XRs>$R z=IFzjq6jA2a#1JV_*l{s?ORJ-92zmWwfU^#p@u!sFtUsZMg;2_sPuJpuf-})V(@_W z-`16Mn>b3^DT=KmBg64fxs-2k1HgQ(A>?m-iuw;BYG#8TaSoBuQ@^Rs5|$3S#N zM;H&|{WIc0M!GSc=+W}-to(wQe};* zn)DkL^M!HUJ6qdnNR$HbaG@s-1E;S;_IhD+sZkK>gx9ShTi^QlRZ{d-mUE2=dR%_RG;cKe~76zLDCZ zRu!^FzhUR5sz5C|;bNpRC9@nFBQJay+Q^w}2#KCSr5ic{uR*Xj-s)^O`=4Q3D6{XN z?rlLH@4{>6@03CX0{wSC+>|~yWj5_u>7>*Q0$C<*}Egf_&=M)0a~Rq$Vv zE95ZlF~M&@&vo9(7^ zfzOtkpW+jdJ@5u)e=%&v=1bm>BFgGbk{7F>;+t)R5?er)0^P{VFqFN&wWO;@=zaWn zM7+iv-$@8$)AnmOHK5Vj)*Fhs zn;fVJ<>Y}*`Up3th&&!Cx5QjV;NQC3P@tK`y>z%aGQwfwxn0^msApGiQ^3;HdYe{7 zma+Z@5HkHRahUiH|A5qcG&=b!04ws$m(TIZ)X^ycII6JTar&bPUi^wLQBh8^0X=|) zCCMM7yj{KcNFrEaT&}+D3cgBi?YK}~sy{QWc1N~t4Xfaf+h{hkuOy2_w?hx%+?qCibj7kj{+O#JYfk zz^nKBM~w=d8iVGOp03@4a_8hGnIqxvX#(k^8fpf+>(y$l5wq(nR^2d2FA5XY?@&Jg z?Rnmj9=tHib?NE9nJ#GM5OMq5Io=yH`Y^Y%3rvE;;(h~lH@LeAsZJW)>KiCM>@9h& zTUT@XpGh1^!o{--+WT)E>*ckpJ3jWT_Z|-lDu4BY-1ZMd4!V~pn7)}dX3s_xj=#cR zFT|{Ttw31vaVrGsvTv}u(8ZfmR=2i*o1S|!X_F<)p!AykRtZZ?yv4m4Ol}Wvt9Mv} zhmhEtZqD+S+EG4c)q|e(-52u8pC87_)NQL;@^eZVzeC;*+T%^L31%Q!ZpWu^QbpZh z+ZU6jDdgnV^c3n6^F?;@#n^GGY2H*9OZgCs4t^Q6crKZppWpcPqy;e1Zc}>#vWFL@ zNy;W!>PU(v@*&?t+@*#4{-)oGxiGGJ3f8R_Ty8Jx6;xH+5r}uVyQ}G)jITlkT7{lf zEy8m4yJcDhqj2Y-EX*_{9yNq8Mg5<>c#9JhX>LtRc~4rP^?|RuB;)u-%3R$ z>PI7*n+gdV#*|NTlH{*Z3DH^sKAzW3^@6uoQ*Y^Wt)6Cn$db0U9|rDUrYZ;i#~&n?@47156Jj?Z$YZ;Jusp8G7%+_k_MDeH9Z^5!qx}Cs$F^k%RRe8 zEUWAgm`JG}p4?Gr+0IPSBD1u17lBL@LePmLEr$58fUd2 z3dBPi%H6Kb$k7GFJW~@8qT_^xvhn6gb3VL=g33zqNTnVc2uGbQH!VWT!f~hXrdCTFWLWm>cl-xv%D4b&B3o-u;>^t^aU%e3vu7XR;|_* z8K&ky#`WU&!~2qSA|XI-FFb%^=s-QgVkCW3E1fN{)|zzxxXeL#I2jIEI)Sd{bF$(Q zosG`Sulh}ebXI`q&>JvBBnMjG3-LCxJyl2{`%BIGPWd*?NR5?;)hHXXe4aRMSQq-P zx&WswHWeb0)r|ybyM}u2c-4qwULClre!tPMT*^jtkIfN?XfWJMyHtV16|kY;i0nlm z=SJcH$?eZ;j+dG}O;~S`X`_ObE#sQLL%A<}D7pdbAFH&zpiJFJ;!rH-U=Jg-7wwlE z374>Y_(S%aF>!HoYH!=PVI#4I4INeMWYKvb*FyN)zGnrH*7QPs{QC86NtY95>`^n1 zwVfTk>r{euY16fTiZ5wdzwmd830Rf0__@7~%b?xr;MIjo*gLB7n-)kLe)Qx6*3A zm=Yy7XtpPMY_PH4P`NX!76|TlQDevFeeb>$c2pvTWvsrrLh)XliVja6ZPla-oL9Tk zIbyURDk&lM=Ex6VY#WK4DJu&Op_;3dOz{y6McW1TN32rKE=w3-qqihY?jzY!3c%3W z!IWA`yIDzq@BYIlXY9;U_!M*=o?2CUFc2yneIp(++bkWruqkSojSnK^qs7qWzq2oO z_0flvHy^WBJQ$=3|3T~iVtq7PC9+^&w#zN%?0+Sog&=ZW?kzoLV9sO8-j&v|0YEn-Z4^rl>7Lt7u|@s_riflvQV%Ti6)+2n%%0E+aP zD{0*3MZq;RE;-X96&7Qf`YekMMw(7XdJlm7{m{}vh)O7RA0`ucBpHN5EO5NDYdG)yG~~*btQ$rFIX+i1S-$-I=aRu{)GgwHMK{C=vm% zYuzBPM~N#+elm?e)_e{nvWR>|mf3{VE2#@2;tD^};6 z4#C!4YXKyO(}6RUV|gyDu{NKeSkuT}3fVB&(GLTfLN88NjU=;{7O)ZK$V`F9{(ol* z^bTsJt-b2g*VEI}w#Wo`NRd!+yWYU(;cBk|HUg!g+fo*v@kpODVP}S3_=G_Ck4z8q zP95(K+;)%8QjcSCOS%7acp{=)((LPaS@R!0K1T5#f?joBoHdl}#fKxJTE|R=eb(E; zJF4fTDha-fD5iJ%;S(F)Z+8!yl8hH!XO?ihYoehO-G zaqDQQ}s!+zUpGjE9 z_WA{E-w$|K>JK^~+sGJ7Ub_063dk=!@yx@A=gdMv3<$yRdPfz|^zzV?W(xyM>t^G; zAc}NARbU8-$G$%q22p^M2CC)Z;dSdx^R~-2Ru9>21q@S+V&sRuPg!Vsn8wIIM?DPl3=a$>ZT+TIRGymX+~!jE(fj1E?MqaDGmgN}eBa$O7Bq z0yuzKn8x<;h6=@_9vjl*%G<=_(e`qG%5JlH;lZ|w=}cToUul;2LevbEypCyL|Do5l ze)HuN`y$}%eKbb zPpFQ^1tk6$(SEK!ODaPueQao2(!2?&#Y8}8^E{u_g}4ygWoQ%)n)?`9Jb1@-h6u&C z?I0sxXBdo$H!`?lYi$WO>{11C7G>-;>Am2?VtUgAu>><}Mjg08BugjGXv>o*PxVIw znpRo9rt-4M3bPCIIJ1&fqY6UEZqr$b_ll;>0Y%~6JO^B8si}^912B48IhAyg3MoK& z+uXF$xQZI_E&P!6T=`-v$os(G=Q~Dw+Re+2iscF1$Vd!Sa_5=RA}aSLEoZ4Kg}1N2 z>0(^PQ*_V~>xjQ#UNh2#|7mLFlAV0Sn}U0fCd^9Bi=7Fab-%YSC)#F)0UYkn=?(+Q z5UE=tIJb=V*x0gxSLenZ06#BIp=Y74{_Dygkz=4 zW4-2=|Eu8x$1C!8EYo{A&180KbYqe{?cuRsdYw3baz8Bc$&J*6ogU!uWW05_QLH--fAU*ip`lZ)_;G-X#6Dqc3Yf_iu~Y2=C7ivKRlHoGCWH+ zY#k-8a4|oS6)BMqP61S6sqgg8zj99M>$RhDUiO`jOJ5(qmmW5%>t3ITwY4^5YAl~o>2#Kc`(C?3V3|jKrP-nNL5H#=wmsy(TOwZ_EyrbgF4P%!p{XuS z&F!7TwIzPUZ!3?zOQ4B+i)H7d9ZJf;l4z$o;0vi_()`z4lN^PoOh*D#{q#wzuhYdgG!b>dGomXR=iIp@Hs$h0>1+J3-WQbuWy2ixOiJ%g zfhI7b{L6vF+&g|Ko4!UQu-JqyutUY0aD1-nQ${0g9vni=@}6NQpUwbD4p51UjO z21)=vWoWOR#4z5hUaH`XNyrIIwma5z!SEyTG?&YWZRQni7jzn;QW)ML1cwz;=&t&9 zIy`kL(&gS(tK+U?vi@!0SM(hK1LoolFLbUJ(%B=E+JK{}+-{EFdeTls5T{8JAX{lb zQhOq$+XOfrw{Va9=~6$?M^cgb7|($l@e?7hvXG0z5>83?xRO$dCt7Wk{_0_oP?AE` z`Z|5cCv8_mS3pUG%hF4z)|wO7QW& z&v&{~L_fmzh7%&3HIkmUi8Fv`9;72u^!pI4hyuK>>M#X5%_<5^B#F=&9)u}rTOulM z^kIsraP1~!1RYEx*NIqbtOjX>`?5tYDaGLM0_zl#2b>IR^U#>f=uFBvzD$>5AjxP6 zb(ZNWIHX9U2`?rML^c>6{ozVHzrjL?Nz3W=*LxDo#t$&M{?RQa`eck*OZ-qc=i{+mqR*FP38Uoqid(qOFDat5-%dL;5OeSx1x+^S@QIgu67$wf zh1HVcjJK}P@dHiAF;gpfY)V5Z&VCa9nF4mG`F(Wck^e>9D1hrvL^$`~2SZbS`+kzW z9w{0%JZ2DNfQ}0O_uCA<#bUY*+=H7GP9mU~8Qm1$SDirBP%L2r$%21~oM`6Ebk`Ln z?k*uzCxvQLbY);ExHh*4ip9Z))P4RHuhJY2UjCVbW zHkoBP_INj%36E~-cI-L`OY*q!b{>>_G)jo>ibBQanCBq?H3qe6AeFnPbk@n0|M?;g zpd5stWl!rvl+oD?z@pU+kf z!CO}Ui?4HR&Ma)uU~JpAZQHi(i9Jc4*tTuknb@|?iEZrcTeVfYU-n0wb02i~l>tlb zzI0uruD{bSX@yJEej@V{sW_y{!n#e#4_PS`tad0s*)RH8$(Lvn2!9ZK$>W57DrCx! zFLknE{e+T)av2Z!`kOrMrsH9LS63UWI9oS$5anvZzg{by`Whg0i6x)=q)3n_>mdos zvSuMvsl8pwQKKPxwrryxd1_8{oN)_u#@`3Ia6d$vpPR9TaYI#QoYgbJmbxLn^>De! z1;Z_!dvYC|JU_a7i5&#~A$?zu*Cu4y?=O91j0)A>#tjc3kIo?SKsha5OV3)o2a}hR zsMa4(+!qB3&t6BwIRWzgO+;7@$8^hccqR-o~97b2_*2AWV# z=qzZQZ62CRA*Ap0Y)&QwM6VxEA%6RL!T1x?#_X8cC9r4LV!%#eW8G$e8870zCf(mM zvLI$egb4+3cS$m!KRi@MJr^6R8+nX4WUSr}jT_;zos6(*G9`2sXElD#w{qKX(`K8B z8N+K|;emDG5lhg_K=4mljizK$=U1c2^7Mm2!I;n;9M{FsjOq|KEv6uw<6$s#**c|` zF`yBElz$DELxmH16=i5xZH~<--SpP^eQt?pvcC;@6xJ!^K0Na4LmB)IYC;*F`jtaP zguT2_!|8eq)tXqpTuoe8p&hc^$Ma?|lY!EBu>7DyPnH7O6d5#nNXW7O|PV)LT zQgRvok>nm0N7RyOf32pERDH?rps5XVyW(0jIuz5oXsH$(OMLmI>E!)4%_3pDkX;=WMy>!;gEw*ds%(NmuMlQoeL~ zbsLJk2Y^{&5v^|0(rH`zFCo2b7s3vKFBPNVzC!wGN?kZ;hq1?fnXDirH@+T0ZC7T9 zmza8&t*aj9=A$bsr`%P~$~_D<4Hgl^DX!*dJrl9Ub;fsnPnG39n3>rO8OJhJNPq}f z3cABP@1;vr8!X`zMM*3O02Ae{*zQvsl?a=*c{iXyu?03WBJj?2aAsn`ZnD5BGF^&e zN;%GoaXC_1mi^06&Q2u9OfgKVHrkh(y^rJ4@GD)WLR!R}Lwt%_eWs1Xq!P|F`VnW) zCCPH-o^D-5N3@V-qOMZNlyb%*Dq0TUHqiu*oCCIX4yoE15s+I#2}}Ys2Vf1<(ZKG9mum zave(s)Cil<&*>Kc4~K6k^w0_Pj(`zX%p0S-2Le~n56KWgWiniH7`d7}9eOA?v!%|n zTR5fw1CA=IXwAxGrsyZdVVX6^!!YtQI9vw3l+6JiQKX97N#B1%4zExo?ITFUweOm6 zf05pZ<>u;LCNJXK&HlBoW_t(xz|BY}3^A>9r&A?8X7<0uyMC{1Mg8SCGYF?t90vWu zasBYk`O#8U(3S_zO5Q)#U@nx)Nx*o@*AvLWa?XWv5Vg`*@9zFSn{ru#n2&2VC3$3?p%ZGCUw4i^yQ5#$%dxs(Gua)@tZ#+Z9e zmyE5G%4|^Rks1Blt!*p8S+N;wZzV_rafjq0E+K816=pk`qpVz6_R&{C7VT4tAFMrm z9h6}7)dTA?146>^nn=he13SS%B?6*G`!l^KS7t#})89n^t6cwF7E`joc7k5+@OAYVVu@?H;ylk3xZ}&INAfg1g4uUx zjx`?~mN|yTiHX~oFddzNecu$Nny5mb>?mlwVZD8&>p{%wrcF{ zx$+jVu!GsMwPrWlrIrPK)4&6b+z@DP3z;h?Juf$R%5_7P`*I`T*NCy99gWGab5~Jm z1(~H;oAC&is`=3FM1j)GH)nkdr~o5nM^SUMY-;HXEURCB(wM_8L&DT@@QI)@x)2n* zg-r;*_FjHx5mqA_!=9HVTWYY7Z?*2A(One8wwmFQR_m3v=%2;NyOm1t!YigGHrcwO zLhPYV(vU<}wf!Rq`FeY7S~lFHyD;TnZ|!qZb?3jpa6NU80<|5rfn2|r?56D-&k`V8&qN+JGhbN736p(TD0>5*Mjn+_Sk+7F&w@fN3GR_; zCGQ8MsHPj>^0_DG7<9AX;*Vz&Ygm@p>1?yHXdom(j^Q+o!z1W?4S$VQb33({&b?!n z%}3{R@0x+k$!w`Sy?o+Rer7aUyd{7uVxiiZELzgdsXu6^dUp~<%*u7ux~_+3q({dn zoKnmMGLQH5v$}y8Pp7a^NB7IBh1)@DeqoOY@oEBK5yl&eRF|=Y zO;e#_E)cx8dmi{VY@6JP3Z(xZ>Rzd*kUsWhQM)`j9`UqMj)V8u9*b2QeSR}nCfYv* zd+8E;9~f))TJ@kV9|uaX6wCGy0$}tp&C1!dNyTg>gxLu6v-U`o@q{5vq(Wd&tsdO% zHq5br6B_Je;P?n$;(C>(mn{FbDfrCnT<)NXY}2-o13f~^3Fb|(8&K5|o4Eaod~M5~ zbjGK~m3YJ+otiF|9xur~5$90WDwRQLlR~o-RT~4%h`VsC(u0Pxp|HiqOOMEA^4&98 zG*E0oGuM*~nW91#d0qN|lWd@26EHjWrlC`Ss_O3o3xu!elg<5cP3`3-(;Gab-=cr4 zyVThaKA7A+^N6Wmo96X27REDic_nsZ4|6L?JurO*))V`5e@ohNs^kAij1FOHqVH{i2=W*bUm32Q^*4UoAy6G1#S-`bJ@0*E$;}xo_8A9zsNJvxw zWF)u3r=T<%A*|cmj%|X)+d1Bs5OxxZ`x7c!X-pV?6_oqMR1ZSs zX^=+r)3Rd^3C&$aS$b<ECZ4rgNl*p2E+2K8sSG(=>}^R%%Sy*{mSXGtuC8-Dym zBP}Rc8k6SI>r)`pam#77e1JE-fj`?H@SX&J$ zHKJN<%=hQEpz$Y8Tp_A|HMkb8*J^{g`S{FxEGy(*xo(DZxV?mCZSWszG$06eg_%lmd8Fmv7hDXwCc+El&ubmK}OlFFG}9`Wa_nE7gZJ|`UN zdHPGuEsavg?N3VfE?>ct*i&D2?KKc$Q&IKG0<(7B?mss-jklNaS*dGw*o6*9q=q%z)Ty+mQ(r(-V!&@Smdxn)Ae z@e`W#FKDDyhH0?t>vQl)5H_5sr896~=lgeT<198MfcOTGC)0b-aF7fNmEb#kq zp+|Gj5hXblbs)=Spl_Q5Sfdnv?>TljrDgBd=6*{}B$T~lcBM}?(VsXUYErO0F~5}9 zVA7^m?oFrdo#6ofMCkc6!RMvbS5wvLU`o3g?bE@OfAf^wm{46~VjiPXh&0beO+r9+ zLVVVPMV#1T4Fw?I=-1tQ*1o(R9GR5CL^fS1PfvI#u(5w?F$E<7h9%e^$#pA#KorL# z9%@oi6w{kfekAo%62p9+EUCqnzQrLS@@^)6y}#=WFB~t!i#kY)(+zN<;)D((d-m*_6a`) z2Bckd$)05Kb+StVAR@~R$^|N82B*Ae%2YE)C9;%=<*Mx02`|tU-Z5)kXpV30U{e47 zQZ_ZyO>e2&7>RVWj%u%>SKG77chd>G|W@^=wE>5r48?eJQv zRnt^=YtrTiGk9yFsW^Ggr(jCQGwE6*&MedXaApLuz4L=iF&OLTwGJT8N zFlp{7je-Eq?!ybMkeSB-@_4UOU=;+?bhXg3O>H;BEK1_E+1lBe>mDYAAZ8 zq0kbkLOR)~vjMln8h<;EF^FvpcOn(CeZ2psEPH=pUx++*14%4gL{ctBju~QSVV`H?F#3!wXoXu71Dn`|G>W{b zP)SxJvztU|p~9weoUj$PjGsv@CQ%>rk-1J4^G(g>*Agr8UHd+K92%oZ42oXk(*J-} zKjSw5czF!oyin{@yE3w`;Y@jkfp0jT2H_!l<_y_j?0YK8^aDZA-glY1!%9(ZUAUqLt&MLNYBT=({<0=PiKL^Pygu*|&g(V3kf&N3tJ*E} zVvTUZ6yZ%w{ClokB9CUsZc0IOojn`I3D%J#ow3W`au$Z3)1!Ct33?@{=2T>c^n zurJ`?NtjfQOBv)uYI`{BmoOi=;)08+{( zVGllRh%JrE^vqfhQ<@CzD)}n&?pf5RypR23g@=a&))51vl8M%Z^&qYvF<@?)uTQ;% zZq|I}g5pc85c+p&0umu@q7@W|80TNNq7}%f8ch{x-B=Dv(z!#<1_R2TefAz~7rh&J zsAF%~^B~$s zknaY8AyY4V-axO5pDBsGl@c1Hv4Qo|{W8~xpOu+4zhFEkUXnYIk3E3qsVhhmkR3{< zoN$w(e#ygx${ZIzQL#9MKQn#WKdA3G0L0ywHzyL>3_>>}Yc{|WE)njq*9=iHqmWoXM-HY@aWEnkLDYgH6~uDRO zG7FD}qy0>p^5`Y@Q5ss6)^yb`)oTYpz(Whwpcge(^9C!2(t3 zl52J4Gr%^^qr)oev$rRmS7%fL3J*Bs?uAFuN5%z*6B%R8FyTU<+N7}rf=dP>2t?If z=su$$HR3~`i#R4GyR3{IV!=;`?YNf-ssb1=3MrSWU!=mTu4vVEmYe4H zF5J9It4<$ERn~Cmj)_`E&_D69z6Y=I2X1;-ekfkEkZ%JtuQwcLQ;Y4u9%VHR zgSFCSh74?Tk=*5^j%dJ{8Nm|!zy}jd6_jdW@S=2QCspMS%H2a;#{9m(9S0iDm@CP2c&v(Zl}25t94z-}Gtahg{OJ zoBRo3W`u|#yhsFX_fj?Bi5BH?p>1jhs;5{R z5?}dR&3cWQcIpM`fq+R^@g_HlYihGBs(fO`{Cb^3eHejB{uY0`VL69msRrcADZAYK zuVthlGWhAa%0cmrvCN47Iu`1clgo}Ac^Q3v!Gdw@2Jx{IDYe^wLL6S zM!;33T*X)0!j2|>)T$#KXv-|{u#l>T{ht~COQ{058JlitLNIA!DC}RHB5D(L;X!%{ zg2L-KI@b9}G0C7bQ$Fx+*RU^-(FuLiRD2K%Y$6QReI7ZaIW;Jq26TmB`C1}o6i((g z;Dsc^=uxw-MF3}8EA_hk&-1R!wG%=?0s}bj(P%exM!m{%#~yhbXuZ*)*A6_;V*|3>%yH=MZe20x*hL)M|47T5`A?Ii{V)!>@$k zb$H9rsU-Tqp)g*xrSpqkpKV>$T4THqPbN8tZ7))+x*bxP@=L~0O=Pr3iOk?jw2lc_ zz{$3S%Fw@(q_poNR=iq0zKDMQIZz0pEm6f5HXuMT2&Gc$(4|rK^&#rfMFfPL_wClc z>U}JQNH*OkZW8QPO12uF=!}=e5CZhbw-MISXlZ;GnARV72ArB?cq!r_wi$gD`+h&f z#|DFUps45qEJvE}t;cCU%IqjiDpMV%} zeRuTO%oMB%LwO$4UP(-lYEEC@oSjM_0{CsrR`c3PFO8QY;)3@M=p{3Y>L2h@`mnX6 zFAr=Ux%Co8f$rL=#Z@BDw5GcfOb>7Ss6*!^+i>IyfS3cT{!eysJrmY=mdUejZ8#D_`m!g3Gm>G%OF1#p*yQ*kThYEz$amF8NcSc?JwpjH z<8(?qB!W6I{?-NB!^SkBIkj%1J0f6+no?%{`So&1asb0mSU45|gym zDDEtQ_=+3qz{#NoEM7#hu>SHxs8QFf_xZdvQ2=bwVMS(J zF{eEVf-<=%4=^d^KR$2n@8_d`yEJ!}6>omtjo%0J1U8;3;qS9{BVwGtUC&Q0HJz7s zZhoHMxBu*70765RIP}Hw0XHa2?R(Tnfep#(TVB=yZU7~7lwE~7gWiufw^Jrj+jLC+~%^0@2?(@x51CLU%iV@48z`kwe^6p-XMTx4gTb_ zFJygAfqI+%3VZDohlHlv{ES&W5K&}>at@|v6g6Jtn21cfh2#GHDgwDaS7_M&QaKG5 z-s)XsXY@3%iZ^akwtDsWi7^f8ZBuh1_BcEX`tTsv&A1uvFq`N5``PgS>}TZN{=<>P z|A$%X<+CmUbu9Q#MZ}WDFTa}+2Eh{)T_-^I^2MUzogVIy~i;W9z!JJ!!zKvI`0HmzqTYtncSA^kv(uXmZR&zc~% z%#+F4dKhbpjA1E?!Ph6OAd7OU=L7;Q;3g6<+5BRpl2L4eM{a`gMuX&nZFN>rbh|P` z`Jxc^M~aj){D3(oUSfy~zB$$QAzVC}FFMEy zFpo^APScn^!Z^~kR3t_rTOisH%>EBkazhDWMKApj5ebTbQ4egA-luf7E2EZ=FXswe zUxiYqU~hJ%mf#uU?iRu+8r?XI6|*UeMdmCK?8Ss;x* z2b5ZXRTzMXNt|wow60U&-h=y*+S1eU8-LQu-=S|HUiishb0C-}R$b>2V6%3fVqKS( zuVcJU(S;1*_VE_8t%#Hw#-vj%bLdT&`M{&>5<>!kYJc#YA8~u*82q%uV^gB4>ndmK5R3}LtIe%uVf%TU!+_sNOQY#wmu?L zik+_P%m{==G4Yr7x$YPauy%v=)_5(}eyALzm9P=;u+LC0?xKG6>~*hoS@fV%3kbp~ z=uIFF52!oc2peCe_8?`$&WqM3Ih%)jEsb%}sx_>kP$MrtC+s+kih*_SBgOeD?_PSc zF-mT6^QZKYuN`PdP`oznVLYiForc@D$Ka5ve_Zy1XG5WF$qr5cNVjNS-4A_hGoxpE zb_(d$*meNnrA%9Nl<-y9J{s_sg7C@Ojq>o7mmU1}ua8z7E3D1P==K;K*tIzn2kP~A zsE-9w=3pUIi>nu9{(WWOi>rw8w@TgCyIny04%a{aATI4^!9I#`ki*p`Ou14*`Ed>c zl7A&hWtVNrg4918uz&O~vRzEm3sS0BbflhXmNII}RC*C($$6(}bH&b*#&<1&g%2HOc7>%dF+gtsWX0Tv0I z4LksblE!HtKt^X(EVLK@V~)G{Po!lDCgx%(NBk`HIX!C77j=Y6SY`>Wuge#SzzDAP zrsMbDnjJkO8i-i8MShC_8R=)^gHR^(6*R6jDKCf_Pp*K+pB20K_tPF+KR^hLnQS@Z zq9f$r#i0|&Ii1L4NqW`bQea@QU?suJ2HrWBe+CH-fDMu`(m6Tfik3OS;p7LV(9h)=n1AE-Q7eEfRz1D-{csNCZ2`8-p{PCSej{4mJW=M<;)Rg+$K* zspyWafHA?sHgyikg19{p?sq2r=Rp?}p8<2C)U0 zia~e*#-ckwn1Slz#a!$TcsiGF4)05TOh=rtDK`dxve5cx#WbJm=x2;6N^8jY$>h5+ zLoD4CSet3nCz`ue5N|^tyWuQ_om@UrZ9x#BE`ULdbuG>=0w7;e8lJI^u#R!P>ISH3 z8MzRn$s)-?d@ZJML*|)B#CV1p)*7bN(*ukEVk54tO)iYF5?_bPKSsE`tD8or8Y0V` z8sKxJzRVd4(#8u^FkBU&6v7HHP(g)FhK)>QGWmdnvhjfgcHS_24PV=!9W7^WFf~8k7O3Ui=xhYurTnL9N==sb$jhKep?p6 zq=nuTd(om;|DJ_-@H$XtY9glZ9 zlwdvHMy@NWkbTah^$F>i9ikkv#E3*HsfSDa52nf38Rq=3Xxju>81okj}zS0C|US$LW zJ7-2H95R;TPQ$U2&X3fJoF}5nQ7*f(HiGPgIGPL0MWv79ID}s#0`l3Xh{<#&5@k7k{+ z13U0+CNNo`-+i$k^Hztn+uFE9WmscPq{WPoKN*KGi@@RHUL+50G{d?Jy?E0mc@4rUsTPn6z|B|2Q54fDUX_m zBb8gCxZOT`EJFlO6skAbET7ADSx%FM7_0LyS`g2cz#hQM|1O=M>XUkr%6f$~{cE6u zN^sX0R)x^i3OsIgpBLN*&FVlR00#7sUT_Y!J{*`FT1@{5%BRv@{RNo!JCwr7iDdZp`_d@a_#cPt3Vh1FV$8r2hr0L+y_MVQmEq?RR zlS#0FnGV24!Ae+Ds0wWvlIRTNJs=_dOX$g8pO*X!-_8+ozm^A>OH8sI%Af*@coEoG z-#N5c=xypFnG%hxIZ4F+)A7fU#G;b;xm<4gu~3?kiLrX1=e zVaO*Zdc8vj!c;fLHvaW{BA!bkAwTmTDb?omNsntA3oA}7cdQc81y{ozc?4T_7d^*o zP0vW7&T##Z@+cv--qm%|aXc%Cdw?h$gGjYy7$?2jp@mC%^ZM5XMa@~G>I#RHyacz-0_m%yN^lD*Ss!dYDHXEajgjuZqQ4in5Hg>)xtrC2 zE-ACu*W}945$O^tS+3y;@fbCsQwojIiUh#RhBLc@dE2_VUABk;ehD1ZXSA1lt@2u9 zXR9E7OMBJxl%i+y7dg1e?{iArzqro?SHudnOZa( z^bvXMT|Mfo*19z!eDa-f47ka*}O@d^K!i> zkf=m=i2f_>eLE-yDuhQ62(BXwRO)s&QFeFlX6e{x0Ux(dgMO}uXdhAJeeRVNFCJWX z5vp4w{x94`4|sF&7HwoOHZ%BYMkv6emX|D}P>3WG8jd6k&or6V5 z1tHbHWDrVBH>wihpk{@|`qGGQUg%mcEAF8D##&6Y^&t7>wTWI4a#E4_4^4n|yNsq< z>MR5E$|IZzl0~~pX2ko>G57mqcA>Dn4K43zF8-Xp#M2>fIW`mLKrTp#L**j$JNjB`7nd_SuqF*^We`^uc+UF6f`@0M|06dfoUIx&wTW#JLq^uR5e4M_#m zhnWsM;1^EM;&3UvS;(OW#hZi&FN+t(Jlckia5QQBP}H%3&tp^*-oAkopvNe7lxVE_ zf4edIiWebRhv!#sCGGFuUvH3F*?D44xHUPi|L&+jT(~0jIVRt^vJrrN#wpf?{u16Q zWoYvWc5_RjeqrK_|Z&_a15n)M@$>Jyf!=m)>qlqgAoBX7!A{t^y&uitx(KGx`cm-29D+1%s`uk3qHbs*McrrI84zK z79c~Xp+caT^5oa4uqpr%&7VgUTS6eo9j?`^NHXneoRk-)5;KQ%wd&zY=wKI^vcg%iqEcvgxbq_q&wSZMkD1tO1=5qLkQmPH@c23gS`>b7( zn1_>7B@d1xwzZ=LO0?p?sX&JaQepXcQ8+>?i*O}_tYdOOcdNKvIY&3a9iWlNXh_#! z5v&x3R?1IVH&p@L4Zdss9aARIlVJzo2@C#YeFQ$Txh>0`+yD<>FL#$6o2xG7>)w8F zTg7CSIl^$(7x|5&;3wVNCQh%^PNa?IFw9g3Wh$hi!<*2-}c zMUk1jJe$F7#Cwr4+kHYY5TlCQ2+>@<0ryX$J$+vtYNEs&f)RL(4wQFA->=21ygMk6 z_IwdruDmCI6Incpkl@WQ&hv;y+G%$aI}UN}vMa`~JP?wJnV{!amN0_TXf$|0XvZEQ zRP=+Pog4re5FV#BD&|s2^U#r9pP^o(bO=IU> zwy4Xe^msP+pRVla+1v_i8lDF^8y;TM_zvMVP8dSl?uy+l$jkpGIP*jsAFA^tzZJ)7 zP+4O5lHmxe66pV(dWZ_4WK?#|OGNo%0#|9p$@&IL9a>l^z5MH_RwDEqp!4oXq zkG}x?aZ#7~z#3Xlu69xlj2wMF3jxjQ!npd(pCpX+2CQ0!@f9MFegEd6cMyyjH%y=B zH#FR^|1|9=B}<)z*a+RL_*N^!9*3Vp>DGLZm;>UN)Kyk;;o^otd~Su`@M54X7@CHd zFC0pXipsgb`L>Hmp&IYsetAuZP)6oAB0vK^Z5j3_Wd*daz}WIsN@12DqTzF9t$6x9 zFu^g2>iws_wVE8uL%)KG_E9;jII}_$H&SOLJUcccF8lY5JQq}*aTB(Q>I?6@FH>`0 zUtjlb&xSQN*w=jc?!8u>FT4#Co!9Gd$0iY}q8u8M?z5sGaa7*)mT9$hW&hD&DQN%< z=qnSOD&;D3kr?yuVmcOiIHiPqiBt6YWLwpUA747Dwp!8mf}!3;DCTbK zMDT|r;Qc9h47Z07{1#FVtST*Wy&2iDaZe?QG4YtNEjjK04u^y?#wjrY+aCm&)Y+%@ zircNj!kS8IWA=*0fHI6^svw}SL&LwK5&#JroUe>Wc*oucAjDo4Kygu z`cgVeU7m7m`V8gQ%E~tSm3s+zGjPBaW!-X}evNQwTWbs-y7}F+f2ZMgIyfy=ysyh{ zV+-YKvVh}BXz4%3fA`lGTVM@^P-&UdwdsMUcU@bncGzUzD?SMtY=KBqUlU`Q9Vk&1 z+u~bIKx+3U)~LI+D`mYjD(C8;^vu8@Q4-&fuwJmhZ{JQlbAfmwiIrA)){vKD&o@v605txe6X-|@n1ySL-iyi|aa z-465xA}f9|10gyc@4>~f40m)*`swcuF?xSu8V72^PK_$vGjF67^8{RA!Payd-2Q6C z84s`t&-vW=z9}w2LJulHFZV&`UU`IEg;E?@;D_0m^?1ul_jJq7V+u_=-mC?Z%TAu4 zjnONAip4^KqJ+lF_ib*IYh(Lvl>BQZ(`xq);nqXTLI}rk5KpcZzGPagKL+&4K;lA( z`fp?bN1GG?F*e7KYmZ8h1{JtwncD&R9Smb%LD0#40V5Cj^Vh)(a#NN`w0A*vV{p&<6?7dm-TVK&s;V z1~|X?6TEfYmQM2zQC}u4C8$U9kJ|!xAYScB*hm|1S~mE|$NO8ViJc>jy@Tj`+V}GX zAR?<#EjQ9qP?HjHS--S%_hFbDQFw}5jA??U1n!tz^<~BG@%fy#iPF)c&N>cA^Z3>E z%T)Wh=TD%i5k)KUc&Bz6QHC&$LM42$YDlqnC2p*70ts~!NfeH9W05u3+3s1JSI+L4 zVg_SI6XU4_feK5_l97^^R3+K@t=`MkWlQUiBkIV`&o|4CVrVTS=_$gq6t_yAy^mwobt3g$rimEo6~WY3a6z{n$viFewMmm7ag z4H82#LTYY4^ZJ)M7l;@yhrbG?ipZf?Lu*W;tus6~-Ug z=vE6*$9_IgQy4k!*ua>jD+~*5F$+^aQ)bH?`D2L8zFjkST+GR5OM*)f*HgX{JuN;_ zU5axNBUxm?SLyaKQS8?hYmUxqBpoSg-&O!NDrVK^!h}I;Y+CF}DSL})pVfAMjD18L zz)VMP3cvu?0{e+rS^S)_APe*OlmOxW5`P9D)ikm;$|<~p`(U!5Ln3s2_El`c#*5E4KLl)zDD($QvO~Nb8f$4-!YCgB@U5hKQ3`bJxt|nbd0O_L+fEIk< zRjdoWa+;7O{!MB?^ga_y;=;*>oPSfDK~N0kBDqIvA-r8Rra;zbTQkN~b@_Zm2VM^@U&}RD438Ux6XL#8*+Ns>JK2(Q;S~mH2kyaur=iDYGD$#t=gdYM zU-DkLF6<4#6%|49Xvt*Io9W3=T+abq4^Q1#mx`{9mU5nHo*kDHucOsdSaUi>xnRWG zZ#4%}o|r09N$#Pbmk@=%V_6=ka;KrQhz#cindasP?0lIHM3I5{8|E>H7VU=?{ATQj zbi#i!%@y0pNv4Q1IAT-hM}+C{LV74}_rcOQ6=y@RHBn#Rz_9OSL}<&(UgZEB$c1{u ze;uRqh`~sJNvw)q5ipntPg{=zvQMN%SfH5%C|bWb`Z4SZxqQChc4LKXjZ2A}IED7e z=M6?CMZTwL7K|{>vJ);g^Tk-PRlSMkj(9Tr3lb5%8U!UBFL>NuupPdQ5O;6iLJVdR z>^Nd(f~+|S_@y5;9r`lU_LKpdp3j4lc$QK?`_#c)>3_J{KOla*=2(G$>f!>YAn7Ok zO#ak!><6TH)}cP~e0z!4f6voKG?|YU8Oa+O`0}v49G4YcKoC%k&9#JDg+!g30KZ>d zTWP7P-z1p!1ch%ri3SlNI1O!rF+Os}4M-%bL+zV`Ya{8qir_rc5}u&_J2&)y2zSG z)jHJhLhhkjUn7&Zio=8I9 zJcy|%SfV%Wg1!UqGWZi<7Y5^28=Z{|ZZ^AS2((IoXi#hX`Ah+AK3QjG(v(0Uy>2q6 zONN+9@(JpOAf7;#(GIjMS!0J(UUW}KB{sNnBJg?GYhZnj&&_E-A}{_uL3IDm&UUvk z(~6!_86vo=NpdSASDahw$waQ?d>hpir3irFGVbV&&I$2rOY4*I*q-pJLHFs9^+JE*KYv2 z-P5aF=T1y}zs3OM9hoj6%uFxg;H5$a!|w;6iUdbV)49TB_%WqDI!I&v_W9EUPplJ3 zbIqU=sWbeyF{I#KrDQBmgz3no)hHAp6>l0LGx4Y`5N;uB*Sgvmm3$X9!Cjw>(edMp ze~35v7Y7o{h^So=4)DfC6zBwV`B0YO1OyXSQr0gx&7=U7xShZnWnisR zf{oz1%@G0JMDt@JK5v>Is{hPL0-65PTwkB7uo!1c%H|szmc^IZk08GnW}}T=@~tP| z2@g+{6Jm1EvDu1lIf#E1K`B-}@|Sq;5EqnEWr)w&#c3Py3>_&8Qx^>WN#`BxAqJ-R z=(3M{k+r)I5?cTUe*qT9&IZyhi@1%mGq>tebSD5Je@%gvndR>RvHD72`{pN;8l4Od zdEHwAR?qR?#;Q*i^w8F%7+N_390j%yd$I=nGE?87crB8vBK4~!>%b&(pe?Kn++m1o zZr%Vm-?dzqGTxGmC-&utK}zZ{iw5=!)daq0V?|zbqIr!K7sjlAHZ0O5gt#7!Lj>%N zDmVaa3KzD)*TRt(>_g-j_C}3k=LMOCv_5H;QBSbX+xa-;bCNsXC(UQvFjET0TgI4J zk|p6D9iWdw@jC4K$E8)F)OyM|&}@K+>)$&Z0&pLeZ^mMUjuC3WhaMCXkv>cRp;s(L zH8;gQXVNKZMpR-goow)GtdwCJO&_%bb1ne7K~%GQ6)bnnv55=Km6`9N5>+{Mp z$}=F+ZAybpjrpoWI&mLPetIk(*k9|xm$6yKom`}nP^HISjGu0`f*n)N>~)n9s*~u0PkIc3s?VgAMRv z1LPft6-B#MpG^+@q_{kvNU8I7i6Uw4>CxQXmN2I-JvAyzz<_|e-yc*7Tyd6$Ur0V^ zZfJYP&HeC_aR;Tvt6(eT0Pj7-7FQeT1U!4iZU~shRs=7%uX$rZJzQ#5((p{(B9C{J zRmFxe@09`Bl7~#^d_+a$G;MwG%>m@$J^b{p&<`_*|Ngor-LmP{fca#Pm%!T4W8ujE zYy{gX*l+Il#CqCP6j>*QfeEVY%$vRHDA_u6Lr`rt_?3s5reUYzC8VsdC1aV3(RJJH zd7bh7tzAp1d0{)sd&ZAdz6ztwxj?jCfMS0uIy;cpu^IfwKeuc0`3CZn`W;}R)gkfc zy<~eXg?q>LjIO)k7Rlo%^NZqTGlY=9w*73ad4m~x{oQxxHw{?;()sB8 z_^M({QH*dKKaPU!{5u>WxGbRQVN{m2aITD%B>cp!aw;mbBsWe?;#h7S|td^s)S_KGFLgK-xWQJ*jk;Nk4;ZFDKKh3zI$1Vi4M)}qN_ z>P_vb0+0{#LTO)#grvU|@P;rZ7|n~Fpe%aEz<>M7Cj>+H #EDXx zz&QU^$tZBR=p^JrdGs+yDmt3k^`Qixx=V_{cQBuEn}lh@}P$ES9 z9cYIj^-D$y2smBh)}}=LEC2q?6AWLgd9ozL1R5$8-Xpd zLHdll)eMwn44e&S$GMc7au-ZZi>d$(8VL>c5W!a@3ofE8P4eA_IG!!Axgs?dnYe&=XCUVp)$H~YM}Q49Nn z$UE<;HA#bal|eqso(7$jj21MF(4}P8F|C4rnz)i*EUYK4mOGeGlG~DBID+qmxakx*1P~NqV`e=7cG#!FDUy+dNg0bRS3aUL|K=)ya-f-a}#c&<7PGVz0NJkuHDN>fwL@|x9+eWiOCF+De556bD0y;Zn~Dsp0UBc~R1IfP={e{uwFaUT z2ol09t^YzNY<=HAh44wuStd0Jz+|{ivkV1q>wXdUF4O5VFo0`u+_xBP#C0(HInw5&ZC53qq@MX@XAEWFX z7@Oe(8&^R#0B}4Uh<`#bWXf@r%h-$90dvjxYBDwITqK~KGU9A4=v7I=_^XJ)!SQwh zPrG+Uslkej(9V&OYE>k@{OY{l0% z#=9p&B>k2MR2)=3b5_CujB+^ZxRR#4SE%kPipbWLKcTYmv|;!w_Y zhm^w_0Sqv31%JW7sj&#-D_c23WLvc!BuMcc0)(Ul73d)mWcxSp;UF__ZL$h=uRcmU zAHPpjI1n5wJpA=HBBv}8<8dy7Uk!9|y6CGBjz3Xidt#dE_W1ojIVm9`jEEy(j8uoi zM(+NslKsxfm^iQ}4bm|Y#WZv=#&-d`UU46GJYR=mc%rmnK+H^L%Ze)66jTu`XqwL4 z1BkCq@~RzOlIxHqrR{DEL|W&Kq1c2NYg9KgY=? zP&Ewz@LXLo-#_g_=IEoSS;@MGJ-9@8&FIb>B z(=hg*tD~bFP!1`IQ8su7CxLd{e2C_DzoJ}>73t94Te*qJ)eUf~Chcs(WMRGgx#;W$ zeOR|jQs|{K)rtPzbtf6a%<65m``~^$cN>y%4t)wiQpt`^vt*EX*h-qmU$mWB17ctM zLqz7ly^_9P%fm9jE~>aYi)W?^qnHCC!nA-3nrwYFlOl{3YdE~gWJac$4E18XKw58n zIJ6S^XzpAElkm*k;mPU+2;BrMB)K+0U7MJVw>%p0RuOK8A@d}JRR2`-^yW?1 zB#$1PLDt8)I3+6-^ET^so-9Il*gzjEPlbP!e9!nUSClBi-8j^&YkQdCEC5KE!$G2+ z0zWvvkBDK*aXQ#WKL(sEolBI+L@y(>3KQ~vh)i}Aw(oBx z^>~`jg=yF=s8*o(8{%zmLTjLy2$|TU2=~dHiN$G+`vG6%%&co{4wXrJtfX?lRzvi2 zVy|>0H+qr(AdaesV$h-16E45GM_(9Jaa^u1pt0jR^ulV8>{$>@QGJ)nDmGlqZ+WxO zO@N}^A|fuKA{B=D4E`-2AlQ!aovRU<*-F%Xanf||l9}|ZX41ae=phvu4pHEAui90g z7hPj^q|mX(-Q`Yb@VcZL&=CLK)Xs;DUnBCg50?ajKmb>s`X2WiBtuj4N=bA}m#yx( zgyCR8=>~I(yj4Ky&)>G3!B!wFkSfO|A6jGT(ugx-N8G&H*@7@?fP9yc6Llw+Dc7L7 zM^qzhy8D{X_9-w^^3mvL^;x<(bE$*X7a9WY9;cW-?Fs0z!)wLMzGSx4ir;5wh-_*@|zns?Kak0xrCKjsdM_7@{*76kbJ=4~Y zkiQQ*@;gl8leMS2ALre}kr6+4-JY2a{dzt8jL<_R>W5xAQD8|gM0vJg%xN&hp1i0< z5EQm|0E6Kgjo7=Qej#^hPrFUF+UBSB(t+6P`&&7v@O@^#ah~4$pGsH=A|U1kZ5T}< z718_c1f#x{>rX%5%UzN&0Ue%z-OjJu;%?8Dk6lFm@Iu?LP|Vd#4XR`HV(ZIqt*5p3 zVSvY`bMrD903I&?(w@o6tvbhqgD239Bc>i3(4?!uBT1OEf-|wd*gn2&vjy-}wvq0) zddF@Sa4Vcp$nw}6&Zhs`@JyD;7sX&>Q1yFsxL!o$@3eT*vbs)8Vc-p0KHb!5NTc}r z7{s28)jhc8iQ!nX4~dR8B^7kX2khYoN;YM@ujA$j`;s%HGx~d<-N7#FejY&-w7chj_%6Z_I>4BWn?c?>ZGU{@-gY^X^YMOP zox6O){$Cvif4&cYFlnGbKsDJ|+(n>-|1bEcm+E-bx)@T&{V$xAaQqXy=Sc~l6<>Bc zebVH>T5cTn!P+2}2Oc#=xz`X+gD$c~mY@AZB#@uTRDRpf;Uo^AvRL!j*9tSoqKm%| zecmGcJO7ac%KH&F?64=$Eq9)+_edRiJLwE~FTAq!{B5Zh|3oK`6|16e&Y9-FpA=WT z@*b}iOZ_OL6=`l$At@$DVn0lC=!3bf^ICDTfP16J?vCRRV6U`;|Fkeh5xfg5#eT|PFC-dAZGUMwSX^#m_=Id7Q#~YdrkdG{XsrGkdQ-zXAY~6rfRoqM2vw{=gO_Ew* z-B_(k)dV7V9Y(nWZ+i`xD#vzL z<>98HRILPup&%#+3K2c!xg*I$H#JR{cIfcx?%bX{Z{PH)Fx5|xkrDgFeQA9{7vXII zgiyXW&GUGg_%(Im5A%<>LVrS#JnJ>-k5oeVlQ)fHBs|$O{s^KPxe+T=k=Vb|PDGIK z@#YcJPlgIqEWq7Kj;_WYuFnd2MeBr{vZvk!p~W1yan0y@~}jRbm0lsbHte@ zxP}dZIrp^&s4^$rLe;cbW##ekD6tx8!-0#*o+o*zL8nsBfSG#FeU zVnAdRE{5Xgzi|gGtqtWOqbo# z_igHCYi9CtldiTcO@UZooK*INOOMfwugzB`4JR=>p?@4cB*g37D8GBM-|F$))Ofk< zeH8kY;X{}mVG?S(9mvlTC!ei1IsKwHP%ForxolMUk9+A20z-RhUhFvJOY32j`K{;` z@js97{~8<5@)d#7{7;k6e~gXMp#N7Sl3U!V6ll%{$7XfL22na|k7$}=qtRIvxd*>%1Z1&sU3@ACw{~9iL zOB7auZV~-A0rgNNsQ&*+5c4zYjv1-~h0ytLJ5&Z8K9w035KxK+5RgpvZ9yd{RrWM# zH7M=>>wR|d=i45aBkl{34f1a>iH<-r6wP=v?2>L!4P|jGX=7ydi&)nEyu*ASDGDmo0tCAhM)Rw&*SH zLk`Ees(hZyjM5_SKd2p)QblNCIYJZ#NcG8O0i17^nNjCy;S-n)sUl&(3BXQ`{JLdw zP*P79LbNlZ!DE?d=Ep)eJp z0JvkKRJGy4Gg?{f^LirXKq8ODup}@^)UZ!vi*F;yGV^lvl;&{fN9OJowpViXK5it2 zXvV?6FW~lBTb#d_Bon@FBvL#vezD???YJitf-liZ?6Y#CXzN|g13EVu0z8-_W=~+5 zWa<@mm$x3|Qqbm5wA+I{DE)l_i=nCs02gbgl<4SzuZO6tZKnwhq0VR#!p)OuAVA%$ z0^h0?>@%Rb`!kRt^i-}A@B>Ww4@Ii(rzu2L2X$^DM5H!P9R|#45an;8#xV_g<%7>7z$I?( zY_;2Q4tfy{Q5v2i5opnE(GznVFJIm@nJ?KeqZCS+;P`a`@A4wROuubn6|Ayd>G~jb z0otTP`EEV5CSsEX9UyjA8bhD=&e|En2(&rKP7Drc8U#`&jyLgbV<+n2N zySd!xrnbYv@74F#U8P^FYXO;ioavdABOO}%o&SoyBW zDp0lktV#DUCe>q1(2j%Cc4sy$AawFf}1D*76EtVYQ_o^gRb zh6RJB6WU?Inin#uBd#V5X5nUO+AR3WP?lB>9oAi%Dno_OwCw$~Mm?oQ83FNslKFx4Ca+pE zy{G2k0fQLhOvJ+VPe`$WZ?B~oii&vBYCVe_&^dzp7Z?uE93h#H+zwqo^{u>u?Smi} zVUkuewYNh7j9}x7QFU{fUi>t}9~? zs0;U&cN8U%CR#U1rI9G})Wdl|DJQv6&L6e0^V_Ba#E{d;=0Jscdg0@IKv0Gv{cuj8 zOSTkHw`ZrjWapE2N*x1f*VJx0`V+X5PzF@Ncqu+9R?CenIj$qesJu&xpuS{` zFk##QIn*&irU+WBqAdoAXbKSvDOV=W)=2OkG;7J}E3IER+|5Ya?2Tm^JlB}YCm9?v zSw%HSaO?uH;;fqOI0amnX@Lf+eP7a`lNeEztGCqBiwa;+g+9Et?8SYl#Hu{x^@dZY3I~40VdYKUn%II> zvw>DFhKg4r&i(+CT|Z{hOJSrWE+V6nJ8kgPiQg>)mG{^H zSMJiZN4k?K@MtUI91Sre0&p(a@udV({U%Nz(8B(Asfp+9*dH=V%m1*@3;L13M-vDH z5TQ_o^YArZw~JPt)HB3nGB*5#QF!Rw@fhoJf4&s=aZSdt^#xh09=-{1{^g}l{X4Hj zk9seOW|$fQLnYnJgyi(mlU*z-6q-E-VD{q_N}*)r@TGK0Zy_R+(iXLq$zPA0nj1LF zts+FNlE&EQwI+ko)B%&PTL@A=0Q(Qk zfg~(Tws5gpTQ*MI+#;1yYo*KBaOR40{hr12FT>VuDk6o9u~)1*sgs>hP4*!%|A>U! zA)j4R{rb7@lwNMwkRcG{XZx?fP z8O4`3`HzMD;s!`$5<4Dv4wzCgz=-hgv@*4Y>c4^2f58d!t@3uJEjBHz(d0)b90_4Y z`M@3uI>awmIvQ%^EQU$b@ay<9%U*?nrcglsuB+aCX%l6%o4-5|DNVDp5P6)#N#Mg- zn$Mei{fXe#swB1RK+0G_q7}aIMMt#gjf8J@gZ;X!{N(0m0PXU=&ar2j7*T@&2I$c;wYPx?#4nUCU|#4Mw~0kfZgN}YQ6i;#Y4@(C zRd)J0GeLM!s$W`xByp;Sh?q0*pdkleM5A7{Vfe&N$SBDIn`u`EAklKo)du0#D9 z>3nff2K0m_kmN7oFt@<3*r7?J9+;%|x|qiHc~numP~8yBwghdaMaMBnQk#x)lK6Cl zMg;9FsP^#um}#t&Zea6O8R0PF=+1&?n`o6cuAH970nw0I01|{|+BjCfhjT zbMvU4;j#q(@tD#fd z>!Sgv6#*B^qxf(P04Z{+_g5z&2G@~Bawf4&j{QMXpcG?T-vLaVsgC4X)CPRL`H;